Modelling and Simulation of The Strength
Modelling and Simulation of The Strength
Modelling and Simulation of The Strength
doi:10.56397/IST.2024.01.01
Abstract
The strength performance of UHPFRC with micro and hooked-ends steel fibers was studied in this paper using
Finite Element (FE) modelling and simulation. The specimens were modelled in geometry, material, constraints,
load and boundary conditions using ABAQUS package and dynamic explicit analysis was used to investigate the
crack pattern, failure mode and stress-stain behaviour of the UHPFRC specimens when subjected to uniaxial
compression and tension loading. Results from the FE simulation revealed that the surfaces of the UHPFRC
specimens were still in good shape after attaining their ultimate resistance against compression and tension
loading. The UHPFRC’s ultimate compressive strength of 260N/mm2 was just 7% higher than the experimental
compressive strength. The UHPFRC exhibited both linear compressive stress-strain behaviour up to 89% of its
peak strength and non-linear stress-strain behaviour with strain hardening and strain softening phases. Further
findings also showed that the UHPFRC despite having ultra-high tensile strength did not undergo strain
hardening phase as the vertical direction stresses were mainly distributed around the loaded section of the
specimen. The FE UHPFRC models’ strength in compression and tension only has slight and negligible
variations from the experimental strengths. The deformation and strength performance of the FE specimens are
in perfect agreement with the experimental specimens. Therefore, FE modelling and simulation can be used as a
reliable method of carrying out extensive studies on UHPFRC properties and performances.
Keywords: modelling, simulation, strength properties, ultra-high performance fiber reinforced concrete
1. Introduction
Ultra-high performance fiber reinforced concrete (UHPFRC) is a concrete that has a compressive strength of at
least 150N/mm2 (AFGC, 2002) with better performance in strength, durability and rheology when compared
with normal strength concrete (NSC) (Azmee & Shafiq, 2018). The ultra-high strength of UHPFRC is a result of
its ingredients which include high cement content, powder like silica fume, very fine aggregate like quartz sand,
steel fibers, super plasticizer and a very low water-cement ratio (w/c) (Raja & Sujatha, 2014; Sadrekarimi, 2004).
Researchers are now intensifying interest in the investigation of UHPFRC properties as well as its structural
performance because of its excellent properties. Many studies on the strength of UHPFRC have already been
done by several researchers. For instance, Hashim et al. (2020) investigated the mechanical properties of
UHPFRC containing different types of fiber and used the stress-strain results obtained to develop damage and
constitutive models for UHPFRC. The models when employed in Finite Element Method (FEM) to simulate the
load-displacement property of a hollow concrete column, yielded promising results. Graybeal and Baby (2013)
studied the tensile strength property of UHPFRC and utilized the findings obtained to develop a direct method of
testing the tensile property of UHPFRC. Findings from Graybeal and Baby showed that UHPFRC specimen
subjected to direct tension test has four phases of tensile response: (1) elastic phase (2) multi-cracking phase (3)
crack straining phase (4) localization phase. Krahl et al. (2018) conducted research on the mechanical properties
1
INNOVATION IN SCIENCE AND TECHNOLOGY JAN. 2024 VOL.3, NO.1
of UHPFRC and proposed a mechanical damage and stress-strain models using the obtained results. Verification
of Krahl et al.’s models through numerical simulation showed good correlation. Prem et al. (2012) conducted
research on UHPFRC to evaluate its mechanical properties and findings revealed that change in the volume of
steel fiber and aspect ratio lead to substantial change in the strength properties of UHPFRC. There is no specific
strength value for UHPFRC in compression or tension; and the values depend on the target strength, mix design
and curing condition as researchers like Rossi et al. (2004) has reported that UHPFRC’s compressive strength
and tensile strength may even be as high as 204N/mm 2 and 20N/mm2.
