2013 Young Researchers Conference Proceedings PDF

Sponsors:
The Institution of Structural Engineers | Proceedings of the 15th Young Researchers’ Conference | 14 March 2013 1
Contents
Conference sponsors 4
Welcome 5
Keynote speaker 6
Conference team 7
Programme 8
Poster presenters 9
Synopses
1 Aaron Aboshio - Dynamic study and optimal design of offshore inflatable barriers subject to impact loading 10
2 Sheida Afshan - Efficient use of stainless steel in structural design 12
3 Samah AlHafian - Collapse behaviour of reinforced concrete frames using the applied element method 13
4 Alan Armstrong - Life cycle assessment of Irish residential buildings and typical building envelope analysis 15
5 Bogdan Balan - Development of a strategic design philosophy for structural glass 17
6 Fabio Battistini - Concrete-filled steel columns post-fire behaviour 18
7 Muhammad Bhutto - FRP composites for strengthening of steel bridge members 19
8 Domenico Cecini - Spectrum-compliant accelerograms through harmonic wavelet transform 21
9 Daniel Claff - Medial-lateral gait patterns in healthy adult walkers 23
10 Bernardino D’Amico - Form-finding and structural optimisation of timber gridshell structures 25
11 Hiep Dang - Pedestrian interaction with low-frequency structures 27
12 Marco De Angelis - Robust design optimisation of structural systems 29
13 Athanasios Dimopoulos - Seismic design, modelling and assessment of self-centering steel frames using
post-tensioned connections with web hourglass shape pins 31
14 Marco Dona - Multi-cracked beams with local and non-local elasticity 33
15 Shamsoon Fareed - Impact strength of steel pipes 35
16 Andrew Foster - Numerical analysis of the mechanical behaviour of composite sandwich panels 37
17 Rob Foster - Lifetime extension of reinforced concrete slab-on-beam structures 38
18 Joseph Gattas - Morphing origami panels: geometry and construction 40
19 Monika Grusova - Structural strengthening of existing reinforced concrete slab-on-beam structures in shear
using CFRP sheets and bars 42
20 Filip Gurkalo - Dynamic response of a novel form of water tower shafts composed of interlocked panels 44
21 Nawras Hamdan - A numerical tool for simulating dynamic soil-structure interaction problems 46
22 Aras Jaly-Zada - Building Information Modelling for improving collaborative design 48
23 Majd Khador - Structural behaviour of external diaphragm connection between steel I-section beams and circular
hollow section columns under cyclic loading 49
24 Payam Khazaeinejad - Thermo-mechanical analysis of shells using a hybrid-Trefftz formulation 51
25 Mohamed Kiari - Design of an FRP-reinforced concrete beam system for fire performance 53
26 Mark Kirk - Development and testing of a UK home-grown timber floor truss 55
27 Spyridoula-Iris Koltsida - Long-term fatigue deterioration and life expectancy of masonry arch bridges 57
28 Kaloyana Kostova - A new form of architecture for concrete structures 58
29 Rolands Kromanis - Detecting anomalous bridge behaviour using distributed temperature measurements 60
30 Ying Lian - Web crippling behaviour of cold-formed steel channel sections with web openings 62
31 Shuyuan Lin - Modelling of steel connections at elevated temperatures 64
32 Fernando Madrazo-Aguirre - Behaviour and design of under-deck and combined cable-stayed road bridges with
steel and composite decks 66
2 The Institution of Structural Engineers | Proceedings of the 15th Young Researchers’ Conference | 14 March 2013
33 Mohammed Mahmood - Bending behaviour of column face of concrete filled square hollow sections 68
34 Cristian Maluk - Study of heat-induced concrete spalling using a novel fire testing methodology 69
35 Navroop Matharu - Aspects of bolted connections in fibre reinforced polymer structures 71
36 Louise McCluskey - Blast resilience of large scale urban built infrastructure 73
37 Madison McDonald - Crowd synchronisation on grandstand structures 75
38 Emma McIntyre - Fire performance FRP reinforced concrete: bond deterioration at elevated temperature 77
39 Paul McNulty - Behaviour of a novel skew FlexiArch bridge system 79
40 Jesus Menendez - An innovative design tool for optimisation of timber structures 81
41 Muiris Moynihan - Under-utilisation of steel beams in construction 83
42 Hussam Nasreddin - Concrete failure: maximum tension strain as the controlling parameter 85
43 Tien Thuy Nguyen - Design framework for pultruded FRP beams in bending: a proposal 87
44 Darragh Noble - The dynamic effects of prestress loss on precast, post-tensioned concrete towers 89
45 Ross O’Donovan - Reinforced concrete deterioration of a 100 year old structure in a marine environment 91
46 Olayinka Oladiran - Innovative method for assessing shrinkage cracking of concrete and other cement-based
materials at early-ages: experimental and numerical modelling 92
47 Oluremi Olamigoke - Vulnerability of cable-stayed bridges 93
48 Daryan Othman - The influence of changing temperature on adhesively bonded FRP strengthening 94
49 Max Pearson - Investigations to determine the usability of basalt fibre reinforced polymer 96
50 Natalja Petkune - Assessment of modes of failure in hybrid steel and FRP shear walls 97
51 Klelia Petrou - Experimental and finite element modelling of protected cellular steel beams in fire conditions 99
52 Michael Qapo - Seismic performance of prestressed concrete girders strengthened in shear with CFRP reinforcement 101
53 Ansam Qsymah - Mesoscale modelling of fibre reinforced concrete considering random distribution of fibres 103
54 Jaylina Rana - Structural behaviour of concrete twin walls 104
55 Bo Ren - Noise and vibration from building mounted small wind turbines 106
56 Wenyuan Ren - Multi-scale stochastic fracture mechanics of concrete 108
57 leuan Rickard - A novel, cost effective and simple predictive test for design and optimisation of fire-safe
concrete mixes for tunnel linings 110
58 Ridwan Ridwan - Seismic performance of reinforced concrete beam-column joints strengthened in shear with embedded FRP bars 111
59 Mehdi Rouholamin - Collapse of bridges founded on piles in liquefiable soil: importance of proper representation of soil springs 113
60 Justin Russell - Investigation into the resilience of concrete flat slabs against progressive collapse 115
61 Erfan Shahabpoor - Vibration serviceability assessment of structures under intelligent interactive pedestrian loading 117
62 Jonathan Skinner - Ultra-thin toppings for upgrading existing timber floors 119
63 Holly Smith - Punching shear of reinforced concrete slabs in fire 121
64 David Stephen - Improving design efficiency of high rise structures in mitigating the potential of progressive collapse 123
65 Haolin Su - Properties of low cost concrete with recycled aggregate and tyre rubber for high speed train lines 124
66 Mahmood Tavallaee - High performance repair system for rehabilitation and strengthening of corroded reinforcement
in reinforced concrete columns 126
67 Ajibola Tijani - Low impact pre-stressed concrete sleepers 128
68 Eleni Toumpanaki - Long term performance of CFRP tendons in pretensioned structures 130
69 Osahon Umweni - Structural and sustainable assessment of deployable structures 131
70 Carolos Vidalis - Designing resistance against progressive collapse into framed structures 133
71 Wenjuan Zhuang - An energy approach for the robustness and damage potential of structures 135
Conference sponsors
Arup
A global firm of designers, engineers, planners and business consultants, Arup provides
a diverse range of professional services to clients around the world. With an innovative
and fully-integrated approach, their fundamental aim is to bring together the best
professionals in the world to meet their clients’ needs.
www.arup.com
Atkins
Atkins is one of the world’s leading engineering and design consultancies. We have the
depth and breadth of expertise to respond to the most technically challenging and time
critical infrastructure projects and the urgent transition to a low-carbon economy.
www.atkinsglobal.com
Flint & Neill

Flint & Neill provide innovative, advanced and sustainable solutions to a variety of
engineering projects and clients worldwide. Our creative thinking combined with sound
engineering knowledge and solid customer relationships enables us to deliver world-class
services.
www.flintneill.com
Ramboll
Ramboll is a leading global engineering and design consultancy offering services in
the UK in buildings, environment, transport and energy. Innovation - and a commitment
to sustainable development - informs all our work. We test the limits of design and
construction technology because we have a continual appetite for improvement.
www.ramboll.co.uk
S-FRAME Software (UK) Ltd

At S-FRAME, our expertise lies in developing & supporting innovative, easy-to-use and
reliable software for structural engineers in a wide range of industries around the globe.
S-FRAME’s new London office supports our UK clients and our strong UK academic
program. S-FRAME products include: S-FRAME, S-STEEL, S-PAD, S-CALC, S-CONCRETE,
and S-LINE.
s-frame.com
The Institution of Civil Engineers

ICE is an international membership organisation that promotes and advances civil
engineering around the world. Its purpose is to qualify professionals engaged in civil
engineering, exchange knowledge and best practice, and promote their contribution to
society.
www.ice.org.uk
The Institution of Structural Engineers Research Fund

The Institution of Structural Engineers Research Fund supports the Undergraduate
Research Grant Scheme, the MSc Research Grant Scheme, the Young Researchers’
Conference for PhD students and the Institution of Structural Engineers Research Award.
The fund occasionally awards other research grants. Details of all these schemes are
available on our website. www.istructe.org/about-us/donate/research-fund
www.istructe.org
Welcome
Since its inauguration in 1999, the Young Researchers’

Conference has become one of the highlights of the
Institution’s calendar, and it gives me great pleasure to be
asked to welcome you to this year’s event. As usual, the
programme includes a series of very strong contributions
on a variety of subjects from researchers examining issues
at the cutting edge of our professional endeavours. In the
past, delegates have been treated to a series of fascinating
insights, and I am sure that this year’s Conference will be
no exception.
As a consulting engineer often engaged in projects

requiring the application of engineering first principles and
occasionally dealing with the application of technologies
where merely applying codified rules and standards is not
an option, I fully appreciate the vital importance of thorough
ongoing research. Such research needs to be focussed on
the needs of the profession and developing technologies,
which is why it is so important to maintain a strong link
between research and practice. We all love to innovate; to
do something new. Indeed I would go so far as to say that
innovation is the lifeblood of any vibrant and developing
engineering practice. But without proper research,
innovation quickly stagnates and loses its way, and we
merely start repeating ourselves. Innovation needs to be
grounded in a sound understanding of the engineering
principles and technologies involved; otherwise it is just a
leap in the dark.
We rely heavily on our researchers and research

establishments, and it is wonderful that so many young and
enthusiastic engineers are applying themselves in this way
to develop our capabilities. So I am excited to hear what
our presenters will have to say at this Conference, and
confident that there will be important lessons to learn which
will help us to continue to innovate and develop new and
better ways to meet the structural engineering challenges
which will face us in the future.
Ian Firth
Vice President of the Institution of Structural Engineers
Keynote Speaker
Professor David A Nethercot

OBE FREng BSc(Eng) PhD DSc CEng FIStructE FICE FCGI
Prof David Nethercot OBE, FREng, was for 12 years until

the autumn of 2011 the Head of Department and is now
Emeritus Professor of Civil and Environmental Engineering
at Imperial College London; he was also Deputy Principal
of the Engineering Faculty with particular responsibility for
teaching. Previously, he was on the staff at Cardiff, Sheffield
and Nottingham Universities, including five years as Head
of Department at Nottingham.
He was awarded a DSc degree in 1993 and elected to the

Royal Academy of Engineering in that same year. He is
a past President of the Institution of Structural Engineers
and a former Council Member of the Royal Academy of
Engineering. In 2006 he was awarded an OBE for services
to Structural Engineering; the 2008 Charles Massonet Prize
from ECCS and in 2009 the IStructE Gold Medal. In 2010
he was elected as a Foreign Fellow of the Academy of
Technical Sciences and Engineering of Australia.
For more than 40 years he has been active in research,

teaching and advisory work in the area of steel structures.
He is the author of more than 400 technical papers,
including 10 that have won Institution prizes. A member of
the BSI committee responsible for the UK Steel Buildings
Code for more than 25 years, he chaired the committee
for 10 years. During that time he lectured on more than
200 specialist courses for industry, wrote 5 textbooks,
contributed to 5 design guides and wrote 5 chapters of the
Steel Designer’s Manual.
He has supervised over 40 PhD students to completion,

been responsible for a similar number of externally funded
projects, and spoken as a keynote speaker in more than
50 countries.
David chaired the Institution’s Research Panel for many

years and was instrumental in devising the first Young
Researchers’ Conference in 1999.
Conference team
Chairman: Ian Firth

Ian Firth – Vice President of the Institution of Structural MSc DIC FREng FIStructE FICE
Engineers and Chief Operating Officer, Flint & Neill
Judges:
Dr Mithila Achintha (University of Southampton)
Brian Bell (Network Rail)
Dr Anthony Darby (University of Bath)
Dr Leroy Gardner (Imperial College London)
Prof Costas Georgopolous (Kingston University)
Susan Giahi-Broadbent (Atkins)
Prof Abdy Kermani (Edinburgh Napier University)
Prof Dennis Lam (University of Bradford)
Chris Walker (Flint & Neill)
Dr Christopher White (EPSRC)
Ian joined Flint & Neill in 1979, and became a Partner in

1990 and Chief Operating Officer in 2008. Ian’s main areas
of expertise are in the design, analysis and assessment
of steel and concrete structures, particularly in relation to
major bridges and special structures, and the application
of new materials. He is well known for the innovative
engineering designs of several contemporary landmark
footbridges. He directs engineering projects, large and
small, around the world, and provides specialist advice on
long span bridge design, construction and maintenance.
He also directs the engineering design and assessment of
complex building structures and other projects. Ian is a
Vice President of the Institution of Structural Engineers.
Programme
09.45 Registration and coffee
10.15 Welcome by Ian Firth – Vice President of the Institution of Structural Engineers
10.25 Keynote address by Professor David A Nethercot – Imperial College London
10.55 Session 1
10.55 Under-utilisation of steel beams in construction (Abstract No. 41)

Mr Muiris Moynihan – University of Cambridge
11.10 Efficient use of stainless steel in structural design (Abstract No. 2)

Miss Sheida Afshan – Imperial College London
11.25 Discussion
11.35 Coffee
11.55 Session 2
11.55 Punching shear of reinforced concrete slabs in fire (Abstract No. 63)
Ms Holly Smith – University of Edinburgh
12.10 Investigation into the resilience of concrete flat slabs against progressive collapse (Abstract No. 60)
Mr Justin Russell – University of Nottingham
12.25 Designing resistance against progressive collapse into framed structures (Abstract No. 70)
Mr Carolos Vidalis – Imperial College London
12.40 Discussion
12.55 Membership matters
13.00 Lunch
13.40 Poster session
14.15 Session 3
14.15 Behaviour of a novel skew FlexiArch bridge system (Abstract No. 39)
Mr Paul McNulty – Queen’s University Belfast
14.30 Aspects of bolted connections in fibre reinforced polymer structures (Abstract No.35)
Mr Navroop Matharu – University of Warwick
14.45 A new form of architecture for concrete structures (Abstract No. 28)
Miss Kaloyana Kostova – University of Bath
15.00 Morphing origami panels: geometry and construction (Abstract No. 18)
Mr Joseph Gattas – University of Oxford
15.15 Discussion
15.35 Tea & judging
16:00 Prize giving & closing remarks
16:30 Close
Poster presentations
FRP composites for strengthening of steel bridge members (Abstract No. 7)

Muhammad Bhutto – Heriot Watt University
Numerical analysis of the mechanical behaviour of composite sandwich panels (Abstract No. 16)
Andrew Foster – University of Manchester
Structural behaviour of external diaphragm connection between steel I-section beams and circular hollow section columns
under cyclic loading (Abstract No. 23)
Majd Khador – University of Warwick
Development and testing of a UK home-grown timber floor truss (Abstract No. 26)
Mark Kirk – Edinburgh Napier University
Detecting anomalous bridge behaviour using distributed temperature measurements (Abstract No. 29)
Rolands Kromanis – University of Exeter
Modelling of steel connections at elevated temperatures (Abstract No. 31)

Shuyuan Lin – Brunel University
Study of heat-induced concrete spalling using a novel fire testing methodology (Abstract No. 34)
Cristian Maluk – University of Edinburgh
Fire performance of FRP reinforced concrete: bond deterioration at elevated temperature (Abstract No. 38)
Emma McIntyre – University of Edinburgh
Concrete failure: maximum tension strain as the controlling parameter (Abstract No. 42)
Hussam Nasreddin – University College London
Investigations to determine the usability of basalt fibre reinforced polymer (Abstract No. 49)
Max Pearson – Kingston University
Experimental and finite element modelling of protected cellular steel beams in fire conditions (Abstract No. 51)
Klelia Petrou – University of Ulster
Structural behaviour of concrete twin walls (Abstract No. 54)

Jaylina Rana – University College London
Multi-scale stochastic fracture mechanics of concrete (Abstract No. 56)

Wenyuan Ren – University of Manchester
Collapse of bridges founded on piles in liquefiable soil: importance of proper representation of soil springs
(Abstract No. 59)
Mehdi Rouholamin – University of Surrey
Vibration serviceability assessment of structures under intelligent interactive pedestrian loading (Abstract No. 61)
Erfan Shahabpoor – University of Sheffield
Properties of low cost concrete with recycled aggregate and tyre rubber for high speed train lines (Abstract No. 65)
Haolin Su – University of Birmingham
This booklet contains synopses from several students taking part in the 15th Young Researchers Conference, organised by the Institution of
Structural Engineers. The Institution bears no responsibility for the presentation or technical accuracy of the content in this booklet.
Dynamic study and optimal design of offshore inflatable
barriers subject to impact loading
01 Aaron Aboshio
Lancaster University
Background • To develop a material constitutive model that could be

used in finite element software that characterise the
Offshore oil and gas facilities alongside Naval facilities
membrane of the Fender barriers.
in oceans across the world has over the years suffered
a number of attacks from terrorist and militant groups
• Study the dynamic response of the components and
thereby leading to considerable loss of billions of dollars
system of the barriers against impact loads of various
of investment of these facilities. Example of such terrorist
vessel hull forms at different impact velocities.
attacks are the USS Cole in the port of Aden on 12th
October 2000 (Nixon et al, 2001), Exxon Mobil’s offshore
• Study the pneumatic fluid-structure interaction of the
production platform in Nigeria on 15th November 2010
barrier structure and the external response of the
(Sahara, 2010), the Royal Dutch Shell’s Bonga platform and
mooring buoy and anchor at sea bed before, during
its attending floating, production, storage and offloading
and after impact
vessels on 19th June 2008 (Jamestown, 2008) and a lot
of other losses incurred from terrorist attacks across the
Methodology
globe.
The following course of actions was undertaken:
In order to protect these facilities from concerted attacks,
effort by researchers and other concerns lead to the • Laboratory tests were carried-out on the composite
development of easily deployable counter-terrorist fence material samples based on Uniaxial, Biaxial and shear
around these facilities to limit small ship and boat access. modes of failures.
One of the common barriers being used around the world
is the Dunlop Ship Fender Barrier System (DSFBS). The • Using continuum mechanics theory and pragmatic
Dunlop Ship Fender barriers are inflated floating cylinders approach an anisotropic constitutive material model
made of rubber-coated textile, measuring 25 metres was developed and validated in Abaqus FEA
(82-feet) long by 2.4 metres (eight-feet) in diameter with
a central steel reinforcing rod and a supporting end plate. • A 3D, full numerical modelling of the Barrier system
They are linked together or to a mooring buoy round the was carried out in Abaqus CAE, using the pneumatic
protected facilities (Dunlop-GRG 2011). fluid structure interaction capabilities and coupled
Eulerian-Lagrangian Finite Elements mesh to simulate
Full scale test on the structural system, though carried the sea water, barrier and mooring system interaction
out on a smaller section of the barrier in 2001(Nixon et al, before, during and after impact under the following
2001) is apparently very difficult and expensive to perform. loading conditions:
To avoid the exceptional high cost of physical testing
and due to the complex geometry, loadings, boundary • Injected air pressure into the barrier structure
conditions and contact interactions of the structural to the currently arbitrary initial pressure of 1 psi
system, finite element method is herein adopted to assess (6894.76N/m2) and up to 2psi (13789.52 N/m2) at
the capacity and effectiveness of the structure as offshore constant temperature
barrier against small ships and boats.
• Small ship/Boat of 1874kg impacting at a
Aim and objectives maximum velocity of 24m/s
Aim
• Environmental loads of wind and water current
To assess the structural capacity of offshore fender barriers
against boat impact and environmental loads with a view to Results
optimise its design.
From the material test carried out and continuum
mechanics theory; the total strain energy density potential
Objectives
Results (U) in the neoprene and nylon composites used for the
• To understand the Uniaxial, Biaxial and Shear modes barrier membrane is obtained as presented in equation 1:
From the material test carried out and continuum mechanics theory; the total strain energy density potential
of failures of the barrier composite material
(U) in the neoprene and nylon composites used for the barrier membrane is obtained as presented in
equation 1:
(1)
(1)
WhereI1,I1I4,,II46, and
Where I6 and areirreducible
I8 the
I8 are the irreducible strain invariants
strain invariants
Results of initial simulation of a segment of the structure show that:

10 The-Institution
The of Structural Engineers | Proceedings of the 15th Young Researchers’ Conference | 14 March 2013
2.3m diameter fender is capable of resisting boat of 1847kg moving at up to 15m/s speed.
Results of initial simulation of a segment of the structure References
show that:
Dunlop-GRG (2011): Boat Barriers, Retrieved December
21, 2011, from Dunlop-GRG web site: http://www.
• The 2.3m diameter fender is capable of resisting boat
trelleborg.com/en/Dunlop-GRG/Products-and-Solutions/
of 1847kg moving at up to 15m/s speed.
AntiTerrorist-Boat-Barriers
• The internal pneumatic fluid structure interaction of the
Moneey, M. (1940): A Theory of Large Elastic Deformation;
barrier structure under a 21m/s speed boat impact
Journal of Applied Physics, published by the American
indicated a sharp increase in air pressure from initial
Institute of Physics, doi:10.1063/1.1712836
inflated pressure of 7kpa to 30.12kpa after impact.
Nixon L.P.E, Synnes D.M.S, and Brooks R. (2001): Full
• Von misses stress of 22.1Mpa at the time of initial
scale testing of Dunlop Fender Barriers, Unpublished
impact to a maximum of 244.1Mpa on the barrier
Draft technical memorandum, Naval Facilities Engineering
carcass was obtained for the 21m/s speed boat. This
Services Centre, Port Hueneme, California 93043 – 4370,
is well below the ultimate strength (failed stress) of the
USA
material as obtained in laboratory
Offshore Barriers (2011): Port Security Barrier, Retrieved
• Results of logarithmic strain on the structure due to
December 21, 2011, from Harbour Offshore Barriers web
the boat impact is shown in Figure 1
site: http://www.offshorebarriers.com/psb.html
Potential for application of results
Sahara Reporters (2011): Militants attacks Exxon Mobil
(1) Further parametric study on the capacity of the barrier offshore facilities, News and Report, Retrieved November
system under varied likely loading conditions would 8, 2011, from Sahara reporter’s web site: http://
identify weak zone(s) in the current design and an saharareporters.com/news-and-reports
optimal design of the structure of will be proposed.
Seelig, W.P.E (2001): Numerical Modelling – Predictions by
(2) A safe working inflated air pressure under varied William Seelig P.E, Unpublished Report submitted to the
environmental conditions typical of offshore sites would Naval Facility Unit, USA
be proposed as against the current arbitrary chosen
pressure of 1psi. Funding body
Petroleum Technology Development Fund-Nigeria
(3) A safe and effective mooring system of the barrier
structure would also be proposed from the study.
Further information
Aaron Aboshio (E: a.aboshio@lancaster.ac.uk) or Prof
Jianqiao Ye (E: j.ye2@lancaster.ac.uk)
Fig 1: Logarithmic strain

distribution of a 2.3m
diameter inflatable barrier
structure under impact of
a 1847kg boat moving at
21m/s
Efficient use of stainless steel in structural design
02 Sheida Afshan
Imperial College London
Stainless steel is being increasingly used as a construction The continuous strength method for structural stainless
Oral presenter
material in various structural applications, taking advantage

steel design
of its well-known corrosion resistance, fire resistance
and material properties. Given the high initial material Current international generation of stainless steel design
costs of stainless steel, associated primarily with its standards are based on elastic, perfectly plastic material
alloying elements, and increasing emphasis being put on behaviour providing consistency with carbon steel design
the sustainable use of resources, it is essential that its expressions, but often leading to overly conservative
distinctive properties are recognised in the development of results, particularly in the case of stocky elements.
structural design rules. Unlike the traditional carbon steel More economic design rules in accordance with the
which has an elastic response, with a clearly defined yield actual material response of stainless steel, which shows
point, followed by a yield plateau and a moderate degree a rounded stress-strain curve with significant strain
of strain hardening, stainless steel has predominantly hardening, are required. Hence, the continuous strength
non-linear stress-strain behaviour with substantial method (CSM) was developed. The CSM replaces the
strain hardening. This research project focuses on key concept of cross-section classification with a cross-section
characteristics of stainless steels’ material stress-strain deformation capacity and replaces the assumed elastic,
behaviour, in particular strain hardening. Methods for perfectly plastic material model with one that allows for
harnessing the strain hardening effects, occurring during strain hardening. Comparison of the predicted capacities
both the production stages and the in-service use of with over 140 collected test results shows that the CSM
stainless steel structures, are proposed. offers improved accuracy and reduced scatter relative to
the current design methods. The reliability of the approach
Strength enhancements in cold-formed stainless steel has also been demonstrated by statistical analyses.
structural sections
The CSM is currently under consideration for inclusion
Current stainless steel design standards employ in European and North American design standards for
material strengths that are significantly lower than the stainless steel structures. Also, it is envisaged that the
actual strength of the cross-section material, leading to proposed strength enhancement predictive model will be
conservative design and an inefficient use of the material. incorporated in future revisions of Eurocode 3.
This disparity in strength is due to the occurrence of
significant strength enhancements induced during References
manufacturing of cold-formed stainless steel sections.
[1] Cruise RB, Gardner L. Strength enhancements induced
The aim of this research is to develop simple and accurate
during cold forming of stainless steel sections. Journal
models for predicting this strength increase, enabling its
of Constructional Steel Research, 2008;64(11):1310-6.
incorporation in structural design.
[2] Rossi B. Mechanical properties, residual stresses and
Predictive models from the literature for determining the
structural behavior of thin-walled stainless steel profiles.
strength enhancements observed in cold-formed metallic
PhD Thesis. Belgium: University of Liège; 2008.
sections were reviewed. Two recently proposed predictive
models, developed by Cruise and Gardner [1] and Rossi
Funding body
[2], were assessed extensively. Improvements to the
existing models were made and a new predictive model Outokumpu Research Foundation, Steel Construction
is proposed. A laboratory testing programme was also Institute.
conducted to measure the level of strength enhancement
induced in cold-formed structural sections. Material tests Further information
on a total of 67 coupons, extracted from the flat faces and
Sheida Afshan (E: sheida.afshan06@imperial.ac.uk) or
corner regions of austenitic, ferritic, duplex and lean duplex
Dr Leroy Gardner; (E: leroy.gardner@imperial.ac.uk)
stainless steel tubular sections were conducted. The
generated tensile coupon test results, combined with those
from existing experimental programs, were used to validate
the predictions from the models and make comparisons
between the presented predictive equations. Statistical
analyses were carried out to ensure that the current level of
reliability of the European design standards is maintained
when the new predictive model is incorporated in design.
The new proposed model provides good predictions of
the test data and is simple to use in structural calculations.
Due to the universal formulation of the model, it is also
applicable to other metallic structural sections.
Collapse behaviour of reinforced concrete frames using the
applied element method
Samah AlHafian
Heriot Watt University 03
Introduction Parametric study
The accurate prediction of the potential collapse A deterministic sensitivity analysis using tornado diagram
mechanisms of structures under seismic loading is analysis has been carried out to invistigate the influnce of
crucial for collapse assessment. Several failure modes variation of material properties on the time at the collapse
can be triggered by earthquakes such as vertical and onset. In general, the structural collapse does not occur
side way collapse modes. Estimating the collapse due to the loss of a single or several structural members.
capacities of the structures and thus their safety margin Collapse occurs when the structure becomes unstable.
can be overestimated due to ignoring some of these Thus, the time at collapse onset is defined as the onset
possible failure mechanisms in the collapse assessment. of the unrestrained increase of either vertical or horizontal
Furthermore, modelling uncertainties can strongly influence displacements at a single or several storey levels. An
the collapse process. The Applied Element Method, artificial record has been employed for the seismic loads.
AEM used in the computer software ELS, allows explicit Two intensity levels namely 0.7g and 0.87g, which are
modelling of the progressive collapse as it automatically strong enough to initiate the collapse of this frame are
accounts for strong geometric nonlinearity, element adpoted. The random parameters of concern are the mass
separation and impact effects of falling debris. of the structure, the concrete compressive and tensile
strength, the concrete initial modulus of elasticity, the yield,
Project goals and objectives ultimate strength and the ultimate strain of the longitudinal
and transverse reinforcing steel. Different material
The first aim of this study is to validate the AEM for seismic
parameters have been used for the beams and columns
collapse simulation is. The second aim is to investigate
resulting in 40 random variables.
the effects of the uncertainties in the structural parameters
on the time of onset collapse and the potential collapse
The effects of modelling uncertainties on the time at
modes based on the AEM.
incipient collapse
The Applied Element Method
The results of the tornado diagram analyses provide
The AEM is a new method based on the discrete cracking insights into the relative importance of the uncertain
approach and developed by Tagel-Din and Meguro. variables, see Fig 2. The order of the importance of the
By using this method, the structure response can be variables varies according the seismic intensity level.
followed all the way to collapse with acceptable accuracy. The most important variables are those can change the
Furthermore, different collapse modes can be directly collapse process and lead either to more uniform damage
simulated. In the AEM, the structure is modelled as an distribution with a larger number of structural members
assembly of relatively small rigid elements connected involved in the collapse mechanism or to more damage
by sets of nonlinear normal and shear springs located localisation and a single storey mechanism. In general, the
at contact points distributed on the element surfaces, parameters that control the column-beam strength ratio are
as shown in Fig 1. These springs represent the element the most important variables.
stresses and deformations.
The effects of modelling uncertainties on the collapse
Validation of the AEM model
modes
The AEM model of a scaled six-storey three-bay reinforced
The dominant collapse mode is a first-storey collapse
concrete irregular frame with a discontinuous column at
mode. However, several failure modes have been observed
the first storey is validated by comparing the measured
from the sensitivity analyses. A fifth-storey failure mode and
storey-displacement histories obtained from a previous
two other side-sway collapse modes have been predicted
shaking table test with the AEM results at different levels of
namely (S3+4) and (S4+5). The former is a soft storey
Peak Ground Acceleration, PGA (e.g, 0.1, 0.3, 0.6, 0.9 and
mechanism over the third and fourth storey together and
1.2g). The results show that the AEM can reliably predict
the latter is a side-sway mode over the fourth storey and
the highly nonlinear response of this frame.
fifth storey together. As far as vertical collapse mode is
concerned, pancake collapse occurs due to the impact
of the falling beams on the lower parts of the structure.
The secondary effects are the primary contributor to the
collapse in several cases and play an important role in
determining the next structural members to progressively
separate in most of the collapse cases.
Conclusions
The Applied Element Method can predict the progressive
collapse behaviour of RC structures with acceptable
accuracy.
The results of the sensitivity study show that the seismic

intensity level is the most important parameter, followed
by the parameters that control the column-beam strength
ratio. Different failure modes involving side-sway and
pancake collapse modes have been predicted due to
variations in the selected variables. The collapse modes
strongly depend on the location and the type of the first
removed member. The impact forces resulting from falling
debris are the main cause of progressive collapse in several
cases.
References
Tagel-Din, H., and Meguro, K. (1999). Applied Element
Simulation for Collapse Analysis of Structures. Bulletin of
Earthquake Resistant Structure Research Center, 32, 113-
123.
Acknowledgment
The author is grateful for the use of the academic license
of the Extreme loading for structures software provided by
Applied Science International LLc.
Further Information
Samah AlHafian (E: Sa390@hw.ac.uk) or
Prof Ian May (E: I.M.May@hw.ac.uk)
Fig 2: Result of Tornado diagram
Fig 1: AEM model
Life cycle assessment of Irish residential buildings and
typical building envelope analysis
Alan Armstrong
National University of Ireland, Galway 04
Project objectives and goals Various wall, roof and floor building material scenarios were
studied in isolation to reflect typical Irish residential building
The objective of the project is to present results of an
structural features. Concrete masonry cavity walls and
economical and environmental life cycle assessment (LCA)
timber framed walls were found to be very similar in terms
of selected Irish residential building structural assemblies
of financial cost. Over the lifecycle of a building, a concrete
studied in isolation over a 60 year lifespan. Environmental
masonry wall with 150mm cavity rigid board insulation
LCA was conducted quantifying lifecycle operational heat
performs better by €5/m2 of wall compared to a standard
loss through structural envelopes and associated building
masonry cavity wall containing 100mm of rigid board cavity
material carbon footprint, in particular using embodied
insulation. LCA of timber roof designs demonstrated the
energy (EE) and embodied carbon (EC) as metric
significant negative effects of no insulation on the internal
indicators. With the pending introduction of Nearly Zero
environment and financial lifecycle costs. Selection of
Energy Building (NZEB) in Europe (European Commission,
analysed roof covering types found potential lifecycle
2010), more construction industry focus will turn to
cost savings based on assumptions. Material choice for
reducing a construction project’s lifecycle carbon footprint
floor coverings and floor insulation depth were found to
and special attention will be placed on construction
contribute significantly to both the lifecycle carbon footprint
material EE and EC. Therefore, this research emphasises
and financial cost.
studied structural envelope options which reflect poorly and
favourably from a life cycle energy and carbon perspective
in an Irish residential context.
Resulting from LCA of building envelope scenarios, project
Furthermore, economic analysis was performed on each designers may begin to consider EE or EC qualities in
assessed structural envelope to allocate lifecycle financial building material selection. Alternative building assembly
costs attributed to constructing each assembly. A structural systems could be further assessed from an environmental
envelope may perform favourably at the construction lifecycle perspective. A simple building envelope calculation
stage of a project but not be a worthwhile choice over mechanism could be developed for project design
the project’s lifecycle. Resulting from Irish residential calculations to be quantified in terms of EE and EC.
building envelope LCA, this research highlights the need
for designers to consider various lifecycle qualities when Further adaptation of project results could influence
selecting building materials. Providing life cycle economic Irish policy makers to introduce more legislative building
analysis on each structural envelope allows design based regulations regarding lifecycle energy and carbon usage
decisions to be communicated easily to clients. for building materials, with EE and EC intensities used as
potential environmental assessment metrics.
Description of method and results
Due to customer awareness and regularity bodies of on
Environmental LCA was carried out in two stages on
energy consuming material options, product manufacturers
various structural assemblies found in Irish residential
will be forced to adopt the most energy efficient production
houses. Firstly, EE and EC analysis was conducted in
methods in order to satisfy selection standards.
conjunction with the Inventory of Carbon and Energy (ICE)
database v2.0 (Hammond and Jones, 2011). Secondly,
operational energy and carbon heat loss values for various
building envelopes were quantified using calculated U
values. Economic analysis was conducted utilizing data
from Spon’s Irish Construction Price Book – 3rd Edition
(Talyor & Francis Group, 2008). Recorded prices were
updated to 2012 values using a consumer price index
(CPI).
References Funding body
European commission 2010. Directive 2010/31/EU of the Irish Research Council
Europena Parliament and of the Council of 19 May 2010 on
the energy performance of buildings (recast). In: UNION, E. Further information
(ed.). Official Journal of the European Communities.
Alan Armstrong (E: a.armstrong1@nuigalway.ie) or Dr Jamie
Goggins (E: jamie.goggins@nuigalway.ie)
Hammond, G. & Jones, C. 2011. Inventory of Carbon and
Energy (ICE) v2.0 [Online]. Available: <http://people.bath.
ac.uk/cj219/ > [Accessed 20th September 2011].
Talyor & Francis Group 2008. Spon’s Irish Construction

Price Book - 3rd Edition, Talyor & Francis Group.
Fig 1: Embodied energy and embodied carbon per surface

area values of various wall assemblies
Fig 2: Initial lifecycle heating and material costs per m2 of

wall based on Irish construction
Development of a strategic design philosophy for
structural glass
Bogdan Balan
University of Southampton 05
The unique combination of transparency, durability, References
relatively low cost, and the significant aesthetic potential
Achintha, M., Novell, D., 2011. Eigenstrain modelling of
of glass should make an attractive “structural” material
residual stresses generated by laser shock peening, J. of
for staircases, floors and roofs in buildings. Despite its
Mat. Processing Technology, 21, 1091-1101
potential, a limited research of “structural glass” was
carried out in the literature. A comprehensive research
Further information
which provides a good understanding of the fundamental
material behaviour of glass, mainly its brittle fracture, is Bogdan Balan (E: B.A.Balan@soton.ac.uk) or
required to develop routine design guidelines. Dr Mithila Achintha (E: Mithila.Achintha@soton.ac.uk)
The first objective of this PhD project is to correctly

predict the residual stresses generated in the glass during
the manufacturing process. This will be done using the
eigenstrain technique based finite element models which
taken into account of the misfit strains generated during
the heat treatment processes of glass. As shown by
Achintha & Nowell (2011), the eigenstrain technique is an
accurate tool to model residual stress states present in
materials. The key challenge during the first year of this
PhD project is the modelling of the misfit strains generated
in various types of glasses (e.g. annealed glass, heat
strengthened glass, fully tempered glass, etc.) during the
relevant heat treatment processes. Accurate analyses,
by considering actual physics involved in the process
of glass manufacturing will be developed. The model
predictions for the residual stress may be validated against
the experimental results determined using the “contour
method”.
An improved understanding of residual stresses present in

glass will enable the use of glass as a structural material in
the construction industry. A number of practical structural
designs will be developed during the latter part of this
PhD study. The project will also investigate a novel idea
of the use of glass-fibre-reinforced polymer sheets as a
prestressing layer in laminated glass. Laminated glass
will have an improved post-fracture response during a
breakage and hence it is appropriate to use in structural
applications. The current project will study the improved
strength and ductility of prestressed laminated glass
and will develop efficient novel structural forms over
conventional designs. The project will also study novel
manufacturing techniques for laminated glass, however
whether this work can be carried out or not will depend
on the availability of a glass laminating furnace within the
university.
Concrete-filled steel columns post-fire behaviour
06 Fabio Battistini
BRE Centre for Fire Safety Engineering, University of Edinburgh
Project objectives and goals filled steel columns, contact between the steel tube
and the concrete core is modelled by elements with
The objective of this project is to examine how composite
frictional behaviour between their two faces. The friction is
concrete filled steel tube, CFS, columns behave after
maintained as long as the surfaces remain in contact.
exposure to fire. Particular attention will be paid to the
overall mechanical properties of the column. The aim of
Validation of the numerical model
this work is to gain a greater understanding of the residual
post-fire behaviour of the columns and their response to an Sensitivity analyses will be conducted to evaluate the
axial load. It is hoped that it will be possible to model the effects of different inputs on the model’s accuracy. This is
columns well enough to be able to predict both local and necessary in order to be able to select the optimum input
global buckling. values for the model. The model will be created using
results from experimental tests carried out at the University
Much research has been carried out on the behaviour of Edinburgh. If possible, further experimental verification of
of CFS columns at ambient temperatures but there is the model will be carried out.
currently a lack of knowledge of their behaviour at elevated
temperatures which should be addressed in order to Potential for application of results
ensure safety during and after fires. . Currently research is
It is vital to know the post-fire behaviour of a composite
focused on the fire response of CFS columns in fire (Rush
concrete and steel structure in order to be able to properly
et al. 2010). There is very little knowledge of this area.
evaluate the residual strength and specify repair and
There is an even greater lack of knowledge of the post fire
strengthening methods post fire.
residual strength.
References
Rush D., Bisby L., Jowsey A., Lane B., (2010). Structural
The post-fire residual behaviour of a reinforced concrete
fire performance of concrete-filled steel hollow sections :
structures is dependent on the temperatures experienced
state-of-art and knowledge gaps. Interflam 2010, Volume
by its individual components and on the recovery time after
1, pp. 57-70
cooling down to room temperature (Drysdale et al. 1990).
For the purpose of modelling it is assumed that the steel is
Wang Y.C., Orton A.H. (2008). Fire Resistant Design of
at the temperature of the fire. The temperature through the
Concrete Filled Tubular Steel Columns. The Structural
concrete varies with depth (Wang et al. 2008).
Engineer, October, pp. 40-45
The target is to use ABAQUS to model the non-linear
Drysdale, D.D., Schneider, U., Babrauskas, V., and
buckling response of post-fire concrete filled steel columns.
Grayson, S.J. (1990). Repairability of Fire Damaged
Both square and circular cross-sections will be considered.
Structures CIB W14 Report, Fire Safety Journal, 16(4), pp.
251-336.
Material Model
The steel shell during fire is assumed to have uniform Further information
temperature and its mechanical behaviour is modelled with
Fabio Battistini (E: v1fbatti@staffmail.ed.ac.uk) or
two multi-linear stress–strain curves.
Dr Luke Bisby (E: Luke.Bisby@ed.ac.uk)
The mechanical properties of the concrete core after
fire depend completely on the temperature distributions
experienced during fire and the cross-section profile. In the
case of the circular columns the concrete cross-section is
divided into radial segments of equal thickness. Different
non-linear material models must be used based on the
post-fire material properties.
Load and Interaction

Different types of concrete have a different behaviour under
eccentric load. The available data shows that concrete
filled steel columns are sensitive to load eccentricity and
they can suffer reductions in load capacity under increasing
eccentricity. Therefore eccentric loads will be analysed in
depth.
In order to accurately simulate the behaviour of concrete-
FRP composites for strengthening of steel bridge members
Muhammad Aslam Bhutto