The cost of producing UHPFRC is very high when compared with NSC due to its special ingredients and testing
conditions; and this high fund requirement has hindered researchers from carrying out aggressive experimental
studies on UHPFRC. However, a less costly way (which researchers rarely take advantage of) of investigating
the properties and performance of UHPFRC which produces highly correlated results with experimental results
is Finite Element (FE) numerical modelling and simulation. Numerical modelling basically involves the
modelling of UHPFRC’s geometry, material (using mechanical properties obtained from experiment),
constraints and boundary conditions. While numerical simulation involves the computer analysis (static or
dynamic analysis depending on the property or behavior being investigated) of the UHPFRC specimen’s
property or element’s performance. The use of FE model to study UHPFRC properties is not straight forward as
it requires the use of damage and stress-strain models peculiar to UHPFRC to model its material properties; and
this may also be one of the reasons why there are limited studies in this area but many researchers have
developed models that can be used to capture the damage of UHPFRC under compressive and tensile loading as
well models that can be used to predict its stress-strain response. For instance, Tian et al. (2023) studied the
behavior of UHPFRC subjected to cyclic and monotonic tensile loading; and utilized the study findings to
propose a tensile damage and stress-strain models for UHPFRC. Hashim et al. (2020) also proposed simplified
damage plasticity and stress-strain models for UHPFRCs and findings revealed that the models perfectly predict
the damage as well as the stress-strain behavior of the studied UHPFRC.
Many researchers have not shown much interest in using numerical modelling and simulation to study the
properties of UHPFRC even though several literatures that employed the use of numerical simulation are
available. Some of which include: Shi et al. (2023) study on the response and constitutive model of UHPFRC
under tension loading whose numerical results showed good prediction of the UHPFRC’s tensile stress-strain
behavior. Bahraq et al. (2019) numerical research on the shear performance of reinforced concrete beams
strengthened with UHPFRC in which the obtained failure mode and ultimate load matched perfectly with those
obtained from experiment. The stress-strain behavior of UHPFRC subjected to uniaxial compression and tension
was studied by Naeimi and Moustafa (2019) using DIANA package; and findings revealed that the experimental
elastic and inelastic behaviors of the UHPFRC under both compression and tension were reproduced
successfully through numerical modelling and simulation. Chowdhury et al. (2016) modelled steel fiber
reinforced concrete (SFRC) cylinder and simulated its compressive and tensile properties using ANSYS package;
and after deep analysis and optimization of the main parameters concluded that results from the numerical
models have good correlation with the results from experiment.
Although some studies have been reported on numerical modelling and simulation of UHPFRC’s properties,
they are still far below the percentage of studies that should have been conducted using numerical simulation in
order to understand the properties of UHPFRC that have not been studied due to the high cost of experimentation.
So, the aim of this study is to investigate the compressive and tensile strength performance of UHPFRC with
micro and hooked-ends steel fibers using FEM and to also check its performance variation level from that
obtained through experiment.
2. Modelling
ABAQUS package was used to model and simulate the compressive and tensile properties of the UHPFRC
specimens. The UHPFRC cube specimen of 100x100x100mm3 in dimension, prism specimen of
100x100x300mm3 in dimension and dog-bone specimen with length of 200mm, 50mmx50mmx25mm end
sections, 26mmx50mm middle portion, 12mm length tapered section with a radius of 12mm, 126mm length
notched cross-section were modelled using the three dimensional (3D), solid elements with three degrees of
freedom (DOF) at each of its 8-nodes. The geometry of the specimens is presented in Figure 1.
2
INNOVATION IN SCIENCE AND TECHNOLOGY JAN. 2024 VOL.3, NO.1
The UHPFRC material was modelled using concrete damage plasticity model (CDPM) developed by Lubliner et
al. (1989) and Hashim et al. (2020) stress-strain model as expressed in (1).
𝜎 = (1 − 𝑑)𝐸𝑜 (ɛ − ɛ𝑝𝑙 ) (1)
where σ is the stress in compression/tension, d is the damage variable in compression/tension, E0 is the elastic
modulus, ε is the total strain and εpl is the plastic strain.