Poster presenter
Project goals and objectives Test results
Overview The ultimate loads of the FRP-strengthened specimens
were increased by 1.20 to 1.54 times, compared to the un-
There are several situations in which steel bridge members
strengthened control specimens. The failure of the fabric-
may require remedial measures to increase their strength.
strengthened specimens B3, B4 and B7 was initiated by
Due to the high strength, high stiffness, durability and light
breakdown of the bond between the steel and fabrics.
weight, fibre-reinforced polymers, FRP, composites have
There was no bond breakdown, at the ultimate load, in the
been used successfully to strengthen and repair steel
specimens strengthened using the pultruded sections.
bridge members. Little attention has been given to the
use of composite materials for increasing the strength of
Finite element, FE, analyses
thin walled steel sections, such as steel plate girders with
slender webs, where failure is initiated by shear buckling of Before testing, FE analyses which included both material
the slender web panels. and geometric nonlinearities were carried out using LUSAS
version 14.3 to predict the ultimate loads, modes of failure
Objective and load-deflection responses of the test specimens.
The objective of this research is to investigate the use of
Comparison of test and FEA results
glass and carbon FRP composites to increase the ultimate
strength of steel plate girders by increasing the shear The test and FE ultimate loads, modes of failure and
strength of the slender web panels. load-deflection responses for the un-strengthened control
and GFRP-strengthened specimens in groups G1 and
Description of method and results G2 were in good agreement. The test and FEA loads and
load-deflection responses for the FRP fabric-strengthened
Methodology
specimens in group G3 were in agreement up to
The method used was to carry out studies of un- breakdown of the fabric-steel bond that occurred only in
strengthened and FRP-strengthened steel plate girders the tests but was not modelled in the FE analyses.
using finite element, FE, modelling followed by the
experimental investigations of the test specimens. The test Potential for application of results
and FEA results were compared and conclusions were
The research carried out has provided an understanding
arrived.
of the behaviour of the un-strengthened and FRP-
strengthened steel plate girders. The test and FEA results
Methods used for FRP-strengthening
have been used to develop guidelines for the design of
By using GFRP pultruded sections as additional stiffeners FRP-strengthened steel plate girders and the design of
to the web panels, Fig 1(a). the GFRP pultruded section stiffeners. The study has
also helped in addressing the issues of bond breakdown
By applying layers of carbon and glass fabrics to the web between the steel and FRP composites, optimal orientation
panels, Fig 1(b). of GFRP stiffeners and effectiveness of using different types
of FRP composites for strengthening.
Test programme
Eight specimens, one un-strengthened control B1 and
seven FRP strengthened, B2 to B8, were tested. Based
upon type of the FRP strengthening of the end web panels,
the specimens were divided into three groups, G1, G2
and G3. Group G1 comprises un-strengthened control
specimen B1. In the group G2 specimens, two, B2 and
B8, had a vertical GFRP stiffener on each side of the web.
The other two specimens, B5 and B6, had a vertical and
a diagonal GFRP stiffener on one side of web respectively.
The three G3 specimens, B3, B4 and B7, had four layers of
the carbon and eight and four layers of the glass fabric on
one side of the web panel respectively.
Acknowledgment Further information
The author is grateful to Brian Bell, Network Rail, for Muhammad Aslam Bhutto (E: mb284@hw.ac.uk) or
providing assistance throughout the project. Prof Ian May (E: i.m.may@hw.ac.uk)
Funding body
Department for Transport (DfT), UK
Fig 1(a): Specimen B5 with GFRP stiffener Fig 1(b): Specimen B7 with 4 layers of glass fabric
Fig 2(a): Test of specimen B6
Fig 2(b): FE failure mode of specimen B6
Spectrum-compliant accelerograms through harmonic
wavelet transform
Domenico Cecini
Loughborough University 08
Project objectives and goals As first part of the study, the proposed deterministic
spectral matching method has exploited the HWT to
Modern building codes allow the analysis and design
iteratively introduce localised time-frequency modifications
of earthquake-resistant structures with recorded and/
in a given accelerogram until the updated response
or generated accelerograms, provided that they are
spectrum is fitted within the interval defined by given
compatible with the elastic design spectrum (EDS) used
bounding tolerances. The method attempts to minimise
to define the seismic action. Different methods of analysis
the impact of the modifications done on the recorded
of structural performance are allowed, depending on the
accelerogram by taking advantage of the fact that, once
specific design situation. The non-linear dynamic analysis
the maximum response value has been reached by a
is the most general approach, and offers very detailed
single-degree-of-freedom (SDoF) oscillator, what happens
information, provided that all the sources of non-linearity
thereafter does not affect the response spectrum’s
are accurately represented and the expected seismic event
ordinate, and hence there is no reason to modify that
is properly modelled with a set of spectrum-compatible
part of the record. In order to achieve this, the instant
accelerograms. The problem then arises to generate
at which the maximum response is attained must be
such set of accelerograms while maintaining realistic
identified, allowing adjustment of the record in the vicinity
non-stationary characteristics, which in turn may play an
of this time instant. Noticeably, while traditional methods
important role in the non-linear seismic response.
bring corrections localised in frequency domain only,
the proposed HWT-based method applies modifications
The Harmonic Wavelet Transform (HWT) has emerged in
localised in both time and frequency domain.
recent years in the field of signal processing as a potent
tool to describe and investigate the characteristics of
In the second part of the study, a stochastic generation
signals and has found many engineering applications. Its
method has been conceived, looking at the HWT as a
distinguishing feature is the capability of providing a joint
generalisation of the Fourier Transform (FT), capable of
time-frequency representation, which makes it attractive for
representing a signal by a compromise in terms of time
dealing with non-stationary signals like seismic records.
domain and frequency domain information. In fact, a
popular method for generating samples of stationary
On these bases, the project objectives were the use of
Gaussian processes is to sum a large number of harmonic
the HWT to (i) deterministically modify an accelerogram to
waves with random phases [2], and to modulate the
match the target spectrum and (ii) to randomly generate an
resulting signal to reflect the statistically assessed
arbitrary number of signals with the same non-stationary
frequency distribution and the time evolution of the
characteristics of the parent accelerogram. The overall
simulating phenomenon. Following this approach, the
aim was to demonstrate through numerical examples the
effect of a phase randomisation on the HWT transform
performance of the proposed approaches, which can be
complex-valued coefficients was investigated, and an
effectively used in the design practice allowing enhanced
efficient tool was devised to generate stochastic signals for
assessment on the seismic performance of structures.
earthquake engineering purpose. Importantly, the proposed
procedure requires a single parent accelerogram (which in
turn may have been previously made compatible by means
Wavelet analysis allows overcoming the main limitation of the proposed deterministic matching procedure) as the
of the classical Fourier analysis, which does not possess only source of information for the joint variability of the
localisation capabilities in the time domain. The wavelet random process in the time and frequency domains.
analysis consists of projecting the signal on a convenient
basis of functions, called wavelets, which can be generated Potential for application of results
by scaling and shifting a mother function. Unlike a
Besides the need for further studies, the results have
harmonic wave, which is an ever-lasting periodic function,
clearly shown the advantages of the proposed HWT-based
a wavelet is a decaying function, and this feature gives
methods, outlining the potential for an enhanced definition
to the wavelet analysis the time localisation missing in
of the seismic action and therefore for getting more reliable
the Fourier analysis. Interestingly, families of wavelets
output from the structural analysis. In fact, the matching
can be generated in a way to form orthogonal bases, so
procedure eases the practitioners’ task of using recorded
that the wavelet transform is bi-jective, giving a unique
accelerograms, still complying with code prescriptions. As
representation for any signal. This is the case of Newland’s
for the proposed generation procedure, it helps to deal
harmonic wavelets [1], which are complex-valued functions
with the inherently uncertain nature of the seismic action by
with a box-shaped Fourier transform.
adopting Monte Carlo analyses, as the procedure is meant
to generate the required time histories of ground shaking.
References
[1] D. Newland, “Harmonic wavelet analysis,” Proc. R.
Soc. Lond. A, vol. 443, pp. 203–225, 1993.
[2] M. Shinozuka and G. Deodatis, “Simulation of

stochastic processes by spectral representation”
Applied Mechanics Reviews, vol. 44, no. 4, pp.
191–204, 1991
Further information
Domenico Cecini (E: D.Cecini@lboro.ac.uk) or
Dr Alessandro Palmeri (E: A.Palmeri@lboro.ac.uk)
Medial-lateral gait patterns in healthy adult walkers
Daniel Claff
University of Oxford 09
Project objectives & goals of the population mean – a deterministic model – may
not be sufficient for capturing the unique fingerprint-like
The construction of long-span, low natural frequency
oscillations of individual pedestrians. Further work needs to
bridges subjected to densely packed pedestrian loading
be conducted to be able to quantitatively characterise the
has led to instances of significant lateral structural
degree of variability between individual samples.
excitation, a serviceability failure. To understand the
human-structure interactions that lead to such failures, the
The second series of results compares the gathered lab
dynamic nature of human loading during walking must first
data to an established deterministic model known as
be understood. Therefore, the goals of this research were
the inverted pendulum model. The inverted pendulum
two-fold. First, the study sought to characterise pedestrian
was developed over 20 years ago for biomechanical
loading as deterministic or stochastic by comparing
applications – primarily in the ante-posterior plane instead
individual force-time data plots to a population mean
of the medial-lateral plane – but has been adopted for civil
loading. Second, a commonly accepted deterministic
engineering purposes in the last decade by Hof et al. [1]
pedestrian model was applied to kinematic data sets to
and Bocian et al. [2]. The model allows for the proposal
analyse whether the model was a sufficient predictor of
that the normalised lateral force, F’L, is equal to the
population and individual lateral walking forces.
difference between the centre of mass (y) and centre of
pressure (u) locations, divided by the pendulum length, L
(approximated by the height of the centre of mass):
Kinematic and kinetic testing was undertaken at the
Nuffield Orthopaedic Centre, Oxford University Hospital F’L = (u –y) / L (1)
NHS Trust. The gait laboratory consisted of a 10 metre
indoor track, two AMTI force plates, and 12 Vicon Using the kinematic data from the gait lab, the inverted
infrared motion tracking cameras. Over 40 participants pendulum lateral force F’L can thus be calculated. These
walked a series of repeated trials along the track while results were compared to the gait lab kinetic data in order
wearing reflective motion tracking markers. The system to verify the inverted pendulum model. The model was
simultaneously recorded ground reaction force (kinetic) found to be inadequate for predicting the lab data. When
and anthropometric (kinematic) motion as each participant applying the model to all of the participant samples, the
crossed the force plates. Using Vicon Nexus software, population’s mean lateral force was over-predicted by a
data sets were omitted in which the participant did not factor of 1.25 to 5. Furthermore, where the lab population
land exactly one foot with no overhang on each force plate. data mean portrays a ‘w’ shaped lateral force path, the
This resulted in a sample size of 21 women and 17 men inverted pendulum calculations result in a simple concave
contributing 158 successful two-step trials, or a total of downward curve (Fig 2). This shows that the inverted
316 footsteps for analysis. The raw data was manipulated, pendulum does not represent the population data mean.
filtered, and reformatted for analysis using MATLAB 2011b.
Upon analysing the inverted pendulum model results for
The first series of results pertains to the lateral force individual participants, it is further seen that the model
patterns of population subsets in comparison to the mean is much different from the gait lab force plate data. The
lateral force of the whole population. The data means inverted pendulum did not predict any of the acute
of the men, women, all left feet, and all right feet were oscillations that characterised the precise stepping patterns
plotted against the mean of the whole population. Variation of the real participants. This suggests that the inverted
between the subset means was lower in the first half of the pendulum is insufficient for modelling the lateral forces of
footstep than in the second half. Nonetheless, the overall individuals; the model should be refined or replaced if it is
variation was quite low, indicating that the population mean to be used in broader design applications.
is a reasonable representation of each subset.
Conclusions
When the population mean was compared to the samples
The two primary goals of this research were to characterise
of individual participants, much greater variation was found
whether or not individual lateral walking forces could be
than between the subsets. Individual participants were
modelled deterministically, and to determine whether
classified as having either of two types of variation. In cases
an existing deterministic model could predict individual
where a participant’s forcing patterns did not fall within
lateral walking forces. Not only was it determined that a
one standard deviation of the population mean, their steps
stochastic model would be superior to a deterministic
were classified as ‘inaccurate’ (Fig 1). By contrast, when
population mean for predicting acute oscillations, but that
a participant’s force patterns were not consistent over
the inverted pendulum model is insufficient for predicting
successive trials, their steps were classified as ‘imprecise.’
lateral forcing behaviour. This research reveals the need
Different participants exhibited different combinations of
to develop a new or improved pedestrian loading model,
accuracy and precision in their lateral forcing strategy.
based both on empirical and theoretical evidence, instead
Overall, most pedestrians’ trials were relatively precise,
of theoretical reasoning alone.
but not accurate. This indicates that the broad ‘w’ shape
[1} A.L. Hof, M.G.J. Gazendam, W.E. Sinke, “The condition The Clarendon Fund (Oxford University Press)
for dynamic stability,” J. Biomechanics, vol. 38, no. 1,
pp. 1-8, January 2005. Further information
Daniel Claff (E: daniel.claff@eng.ox.ac.uk) or
[2] M. Bocian, J.H.G. Macdonald, J.F. Burn,
Prof Martin Williams (E: martin.williams@eng.ox.ac.uk) and
“Biomechanically inspired modelling of pedestrian-
Prof Tony Blakeborough (E: tony.blakeborough@eng.ox.ac.uk)
induced forces on laterally oscillating structures,” J.
Sound and Vibration, vol. 331, no. 16, pp. 3914-3929,
July 2012.
Fig 1: Body weight-normalised lateral force vs. step Fig 2: Inverted pendulum model left (red) and right (blue)
duration-normalised time. footsteps with kinetic lab data (grey) for the participant in
Fig 1. model fails to predict behaviour of laboratory data.
‘W’ shaped population mean +/- SD (black). One
participant’s left foot trials (red) and right foot trials (blue).
High ‘precision’ depicted by consistency of steps; low
‘accuracy’ depicted by oscillations outside of population
SD.
Form-finding and structural optimisation of timber gridshell
structures
Bernardino D’Amico
Edinburgh Napier University 10
Introduction The subsequent Project’s goal is the definition of a form-
improving procedure for timber grid-shells. Although
The double-layer timber grid-shell technique, despite
the term form-finding has been generally used to define
being able to achieve large spans with lightweight and stiff
those procedures applied to find the best/optimum
construction, has only rarely been used. The reason for the
structural shape regarding a defined objective (e.g. material
apparent lack of enthusiasm may stem from the unique
minimisation) here we prefer to call those procedures as
challenges associated with the design and formation
form-improving in order to highlight the occurring difference
process (Kelly et al. [1]) to deform an initial flat grid made
between the procedure described above and a method
of continuous straight laths joined together. To design the
which pursues to optimise one (or more than one) shape
Downland grid-shell (Sussex 2001) physical modelling
parameter.
of the structure was a central element of the design
process, which was used to determine the boundary
The third research’s objective is the definition of optimal
condition for the form-finding model (Harris and Kelly,
material and connection systems by implementation of
[2]). The final position of the boundary nodes detected
numerical and experimental models.
from the 1:30 scale model was used as target to the
imposed displacement of the boundary nodes of the flat
grid. However, the process of form-finding for grid-shells
is not straightforward and involves numerous interactions The form-finding tool is a script implemented by Python
[2]. In fact, as the input parameters for the form-finding programming language. The script presents two main
simulation come from physical model, a change in the functions:
shape requires the making of a new scale model and so
on. A method to bypass the need of physical models has • Parametric modelling of grid-shells within commercial
been proposed by Kuijvenhoven [3] consisting of a tool CAD software. This function allows quick modelling
which performs, in a first step, the forcing of a general and changes the grid-shell’s geometry, avoiding the
flat grid onto an imposed surface while in a second step, making of scale models during the concept design
boundary constraint are added and the exceeding grid and phase.
external forces (springs jointed between grid and target
surface) are deleted so the grid settles again to its final • Elaboration of the geometrical outputs previously
shape. Although this tool represents a novel approach to found and automatic definition of the FEA model for
the problem, it remains as general design method, not the nonlinear large displacement simulation of the
usefully applicable for professional practice. In fact only construction process. Once the 3D grid-shell model
grid-shell with a “continuous” and “plane” boundary curve has been defined in the CAD software, the script
can be performed, which means that no gates and no generate the equivalent flat geometry of the grid and
spatial boundary curve (like in the Savill grid-shell) can be corresponding boundary displacements values to
considered. perform the simulation in order to find the real shape.
Project objectives and goals The method for the proposed form-improving (structural
optimisation) will make use of genetic algorithms (GA)
The main objectives of the research comprise development
in order to maximize stiffness/strength performances
of a form-finding and form-improving (structural
while minimising the amount of material. The procedure
optimisation) procedure, as well as development of optimal
will follow the work of other authors regarding grid-shell
material and connection systems.
optimisation based on GA [6, 7]. However, no investigation
of this type has been carried out for grid-shells made of
The form-finding procedure is pursued by implementing
timber. For this reason, the methodological framework will
a tool on the basis of previous works [3, 4, 5] which
be the same of the previous works, but shape and residual
allows determination of the grid-shell geometry by simply
stresses due to the forming construction process will have
assigning an initial reference surface and parameters
to be taken into account.
regarding the grid pattern. Once the grid has been mapped
(drawn) on the surface, the correspondent flat silhouette
and boundary displacements (form-finding’s inputs) will be
automatically compiled and inputted to the nonlinear finite
element analysis (FEA) which simulate the forming process
so, giving as output the grid-shell’s final shape.
Fig 1: Grid-shell in Lecce, Italy. 2010. Large displacement FE analysis of the formation process.
Potential for application of results [5] Adrianessens S. Barnes M. R. & Williams C. J. K. 1999.
A new analytical and numerical basis for the form-
The potential application of the research can be
finding and analysis of spline and grid-shell structures.
summarised as follow:
In B. Kumar and B.H. V. Topping (eds.) Computing
developments in civil and structural engineering.
• Produce optimal solutions for free-form timber
Edinburgh: CIVIL-COMP.
construction systems such as grid-shell structures.
[6] Winslow P. Pellegrino S. Sharma S.B. 2008. Multi-
• Provide useful design tools to facilitate design and
Objective Optimization of Free-Form Grid Structures.
construction of complex structural systems.
International Conference on Engineering Optimization.
Rio de Janeiro. Brazil.
When dealing with timber grid-shells, the required design
efforts overcome the advantages in terms of structural
[7] Dimcic M. 2012. Structural Optimization of Grid Shells
performance and material efficiency and sustainability
based on Genetic Algorithms. PhD Thesis. ITKE.
(timber). For this reason, when designing large span
Stuttgart University.
roofs, designers often choose more traditional solutions.
The creation of supporting tools can increase designer
Further information
confidence in this type of structures, facilitating the
knowledge sharing between concept designer (e.g. Bernardino D’Amico (E: 40096810@live.napier.ac.uk) or
architects) and structural engineers. Prof Abdy Kermani (E: A.Kermani@napier.ac.uk)
References
[1] Kelly O.J. Harris R.J.L. Dickson M.G.T. & Rowe J.A.
2001. Construction of the Downland Gridshell. The
Structural Engineer 79: 25-33.
[2] Harris R.J.L. & Kelly O.J. 2002. The structural

engineering of the Downland Gridshell. In: 7th World
Conference of Timber Engineering. Shah Allam.
[3] Kuijvenhoven M. 2012. Particle-Spring Method for

form finding grid shell structures consisting of flexible
members. Journal of the International Association for
Shell and Spatial Structures. 171: 31-38.
[4] Toussaint M.H. 2007 A Design Tool for Timber

Gridshells. M.Sc. Thesis at Delft University of
Technology.
Fig 2: Grid-shell in Lecce, Italy. 2010.
Pedestrian interaction with low-frequency structures
Hiep Vu Dang
University of Warwick 11
Introduction the application of an impulse. Parameters of IPM used in
simulations are chosen from the typical range of humans:
Long span and slender civil engineering structures are
[4] body mass mh = 76.9 kg, pendulum leg length l = 1.045
nowadays more prone to excessive vibrations due to
m, and attack angle θ = 70˚. The model is set in motion
o
inconvenient dynamic properties often encountered in .
by being provided an initial horizontal speed ( xo= 1.55
modern structures made of light high-strength materials.
m/s). The simulation of IPM alone generates an average
These problematic properties are low mass and damping,
walking speed of 1.33 m/s with pacing rate of 1.86 Hz.
as well as natural frequencies coinciding with the range
generated by human activities. An important aspect in
The supporting structure is modelled as a single degree of
reliable assessment of vibration serviceability under walking
freedom system with following dynamic properties: modal
is to account for pedestrian interaction with perceptibly
mass ms = 10,000 kg, damping ratio ζ = 0.5 % and
vibrating surface. In the literature, Pedestrian-Structure
the fundamental frequency fs = 1.86 Hz (coincident with
Dynamic Interaction (PSDI) is described only qualitatively
the pacing rate of the walking model). The equations of
either through observations of changes in dynamic
motions of the combined system are set up and solved for
properties of the pedestrian-structure system [1, 2] or
every time instant. Fig. 2 illustrates the simulation results.
through variations in human-induced forces compared
with walking on rigid surface [3]. Due to lack of studies of
In the model that do not account for PSDI, the structural
walking on vibrating surface, PSDI is neglected in current
response in resonance would experience a steady build-
design guidelines, which often results in significant errors in
up until a steady state is reached. Fig. 2a, however,
vibration response estimates.
shows that due to PSDI, the acceleration of the structure
increases to reach 1.7 m/s2, and decreases to a lower
Project objectives
level of 1.2 m/s2. This result shows that PSDI could lead
To provide accurate vibration predictions, structural to lower than expected structural response, which is in
engineers need to model the combined pedestrian- line with previous experimental observations [3]. Through
structure system as close to reality as possible. The aim varying the initial condition, it is possible to simulate a
of this project is to develop a numerical model that takes range of walking conditions using IPM. Fig. 2b shows the
into account PSDI in the vibration assessment of lively pedestrian’s pacing rates and the corresponding maximum
structures. accelerations of the structure. As a result, the simulation
of the pedestrian-structure system has the potential
Methodology and results application for vibration assessment of lively structures.
The traditional approach to separately model structures
Further works
and human-induced loads neglects the dynamic
interaction. A better strategy would be to model the To further improve modelling, the IPM is currently being
loads and the structure as two components of a dynamic upgraded to a model that better depicts the walking
pedestrian-structure system that interacts with each other. mechanism. This model consisting of spring-like legs
In particular, the loads induced by the walking mechanism and rocker feet is shown to better represent the human-
are influenced by the vibrating surface [4, 5] and the induced loads. The modelling will be followed by
pedestrian body dynamics influences dynamic properties of experimental verifications of numerical simulation in well
the joint system [5, 6]. controlled laboratory conditions on a 20-metre long bridge
and with the use of a motion capture system to record
The simplest walking model in literature that can be used walking kinematics and kinetics of test subjects.
to model PSDI is the Inverted Pendulum Model (IPM),
which contains a point mass, representing the body’s
centre of mass, on top of massless rigid legs (Fig. 1). Even
though IPM contains many shortcomings, [4] its simplicity
is of great help to promote understanding of PSDI. The
simulation starts when one leg makes contact with the
ground with a specific attack angle. The centre of mass
vaults over the ground, making an arc-shaped trajectory.
By the end of each step, the mass has the tendency of
moving downwards and is redirected to the next step by
References [6] Caprani, C.C., Keogh, J., Archbold, P., and Fanning, P.,
Characteristic Vertical Response of a Footbridge Due to
[1] Brownjohn, J.M.V, Energy dissipation from vibrating
Crowd Loading, In Proceedings of the 8th EURODYN,
floor slabs due to human-structure interaction, Shock
2011.
and Vibration, 8(6), 2001.
Funding body
[2] Živanović, S., Diaz, I.M., and Pavic, A., Influence of
Walking and Standing Crowds on Structural Dynamic EPSRC (project Pedestrian Interaction with Lively Low-
Properties, In Proceedings of the IMAC XXVII, 2009. Frequency Structures)
[3] Živanović, S., Pavic, A., and Reynolds, P., Human– Further information
structure dynamic interaction in footbridges, Bridge
Hiep Vu Dang (E: h.v.dang@warwick.ac.uk) or
Engineering, 158(4), 2005.
Dr Stana Živanović (E: s.zivanovic@warwick.ac.uk)
[4] Bocian, M., Macdonald, J., and Burn, J., Modelling of
self-excited vertical forces on structures due to walking
pedestrians, In Proceedings of the 8th EURODYN,
2011.
[5] Qin, J.W., Law, S.S., Yang, Q.S., and Yang, N.,
Pedestrian–bridge dynamic interaction, including
human participation, Journal of Sound and Vibration,
332(4), 2013.
Fig 1: Pedestrian-structure system.
Fig 2: Simulation results of structural response (a) for pacing rate at resonance and (b) under a range of pacing rates.
Robust design optimisation of structural systems
Marco De Angelis
University of Liverpool 12
Over the last few centuries, one aspect that has been The aim of the project is to illustrate how uncertainty
characterising the engineering progresses in science quantification can be applied on computer models,
and technology is the ability of engineers to develop regardless the number of parameters involved (problem
models capable of reproducing the behavioural features dimension), in order to provide a measure of the variance.
of a real object. This ability has revolutionised the way Further, analysis of variance may be performed in order
of constructing, with appreciation from the whole public to identify the set of parameters that mostly causes
society, since losses and costs related to manufacturing the system to vary, and to establish to what extent the
and construction process, might be substantially reduced. variance may be reduced by changing the settings of
the experiment. Once the analyst is given the freedom to
Today, thanks to the advances of high performance vary the numerical values that input the model, by means
computing technology, engineers are able to reproduce of specific strategies, it is even possible to look at the
rather complex structures and systems to a detailed combination of values that minimises the outputted amount
degree of accuracy. The idea of using models - also called of uncertainty, thus achieving robust design.
virtual models in the context of computer technology –
for representing real objects has been widely embraced The case of study is a fatigue prone system of riveted
by industries in the fields of automotive and aerospace connections composing the flooring system of a steel deck
engineering, with the intent of removing prototypes, which truss bridge. Repetition of high load cycle (induced by
are too costly and hard to manipulate. The use of advanced traffic over the bridge) causes the connection (made up of
computer techniques not only has allowed reducing losses steel clip angle) to trigger and to hazardously propagate
related to hazardous construction processes, but also has cracks. The amplitude of the cracks depends on the stress
promoted interesting questions regarding the optimisation intensity - which the connection is subject to - as well
of the process itself and the prediction of costs spread over as the number of cycles of such intensity. The number
the system’s life cycle. Virtual models, mainly based on of cycles makes the problem time-dependent if one
Finite Element Analysis (FEA) models, are widely used also considers the average daily traffic (ADT) over the bridge.
for the design of maintenance programs of systems prone The complexity of the problem stems from predicting
to damage propagation, such as corrosion and fatigue. the state of the system, such as the cracks amplitude
at a given time, in order to decide whether a mission of
The large use of virtual prototypes has brought into inspection is required or not. To ease the decision-making
question to what extent engineers can rely on such process, a computer model of the connections system is
models, as far as the uncertainty is ubiquitous. Since constructed for the purpose of computing the probability of
virtual prototypes may be always defined as an input/ failure of each component, which is given by the probability
output relationship, one way to deal with uncertainties that the crack amplitude exceeds the thickness of the
consists in treating the input quantities (also named clip angle it is growing in. Once the probability of failure of
parameters) as uncertain quantities. This may allow the system as a whole has been obtained by working out
either investigating the sensitivity with respect to the each component, and once a level of reliability has been
parameters or studying the statistical properties of the established in liaison with the bridge owner’s specifications,
model itself. As long as the problem is defined by a a program of inspection can be decided. During an
given set of numerical values, computer calculations are inspection one may either detect or not detect any crack
possible no matter how complicated the problem. But whatsoever depending on the size of the crack and on
if one wishes to find the changes in output when initial the quality of the inspection, leading to the definition of
conditions, boundary condition and parameters are varied, probability of detection. In case the inspection succeeds
the number of combinations becomes enormous and the in detecting the crack, the considered connection is
calculation is impossible to carry out even by the most replaced with a new one. Therefore, in the planning stage
powerful computer. Therefore, it is essential to design the question arises as to what type of connection one shall
specific strategies for the simultaneous arrangement replace the old one with, in order to optimise the program
of all parameters, in order for the investigation to be as and to minimise the number of inspections. This problem
informative as desired. An attractive way of performing this is addressed by preparing a set of different solutions,
analysis is provided by methods of experimental design, characterised by different options, such as thickness of the
introduced by the original work of Fisher R.A. “Design of clip angle and number of rivets, which are tested further
Experiment” 1951. The experimental character of such on the virtual prototype, in order to select the optimum
studies is related to the fact that many executions of the design. The response in terms of stresses is the result of
same computer model with different settings are needed. the loading history over the bridge filtered by the clip angle
The concept of running experiment by means of computer connection. The characteristics of the response in the time
is not a novelty and the literature usually refers to this domain, such as amplitude and number of oscillations,
procedure as to desktop experiment. affect the fatigue behaviour of the whole system. The
optimum design of such a system is the one capable of
reducing the amplitude of oscillations filtered by the single
connection, thus the one that minimises the variance of the
process.
Further information
Marco De Angelis (E: marco.de-angelis@liverpool.ac.uk) or
Prof Michael Beer (E: mbeer@liverpool.ac.uk)
Seismic design, modelling and assessment of self-
centering steel frames using post-tensioned connections
with web hourglass shape pins
Athanasios Dimopoulos
Project objectives and goals The connection performance was experimentally validated
under quasi-static cyclic loading and a nonlinear model of
A challenge of modern earthquake engineering is the
the proposed connection was developed in OpenSees,
development of resilient minimal-damage structures that
compared against the experimental data and found to
can minimise inelastic deformations and residual drifts. A
accurately simulate the connection behaviour Fig 2(a). A
number of minimal damage structural systems, including
5-storey steel frame was designed both as a conventional
frames equipped with passive dampers [1], and self-
MRF and a SC-MRF with WHPs, modelled in OpenSees
centering moment resisting frames (SC-MRFs) [5] with
and subjected to pushover and dynamic analyses. Both
post-tensioned (PT) connections have been proposed.
the analytical and the experimental results show that the
proposed connection eliminates residual drifts and beam
SC-MRFs eliminate inelastic deformations and provide self-
damage for drifts higher than those correspond to the
centering capability under strong earthquakes as the result
Design Basis and Maximum Credible earthquakes (DBE,
of gap openings developed in beam-to-column interfaces
MCE) Fig 2(a), Fig 2(b).
and elastic pre-tensioning elements (e.g., high strength
steel bars) which clamp beams to the columns. In addition
energy dissipation devices (EDs) are inserted in the PT
connections which are activated when gaps open. Several The proposed PT connection with WHPs avoids damage
EDs have been proposed, which in general dissipate in other connection parts as well as in beams and columns
energy through inelastic deformations [2,3] or friction [4]. and eliminates residual drift. For this reason the design of
a SC-MRF with WHP tends to be very promising against
A new steel PT connection using web hourglass shape that of a conventional MRF which results in inelastic
steel pins (WHPs) has been recently developed and deformations and residual drifts, due to the development
experimentally validated under cyclic loading tests [5]. This of plastic hinges at the ends of the beams and at the base
connection was used for the design of a 5-storey SC- of the columns. The use of WHPs also provides many
MRF, and a nonlinear model of this frame was developed advantages against other proposed EDs, because they do
in OpenSees. The seismic behaviour of the SC-MRF was not interfere with the composite slab and are very easy to
assessed and compared against that of a conventional replace without the need for bolting or welding.
moment resisting frame (MRF).
The benefits of the proposed self-centering systems
Description of method and results include sustainable construction and superior minimal-
damage seismic performance. Therefore, the proposed
Fig 1(a) shows an exterior PT connection with WHPs. Two
systems have an inherent potential to meet all the
high strength PT steel bars are located at the mid-height
important socio-economic requirements of modern
of the beam, pass through holes drilled on the column
societies worldwide and address future challenges related
flanges and anchored to the exterior column flange. Four
to sustainability and resilience.
WHPs are inserted in holes on the web of the beam and on
strong supporting plates, welded on the column flanges.
The beam web is reinforced with steel plates to increase
the contact surface of the WHPs with the web. Also beam
flange reinforcing plates are installed in the connection to
prevent early yielding in the beam flanges under the high
PT bars forces.
When the axial tension force of the PT bars exceeds

their initial PT, because of a moment, which is lower than
the plastic moment of the beam, a gap opens between
the beam and the column. In this way any inelastic
deformations at the beams are avoided. Fig 1(b) shows the
gap opening mechanism in the connection where d1u and
d1l are the distances of the upper and lower WHPs from
the centre of rotation (COR); d2 is the distance of the PT
bars from the COR; FPT is the total force in both PT bars;
FWHP,u and FWHP,l are the forces in the upper and lower
WHPs, respectively; and CF is the compressive force on
the beam-column bearing surface. After the gap opening
the WHPs are subjected in flexure which results in yielding
of their material, hence energy dissipation is provided.
References Further information
[1] Karavasilis TL, Kerawala S, Hale E (2012). Model for Athanasios Dimopoulos (E: A.Dimopoulos@warwick.ac.uk)
hysteretic behaviour of steel energy dissipation devices or Dr Theodore Karavasilis (E: T.Karavasilis@warwick.ac.uk)
and evaluation of a minimal-damage seismic design
approach for steel frames. Journal of Constructional
Steel Research 70:358-367.
[2] Christopoulos C, Filiatrault A, Uang CM, Folz B (2002b).

Post tensioned energy dissipating connections for
moment-resisting steel frames. Journal of Structural
Engineering 128(9): 1111-1120.
[3] Garlock M, Ricles JM, Sause R (2005). Experimental

studies of full-scale post tensioned steel connections.
Journal of Structural Engineering 131(3): 438-448.
[4] Rojas P, Ricles JM, Sause R (2004). Seismic

performance of post-tensioned steel moment resisting
frames with friction devices. Journal of Structural
Engineering 131(4): 529-540.
[4] Vasdravellis G, Karavasilis TL, Uy B (2012). Large-

scale experimental validation of steel post-tensioned
connections with web hourglass pins. Journal of
Structural Engineering, in press.
Fig 1(a): Exterior PT connection details Fig 1(b): Gap opening mechanism in a PT connection
with WHPs
Fig 2(a): Comparison of experimental hysteresis and model Fig 2(b): Statistics of residual storey drifts of the
in OpenSees conventional MRF and the SC-MRF under DBE and MCE
Multi-cracked beams with local and non-local elasticity
Marco Donà
Loughborough University 14
Project objectives and goals the stress-strain field around the crack tips, cannot be
removed by using classical theories of elasticity, leading to
The presence of cracks is an important issue in
unacceptable results. Based on such considerations, an
many applications of civil, mechanical and aerospace
enriched formulation for the two-node MCB FE is currently
engineering, as they can occur in the load-bearing
being developed in which, more realistically, the stress
elements, induced by material defects, installation errors
state in the damaged member is a function not only of the
as well as external loads, impacts or fatigue. The aim of
point value of the stress, but also the associated gradients
the PhD project is “to develop a computational method for
(Stage 2).
the static and dynamic analysis of beams with non-local
elasticity in presence of crack singularities”.
An extensive review of the existing literatures concerning The stiffness matrix of the MCB FE formulation has been
methods of damage investigation has been carried out, derived for both Euler-Bernoulli and Timoshenko kinematics
focusing mainly on the different models used to describe models, by imposing static unitary settlements to the
the damaged structures. This survey has allowed exact closed-form solutions of the differential equations
identifying fields not yet covered by the current research ruling axial and transverse deflections. For this purpose,
(Stage 0). the flexibility modelling, where the damage-induced
singularities in the beam’s flexibility are represented by
Cracks are usually represented in the computational Dirac’s delta functions, has been adopted. The equivalent
model with a localised reduction in the flexural stiffness, load vector and both lumped and consistent mass matrices
which is macroscopically equivalent to a beam where the including rotatory inertia have been defined by using the
two undamaged parts are hinged at the position of the same shape functions adopted for the stiffness matrix, and
damage, and a rotational spring takes into account the the results of the proposed approach for both slender and
residual stiffness. This discrete spring model (DS) is also short beams under static and dynamic loads have been
equivalent to an inhomogeneous Euler-Bernoulli beam validated against those of a commercial FE program (Stage
in which a Dirac’s delta function is added in the bending 1).
flexibility at the position of the crack: that is, the severer the
damage, the larger is the resulting impulsive term. Given Also, the enriched constitutive law has been included into
the existing lack of a computational model able to describe the closed form solution for the deflection of multi-cracked
with a single beam finite element (FE) the presence of beam. The differential equation so obtained has been
multiple concentrated damages, a new two-node multi- solved and the formulation has been investigated through
cracked beam (MCB) FE has been developed using a a parametric analysis to draw deductions on the range of
recently formulated closed-form solution for the transversal validity of the parameters involved (Stage 2).
deflection of beams with singularities. The proposed
approach exploits the classical continuum theory, and is The dynamic part of the proposed multi-cracked
able to describe damaged structures without enlarging the formulation including microstructural effects will be further
size of the FE model (Stage 1). developed and investigated through parametric analysis, as
to gain information about the influence of the parameters
Since the DS model concentrates the increased curvature involved, especially the length scales microstructural
in the neighbourhood of the damage at a single abscissa, parameters. To validate the developed equations a set of
a jump discontinuity appears in the field of rotations representative FE models will be built using commercial
experienced by the beam. This is clearly unrealistic, FE software, and the results will be compared with those
and is due to the local elasticity model used for the obtained with the proposed model (Outstanding work).
constitutive law of the beam. Aimed at overcoming this
physical inconsistency, and to develop an improved
representation of cracked beams, a non-local elasticity
model has been adopted in the final stage of the PhD
project, and both stress gradient and strain gradient have
been considered with two independent length scales.
Indeed, while problems and applications in structural and
mechanical engineering are mainly addressed using the
tools of the classical continuum mechanics, where the
stress is related to the strain at a single point (without
higher-order derivatives of the field variables), these tools
do not allow handling properly many size-dependent
problems (including crack initiation and propagation),
where the structural response strongly depends on the
microstructural pattern. Furthermore, the singularities due
to the application of point loads, the dislocation lines and
Potential for application of results Funding body
The PhD project will end with the publication of a doctoral Loughborough University, School of Civil and Building
thesis on the formulation of MCB FE with and without Engineering (doctoral studentship programme)
microstructure. In particular, thanks to the enriched
equations of the gradient elasticity, it will be possible Further information
to have a more accurate description of the structural
Marco Donà (M.Dona@lboro.ac.uk) or
behaviour in the neighbourhood of the cracks.
Dr Alessandro Palmeri (E: A.Palmeri@lboro.ac.uk) and
Dr Mariateresa Lombardo (E: M.Lombardo@lboro.ac.uk)
Possible applications concern engineering problems
where the size dependence is not negligible as observed
in metals, polymers and polysilicon, or nanotubes and
microstructural devices such as sensors or small actuators
(which can be modelled in the form of cantilever beams
exhibiting an elastic response).
This is not the only range of applications. Indeed, it has

been recently shown that the nonlocal constitutive law
leads to the same equations ruling the deflection of
composite beams with interlayer slip (layered beams,
shear connectors or sandwich constructions). This further
extension of the applications to composite structure will
allow describing in a very efficient computational model the
static and dynamic behaviour of composite beams with a
given number of damaged sections.
Impact strength of steel pipes
Shamsoon Fareed
Project goals and objectives Finite element analysis
Pipes that are laid in bed of the sea, in addition to the harsh Dynamic implicit analyses have been carried out for a
marine environment, may encounter a situation when they pipe impacted with sharp and curved impactors using
are subjected to an impact. The situation can be worse for finite element software ABAQUS. 3D models have been
pipes that are used to transport gas compared to pipes developed. As the model is symmetric about the three
that are used for oil, as in the latter case a coating (e.g. axes, only a quarter of the pipe was used to carry out the
polyurethane) is usually provided for thermal installation analyses. Initially two analyses, one for each impactor,
and this can assist in absorbing impact energy. There are were run with elastic properties; this was followed by
a number of parameters that influence the behaviour of plastic analyses for each impactor. In order to validate the
pipes under impact, these include the material properties analyses, each of the four analyses was run with three
and the dimensions of pipes, the way in which the pipes different elements, the 8 node linear brick, the 20 node
are supported, the location of the impact and the initial quadratic brick and the 24 node cylindrical brick with time
velocity and mass of the impactor. The behaviour of a pipe increments of 0.00001 sec. Excellent agreement has been
is also influenced by the shape of the face of the impactor found between the contact forces obtained for the different
which strikes the pipe. The behaviour is also influenced by elements used in the analyses.
whether the pipes are empty or filled with liquid or gas. The
objective of this research is to investigate the behaviour Comparison of impact tests and finite element analyses
of pipes under high mass, low velocity impacts and to
Excellent agreement has been also found between the
determine the effects of the various parameters both
deformed shape of impactor obtained from full scale
experimentally and using finite element analysis.
impact tests and finite element analyses, Fig 2.
The approach used was to carry out the full scale impact
Further full scale impact tests will be carried out with
tests on pipes with curved and sharp impactors with
impact energy of up to 50KJ. Full scale impact tests will
an impact energy of 16kJ, impacting the pipe surface
also be carried out with impactor impacting the pipe weld
transversely. Finite element analyses, have also been
transversely and longitudinally. Finite element analyses will
carried out to analyse the behaviour of pipe under high
be carried out to study the behaviour of pipe with internal
mass, low velocity impact.
pressure under impact. The outcome of the research will
improve the understanding of behaviour of steel pipes
Impact tests
under impact.
Previous research into the behaviour of pipes subjected
to an impact has been with relatively small pipe cross- References
sections and span lengths and with relatively low levels
[1] N Jones, S E Birch, R S Birch, L Zhu and Brown,
of energy [1, 2]. In the current research, full scale impact
An experimental study on the lateral impact of fully
tests were carried out using the facility shown, Fig 1. The
clamped mild steel pipes, Institute of Mechanical
mild steel pipes have length of, 2000mm, outer diameter
Engineering, Vol.206, pp.111-127, 1992.
of, 457mm and thickness of, 25mm were impacted
transversely. The impactor has a mass of, 0.5 tonnes and a
[2] Kaisheng Chen and Wei Qin Shen, Further experimental
velocity of, 3870 mm/sec vertically. Both curved and sharp
study on the failure of fully clamped steel pipes,
impactors were used. The pipes were restrained over their
International Journal of Impact Engineering, Vol.21,
entire length. High speed video recording was used to
pp.177-202, 1998.
record the behaviour during the tests.
Funding body Further information
Total E&P UK Limited, NED University of Engineering & Shamsoon Fareed (E: sf5@hw.ac.uk) or
Technology Prof Ian May (E:I.M.May@hw.ac.uk)
Fig 1: Test setup for impact tests.
Fig 2: Deformed shape of sharp impactor obtained from experiment and finite element analysis.
Numerical analysis of the mechanical behaviour of
composite sandwich panels
Andrew Foster
University of Manchester 16
Poster presenter
Project description Potential for application of results
In this research, finite element analysis is carried out on The results from this finite element validation study are
simply supported sandwich beams subjected to three point significant as they show that it is possible to simulate a
bending. The sandwich beams consist of glass fibre-epoxy full range of elastic and plastic mechanical behaviour of
face sheets bonded to PVC foam cores. It is recognised sandwich beams including a number of failure modes
that sandwich panels and beams fail by a number of within Abaqus. The findings here give us confidence in the
different mechanisms depending on the geometry of the capabilities of the Abaqus crushable foam plasticity model,
system and the properties of the materials used. Steeves meaning that models can be created relatively quickly
and Fleck (2004) tested a number of sandwich beams without the need for further complex user defined material
with varying geometry to study each individual collapse models.
mechanism. The measured collapse responses have been
used to validate predictions from proposed analytical References
models (Steeves and Fleck, 2004) and our finite element
Steeves and Fleck, 2004. Collapse mechanisms of
simulations.
sandwich beams with composite faces and a foam
core, loaded in three-point bending. Part II: experimental
The goal of this research was to validate the capabilities of
investigation and numerical modelling. International Journal
the finite element software Abaqus to capture the full elastic
of Mechanical Sciences, 46(2004), pp.585-608.
and plastic mechanical behaviour of the sandwich beams
as well as each failure mode. A crushable foam plasticity
Funding body
material model within Abaqus has been utilised and plays
an important role in the way the rigid polymer foam core EPSRC and Tata Steel (Industrial CASE Award)
response is simulated. An important aim of the research
was to discover if the crushable foam plasticity model Further information
within Abaqus was accurate or whether further complex
Andrew J Foster (E: andrew.foster-2@manchester.ac.uk) or
user defined material models would be needed.
Prof Yong C Wang (E: yong.wang@manchester.ac.uk)
A number of sandwich beams have been tested with
varying face thickness, core thickness and of varying
lengths (Steeves and Fleck, 2004); however three beams
have been chosen specifically that are expected to fail in
different mechanisms. Since the main goal of this study
was the validation of our finite element models, beams
were chosen that would fail differently so that we could see
if our models could accurately predict each failure mode.
The critical failure mode of a beam under three-point