The damage parameters in compression and tension used for representing crack formation in the specimens are
expressed in (2)-(3) based on Tao and Chen (2015) and Birtel and Mark (2006):
𝜎𝑐 𝐸𝑐−1 𝜎𝑐 𝐸𝑐−1
𝑑𝑐 = 1 − 𝑝𝑙 1 𝑜𝑟 𝑑𝑐 = 1 − (2)
ɛ𝑐 ( −1)+𝜎𝑐 𝐸𝑐−1 ɛ𝑖𝑛 −1
𝑐 (1−𝑏𝑐 )+𝜎𝑐 𝐸𝑐
𝑏𝑐
𝜎𝑡 𝐸𝑡−1 𝜎𝑡 𝐸𝑡−1
𝑑𝑡 = 1 − 𝑜𝑟 𝑑𝑡 = 1 − (3)
𝑝𝑙 1
ɛ𝑡 ( −1)+𝜎𝑡 𝐸𝑐−1 ɛ𝑖𝑛 −1
𝑡 (1−𝑏𝑡 )+𝜎𝑡 𝐸𝑐
𝑏𝑡
𝑝𝑙 𝑑 𝜎𝑐
ɛ𝑐 = ɛ𝑖𝑛 𝑐
𝑐 − (1−𝑑 (4)
𝑐 )𝐸𝑜
𝑝𝑙 𝑑 𝜎𝑡
ɛ𝑡 = ɛ𝑐𝑟 𝑡
𝑡 − (1−𝑑 (5)
𝑡 )𝐸𝑜
𝜎𝑐
ɛ𝑖𝑛 𝑒𝑙
𝑐 = ɛ𝑐 − ɛ𝑐 𝑜𝑟 ɛ𝑐 − (6)
𝐸𝑜
𝜎𝑡
ɛ𝑐𝑟 𝑒𝑙
𝑡 = ɛ𝑡 − ɛ𝑡 𝑜𝑟 ɛ𝑡 − (7)
𝐸𝑜
where dc is compressive damage parameter which ranges from 0 to 1, dt is tensile damage parameter which
ranges from 0 to 1, σc is compressive stress of concrete, σt is tensile stress of concrete, Eo is modulus of elasticity
𝑝𝑙
of concrete, εc is total concrete strain in compression, εt is total concrete strain in tension, ɛ𝑐 is plastic strain
𝑝𝑙
corresponding to compressive strength of concrete, ɛ𝑡 is plastic strain corresponding to tensile strength of
𝑒𝑙
concrete, ɛ𝑒𝑙 𝑖𝑛
𝑐 is elastic strain of concrete in compression, ɛ𝑡 is elastic strain of concrete in tension, ɛ𝑐 is inelastic
strain of concrete in compression, ɛ𝑐𝑟
𝑡 is cracking strain of concrete in tension, bc and bt are constant parameters
with values 0 <bc, bt ≤ 1 (bc = 0.7 and bt = 0.1).
The UHPFRC’s material used for tensile strength simulation was also modelled using ductility damage to show
its damage evolution in terms of deformation and the variable employed for the ductility damage is shown in
Table 2. The mechanical properties of the UHPFRC based on Hashim et al. (2020) and other modelling
parameters are presented in Table 1:
3
INNOVATION IN SCIENCE AND TECHNOLOGY JAN. 2024 VOL.3, NO.1
The top and bottom faces of the cube specimen were modelled with rigid body constraint; while the dog-bone
specimen was modelled using coupling constraint. In terms of load and boundary conditions, fixed support was
applied to the base of the cube and prism while vertical displacement load was applied to the top of the cube and
prism using dynamic explicit analysis as shown in Figures 2(a-c). Vertical displacement load was applied at the
top face of the dog-bone specimen using dynamic explicit analysis as shown in Figure 2(d); while the bottom
cell of the dog-bone specimen was modelled with a fixed boundary condition and the top end section was
modelled with vertical DOF as shown in Figure 2(d).
(c) Prism’s load and boundary condition (d) Dog-bone load and boundary condition
Figure 2. Load and boundary conditions
The cube specimen, prism specimen and dog-bone specimen were meshed using mesh size of 10mm, 20mm and
6mm respectively; and the dog-bone specimen’s end sections were assigned a local seed of 3mm to prevent
element meshing error which usually leads to abortion of analysis. The meshed specimens are shown in Figure 3.