bending is taken to be the weakest mode. The core may
fail in shear, once the shear strength of the core has been
reached. An indentation failure of the beam is caused
by a local elastic instability of the face sheets which is
accompanied by a local compressive yielding of the core.
Lastly, face yielding will occur once the axial stress in the
compressive face reaches the microbuckling strength.
The three beam models A, B and C have been shown

to fail correctly by indentation, microbuckling of the face
sheet and core shear, respectively. Figures 1 and 2 show
the experimental and simulation load vs. displacement
responses for specimens A and B, along with the proposed
analytical models. Also shown in the figures are the finite
element models, where the colour contours represent the
von Mises stress distribution.
Fig 1: Load vs. displacement graph for specimen A (indentation failure)
Fig 2: Load vs. displacement graph for specimen B (microbuckling failure).
Lifetime extension of reinforced concrete slab-on-beam
structures
Rob Foster
University of Cambridge 17
Project objectives and goals The push-off testing approach provides a method for
investigating shear transfer across a dominant crack and
The project investigates the behaviour of reinforced
test results for realistic scale strengthened T-beams will
concrete slab-on-beam structures strengthened in shear by
be valuable in extending the currently limited data set.
externally U-wrapped and surface bonded fibre reinforced
Improved understanding of the possible scaling effects will
polymer (FRP) fabrics. Recent work [1] has indicated that
be critical in bridging the gap between the understanding
current guidelines for strengthening may overestimate the
gained from small scale experimental work in a research
contribution of externally bonded FRP fabrics leading to
environment and its application to full size structures in
the possibility of unconservative design in practice. The
practice.
latest edition of The Concrete Society Technical Report 55
(TR55) [2] has taken some steps to accommodate these
The detailed modelling of strengthened T-beams,
findings but further work is needed to ensure that current
benchmarked against the experimental data from the full
design methodologies are appropriately conservative. The
scale beam tests, will improve and validate the accuracy
objective of this project is to improve upon our current
of numerical modelling of full scale beam behaviour. This
understanding of strengthened concrete behaviour and
will provide confidence in assessing the capabilities and
provide a more rational design method validated using
limitations of computational analytical approaches to
sound empirical evidence.
support the design of strengthened elements in practice.
References
The behaviour of concrete slab-on-beam structures
[1] Dirar, S., Lees, J. & Morley, C. (2012) Precracked
strengthened in shear by externally U-wrapped and
Reinforced Concrete T-Beams Repaired in Shear with
bonded FRP fabrics will be investigated experimentally by
Bonded Carbon Fiber-Reinforced Polymer Sheets, ACI
carrying out a series of push-off tests. Push-off tests have
Structural Journal, Vol. 109 No. 2, pp. 215-224
been carried out routinely on unstrengthened concrete
since the 1960s, notably by Hofbeck [3], Mattock [4] and
[2] The Concrete Society (2012) Technical Report
Walraven [5]. In the current work, push-off specimens with
55: Design Guidance for Strengthening Concrete
varying reinforcement ratios, FRP thicknesses and FRP end
Structures Using Fibre Reinforced Composite Materials
conditions will be tested. Of particular interest will be the
relationship between strains in the internal reinforcement
[3] Hofbeck, J.A, Ibrahim, I.O. & Mattock, A.H. (1969)
and the external FRP with the slip and dilation of the shear
Shear Transfer In Reinforced Concrete, Journal of the
crack.
American Concrete Institute, Proceedings, Vol. 66 No.
2, pp. 119-128
A second experimental phase will involve four-point
bending tests on a series of two control and six FRP
[4] Mattock, A.H. & Hawkins, N.M. (1972) Shear Transfer
strengthened concrete T-beams. These test results will
in Reinforced Concrete - Recent Research, Journal of
be combined with a coordinated test series being carried
the Prestressed Concrete Institute, Vol. 17 No. 2, pp.
out at the University of Bath, to provide a data set of
55-75
strengthened T-beams of three different depths. This will
allow scale effects to be investigated and the implications
[5] Walraven, J.C. & Reinhardt, H.W. (1981) Theory and
for design assessed.
Experiments on the Mechanical Behaviour of Cracks
in Plain and Reinforced Concrete Subject to Shear
Detailed finite element modelling of the T-beams will also
Loading, Heron, Vol. 26 No. 1, pp. 1-68
be carried out and benchmarked against the experimental
results.
Funding body
Potential for application of results EPSRC
The conclusions from the experimental and analytical
Further information
programme are expected to inform technical guidance in
future editions of TR55 and design codes addressing the Rob Foster (E: rmf41@cam.ac.uk) or
use of FRPs to strengthen existing concrete structures. Dr Janet Lees (E: jml1010@hermes.cam.ac.uk)
Conclusions with regard to failure mechanisms associated
with the de-bonding of the external FRP reinforcement and
incipient near-surface failure of the concrete are thought to
be of particular relevance in explaining the possible lack of
conservatism in the current guidance.
Morphing origami panels: geometry and construction
18 Joseph Gattas
University of Oxford
Oral presenter
Introduction cheap materials such as cardboard, stored in a folded

position, and rapidly erected by when needed. However,
In recent years, origami folding patterns have been
constructability of the structure remains a critical concern,
increasingly used in numerous engineering and design
as the strength, stiffness and kinematic freedom of the
disciplines to create a wide range of novel folded plate
structure is significantly affected by panel thickness and
structures and devices. Particular attention has been drawn
hinge stiffness considerations. This project explores three
to rigid origami, a subset of origami that permits continuous
inter-related concepts needed for successful application
motion between folded states without the need for twisting
of morphing panels: optimum core and face geometry,
or stretching of the facets between the creases. This
hinge connections for individual origami panels and hinge
allows rigid origami patterns to be readily parameterised
connections between core and face patterns.
and built from rigid sheet materials suitable for engineering
applications, such as cardboard, timber, plastic, metal, or
Method & conclusions
carbon-fibre sheeting.
Core and face parameterisations required to form a
There are numerous existing engineering applications of morphing panel are presented. A number of hinge
origami. Miura (2006) developed several pioneering rigid connection methods are explored for three sheet materials:
origami structures. His designs include one of the largest steel, timber, paper honeycomb. Possible connections are
origami structures, a deployable space solar panel, and subjected to tension, shear, and peel tests to establish the
one of the most widely used origami devices, the Japanese most suitable for hinged origami pattern. Morphing panel
subway map. Both were developed from the rigid, planar, prototypes are constructed out of honeycomb cardboard
Miura pattern. The Miura pattern is also used as the core and steel, with a smooth kinematic mechanism achieved
material of a foldcore structural sandwich panel. A foldcore for both during deployment, Figure 2. Preliminary numerical
consists of a partially folded Miura pattern sandwiched models were analysed with finite element solver Abaqus .
between two stiff facings, and possesses numerous useful The results provide insight into the structural behaviour and
structural attributes, including isotropy (Miura 1972). stress distribution of the morphing panel. The next stage
Several authors have looked adapting origami patterns of work will involve load tests of steel panels to validate
for folded plate shelters (De Temmerman et al. 2007, Buri numerical models.
and Weinard 2008, Gioia et al. 2012, Tachi 2009). Such
structures are advantageous in that they can be designed
to be deployable, have a continuous watertight surface,
and are lightweight. However these structures all use single
origami layer in construction, and so suffer from extreme
stress-concentration at sharp joints and correspondingly
poor mechanical performance.
Project & objectives

A new type of morphing origami panel is proposed that
consists of a layered origami panel assembly. Deemed
a morphing sandwich panel, it is simply constructed by
introducing secondary origami patterns into the faces
of a typical foldcore sandwich panel. The secondary
patterns are designed to have a single degree-of-freedom
kinematic mechanism that is compatible with the Miura
core kinematic mechanism, until they reach a completely
unfolded state, at which point they constrain the core in a
partially-folded, rigid configuration, Figure 1.
Such an assembly retains the developability seen

in previous origami structures, but with a much-
improved load bearing mechanism, where faces carry
predominantly in-plane loads and the lightweight core
carries predominantly shear. The morphing panel is suitable
for a variety of architectural and structural engineering
applications. Foremost amongst these is as an emergency,
temporary structures, as they can be constructed from
Miura, K. (2006). The science of miura-ori: A review. In The General Sir John Monash Award.
Origami4 : Fourth International Meeting of Origami Science,
Mathematics, and Education (ed. R. J. Lang), pp. 87–99. A Further information
K Peters, Ltd.
Joseph Gattas (E: joe.gattas@eng.ox.ac.uk) or Dr Zhong
You (E: zhong.you@eng.ox.ac.uk)
Miura, Koryo (1972). Zeta-Core Sandwich-Its Concept and
Realization. Technical Report 480.
De Temmerman, N, Mollaert, M, Van Mele, T, and De Laet,

L (2007). Design and analysis of a foldable mobile shelter
system. International Journal of Space Structures, 22(3),
161–168.
Buri, H. and Weinand, Y. (2008) Origami - Folded Plate

Structures, Architecture [online] WCTE Miyaaki, Japan,
2008 conference proceedings.
Gioia, F, Dureisseix, D, Motro, R, and Maurin, B (2012).

Design and analysis of a foldable/unfoldable corrugated
architectural curved envelop. Journal of Mechanical
Design, Transactions of the ASME, 134(3).
Tachi, T. (2009). Generalization of rigid-foldable

quadrilateral-mesh origami. Journal of the International
Association for Shell and Spatial Structures, 50(162),
173–179.
Fig 1: Typical foldcore, deployed morphing foldcore, and flat-folded morphing foldcore
Fig 2: Morphing panel prototype constructed from 6mm thick paper honeycomb panels.
Structural strengthening of existing reinforced concrete
slab-on-beam structures in shear using CFRP sheets
and bars
19 Monika Grusova
University of Bath
Project objectives and goals The experimental results demonstrated that externally
bonded CFRP reinforcement significantly improves the
Currently, there is a global need for accurate and well-
overall shear resistance over unstrengthened control
tested design guidelines for large reinforced concrete
specimens and that this increase in shear capacity is highly
beams strengthened in shear with externally bonded fibre
dependent on the anchorage length provided for the CFRP
reinforced polymer (FRP) reinforcing fabrics. In order to
reinforcing fabric. Provision of insufficient anchorage length
meet this need a greater understanding of local and global
in U-wrapped specimens resulted in a more catastrophic
mechanical behaviour of such strengthened systems
failure mode with concrete cover separation. In both
is essential as well as the ability to identify the most
the fully wrapped specimens and those with sufficient
appropriate strategy for structural analysis of this complex
anchorage length, crack induced interfacial debonding
problem. Despite the increasing amount of interest in
of the CFRP sheets was observed. Furthermore, the
this area of structural strengthening, there is a lack of
results suggested, that the ultimate strength of the push-
research considering the issues of FRP debonding under
off specimens is independent of the ratio of the CFRP
oblique stretching, which is invariably the case in shear-
reinforcement as well as the steel ratio across the shear
strengthening practice.
plane in fully wrapped solutions. Similar ultimate strength
recorded for fully wrapped specimens and specimens with
This context provides two key research objectives:
sufficient anchorage length when bonded on three sides
demonstrated that a minimum increase in shear strength
• To determine effective FRP contribution and
is possible beyond certain anchorage length. The ultimate
debonding effects addressing the influence of the
shear capacity for the specimens with deep embedded
specimen geometry and size; and
CFRP bars showed comparable increase in shear strength
to the specimens with the same ratio of externally bonded
• To deliver robust, mechanics-based design code
CFRP fabrics. Furthermore, the results elucidated, that the
formulations for FRP shear strengthened reinforced
commonly used assumption in current design guidelines
concrete slab-on-beam structures.
that the ultimate shear capacity of a strengthened RC
element can be expressed as the sum of concrete (Vs),
Description of methods and results
internal steel (Vs) and FRP (Vf) resistances is simplistic in its
FRP materials are widely used for strengthening of application and therefore not fully appropriate.
reinforced concrete structures to ensure sustainability and
to meet the increasing load demands on existing concrete Potential for application of results
infrastructure. Bonded fabrics represent the most versatile
Intermediate observations and results from the preliminary
and desirable shear-strengthening solutions for deficient
experimental investigation contribute directly to the
concrete beam bridges where the web of the beam is fully
future experimental and analytical work on strengthening
accessible. However, where there are accessibility issues,
of realistically sized reinforced concrete T-beams (see
another strengthening solution, such as deep embedded
Fig. 2). Subsequent analytical work on development
FRP bars, might be more appropriate. A targeted
of reliable models for calculation of the ultimate shear
experimental and analytical research programme has
strength, Vu, with realistic values of the strain limits from
been developed in order to observe and model structural
strain measurements on internal steel and external CFRP
behaviour of reinforced concrete beams strengthened in
reinforcement is currently under way. Prediction models
shear using various strengthening schemes.
based on understanding of the individual contributions of
the concrete (Vc), steel (Vs) and FRP (Vf) are necessary, and
A preliminary experimental study conducted on a series
now indeed possible.
of reinforced concrete push-off specimens, illustrated in
Fig. 1 (a), with varying ratios of internal steel reinforcement
(stirrups) and carbon FRP (CFRP) reinforcement across a
known shear plane, investigated effectiveness of various
strengthening solutions, see Fig. 1 (b). This study also
considered the effects of the ratio of the bonded area to
the force in the external CFRP reinforcement by varying
the anchorage length from insufficient to fully wrapped
schemes. Deep embedment technique was included for
comparison of the two different strengthening techniques
through provision of the same ratio of the FRP across the
shear plane.
Industrial partners and sponsors Funding body
Parsons Brinckerhoff, Tony Gee consultants, Ove Arup & EPSRC
Partners Ltd, LG Mouchel & Partners, Highways Agency,
Concrete Repairs Ltd, The Concrete Society, Fyfe Europe Further information
S.A., Fibrwrap UK and Hughes Brothers
Monika Grusova (E: M.Grusova@bath.ac.uk) or
Prof Tim Ibell (E: T.J.Ibell@bath.ac.uk)
Fig 1: Push-off specimens
Fig 2: T-beams strengthened in shear with CFRP fabrics
Dynamic response of a novel form of water tower shafts
composed of interlocked panels
20 Filip Gurkalo
Anglia Ruskin University
Many countries around the world experience destructive Research impact and vision
effects of seismic activities therefore, a water supply is
In the light of the unpublished research findings it appears
essential for controlling fires that may occur during the
that the unique AcerMetric© interlock mechanism provides
earthquakes. Without a secure water supply, uncontrolled
an opportunity for a unique and original research study on
fires can cause substantially more damage than the
the structural response of buildings (made of panels) under
earthquake itself. Moreover, it is essential that the
dynamic loading.
availability of a safe supply of water is available after the
effect of an earthquake.
The main benefits of using interlocked panels are; firstly,
while the maximum displacement of the vessel is similar in
Project objectives and goals
both models, the deformation in the staging constructed
The overall aim of the proposed research is to develop with interlocked panels results in the flexibility of joints, thus
a numerical model to simulate the structural behaviour reinforced concrete plates become nearly unreformed.
of a building composed of interlocking panels under
earthquake loading. The model will then be used to Secondly, the construction mechanism allows the
compare the performance of buildings formed by modular replacement of the panels due to cracking or damages
and conventional construction when subjected to seismic without significant efforts. Thirdly, interlocked mechanisms
actions. with different stiffness’s can be used to satisfy conditions
for particular regions. Moreover, gaskets between plates
Traditionally, engineers have designed structures to resist can be provided to reduce friction and increase damping of
earthquake forces, so that the strength of structural the structure.
members is equal to or greater than the seismic demand
placed on them. During the 1980s to 1990s, new Finally, this kind of structure is easily erected on site and
approaches to seismic design emerged, involving the the interlocked mechanism can be applied to various
modification of structural responses to reduce earthquake structures reducing construction time and improving
induced loads to more tolerable levels (Scawthorn, flexibility. With the increased magnitude of earthquakes, all
2007). These methods included; moment frame, braced the models sustained increased deformation, this occurred
frame, shear walls (AC Consulting Group Ltd, 2006), extensively in the models with the heaviest masses within
base isolation (Current Science, 2004), supplemental the water tanks. Moreover, under the same conditions
damping (Engineering Structures, 2005) and active control the greatest increase in deformation was sustained by
(Engineering Structures, 2004). the monolith model, with increasing masses within the
water tanks. The dynamic properties of the water towers
However all these techniques are expensive and the assembled from interlocked panels were improved to a
average family cannot afford to build an earthquake greater degree compared to the monolith model.
resistant house.
Across all the tested seismic conditions, the increasing
Innovation mass of the water tanks results in greater deformation
of the water towers. This is most evident in the monolith
The AcerMetric© interlock system is a new method
water tower model. With the increasing mass of the water
for constructing dwellings that is self-supporting and
tanks across all earthquake conditions, the performance of
comprised of; pre-tensioned inter-locking lightweight
water towers composed of interlocked panels is superior to
panels, used to construct the walls, floors and ceilings.
that of the monolith water tower in regards to withstanding
The panels are readily transported as a flat pack or in
seismic loads.
pre-formed modules that can be quickly assembled on
site. Based on preliminary unpublished research findings
Further research is required to investigate the behaviour of
this unique interlock mechanism has a beneficial effect on
water towers under other loading conditions, varying the
the structural behaviour of the structure formed from these
height and geometric properties.
panels (www.acermetric.co.uk).
Potential for application of results References
The results obtained from the analysis of the three models AcerMetric Ltd versatile Elemental Building system [Online]
are compared. The study demonstrates that the ground Available at: <www.acermetric.co.uk> [28th March 2011]
and spectrum type significantly affects the behaviour of the
water towers depending on the masses contained within Gregory A. J. Szakats (2006) Improving the Earthquake
the water tanks and the height of the shafts, considerably Resistance of Small Buildings, Houses and Community
altering the lateral displacement of the water towers. Infrastructure. AC Consulting Group Ltd, October 2006
However, the performance of water towers assembled of
interlocked panels is more efficient in comparison to the Charles Scawthorn. (2007) Earthquake Engineering.
monolith water tower as the height of shafts and mass AccessScience, DOI 10.1036/1097-8542.800400
inside the water tank increases.
Javad Marzbanrad, Goodarz Ahmadi, Ratneshwar Jha.
(2004) Optimal Preview Active Control of Structures During
Earthquakes. Engineering Structures, Volume 26, Issue 10,
August 2004, Pages 1463-1471
Dimitri V. Val, Fred Segal. (2005) Effect of Damping Model

on Pre-Yielding Earthquake Response of Structures.
Engineering Structures, Volume 27, Issue 14, December
2005, Pages 1968-1980
Sajal Kanti Deb. (2004) Seismic Base Isolation – an

Overview. Current science, vol. 87, no. 10, 25 November
2004
Funding body
Anglia Ruskin University
Further information
Filip Gurkalo (E: filip.gurkalo@anglia.ac.uk) or
Dr Konstantinos Poutos
w(E:Konstantinos.poutos@anglia.ac.uk)
Fig 1: Proposed mathematical model of the Fig 2: Two degree of freedom (2DOF) spring-system of
water tower assembled with interlocked panels. fluid-structure interaction inside the water tank.
A numerical tool for simulating dynamic soil-structure
interaction problems
21 Nawras Hamdan
Heriot Watt University
Objectives Application: rigid foundation over inclined bedrock

The aim of this project is to develop a numerical tool, in the A rigid surface strip foundation with a width of 2B is
frequency domain and under plane strain conditions, for overlying a homogenous soil layer and subjected to a
modelling dynamic soil-structure interaction problems with unit vertical harmonic load applied at its centre. The
emphasise on industrial-scale models. The numerical code assumption of horizontal bedrock is very useful; however,
considers both homogenous and multi-layered soil profile. this is not always the case. Therefore, it is very important
The soil may overlay rigid bedrock or half-space, in which to investigate the effect of inclined rigid bedrock on the
case the domain extends to infinity with respect to depth. dynamic behaviour of the foundation. This is achieved by
The code may use unstructured triangular mesh grids exploiting the benefits of an automatic mesh generator [2]
which allow us to deal with complex geometry such as to create the mesh grid using 3-node triangular elements.
including various types of surface or embedded structures.
For comparison purpose, the rigid bedrock is first
Method of analysis assumed to be horizontal, θ=0º, and the base is thereafter
inclined by 5º, 10º and 15º with respect the horizontal.
Modelling dynamic soil-structure interaction problems,
The results are presented in terms of the normalised
which may arise from various types of loadings such
vertical displacement underneath the foundation and the
as machine foundations, earthquakes, moving loads
dimensionless frequency. The natural frequency of the
on the surface and in depth and many others, requires
soil layer, over horizontal bedrock, is obtained from the
handling finite-size models. This is achieved by truncating
following expression:
the domain of interest and applying special transmitting cp
boundary conditions at the artificial boundaries to prevent f nv = (2n − 1) , n = 1,2,3,.......,
fictitious waves from reflecting back to the domain. In 4H
such a way a finite element model; which consists of a
where cp is the compression wave velocity and H is
central region and two lateral regions is obtained as in
the height of the soil layer. For the current example, the
Fig.1. The thin layer method, a semi analytical numerical
fundamental frequency is around 0.047Hz. The numerical
technique in the frequency domain, developed by Waas
code predicted its value around 0.045Hz where resonance
[1] is adopted in this project to simulate the semi-infinite
occurs if the excitation frequency approaches this value as
extent of the medium in the horizontal direction. It is based
depicted in Fig. 2.
on sub-dividing the layers into thin sub-layers with the
assumption of rigid bedrock at the base. The displacement
In the numerical simulations, the angle of inclination of the
field is assumed to vary linearly with respect to depth
soil layer is varied such that it is reducing the volume of the
and exponentially in the horizontal direction. Solving the
soil in the model. Thus, the overall height of the soil layer is
appropriate governing equation of motion leads to a
decreased. According to the given expression above, the
second order eigenvalue problem; see ref. [1], for which
fundamental frequency will increase. This is demonstrated
eigenvalues and eigenmodes are used to formulate the
in our simulations and displayed in Fig. 2. The numerical
nodal force vectors PR and PL to be applied on the lateral
code predicted fundamental frequency to be 0.048, 0.05
boundaries of the central domain. This allows modelling the
and 0.052Hz for the angles 5º, 10º and 15º, respectively.
unboundedness of the medium while treating the interior
domain by the finite element method. The system equation
Conclusion
of motion is expressed as:
A numerical code is developed in this project to model
((1+i β)K – ω2M)U = P + PR + PL dynamic soil-structure interaction problems. It is validated
by considering the dynamic behaviour of a rigid foundation
β represents the hysteretic damping, ω stands for the over a soil layer overlaying horizontal bedrock. As a
circular frequency, K and M represent the global stiffness practical application, the bedrock is then inclined and
and mass stiffness matrices of the domain and P is the the dynamic response of the foundation is computed to
external force vector. Complex vertical and horizontal demonstrate the ability of the code to deal such cases.
displacements are obtained at each node by solving the The numerical results have shown that as the angle of
obtained system. inclination increases, the overall soil volume decreases,
and hence the fundamental frequency of the soil layer
increases. This is of significant importance as resonance is
associated with the fundamental frequency.
[1] G. Waas. Linear two dimensional analysis of soil Nawras Hamdan (E: nh60@hw.ac.uk) or
dynamics problems in semi infinite layered media. PhD Prof Omar Laghrouche (E: o.laghrouche@hw.ac.uk)
thesis, University of California, Berkeley, 1972.
[2] C. Geuzaine and J. F. Remacle. Gmsh. A 3-D finite

element mesh generator with built-in pre- and post-
processing facilities. International Journal for Numerical
Methods in Engineering, 79:1309-1331, 2009.
λ1 µ ρ1 λ1 µ1 ρ1
1
λ 2 µ 2 ρ2 λ2 µ2 ρ2
∝ ∝
H
λi µ i ρi hi λi µi ρi
λn µ n ρ n λn µn ρ n
rigid bedrock
Fig 1: Schematic diagram of 2D soil medium.
1.2 0
θ=0
Normalised Vertical Displacement
0
θ=5
0
1 θ=10
θ=15 0
0.8
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1
Dimensionless Frequency
Fig 2: Variation of amplitude of vertical displacement underneath foundation
Building Information Modelling for improving
collaborative design
22 Aras Jaly-Zada
University of Nottingham
Design is a teamwork process and the project depends References

upon a strong weave of multi-disciplinary involvement into a
Philp, D. 2012. BIM and the UK Construction Strategy
collaborative team. Collaborative design is not only sharing
[Online].
idea based on the designer specialty but also achieve a
seamless information flow among the collaborative team
Shen, W., Hao, Q., Mak, H., Neelamkavil, J., XIE, H.,
participants. With the rapid development of information
Dickinson, J., Thomas, R., Pardasani, A. & XUE, H. 2010.
and communication technologies (ICT), different systems
Systems integration and collaboration in architecture,
of integration and collaboration technologies have been
engineering, construction, and facilities management: A
innovated and deployed during the process of design
review. Advanced Engineering Informatics, 24, 196-207.
to support and manage the collaborative environment,
to integrate people, information and processes more
Tizani, W. & Mawdesley, M. J. 2011. Advances and
efficiently and to use and distribute information through the
challenges in computing in civil and building engineering.
entire project (Shen et al., 2010) .
Advanced Engineering Informatics.
In recent years, Building information modelling (BIM)
Xiaolong, X., Qiping, S., Hongqin, F., Heng, L. & Shichao, F.
based design process has become an active research
2012. IT supported collaborative work in A/E/C projects: A
area in order to address problems related to information
ten-year review. Autom Constr, 21, 1-9.
integration, communication, and interoperability (Tizani and
Mawdesley, 2011). It is a new way of creating, sharing,
Funding body
exchanging, and managing information at all stages of the
design process using collaborative environments (Xiaolong Ministry of Higher Education and Scientific Research /
et al., 2012). With all these capabilities with BIM, the UK Kurdistan Regional Government, Iraq
government has been adopting a five-year strategy until
2016 to restructure the AEC sector to use collaborative Further information
BIM (Philp, 2012).
Aras Jaly-Zada (E: evxaj2@nottingham.ac.uk) or
Dr Walid Tizani (E: evzwt@exmail.nottingham.ac.uk)
This research proposes utilizing building information
modelling to build an integrated engineering information
environment to support the collaborative design process.
This happens when information technology, different
distributed systems, and multi-disciplinary teams need
to collaborate on a shared design activity. Therefore,
integration, interoperation, and communication of
engineering information are the most critical technical
factors in ensuring successful collaboration.
The aim of this research is to investigate whether building

information models can be used to achieve a better
collaborative design solution. To achieve this, the objectives
have been started with an investigation and assessment
of the state of the art and of the knowledge use of building
information models. Specific emphasis is given to the
collaborative design process to identify and develop a
strategy to address the problems of the collaborative
design process and the information technology.
Identify suitable technologies for creating a virtual

collaboration design system using computer based
collaborative work and computer based design to build
a modelling framework by object oriented programming
language. The modelling framework will be developed and
structured through using building information modelling
and a 3D model is proposed for visualisation and for match
and check of clash detection between the architectural and
structural designers, which allows multi-disciplinary teams
to share, exchange information and collaborate.
Structural behaviour of external diaphragm connection
between steel I-section beams and circular hollow section
columns under cyclic loading
Majd Khador
Poster presenter
Project objectives and goals Potential for application of results
The objectives and goals for the research project are: The main target of this research was to develop a joint
configuration with multi-fuse energy dissipative mechanism
(1) To determine the hysteretic (energy dissipative) where the maximum portion of energy is dissipated by the
behaviour of the external diaphragm connection type replaceable components of the connection (TCPs in this
by full scale laboratory tests research). The results will inform engineers the efficient
and effective design of joint between I-section beams and
(2) To calibrate the non-linear finite element models tubular columns in seismic-resistant moment frame.
(ABAQUS) against experimental results for parametric
analyses on key geometric parameters with an aim of References
achieving optimum design arrangement
CEN (2002). Eurocode – Basis of structural design, EN
1990, Brussels, Belgium.
(3) To derive design expressions in line with CIDECT
(Kurobane et al. 2004) and Eurocode 3 Part 1-8 (CEN
CEN (2004). Eurocode 8: Design of structures for
2005) which will be calibrated against experimental
earthquake resistance. Part 1, General rules, seismic
data in accordance with Eurocode 0 (CEN 2002)
actions and rules for buildings. EN 1998-1, Brussels,
Belgium
The proposed joint includes two diaphragm plates (DPs) CEN (2005). Eurocode 3: Design of steel structures – Part
welded to the outer circumference of the circular hollow 1-8: Design of joints, EN 1993-1-8, Brussels, Belgium.
section (CHS) column and bolted to the I-beam flanges
with two tapered cover plates (TCPs). A web stub is Wang, W., Chen, Y., Li, W., and Leon, R.T. (2011).
welded to the CHS column and bolted to the I-beam Bidirectional seismic performance of steel beam to
web. Figure 1 illustrates the different components of the circular tubular column connections with outer diaphragm.
proposed joint arrangement before assembly. Earthquake Engineering & Structural Dynamics, 40,
1063–1081
Design calculations were made for detailing the joint in
accordance with Eurocodes (CEN 2004 and 2005). A Kurobane, Y., Packer, J.A., Wardenier, J., Yeomans, N.
parametric study using the finite element analysis software (2004). Design guide for structural hollow section column
(ABAQUS) was conducted to investigate the energy connections, CIDECT 9.
dissipative performance of the proposed joint under cyclic
loading. The developed joint creates a multi-fuse energy Schneider, S.P. and Alostaz, Y.M. (1998). Experimental
dissipation mechanism for seismic actions in which the behaviour of connections to concrete-filled steel tubes,
column web panel distortion is eliminated (Wang et al. Journal of Constructional Steel Research, 45(3): 321–352
2011). Most of the seismic energy is to be dissipated by
the TCPs which can be considered as replaceable links in Funding body
the connection after a seismic action. The developed joint
Syrian Ministry of Higher Education, EPSRC, Tata Steel
also eliminates stress concentration in the DPs (Schneider
and Alostaz 1998).
Further information
Six laboratory tests were conducted to validate the Majd Khador (E: M.Khador@warwick.ac.uk) or
numerical simulation results. Figure 2 illustrates the Dr Tak-Ming Chan (E: T.M.Chan@warwick.ac.uk)
general test arrangement. All six tests had the same joint
arrangement and components except the TCPs that
connect the DPs to the I-beam flanges. Six different types
of TCPs in the joint tests were adopted to investigate the
effect of 1) using TCPs with/without stiffeners, 2) different
steel grades for TCPs and 3) oversize/normal-size holes in
TCPs on the overall energy dissipative performance of the
joint.
Fig 1: General view of the components of I-beam to CHS joint
Fig 2: General layout of the test arrangement
Thermo-mechanical analysis of shells using a hybrid-Trefftz
formulation
Payam Khazaeinejad
The University of Edinburgh 24
Brief description Description of method and results
Over the past three decades, hybrid-Trefftz (HT) finite Since special attention is given to investigate the nonlinear
element strategy has been recognised as a highly behaviour of shell structures, geometrically nonlinear
efficient computational technique for solving complex theory was employed to include the effect of membrane
solid mechanics problems [1]. Its advantages over other forces and coupling between the axial reactions and the
numerical methods include higher-order continuity between transverse deformation of shells. Furthermore, in response
elements and the possibility of integration on the element to the need for new structural materials, materially nonlinear
boundary by defining independent internal and boundary theory was developed by introducing nonlinear constitutive
displacement fields in the domain and on the element relations for constitutive relations. Dependency of material
boundary. Thus, numerous degrees of freedom on the properties on temperature filed was also considered in
element boundary could be defined without involving obtaining novel solutions for shell problems. Excellent
complexity in the problem. It also benefits from using initial analytical results have already been achieved and
appropriate analytical solution functions to construct the used for assessment of OpenSees new developed codes
interior shape function. The application of this concept to (as illustrated in Fig. 1) implemented in the last version
deal with uncertainty in modern built environments seems of software framework [4]. The new developed solutions
straightforward for practical and effective frameworks such will also be tested extensively and their accuracy will be
as PEER-PBEE (developed at the “Pacific Earthquake measured against reliable finite element solutions. These
Engineering Research” centre for “Performance-Based solutions would be useful as a trial for shape functions in
Earthquake Engineering” in California) [2] and OpenSees HT formulation. HT solutions will also be developed for
(object-oriented and open source software for earthquake considered problems and will be verified against existing
engineering simulation) [3]. reliable sources to implement in OpenSees and PBSE
frameworks.
In a general sense, finite element methods have been
widely adopted by the scientific community in order to The research will provide a fit for purpose design
deal with nonlinearity in thermo-mechanical modelling of methodology that will enable designers to establish a more
shell structures. However, even though from a point of unified approach for dealing with hazards to infrastructure
view this method could give the most available accurate (including fire, blast, wind, wave etc.). The work will
solutions, from the perspective of computation costs, it especially impact developers of OpenSees and PBSE
is not reasonable. The principle objective of the research frameworks by supplying a hereto unavailable data set with
project is to develop a range of consistent methodologies which to validate models. In particular, if the deterministic
to deal with common shell structures (i.e. beams, plates models could be made, computationally efficient and
and shells) subjected to thermo-mechanical loads. This will variance reduction techniques can be used to significantly
be accomplished via the following three actions: reduce the number of model runs for the PEER-PBSE
approach (expensive computational approaches, e.g. CFD
(1) Development of high quality analytical solutions for and FEM, need to be run thousands of times to be able
shell structures under thermo-mechanical loads by to cover the full range of probabilities) it could become
simulating extreme loads such as fire. a feasible future tool for both researchers and practising
engineers.
(1) Combine the advantages of finite element and
analytical methods in the HT finite element strategy
to establish a highly efficient formulation for shell
structures.
(1) Implementation of the developed methodology in

OpenSees and PBSE frameworks, respectively.
It should be noted that effects of several parameters,

like, influence of thermal shock, boundary conditions and
constitutive material models on coupled thermoelastic
responses of shells needs to be included.
[1] Qin, Q.-H., (2005). “Trefftz finite element method and its Edinburgh Research Partnerships (ERP)
applications.” ASME Applied Mechanics Reviews, 58,
316-337. Further information
Payam Khazaeinejad (E: p.khazaeinejad@ed.ac.uk) or
[2] Porter, K.A., (2003). An overview of PEER’s
Prof Asif Usmani (E: asif.usmani@ed.ac.uk)
performance-based earthquake engineering
methodology, Proceedings of the Ninth International
Conference on Applications of Statistics and Probability
in Civil Engineering (ICASP9), July 6-9, San Francisco,
CA. Civil Engineering Risk and Reliability Association
(CERRA).
[3] McKenna, F.T., (1997). “Object-oriented finite element

programming: frameworks for analysis, algorithms and
parallel computing.” PhD thesis, University of California,
Berkeley.
[4] http://opensees.berkeley.edu.
Fig 1: Comparative study for plates under uniform distributed

load and temperature change
Design of a FRP-reinforced concrete beam system for fire
performance
Mohamed Kiari
BRE Centre for Fire Safety Engineering, University of Edinburgh 25
Problem Statement Project objectives
Concrete members that are reinforced with fibre-reinforced The following four objectives are required to meet the aim
polymer (FRP) have many advantages over traditional steel of this research:
reinforcement, such as the lack of corrosion, lightweight,
and electromagnetic transparency. A substantial obstacle • Produce closed continuous fibre loop FRP
to their use, however, is the poor performance of an FRP- reinforcement, by wrapping carbon fibre tows that are
reinforced concrete member in fire. The bond between saturated with resin around a mould. Several windings
the FRP reinforcement and the concrete degrades due to will be required to create the reinforcement loops
resin softening at temperatures around its glass transition (typically Ø6mm).
temperature. As a consequence of this bond degradation,
all means of force transfer between FRP bar and concrete • Conduct small-scale tests upon single loops of FRP
are lost, the reinforced concrete member is unable to carry reinforcement to understand and characterise its
the required load, and a brittle failure results from the loss behaviour at both ambient and high temperatures.
of the tension reinforcement. Push off tests will be performed in which the ends of a
reinforcement loop are embedded inside two concrete
Previous research blocks, a hydraulic jack is used to push the concrete
blocks apart so as to pull the reinforcement from the
Force transfer between concrete and FRP is a vital aspect
concrete. These tests will be used to characterise
of concrete-FRP composite performance, and has been
the reinforcement performance at a variety of
the focus of much research. However, very little of this
temperatures. By contrasting the performance of
work covers the bond behaviour of FRP bars at high
FRP loop reinforced specimens with those containing
temperature. The limited previous studies show that at
straight reinforcing bars, the relative importance of
elevated temperatures (180-200°C) a severe reduction
bond and fibre continuity for load carrying will be
in bond strength (up to 90%) occurs (Katz et al. 1999).
demonstrated.
Research has shown (Bisby et al. 2005) that structural
collapse of concrete members reinforced with FRP in
• Develop analytical models that describe load-transfer
fire typically occurs due to loss of bond at temperatures
within a concrete beam reinforced with FRP loops,
exceeding glass transition temperature of resin. Although
and how this is affected by temperature. These
several design codes [ACI 440.1R-04, CAN/CSA s806,
models will utilise the results of the small-scale tests
CNR-DT203/2006] are available for the design of concrete
to predict failure of larger-scale beams. They are
structures incorporating FRP, few of them provide
expected to be based upon established reinforced
provisions for fire design of FRP reinforcement in fire, due
concrete design methods combined with new
to the limited research and expertise in this field.
methods for modelling the degradation of FRP to
concrete at high temperatures, although the plan-
Project goal
section assumptions of traditional reinforced concrete
This research project aims to enhance the fire performance design may break down requiring a new approach to
of FRP-reinforced concrete by developing a new the analysis.
arrangement of internal reinforcement that does not rely
only upon bond for load transfer. Instead of using straight • Lab-scale concrete beams will be fabricated using
separate FRP bars as reinforcement (an arrangement that FRP loop reinforcement. These beams will be tested
has been copied from steel-reinforced concrete design), under load and temperature to demonstrate the
closed FRP loops will be utilised as the longitudinal viability of the new reinforcement method, and to
reinforcement. These loops are filament wound from long validate the analytical modelling work.
continuous fibres. The proposed design takes advantage
of the fact that the FRP fibres are capable of sustaining a
large proportion of their original strength at relatively high
temperature (up to 1000°C for CFRP (Blontrock et al.
1999)). When the interlock and friction bond mechanisms
of force transfer are lost due to softening of the resin,
tensile forces in the reinforcement (FRP fibres) can still be
resisted through the loops.
Application of results References
This research will demonstrate the feasibility of FRP loop Bisby, L. A., Green, M., Kodur, V. (2005). “Response to fire
reinforced concrete for carrying load at fire temperatures. of concrete structures that incorporate FRP.” Progress in
It will also provide fundamental data (such as the bond Structural Engineering and Materials 7(3):136-149.
performance of hooked FRP bars at high temperatures,
and the pull-out of continuous loop reinforcement) that Blontrock, H., Taerwe, L., Matthys S. (1999). “Properties
is required for the analysis, design, and application of Fiber Reinforced Plastics at Elevated Temperatures
of this new form of reinforced concrete. This new with Regard to Fire Resistance of Reinforced Concrete
technology will allow FRP reinforced concrete to be used Members.” ACI Special Publication 188:43-54
in situations where its fire performance is important.
It could consequently remove the last obstacle to its Katz, A., Berman, N., Bank, L. (1999). “Effect of high
widespread use in buildings, bringing substantial benefits temperature on bond strength of FRP rebars. “ Journal of
for sustainable construction. Composites for Construction 3(2):73-81.
Further information
Mohamed Kiari (E: m.kiari@ed.ac.uk) or Dr.Tim Stratford (E:
tim.stratford@ed.ac.uk)
Development and testing of a UK home-grown timber
floor truss
Mark Kirk
Poster presenter
Overview Analysis
Development and testing of a UK home-grown timber floor In conjunction with the full-scale testing, Truss 1 and 2
truss is a research project to identify the suitability of UK were evaluated through linear elastic analysis using the
home grown C16 timber to be used in the manufacture of structural analysis package Staad.pro and first principle
timber floor trusses for use in timber frame construction. calculations created within TEDDS design software. Both
The research analysed two trusses using hand calculations, methods were subjected to a series of uniform loads, with
structural analysis software and full scale testing. the mid-span deflection compared in each case. There was
a correlation between the results from Staad.pro and the
The results showed that the use of home grown C16 validation methods, as each method fell within the 10%
timber in floor systems is a viable option however care variance limit set. A degree of variance in results between
must be taken to ensure that large defects are not located the methods is unavoidable due to differing assumptions
in crucial sections of the system. An effective correlation during the calculations. (Fig. 2)
between the analytical model and hand calculations was
achieved. Further research into the consistency of the The plot shown in Figure 2 highlights the close relationship
timber quality is required, along with further investigation between the fully pinned system calculated by hand and
into the predicted performance of the trusses to allow the Staad.pro fully pinned system, which have a variance
design optimisation to take place. of only 2%. The hand calculations for the semi-rigid system
have a variance of 6% over the range from the Staad.pro
Method analysis with the values diverging slightly as the loading
increases.
In order to determine the suitability of UK home grown C16
timber, two truss configurations were evaluated. Truss 1
Conclusions and further work
was initially designed by Kraft Architecture using 95 x 45
mm C16 timbers connected with Heco-fix fasteners using The testing to date proves that home grown timber can
18 mm thick plywood gusset plates with a span of 4500 be used successfully in structural floor systems. The work
mm and a depth of 254 mm. These were constructed by provides insight on the potential of home grown products
Oregon Timber Frame as a comparison to commercially with specific emphasis on the importance of internal
available products. As an experiment control, Truss 2, configuration and connection detailing in truss design. It
an ITW FloorTrus specimen provided by Ochil Timber is clear that the quality of home grown timber is higher
Products was evaluated. This truss design is a widely used than industry perception, although it is not an equivalent to
product utilised in the construction of affordable housing imported timber. In order to determine the quality of timber
product in the UK and measures 5100 mm in length and and utilise the available UK resource in the most suitable
250 mm in depth using 75 x 45 mm C16 timbers. (Fig 1) manner the research has continued by looking at the
construction of a timber truss to be used as a ridge beam,
The first stage of research was to test the components constructed using acoustically graded Scottish timber.
of Truss 1 in isolation to identify the individual mechanical This allows the critical members to be constructed of the
properties of the fasteners and C16 Home-grown interface. highest quality members with the less critical members
The second stage of this research was to further develop being of a lower grade.
both Truss 1 and 2 designs by full-scale testing, detailed
calculation structural analysis. Currently the design methodology adopted for this
research is being adapted to carry out design to Eurocode
Testing 5 for the graded truss. In order to validate the design and
behaviour of the system the truss will undergo testing upon
The trusses were tested in accordance with BS EN
completion early in 2013, with the results to be collated
408:2010 to identify their flexural rigidity (EI). Truss 2 results
and presented in March 2013.
from the test samples were satisfactory with a mean EI
value of 5.65E+11 Nmm2. Demonstrable failure in the
connections along with a consistency of strength evidences
C16 as an appropriate and successful floor system material
in this configuration.
The Kraft truss tests did not provide such satisfactory

results. During each test a lack of overall stiffness was
clearly apparent, with a mean EI value of 8.12E+10 Nmm2
obtained. This was largely down to insufficient connection
stiffness within the system, a factor which has a number
of possible sources, most likely from moisture penetration
during transport and a large number of defects within the
timber.
References Funding bodies and industrial partners
BSI. EN 408:2010 - Timber structures - Structural timber Edinburgh Napier University, Edinburgh Napier University
and glued laminated timber - Determination of some Student grant initiative, Wood Products Innovation
physical and mechanical properties. s.l. : British Standards Gateway, Ochil Timber Products, Aptus Fastener Systems,
Institute, 2010. Kraft Architecture, Oregon Timber Frame Ltd
Kermani, and Porteous. Structural Timber Design to Further information