4
INNOVATION IN SCIENCE AND TECHNOLOGY JAN. 2024 VOL.3, NO.1
5
INNOVATION IN SCIENCE AND TECHNOLOGY JAN. 2024 VOL.3, NO.1
250
Compressive stress
200
(N/mm2)
150
100
50
0
0 0.005 0.01 0.015
Compressive strain
(a) FE prism model’s compressive stress-strain
The stress-strain relationship presented in Figures 6(a-b) revealed that the FE dog-bone specimen underwent
similar deformation like the experimental dog-bone specimen. The FE dog-bone specimen has a linear stress-
strain curve up to a tensile stress of 8.8N/mm2 against the experimental tensile stress of 8.5N/mm2; and this
means that even after the appearance of crack in the critical section of the dog-bone specimen at 7.05N/mm2
6
INNOVATION IN SCIENCE AND TECHNOLOGY JAN. 2024 VOL.3, NO.1
stress and 0.003 strain, the deformation of the specimen was still elastic. The FE dog-bone specimen’s stress-
strain behaviour became non-linear beyond the tensile stress of 8.8N/mm 2 until it attained its ultimate tensile
strength of 9.33N/mm2 at a strain of 0.008. The ultimate tensile strength of the FE dog-bone specimen was just
0.1% lower than the experimental ultimate tensile strength of 9.34N/mm 2; and this showed the high correlation
between the FE model and the experimental specimen. The FE dog-bone specimen after attaining its ultimate
tensile strength then underwent strain softening until it could no longer bear the applied load at a tensile stress
and strain of 4.6N/mm2 and 0.042 respectively. Critically comparison analysis of the FE dog-bone performance
curve (Figure 6(a)) with the experimental dog-bone performance curve (Figure 6(a)) revealed that the stress-
strain behaviour during strain softening immediately after attaining ultimate tensile strength was slightly
different. Whereas the FE model exhibited smooth strain softening, the experimental specimen exhibited zig-zag
strain softening; and this may be due to the unavoidable imperfection with experiment in terms of the specimen
shape, load application and boundary conditions among many other experimental errors.
10
9
8
Tensile stress (N/mm2)
7
6
5
4
3
2
1
0
0 0.02 0.04 0.06
Tensile strain
(a) FE dog-bone model’s tensile stress-strain
The vertical direction stress of the prism shown through its contour in Figure 7(a) revealed that the vertical
stresses were basically concentrated around the middle part of the prism with gradual distribution to other parts
7
INNOVATION IN SCIENCE AND TECHNOLOGY JAN. 2024 VOL.3, NO.1
of the prism. This vertical direction stress distribution may be the reason why the prism’s crack pattern appeared
the way it was in Figure 4(b). For the dog-bone specimen, the vertical direction stress contour presented in
Figure 7(b) showed that the stresses were distributed around the same location within the notched section of the
specimen where the ductile damage of the UHPFRC occurred. This stress contour of the specimen densely
concentrated close to the loaded end section means that there is low rate of distribution to other part of the
specimen; and this may be the reason while the tensile stress-strain performance curve of the FE dog-bone
specimen shown in Figure 6(a) failed to exhibit ductile failure through strain hardening.
4. Conclusion
This study used ABAQUS package to model and simulate the strength properties of UHPFRC with micro and
hooked-ends steel fibers; and the conclusions drawn from the compressive and tensile performance of the
simulated specimens are presented as follows:
(1) The surface of the UHPFRC specimens remained largely un-damaged even after the applied load exceeded
their ultimate compressive and tensile strength.
(2) The FE UHPFRC cube has an ultimate compressive strength of 260N/mm 2, about 7% higher than the
experimental strength
(3) The FE prism specimen exhibited linear stress-strain behaviour up to 89% of its peak strength as well as non-
linear stress-strain behaviour with both strain hardening and strain softening.
(4) The FE dog-bone specimen despite having ultra-high tensile strength did not undergo strain hardening phase
as the vertical direction stresses were mainly distributed around the loaded section of the specimen.
(5) The FE specimens only had slight deviations from the experimental specimens in terms of ultimate
compressive and tensile strength.
(6) The overall behaviour of the FE specimens in terms of deformation and strength performance are in perfect
agreement with the experimental specimens. So FE modelling and simulation can be used to conduct extensive
studies on UHPFRC properties and performances.
References
AFGC, (2002). French interim recommendations of ultra-high performance fiber reinforced concrete (UHPFRC),
Paris: French Association of Civil Engineers.
ASTM C39/C39M, (2014). Standard test method for compressive strength of cylindrical concrete specimens,
West Conshohocken: American Society for Testing and Materials International.