Eurocode 5. Oxford : Blackwell Publishing, 2007.
Mark Kirk (E: M.Kirk@napier.ac.uk) or
Dr Robert Hairstans (E: r.hairstans@napier.ac.uk)
MacLeod, Iain A. Analytical Modelling of Structural
Systems: An Entirely New Approach with Emphasis
on Behaviour of Building Structures. Chichester : Ellis
Horwood, 1990.
Blass, H. J. Timber Engineering Step 1. Almere : Centrum

Hout, 1995.
Fig 1: Images of the Kraft truss and the ITW FloorTrus respectively
Fig 2: Comparison plot of validation method and analysis software results
Long-term fatigue deterioration and life expectancy of
masonry arch bridges
Spyridoula-Iris Koltsida
University of the West of England 27
Project objectives and goals References
The aims of the project are to develop deeper Melbourne, C., Tomor, A.K., (2006) Application of acoustic
understanding of the fatigue deterioration process of emission for masonry arch bridges. Strain, 42(3), pp. 165-
masonry, quantify the long-term fatigue capacity and 172.
remaining service life of masonry arch bridges and propose
life-cycle management techniques for the use of the road Wang, J., Tomor, A.K., Melbourne, C., (2013) Critical
and railway industry. review of research on high-cycle fatigue behaviour of brick
masonry. Construction and Building Materials, 38, pp.
While there is a wealth of knowledge available on the 602-609.
fatigue behaviour of steel and concrete bridges, very little
information is currently available on the fatigue behaviour Melbourne, C., Tomor, A.K., (2004) Cyclic load capacity
and deterioration process of masonry. The proposed and endurance limit of multi-ring masonry arches.
project will seek to help identify the expected fatigue life Proceedings of the ARCH04 Conference, Barcelona.
for masonry based on a range of small-scale laboratory
test data. Although it will not be possible to provide a Wang, J., Tomor, A.K., Melbourne, C., (2010) Study
comprehensive range of numerical data for a wide range of of high-cycle fatigue behaviour of brick masonry. 6th
masonry types, an initial model will be suggested together International Conference on Arch Bridges, Fuzhou, Fujian,
with working examples as a first step towards estimating China.
the life expectancy of masonry arch bridges.
Ronca, P., Franchi, A., Crespi, P., (2004) Structural failure of
Description of method and results historic buildings: masonry fatigue tests for an interpretation
model. Structural Analysis of Historical Constructions, pp.
The literature review will provide an overview on available
273-279.
test data, fracture mechanism, available measurement and
monitoring techniques, materials, numerical and modelling
Oliveira,D V., Lourenço,P B., Roca,P., (2006) Cyclic
approaches.
behaviour of stone and brick masonry under uniaxial
compressive loading. Materials and Structures, 39(2), pp.
Laboratory tests will be carried out on small-scale
247-257.
masonry samples under static and cyclic loading, under
compression or shear. The crack growth process will
Kashefi, K., Zandi, A.P. and Zeinoddini, M. (2010) Fatigue
be monitored by means of digital image correlation and
life evaluation through field measurements and laboratory
acoustic emission techniques.
tests. Procedia Engineering [online]. 2 (1), pp.573-582.
[Accessed 23/10/2012].
Finally, analytical tools like Finite element (ANSYS) and
limit state analysis (Limitstate Ring and Archie M) software
Pippard, A.J.S. (1941) Repeated load tests on a voussoir
will be used to perform parametric studies on small-scale
arch. Journal of the Institution of Civil Engineers [online]. 17
samples as well as the behaviour of full-scale masonry arch
(1), pp.79-86. [Accessed 23/10/2012].
bridges with regards to various loading cases.
Tung, S., Shih, M. and Sung, W. (2008-12-01)
Potential of application of results
Development of digital image correlation method to analyse
Results will be used directly disseminated to the European crack variations of masonry wall. Sadhana. 33 (6), pp.767-
railway authorities through the International Union of 779.
railways that is part-funding the project with the aim
to optimise and improve the life-cycle management of Funding body
masonry arch bridges that represent ca. 40% of the
The project is half funded from the University of the West
European bridge stock.
of England and half funded from the International Union of
Railways (UIC).
A more informed long-term view on bridge deterioration
processes, estimated remaining service life and feasibility
Further information
of interventions will help bridge owners, bridge managers,
engineers and designers to reduce maintenance costs and Spyridoula-Iris Koltsida (E: Iris.Koltsida@uwe.ac.uk) or Dr
expand life-span of masonry arch bridges without inducing Colin Booth (E: Colin.Booth@uwe.ac.uk) and Dr Adrienn
a negative impact on the operation reliability and the Tomor (E: Adrienn.Tomor@uwe.ac.uk)
structural safety.
A new form of architecture for concrete structures
28 Kaloyana Kostova
University of Bath
Oral presenter
Project objectives and goals A novel splayed anchorage system is proposed for
reinforcing thin optimised fabric-formed concrete elements
A wide range of architectural forms and applications are
where the use of conventional anchorage methods,
possible using fabric formwork, all combining efficiency,
such as hooks or bends, becomes impossible. Previous
elegancy and economy. Inexpensive synthetic textiles in
research studies have demonstrated that the system can
the form of reusable flat sheets offer an easy construction
resist significant pull out forces through the mechanical
method for creating a variety of structurally efficient
action of a wedge, formed by splaying the ends of a
concrete elements [1]. Fabric-formed concrete elements
reinforcing bar [4]. A new experimental investigation,
also benefit from an effective compaction by external
comprising a series of hinged-beam bonded tests, has
vibration methods, whilst the use of permeable textiles
confirmed that the splayed anchorage can be successfully
has the already known effect of improving the quality of
applied across a variety of tested specimens and a model
concrete surface, due to the reduced water-to-cement
capable of predicting the behaviour of the anchorage has
ratio near the fabric face, adding to a better appearance
been developed.
and durability of concrete [2]. Nevertheless, fabric-formed
concrete structures differ significantly from those built
Potential of application of results
in a traditional way and the conventional design and
construction approaches for reinforced concrete may not The findings of this research will be integrated in an
always be applicable. automated design process, combining structural
optimisation, form-finding and design of fabric-formed
The aim of this research is to demonstrate the feasibility reinforced concrete structures. This will allow further
of fabric-formed concrete structures through experimental exploration of the fabric formwork applications by
and analytical work. The objectives of the study include engineers and architects and will enable them to design
providing design methodology verified by physical testing, sustainable fabric-formed concrete structures through
applying innovative materials and suitable construction optimised use of materials.
techniques for fabric formwork construction, development
of a novel anchorage system to be used in fabric-formed References
members and creating a design tool offered for free to the
[1] D. Veenendaal, M. West and P. Block, “History and
industry.
overview of fabric formwork: using fabrics for concrete
casting”, Structural Concrete, 12 (3), 164-177 (2011).
Experimental investigation of the structural behaviour and [2] W.F. Price, Controlled permeability formwork (C511),
buildability of fabric-formed concrete beams has been CIRIA (2000).
undertaken in order to develop design methodology and
guidance for the construction of fabric-formed members [3] K.Z. Kostova, A.P. Darby, T.J. Ibell and M. Evernden,
(Fig. 1). The design of fabric-formed members requires “Advanced composite reinforcement for fabric-
an ability to predict the geometry of the produced cross- formed structural elements”, In: Proceedings of
sections, a structural optimisation process, a method for 2nd International Conference on Flexible Formwork
design of variable sections and provisions for adequate (icff2012), Bath, UK, June, 2012.
anchorage and protection against corrosion of steel
bars, which may be difficult to achieve with the use of [4] A.P. Darby, T.J. Ibell, S. Tallis and C. Winkle, “End
flexible fabric formwork. Innovative materials, such as Anchorage technique for internal FRP reinforcement”.
Fibre Reinforced Polymer (FRP) reinforcement, which In: Proceedings of 8th International Symposium on
has high resistance to corrosion and less stringent cover Fiber Reinforced Polymer Reinforcement for Concrete
requirements, can offer a suitable alternative to the Structures (FRPRCS-8), Patras, July, 2007.
conventional steel reinforcing bars for fabric formwork
construction. Furthermore, the lightweight and low stiffness
of these materials allow new approaches for installation,
which contribute to savings in time and to ease of
construction [3].
Funding body
The Leverhulme Trust
Further information
Kaloyana Kostova (E: K.Z.Kostova@bath.ac.uk) or
Prof Tim Ibell (E: T.J.Ibell@bath.ac.uk)
Fig 1: Fabric-formed beam
Detecting anomalous bridge behaviour using distributed
temperature measurements
29 Rolands Kromanis
University of Exeter
Poster presenter
Motivation Aim and objectives

Aging infrastructure is drawing vast interest among The aim of this study is to develop a fast and robust
structural engineers. Many structures constructed during method for anomaly detection by taking advantage of
the 20th century are inevitably reaching the end of their the correlation between temperature distributions across
design life. Bridges are an important component of our civil a structure and the measured structural response. The
infrastructure and play a vital role in today’s busy society. research focuses on the development of data-driven
They are often taken for granted to provide a permanent strategies for integrating the thermal response of bridges in
means of transport connectivity between artificially the measurement interpretation process. It is, in principle,
or naturally separated land-sites. Traffic disruptions a first step towards using distributed temperature and
that may arise from temporary closure of bridges due response measurements for structural performance
to maintenance works often result in frustration and monitoring. The objectives are to (i) demonstrate that a
confusion for daily commuters. For this reason, novel data-driven strategy could accurately predict the thermal
bridge management models that reduce total lifecycle response of a structure from distributed temperature
costs are required. An example of such a model is shown measurements and (ii) such a strategy could then form the
in Fig. 1. With the aid of continuous monitoring and visual basis of an anomaly detection methodology.
inspections the onset of damage can be detected at an
early state, therefore preventing further deterioration of the Methodology and early results
structure and thereby leading to savings in maintenance
The premise is built upon the application of a recently
time and costs.
proposed regression-based thermal response prediction
(RBTRP) methodology which aims to forecast the real-time
In recent years, Structural Health Monitoring (SHM)
response from distributed temperature measurements
systems have been deployed more frequently on bridges
with the least prediction error (PE). For this purpose,
with the objective of tracking their real-time performance.
multiple regression-based algorithms (e.g. support vector
Emerging developments in sensing systems are envisaged
regression (SVR), multiple linear regression (MLR)) are
to form the underpinning technologies for the smart
considered. Their performances are compared on multiple
infrastructures of the future. Sensors are becoming more
measurement data sets. Initially, data is generated using
affordable and robust, and energy harvesting and wireless
a wide range of numerical models and, subsequently, a
sensing technologies are increasingly finding penetration
laboratory truss structure representative of a bridge section
in bridge monitoring. While these technological advances
found in highway bridges is built to continuously monitor
have greatly simplified measurement collection, there still
structural response (strain) and temperature distribution.
remain a number of challenges in managing and analysing
The lowest PE is achieved with SVR. The difference
the collecting data. This paper aims to address these
between predicted and real-time responses results in PE
challenges in measurement interpretation by pursuing the
signals which are further inspected for patterns that could
following objectives:
indicate the occurrence of anomalous structural behaviour
(e.g., damage, unusual event). Figure 2 illustrates the
• To develop fundamental methodologies for processing
correlation between temperature and response as well as
and interpreting the deluge of measurements
the predicted and real-time response.
generated by sensing systems;
Preliminary results from the continuous monitoring of
• To quantify the effects of variations in ambient
a laboratory truss structure have led to the following
conditions.
conclusions:
Previous long-term monitoring studies have illustrated that
• The relationship between distributed temperature
daily and seasonal temperature variations have a great
and response measurements can form the basis for
influence on the structural response of bridges (Yuen and
anomaly detection techniques.
Kuok, 2010; Hua et al., 2007), and that this influence may
even exceed the response to vehicular traffic (Catbas et
• Thermal responses of structures are accurately
al., 2008). This paper thus aims to evaluate the changes in
predicted with the application of SVR.
structural response due to temperature variations and then
employ this understanding for damage detection.
• The prediction error, which is the difference between
a prediction from a SVR model and a corresponding
measurement, is a reliable indicator of damage.
Methods would be applicable to the real-time continuous Funding body
monitoring of full scale bridges. This could help to improve
University of Exeter
quality of bridge inspection, detecting damage at an early
state. This would reduce maintenance costs, prolong
Further information
service life and improve its safety.
Rolands Kromanis (E: rk296@exeter.ac.uk) or
References Dr Prakash Kripakaran (E: P.Kripakaran@exeter.ac.uk)
Catbas, F. N., Susoy, M. & Frangopol, D. M. (2008).
Structural health monitoring and reliability estimation: Long
span truss bridge application with environmental monitoring
data. Engineering Structures, 30, 2347-2359.
Hua, X. G., Ni, Y. Q., Ko, J. M. & Wong, K. Y. (2007).

Modelling of temperature-frequency correlation using
combined principal component analysis and support
vector regression technique. Journal of Computing in Civil
Engineering, 21, 122-135.
Yuen, K. V. & Kuok, S. C. (2010). Ambient interference in

long-term monitoring of buildings. Engineering Structures,
32, 2379-2386.
Fig 1: A framework for bridge management
Fig 2: Time series of strains (real-time and predicted) and temperature
Web crippling behaviour of cold-formed steel channel
sections with web openings
30 Ying Lian
Queen’s University Belfast
Project objective and goals Web crippling (or crushing) at points of concentrated, or
localised, load or reaction in thin wall beams is well known
The primary aim of this project is to develop a better
to be a significant problem, particularly in the case of
understanding of the web crippling behaviour of both cold-
beams with slender webs, and is of high significance in
formed standard steel channel sections and cold-formed
the field of cold-formed steel members, as such members
stainless steel channel sections due to the presence of
are generally not stiffened against this type of loading (see
web openings. Design recommendation will be developed.
Fig. 2). At points of concentrated loading and support,
severe lateral loading can result in local buckling in the
Specific objectives of the research are as follows:
web (Rhodes and Nash, 1998). The American Society of
Civil Engineers Specification (ASCE, 2005) for the design
• To conduct experimental investigations to study
of cold-formed steel structural members provides design
the effects of web openings on the web crippling of
rules for the design of flexural members against web
sections.
crippling based on the American Iron and Steel Institute
Specification (AISI, 2005). Eurocode-3 (1996) provides a
• To develop nonlinear finite element analysis (FEA)
set of equations to avoid crushing, crippling or buckling
models for interior-one-flange (IOF) and end-one-
of webs subjected to support reactions or other local
flange (EOF) loading conditions.
transverse loads applied through the flanges. However,
there is no recognised design specification for the influence
• To correlate the experimental results with the results of
of the hole on the structure integrity (i.e. reduction in
the FE model.
strength). For this reason, a comprehensive experimental
and numerical research programme investigating the web
• To undertake an extensive FE parametric study to
crippling behaviour of cold-formed steel was initiated in
investigate possible parameters which are expected
QUB.
to affect the web crippling strength of cold-formed
channel sections. The parameters include: web hole
Method and results
sizes and position into the web of the section, section
geometry, section thickness and different lengths of Experimental testing
bearing plate.
Cold-formed steel channel sections without holes and
with holes under interior-one-flange (IOF) loading and
• To propose design recommendations of channel
under end-one-flange (EOF) loading for both unfastened
sections with web openings various loading conditions
and fastened condition are being conducted in laboratory
and cases.
study. For both cases, the holes are cut centred beneath
the concentrated load, the weakest part of the specimen.
Background
There is no recommendation covering these cases.
Cold-formed steel members (also known as thin-walled
steel members) have traditionally been used as secondary FE model
non-loading structural members in low-rise commercial and
A non-linear shell finite element model of cold formed steel
light residential applications, such as wall studs, floor joists,
with and without holes subjected to web crippling under
and roof purlins. As the strength of steel that can be cold-
interior-one-flange (IOF) and end-one-flange (EOF) loading
formed increases, cold-formed sections are also now used
conditions has been set up in ABAQUS (2006).
as primary loading bearing members, such as floor beams
in buildings and column and rafter members in portal
frames. These members are commonly manufactured with
pre-punched holes in the web. Such openings provide for The results of research will lead to better understanding of
easy installation of electrical and plumbing systems (Fig. 1). the web crippling behaviour of cold-formed steel channel
The hole can either be pre-punched in the web following sections with web openings and it could develop design
the cold-rolling process, or the hole can be field cut. recommendations. It can also help to identify the good or
bad practice on current cold-formed steel sections with
web openings.
ABAQUS (2006). Standard user’s manual. Version 6.6. Queen’s University Belfast
Vols. 1-3. Hibbitt, Kalsson and Sorensen, Inc. USA.
Further information
AISI (2005). Specification for the design of cold-formed
Ying Lian (E: ylian01@qub.ac.uk) or
steel structural members. Washington, D. C: American Iron
Dr James Lim (E: j.lim@qub.ac.uk)
and Steel Institute.
ASCE (2005). Minimum design loads for buildings and

other structures. New York: American Society of Civil
Engineers.
Eurocode-3 (1996). Design of steel structures: part 1.3:

General rules- Supplementary rules for cold-formed thin
gauge members and sheeting. ENV 1993-1-3. Brussels,
Belgium: European Committee for Standardization.
Rhodes, J & Nash, D. (1998). An investigation of web

crushing behaviour in thin-walled beams. Thin-Walled
Structures, 32, 207-230.
Fig 1: Cold-formed steel beams with holes
Fig 2: Web crippling at a support point (Rhodes J and

Nash D, 1998)
Modelling of steel connections at elevated temperatures
31 Shuyuan Lin
Brunel University
Poster presenter
Project objectives and goals A total of 22 end-plate connection tests have been used
for validation of the new 2-noded connection model. The
Structural engineers and architects have a responsibility
validation includes 8 connection fire tests without the axial
for incorporating fire safety into their building designs in
force, 12 tests subjected to axial force at both ambient and
order to minimise loss of life and property. One aspect of
elevated temperatures and 2 fire tests on the beam-to-
this is to ensure that structural stability is maintained if a
column sub-frame. The comparison results show that the
fire develops. For last two decades, extensive research has
predictions provided by this new model match well with the
been carried out on the behaviour of steel-framed buildings
experimental data.
under fire conditions. It is well known that robustness of
steel connection is vital important to the fire resistance of
steel-framed composite buildings. The development of
effective connection models is a key issue in this research In this study a new connection model is developed for
field. Hence, the main objective of this project is to develop accurately predicting the behaviour of the end-plate
a robust 2-noded connection element for modelling the connections at elevated temperatures. In the model the
connections between steel beam and column at elevated connection failure due to bending, shear, axial tension and
temperatures. After the new development, a series of compression are considered. Also the influence of the
comprehensive parametric studies will be conducted to axial force of the connected beam on the connection’s
deeply understand the influence of the connection on the behaviour is taken into account. This model is very
fire resistance of steel and composite structures. Based computationally effective and can be applied to whole
on the results achieved in this research some constructive building modelling in the real structural fire engineering
design recommendations will be proposed. design. The next phase of the research will be mainly
focused on the developments of connection models
Description of method and results for simulating the behaviour of partial end-plate and
fin plate connections in fire. Those connections will be
At present, the component-based model is of great
represented as two-noded connection elements. After
popularity for modelling the behaviour of connections at
those developments, comprehensive parametric studies
elevated temperatures. This model is based on dividing the
will be conducted to assess the influence of connection
connection into its basic components, such as end plate,
robustness on the fire resistance of steel and composite
column flange, bolts, etc., and each component is idealised
structures. Based on this study some constructive design
as a spring. Each spring, regarded as the individual
recommendations will be proposed.
basic component, has its own strength and stiffness
characteristics in tension, compression or shear. Then the
References
overall behaviour of the connection can be represented by
the combination of these springs. European Committee for Standardization CEN (2005), BS
EN 1993-1-8, Eurocode 3: Design of steel structures, Part
During the first year, the behaviour of bolted end-plate 1.8: Design of joints, British Standards Institution.
connection between steel beam and column at elevated
temperature was thoroughly investigated and a robust Huang, Z. (2011), “A connection element for modelling
2-noded connection element for modelling flush and end-plate connections in fire”, Journal of Constructional
extended end-plate connections has been developed. Steel Research, 67(5), 841-853.
The model is the further development of the connection
element proposed by Huang (2011). In Huang’s model, the Yu, H., Burgess, I.W., Davison, J.B. and Plank, R.J.
resistances of tension and compression for the end-plate (2009), “Experimental investigation of the behaviour of fin
connection are found to be conservative compared to the plate connections in fire”, Journal of Constructional Steel
test results. This is because Huang’s models are based on Research, 65, 723-736.
the formulas proposed in Eurocode 3 Part 1.8(2005) with
the extension to fire. Hence, in this research a robust model
was developed for more accurately predicting the tension
and compression resistances of end-plate connections
under fire conditions. In this model the connection failure
due to bending, axial tension and compression, shear
is considered. Also the influence of the axial force of the
connected beam on the connection has been taken into
account. The new developments were incorporated into
Huang’s two-noded connection element for predicting
the behaviour of end-plate connections at elevated
temperatures.
Further information
Shuyuan Lin (E: Shuyuan.Lin@brunel.ac.uk) or
Dr. Zhaohui Huang (E: Zhaohui.Huang@brunel.ac.uk)
Fig 1: Comparison results of tests EP_450_35_23-11-07 (450°C, θ=35°)

and EP_450_45_23-10-07 (450°C, θ=45°) (Yu, et al. 2009).
Behaviour and design of under-deck and combined
cable-stayed road bridges with steel and composite decks
32 Fernando Madrazo-Aguirre
Imperial College London
Introduction • Stresses due to anchorage rotations [8] may be

relevant for the stays, whose cross-sectional areas are
Under-deck and combined cable-stayed bridges (UDCSB
governed by fatigue [5, 6].
and CCSB) are innovative cable-stayed bridges that
have been built since the 1970s [1]. In these bridges the
stays follow non-conventional layouts in comparison with
conventional cable-stayed bridges [2]. In UDCSB stays are The objectives of the current research are:
located under the deck; and in CCSB the stays are located
both above and under the deck [3]. • To develop suitable finite element (FE) models for the
analysis of UDCSB and CCSB road bridges with steel
These two bridge types have been studied in depth and composite decks.
for road bridges with prestressed concrete decks [3-
6]. The main features are: (1) highly efficient structural • To analyse the static response of these bridges under
behaviour; (2) significantly larger deck slendernesses than persistent and transient (construction) situations.
in conventional bridges (1/80 and 1/100, for single-span Special attention will be paid to the prestressing
and continuous bridges respectively); (3) smaller amount of action, and how much permanent and live load should
materials than in conventional bridges are required, which be compensated.
allows for a more sustainable design; (4) unconventional
design criteria (e.g. the serviceability limit state (SLS) of • To investigate the dynamic behaviour of these bridges,
vibrations determining the deck depth); (5) strong aesthetic and how the vertical accelerations can be effectively
characteristics. controlled with either the structural configuration or
the introduction of dampers. Dynamic live load due to
Steel or composite (steel and concrete) decks seem prima traffic will have to be selected, since codes usually do
facie to be very appropriate for these bridge types when not state the dynamic component for road bridges.
employed for road viaducts. Apart from being lightweight
solutions with high durability and aesthetics, steel and • To analyse the stability of these bridges during erection
composite decks allow for high prefabrication degree by taking into account imperfections and by paying
with its obvious advantages: quality, precision, safety and special attention to post-buckling behaviour.
construction speed. In addition, the following features
would enhance their structural behaviour in comparison • To investigate the linear and nonlinear behaviour, and
with UDCSB and CCSB with prestressed concrete decks: how both material and geometric nonlinearities can
affect the structural response.
• Larger prestressing of the stay cables could be
introduced (absence of significant time-dependent • To identify optimal structural configurations that enable
effects) in order to compensate part of the live load in all relevant limit states to be satisfied.
addition to the dead load [4].
• To provide design recommendations and guidelines.
• The lower stiffness of the deck relative to the cable-
staying system would increase the efficiency of the Description of method and results
cable-staying system under live load [4].
FE models will be developed to perform parametric
analyses in order to seek for the optimal structural
• As a consequence, larger deck slenderness
configuration.
and smaller amounts of material are expected,
consequently leading to a more sustainable design.
Some initial analyses have been performed with a single-
span double I-beam composite UDCSB. These initial
• On the other hand, the following features may limit the
results have shown that:
performance of steel and composite decks:
• High efficiencies are obtained (85%; portion of the
• Steel decks with large slenderness ratios are prone
external bending moment due to live load resisted by
to buckling under compression loads [7]. This may
the cable-staying system).
determine how the prestressing load is applied during
the construction stage.
• SLS of vibrations governs the deck depth.
• Steel and composite decks are prone to larger vertical
• When loads are applied with an eccentricity,
accelerations due to live loads.
accelerations are increased by 40% due to the low
torsional stiffness of these types of cross-sections
• Larger prestressing forces would increase the second-
(I-beams).
order effects.
• Stays cross-section is governed by fatigue, and the Funding body
stresses due to the bending rotation of anchorages
Department of Education, Universities and Research of the
are not negligible.
Basque Government (link) (Spain)
• The effects of some parameters (stays cross-
Further information
sectional area, stay transverse layout, beam depth,
slab thickness) on main dynamic modes have been Fernando Madrazo-Aguirre (E: fmadrazoaguirre@gmail.
studied. com) or Dr Ana Ruiz-Teran (E: a.ruiz-teran@imperial.ac.uk)
and Dr Ahmer Wadee (E: a.wadee@imperial.ac.uk)
The lack of research and guidelines for the design of
UDCSB and CCSB with steel and composite decks, may
prevent designers, contractors and clients from projecting
these bridge types.
The current research aims to gain understanding on the

structural response of these bridge types, and to define
a set of design criteria. Consequently, a highly useful
tool may be produced for the design and construction of
UDCSB and CCSB with steel and composite decks.
References
[1] Ruiz-Teran AM, Aparicio AC. General overview of two
new types of cable-stayed bridges: Under-deck cable-
stayed bridges and combined cable-stayed bridges.
Proceedings of ICE Bridge Engineering. Special Fig 1: Weitingen Viaduct (first UDCSB, designed by F.
issue about ‘Bridge Engineering: Cable-Supported Leonhardt, Germany, 1978)
Structures’. June 2010; 163(2): 67–78.
[2] Walther R, Houriet B, Isler W, Moïa P, Klein JF. Cable-

stayed bridges. Thomas Telford, 1988.
[3] Ruiz-Teran AM, Aparicio AC. Two new types of bridges:

under-deck cable-stayed bridges and combined cable-
stayed bridges. The state of the art. Canadian Journal
of Civil Engineering. 2007; 34(8): 1003–1015.
[4] Ruiz-Teran AM, Aparicio AC. Parameters governing

the response of under-deck cable-stayed bridges.
Canadian Journal of Civil Engineering. 2007;
34(8):1016–1024.
[5] Ruiz-Teran AM, Aparicio AC. Structural behaviour and

design criteria of under-deck cable-stayed bridges and
combined cable-stayed bridges. Part 1: Single span
bridges. Canadian Journal of Civil Engineering. 2008;
35(9): 938–950.
Fig 2: Obere Argen Viaduct (first CCSB, designed by J.
[6] Ruiz-Teran AM, Aparicio AC. Structural behaviour and
Schlaich, Germany, 1991)
design criteria of under-deck cable-stayed bridges and
combined cable-stayed bridges. Part 2: Multi-span
bridges. Canadian Journal of Civil Engineering. 2008;
35(9): 951–962.
[7] Saito D, Wadee MA. Optimal prestressing and

configuration of stayed columns. Proceedings of the
Institution of Civil Engineers – Structures and Buildings.
2010; 163(5): 343–355. ISSN:0965-0911.
[8] Virlogeux M. Fatigue in cable-stayed bridges. In: XII

Congress FIP. 1994 Washington DC, USA.
Bending behaviour of column face of concrete filled square
hollow sections
33 Mohammed Mahmood
The University of Nottingham
The use of tubular sections in structural applications is References

attractive for both structural engineers and architects
Al-mughairi, A., Tizani, W. & Owen, J. Year. Connection
due to their high strength to depth ratio and aesthetic
to Concrete Filled Hollow Section using Extended
appeal. However, the difficulty of providing fully bolted site
Hollobolt (Moment Connection Tests). In: LAM, D., ed. 9th
connection limits their use. To overcome this problem,
International Conference on Steel Concrete Composite and
different types of blind bolting systems were proposed.
Hybrid Structures, 2009 Leeds, UK. 599-604.
The Extended Hollow Bolt is a promising blind bolt system,
which was introduced in the University of Nottingham
Tizani, W., Wang, Z. Y. & Hajirasouliha, I. 2013. Hysteretic
especially for concrete filled hollow sections. This bolt
Performance of a New Blind Bolted Connection to
was developed to improve the connection stiffness by
Concrete Filled Columns under Cyclic Loading: An
involving the concrete in the pullout resistance of the
Experimental Investigation. Engineering Structures, 46,
bolt by embedding the bolt shank and nut inside the
535-546.
concrete. The Extended Hollow Bolt was proven to
be used adequately in moment resisting connections
Funding body
(Al-Mughairi et al., 2009, Tizani et al., 2013). However, the
behaviour of the joint components when using this type The Iraqi Government
of blind bolt has not been investigated yet, and one of
these aspects is the column face bending. Therefore, this Further information
research aimed at investigating the bending behaviour of
Mohammed Mahmood (E: evxmsma@nottingham.ac.uk)
column face component of Concrete Filled Square Hollow
or Dr Walid Tizani (E: walid.tizani@nottingham.ac.uk)
Sections (CFSHS) connected using Extended Hollow
Bolts. To achieve the aim of this research, an experimental
programme will be carried out to examine the bending
behaviour of the column face by varying influential design
parameters. Around thirty specimens of CFSHS will be
tested under tensile load single row and double row of
Extended Hollo Bolts.
A finite element model will also be developed using

simulation software to predict the structural behaviour of
the tested specimens. The first step will be validating the
model with the experimental results. After that a parametric
study will be performed to further understand the behaviour
of column face bending. Finally, based on the results of
the experimental program and the finite element analysis,
simplified theoretical approach will be established. This
approach could be used for the design purpose of this type
of connection using the component method.
Study of heat-induced concrete spalling using a novel fire
testing methodology
Cristian Maluk
Poster presenter
Project objectives and goals technique particularly in replicating the thermal loading
imposed by a given standard testing furnace. Furthermore,
This research project studied the effects of including
the high repeatability of H-TRIS tests was demonstrated as
different types and amounts of polypropylene fibres
compared with the scatter seen in the otherwise identical
(length, diameter, shape, and dosage) on the occurrence
furnace tests for a range of measured temperatures in
of heat-induced concrete spalling. This was accomplished
identical specimens.
using a novel fire testing methodology named the Heat-
Transfer Rate Inducing System (H-TRIS). The methodology
Eleven concrete mixtures were tested (refer to Fig 2) under
was developed in an attempt to tackle some of the
sustained compressive stress or under a free-to-expand
shortcomings of ‘standard’ furnace fire testing procedures
condition (in triplicate in each case), thus also replicating
in the to study heat-induced concrete spalling using a
the extreme cases of possible mechanical conditions
thermal loading technique in which the thermal exposure
experienced during furnace testing. A study of the effects
is actively controlled, not using the traditional gas phase
of polypropylene (PP) fibre type on the occurrence of heat-
temperature, but rather by controlling incident heat flux
induced concrete spalling was performed by assessing
(i.e. thermal energy) measurements at the test element’s
time to first spalling, volume and mass spalled, and the
heat-exposed surface. H-TRIS was developed in order
amount of thermal energy absorbed up to the moment of
to generate rational fire resistance and spalling studies
first spalling. Parameters such as PP fibre diameter, length,
with high thermal repeatability, proper, realistic boundary
dosage, and shape were assessed, aiming to further
conditions, and statistical confidence, all at low economic
understand the mechanism(s) by which PP fibres reduce or
and temporal cost.
prevent the occurrence of heat-induced concrete spalling.
The new testing methodology, H-TRIS, uses a mobile
It is hoped that in the near future, others within the
array of propane-fired high performance radiant panels,
structures and fire engineering communities (both
along with a mechanical linear motion system (refer to
practitioners and researchers) are encouraged by the
Fig1). Thermal loading of test specimens is controlled
high repeatability, good statistical confidence and low
using incident heat flux (i.e. thermal energy) measurements
economical and temporal cost demonstrated by this study,
taken by a heat flux gauge during a pre-programmed
into developing more research projects that can properly
calibration procedure. Calibration can thus be performed
understand and subsequently mitigate heat-induced
before each new test, relevant to the specific conditions of
concrete spalling, which are marginal to date. Furthermore,
the laboratory on any given day, if desired. The computer-
H-TRIS has proved to be an advantageous experimental
controlled linear motion system is used during testing to
tool for the research and development of intumescent
adjust the radiant panels’ location (i.e. distance from the
systems, timber-based materials and others building
specimen) to follow any pre-defined time versus heat flux
construction materials and systems.
relationship. H-TRIS thus allows accurate quantification
of the thermal energy absorbed by a tested element with
precision and repeatability; all at negligible economical and
temporal costs in comparison to a standard fire resistance
(i.e. furnace) test [1]. It also allows rational simulation of any
given standard furnace exposure condition.
The experimental study discussed herein was intended

to replicate the thermal conditions (i.e. internal thermal
gradients) experienced within concrete specimens during
standard fire resistance tests performed in the large scale
floor testing furnace [2] at EMPA, Switzerland. Experimental
internal temperature measurements (at depths of 10, 20
and 45 mm from the exposed concrete surface) were used,
along with an inverse modelling procedure, to theoretically
calculate the time history of thermal energy absorbed
by the specimens in the furnace test; this was then
replicated and reproduced using H-TRIS. The comparison
of internal thermal gradients between effectively identical
concrete elements tested in the furnace and in H-TRIS
was satisfactory, thus partially validating the use of this
[1] Maluk C., Bisby L., Terrasi G., Krajcovic M., Torero Cristian Maluk (E: c.maluk@ed.ac.uk) or
J.L.; Novel Fire Testing Methodology: Why, How and Dr Luke Bisby (E: luke.bisby@ed.ac.uk)
What Now?, Mini-symposium on Performance-based
Fire Safety Engineering of Structures as part of The 1st
International Conference on Performance Based and
Life Cycle Structural Engineering (PLSE), Hong Kong,
December 5-7 2012, 11pp.
[2] Terrasi G.P., Bisby L., Barbezat M., Affolter C. and

Hugi E.; Fire Behavior of Thin CFRP Pretensioned High
Strength Concrete Slabs, Journal of Composites for
Construction, 16 (4), 2012, pp. 381–394.
Fig 1: Heat-Transfer Rate Inducing System (H-TRIS)
Fig 2: Layout of concrete specimens tested with H-TRIS
Aspects of bolted connections in fibre reinforced polymer
structures
Navroop Singh Matharu

The emergence of advanced materials, such as Pultruded Description of method and results
Oral presenter
Fibre Reinforced Polymer (PFRP) composites, that
My research project has involved a comprehensive
can compete with traditional building methods in civil
program of experimental testing of bolted connections
engineering applications has been steadily growing for
in PFRP materials, with the building blocks of structural
several decades. As with any relatively ‘young’ technology,
connections being investigated. These fundamental
the wider structural engineering community can be prudent
connection criteria in PFRP materials provide fabricators,
if not reluctant to fully adopt and widely utilise it without
designers and contractors confidence in using the material,
a solid foundation of guidance and design data. These
and critically, are statistically verified using accepted
materials can often be found in standard structural shapes
standards.
that mimic steel sections and are to be used in primary
load bearing structures for example in non-sway braced
The characterisation of pin-bearing strengths has been
frames as found at the East Midlands Parkway Railway
conducted by way of a non-standard test method,
Platform [1]. Within the last decade, the application of
developed at the University of Warwick, and presents
pedestrian and road bridges with PFRP components has
advantages compared to other non-specific methods
rapidly increased, with over 2000 bridges being opened
especially with regards to evaluating targeted resistance
in The Netherlands since 2001 alone [2]. The material has
values for PFRP materials [4]. The test program, which
been utilised in these applications for its key advantages of
receives external support from PFRP fabricators Access
high strength-to-weight ratio, potential for low maintenance
Design and Engineering (and Creative Pultrusions Inc.) by
over the structures lifetime (corrosion resistant) and good
way of materials and structural engineering support from
electrical and thermal properties. Furthermore, PFRP
consultants Mott Macdonald, has allowed the specific
structural shapes can be easily connected and assembled
needs of the industry to be addressed. In particular, the
with conventional mechanical fasteners, often stainless
study focuses on bolt thread in bearing and issues of long-
steel bolting, hence reducing construction times and costs.
term performance through accelerated aging. Tests results
have been produced and analysed using both European
Objectives and goals
and American standard method to determine characteristic
The design and verification of bolted connections in PFRP values for both plain and thread bearing strength of PFRP,
frames is a complex exercise that has considerable gaps and associated effects of change in pin diameter, material
in knowledge. It is the aim of the author’s PhD work to orientation with respect to loading direction and aged
address some of these issues by way of experimental conditioning.
testing, analysis and evaluation of targeted test results to
either revise or modify current design provisions so that The study into plate-to-plate connections will utilise both
risk of structural failure can be minimised. Specific goals the single and double-lap shear configuration and consider
include: the failure modes of net-tension, bearing and shear-out;
this second part to the comprehensive testing programme
• Development of the test methodology used for in progress.
the determination of pin‐bearing strength (without
lateral restraint), and to characterise this key strength
property for use in the design of bolted connections in
FRP shapes;
• A series of plate‐to‐plate tests that characterise the

resistance of bolted connections of Pultruded material,
which shall lead to the proposal of new or improved
design guidance.
Furthermore, these goals were developed with regards to

addressing the pressing knowledge gaps [3] found in PFRP
bolted connections. The results are to be further evaluated
with due consideration of specific failure modes found in
PFRP materials and their associated design equations
in order to develop their appropriate reliability for use in
structural connection design.
Potential application of results References
The immediate benefactors of the project results and [1] Anon. (2012) ‘East Midlands Parkway Platform’, Case
recommendations include clients that are currently Study, Redman Composites, Telford, UK, (2012) http://
implementing designs using PFRP materials including www.redmancomposites.co.uk/case-studies/east-
Network Rail in forms of railway platforms, and National midlands-parkway.asp?type=overview .
Grid. Further dissemination of the results will assist a
growing industry by way of the preparation of recognised [2] Anon. (2011) ‘More than 2000 bridges with Fiberline
design rules, such as to be found in the publication of a decks’, News and Press Room, Fiberline Composite,
2013 American LRFD design standard (supervisor is a Kolding, Denmark, http://www.fiberline.com/news/
drafter) for PFRP structural sections and the new-build konstruktioner/more-2000-bridges-fiberline-decks .
sections of an anticipated future structural Eurocode
for FRP materials. Design standardisation for structural [3] Mottram, J.T. (2009) ‘Design Guidance for Bolted
materials is needed to improve and sustain confidence in Connections in Structures of Pultruded Shapes: Gaps
a technology. Hence, upon completion there will hopefully in Knowledge,’ 17th International Conference on
be an increased level of confidence with respect to using Composite Materials (ICCM17), Edinburgh, 2009.
bolted connections in PFRP structural frames for buildings
and bridges. Ultimately, leading towards wealth creation [4] Mottram, J.T. and Zafari, B. (2011) ‘Pin-bearing
and future innovation in structures constructed using these strengths for bolted connections in fibre-reinforced
materials, which envisage having an overall performance polymer structures’, Structures and Buildings, 164 (5),
that satisfies the drivers of sustainability and a Green 291-305.
Economy – some of the key advantages, particularly in
the more environmentally conscious society of today, in Funding
exploiting the use of pultruded shapes in construction.
EPRSC
This project forms part of the larger research project for
Further information
Connections and Joints for Buildings and Bridges of Fibre
Reinforced Polymer. Further details can be found at: http:// Navroop Matharu (E: n.s.matharu@warwick.ac.uk) or
gow.epsrc.ac.uk/NGBOViewGrant.aspx?GrantRef=EP/ Dr J. Toby Mottram (E: j.t.mottram@warwick.ac.uk)
H042628/1.
Blast resilience of large scale urban built infrastructure
Louise McCluskey
Queen’s University Belfast 36
Background any spalling that would normally occur when an ordinary
concrete sample undergoes impact loading [5] and this
When designing any structure it is reasonable to assume
is mainly due to the ability of UHPFRC to dissipate large
that for the majority of its life-span a structure will be
amounts of energy whilst having a very large deformation
subjected to long term static loading, however it is also
capacity [3]. Since the flexural tensile strength and fracture
important to consider the effect that short term dynamic
energy properties increase at high strain rates then
loads can have [1]. Examples of dynamic impulsive or
UHPFRC has been found to be a promising material for
shock loading are explosions or blast loads, accidental
use in blast resistance structures [3].
falling loads, projectiles/missiles striking a structure, or even
an airplane or a road vehicle crashing into a building [1].
Murtiadi et al [10] claim “the use of high-strength concrete
Impact loading is a type of dynamic loading which occurs
is increasing faster than the development of appropriate
over a very short duration but produces very high strain
design code recommendations”, therefore it is vital that
rates within the concrete compared to static and seismic
research is carried out in order to fully understand the
loads, and the deformations that the structure undergoes
complex behaviour of this material so that guidelines for the
are also different [1].
use of UHPFRC can be developed. UHPFRC differs from
ordinary concrete and conventional FRC because there is a
Recently the development of ultra-high performance fibre
pronounced strain hardening phase straight after the initial
reinforced concrete (UHPFRC) is a promising solution
elastic phase.
to prevent the damage caused by high impact loads.
This is mainly due to its high compressive and flexural
Objectives
strength when compared with normal strength concrete
[2]. Enhanced material properties under impact loading The main aim of the project is to develop completely new
mean that UHPFRC is an excellent option for structures at concepts and technology to improve resilience of large
increased risk of terrorist attacks or for military buildings [3]. scale urban built infrastructure to natural and man-made
threats.
It is the relatively low tensile strength and poor energy
absorption properties which can be attributed to the low The proposed research has the following objectives:
impact resistance of normal strength concrete [4]. Under
impact loading, normal strength concrete tends to spall • To develop a material which is resistant to blast and
due to its brittle nature and low energy absorption capacity impact loading, which can resist extremely high
[2]. Even the presence of steel reinforcing bars does not strain loading rates with minimal scabbing. The main
prevent spalling [5]. applications will relate to the design of new structures
and the strengthening of existing structures.
Ultra high performance fibre-reinforced concrete (UHPFRC)
has a cementitious content three times that of ordinary • Experimental tests will be carried out to determine
concrete [5]. It has a very low water/binder ratio and the material properties of UHPFRC, such as the
there is no coarse aggregate used within the mix [2]. compressive, flexural and tensile strengths, the moduli
With no coarse aggregate the concrete is very dense and of elasticity and the energy absorption capacity.
the homogeneity is much higher compared to ordinary Accurate determination of these parameters is
concrete [5], and this also helps to reduce the size of any essential in order to carry out accurate finite element
microcracks which form under different types of loadings modelling of UHPFRC structures and elements.
[6]. The increased density also means that UHPFRC has
a very low permeability and therefore improved durability • Develop non-linear explicit finite element analysis
compared to ordinary concrete [3]. Curing at 90°C results models using Abaqus software
in very high compressive strength values which normally
range between 150-200 MPa [5]. The inclusion of steel • The experimental tests and analytical models will be
fibres within the UHPC mix increases the flexural toughness used together to predict and investigate the behaviour
to values up to 200 times greater than conventional FRC of UHPFRC samples under blast and impact loading
[7]. The fibres within UHPFRC resist the lateral spreading
of cracks by bridging across areas of lower strength, thus • Develop best practice guidelines for improving the
usually leading to a higher failure strength under quasistatic resilience of large scale urban infrastructure to both
loading [5]. natural and man-made threats.
The behaviour of UHPFRC is strain rate dependent, and • It is hoped the use of a material such as UHPFRC
its strength at very high strain rates increases significantly within large scale urban infrastructure will greatly
[8]. Tests carried out [2,3,5,7–9] have shown that UHPFRC improve resistance against explosions and also
has excellent blast and impact resistance characteristics possibly improving resilience to seismic loading at no
and has potential to be used in the field of infrastructure additional cost.
protection [7].The use of UHPFRC will help minimise
Methodology [4] Banthia, N. Impact Resistance of HPFRCC.
International RILEM Workshop on High Performance
Experimental testing
Fiber Reinforced Cementitious Composites in Structural
• Creation of UHPFRC specimens - A number of Applications 479–488 (2006).
UHPFRC specimens will be cast and cured (at up to
90°C using a specially designed curing tank). [5] Millard, S. G., Molyneaux, T. C. K., Barnett, S. J. &
Gao, X. Dynamic enhancement of blast-resistant ultra
• Laboratory Tests - Compression, flexural and tensile high performance fibre-reinforced concrete under
tests will be carried out on the UHPFRC specimens flexural and shear loading. International Journal of
Impact Engineering 37, 405–413 (2010).
• Impact Testing - A drop-hammer rig has been
designed to simulate a blast load on UHPFRC [6] Richard, P. & Cheyrezy, M. Composition of Reactive
samples. This equipment will be used to applied high Powder Concrete. Cement and Concrete Research 25,
strain rates to UHPFRC tiles and notched beams 1501–1511 (1995).
in order to determine the behaviour under dynamic
loading. [7] Rebentrost, M. & Wight, G. Investigation of UHPFRC
Slabs Under Blast Loads Previous Research. UHPFRC
Numerical modelling and finite element analysis 1–8 (2009).
Non-linear explicit finite element analysis will be carried out
[8] Cavill, B., Rebentrost, M. & Perry, V. Ductal ® - An
using Abaqus software in order to predict the behaviour of
Ultra-High Performance Material for Resistance to
UHPFRC samples under blast and impact loading.
Blasts and Impacts. 1–10 (2006).
Design recommendations
[9] Yi, N.-H., Kim, J.-H. J., Han, T.-S., Cho, Y.-G. & Lee, J.
As a result of the combined experimental and analytical H. Blast-resistant characteristics of ultra-high strength
modelling of UHPFRC it is hoped that there will be a concrete and reactive powder concrete. Construction
much better understanding of its behaviour under impact and Building Materials 28, 694–707 (2012).
and blast loading conditions. It is hoped that results will
be consistent enough so that guidelines for the use of [10] Murtiadi, S. Ã. & Marzouk, H. Behaviour of high-
UHPPRC within urban infrastructure can be created, or at strength concrete plates under impact loading.
least pave the way for further research. Magazine of Concrete Research 53, 43–50 (2001).
References [11] Lappa, E. S. High strength fibre reinforced concrete:

static and fatigue behaviour in bending. (2007).
[1] Elavenil, S. & Knight, G. M. S. Impact Response of
Plates Under Drop Weight Impact Testing. Daffodil
Funding body
International University Journal of Science and
Technology 7, 1–11 (2012). EPSRC
[2] Millard, S. G., Barnett, S. J., Tyas, a. & Schleyer, G. K. Further information
Briefing: Blast tests of fibre-reinforced concrete panels.
Louise McCluskey (E: lmccluskey04@qub.ac.uk) or
Proceedings of the ICE - Construction Materials 163,
Prof Marios Soutsos (E: m.soutsos@qub.ac.uk) and
127–129 (2010).
Dr James Lim (E: j.lim@qub.ac.uk)
[3] Habel, K. & Gauvreau, P. Response of ultra-high
performance fiber reinforced concrete (UHPFRC) to
impact and static loading. Cement and Concrete
Composites 30, 938–946 (2008).
Crowd synchronisation on grandstand structures
Madison McDonald
Objectives To overcome this problem Thorton-Trump et al [2]
developed a novel approach using motion capture
Dynamic loading due to human movement is not well
systems, previously used for biomedical capacities to
understood, the most extreme of these cases can be seen
calculate the GRF of a subject. This work was then
within stadia where large crowds assemble. At stadium
furthered and introduced to the field of structural
events individuals within a crowd will often synchronise
engineering by Racic, et al [3]. This ‘free field’ method uses
their movements, this is expatiated by a beat or music and
markers placed on anatomical landmarks of the human
potentially by visual triggers.
body to capture the movement of each body segment
using the VICON system of cameras. The GRF applied to
There is a lack of reliable information regarding the degree
the structure is obtained by subtracting the person’s weight
of synchronisation seen within crowds, leading to difficulties
from the product of the mass and acceleration experienced
in predicting and evaluating structural vibrations during the
by the person’s centre of mass as they move.
stadium design phase.
As the movement of the centre of mass cannot be
The aim of this project is to investigate the dynamic forces
directly measured indirect measurement is used. This
induced by humans during different activities (e.g. jumping,
is achieved by dividing the body into segments and
bouncing and stamping), and to quantify the degree
using a mass distribution model to take account of the
of synchronisation seen within a crowd in response to
displacement of each segment and the effect on the CoM
different audio and visual stimuli.
[4]. By differentiating the displacements of the markers
collected by the cameras twice, the acceleration of the
To achieve this aim, a motion capture system VICON
body segments is found and the GRF can be calculated.
within the Gait Lab at the University of Warwick will be
The GRF is equal to the sum of all the segment masses
used to quantify the movements of individuals and the
multiplied by the segment accelerations.
corresponding ground reaction force (GRF). In addition
to this, the levels of synchronisation between groups of
The above method has worked successfully to calculate
individuals will be measured and quantified.
the GRF when multiple markers have been used on
each subject [3]. Within a crowded environment the
This work is divided in two objectives:
liberal use of markers is not suitable, as subjects may
obscure the camera’s view of each other hiding markers
(1) To develop a motion capture method suitable for the
and movement. The motion capture method requires
observation of kinematics and GRFs for both individuals
adaptation for application in group situations. The
and crowds. The key challenge is to identify minimum
adaptation of this method is the first development made
number and locations of monitoring points on the
within this project.
human body that are required for reliable calculation of
the GRF.
17 prospective marker locations which had the potential to
describe the movement of body’s CoM mass were chosen
(2) To use the methodology developed in the first objective
on the front, hips and back (Figure 1). Eight subjects took
to track movements of individuals and crowds in a
part in these experiments and were instructed to jump at
range of different conditions and quantify the total GRF
different frequencies between 1-3Hz using a metronome
and the achieved level of synchronisation between
beat. To verify that an accurate force profile was produced
people in crowds. Successful completion of this
from the cameras the trials took place upon a force plate;
objective will increase our understanding of dynamic
allowing the two force profiles to be compared. The
loads allowing for less conservative, more controllable
displacements of the markers were recorded at 200Hz by
and cost effective stadium designs which can safely
the VICON cameras; while the GRF profile was sampled at
support the expected crowd and their actions.
a rate of 1000Hz using an AMTI OR6-7force plate.
Method
The trajectories of each marker were used to indirectly
Quantifying the dynamic load produced by people jumping, calculate the generated force. The accuracy of this force
bouncing or stamping on a structure is challenging. Most was then evaluated against the benchmark GRF acquired
often accelerometers are used to measure the structure’s directly using the force plate.
response [1], but relating this to the applied force is difficult
as the level of crowd participation and synchronisation is
unknown.
It was found that the marker at the top of the back on the
7th Cervical vertebra consistently between all subjects and
frequencies preformed the best, producing an excellent
agreement with the indirectly computed GRF when
compared with the force plate results (Figure 2). Therefore,
Objective 1 has been successfully completed in the first
year of the project, allowing the researcher to proceed with
Objective 2.
For the second objective, further experiments using a single

marker per subject in groups of 2, 4 and 8 individuals
are under way to observe the levels of synchronisation
between members of the group presented with different
audio and visual stimuli.
References
[1] B.R. Ellis, T. Ji, The response of structures to dynamic
crowd loads, BRE Digest 426, 1997.
[2] A.B. Thorton-Trump, R. Daher, The prediction of

reaction forces from gait data, Journal of Biomechanics
8, 1975, pp. 173–178.
[3] V. Racic, J.M.W. Brownjohn, A. Pavic, Reproduction

and application of human bouncing and jumping forces Fig 1: The front marker positions
from visual marker data, Journal of Sound and Vibration
Vol. 329, 2010, pp. 3397-3416.
[4] P. De Leva, Adjustments to Zatsiorsky-Seluyanov’s

Segment Inertia Parameters, J. Biomechanics, Vol. 29
(9), 1996, pp. 1223-1230.
Funding body
EPSRC
Further information
Madison McDonald
(E: Madison.McDonald@Warwick.ac.uk) or
Dr Stana Zivanovic (E: s.zivanovic@warwick.ac.uk)
Fig 2: The time history graph of Subject 1 jumping at

2Hz produced by marker B6, comparing the indirectly
measured force using the motion capture system and the
directly measured force collected by the force plate
Fire performance FRP reinforced concrete: bond
deterioration at elevated temperature
Emma McIntyre
The constantly evolving nature of the built environment Objective 2
Poster presenter
requires a clear evaluation of novel materials which are
To carry out a thorough investigation evaluating Tg for both
used in the design and construction of modern buildings.
glass and carbon FRP bars, and assessing the tensile
The introduction of fibre reinforced polymer (FRP) bars
behaviour and physical changes of the bars at elevated
into concrete construction raises a number of structural
temperatures, up to and exceeding Tg. This data can be
benefits, including resistance to corrosion and optimisation
used as an input parameter for test methodologies of
of slab thickness. Despite these benefits, the polymer
concrete pullout specimens with embedded FRP bars.
resins used as the FRPs’ matrices are problematic at
high temperatures due to their mechanical degradation
Objective 3
at temperatures in the range of the resins’ glass transition
temperature (Tg). Research has shown that there is good To understand the variation in bond properties at elevated
correlation between material degradation and a decrease temperatures of FRP bars embedded in concrete. In
in the bond strength and stiffness of FRP bars. At elevated comparison to the high thermo-mechanical sensitivity
temperatures there is differential expansion between the of FRP, understanding of bond properties is not well
FRP and the concrete reducing the bond capacity and documented. The change in bond strength will provide
exposing the fibres. Furthermore the reductions in the bond insight into the behaviour of FRP reinforced slabs when Tg
strength at these temperatures may cause slip between the is exceeded.
FRP and the concrete if cool anchorage of the bars cannot
be ensured or maintained. Methodology
The investigation begins with the determination of the glass
The current approach to reinforced concrete slab design
transition temperatures of three commercially available FRP
for fire essentially assesses a 50% reduction in the
reinforcing bar products through thermogravimetric analysis
tensile strength of the steel reinforcement as leading
(TGA), differential scanning calorimetry (DSC) and dynamic
to failure, and thus defines the critical temperature of
mechanical analysis (DMA).
the reinforcement using this criterion. In comparison to
steel the composition of FRP includes polymers which
Tension tests will be carried out at elevated temperatures
soften at relatively low temperatures and thus the critical
to assess the physical changes of the FRP, and the effect
temperatures of these materials may be is much lower
of maintaining cold anchorage on the tensile capacity of
and may appear to be inferior. However, it is hypothesised
the bars. A series of sixty-six, 150mm cubes (Fig. 1) with
that by fully understanding the effects of temperature
embedded glass and carbon FRP bars are to be tested
on the tensile and bond strength reductions of FRP
for bond properties under both steady state and transient
reinforcement for concrete and optimising the placement
thermal testing regimes to examine the reduction of bond
and anchorage of the bars, FRP may be a viable and fire
strength with increasing temperature. They have been
safe alternative to steel reinforcement in concrete. This
designed in accordance with CSA S806—12 Annex H. The
will require demonstrating that the criticality of FRP should
temperatures in the steady state testing are dependent on
not be defined by a 50% reduction in tensile strength, but
the glass transition temperature.
rather as a reduction in a structural element’s load bearing
capacity sufficient to cause structural collapse under
Digital image correlation analysis will be used to assess
service loads.
bond slip during testing and to understand the potential
roles of the bond strength during fire, irrespective of the
Aims and objectives
tensile strength of the bars. This will allow the development
This project seeks to evaluate the influence of temperature of bond-slip models at high temperature which can be
on the mechanical and bond properties of FRP bars and used by designers to assure satisfactory performance of
their usefulness as internal reinforcement in concrete slabs FRP reinforced concrete elements in fire.
under realistic fire and loading conditions.
Further work
Objective 1
The assessment of Tg and bond properties is essential in
To undertake a comprehensive literature review of past understanding the performed of FRP reinforced concrete in
and current research into the use of FRP bars as an a whole building context. This project will progress to the
internal reinforcing material in concrete. By investigating heating of eight metre, three span slabs under full service
the experiments that have been carried out it will allow loads, and allowing a determination if the use of FRP bars
an insight into the most appropriate method to further in concrete slabs is realistic with the risk of fire. Rigorous
investigate FRP bars and progress accordingly. testing and analysis will allow for the use of performance
based design to provide reasonable resistance to fire in line
with that of steel to ensure a realistic timeline for building
evacuation and fire fighting activities while maintaining
structural integrity.
References
Bank, L; Puterman, M & Katz, A (1998) The Effect of
Material Degradation on Bond Properties of Fiber-
Reinforced Plastic Reinforcing Bars in Concrete, ACI
Materials Journal 95(3):pp132-143
Bisby, L.A. and Kodur, V.K.R. (2007), Evaluating the

fire endurance of concrete slabs reinforced with FRP
bars: Consideration for a holistic approach. Journal of
Composites, 38 (5-6): 547-558
Bisby, L.A. and Stratford, T.J. (2011), Performance-based

fire-safe design of FRP reinforced concrete slabs: Holistic
design through consideration of real structural response t
to real fires. Proposal for Research, University of Edinburgh.
Nigro, E, Cefarelli, G., Bilotta, A., Manfredi, G. and

Cosenza, E. (2011), Fire resistance of concrete slabs
reinforced with FRP bars: Part A and Part B, Journal of
Composites: Part B 42(6): 1739-1763
Robert, M and Benmokrane, B (2010), Behaviour of GFRP

Reinforcing Bars Subjected to Extreme Temperatures,
Journal of Composites for Construction, 14(4):353-360
Funding body
EPSRC, Pultrall Inc. & BP Composites
Fig 1: Concrete pullout specimen with embedded Glass/
Further information Carbon FRP bar.
Emma Reid-McIntyre (E: emma.mcintyre@ed.ac.uk) or
Dr Luke Bisby (E: luke.bisby@ed.ac.uk)
Behaviour of a novel skew FlexiArch bridge system
Paul McNulty
Queen’s University Belfast 39
Oral presenter
Brief description The models are verified against the results obtained from
an experimental programme, allowing a parametric study
Queen’s University Belfast in collaboration with Macrete
on a wider variety of variables to be conducted, eliminating
Ltd have developed the FlexiArch bridge system, which is
the need for further physical testing.
a “flat pack” precast concrete solution that allows durable
arch bridges to be constructed quickly and easily. Individual
Project objectives:
concrete voussoirs, precast with the correct taper for a
given span and rise, are connected together by a flexible
Conduct an experimental programme to investigate the
polymeric membrane which carries the self-weight of the
behaviour of the skew FlexiArch during lifting and under
arch when it is lifted into position.
loading. The results obtained from the experimental
programme are compared against 3D NLFEA models
This project involves researching the behaviour of the
of the skew FlexiArch system under loading, and an
skew FlexiArch bridge system and builds upon previous
analytical model used to predict the forces in the polymeric
developmental work in the collaboration. Square spanning
membrane during lifting.
arches can be regarded as a structure used to bridge a
crossing at ninety degrees, however, it is common for two
systems to cross at an oblique angle. Square arches may
still be used but the more efficient and sustainable choice is It is the objective of this research project to analyse the
a skew arch, as shown in Figure 1. A square arch spans a behaviour of the skew FlexiArch through the experimental
distance x, with the equivalent skew arch bridge spanning programme detailed in Table 1. To allow a comparison of
a lesser distance y, and is therefore more efficient. the research conducted on a third scale 5m x 2m (span
x rise) square FlexiArch, the three core experimental tests
Skew FlexiArch systems form an S-shape in flat form are of the same arch dimensions but with varying angles of
(Figure 2) and have torsional forces that do not occur in skew; 15o, 30o and 45o.
the equivalent square arch. Laboratory tests at model
scale have been carried out on FlexiArch rings which were A monitoring regime has been established to capture
backfilled with carefully graded backfill. The tests aim to the behaviour of the FlexiArch during testing. Electrical
investigate the behaviour of the FlexiArch with a change resistance strain gauges have been used to monitor
in skew angle. The results are compared against three- the strains in the polymeric membrane. Displacement
dimensional (3D) non-linear finite element analysis (NLFEA) transducers and vibrating wire strain gauges (VWSGs)
models of the arch. have been positioned at selected points on the FlexiArch
concrete ring intrados during loading, with the VWSGs
Project aims & objectives measuring the compressive strains in the voussoirs.
Aims:
The results from the experimental programme have shown
the skew FlexiArch to be a viable structural solution for
• Develop the following models for the skew FlexiArch
medium span bridges, with strengths far in excess of the
system:
equivalent maximum axle load from the design manual
for roads and bridges (The Highways Agency, 2004). The
• A 3D NLFEA model to accurately represent the
failure modes of the arches varied between low and high
behaviour of the skew arch under loading
skews. The structural behaviour and failure load of the 3D
NLFEA compared well with the experimental tests.
An analytical model to predict the forces in the polymeric
membrane for the arch during lifting
Table 1
Experimental programme
Experiment Name Variable

Angle of Skew Span to Rise Square Span Skew Span (m) Rise (m)
(degrees) Ratio (m)
ARCH:1/3(5*2):15o 15 2.5 1.67 1.73 0.67
ARCH:1/3(5*2):30o 30 2.5 1.67 1.93 0.67
ARCH:1/3(5*2):45o 45 2.5 1.67 2.36 0.67
There are approximately 73,000 masonry arch bridges in Harvey, W. J. (2006). Some problems with arch bridge
the UK, many of them forming an important part of the assessment and potential solutions. The Structural
UK road and rail network, equating to approximately 40% Engineer, 84(3), 45-50.
of the total used within the UK (Page, 1993). In recent
years the repair and maintenance of bridge structures Page, J. (1993). State of the art review - masonry arch
has become a major concern, with the UK Highways bridges (First ed.). London, UK: HMSO.
Agency (2001) recommending the use of the arch form and
also stating “consideration shall be given to all possible The Highways Agency. (2001). Design manual for roads
means of reducing or eliminating the use of corrodible and bridges, volume 1, section 3, part 7, BD 57/01, design
reinforcement”. for durability. London, UK: The Highways Agency.
The research associated with this project is ground The Highways Agency. (2004). Design manual for roads
breaking given that no method currently exists to analyse and bridges, volume 2, section 2, part 14, BD 91/04,
the behaviour of skew arches. Harvey (2006) raised serious unreinforced masonry arch bridges. London, UK: The
concerns about the use of two-dimensional (2D) analytical Highways Agency.
methods to model arch behaviour. A 2D analysis represents
a 3D structure by adding rules to define the effects of the Funding body
third dimension. Harvey states that these rules contain
Department of Employment and Learning (DEL)
errors which cause the capacity of small span bridges to
be grossly over-estimated. A 3D analysis method, such as
Further information
the one proposed in this research project, will give more
accurate results, for both square and skew arches. Paul Mc Nulty (E: pmcnulty06@qub.ac.uk) or Dr Susan
Taylor (E: s.e.taylor@qub.ac.uk)
Fig 1: Advantage of the skew arch – plan view
Fig 2: Lifting of 45o skew FlexiArch
An innovative design tool for optimisation of timber
structures
Jesus M Menendez
Project objectives and goals The unified UK lateral racking resistance method is
dependent of the wall length (L) and geometry (Ki,w), the
The purpose of this research project is to provide structural
shear capacity and spacing of the sheathing fasteners
engineers with a user-friendly tool for the design of timber
(fp,d,t), the existence of openings (Kopening) and the sole plate
structures. This application also facilitates real-time
fixing detail (Ki,w and fi,sp,d). Findings from internal research
structural optimisation by updating the latest research
work [5] have contributed to the British Standard Institution
findings on innovative products and design methodologies.
(BSI) committee in the development of a new racking EC5
The final aim is not only to improve the use of timber in
design method [6]. Equation 1 shows the expression used
construction but also to increase the value of home-grown
to calculate the racking resistance of a shear wall, i:
timber.

Introduction
The UK has a housing deficit, requiring 3 million new
homes by 2016 that will have to comply with building
regulation requirements. The economic down-turn (1)
has exacerbated this problem and as a result, the
government has encouraged the use of modern methods Providing the required uplift resistance is held by the sole
of construction (MMC) to produce a higher quantity and plate detail, the optimal structural solution is given by
quality of houses [1]. The inherent sustainable credentials, equation 2, where С is a shape dependent constant:
speed of construction and off-site lean manufacturing
possibilities make timber a suitable construction material
for MMC [2].
(2)
Frequently, the design of timber structures is over specified
with an inefficient use of materials. Also, the UK timber Laboratory testing of key structural components in
sector is defined as a fragmented industry heavily reliant combination and in isolation are being conducted at
on imports due to the low quality of the wood harvested Edinburgh Napier University and at a United Kingdom
in the British forests. The transformation of this wood into Accreditation Service (UKAS) external centre in order to
engineered wood products (EWP) can add value to the fully understand the behaviour of every element of the
forest and furthermore, optimise the structure. panel and to verify the EC5 racking design method (Fig 1).
Structural timber design has been marginalised relative This UK unified method has been developed into a Tedds
to steel and concrete design. Furthermore, the current application which calculates the racking resistance of a
design code for timber structures, Eurocode 5 (EC5), is shear wall in real-time. The interface is also linked to the
an analytical approach which requires additional design Tedds material database increasing the resistance values.
considerations. However, EC5 allows for a more informed An additional advantage of this application is the capability
design, facilitating properly applied innovation and to inform the user of the methodology employed, the
optimisation. Spread sheets are still generally used for factors applied or the variables considered in the design
engineering calculations due to the lack of affordable timber via explanatory notes, active drawings and captions. The
design software. Nevertheless, concerns about the integrity application follows a transparent approach of descriptive
quality control, transparency and partial check of spread calculation sections controlled by a process flow chart. The
sheets have been identified [3]. Other available resources result is a user-friendly tool that assists engineers in the
are currently supplied by manufacturers of EWP. However, design of shear walls according to the latest design code
these “black-box” tools do not provide design flexibility (Fig 2).
as they are limited to the products of the manufacturer.
This restricts engineering judgement and commoditises The validation of this application has been carried out
the design process. Hence, there is a perceptible need by comparing the results of twenty different wall panels
for transparent timber design tools with integrated flexible with PTC Mathcad® and MS Excel. The resistance values
databases [4]. obtained in the Tedds application were, for all of the
wall panels, within the satisfactory deviation limit of ±2
Description of methods and results percentage.
Tedds® for Word (Tedds), from CSC (UK) Ltd, was selected
as a programming platform due to its compatibility with
Microsoft® Office, wide distribution and recognised
experience in structural design. Although the project will
include several timber design applications, this paper only
describes the methodology and results for the design of
timber shear walls.
Potential for application of results Funding body
A comprehensive parametric analysis of more than two EPSRC
hundred wall panel configurations will be undertaken by the
Tedds racking application in order to investigate the effect Further information
of the different design variables in the lateral capacity of
Jesus M Menendez (E: j.menendez@napier.ac.uk) or
shear walls. These analytical results will be also compared
Dr Robert Hairstans (E: r.hairstans@napier.ac.uk) and
with the outcome from standard racking tests on walls to
Kenneth Leitch (E: k.leitch@napier.ac.uk)
evaluate the existing racking design methodology.
The developed Tedds racking application, once externally

certified, will be embedded into the proprietary Tedds
library of calculations. As a result, the application will be
distributed to over 14,000 engineers in the UK who are
currently using this software. Further timber engineering
calculations will be added to the library in subsequent
phases.
Finally, a centralised Tedds-compatible database of timber

products is being created. This database contains technical
data from laboratory UKAS testing and other information
such as construction details, thermal performance and
environmental facts. The ultimate aim is to create a building
information modelling (BIM) enabled database for future
use.
References
[1] Hairstans R. Off-Site and Modern Methods of Timber
Fig 1: Racking wall panel in UKAS test
Construction: A Sustainable Approach: TRADA
Technology Limited 2010.
[2] Egan J. Rethinking construction. London: Department

of the Environment, Transport and the Regions; 1998.
[3] IStructE. Guidelines for the use of computers for

engineering calculations. London: The Institution of
Structural Engineers 2002.
[4] Osterrieder P, Richter S, Fischer M. A product data

model for design and fabrication of timber buildings.
2004; Lahti: WCTE 2004 Proceedings; 2004.
[5] Leitch K, Hairstans R. The development of a hybrid

racking panel – appraisal of methods for strength and
stiffness calculation. World Conference on Timber
Engineering; 2010; Trentino, Italy; 2010.
[6] BSI. PD6693 Complementary Information to Eurocode

5 Design of timber structures Part 1 General Common
rules and rules for buildings. London: British Standard
Institution 2012:66.
Fig 2: Representation of the racking design tool procedure
Under-utilisation of steel beams in construction
Muiris Moynihan
University of Cambridge 41
Oral presenter
Project objectives and goals Across all projects, it is found that average utilisation by
piece is 40%, ranging 18%-64%. When averaged instead
A review of published literature reveals that, while
by mass this increases to 52%, ranging 37-68%. Figure
engineers’ understanding of steel behaviour is continually
1 shows an example of the frequency distribution of
improving (Beal, 2011), modern design practice is to
utilisation ratios for a building. The results of mapping,
provide additional steel – above that which is necessary
an example of which is shown in figure 2, show that
for adequate structural performance – in order to reduce
secondary beams and beams along a perimeter are most
labour costs during fabrication and construction (Gibbons,
likely to have low utilisation ratios. These are also the areas
1995). This practice, referred to as ‘rationalisation’, aims
where many beams commonly have the same section size.
to output a lowest-cost solution, which Needham (1971)
The standard beam section catalogue was analysed to
states is rarely a lowest-weight design. Nowhere in the
reveal what ‘step’ in capacity, hence ‘enforced’ reduction
literature, however, is it assessed as to how much extra
in utilisation, is present between consecutive beam sizes.
steel is being used in typical buildings.
It was found that a section size typically has 80% of the
capacity of the one above it; however for the lightest
Approximately one-third of steel produced globally in one
sections the drop in capacity is larger.
year is used in the construction of buildings (Moynihan &
Allwood, 2012). Making this steel causes carbon dioxide
Undoubtedly there is uncertainty attached to the results,
emissions, which IEA (2008) predict will double by 2050
two sources of which are identified: incorrect data, and
as demand for steel rises. Climate change experts
excluded design criteria. The data are taken from computer
recommend emissions be halved in the next 38 years to
models of the buildings, in every case the ‘most final’
avoid the worst consequences of global warming (Metz et
model available is used even though some contained
al., 2007). Allwood et al. (2010) show that remaining steel
failing beams (i.e. with utilisation > 1) – such beams
production technology improvements cannot deliver the
are omitted from the analysis as it is not known what
required reduction. Smil (2011) outlines the considerable
their redesigned state is. The design criteria used in the
logistical challenges to installing sufficient carbon capture
analysis encompassed strength, stability and deflection
and storage or renewable energy networks within the time
parameters. To account for excluded design criteria,
frame. Instead Allwood et al. (2012) propose ‘material
interviews are held with the lead designers of each project,
efficiency’ - providing services with less production of new
ascertaining additional governing criteria (e.g. maximum
material – as an option to achieve this reduction. They
depth allowances or floorplate vibration) to help interpret
suggest six material efficiency strategies, one of which is
the results.
to design and build using less steel to deliver the same
structural functions, so reducing steel production and
hence carbon emissions.
The surprising results indicate that most steel beams in
Description of method and results buildings are not highly utilised – in fact buildings contain
almost double the amount of steel necessary. This implies
To estimate how much steel could be saved by building
that substantially less steel could be used in buildings,
with less steel, 20 steel-framed typical building designs
causing reduced steel production and hence decreased
are studied (comprising over 10,000 steel beams in total),
carbon dioxide emissions. This material reduction could
sourced from leading UK engineering design firms. A
occur without any loss in service to the end-user.
‘utilisation ratio’ is defined as the actual performance (e.g.
deflection) divided by the allowable performance (e.g.
The mapping results suggest that rationalisation is the
deflection limit). Utilisation ratios are calculated for the
cause of low utilisation ratios, however producing lower-
worst-case locations along each beam, and the highest
mass steel designs may require additional design and
utilisation ratio across 16 design criteria is selected as
construction time and cost. This may be compensated for
the beam’s governing utilisation ratio. For each building a
by: steel tonnage savings; the client, persuaded to accept
graph is plotted showing the frequency of decile bands of
the added cost by the environmental benefit (an average
utilisation ratio, marking average utilisations – calculated
utilisation could be specified in the brief); advances in
both by piece and by weight. Floorplates within each
design and construction technology (including fabrication),
building are mapped to identify areas of high and low
which might allow optimised designs to be built at
utilisation, with results compared within and between
comparable cost to rationalised ones. There may also be
projects to determine recurring patterns. Similar maps
scope for manufacturers of steel beams to extend their
are plotted showing the different section sizes used per
catalogues of light sections to prevent trimming steel being
floorplate, and also analysed for patterns.
overdesigned.
References
Allwood, JM, Cullen, JM, Milford, RL, 2010. Options for
Achieving a 50% Cut in an Industrial Carbon Emissions by
2050. Environmental science and Technology, 44(6), 1888-
1894.
Allwood, JM, Cullen, JM, Carruth, MA, Cooper, DR,

McBrien, M, Milford, RL, Moynihan, MC, Patel, ACH, 2012.
Sustainable Materials: with both eyes open. Cambridge,
UK: UIT Press.
Beal AN, 2011. A history of the safety factors. The

Structural Engineer, 89(October).
International Energy Agency, 2008. Energy Technology

Perspectives – Scenarios & Strategies to 2050. Paris,
France: IEA Publications.
Metz, B, Davidson, OR, Bosch, PR, Dave, R, Meyer, LA,

Fig 1: Graph of governing utilisation ratio against frequency
2007. Climate Change 2007: Mitigation: Contribution of
of occurrence for beams in a building
Working Group III to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change. Cambridge,
UK: Cambridge University Press.
Moynihan, MC, Allwood, JM, 2012. The flow of steel

into the construction sector. Resources, Conservation &
Recycling, 68. pp. 88-95
Needham, FH, 1971. The economics of steelwork design.

The Structural Engineer, 55(9):367-373.
Smil V, 2010. Energy myths and realities: bringing science

to the energy policy debate. Washington, DC, USA: AEI
Press.
Funding body
CASE Studentship from EPSRC and Arup
Further information
Muiris Moynihan (E: mm521@cam.ac.uk) or
Dr Julian Allwood (E: jma42@cam.ac.uk)
Fig 2: Plan view of floorplate with beams coloured to indicate

utilisation ratio
Concrete failure: maximum tension strain as the controlling
parameter
Hussam Nasreddin
University College London 42
Poster presenter
Project description and objectives In 1978, Lowe defined concrete behaviour under loading
as a linear elastic material up to a critical value of principal
The current state of concrete design practice relies on
tension strain, εt. The equations that define maximum
the stress strain relation obtained from a standard cube
extension strain are those of Hooke’s law. Lowe plotted
or cylinder unconfined compression test. This is quite
the strength envelop for different values of Poisson’s ratio
satisfactory when describing structural performance under
in stress space where the horizontal axis represents equal
service loading and in ultimate strength assessment, but far
circumferential pressure applied on a cylinder, (σ2=σ3).
from it when structural ductility is under scrutiny. Modelling
The vertical axis is the axial stress σ1 and both axes are
the nonlinear behaviour after the ultimate strength is
normalised to the uniaxial compressive strength. Figure
met has been a difficult task. Currently, there are several
1 shows the plot with tension being positive. Some
models that engineers could use to estimate the strength
experimental results by Hobbs et al (1977), Richart et al
degradation of concrete structures during the collapse
(1928) and Theocaris and Prassianikis (1974) are also
stage. Most of these are based on plasticity theory, which
included in the figure showing a good agreement with
was originally developed for metals, which requires some
most of the theoretically based model. Contrary to the
assumptions about the behaviour that concrete does not
theoretical concept of the theory, Lowe suggested that
obey.
two cut-off lines extending from the tension-compression
zone to the pure tension zone exist. This suggestion was
The stress strain graph of a compression test on a
entirely grounded on the triaxial tension-compression test
concrete specimen is often misinterpreted. The slope of the
results obtained by Theocaris and Prassianakis and no
post ultimate descending branch of the curve is dependent
theoretical explanation was provided for it nor to how these
on the specimen design and the testing technique and
lines translate to planes in the three dimensional surface.
thus, must be considered as a structural rather than a
In the current research work, these results are questioned
material response. The standard cube or cylinder uniaxial
since applying tension to a concrete specimen could incur
compression test result, for instance, is not representative
secondary effects that could change the state of stress
of the material behaviour when other structural members
developed in the specimen.
such as beams or columns are in question. A clear
differentiation between structural and material performance
In light of what has been stated above, this research aims
is important for the understanding of the observations
mainly to investigate thoroughly maximum tension strain
on concrete failure. In addition, it is unfortunate that
failure criterion on a theoretical and rational basis. As
shear failure is often referred to when concrete structures
part of the main objective, a triaxial tension-compression
collapse under dominantly compressive forces. An example
test is being conducted at University College London
of shearing behaviour is rather apparent in ductile metals
to investigate the results obtained by Theocaris and
in which the material crystals slide against each other
Prassianakis. The nonexistence of the cut-off lines would
appearing to flow to form new structural geometries with
allow for a strong case to move forward with the potential
the material continually intact up to fracture. It is important
applications of the theoretical model.
to mention that this kind of shearing has never been
observed in concrete structures.
Test method
The onset of the failure process in any concrete structure Figure 2 shows a photo of the set up for a triaxial
is cracking which demonstrates that the material is failing tension-compression test. 38 x 70 mm cylindrical mortar
in tension. According to Neville (2011, p.299) Mather specimens are subjected to equal lateral pressure and axial
argued that; ideally, a failure criterion of a single stress tension. The lateral pressure is applied using hydraulic fluid
parameter could be sufficient to express concrete failure by means of a Hoek cell. On the other hand, axial tension is
under all possible stress states. Such a stress parameter applied by a 5kN INSTRON machine. Mortar samples are
could be thought of as the uniaxial tensile strength. Much bonded to steel end caps using an epoxy resin adhesive.
of the evidence in the relevant literature suggests, however,
that it’s a tension strain rather than a tension stress that Potential applications of the research outcomes
dictates failure. Kaplan (1963) found that the value of the
Since concrete is considered an elastic material in
maximum extension strain that initiated cracking not only
this work, geometric nonlinearity will be explored and
was similar in uniaxial compression, flexure, direct tension
incorporated to develop a computational tool to enable
and splitting tests, but was also independent of two types
modelling of structural ductility. Such advancement would
of aggregates and two water cement ratios. Many other
lead to better procedural and design decisions particularly
researchers support that a single tension strain criterion
for concrete structures that are subject to environmental
seems plausible.
cyclic damage loadings such as earthquakes and
tsunamis.
Hobbs, D.W., Pomeroy, C.D., & Newman, J.B., 1977. Hussam Nasreddin (E: uceshsn@ucl.ac.uk) or
Design stresses for concrete structures subject to multi- Dr John R Eyre (E: j.eyre@ucl.ac.uk)
axial stresses. Structural Engineer, 55 (4), pp.151-164.
Kaplan, M.F., 1963. Strains and Stresses of Concrete at

Initiation of Cracking and Near Failure. ACI Proceedings, 60
(7), pp.853-880.
Lowe, P.G., 1978. Deformation and Fracture of Plain

Concrete. Magazine of Concrete Research, 30 (105),
pp.200-204.
Neville, A.M., 2011. Properties of concrete. 5th ed.