Azmee N M, Shafiq N, (2018). Ultra-high performance concrete: from fundamental to applications. Case Studies
in Construction Materials, 9, pp. 1-15. https://doi.org/10.1016/j.cscm.2018.e00197.
Bahraq A A, Al-Osta M A, Ahmad S, Al-Zahrani M M, Al-Dulaijan S O, Rahman M K, (2019). Experimental
and numerical investigation of shear behavior of RC beams strengthened by ultra‑high performance
concrete. International Journal of Concrete Structures and Materials, 13(6), pp. 1-19.
https://doi.org/10.1186/s40069-018-0330-z.
8
INNOVATION IN SCIENCE AND TECHNOLOGY JAN. 2024 VOL.3, NO.1
Birtel V, Mark P, (2006). Parameterised finite element modelling of RC beam shear failure. Proceedings of 2006
ABAQUS users’ conference, pp. 95-108.
Chowdhury M A, Islam M M, Zahid Z I, (2016). Finite element modelling of compressive and splitting tensile
behavior of plain concrete and steel fibre reinforced concrete cylinder specimens. Advances in Civil
Engineering, 2016(6579434), pp. 1-11. https://doi.org/10.1155/2016/6579434.
El-Helou R G, Haber Z B, Graybeal B A, (2022). Mechanical behaviour and design properties of ultra-high
performance concrete. ACI Materials Journal, 119(1), pp. 181-194.
Graybeal B A, Baby F, (2013). Development of a direct tension test method for UHPFRC. ACI Materials
Journal, 110, pp. 177-186.
Hashim D T, Hajezi F, Lei V Y, (2020). Simplified constitutive and damage plasticity models for UHPFRC with
different types of fiber. International Journal of Concrete Structures and Materials, 14(45), pp. 1-21.
https://doi.org/10.1186/s40069-020-00418-9.
Krahl P A, Carrazedo R, El Debs M K, (2018). Mechanical damage evolution in UHPFRC: experimental and
numerical investigation. Engineering Structures, 170, pp. 63-77.
https://doi.org/10.1016/j.engstruct.2018.05.064.
Lubliner J, Oliver J, Oller S, Onate E, (1989). A plastic-damage model for concrete. International Journal of
Solids and Structures, vol. 25 no. 3, pp. 299-326.
Naeimi N, Moustafa M, (2019). Uniaxial compression behavior of ultra-high performance concrete confined by
steel spirals. Proceedings of the 2nd International Interactive Symposium on UHPC, Albany, New York,
United States, pp. 1-8.
Prem P R, Bharatkumar B H, Iyer N R, (2012). Mechanical properties of ultra-high performance concrete. World
Academy of Science, Engineering and Technology International Journal of Civil and Environmental
Engineering, 6(8), pp. 676-685.
Raja L V N, Sujatha T, (2014). Study on properties of modified reactive powder concrete. International Journal
of Engineering Research and Technology, 3(10), pp. 937-940.
Rossi P, Arca A, Parant E, Fakhri P, (2004). Bending and compressive behaviors of a new cement composite.
Cement and Concrete Research, 35, pp. 27-33.
Sadrekarimi A, (2004). Development of light weight reactive powder concrete. Journal of Advanced Concrete
Technology, 2(3), pp. 409-417.
Shi Z, Su Q, Kavoura F, Veljkovic M, (2023). Uniaxial tensile response and tensile constitutive model of ultra-
high performance concrete containing coarse aggregate (CA-UHPC). Cement and Concrete Composites,
136(104878), pp. 1-17. https://doi.org/10.1016/j.cemconcomp.2022.104878.
Tao Y, Chen J, (2015). Concrete damage plasticity model for modelling FRP-to-concrete bond behaviour.
Journal of Composites for Construction, 19(1), pp. 1-13.
Tian X, Fang Z, Zhou T, Xiang Y, (2023). Behavior and constitutive model of ultra-high-performance concrete
under monotonic and cyclic tensile loading. Construction and Building Materials, 389(131634), pp. 1-22.
https://doi.org/10.1016/j.conbuildmat.2023.131634.
Copyrights
Copyright for this article is retained by the author(s), with first publication rights granted to the journal.
This is an open-access article distributed under the terms and conditions of the Creative Commons Attribution
license (http://creativecommons.org/licenses/by/4.0/).