LONGMAN.
Richart, F.E., Brandtzaeg, A., & Brown, R.L., 1928. A study

of the failure of concrete under combined compressive
stresses. University of Illinois Bulletin, 26 (12).
Theocaris, P.S. & Prassianakis, J.N., 1974. Mohr envelope

of failure for concrete: a study of its tension-compression
part. Magazine of Concrete Research, 26 (87), pp.73-82.
Fig 2: Triaxial compression-tension test arrangement at
University College London
Fig 1: A plot of the maximum tension strain criterion and experimental data
Design framework for pultruded FRP members in bending:
a proposal
Tien Thuy Nguyen

Project objectives and goals the test configuration (Fig.2) simply supported beams are
subjected to three-point bending. It also allows the restraint
This study combines advanced computational simulations
for lateral flexure at the beam ends to be fully fixed or free.
(by ABAQUS) and experimental work to obtain research
The concentrated load is applied through wire wrapped
results that can be used to prepare, for a future FRP
around a steel disc fixture at beam mid-span. By varying
Eurocode, provisions towards the design of beams against
the disc’s height position, the vertical load can be made
failure by Lateral-Torsional Buckling (LTB).
to act at the level of the beams’ top flange, shear centre
or at its bottom flange. For a buckling load up to 150 kg
Brief description
dead weight are applied as this ensures the LTB response
Shapes and systems of Fibre Reinforced Polymer is fully unrestrained; previous test series have often had
(FRP) made by the pultrusion composite process have a deficiency with satisfying this test condition. Buckling
applications in civil engineering works, working alongside resistance is determined from measurements using the
construction components of conventional materials following two data reduction methods:
such as steel, concrete and aluminium. Exploitation of
Pultruded FRP shapes in structural engineering requires (1) Southwell plot method when the load-vertical deflection
structural engineers to have reliable design guidelines relationship does not possess a clear buckling point, or
with closed-form equations for the thin-walled sections
that practitioners will routinely use. To facilitate the wider (2) Peak load method when the load-vertical deflection
application of standard PFRP shapes, mimicking the relationship has a bifurcation-like form.
standard shapes we are all familiar with in steelwork,
requires the preparation of structural design guidance that Measured buckling resistances are compared against
has been verified using resistance results from physical analytical predictions using a closed form equation that
testing. has been modified from how it is given in Eurocode 3
for steel beams of the same cross-section shape. The
Owing to the high strength-to-stiffness ratio of PFRP comparison shows that the semi-empirical equation is
material, the design of beam members (in braced frames very good at predicting LTB resistance when beams are of
of simple construction) is normally governed by elastic pultruded FRP material. A framework for design provisions
deflections and/or elastic buckling instabilities and rarely for pultruded FRP beams failing by LTB can therefore be
by material strength limitations. In other words, to execute prepared.
the Ultimate Limit State (ULS) design approach the different
instability modes must be quantified by a combination of Further information
rigorous numerical analysis and physical testing.
Tien Thuy Nguyen (E: T.Nguyen-Tien@Warwick.ac.uk) or
Associate Professor Tak-Ming Chan
Lateral-Torsional Buckling (LTB) is a commonly observed
(E: T.M.Chan@warwick.ac.uk) and
type of global instability in laterally unrestrained beams
Prof J. Toby Mottram (E: J.T.Mottram@warwick.ac.uk)
of open thin-walled sections (e.g. I, H, channel) that are
subjected to flexure about their major axis. The beam loses
stability in the LTB mode when the member bends laterally
and twists along its length, without any cross-section
distortion. Because a beam’s moment of resistance to LTB
is influenced by having relatively small lateral (minor-axis)
flexure and torsional stiffness, it is observed that PFRP
shapes are susceptible to this failure mode governing in
ULS design. The presenter’s study combines advanced
computational simulations (by ABAQUS) and experimental
work to obtain new information to develop calibrated close
form formulae for design against LTB. In the non-linear
Finite Element (FE) analyses (Fig.1) imperfections, including
out-of-straightness and twist, have been included into the
model to simulate reality as closely as is feasible. A new
data reduction method has been proposed to approximate
the buckling resistance of beams when the load-deflection
curve does not show a distinct (bifurcation) buckling point.
It is shown that the developed FE modelling methodology
will enable parametric studies to be performed for the
development of closed form formulae that will be reliable.
96 individual LTB resistance tests have been conducted on
four different PFRP shapes at four different beam spans. In
Fig 1: Pultruded FRP beam from unload to non-linear failure
Fig 2: Testing configuration allows for various boundary and load conditions
The dynamic effects of prestress loss on precast, post-
tensioned concrete towers
Darragh Noble
Trinity College Dublin 44
Project objectives and goals Some research has been conducted into the effect that
the magnitude of the prestressing force has on the natural
Precast post-tensioned concrete tower structures are
frequencies and modal properties of prestressed concrete
most notably used as support structures for both onshore
structures. Following a thorough literature review, it is
and offshore wind turbines, but can also be used as
clear that significant disagreement has arisen between
support structures for water towers and for the power
researchers in the field. Many researchers such as Tse,
transmission grid. They have become increasingly popular
Morse et al. (1969), Chan and Yung (2000), Dall’Asta
as the support structures for wind turbine towers for a
and Leoni (1999), Raju and Rao (1986), and Breccolotti,
variety of different reasons. Firstly, the current trend in
Ubertini et al. (n.d.) all argue that the natural frequency of
the wind energy industry is to have larger, more powerful
prestressed concrete beams decreases as the magnitude
units, stretching higher into the atmosphere to make use
of the prestress force is increased. This is known as
of greater and more stable wind speeds. This trend is only
the “compression softening” argument and was first
expected to increase as the industry strives to produce the
put forward by Tse, Morse et al. (1969). Others such as
first 10 MW installation and beyond. Traditionally, tubular
Saiidi, Douglas et al. (1994), Jain and Goel (1996), Kim,
steel towers had been used as support structures for
Yun et al. (2004), Hop (1991) and Zhang and Li (2007)
wind turbine towers up to approximately 90 m in height.
argue, mainly through dynamic testing, that the natural
However, beyond 100 m in height, there are serious
frequency increases with increasing prestress force. This
manufacturing and transportation difficulties associated
is based on the “cracked model” for prestressed concrete.
with the traditional rolled steel solution.
Furthermore, others such as Hamed and Frostig (2006),
Dall’Asta and Dezi (1996), Deák (1996), Kerr (1976),
As a result, concrete towers have come to the fore as an
Bartlett (1987) and Dai and Chen (2007) all argue that
alternative solution to the issues encountered with the
the magnitude of the prestress force has no effect on the
traditional rolled steel sections. Prestressed concrete, in
modal frequencies.
particular, has a high fatigue resistance, providing more
tolerance and less risk from dynamic failure. Precast
Description of methods & results
concrete segments have advantages over in-situ concrete
in that they can be manufactured and cured in a controlled The project aims to fully investigate this multi-layered
factory environment, ensuring the required durability and problem to determine the effect that the magnitude of
strength of the concrete. Furthermore, they allow for easy the prestressing force has on the modal properties of
transportation to site, with smaller, easier to manage post-tensioned, precast concrete towers. A mathematical
sections. They also allow for quick and easy erection, model will be formulated and some dynamic testing will
maximising construction time and minimising the amount be conducted followed by finite element modelling to
of labour required on site. Ultimately, this is economically determine the effect that temporal prestress loss has on
beneficial in terms of the capital cost of manufacturing and the modal frequencies. Following a full investigation into
erecting the tower. Precast construction also allows for thin the effect that the magnitude of the prestress force has
walled shell units to be manufactured, which is beneficial on modal frequencies, a non-linear time history analysis
as it keeps the self-weight of the structure to a minimum. will be conducted, including the combined effects of gain
Not only is this beneficial for the dynamic performance of in stiffness due to crack closing and the compression
the tower, it is also important to minimise cost of erection softening effect.
when cranage is required. Post-tensioned precast concrete
towers are becoming more and more prevalent in the
industry due to their ease of fabrication, transportation and
erection. Post-tensioning the precast tower enable the
tower to act monolithically.
Due to the great height of these structures (> 100 m), they
are subjected to significant dynamic loads in the form
of wind loading. If they are located offshore (i.e. offshore
wind turbines) then they are also subjected to significant
hydrodynamic loads in the form of wave and current
loading. Furthermore, these structures may also be located
in zones with a significant seismic history and risk, such
as the USA, Japan or China to name but a few. They also
must be designed to survive extreme earthquake events.
As a result, accurate calculation of the dynamic properties
of these post-tensioned precast concrete towers is of
significant importance in design to ensure that the natural
frequency of the tower structure does not coincide with the
frequency of dynamic excitation.
The potential for application of the results of this project Contact Darragh Noble for details of above references.
are vast. It will help to better understand the dynamic
behaviour of post-tensioned precast concrete towers and Funding body
in the efficiency of their design. It also will act to add to the
Irish Research Council (IRC)
current body of knowledge and help to alleviate the clear
confusion that has arisen in the subject. Finally, it will allow
Further information
structural engineers to accurately compare the dynamic
behaviour of post-tensioned, precast concrete towers with
Darragh Noble (E: nobleda@tcd.ie) or
their cast in situ concrete, steel, lattice and even timber
Dr Alan O’Connor (E: alan.oconnor@tcd.ie)
counterparts.
Reinforced concrete deterioration of a 100 year old
structure in a marine environment
Ross O’Donovan
Cork Institute of Technology 45
Project objectives and goals used in the bridge would have been in the order of 18N/
mm². The compressive test results indicated that the
The current study investigates the deterioration of a 100
compressive strengths had doubled over the 100 year
year old bridge located in a harsh marine environment.
period.
Constructed in 1908 – 1909, the Mizen Head Bridge (Fig
1) was one of the oldest reinforced concrete bridges in
Anaerobic corrosion is a rare form of corrosion that occurs
Ireland. The demolition of the bridge in 2009 provided a
when the oxygen supply is limited at the active anode. The
unique opportunity to examine its service life and to extract
literature review indicated that anaerobic corrosion may
concrete samples for analysis and investigation. This
result through the use of concrete protective membranes.
research may provide valuable advice and guidance for the
Anaerobic corrosion may also occur during an advanced
asset management of similar structures.
stage of the corrosion process, as the rate of oxygen
diffusion to the corroding bars through the rust layer is so
low that anaerobic corrosion activity occurs. Both of these
The construction of the bridge consisted of precast, conditions were present on the Mizen bridge.
composite and in-situ concrete. Noel Ridley, the bridge
designer, incorporated his patented Ridley-Cammel Potential for application of results
corrugated dovetail sheeting in the composite members.
A paper “Reinforced concrete deterioration of a 100 year
The service life, maintenance regime and data collected
old structure in a marine environment” has been presented
from previous bridge inspections were examined.
at the Bridge and Concrete Research in Ireland Conference
in 2012. Another paper “Anaerobic Corrosion” is due to be
In 1972 following visible deterioration in the reinforced
presented at the 10th International Conference on Damage
concrete, significant repair work was undertaken on the
Assessment of Structures in July 2013. Ultimately this
bridge. The repair work consisted of concrete patch repairs
research may provide valuable advice and guidance for the
to defective areas, followed by a coating in a bituminous
asset management of similar structures.
product. The purpose of the bituminous coating seems to
have been an attempt to stop further chloride ingress into
References
the concrete. A later bridge inspection in 2002 identified
the presence of a rare form of corrosion (anaerobic Contact Ross O’Donovan for details of above references.
corrosion) in the bridge.
Funding body
During the demolition of the bridge, reinforced concrete
National Road Authority (Ireland), RPS Group, The
samples were selected and extracted from the bridge
Institution of Structural Engineers
members. The samples were labelled according to the
bridge member, position and orientation. A suite of
Further information
concrete tests was then carried out on the concrete
samples. Ross O’Donovan (E:ross.odonovan@rpsgroup.com)
or Kieran Ruane (E:Kieran.ruane@rpsgroup.com)
Concrete compressive tests indicated that compressive
strengths were generally lower in flexural and tension
members than in compression members.
High chloride levels (in the order of 2.0% by weight of

cement) were found in the concrete. Normally, higher
concentrations of chloride concentration would be
expected at the external surface and chloride concentration
would reduce with increasing depth. In this instance the
measured chloride concentrations are uniform throughout
the concrete sections.
A literature review of historic concrete was carried out and

current testing involves re-creating the 1908 concrete mix
to determine if saltwater was used in the concrete mix. The
use of saltwater would explain the high uniform chloride
levels present in the concrete. Another explanation may be
attributed to leaching of concrete following the application
of the bituminous coating in 1972.
Fig 1: The Mizen Bridge
The 28 day concrete compressive strength for the mix
Innovative method for assessing shrinkage cracking of
concrete and other cement-based materials at early-ages:
experimental and numerical modelling
46 Olayinka Oladiran
Brunel University
Project objectives Results

• Investigate how the elliptical ring specimen can Specimens with shorter minor axis (i.e. 300*150) showed
provide information for likelihood of cracking; crack initiation at an earlier age of 10 days as expected,
and subsequent propagation appeared closer to the
• To evaluate early age properties; Cracking criteria major principal axis compared to specimens with larger
under sustained loading in restrained conditions; minor principal axes (i.e. 300*250 & 300*200) indicating
stress developed at the axis as expected. Once cracks are
• To develop a structural model to predict the position initiated, it is assumed that no further cracks are generated
and time of initial cracking in the elliptical ring test. as the energy is now concentrated on the cracks already
generated. The experimental and numerical results so
Project overview far suggest that elliptical rings with a decreased minor
principal axis can increase the likelihood of crack initiation
In a bid to overcome the problems from traditional circular
in concrete specimen at expected positions and seen to
ring tests and to improve the ability of generating a visible
crack at an earlier time. This study indicates that early-age
crack within a short period, a novel elliptical ring test [1] is
cracking behaviour of a cementitious system is geometry
used for investigating the cracking potential in concrete.
dependent.
This project is focused on testing and modelling shrinkage
cracking of concrete and other cement-based materials
at early ages, the study introduces a new cracking test by
using a proposed elliptical ring specimen to assess the This study proposes a novel and faster test method of
cracking potential of concrete as well as the sensitivity assessing cracking potential and evaluating early-age
of the steel ring. A combined thermo-mechanical 3-D concrete cracking properties which can be attributed to the
numerical model, based on ANSYS FE code, is also being elliptical geometry adopted. It is however fully anticipated
developed for describing early-age shrinkage cracking of that the approach presented in this study can be extended
concrete in the elliptical ring test already being investigated. for general use in quantifying the cracking potential of
concrete under drying as well as the effects of the degree
Applications of restraint on cracking potential of concrete. As a test
method for assessing cracking potential of concrete
• AASHTO (American Association of State Highway and
mixtures, the ability of generating a visible crack within a
Transportation Officials) uses a 12.5-mm-thick steel
short period is desirable for a large amount of assessment
ring, a 75-mm-thick concrete ring, and allows the
of cracking tendency of concrete.
concrete to dry from the outer circumference.
References
• ASTM (American Society for Testing Material) uses a
12.5-mm-thick steel ring; however the concrete wall [1] Z. He, X.M. Zhou, and Z.J. Li (2004), “New
thickness was reduced to 37.5mm to encourage Experimental Method for Studying Early-Age Cracking
cracking to develop at an early age. of Cement-Based Materials,” ACI Mater. J., 101(1), pp
50-56.
Further information
Test Method
Olayinka Oladiran (E: Olayinka.Oladiran@brunel.ac.uk) or
The test setup typically comprises of steel rings having
Dr Xiangming Zhou (E: Xiangming.Zhou@brunel.ac.uk)
different geometries, with four strain gauges attached to
the inner surface of the steel rings. The circular and elliptical
concrete ring specimens were cast around the structural
steel ring in moulds. After been moist cured for 24 hours
under initial conditions, the ring series were immediately
sealed at the top and bottom surfaces with aluminium
tape while the ring was permitted to dry from the outside
circumference cylindrical surface of the concrete ring, as
suggested by the standard testing procedures ASTM C
1581-04 and AASHHTO PP34-99. The strain gauges were
then connected to the data acquisition system and the
instrumented specimens were subsequently moved into
the well-controlled environmental chamber for continuous
drying. The chamber was kept at a temperature of 23°C
and 50% relative humidity, where all the specimens were
placed to dry for strain monitoring throughout the duration
of the test for regularity.
Vulnerability of cable-stayed bridges
Oluremi Olamigoke
University Of Surrey 47
Brief description Description of method and results
Over the last half-century, there has been an increase in A non-linear global dynamic analysis rather than the quasi-
the construction of cable-stayed bridges around the world. static analysis that is recommended by codes and major
This development has been due to its economical solution guidelines has been carried out by varying the traffic load,
for carrying traffic loading for spans from 200m and rupture time and number of broken stays using Abaqus
above, high strength materials, development of computer 6.10.
technology and better construction methods.
My future work will concentrate on the loss of the pylon
The cables are vital to ensure stability and load carrying and further analysis on this aspect will also be considered.
capacity of the bridge structure. These cables can break Local analysis will also be carried out to consider the effect
due to accidental or intentional extreme events, lack of these accidental actions on the anchorage zone of the
of maintenance over a long period of time, excessive cable stayed bridge.
corrosion of the connection or loosening a cable before
replacing it. Likewise, the pylon can be subject to Potential for application of results
blast, impact or fire loading which can also lead to the
It is therefore of importance to study whether these highly
vulnerability of this structure thus progressive collapse.
redundant structures which are huge in size and exhibit
complicated nonlinear structural behaviour can withstand
Previous work, international guidelines and codes has
this accidental/intentional loads and fulfil their economic
focused on the effect of the loss of one cable on the bridge
benefit to the society.
and individual work has also been done on the effect of the
loss of the pylon of cable-stayed bridges to blast load.
The results obtained will also help in developing a guideline
on how to carry out a dynamic analysis of the cable-stayed
bridge to accidental loads as the loss of the cable or pylon.
The objective of this research is:
This research will also benefit the code of practice due
• To determine if a cable-stayed bridge will remain stable to the present lack of knowledge in carrying out dynamic
after damage or loss of a major element. analysis for the sudden and gradual loss of a cable. This
will provide potential improvement to designing robust
• To further understand the dynamic response of cable- cable-stayed bridges.
stayed bridges to loss of a major element.
Further information
• To establish the likely rupture time of the cable
Oluremi Olamigoke (E: oo00033@surrey.ac.uk) or
members used in a typical cable-stayed bridge.
Prof Gerry Parke (E: g.parke@surrey.ac.uk) and
Dr Boulent Imam (E:b.imam@surrey.ac.uk)
• To determine the local effects of the loss of a cable
and pylon.
• To determine the effect of damage to a pylon on the

stability of a cable-stayed bridge.
• To evaluate the effectiveness of different analysis

procedures suitable for nonlinear dynamic analysis.
The influence of changing temperature on adhesively
bonded FRP strengthening
48 Daryan Othman
University of Edinburgh
Project objectives and goals: After the small scale tests have been completed, the
adhesive property data will be used to develop an
Externally Bonded FRP (EB FRP) is used to strengthen
analytical model that examines the behaviour of the time
reinforced concrete and steel structures, and this project
and temperature dependent behaviour of the adhesive
investigates the impact of service temperatures upon EB
interface, by extending previous bond models [5,6].
FRP. The durability and performance of EB FRP depends
directly on the behaviour of the bonding adhesive, which is
Finally, a small series of FRP-strengthened steel beams will
highly sensitive to temperature. Current design guidance [1,
be tested under combined heat and load, and the results of
2] recommends that the glass transition temperature (Tg) of
these tests will be contrasted with the analytical results.
the adhesive is 15°C above the maximum design service
temperature.
Potential for application and results:
The aims of this project are to investigate the effect of This project will offer a better understanding of the effects
elevated temperature on adhesive properties, and to of service temperatures on the adhesive joint in EB FRP.
provide reliable information on temperature effects for Fundamental information on curing temperature and
strengthened steel members. The project will: humidity effects will demonstrate the importance of an
appropriate curing regime for quality control samples, and
• Characterise widely used adhesives near Tg Fig 1. will be used in contrast with the elevated temperature
curing regimes currently given on the data sheets of some
• Investigate the impact of the curing environment (time, manufacturers.
temperature and humidity) on adhesive behaviour Fig
2. The project will allow a designer to examine the effects
of temperature upon EB FRP, including the potential for
• Investigate the creep behaviour of a typical bonding damage due to daily or annual temperature cycles, as well
adhesive. as extreme temperatures. It will also contribute to research
into the fire performance of EB FRP, although this is not the
• Apply the adhesive property data to analyse the primary aim of the project. The analytical bond model will
behaviour of steel beams strengthened using FRP at allow the designer to study these effects.
temperature.
This study has the potential to become an important
• Perform experimental work to confirm the effect of reference in the strengthening of steel and concrete
temperature upon an FRP strengthened beam. structures, and will be published as design guidelines to
allow its use by the industry.
Description of method and results:
The first part of the project is to develop data on the
performance of a typical epoxy-bonding adhesive at
elevated temperatures, after they have been subjected to
a variety of cure conditions. This is currently under way
using Dynamic Mechanical Analysis (DMA) Fig1. Adhesive
samples are being cured at a range of temperatures,
humidity and times to determine how these affect Tg,
to establish fundamental data for use in design, and to
help understand the correct cure environment for quality
control specimens that are used to check that on-site
strengthening work has been successful.
The second series of tests examines the shear behaviour of

the adhesive at elevated temperature. Adhesive joints will
be tested in torsion to give pure shear to give information
on shear stress, strain and deformation of the adhesive.
The time-dependent behaviour of the adhesive will also be
studied using the same arrangement.
References: Further information
[1] American Concrete Institute (ACI) ACI 440.2R-08, Daryan Othman (E:D.Othman@ed.ac.uk) or
“Guide for the Design and Construction of Externally Dr Tim Stratford (E: Tim.Stratford@ed.ac.uk) and
Bonded FRP Systems for Strengthening Concrete Dr Luke A. Bisby (E: Luke.Bisby@ed.ac.uk)
Structures.,” Farmington Hills, MI.2008.
[2] Concrete Society, “Design guidance for strengthening

concrete structures using fibre composite materials,”
Camberley, U.K.2012.
[3] ASTM E1640, “Standard Test Method for Assignment

of the Glass Transition Temperature By Dynamic
Mechanical Analysis,” 100 Barr Harbor Drive, West
Conshohocken, PA 19428-2959, United States.1999.
[4] ISO. (2012), “Plastics. Determination of dynamic

mechanical properties. Part 11 - Glass Transition
Temperature “, ISO 6721-11, British Standards
Institution, London, UK. 2012.
[5] T. Stratford and J. Cadei, “Elastic analysis of adhesion

stresses for the design of a strengthening plate bonded
to a beam,” Construction and Building Materials, vol.
20, pp. 34-45, 2006.
[6] T. Stratford and L. A. Bisby, “Effect of Warm

Temperatures on Externally Bonded FRP
Strengthening,” Journal of Composites for
Construction, vol. 16, pp. 235-244, 2012
Fig 1: Typical Storage modulus, Loss modulus, and tan δ Fig 2: Storage modulus response with temperature of
for structural epoxy cured at 50ºC, dry and 28 days with typical structural epoxy
recommended analytical Tg points
Investigations to determine the usability of basalt fibre
reinforced polymer
49 Max Pearson
Kingston University
Through a varied series of tests and mathematic models A total of eight C30/35 concrete beams were cast with
Poster presenter
this PhD project has accomplished in-depth analysis on ducting to allow for the post tensioning of BFRP and high
Basalt Fibre Reinforced Polymer (BFRP) for its use in Civil yield reinforcing steel. All beams were subjected to two
Engineering projects. BFRP is a relatively new composite point loads until failure of the beams was established.
material in the FRP family and has a high potential for more The samples with various types of BFRP and steel
economical, sustainable, ease of manufacture and better reinforcement and differing degrees of prestressing
life spans in comparison to other FRPs and steel. were analysed and compared with the results from
unprestressed control samples.
BFRP in this project has been investigated in relation to
tensile testing, varied durations of saturation in aggressive Finite element analysis has also been performed in ANSYS
environments which were both acidic and alkali, scanning and is currently being undertaken in the Multiphysics
electron microscopy, energy dispersive x-rays, long term modelling software COMSOL for the prediction of the
creep testing and prestressing BFRP in post-tensioned behaviour of the prestressed beams.
concrete. These tests have all been performed with
encouraging results. These investigations are compared Further information
with high yield steel reinforcing bar in the same conditions
Max Pearson (E: maxpearson@me.com) or
to establish a good and unbiased argument on the field
Dr Ted Donchev (E: t.donchev@kingston.ac.uk)
potential of BFRP.
The aggressive solutions investigation took the form of a

preliminarily study on the tensile degradation, and a full
secondary study on the estimation of the influence over
these environments. Over 100 samples were tested over
a period of 7 months for their tensile capacity as well as
scanning electron microscopy to see how the surface
of the bars were being effected in these conditions.
Employing an energy dispersive x-ray to see the chemical
changed in reactions to the bar, to show how the polymer
fibres and matrix were interacting with the environment. It
was established that the environments less affected BFRP
in comparison with steel, and that tensile strength was
maintained for BFRP.
The creep testing was performed over a period of over

100 days. BFRP, Steel reinforcing bar and steel cable was
loaded to 50% of their ultimate tensile capacity and allowed
to creep and relax while the effects were monitored using
National Instruments DAQ (data acquisition) and LabView
for Data analysis. Results showed similar creep behaviour
to the steel reinforcing bars, showing that BFRP has a high
potential to be prestressed for use in concrete.
Assessment of modes of failure in hybrid steel and FRP
shear walls
Natalja Petkune
Kingston University 50
Background Project objectives and goals
Steel shear walls are a popular lateral resisting system The aim of the study is to investigate earthquake resistance
because they have many advantages. The main advantage of hybrid steel and FRP shear walls. Several types of
of the steel shear wall is that they are much lighter than the shear wall samples will be tested consisting of steel
reinforced concrete shear walls, which means that wall and FRP infill plate. Hybrid shear walls are designed by
thickness, weight of the building and dynamic forces and attaching carbon prepreg layers to the steel panel in
loads on foundation can be substantially reduced (Roberts, accordance to design instructions. Carbon prepreg are
1995). In addition, they have enhanced stiffness, strength, based on 120oC curing toughened, epoxy matrix resins.
ductility and large energy dissipation capacity.
Description of method
An innovative type of the shear wall is hybrid steel shear
Experimental investigations will be conducted to evaluate
wall which combines the application of the fibre reinforced
the lateral seismic characteristics of composite shear
polymers (FRP) layers and steel plate. Recent investigations
wall system under quasi static regime in accordance to
about hybrid shear walls have shown excellent results in
the ATC-24 protocol. ATC- 24 protocol was used as a
terms of the initial stiffness and reduction of structure’s
guide for cycling loading. A number of amplitudes up to
weight. Hatami and Rahai (2009) conducted experiments
25mm will be applied for three cycles to evaluate energy
on steel shear wall with CFRP. Hybrid shear wall was
dissipation of the system.
achieved by roughening steel plate surface with sand blast
and then attaching FRP layer on both sides of steel plate
To test a rectangular hybrid steel plate shear wall under
with epoxy resin. Experimental tests showed improvements
cyclic loading, a testing rig had been previously set up
in shear stress results and higher energy absorption
at Kingston University Structural Laboratory. Figure 1
capacity in comparison to ordinary steel shear plate.
demonstrates different parts of the testing rig. Testing
equipment consists of different components which are
Further experimental and numerical studies carried out
responsible or assist specific actions: loading, measuring
by Hatami, Ghamari and Rahai (2012) compared the
and recording. Infrared thermography (IRT) will be
behaviour of steel shear wall (SSW) and steel shear wall
implemented to monitor and evaluate the effect of the
with carbon fibres (CSSW). For experimental study, one
delamination between steel and FRP layers.
layer of CFRP was applied on each side of the 3mm steel
plate. They looked at fibre content/ angle and width of the
panel to investigate how it influences the behaviour of the
steel shear wall. It was found that when higher contents of Studies on Hybrid steel and FRP shear wall have
fibre are used, energy absorption, stiffness, over- strength demonstrated good results in terms of the improvement
and capacity of the steel shear wall is increased, and of the energy dissipation and stiffness of the system,
ductility is decreased. which is of a great importance in seismic design. Further
studies would allow investigation of the behaviour of the
Maleki et al. (2011) investigated the application of glass steel and FRP shear walls at a number of amplitudes and
fibre reinforced polymers (GFRP)- steel sandwich shear provide further knowledge about the development of the
wall in order to determine how GFRP sandwich plates delamination.
can enhance the behaviour of shear wall under seismic
loading. It was found that steel shear wall experiences
sudden buckling of the infill plate at the early stage of
seismic loading. In this case, it is necessary to increase
stiffness of boundary elements. However, when GFRP
steel shear wall system was tested, uniform tension fields
are developed within infill plate during the test and initial
stiffness of the system is preserved. It was demonstrated
that use of GFRP laminates affects in-plane stiffness and
out-of-plane deformation of the infill plate. This system has
higher loading capacity. However, currently there is still lack
of the knowledge of this type of the shear wall, extensive
study is necessary to provide an important contribution to
the design and inspection of these walls.
Hatami, F. and Rahai, A, “Effect of carbon fiber reinforced Natalja Petkune (E: N.Petkune@kingston.ac.uk) or Dr
polymer on steel plate shear wall”. Proceedings, Annual Ted Donchev (E: T.Donchev@kingston.ac.uk), Dr Homa
Conference - Canadian Society for Civil Engineering. 27-30 Hadavinia (E: H.Hadavinia@kingston.ac.uk), Dr David
May 2009. Wertheim (E: D.Wertheim@kingston.ac.uk) and Prof
Mukesh Limbachiya (E: M.Limbachiya@kingston.ac.uk)
Hatami, F., Ghamari, A. and Rahai, A. (2012) “Investigating
the properties of steel shear walls reinforced with Carbon
Fiber Polymers (CFRP)”, Journal of Constructional Steel
Research, Vol.70, pp.36-42.
Maleki, A., Donchev, T., Hadavinia, H., Limbachiya, M.,

“Improving the seismic resistance of structure using FRP/
steel shear walls”. Proc. 6th International Conference CICE
2012, Rome, Italy, 13-15 June 2012.
Roberts, T.M., “Seismic resistance of steel plate shear

wall”, Engineering Structures, Vol. 17, No.5, 1995, pp.
344-351.
Fig 1: Testing rig
Experimental and finite element modelling of protected
cellular steel beams in fire conditions
Klelia Petrou
University of Ulster 51
Poster presenter
Project objectives and goals Temperature distribution on the composite cellular beams
was recorder by 33 thermocouples disposed at various
The present paper describes an experimental and finite
locations along the cellular beam in different zones and
element analysis, study at elevated temperatures on
by 10 thermocouples disposed along the composite slab.
the behaviour of three composite cellular steel beams
Deflections and axial displacements were recorded using
protected with intumescent coating having different size
5 Linear Differential Transducers LVDT’s, placed in different
and openings shape. This research has been propelled by
locations on the cellular beams.
the growing popularity of the use of cellular steel beams in
composite floors and the increasing demand of attention to
The finite element modelling software TNO-Diana was used
the fire safety engineering design. In fire, the temperature
to complete the numerical investigation. Comparison of the
distribution across a composite member is non-uniform,
experimental and FEM results is presented.
since the web and bottom flange have thin cross-sections
and a greater exposed perimeter than the top flange.
The deterioration of the material properties of the web
will therefore become an important effect on the overall Figure 1 present the deflection recorded in the three fire
performance of the member in the event of fire. tests of protected composite cellular beams compared
with the deflections obtained from the FEM analysis, using
Fire resistance and protection of cellular-beams has been TNO-Diana. In the FEM two approaches were used. In the
very controversial concerning their behaviour in elevated first case the top and bottom web were modelled using the
temperatures, the fire protection material and the required maximum recorded temperatures (red line on the graphs)
thickness. The stability of a building, having intumescent and in the second case we used the average temperatures
coating protected steel structural elements, in case of fire recorded from the experiment. In addition the photos
depends on the thickness of the coating, the depth and present the buckling effect observed on the composite
insulation properties of the char produced of it [1, 2]. cellular beams after the fire test.
The applied required insulation thickness of the References

intumescent coating is in respect of the required fire
[1] H. van de Weijgert “Protecting structural steel” Fire
resistance, given in minutes, for example 60 or 90 minutes.
Safety Engineering, (2001)16-19,
The European Standards on the other hand provide
four different methods of evaluating the contribution
[2] P.Morris “Test of Strength” Fire Safety Engineering,
of intumescent coating. For the current paper it was
(2005)17-20,
decided to use the differential equation provided from
Eurocode 3, EN 1993-1-2, for calculating the temperature
[3] R.P.Krishnamoorthy, C.G.Bailey “Temperature
increase Δθa,t of a protected steel member, in the case
distribution of intumescent coated steel framed
of uniform temperature distribution in a cross-section
connection at elevated temperature”, NSCC(2009),
[3] using variable, λ, approach [4]. The finite element
572-579
analysis model, using TNO- Diana, having as heat load the
calculated temperatures from Eurocode 3 approach was
[4] Committee of European Normalisation CEN (1995)
compared with the experimental results for all the three
Eurocode 3: design of steel structures, part 1- 2:
cellular steel beams.
general rules-structural fire design, ENV 1993-1-2.
British Standards Institution, London
In fire tests conducted in FireSert facilities in University Funding Body
of Ulster, were tested simply supported composite slabs
VCR scholarship
with downstand cellular beams in furnace. The protected
composite cellular beams were asymmetric with a large
Further information
web opening (Beam1), symmetric with two large web
openings (Beam2) and asymmetric with multiple circular Klelia Petrou (E: petrou-k@email.ulster.ac.uk) or
openings (Beam 3). Thermal loading, represented by the Prof Ali Nadjai (E: a.nadjai@ulster.ac.uk)
ISO parametric curve and vertical load 200kN was applied
for Beam1 test and 150kN for Beam2 and Beam3 tests.
Fig 1: Composite cellular beams 1,2, 3 deflections and buckling effect, experimental, and TNO-Diana results for
maximum web temperatures and EC3 analytical method
Seismic performance of prestressed concrete girders
strengthened in shear with CFRP reinforcement
Michael Qapo
University of Birmingham 52
Background Methodology and preliminary results
Earthquakes can have detrimental effects on and cause Non-linear finite element (NLFE) analyses are being carried
serious damage to bridges. In the past, several bridges out to simulate the structural response of experimental
in seismic-active areas, e.g. Northridge (1994) and Kobe specimens from the published literature. Appropriate
(1995), were not able to withstand seismic actions and constitutive relationships are employed to model the
eventually collapsed [1, 2]. The collapse of a bridge can behaviour of concrete, internal steel reinforcement, CFRP
lead to serious economic repercussions, very high user reinforcement, and CFRP-to-concrete interface.
delays and significant detour costs. Therefore, bridges
deficient in seismic resistance should be strengthened to Preliminary results for specimens subjected to static
minimise the possibility of their collapse due to seismic load show good agreement between the FE-predicted
actions and preserve them in good service conditions. and experimental behaviour. Figs. 1 and 2 compare,
respectively, the crack pattern at failure and shear force-
Carbon fibre reinforced polymer (CFRP) reinforcement deflection curves for specimen T4-18-control tested by You
has been successfully applied as external and/or internal et al. [5]. It can be seen that the experimentally observed
reinforcement to strengthen existing concrete structures shear failure was well captured by the FE model developed
such as bridges [3]. CFRP reinforcement has several as part of this project. The FE model also predicted the
advantages over conventional strengthening techniques initial stiffness of the tested beam quite well, suggesting
utilising steel plates bonded to the soffit and/or sides of a that the elastic constants and boundary conditions were
bridge. Such advantages include high strength-to-weight well modelled.
ratio, corrosion resistance and ease of application [4].
However, CFRPs are brittle. This implies that conventional Potential for application of results
design approaches based on plasticity theory cannot be
As stated above, the results of this project will identify the
used to predict the capacity enhancement offered by CFRP
effects of several parameters on the seismic response
strengthening systems.
of prestressed concrete girders strengthened in shear
with CFRP reinforcement. Hence, the outcomes of this
This research project aims to develop predictive tools
project will help improve our general understanding of
for the seismic response of prestressed concrete girders
such structures. The results will also help evaluate and, if
strengthened in shear with CFRP reinforcement. The
necessary, update current seismic design approaches for
project will identify the influence of several parameters such
prestressed concrete girders.
as CFRP reinforcement ratio, prestressing force in tendons,
and shear span to effective depth ratio on the strengthened
References
behaviour.
[1] Agarwal, P., and Shrikhande, M., “Earthquake Resistant
Project objectives Design of Structures”, PHI Learning Private, New Delhi,
India, 2010.
(1) To develop FE predictive tools capable of simulating
the seismic response of prestressed concrete girders
[2] Priestley, M. J. N., Seible, F., and Calvi, G.M., “Seismic
strengthened in shear with CFRP reinforcement.
Design and Retrofit of Bridges”, 2nd Edition, Wiley-
IEEE, Canada, 1996.
(2) To validate the accuracy of the FE models using
experimental results from published literature.
[3] Belarbi, A., Bae, S. W., Ayoub, A., Kuchma, D.,
Mirmiran, A., and Okeil, A., “Design of FRP Systems
(3) To carry out a parametric study to identify the influence
for Strengthening Concrete Girders in Shear”, Technical
of CFRP reinforcement ratio, prestressing force,
Report of NCHRP, Project 12-75, National Cooperative
concrete strength, shear span to depth ratio, and
Highway Research Program, Transportation Research
internal shear reinforcement ratio on the strengthened
Board, Washington, USA, 2010.
behaviour.
[4] ACI Committee 440, “State-of-the-Art Report on Fiber
(4) To enrich the general understanding and enhance the
Reinforced Plastic (FRP) Reinforcement for Concrete
weak archival data in the field of seismic performance
Structures”, ACI 440R-96, American Concrete Institute,
of prestressed concrete structures strengthened in
Farmington Hills, Michigan, USA, 2002.
shear with CFRP reinforcement.
[5] You, Y. M., Ayoub, A., and Belarbi, A., “Three-
Dimensional Nonlinear Finite-Element Analysis of
Prestressed Concrete Beams Strengthened in Shear
with FRP Composites”, Journal of Composites for
Construction, ASCE, Vol. 15, No. 6, December 2011,
pp. 896-907.
Further information
Michael Qapo (E: mkq118@bham.ac.uk) or
Dr Samir Dirar (E: s.m.o.h.dirar@bham.ac.uk)
Fig 1: Crack patterns at failure (a) experimental (b) current FE work
Fig 2: Shear force – deflection curves of girder T4-18-control
Mesoscale modelling of fibre reinforced concrete
considering random distribution of fibres
Ansam Qsymah
University of Manchester 53
It is well known that, the mechanical performance will be inserted along the edges of the matrix elements to
of concrete enhances significantly by adding short represent potential crack paths[4].
random distributed fibers to the concrete mixture. This
enhancement related to the ability of added fibres to bridge X- ray microtomography along with high resolution 3D
the opposite faces of the cracks and thus stresses still able imaging technique will be utilised to calibrate and validate
to transfer between the crack faces up to relatively large these models. Small specimens of the UHPFRC will be
crack mouth opening displacement. This phenomenon, scanned to determine the internal structure of the fibres
which usually called the bridging effect of fibres, greatly through the matrix. Then in-situ loading will be applied
increases the tensile and compressive strength of fibre incrementally based on a wedge splitting test. The fracture
reinforced concrete (FRC) as well as its fracture toughness process of these specimens will be monitored from crack
and ductility. For instance, ultra high performance fibre initiation at microscopic scale up to complete failure.
reinforced concrete (UHPFRC) is relatively a new class of This advanced technique will provide us with detailed
the FRC that has compressive strength more than 150MPa description for the internal structure of the specimen as
and tensile strength more than 10MPa [2]. well as quantitative measurements of internal damage
progress.
UHPFRC is increasingly used in many structural
applications such as slabs, beams and bridge deck In the initial model, a 2D square specimen was subjected
overlays. Nevertheless, its applications are still limited with to uniaxial tensile load. The concrete matrix is represented
respect to its potential. This is mainly due to lack of reliable using plane stress elements (CPS3 and CPS4R) and fibres
design codes, high fibre cost and limited experience with are represented as truss elements (T2D2) embedded in the
its behaviour [1]. The objective of this project is to study concrete matrix. The initial results of the developed model
the fracture behaviour of this material in order to improve show that this model can simulate the fracture process of
our understanding for this material. The results of this the UHPFRC. It is able to give good indication about the
study can hopefully help in increasing the applications of effect of fibre content on the load bearing capacity of the
UHPFRC and exploit most of its outstanding merits. The material. Also, uncertainties related to fibre orientation can
specific goals of this study are: be investigated using this model to improve the reliability of
the UHPFRC.
• To develop mesoscale finite element models that are
able to simulate the fracture of the FRC with explicit References:
representation of short random fibres based on
[1] Di Prisco, M., Plizzari, G. & Vandewalle, L. (2009) Fibre
cohesive crack model.
reinforced concrete: new design perspectives. Materials
and Structures, 42, 1261-1281.
• To incorporate X-ray computed tomography XCT
technique into finite element method to calibrate the
[2] Hassan, A. M. T., Jones, S. W. & Mahmud, G. H.
model and validate its results.
(2012) Experimental test methods to determine the
uniaxial tensile and compressive behaviour of ultra
• To perform a parametric study on main design factors
high performance fibre reinforced concrete (UHPFRC).
in order to optimise these factors and minimise the
Construction and Building Materials, 37, 874-882.
material cost.
[3] Suuronen, J. P., Kallonen, A., Eik, M., Puttonen, J.,
It has been proven that the distribution and orientation
Serimaa, R. & Herrmann, H. (2013) Analysis of short
of fibres through the concrete matrix has a considerable
fibres orientation in steel fibre-reinforced concrete
influence on the mechanical properties of the material
(SFRC) by X-ray tomography. Journal of Materials
[3]. This influence must be considered carefully before
Science, in press.
employing this material confidently in practical structures.
To study the effect of fibre structure on the FRC
[4] Yang, Z., Su, X., Chen, J. & Liu, G. (2009) Monte Carlo
performance numerically, fibres need to be represented
simulation of complex cohesive fracture in random
explicitly. There are very limited studies that modelled
heterogeneous quasi-brittle materials. International
fibres explicitly in fibre reinforced concrete. This is due
Journal of Solids and Structures, 46, 3222-3234.
to the high computational cost associated with the
large number of fibres and difficulties in considering the
Further information
complicated interaction mechanisms between fibres and
the surrounding matrix. In this study, these difficulties will Ansam Qsymah
be addressed by developing finite element model in 2D and (E:ansam.qsymah@postgrad.manchester.ac.uk) or
3D. In these models the UHPFRC will be modelled as a two Dr Zhenjun Yang (E:zhenjun.yang@manchester.ac.uk)
phase material; matrix and fibres. Fibres will be modelled
as embedded elements into the background mesh that
represents the concrete matrix based on cohesive crack
model. In this model zero-thickness interfacial elements
Structural behaviour of concrete twin walls
54 Jaylina Rana
University College London and Laing O’Rourke
Poster presenter
Project objectives and goals Thus to help clarify these points a series of push-out tests
aimed at elucidating the effect on shear stress of the
The overall aim of this project is to develop a deeper
following parameters were carried out:
understanding of the structural behaviour of the twin wall
system developed by Laing O’Rourke (LOR). The twin
(1) Surface roughness of the concrete at the core/biscuit
wall system is a relatively new form of hybrid concrete
interface – (1) very smooth, (2) smooth, (3) rough
construction consisting of two precast concrete biscuits,
reinforced with longitudinal and transverse reinforcement,
(2) Compressive strength of core and biscuit concrete –
which are connected by shear connectors in the form of
Biscuit 1/Inner Core/Biscuit 2: (1) 50 N/mm2 / 50 N/
3-dimensional triangular steel lattices, partially embedded
mm2 / 50 N/mm2, (2) 50 N/mm2 / 10 N/mm2 / 50 N/
in the inner faces of the biscuits. The void between the
mm2, (3) 10 N/mm2 / 10 N/mm2 / 10 N/mm2.
biscuits is then filled with in-situ concrete.
(3) Embedment depth of lattice shear connectors – (1) 15
To-date, twin walls have been primarily used as
mm, (2) 20 mm, (3) 25 mm
compression elements including shear walls, bearing/
non-load bearing walls and cladding panels. LOR is now
(4) Diameter of bars in lattice shear connectors - (1) 3 mm,
seeking to extend the range of applications of this element
(2) 6 mm (not included in this report)
to other forms of structure, in particular the walls of water
tanks. In this situation the walls act as cantilevers and will
Results
be subject to combined bending and shear. The bending
strength is a function of the strength of the outer biscuits • Effect of Surface Roughness: It can be seen that
and the degree of composite action, whereas the shear the failure load increases with increasing surface
strength is a function of the lattice connector. The overall roughness. The results show that there is a good
aim of this project is to develop a deeper understanding of agreement between the experimental and theoretical
the shear and flexural behaviour of twin wall elements both values of failure load for ‘smooth surface’ specimens
of which depend on the longitudinal shear capacity. but poor and very poor agreement for the ‘very
smooth’ and ‘rough’ surfaces. The results therefore
Methodology confirm that the values recommended by EC2 are
largely insensitive to failure load.
The research for this project was undertaken in two distinct
stages to investigate the shear and flexural behaviour of the
• Effect of Concrete Strength: It is observed that the
twin wall system respectively. Theoretical, experimental and
failure load increases with increasing strength of
finite element analyses were carried out. For the purpose
concrete in the outer biscuits but is not affected by the
of this synopsis, the methodology and results obtained for
strength of the concrete in the core. Cl 6.2.5 of EC2
the analysis of the shear behaviour of the twin walls are
advises that the tensile strength of the concrete should
summarised.
be taken as that of the new concrete, in this case the
core. However, the test results suggest that it would
Section 6.2.5 of Eurocode 2: Part1.1 (BS EN 1992-
be more sensible to base the tensile strength on the
1-1:2004) (EC2) suggests that the shear stress at the
concrete used in the biscuits rather than the core.
(precast/in-situ) concrete interface is primarily dependent
on the nature/ roughness of the biscuit/in-situ concrete
• Effect of Embedment Depth: The test results also
surfaces, concrete strength of the biscuits and dowel
show that the failure load increased when the
strength of the shear connectors. Use of the EC2 equation
embedment depth is increased from 15mm to 20mm
also assumes that the steel is fully anchored on both sides
but no further increase is obtained by increasing the
of the interface. However, no work has been carried out
embedment depth to 25mm. This suggests that the
to determine the influence of the above parameters on the
lattice is fully anchored at an embedment depth of 20
shear strength of the LOR twin walls.
mm but there would not appear to be any easy way
of estimating the embedment depth necessary to
achieve anchorage of the lattice in other situations.

The results are intended to provide design values and to
assist in the understanding of the general behaviour of
the LOR concrete twin wall system. The aim is to use the
results as a basis for extending the use of twin walls to
other applications such as retaining walls.
British Standard Institute. (2004). BS EN 1992-1-1: 2004, Jaylina Rana (E: j.rana@ucl.ac.uk) or
Eurocode 2: General – Common rules for building and civil Dr Chanakya Arya (E: c.arya@ucl.ac.uk)
engineering structures. London: BSI Group.
Funding body
EPSRC
Fig 1: Details of test specimens and test set-up
Fig 2: Effect of (1) surface roughness, (2) concrete strength and (3) embedment depth on failure load
Noise and vibration from building mounted small wind
turbine
55 Bo Ren
Coventry University
Project objectives and goals The FRF was measured by impact hammer hitting, say,
MP1 (reference point) and measuring accelerations at other
When a small wind turbine (SWT) is installed on a building
mass positions in the structure. This would yield FRF1,1,
there is potential for airborne noise, structure borne noise
FRF1,2, FRF1,3…, FRF1,n. Given the frequency response
and vibration of the structure. Methodologies for the
functions, it is possible to determine the natural frequencies
assessment and reduction of the structure borne noise
and mode shapes by modal analysis techniques (Karadelis,
and vibration are not currently available. As a result, the
2011). If the modal analysis is completed, the experiment
installation of SWT on a variety of dwellings and other
results (models) could be used to validate the theoretical FE
commercial and industrial structures in the UK has taken
models. This part of the results is shown in Fig.2.
the back seat and is not included in the “General Permitted
Development Order”.
The aim of this research project is to develop a method The project hopes to provide a robust theoretical FE model
of prediction of vibration, and structure borne noise, for a small wind turbine - wall system. Thus anyone else
applicable to a wide variety of installations in the UK by could decide whether they could mount a wind turbine
using primarily advanced numerical analysis techniques. on the building by the prediction results of our validated
The methodology to predict these impacts should be models to avoid vibration problems and it is easy to predict
sufficiently robust that decisions relating to the amenity the vibration levels accurately without any experimental
(and not merely) of neighbouring properties can be testing. Moreover, the experimental results will provide a
predicted. method to reduce the vibration levels and eventually claim
the “General Permitted Development Order”.
References
Numerical modal analysis
BWEA (2008) Small Wind Turbine Performance and Safety
Modal Analysis is a linear analysis as the structural
Standard.
properties (natural frequencies and mode shapes)
obtained, characterise the linear response of a structure
Karadelis, J.N. (2009a). “Concrete Grandstands. Part I.
under excitation. The numerical modal analysis is to set
Experimental Investigation”, Proceedings of the Institution
up a Finite Element (FE) Modal to perform the above linear
of Civil Engineers, Engineering and Computational
analysis using ANSYS code. The model was based on the
Mechanics Journal 161, Issue EM1, doi: 101680/
SOLID65 element chosen to represent brick wall. This is
eacm.162.1.3, 3-9.
a 3D, eight node, solid, isoparametric element with three
translational degrees of freedom (DOF) per node. When in
Karadelis, J.N. (2009b). “Concrete Grandstands. Part II.
a non-linear domain, it is capable of simulating the dynamic
Numerical Modelling”, Proceedings of the Institution of
behaviour of a free-standing wall (Karadelis 2009a, 2009b).
Civil Engineers, Engineering and Computational Mechanics
The Modulus of Elasticity was taken from relationships and
Journal 162, Issue EM1, doi: 101680/eacm.162.1.1, 11-
estimated by routine laboratory tests (Karadelis 2009a).
21.
This part of the FE analysis results is shown in Fig. 1.
Karadelis, J.N. (2011). “Reliability Pointers for Modal
Experimental Modal analysis
Parameter Identification of Precast Concrete Terraces”,
Experimental Modal analysis will take the response data Computers and Concrete.
of the structure, which is associated with the excitation
force properties (such as the excitation frequency) and from Funding body
these extract the modal parameters (natural frequencies,
Faculty of Engineering and Computing of Coventry
mode shapes etc). The free-standing wall was modelled
University (RAE MONEY)
by dividing it into a set of lumped masses. These masses
are intended to be connected to a set of spring/damper
Further Information
elements (Karadelis, 2011). In modal analysis, masses are
used as measurement points (MPs) and the properties Bo Ren (E: renb2@uni.coventry.ac.uk) or
of the structure are determined by measuring the FRF Dr John N Karadelis (E: aa9118@coventry.ac.uk)
(Frequency Response Function) at each of the MPs
(Karadelis, 2011).
Fig 1: Mode shape of free-standing wall
Fig 2: FRF of a cantilever steel beam
Multi-scale stochastic fracture mechanics of concrete
56 Wenyuan Ren
University of Manchester
Poster presenter
Project objectives and goals A 2D simulation was conducted by extracting a slice

image from the 3D model. The 2D mesh was directly
Multiphase composite materials, such as concrete,
generated by considering every pixel as a quadrilateral
are widely used in engineering structures. Due to
element. For each square grid, the four corner coordinates
random distribution of multiple phases from nano-,
can be calculated by the corresponding positions in the
micro-, meso- to macro-scales, multiphase composite
image. The 4-node cohesive elements with zero in-plane
materials have intrinsically heterogeneous nonlinear
thickness were implemented by a self-programmed Matlab
mechanical properties, which in turn directly determine
code. Figure 2 shows the 2D mesh and the simulation
the performance and reliability of structures and systems.
results under tension load. The following research will be
Therefore, understanding their mechanical properties
concentrated on the 3D simulations and validation with the
including damage and fracture at different scales, through
XCT images.
both experimental studies and computational modelling,
becomes one of the most critical and challenging
engineering and scientific problems [1, 2].
A better understanding of composites’ mechanical
This project aims to develop an innovative Multiscale behaviour, especially on fracture and damage, can
Stochastic Fracture Mechanics (MustFrame) strategy for lead to development of materials with higher loading
accurately modelling damage and fracture in concrete resistance, cost-effective manufacturing processes and
and critically evaluating the reliability and performance of optimal structural designs. This understanding promises
normal- to large-scale structural systems, with an ultimate to contribute a substantial impact to the improved
view of optimising the material microstructure for better economy and sustainability of our society. It is vital that
quality, greater reliability and lower cost. modern structures are designed efficiently and with
long term durability and serviceability uppermost in the
Description of method and results design process to meet society’s needs. An enhanced
understanding of the behaviour of multiphase composite
In-situ 3D images of small-sized specimens under
materials is a clear way forward to achieve this goal.
continuous loading steps, showing internal micro/meso-
structures and evolution of damage and fracture, will
Simulations on real normal/large-sized structures by using
be obtained by the state-of-the-art X-Ray Computed
aforementioned MustFrame can be used to assess their
Tomography (XCT) technique. The detailed internal
structural integrity and reliability, which can also make
structure can be characterised and visualised directly
recommendations of modifications to modern structural
from XCT images. Then the 3D finite element meshes
design codes based on the research outcomes.
of real structure are then automatically transformed to
model different phases and interfaces. Complicated crack
References
propagation will be modelled by embedding cohesive
interface elements into these finite element meshes [3, 4]. [1] Oden JT et al, (2003), Research directions in
The numerical models will then be directly validated and computational mechanics. Computer Methods in
calibrated by comparing the modelled failure process with Applied Mechanics and Engineering 192:913-922.
3D XCT images.
[2] Kassner ME et al., (2005), New directions in
Concrete cube specimens of size 40 mm were tested mechanics. Mechanics of Materials 37:231-259.
under progressive loading. The in-situ experiment was
carried out at the Manchester X-ray Imaging Facility (MXIF), [3] Yang, Z.J., Su X.T., Chen J.F., Liu G.H., (2009),
the University of Manchester. The initial segmentation and Monte Carlo simulation of complex cohesive fracture
visualisation procedure was finished. The 3D segmented in random heterogeneous quasi-brittle materials.
aggregates, cement, cracks and voids, and the combined International Journal of Solids and Structures. 46:3222-
concrete of the specimen without load can be shown in 3234.
Figure 1, which indicates the realistic random shapes and
positions of each phase in concrete. The fraction of three [4] Su X.T., Yang Z.J. Liu G.H., (2010), Monte Carlo
phases can be calculated, among which, the voids and simulation of complex cohesive fracture in random
cracks fraction reflects the evolution of the cracks under heterogeneous quasi-brittle materials: A 3D study.
compression. International Journal of Solids and Structures. 47:2336-
2345.
Funding body
Royal Society, EPSRC
Further information
Wenyuan Ren
(E: wenyuan.ren@postgrad.manchester.ac.uk) or
Dr Zhenjun Yang (E: zhenjun.yang@manchester.ac.uk)
Fig 1: 3D Segmented phases without load Fig 2: 2D Simulation results
A novel, cost effective, and simple predictive test for
design and optimisation of fire-safe concrete mixes
for tunnel linings
57 leuan Rickard
BRE Centre for Fire Safety Engineering, University of Edinburgh
Background • To produce rational guidelines for the design of

optimised, high-performance, self-consolidating
Explosive spalling of concrete is a phenomenon in which
concrete mixes which can be used to limit, or
the concrete surface breaks off violently as a result
preferably completely eliminate, explosive spalling from
of internal stresses caused by thermal gradients and
concrete tunnels in a fire of any credible worst case
rising pore pressures on heating during a fire. Spalling is
scenario (or other types of concrete structures).
notoriously difficult to predict for design as it is affected
by many different and interrelated factors, including
Methodology
aggregate type, moisture content, mix design, age,
strength, mechanical stress, size, etc. of the concrete as The project will test many different candidate concrete
well as the thermal gradient and time history of absorbed mixes which will be carried out to establish their
energy experienced in a fire. In tunnel fires, explosive propensity for spalling under a range of thermal, hygral,
concrete cover spalling often governs the design and and mechanical conditions. This type of characterisation
specification of the thickness of concrete cover to the represents the practical limit for design purposes for the
steel reinforcement and/or steel prestressing tendons; as design fires which are applicable to tunnels. The eventual
a result this potentially drives the cost of a tunnel project goal is to develop both a predictive test method and a
up and has considerable adverse effect on its embodied mix design procedure to ensure that no spalling occurs
energy use, carbon footprint, and sustainability credentials. whilst still meeting the other mechanical and rheometric
Uncertainties around the degree of spalling likely to be properties required of a concrete mix for use in tunnelling
experienced during a potential tunnel fire are also a major applications. The study can be broken down into smaller
risk element during design, which expands as a project component parts:
continues.
(1) Background reading and literature review
Micro-polypropylene (PP) fibres can be added to the fresh
concrete mix during mixing and casting to reduce spalling (2) Development of test procedures, adapting a pre-
once the concrete has set, however this may reduce existing heat transfer rate inducing system (H-TRIS)
the overall strength and workability of the concrete mix – developed previously at the University of Edinburgh
and thus needs to be considered during concrete mix by PhD student Cristian Maluk – so as to be able to
design and materials specification. Passive protection replicate different design fire curves and possibly to test
(i.e. supplemental fire protection applied to the exterior larger samples .
of a concrete tunnel lining element) is often prohibitively
expensive, and addition of PP fibres is now generally (3) Development of mix designs and casting and testing
accepted in industry as best practice. However, as of two sets of samples. Making the second set of
concrete mixes with PP fibres are often proprietary and samples after testing the first set will allow the tests to
unique, any contractor wishing to use PP fibres in a given be modified to best investigate the variables which are
mix or given project must ‘prove’ that the specific mix deemed as most important after initial testing
being proposed for use has low-spalling properties. This
currently requires costly, time consuming, and uncertain (4) Testing of the samples using H-TRIS. If possible it
large-scale furnace tests on large scale samples of the would be desirable to be able to test identical samples
proposed tunnel lining. As a result, passive protection using both the H-TRIS method and a standard furnace,
is often used at an additional cost and project risk. This as this would enable at least partial validation of the
research project seeks to begin to address many of these experimental method and testing outcomes
issues.
The project will link closely with partners Arup and Propex
Aims and objectives to ensure that the outcomes of the research are of practical
use in industry. Close collaboration is key to gaining an
The aims of the project are:
understanding of the main constraints and drivers when
designing a concrete mix for a tunnel lining and ensuring
• To develop and validate a new test method which
that the research is relevant.
can be used for the prediction of the propensity for,
and extent of, explosive spalling due to fire (or high
Funding body
temperature) in tunnel linings or related underground
structures constructed in concrete; EPSRC Industrial CASE award, Arup
• To create a wealth of empirical knowledge on the Further information

factors influencing explosive spalling in concrete tunnel
Ieuan Rickard (E: I.Rickard@ed.ac.uk) or
linings; and
Dr Luke Bisby (E: luke.bisby@ed.ac.uk) and
Dr Tim Stratford (E: tim.stratford@ed.ac.uk)
Seismic performance of reinforced concrete beam-column
joints strengthened in shear with embedded FRP bars
Ridwan
University of Birmingham 58
Background Project aims and objectives
Beam-column joints play an important role in the seismic This project aims to provide a fundamental understanding
resistance mechanism of some existing reinforced concrete of the seismic response of shear critical exterior RC beam-
(RC) frame structures. Without proper design and detailing column joints strengthened using the ETS FRP system. The
of the joint shear reinforcement, as the case is with RC objectives of this project are:
buildings designed according to the pre-1970s building
codes, RC beam-column joints can be the most vulnerable (1) To identify the effect of axial column load and FRP
elements during an earthquake and can undergo joint reinforcement ratio on the seismic response of RC
shear failure. beam-column joints.
Several techniques; e.g. RC jacketing, steel jacketing, and (2) To develop finite element (FE) predictive tools for the
steel plate bonding; have been proposed for improving behaviour of the tested specimens.
the seismic performance of RC beam-column joints.
However, these techniques have been criticised for being (3) To investigate the accuracy of the current FRP design
labour intensive, increasing the dead load of the structure guidelines and to improve their predictions where
(RC jacketing), and being susceptible to corrosion (steel needed.
jacketing and steel plate bonding). On the other hand,
strengthening techniques utilising advanced composite (4) To enrich the general understanding and develop a test
materials, such as fiber-reinforced polymers (FRPs), have database of RC beam-column joints strengthened with
high strength to weight ratio, high corrosion resistance, no the ETS FRP system.
significant increase in size of member, easy applicability,
and limited disruption to building occupancy (Engindeniz et Research methodology
al. 2005).
In order to meet the objectives outlined above, two
approaches will be followed.
Externally bonded FRPs have gained popularity in the
retrofitting of existing RC structures. However, experimental
(1) Lab testing: RC beam-column joints with varying axial
results have shown that FRP de-bonding remains the
column loads and FRP reinforcement ratios will be
main drawback preventing the utilisation of the high tensile
subjected to reversed-cycling loading.
strength of the FRPs. This failure mode is related to the
relatively low tensile strength of the concrete surface that
(2) Numerical modelling: the Finite Element Method
limits the bonding force between the FRP and concrete
will be used to simulate the behaviour of the tested
(Mofidi and Chaallal, 2011).
specimens.
A new method, utilising embedded through section
(ETS) FRP bars, for strengthening RC beam-column
joints deficient in shear will be investigated. Experimental The concepts encompassed in this work will contribute
investigations on strengthening RC beams deficient in to our general understanding of the seismic performance
shear using the ETS method were successfully conducted of RC beam-column joints strengthened in shear with
by Chaallal et. al (2011), Mofidi et. al (2012) and Godat FRPs. The combination of lab testing and FE modelling will
et. al (2012). The test results confirmed the applicability of provide a detailed description of the structural behaviour
the ETS method and revealed that the performance of the and ensure an integrated modelling approach that can feed
strengthened beams was significantly superior compared directly into the development and/or updating of design
to that of beams strengthened with externally bonded or guidance.
near-surface mounted FRP reinforcement (Chaallal et.
al 2011). The ETS method needs less adhesive and is
less susceptible to debonding compared to other FRP
strengthening systems.
Chaallal, O., Mofidi, A., Benmokrane, B. and Neale, K. Directorate General of Higher Education, Ministry of Culture
(2011) ”Embedded Through-Section FRP Rod Method and Education, Republic of Indonesia
for Shear Strengthening of RC Beams: Performance
and Comparison with Existing Techniques.” J. Compos. Further information
Constr., 15(3), 374–383.
Ridwan (E: rxx211@bham.ac.uk) or Dr Samir Dirar (E:
s.m.o.h.dirar@bham.ac.uk)
Engindeniz, M., Kahn, L.F. and Zurieck, A. (2005) “Repair
and strengthening of reinforced concrete beam-column
joints: State of the art.” ACI Struct. J., 102(2), 187–197.
Godat, A., L’Hady, A., Chaallal, O. and Neale, K. (2012)

”Bond Behavior of the ETS FRP Bar Shear-Strengthening
Method.” J. Compos. Constr., 16(5), 529–539
Mofidi, A. and Chaallal, O. (2011) ”Shear Strengthening

of RC Beams with EB FRP: Influencing Factors and
Conceptual Debonding Model.” J. Compos. Constr., 15(1),
62–74.
Mofidi, A., Chaallal, O., Benmokrane, B. and Neale, K.

(2012) ”Experimental Tests and Design Model for RC
Beams Strengthened in Shear Using the Embedded
Through-Section FRP Method.” J. Compos. Constr., 16(5),
540–550
Collapse of bridges founded on piles in liquefiable soil:
importance of proper representation of soil springs
Mehdi Rouholamin
University of Surrey 59
Poster presenter
Project objectives and goals if the shape of the p-y curve is that of the “S-curve”
(proposed by Dash 2010) as shown in Fig.1-2b; the pile
It has been observed that many bridges founded on pile
response as described above for small and large amplitude
foundations have suffered during strong earthquakes.
vibrations will alter significantly. The lack of initial stiffness
There are many examples of failures e.g. Luanhe Bridge
and strength of the liquefied soil will increase the P-delta
(Tangshan Earthquake, 1976), Pontleo Gallardo Bridge
effect in the small amplitude vibration, and will promote
(Chile Earthquake, 1985), and Minjiang Bridge (Wenchuan
the buckling mode of failure of the piles. However, the
Earthquake, 2008). By studying these collapses, the
advantage achieved by using the later model (Fig.1-2b) is
following hypothesis may be drawn on the failure
the higher strength and stiffness at large differential pile-soil
mechanism of the bridge structures. Middle of the bridge
movement, which may prevent a complete collapse of a
has low stiffness and hence the bridge starts to fail from
structure.
the middle during strong shaking i.e. more prone to
higher deflection making it vulnerable to unseating. This
In this paper, empirical p-y curve from API 2000 and
hypothesis may be explained with the help of the following
mechanism based p-y curve (Dash 2010) have been
Fig. 1-1. Owing to the river profile, the unsupported length
employed for studying behaviour of piles in liquefying soil.
of pile above the river bed at the middle span of the bridge
For this study, pile details and a soil profile of ShowaBridge
(L1m) is generally greater than the unsupported length of
have been adopted.
pile at other spans of any bridge (L1s). Generally, it is also
observed that the liquefaction profile through the section of
Results
the river does not alter much. Hence for practical purpose
liquefaction depth at the middle span (L2m) equals to the Low amplitude vibration
liquefaction depth at other spans (L2s) of the bridge. It is
Harmonic base acceleration of amplitude 0.03g was
also observed that the piles of similar lengths are used
applied for six different frequencies for the two models
typically throughout the length of the bridge. So it is evident
developed with p-y springs obtained by empirical
that the unsupported pile length of the middle span of the
and mechanism based respectively. The pile head
bridge (L1m+L2m) is generally higher than the unsupported
displacements for the two cases are compared in Fig.
length of piles at other spans of the bridge (L1s+L2s) making
2-1. It may be observed that the pile head displacements
the piles at middle span of bridge more flexible than the
increase when the excitation frequency falls in the range
other spans of bridge. As a result, for a similar lateral load,
of 1-3 Hz. Again it may be observed that the pile head
the piles in the middle will have higher deflections i.e.
displacement increases significantly for the model with
by definition low stiffness. This constitutes a necessary
springs obtained from Dash 2010 due to the lack of initial
condition to unseat the bridge deck initiating collapse.
stiffness of those springs as is evident from Fig. 2-3. This
increases the P-delta effect and promotes a buckling mode
of failures in the piles. Due to this the deflection required to
With this hypothesis in mind, numerical analyses have been unseat the bridge deck may cross the allowed limit initiating
carried out to analyse the pile-soil interaction. In practice, the collapse of the bridge.
laterally loaded piles are most often modelled as beams on
non-linear Winkler foundation, also known as “p-y spring” Large amplitude vibration
approach. The stiffness of these p-y springs are generally
In this case, harmonic base acceleration of amplitude
obtained empirically. A reduction factor or a p-multiplier
0.3g was applied for six different frequencies for the two
is generally applied on the p-y curve for the non-liquefied
models with different p-y springs. The comparison of pile
condition to obtain the p-y curve liquefied soil condition.
head displacements are shown in Fig. 2-2. Here also it
Since proper representation of soil-pile interaction is
is observed that the pile head displacement resonates
important for determining the dynamic properties of the
in the frequency range of 1-3 Hz establishing the fact
soil-pile system, a failure mechanism based p-y curve for
that frequency of the excitation plays an important role
liquefying soil condition (Dash 2010) is employed in the
in determining the behaviour of the soil-pile systems.
study. The importance of adopting proper mechanism
In contrary to the low amplitude vibration, the pile
based p-y curve over empirical p-y curves in predicting
head displacement is more for the model with springs
deflection for performance based design are highlighted.
obtained from API 2000. This may be explained with
the higher strength and stiffness of the p-y springs with
Fig.1-2 shows the possible response of a pile due to the
the mechanics based approach (Dash 2010) at higher
two types of p-y curves. For the empirical curve, when the
displacement values as also evident from Fig. 2-3. Hence it
differential soil-pile movement is small (i.e., the soil is not
may be stated that the mechanics based p-y springs may
pushed to its full capacity), the resistance on pile depends
be advantageous for large differential pile-soil movement
on the initial stiffness of the soil and the value of deflection
and may also prevent complete collapse of the bridge
(Fig.1-2a). When the differential soil-pile movement is large,
structure.
the resistance offered by soil to the movement of the pile
is governed by the ultimate strength of the soil. However,
Potential for application of results Further information
These results can be used to design pile foundations in Mehdi Rouholamin
liquefiable soils in future. (E: m.rouholamin@surrey.ac.uk,
Mehdi.rouholamin@gmail.com) or
References Prof Subhamoy Bhattacharya
(E: subhamoy.bhattacharya@googlemail.com)
American Petroleum Institute (2000), Recommended
practice for planning, designing and contracting fixed
offshore platforms-Load and resistance factor design, API-
RP-2A, 21st Edition.
Bhattacharya, S. (2003). Pile instability during earthquake

liquefaction. PhD thesis, University of Cambridge, UK.
Bhattacharya, S., Madabhushi, S.P.G., (2008), A critical

review of methods for pile design in seismically liquefiable
soils. Springer, Bull Earthquake Eng(6). 407-446.
Dash S. R. (2010), Lateral Pile-Soil Interaction in Liquefiable

Soil. Ph.D. thesis, University of Oxford, UK. SAP2000
Software, Structural analysis program, Computers and
structure Inc.
Fig 1: Typical sectional view of pile supported bridge and Fig 2: Comparison of pile head displacements in different
two types of p-y curves amplitudes and comparison of p-y curves
Investigation into the resilience of concrete flat slabs
against progressive collapse
Justin Russell
University of Nottingham 60
Oral presenter
Project objectives and goals Further studies into the influence of the nonlinear effects
experienced at high deflections have been conducted
Ever since the collapse of Rowan Point tower building in
with ABAQUS [3] to model the slab more accurately. This
1968 there has been significant interest in progressive
has shown that both material and geometric nonlinearity
collapse of structures after a relatively small damaging
play a significant role in providing additional capacity to a
event. Current approaches for conducting such analysis
damaged structure. These primarily are due to the post
allow a choice between a simplified linear assumption
yielding strength of the steel and the formation of a tensile
for material and geometric behaviour or a full nonlinear
membrane in the slab. Therefore a structure may be able to
analysis, as well as either a static or dynamic simulation [1].
maintain integrity and prevent complete collapse.
The aim of this project is to investigate the influence of the
nonlinear and dynamic effects in the analysis of a flat slab
To validate the numerical model a series of experimental
structure after a column loss scenario. To achieve this aim
tests on scaled slab elements will be conducted. The
the following objectives will be completed:
primary goals of these tests is to determine the dynamic
influence on the slab after a column is removed suddenly
• Conduct numerical simulations of column loss
from a loaded floor compared to a static increase in loading
scenarios to consider the potential for progressive
and to observe the damage and failure modes of a slab
failure
after loss of a support.
• Collect data from scaled RC slab elements during
Four series of tests will be carried out with different
a column loss scenario for both static and dynamic
configurations, either removing a corner or central edge
conditions
column and with different reinforcement layouts. Each test
series will include a static push down test and a dynamic
• Validate the numerical model against the results from
test where full load is applied and then a column removed
the experimental tests
and the response recorded. Figure 2 shows an example
of typical test. Reaction forces, deflections, strains at key
• Analyse a range of structural layouts and designs
locations and crack patterns will be recorded throughout
under sudden column loss with different analysis
the experiment. A comparison of the results will provide
approaches
information about the inertial effects and a dynamic
amplification factor can then be assigned. Furthermore,
information can be gained into the additional capacity after
Studies into the behaviour of flat slab structures after cracking occurs and the yield strain has been reached.
column removal scenarios were conducted using the finite
element analysis software OpenSees [2], which is capable Finally, a complete parametric study will be conducted for
of modelling large structures under dynamic loading a range of typical structural configurations with different
conditions. This demonstrated that the redistribution of analysis approaches, i.e. Linear Static, Nonlinear Static,
forces after a column loss can subject the structure to Linear Dynamic and Nonlinear Dynamic. A comparison
conditions far beyond its design limits, either in flexure or between results will then be made.
punching shear, from this it is clear that both the demand
and the capacity need to be determined accurately in order Potential for application of results
to make assessments for progressive failure.
Conclusions from this work will provide valuable data
for designing new structures and retrofitting old ones to
Investigations into the influence of dynamic effects have
ensure they are protected against extreme events such as
shown that there can be an increase of up to 1.8 times
column loss. By understanding the influence of nonlinear
in the deflections and forces experienced in the structure
and dynamic effects, safer and more efficient designs can
compared to a static loading condition, depending on the
be made and simplified analysis can be conducted where
configuration and structural properties. Figure 1 shows the
appropriate.
variation of the Dynamic Amplification Factor (DAF) for four
removal conditions against different span/depth ratios. As
the structure becomes more flexible, i.e. higher span/depth
ratio, the dynamic effects become more critical.
[1] GSA. Progressive collapse analysis and design Justin Russell (E: evxjr@nottingham.ac.uk) or
guidelines for new federal office buildings and major Dr John Owen (E: john.owen@nottingham.ac.uk)
modernization projects. Washington (DC, USA): The US
General Services Administration; 2003.
[2] Mazzoni S, McKenna F, Fenves G. OpenSees

command language manual. Pacific earthquake
engineering research. PEER center. 2005.
[3] ABAQUS. (2010) Analysis User’s Manual Version 6.10.

Inc. and Dassult Systems, Providence, RI
Fig 1: Dynamic Amplification Factor for different conditions.
Fig 2: Photos of typical slab, a) before test b) after load has been applied
Vibration serviceability assessment of structures under
intelligent interactive pedestrian loading
Erfan Shahabpoor
University of Sheffield 61
Poster presenter
Brief description As a major part of this investigation, a mechanical model
of human body has to be developed, capable of simulating
Vibration serviceability problems have been around for
human-Structure interaction (HIS) effects especially in
many decades. Recent advances in design methods and
vertical direction. An extensive set of experiments and
tendency towards using more slender elements, longer
simulations is being done to find a proper model, applicable
span bridges and open plan offices have intensified the
to different loading scenarios and structure types.
problem and have made structures significantly susceptible
to excessive vibration due to human activities such as
On the experimental side, the project includes designing
walking, bouncing and jumping.
and manufacturing a unique data acquisition system called
‘Pedestrian Motion Monitoring System (PMMS)’, capable
Complex kinematics of human motion, stochastic nature
of collecting a complete set of crowd and structure data
of its loading and unknown mechanisms of its interaction
on full scale structures in real-time. Data collection involves
with structure and surrounding environment have
motion tracking of people in crowd, pressure mapping of
significantly limited our understanding of this phenomena
the interaction surface and recording full set of dynamic
and consequently the level of confidence on available
response of structure. Parallel to full-scale structure
design methods. Engineers today face a serious problem,
tests, experimental investigation of kinematics of human
attempting to design a pedestrian structure, as none of
body using visual motion tracking and pressure mapping
current design guidelines can provide reliable methodology
technologies at the same time have provided a great
for vibration serviceability assessment of structures under
insight on underlying interaction mechanisms.
crowd loading [1]. Although considerable research has
been focused today on developing a better understanding
In the next step, a sensitivity analysis is being done on
of the nature of individuals and crowd loading and their
different variables and their statistical distributions to find
interaction with structure, but still none of proposed models
the level of accuracy required for each variable. Based on
is able to simulate properly pedestrians’ dynamic effects on
sensitivity analysis and simulations results, this complex
structure in vertical direction.
system is being converted into a practicable method
for engineers to use in everyday designs. The proposed
Methodology
method will take into account the dynamics of pedestrians
This interdisciplinary research tends to investigate the in a crowd in a dynamic analysis of the structure.
walking-induced dynamic effects of pedestrian traffic
on footbridges in vertical direction by employing a novel Objectives
methodology. The proposed methodology considers
The aim of this combined analytical and experimental
pedestrians as “intelligent” elements of a complex dynamic
research is to investigate thoroughly the pedestrian traffic
system, interacting with itself, structure and its surrounding
walking-induced interactive effects in vertical direction,
environment (including other agents) in real-time.
to develop a mechanical model of human body capable
of addressing human-structure interaction in vertical
The analytical part of the investigations involves modelling
direction and finally to propose a codifiable, practically
intelligent humans, structure and environmental elements
implementable and experimentally verified design
and putting them in interactive plasma to live together.
methodology for modelling groups of people walking on
This process includes mathematical modelling of human
structures.
body kinematics based on biomechanical principals,
FE modelling of structures and updating models using
Potential application for results
experimental data from dynamic tests of full-scale
structures, simulating human behaviour using Artificial Recently built footbridges are increasingly exposed to
Intelligence (AI) and machine learning principals and control vibration serviceability problems. The results of this project
algorithms (hybrid control mostly combination of semi- will work towards developing design guidance so that
active and active control methods), and finally simulating design engineers can confidently model pedestrian traffic
the whole system interactions in real-time, using agent- on footbridges during the design stage and can realise
based methodology. Monte Carlo simulation is being used the potential slenderness benefits that are available.
to generate the pedestrian traffic according to the collected Furthermore, this work will also contribute to understanding
experimental data. better the behaviour of pedestrians in crowds/groups/
streams and their active interaction with structures and
their surrounding environment.
Reference
[1] Shahabpoor, E., Pavić, A., Comparative evaluation
of current pedestrian traffic models on structures,
Conference Proceedings of the Society for
Experimental Mechanics Series. V 26, pp 41-52. 2012.
Further information
Erfan Shahabpoor (E: e.shahabpoor@sheffield.ac.uk) or
Prof Aleksandar Pavic (E: a.Pavic@sheffield.ac.uk)
Fig 1: Conceptual illustration of using bipedal IP model for simulating HSI
Ultra-thin toppings for upgrading existing timber floors
Jonathan Skinner
BRE Centre for Innovative Construction Materials, University of Bath 62
Project objectives and goals The change in the duration of the vibration response was
investigated by identifying the mechanisms by which
In the UK, timber-concrete composites are not an
energy can be dissipated. The most influential mechanism
established construction technique. It is also true that
arises from the shear connectors joining the topping and
designed solutions for refurbishing existing structures are
timber together. As the connectors are subjected to cycles
not as commonplace as for new structures. Research
of loading they exhibit a hysteretic load-displacement
at the University of Bath is countering this situation by
response. The area of the hysteresis is expressed as an
developing a thin cementitious topping upgrade for existing
equivalent viscous damping ratio to evaluate the damping
timber floors. As timber floors often have inadequate
in a single T-beam.
serviceability performance the purpose of the upgrade is to
stiffen and solve vibration problems.
Panels and floors were constructed to validate the
parametric study (Figure 2). They were loaded dynamically
A topping thickness of 20mm or less has been targeted
before and after the topping was added to establish the
so as to reduce the requirement for a floor to be propped
change in vibration performance and then loaded to
during construction, reduce the load added to the existing
failure. Results show that the addition of the topping leads
structure and limit the change in finish floor to ceiling
to an increase in stiffness of 60%, improved dynamic
height. This poses technical challenges such as preventing
performance and a desirable failure mechanism.
shrinkage and cracking of the topping and establishing an
efficient mechanical connection between the timber and
Application
topping.
It is observed that older floors are most likely to require
Methods and results upgrade and in particular in buildings which have
been converted from single to multiple dwellings as
The human perception of footfall induced vibration is
the occupants have a heightened awareness of their
determined by the frequency, magnitude and duration of
neighbour’s activities. The market for this technology is
the response. Vibrations of high frequency, low magnitude
substantial, of the existing housing stock that has been
and short duration are generally considered least
converted to apartments, 85% predates 1919 [1].
perceptible by occupants. Whilst it might be considered
better to have a very thick topping to reduce the magnitude
Whilst this research is directed towards upgrading existing
of the response, this viewpoint does not consider that
floors it still has implications for new timber-concrete
even very thin toppings have a much greater mass than
composites. Vibration response of these floor types has not
a timber floor and as such thin toppings are sufficient. To
previously been well investigated and this research will be
understand the topping thickness at which the greatest
important to the formation of new design guidance.
increase in frequency could be realised, a parametric
study has been undertaken. It considers how the change
References
in frequency (equation 1) varies as the topping thickness
increases from 0 to 100mm for different levels of composite [1] Department of Communities and Local Government.:
action. Figure 1 demonstrates that the most benefit is English House Condition Survey-Annual report. HMSO,
gained when the topping is thin (less than 20mm) and there London, UK, 2007.
is complete composite action between the timber and
topping (γ=1). Funding body
BRE Trust
∆k
fn ∝ (1)
∆m
To achieve composite action shear connectors join the
topping and timber together to resist slip at the interface of
the materials. Possible types of connector were identified
and tested, and the results showed that inclined screws
were the optimal connector for this application. Further
pushout testing of the screw connectors demonstrated
that the inclination of the screw and the thickness of the
topping are both important factors in the stiffness and
failure mode of the connection.
Further information
Jonathan Skinner (E: J.M.Skinner@bath.ac.uk) or
Dr Richard Harris (E:R.Harris@bath.ac.uk)
Fig 1: Change in frequency with respect to topping depth

and composite action
Fig 2: Section of an upgraded floor
Punching shear of reinforced concrete slabs in fire
Holly Kate McLeod Smith

Oral presenter
Project objectives and goals To apply the DIC technique in 3D, I spent two months
working with Dr Andy Take’s DIC research group at
This project examines the mechanisms through which
Queen’s University, Canada. This allowed me to study their
punching shear is transferred in reinforced concrete slabs
application of the DIC method, to examine the potential
during fire. The goal is to better understand the transfer
errors when using the method, and specifically to address
mechanisms and their interactions to improve shear design
errors caused by camera lens distortion, which was
codes for high temperature.
highlighted as significant during my previous undergraduate
testing. This allowed the existing DIC method to be
Research on reinforced concrete structures in fire has yet
adapted to study the behaviour of concrete slabs in fire.
to embrace many of the phenomena that are now exploited
during the performance-based design of steel structures
The tests are currently underway and preliminary results will
for fire. However, there is growing recognition that effects
be presented including the load-deflection response, failure
such as restrained thermal expansion and large deflection
loads, and in particular the effects of edge restraint.
load-carrying mechanisms need to be understood for
concrete structures to ensure safety and to allow efficient
fire design [1, 2].
The fundamental knowledge of punching shear transfer
A particular topic that needs to be addressed is the in a reinforced concrete slab during fire, will provide
punching shear behaviour of concrete flat slabs in fire, industry with design and assessment methods to prevent
highlighted by the failure of the Gretzenbach car park in catastrophic shear failure and improve public safety.
Switzerland. Shear in concrete at normal temperatures is
notoriously complex and remains the subject of research, References
and hence it is no surprise that little is known about shear
[1] Annerel, E., Lu, L., & Taerwe, L. (2012). Punching shear
in concrete subjected to elevated temperatures and
tests on flat concrete slabs exposed to fire. Fire Safety
thermal gradients. The shear behaviour of concrete in fire
Journal. Retrieved from http://dx.doi.org.ezproxy.
is dependent both upon material properties at elevated
webfeat.lib.ed.ac.uk/10.1016/j.firesaf.2012.10.013.
temperature and the effect of in-plane actions that result
from restrained thermal expansion.
[2] Bamonte, P., Fernández Ruiz, M., & Muttoni, A. (2012).
Punching shear strength of R/C slabs subjected to
fire (pp. 689–698). Presented at the 7th International
Fifteen model slab-column reinforced concrete slabs Conference on Structures in Fire.
(1400 × 1400mm) are tested in an inverted orientation
in a purpose built frame that allows the edge restraint [3] Gales, J. A., Bisby, L., & Stratford, T. (2012). New
conditions to be modified, and the restraint forces to be parameters to describe high-temperature deformation
characterised. The specimens are heated by a six-panel of prestressing steel determined using digital image
array of radiant panels from above, allowing greater control correlation. Structural Engineering International, 22(4),
of the applied heating than would be possible using 476–486.
traditional furnace testing.
Funding body
The tests are instrumented with temperature-adjusted
EPSRC DTA award, Royal Society of Edinburgh Lessells
strain gauges on the frame to characterise the restraint
scholarship.
forces, thermocouples within the specimens and around
the frame, and deflection measurement using a bespoke
Further information
3D digital image correlation method, which will allow
displacements and strains across the lower slab surface of Holly Smith (E: H.Smith@ed.ac.uk) or
the slab to be captured. Dr Tim Stratford (E: Tim.Stratford@ed.ac.uk) and
Dr Luke Bisby (E: Luke.Bisby@ed.ac.uk)
The use of digital image correlation (DIC) to record
displacements is motivated by the need for a high spatial
density of deflection measurements to investigate the shear
behaviour, coupled with DIC’s resilience to temperature
effects (unlike strain and displacement gauges). Three
off-the-self digital cameras are used simultaneously to
determine the deflected shape using an academic image-
processing algorithm written in MatLAB [3].
Fig 1: Slab two-way edge restraint testing frame
Improving design efficiency of high rise structures in
mitigating the potential of progressive collapse
David O Stephen
University of Leeds 64
Progressive collapse investigations gained a widespread checks were carried out to ensure that the connections
recognition after the partial collapse of Ronan Point building can withstand a tensile force equivalent to the shear force
in 1968 in London as a result of domestic gas explosion. since the shear force significantly exceeds the minimum
[1] Other landmark events highly referenced in existing recommendation of 75KN tensile force as recommended in
literature are the collapse of Alfred Murrah building in 1995 Eurocode 3.
and the collapse of the World Trade Centre in 2001 all in
the United States. Following these collapses, extensive The corner column, interior column, edge column and the
research investigation in mitigating and limiting progressive corner column at the eight floors were removed suddenly
collapse has been the focus of researchers. Some to investigate the structural response. The changes in
researchers develop strategies in simplifying the analytical stress resultants (shear, catenary force) due to sudden
methods of investigations considering the computational removal of column were recorded. The connections initially
resources required for complex non-linear dynamic designed to within static loading conditions are subjected
analysis. [2] to both static loads and additional forces (catenary force)
developed due to sudden column loss scenario.
High rise structures are often designed to withstand
gravity loading, wind and snow loads depending on the Potential for application of results
geographical location and design intent. In recent times,
Structures having social, economic, and political influence
global terrorism posed a great challenge to the engineering
are basically the prime target of terrorist in making an
community in the choice of frame structures particularly for
impact. The performance of connections under abnormal
buildings with high social, economic, and political influence.
loads determines the survival time of the structure under
Abnormal loads could either be a consequence of natural
progressive collapse. The connection was designed based
disaster or may be due to explosives, bomb detonations,
on static loading and then investigated under progressive
impact, or extreme fire. In view of this, codes and
collapse scenario. The results/design recommendation
design guidelines around the world recommends threat
could serve as a basis for structural design consideration
independent method of analysis and design. Localised
under progressive collapse by integrating connection
failures may trigger complex redistribution of stresses within
performance into design requirements for mitigating
the structural system over a short period of time. [3] This
progressive collapse.
paper is aimed at investigating the performance of simple
connections commonly used in construction industry
References
under progressive collapse scenario and to determine the
structural response of high rise steel structures due to [1] Pearson, C. and N. Delatte, Ronan Point Apartment
progressive collapse. Tower Collapse and its Effect on Building Codes.
Journal of Performance of Constructed Facilities, 2005.
Project objectives and goals 19(2): p. 172-177.
• The aim of the investigation was to assess the
[2] Vlassis, A.G., et al., Design Oriented Approach for
potential of simple connections under progressive
Progressive Collapse Assessment of Steel Framed
collapse scenario.
Buildings. Structural Engineering International, 2006.
16(2): p. 129-136
• To investigate the stress resultants ( shear, catenary
force) developed during progressive collapse scenario
[3] Stephen, O.D, Jianqiao Ye, Lam D. Progressive
collapse evaluation of sudden of high rise structures
• To Investigate the behaviour of tie beams under
due to sudden loss of structural members; in proc.
progressive collapse scenario
Of the 10th Intl. Conf. on Advances in Steel Concrete
Composite and Hybrid Structures, Singapore 2012, pp.
• Finally, parametric evaluation of connections under
1179-1186
progressive collapse scenario
Funding body
Petroleum Development Technology Fund (PTDF) Abuja,
A 10 storey braced frame structure was used for this
Nigeria
investigations. The span along the Y-axis is 6m while the
span along the x-axis is 4.5m. The building has a constant
Further information
storey height of 3.5m with a slab thickness of 130mm.
Static analysis and design of the structure was carried out David O. Stephen (E: cen4dos@leeds.ac.uk) or
using Eurocode 3 built up in SAP 2000 finite element code. Prof Dennis Lam (E: d.lam1@bradford.ac.uk)
The maximum shear force obtained at the connections
was recorded and simple connections (flexible end plate
and angle cleat connections) were designed to satisfy
the provision of Eurocode 3: part 1-8. Further design
Properties of low cost concrete with recycled aggregate
and tyre rubber for high speed train lines
65 Haolin Su
University of Birmingham
Poster presenter
Project objectives and goals (3) Effect of cement type. The rubber inclusion has a
more substantial influence on PFA concrete workability
Recycling scrap tire rubber and recycled aggregate for
than it does on that of the GGBS concrete due to
sustainable development, reducing the capital cost of
the fact that the GGBS particles has smooth, glass-
constructing ballast-less track and reforming the durability
like surfaces. GGBS concrete is more affected by
of concrete slab track.
the inclusion of rubber than PFA concrete in terms of
compressive strength, splitting tensile strength and
flexural strength. Permeability dramatically increases
In this project, sand was replaced with scrap tyre rubber for GGBS concretes containing rubber.
by different content and different shape, crumb and shred;
natural aggregate was replaced with recycled aggregates (4) Effect of recycled aggregate. With the increase of
by different content; three types of cement were used to recycled aggregate from 0% to 100%, slump tends to
compare and polymeric fibre was used as well to improve decrease. Strength suffers when natural aggregates
the strength. Tests were carried out and so far, the results are replaced by recycled aggregates. There is a drop
and conclusion obtained are listed as follows: of 17.2% in compressive strength at 28 days when
recycled aggregate is 100%. Splitting tensile strength
(1) Effect of rubber content and particle size. With the and flexural strength are not affected as much as
increase of rubber content and rubber size, the slump compressive strength. With the increase of recycled
tends to decrease. When rubber replacement is aggregates, air permeability, water absorption and
over 20%, the slump reduces severely. Fresh density water permeability indices tend to increase.
decreases with the increase of rubber content. The
extent of decrease depends more on the replacement (5) Effect of recycled aggregate and rubber particles.
content than the rubber particle size. Density of fresh When rubber is added in, workability of concrete
concrete is not linear with the rubber content because is even worse. Rubber affects more than recycled
of non-polar nature of rubber particles. Addition of aggregates. 20% rubber causes a drop of 14.3%
rubber materials reduces the strength. But the effect compressive strength while 100% recycled aggregates
of rubber on compressive strength is not considerably causes a drop of 17.2%. When they are put in one
significant when rubber replacement is below 20%. mix, there is a significant drop 34.3%, which is more
Most samples do not show explosive failure due to than the total effect of two. As for the splitting tensile
the addition of rubber. Both splitting tensile strength strength and flexural strength, the drop is still not
and flexural strength reduce as rubber is added. It is obvious. The more rubber and recycled aggregates it
found that bigger rubber particles result in a higher contains, the worse permeability resistance it is. But all
loss in strength. As the proportion of rubber within the of the samples in the experiments show good quality
mix increases, so does the permeability. The inclusion when rubber content is below 20%. Both of recycled
of rubber regardless of type increases the concrete aggregates and rubber decrease Young’s modulus
water permeability. But within 40% rubber crumb, the and rubber affects more than recycled aggregates
permeability resistance of the specimens is good. does. When all natural aggregates are replaced by
recycled aggregates, the drop of Young’s modulus
(2) Effect of rubber particle shape. The workability of is 6.6%. However, when 30% rubber is put in, a
concrete with rubber shred is worse than with rubber decrease of 18.3% is found. When 100% recycled
crumb. Difference between rubber crumb and rubber aggregate and 20% rubber is added, the drop is
shred is more obvious than that between rubber 25.9%.
particle sizes. The reduction in fresh and hardened
density shows that the density with different rubber
shape is almost the same with each other. The
strength of concrete with rubber crumb is better
than that with rubber shred. At higher replacements
strength loss is more profound, owing to low density
voids theory. Permeability of concrete with rubber
crumb is better than that with rubber shred.
Firstly, with the development of vehicle industry, deposing
more and more scrap tire is quite a big issue. Landfill
has not been permitted since several years ago in some
countries not only because of the limited land, but also
due to its non-biodegradation which may pose a potential
environment concern. Recycled aggregate is also a big
problem to dispose with more and more construction
waste. In this project, scrap tyre rubber and recycled
aggregates can be used to replace a lot of natural
aggregates. So adding scrap tire rubber and recycled
aggregate in concrete is quite an eco-friendly way.
Secondly, in recent decades, ballast-less slab track has
been utilised in many countries. One of the issues for slab
track is vibration. In order to damp, adopting rubber is a
very promising way because rubber is a kind of very good
elastic material. If rubber is put in concrete, the ductility will
be improved much and the shrinkage/fatigue crack and
brittle failure rate will reduce. Results show that although
strength reduced when recycled materials were added
in, with certain content, the strength was still acceptable.
And the concrete became more elastic which is good
for absorbing energy. Thirdly, it is obvious that scrap tyre
rubber and recycled aggregate are cheaper than natural
ones. So the cost for slab track will be lessened.
References
Esveld, C. (2010). “Recent development in high-speed
track”, First International Conference on Road and Rail
Infrastructure.
Ganjian, E., Khorteza, M. and Maghsoudi, A.A. (2009).

“Scrap-tyre-rubber replacement for aggregate and filler
in concrete”, Construction and Building Materials, 23(5),
1828-1836.
Raghavan, D. and Huynh, H. (1998). “Workability,

mechanical properties, and chemical stability of a recycled
tyre rubber-filled cementitious composite”, Journal of
Materials Science, 33(7), 1745-1752.
Funding body
College of Physical and Sciences, University of Birmingham
Further information
Haolin Su (E: hxs146@bham.ac.uk) or
Dr Jian Yang (E: j.yang.3@bham.ac.uk)
High performance repair system for rehabilitation and
strengthening of corroded reinforcement in reinforced
concrete columns
66 Mahmood Tavallaee
Kingston University
Reinforced concrete is a versatile, durable and economical The testing consisted of:
construction material. However, in some cases it does not
perform adequately due to poor design, poor construction, • Half-cell potential measurements of the steel
inadequate materials selection, severe environment or a reinforcement contained within the concrete.
combination of these factors.
• Determination of the concrete cover to steel
The corrosion of the reinforcing steel in concrete is a major reinforcement.
problem facing civil engineers today as they maintain an
ageing infrastructure. An American estimation indicates • Determination of the electrical resistivity of the
that there is $150 billion worth of corrosion damage on concrete.
their interstate highway bridges due to de-icing and sea
salt induced corrosion. In a Transportation Research Board • Sampling and analysis of concrete dust to determine
report on the costs of de-icing the annual cost of bridge the chloride ion content over a range of depths from
deck repairs were estimated to be $50- 200 million, with the exposed concrete surface.
substructures and other components requiring $100 million
a year and a further $50- 150 million a year on a multi-story • Sampling and analysis of the concrete to determine
car parks. The repair techniques, procedures and materials the depth of carbonation.
necessary to remedy the situation are based on the careful
evaluation of the causes, extent, and consequences of the The purpose of this testing was to ascertain the present
deterioration. The cost and ease of application and the condition of the concrete and associated reinforcement, to
efficiency of the repair process are major considerations in allow for the assessment of the structure to carry out any
choosing the materials and techniques. necessary repairs within the maintenance program.
This paper discusses the use of High Performance In order to protect the reinforcement Migrant Corrosion
Repair System (HPRS) as rehabilitation and strengthening Inhibitor (MCI) was injected to the corroded columns. To
technique for corroded reinforced concrete columns. The address the reduction of the ability to sustain the design
effect of applying the wraps at early ages of the columns load due to the reduction in cross sectional area of the
and its effect on corrosion propagation is addressed. reinforcement, CFRP wrapping has been applied partially
over the column height.
The project main objectives and goals are as follows:
A Fibre Reinforced Polymer (FRP) is a combination of
• To obtain general knowledge about FRP composite matrix of a plastic or resin material. The fibres bring the
materials strength to the composite; the matrix binds the fibres
together, transfers the load between them and the rest of
• To review available data and publications for FRP the structure, and protects the fibres from the environment.
strengthening of RC columns and to analyse the
results The different types of fibres are as follow:
• To introduce new technology and materials in the field • Carbon Fibre

of repair and strengthening of RC structures using
FRP • Glass Fibre
• To draw conclusions and recommendations based on • Aramid Fibre

obtained theoretical and experimental results
It was concluded that applying HPRS technique
The proposed work is including on-site rehabilitation of significantly decreases corrosion activity when applied over
8 reinforced concrete columns which were subjected to specimen. The application of the system before corrosion
aggressive environment. As part of the site investigation propagation will prevent corrosion taking place, while the
conditions of columns were monitored through Non- application of the system after corrosion occurrence will
Destructive Test (NDT). This was carried out to allow for the drop the rate of corrosion sharply.
assessment of the structures for any necessary repairs and
maintenance program.
References
ACI 440R (1996) State-of-the Art Report on Fiber
Reinforced Plastic (FRP) Reinforcement for Concrete
Structure, American Concrete Institute (ACI) Committee
440, Framington Hills, Michigan
Canadian Standards Association Can/CSA-S806-2:

“Design and Construction of Building Components with
Fiber-Reinforced Polymers”, Canada, May 2002.
Ehsani, M.R, Saadatmanesh, H .and Tao, S (1996) ‘Design

recommendations for bond of GFRP rebars to concrete’,
Journal of Structural Engineering, ASCE, Vol.122
Ehsani, M.R, Saadatmanesh, H. and Jin,L.(1997) ’Seismic

retrofitting of rectangular bridge columns with composite
straps’
Further information
Mahmood Tavallaee (E: k0826253@kingston.ac.uk) or
Dr Ted Donchev (E: T.Donchev@kingston.ac.uk)
Fig 1: Corroded RC columns Fig 2: FRP- strengthened RC columns
Low impact pre-stressed concrete sleepers
67 Ajibola I Tijani
University of Birmingham
Objectives Results
The objectives of this novel research include; The workability of concrete reduced considerably with
increase in quantity of recycled aggregate. This was as a
• Determination of fresh and hardened properties result of the ability of recycled aggregate to absorb more
(mostly strength characteristic) of concrete produced water than natural aggregate. The results of compressive
from cement replacement materials (Microsilica, ggbs, strength and tensile splitting test showed gradual reduction
& fly ash) and result comparison with that obtained in compressive and tensile splitting strength with increment
from Portland cement concrete (control result). of content of recycled aggregate without addition of
cement replacement materials. With the addition of cement
• Determination of strength of concrete made from replacement materials (fly ash, GGBS), the strength of
recycled aggregate and result comparison with recycled aggregates concrete increased. The earlier
concrete made from natural aggregate. reduction in workability (slump) was also improved by the
addition of cement replacement materials.
• Determination of the effect of Synthetic ‘Macro’ Fibre
on the strength properties of the prestressed concrete Potential for application of results
sleeper.
The goal of this on – going research is to determine the
strength characteristic of recycled aggregates, cement
Goals
replacement materials, and synthetic ‘macro fibre for
The goals of this novel research are; application in pre-stressed concrete sleepers, which will
give a better understanding on the properties of concrete
• To investigate the possibility of producing low-impact produced from these novel materials and contribute greatly
pre-stressed concrete sleepers through the use of in ensuring that rail transportation remain more eco-friendly
recycled aggregate, cement replacement materials with the production of low impact pre-stressed concrete
(Microsilica, Fly Ash, & GGBS) and fibre. sleepers.
• To provide better understanding on the properties References

of concrete produced from these aforementioned
Hart-Beat. 2006. Welded rail on concrete sleeper
materials and contribute greatly in ensuring that
[Online]. Available: http://www.lightrail-hartbeat.com/
railway transportation remain more eco-friendly
Construction_09.htm.
through the construction of rail infrastructures from
low carbon concrete leading to further reduction in the
Kaewunruen, S. & Remennikov, A. Low-velocity impact
amount of carbon footprint produced within the rail
analysis of railway prestressed concrete sleepers.
sector.
Proceedings of the 23rd Biennial Conference of the
Concrete Institute of Australia: Design, Materials, and
Construction, 2007 Adelaide, Australia. 370.
The use of pre-stressed concrete sleeper as reported by
Kaewunruen and Remennikov (2007) became necessary Peng, H.S., Chen, H.J. & Yen, T. 2002. Effects of
as a result of the need to resist repetitive impact stress Pozzolans on the Strength and Workability of Recycled
waves from dynamic interactions between rail vehicle and Aggregate Concrete. The 6th National Conference on
track infrastructures in services. The materials used in this Structural Engineering Paper No. C07. Pingtung, Taiwan.
research includes fly ash, ground granulated blast-furnace (in Chinese)
slag, microsilica, cement, tap water, river sand, natural
coarse aggregates, recycled coarse aggregates, and Forta- Further information
ferro 54mm synthetic ‘macro’ fibre. Specimens were also
Ajibola Tijani (E: jibolatijani@yahoo.com) or
tested at 1day, 7 day and 28 day old. The investigation
Dr Jian Yang (E: J.yang.3@bham.ac.uk)
was carried out using compressive strength test, tensile
splitting test, slump (workability) test, and flexural strength
test. There were total of five batches of concrete mixes,
which is made up of 25% increment of recycled aggregate
replacement from 0% to 100% and inclusion of Forta-Ferro
54mm Synthetic ‘Macro’ Fibre. All the tests were carried
out according to British Standards.
Fig 1: Welded rail on concrete sleepers (Hart-Beat, 2006)
Fig 2: Effect of fly ash & GGBS contents on strength of concrete (Peng et al., 2002)
Long term performance of CFRP tendons in pretensioned
structures
68 Eleni Toumpanaki
University of Cambridge
Project objectives and goals due to the beneficial role of thermal expansion in the bond
behaviour. On the other hand preconditioning of CFRP
Carbon fibre reinforced polymers (CFRPs) are promising
tendons before casting and testing can lead to a more
alternatives to steel reinforcement for marine and
severe degradation that is not representative of the real
offshore structures. CFRPs do not corrode but, in humid
in-situ conditions.
environments, it is important to ensure that the long-term
durability of the CFRP is not adversely affected due to
moisture absorption in the epoxy-matrix material. This
research project focuses on two main issues: the long- The research will provide insight into the performance of
term durability of CFRP reinforcement in isolation and the CFRP tendons when exposed to moisture. In particular, the
long-term bond performance between CFRP tendons and degradation of the matrix-dominated properties and the
concrete in pretensioned structures. bond performance of CFRP reinforcement in pretensioned
structures will be established. This degradation is not
Description of methods and results necessarily detrimental, since it could lead to a more
ductile behaviour [3,4].
The investigation of CFRP reinforcement subjected to
humid environments provides insight into the structural
References
performance of pretensioned concrete structures under
similar exposure environments. Exposure of CFRP [1] Davalos, J.F., Chen, Y., Indrajit, R. (2008). Effect of FRP
tendons to moisture may be detrimental to the mechanical Bar degradation on interface bond with high strength
performance of the epoxy-matrix material due to the concrete. Cement & Concrete Composites, 30, pp.
porous structure of the matrix. The diffusion properties 722-730.
were initially investigated for CFRP tendons manufactured
using two different curing processes. The diffusion [2] Achillides, Z. & Pilakoutas, K. (2004). Bond Behavior
characteristics were inferred by regularly weighing the of Fiber Reinforced Polymer Bars under Direct Pullout
CFRP samples which had been fully immersed in distilled Conditions. Journal of Composites for Construction,
water. In this way, it was possible to assess the level of 8(2), pp. 173-181.
attack by water molecules in the chemical microstructure of
the epoxy. The manufacturing process via the post-curing [3] Burgoyne, C.J. (1993). Should FRP be bonded to
of CFRPs seems to decrease the rate of diffusion and Concrete? Conference Paper presented at American
shear modulus degradation by creating a denser network Concrete Institute Convention, Vancouver, British
structure in the matrix material. Columbia, Canada.
Accelerated ageing using elevated temperatures is a [4] Lees, J.M. & Burgoyne, C.J. (1999). Experimental
common method to predict the long-term behaviour in a Study of Influence of Bond on Flexural Behavior of
short-term period. The use of 40°C and 60°C will enable Concrete Beams Pretensioned with Aramid Fiber
us to build an Arrhenius relationship for CFRP tendons Reinforced Plastics. ACI Structural Journal, 96(3), pp.
and to establish the time-shift relationship between 20°C 377-385.
(lab conditions) and elevated temperatures. Finally, an
evaluation of this accelerated ageing method will be carried [5] Robert, M. & Benmokrane, B. (2010). Effect of Aging on
out. Bond of GFRP Bars Embedded in Concrete. Cement
and Concrete Composites, 32(6), pp. 461-467.
The long-term performance of sand coated CFRP tendons
is also of interest in this research work. The parameters to Further information
be studied are the percentage loss of sand coating due to
Eleni Toumpanaki (E: et343@cam.ac.uk) or
handling, storage and transportation, the performance of
Dr Janet Lees (E: jmL2@cam.ac.uk)
the sand coating layer and changes in the matrix material
due to exposure to moisture. The role of the loss of bond
performance in pretensioned structures will be assessed in
terms of the ultimate failure load and deflection behaviour.
In an initial study, pull out tests will be carried out due
to the simplicity of the test method and for comparison
with existing research work in the literature [1,2]. The
bond performance in pretensioned beams will then be
modelled and beam tests will be designed. The most
challenging aspect in the experimental series is to evaluate
the appropriate accelerate ageing conditions [5]. On the
one hand, the use of elevated temperatures in concrete
cube specimens immersed in water can be misleading
Structural and sustainable assessment of deployable
structures
Osahon Joshua Umweni

Nottingham Trent University 69
Background to study The objectives of the research are to:
When I was a Masters student at the University of Salford,
• Carryout thorough material investigation of probable
Manchester, United Kingdom I researched on the “folding
sustainable materials deployable structures, analysis
techniques of deployable structures using the pantograph
at high and low temperatures, checks on durability
system and a novel deployable cable-chain system as case
with the aim to find the best material properties and
studies”. I used a combination of both systems and was
possible composite materials.
proven to provide a more resilient and efficient structure
using bridge models for structural analysis in 2-dimension
• Examine extensively the assessment of deployable
and how compact these structures can be folded. A
structures on the grounds of structural efficiency and
resilient, redundant structure was proposed and termed
sustainability.
Hybrid structure (Umweni, 2011).
• Carry out environmental impact assessment to
Currently, I wish to broaden the scope of this research to
know the effects of deployable structures to the
pantographic systems in 3-dimension weight optimisation
environment.
with the aim to answer sustainability and environmental
impact questions during the course of this research.
• Carry out life cycle cost analysis and life cycle
I believe this is a good background to undertake any
assessment to make an appropriate selection of a
research pertaining to Lightweight structures or deployable
construction material for deployable structures.
structures.
• Propose and develop a sustainable deployable
structural concept which is structurally effective,
To accomplish deployment, deployable structures will durable, and environmentally friendly, demonstrated
consist of functional elements that are typically active only using models and analysed using recommended
during deployment. Their application in diverse scenarios structural design and analysis software.
cannot be over-emphasised but how sustainable are they?
What impact do they have on the environment? How Methodology of research
flexible can their structural designs be?
Sustainable techniques applicable in the design of
deployable structures using the various deployable
(Chilton, 2009) reviewed strategies for the deconstruction
structural systems will be investigated, taking case studies,
of conventional construction materials such as masonry,
propose structurally efficient models and putting material
timber, reinforced/prestressed concrete, and steel and
properties into consideration drawn from literature studied.
went on to advise that recent materials, such as tensile,
Confirmatory structural testing of a select construction
coated-fabric membranes, ETFE foils and fibre-reinforced
materials (such as impact test, abrasion test, brittleness
plastics can be employed for structural systems like
test and other forms of structural testing) and make
deployable structures which are easy to deconstruct and
comparison on material properties. Construction materials
designed to utmost possibility for reuse. However, the
studied but not limited to are steel, timber, masonry
sustainability and the impact to the environment of these
and more recently fibre-reinforced plastics. During the
materials when employed on deployable structures have
course of the research, an understanding of the need for
not been fully researched on.
deployable structures, their impact on the environment,
various scenarios for application, their growing demand
Research aims and objectives
in modern times and also the maintenance strategies that
This research will investigate and summarise the key can be employed in the various forms studied drawn from
principles in determining how compactly deployable extensive literature studied.
structures can be folded away for transportation and
develop/propose how structurally efficient they are, The various types of deployable structures will be reviewed;
how sustainable they can be in numerous applications their characteristics, geometry and structural morphology
(Earth and Space applications) and their impact on the deployable structures will be discussed in detail including
environment; from a material standpoint in relation to the various forms they can take during translation.
durability and structural efficiency. These would be studied
using the deployable structural systems.
The aim of this research is to investigate deployable

structures with more emphasis on pantographic deployable
structures for structural stability and sustainability.
The folding theory of deployable structures will also be References
reviewed. The folding theory is intended to provide an
Akgün, Y., Gantes, C. J., Kalochairetis, K. E., & Kiper,
insight into the application of algorithms in the folding
G. (2010). A novel concept of convertible roofs with
of structures. Because only a few of the many types of
high transformability consisting of planar scissor-hinge
deployable structures that have been proposed meet
structures. Engineering Structures, 32(9), 2873-2883.
actualisation, detailed design needs to be carried out.
Structural analysis in the deployed and folded state using
Chilton, J. (2009) Designing structures for Deconstruction
the ANSYS software or any other available structural
[paper 245], Construction information Quarterly, Volume 11
analysis software will be carried out.
issue 3, 121 -126
Actual prototyping of deployable models conceived will
Chilton, J, & Velasco, R. (2005) Application of textile
be carried out putting, specific type and dimensions
composites in the construction industry in Design and
of actuators, power of motors during deployment if
manufacture of textile composites Edited by A. C Long,
necessary; probable covering materials and other essential
CRC press
elements will be defined in detail.
Gantes, C. J. (2001). Deployable Structures: Analysis and
Folding Algorithms of deployable structure will be
Design WIT Press (2001).
researched into so that general and efficient polynomial
algorithm equations can be used in folding the many forms
Gunnar, T. (2002). Deployable Tensegrity Structures for
of deployable structures.
Space Applications. (Phd), Royal Institute of Technology,
Stockholm, Sweden.
Possible outcome of research
Deployable structures have gradually become more The legacy of 1960’s University buildings. A Report.
popular because of their numerous advantages, even Commisioned by AUDE and supported by HEFCE, March
though only few have been built successfully. A sound 2008.
understanding of their intrinsic design parameters, flaws
and tolerances would be known from research so that Li, Y., Vu, K. K., & Liew, R. J. Y. (2010). Deployable cable-
successful deployment and folding/deconstruction can chain structures: Morphology, structural response and
be achieved. Basic issues such as shape of elements, robustness study. Journal of the International Association
type of material, deployment behaviour, joint connections, for shell and spatial structures: J. IASS, 14.
structural and sustainability performance when deployed
and when folded/deconstructed will form the basis of The UK Green Building Council (2009), Making the case for
the research so that the use of the structures proposed a code for sustainable buildings: Executive Summary.
in this research would be used worldwide allowing for
sustainability, and improve the sustainability design Umweni, O. J. (2011). Folding Techniques of deployable
concepts of engineers and help to improve designs for structures using the pantograph system and the novel
disaster ravaged parts of the world. Possible employment deployable cable chain structural systems as case studies
of deployable structures to solve housing needs in the (Masters Degree), University of Salford, Manchester, United
United Kingdom and other parts of the world is another Kingdom.
possible outcome.
Further information
Osahon Joshua Umweni (E: osahonjoshua@gmail.com) or
Dr Anton Ianakiev (E: anton.ianakiev@ntu.ac.uk)
Designing resistance against progressive collapse into
framed structures
Carolos Antonios Vidalis

Imperial College London 70
Oral presenter
Project objectives and goals Description of method and results
The purpose of this project is to develop practical guidance Research was undertaken using a range of methods,
for designing a steel framed structure to withstand a depending on the project phase:
progressive collapse scenario.
• Development of the connection and beam system
According to the most recent Building Regulations, almost models through analytical work
all buildings need to satisfy robustness requirements to
avoid disproportionate collapse. The designer responsible • Validation of the models against FE analysis and
needs to find a way not only to assess the structure’s experimental results
resistance but also to decide on whatever changes are
necessary in order to ensure satisfactory performance. In • Case studies for representative structures – response
the event of an incident, the building should survive and not analysis
suffer the damage of the structure of figure 1 – so what is
available? • Critical evaluation of findings to develop an adaptable
design methodology
Contrary to Codes and Standards for other loading cases
like earthquake or fire, robustness provisions remain largely In the first part of the project, five types of connections
dependent on prescriptive measures. Notwithstanding were modelled based on the solution proposed by
this, guidance [1] for multi-storey buildings has gradually Stylianidis [4], which takes into account the combined
started to encourage the use of deterministic methods. effect of bending moment and axial loading. The systems
However, most of the analytical approaches that have been were chosen to be representative of partially restrained
developed involve the use of complex and demanding (bolted endplate) and fully restrained (flange welded RBS
numerical analysis, which renders them unsuitable for and coverplated) connections used in simply designed and
use in routine design. In addition, previous research moment resisting steel frames.
has focussed on isolated case studies, as opposed to
developing an understanding of how the design of the Subsequently, the solutions were validated against a finite
system is linked to its behaviour. Thus, the potential of a element model built with the ADAPTIC software. Due to the
readily available tool for a quantitative comparison of the cost and time required to experimentally test an assembly
merits of alternative designs provides a strong case in under sudden column loss, most models were compared
favour of further work. against available experimental results of a similar project of
the USA National Institute of Standards and Technology.
Research has shown that the current provisions, which are The validation was performed for the beam system
mainly based on tying elements together, are not directly response with very good agreement as shown in figure 2.
related to progressive collapse resistance [2]. Instead,
behaviour is influenced by the ductility, strength and Using these models with the ICL analysis method, the
energy absorption capacity of the system. Their combined behaviour of six representative structures was examined
effect is recognised by a method developed over the past in detail: the Cardington test frame (bare steel and
8 years at Imperial College London [3], which offers a composite), the SAC Project frames and the NIST
simplified framework for quantitatively evaluating structural Robustness Project frames. Each case-study involved
robustness on the basis of pseudo-static capacity supply extensive parametric tests aimed at determining the link
and demand. between the beam and floor system resistance actions and
system design. The results obtained clearly highlight:
Nevertheless, due to the complex interplay between
features, there has been little work in determining the • The influence of the connection design and frame
influence of connection and frame design on local layout
resistance mechanisms. The lack of such knowledge
prohibits the designers from addressing the needs of the • The contribution of each resistance mechanism to
structure using a streamlined process instead of a trial and system capacity
error approach.
• The role of each structural component during the
In an attempt to overcome these barriers, this project aims response
to explore how this capacity indicator can be translated
into specific remediating recommendations and thus
provide a tool for determining the most effective structural
modifications to ensure robustness.
In addition, the comparison of behaviour of each type of References
design allowed determining:
[1] ODPM, The building regulations 2000, Part A, Schedule
1: A3, Disproportionate collapse. 2004, Office of the
• What the potential vulnerabilities of each structure are
deputy Prime Minister: London, UK.
• Which design features influence these vulnerabilities
[2] Nethercot, D.A., et al., Resisting progressive collapse
by the use of tying resistance, in 4th International
The findings of the project were used to develop a step
Conference on Steel & Composite Structures. 2010:
by step process able to assess resistance in progressive
Sydney, Australia.
collapse and propose modifications to match the required
margin of safety according to the design of the frame.
[3] Izzuddin, B.A., et al., Progressive collapse of multi-
Communication with structural engineers contributed in
storey buildings due to sudden column loss - Part
determining the structural modifications that are more
I: Simplified assessment framework. Engineering
practical to apply in the construction practice.
Structures, 2008. 30(5): p. 1308-1318.
[4] Stylianidis P. and Nethercot D. A., Connection
The proposed methodology can provide a useful tool Modelling for Progressive Collapse Analysis. Steel
for designers, helping to develop a more intuitive feeling Concrete Composite and Hybrid Structures, 2009: p.
on designing against progressive collapse. Also, the 575-580.
enhanced understanding of the main resistance actions in
a progressive collapse scenario opens the way for future Further information
studies on more complex framed designs. In addition,
Carolos Antonios Vidalis (E: c.vidalis10@imperial.ac.uk)
this research has global importance towards enhancing
or Professor David A Nethercot
the present construction codes with guidelines capable of
(E: d.nethercot@imperial.ac.uk)
providing efficient and safe design provisions for routine
design use.
Fig 1: Murrah building collapse Fig 2: Comparison of model and finite element
analysis pseudo-static responses of RBS
beam systems
An energy approach for the robustness and damage
potential of structures
Wenjuan Zhuang
University of Bristol 71
Introduction and objectives Analysis procedure
The catastrophic events such as the collapse of Ronan The analysis of a structure for its damage potential involves
Point apartments and the terrorist attack on the World the following: (a) a model of the loads, (b) an identification
Trade Centre towers have been drawing increasing of the initial damage event, (c) calculation of strain energy,
attention on the design of structures to resist progressive (d) processing of results to obtain the failure sequence. The
collapse for the past decades. There is a wide body predicted sequence of progressive collapse is obtained
of literature on progressive collapse (e.g. Agarwal and through repeated analyses where the elements with highest
England 2008 [1], Dusenberry and Hamburger, 2006 strain energy are destroyed first thus releasing high strain
[2], Starossek, 2007 [3], Nafday, 2008 [4]). Some regard energy. This energy will spread to other elements possibly
robustness as a property of structure (Lu et al. 1999 [5], leading to another damaged element that cannot withstand
Agarwal et al. 2003 [6]) while it has also been defined more energy.
in terms of probability of failure (Lind, 1995 [7]) and risk
(Faber, 2001 [8]). However, seldom do these studies refer Numerical results
to the distribution of energy that must relate to the form of
A 2-bay 4-storey concrete frame is shown in Fig. 1. The
the structure.
members are modelled using element BEAM23 in ANSYS
[10]. A static nonlinear analysis is adopted. Removal of a
The objective of this paper is to present a new approach
central column is considered as an initial damage event.
which examines the distribution of strain energy so that
The vertical loads as recommended in GSA [11] are used.
an assessment of the robustness of the structure and its
damage potential can be made.
Case 1 - original structure
An energy approach for robustness Fig. 1 shows the changes in strain energy densities as the
failure propagates after the initial damage event, i.e. the
Progressive collapse is a phenomenon when the failure
loss of column B1. In every step, the element that gains the
of a structural member spreads to the adjacent structural
most strain energy density is the next one to be damaged.
members, which in turn causes the collapse of a part of
For example in Fig. 1(a), the change in strain energy density
the structure or the entire structure. Beeby [9] argued that
for beam AB1 is 26.4kJ/m3, which is the largest value and
robustness of structures is a very significant way to mitigate
so this is the next damaged element. Similarly in Fig. 1(b),
the progressive collapse.
beam AB2 is the next damaged element with the maximum
change. 205.5kJ/m3. The values in Fig. 1(c) point to AB3
Strain energy
as the next element to be damaged. Due to symmetry, the
As a structure deforms, the external work done by the energy variations of beams and columns in Fig. 1(a) are
loads is transformed into strain energy of the structural symmetrical. So the predicted failure sequence considering
members and an equilibrium state is reached when the gains in strain energy density after the first event is
the external work is equal to the stored strain energy. AB1/BC1  AB2/BC2  AB3/BC3  AB4/BC4, which is
A stiffness matrix contains the inherent properties of the same as initially predicted.
a structure and deformations can be recognised as a
reflection of the external loads. A normalised measure Case 2 – strengthening of beams
referred to as ‘strain energy density’, can be obtained by
In Case 2, with AB1 strengthened and middle column
dividing strain energy by the volume of material.
removed, the resulting changes in the strain energy density
are shown in Fig. 2. There is a difference in the changes
Robustness
for the beams in the left bay between the two cases. The
If a member is highly stressed, it will correspondingly store change in the strain energy density of the strengthened
more strain energy and thus it will require less external beam is much less than the other beams which is also
effort to reach a failure state. Once an element cannot proved by Jennings [12], but the beam adjacent to it shows
sustain itself, strain energy will be released and to follow the most gain in strain energy density.
the conservation of energy, this energy will be transferred
to other elements. During this redistribution of strain
energy some elements will gain energy while others may
lose energy. The changes in strain energy density after
a damage event must relate to the characteristics of the
structure. These characteristics are embedded in the
stiffness matrix of the structure but are difficult to extract
otherwise.
Application of results References
For Case 1, the same sequence is obtained by failure strain Contact Wenjuan Zhuang for details of above references.
as well. This suggests that it is possible to identify the
failure sequence based on changes in strain energy density Further information
after initial event. Such an analysis also helps to reduce the
Wenjuan Zhuang (E: cexwz@bristol.ac.uk) or Dr Jitendra
computational effort.
Agarwal (E: J.Agarwal@bristol.ac.uk)
The results suggest that the strengthening of a beam
inhibits damages in one bay, but this may aggravate
damage in the other bay. It can be concluded that a
strengthened beam causes the redistribution of energy. It
is also worth noting that strengthening may not guarantee
the structure can survive the extreme event, but it may be
possible to influence the propagation of damage.
Fig. 1: Changes to strain energy density (kJ/m3) after damage to column B1 (key: maximum gains highlighted by boxes;
energy losses highlighted by circle; damaged member highlighted by dashed line)
Fig. 2: Changes to strain energy density (kJ/m3) with beam AB1 strengthened - after damage to column B1
Notes
Notes
The Institution of Structural Engineers
International HQ
11 Upper Belgrave Street
London SW1X 8BH
United Kingdom
tel +44 (0)20 7235 4535
fax +44 (0)20 7235 4294
mail@istructe.org
www.istructe.org
Founded 1908 and incorporated by Royal Charter 1934

Registered Charity No 233392

2013 Young Researchers Conference Proceedings PDF

Uploaded by

Copyright:

Available Formats

2013 Young Researchers Conference Proceedings PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2013 Young Researchers Conference Proceedings PDF

Uploaded by

Copyright:

Available Formats

Sponsors:

Flint & Neill

S-FRAME Software (UK) Ltd

The Institution of Civil Engineers

The Institution of Structural Engineers Research Fund

Since its inauguration in 1999, the Young Researchers’

As a consulting engineer often engaged in projects

We rely heavily on our researchers and research

Professor David A Nethercot

Prof David Nethercot OBE, FREng, was for 12 years until

He was awarded a DSc degree in 1993 and elected to the

For more than 40 years he has been active in research,

He has supervised over 40 PhD students to completion,

David chaired the Institution’s Research Panel for many

Chairman: Ian Firth

Brian Bell (Network Rail)

Dr Anthony Darby (University of Bath)

Dr Leroy Gardner (Imperial College London)

Prof Costas Georgopolous (Kingston University)

Susan Giahi-Broadbent (Atkins)

Prof Abdy Kermani (Edinburgh Napier University)

Prof Dennis Lam (University of Bradford)

Chris Walker (Flint & Neill)

Dr Christopher White (EPSRC)

Ian joined Flint & Neill in 1979, and became a Partner in

09.45 Registration and coffee

10.25 Keynote address by Professor David A Nethercot – Imperial College London

10.55 Under-utilisation of steel beams in construction (Abstract No. 41)

11.10 Efficient use of stainless steel in structural design (Abstract No. 2)

12.55 Membership matters

13.40 Poster session

15.35 Tea & judging

16:00 Prize giving & closing remarks

FRP composites for strengthening of steel bridge members (Abstract No. 7)

Modelling of steel connections at elevated temperatures (Abstract No. 31)

Structural behaviour of concrete twin walls (Abstract No. 54)

Multi-scale stochastic fracture mechanics of concrete (Abstract No. 56)

Background • To develop a material constitutive model that could be

Results of initial simulation of a segment of the structure show that:

Fig 1: Logarithmic strain

material in various structural applications, taking advantage

The results of the sensitivity study show that the seismic

Fig 2: Result of Tornado diagram

Fig 1: AEM model

Talyor & Francis Group 2008. Spon’s Irish Construction

Fig 1: Embodied energy and embodied carbon per surface

Fig 2: Initial lifecycle heating and material costs per m2 of

The first objective of this PhD project is to correctly

An improved understanding of residual stresses present in

Load and Interaction

In order to accurately simulate the behaviour of concrete-

Muhammad Aslam Bhutto

Fig 2(a): Test of specimen B6

Fig 2(b): FE failure mode of specimen B6

[2] M. Shinozuka and G. Deodatis, “Simulation of

[2] Harris R.J.L. & Kelly O.J. 2002. The structural

[3] Kuijvenhoven M. 2012. Particle-Spring Method for

[4] Toussaint M.H. 2007 A Design Tool for Timber

Fig 2: Grid-shell in Lecce, Italy. 2010.