Capita Selecta Steel and Aluminium Structures

CIE5122, Lecture S4, 2018

Stability and resistance of columns and shells
M. Veljkovic, Section: Steel and composite structures
Dept. Engineering Structures

Ilustrative image 2

1
REFERENCES FOR SHELL DESIGN
VERIFICATION

• High-Strength Steel Tower for Wind Turbines, HISTWIN+ report, 2015, Ch.3
page 43-50
• European Committee for Standardization: EN 1993-1-6: Eurocode 3 –
Design of Steel Structures, Part 1-6: Strength and Stability of Shell
Structures, Brussels, Belgium, 1993.

2
CONTENT

• Solution of home exercises, S2, S3, S4

• Shell stability
− Influence of imperfections
− Design according to EC3 part 1-6
− Ultimate limit states
− Types of analyses
• Solved numerical example 5
• Concluding questions
• Home exercise, S5

3
BUCKLING RESISTANCE

Indifferent Stable path Unstable path

post-critical path Favourable influence Unfavourable influence
of spatiality of spatiality
20
SNAP-TROUGH INSTABILITY
Eccentric loading on arch
Geometry of the structure changes as the load increase

bifurcation

21
 RELATIVE STRENGTH FOR VARIOUS
STABILITY PHENOMENA

1
σcr 1
= 2 Euler’s curve
0,9 fy λ
0,8 Plate:buckling
lokal local buckling
χ
Beam: lateral
vippning
0,7
χLT tors. buckling
0,6
ρ böjknäck-
Column: global
0,5 ningC-curve
buckling,

0,4
skalbuckling
Tower: shell
buckling
0,3

0,2

0,1

0
0 0,5 1,0
λ 1,5

22
SHELL BUCKLING ASSESMENT

• Hand-calculation
• (Mixed)
• (FEA)

23
BUCKLING OF A CYLINDRICAL SHELL
• Axial compression

The critical stress:

• Bending moment
Flügge: Brazier:

24
IMPERFECT SHELL BUCKLING

• Elastic behaviour
• Elastic-plastic behaviour
• Parameters: α, β, λ0, η

25
ULTIMATE LIMIT STATES OF A SHELL
STRUCTURE
• Plastic limit state (LS1)
• the capacity of the structure to resist the actions is exhausted by
yielding of the material.
• Cyclic plasticity limit state (LS2)
• relatively low number of repeated cycles of loading and unloading
produce yielding in tension and compression at the same point.
• Buckling limit state (LS3)
• loss of stability under compressive membrane or shear membrane
stresses in the shell wall
• Fatigue limit state (LS4)
• low stress ranges in large number of cycles leading to cracking of
the structure components (welds, bolts,...)

26
CONTENTS OF EC3 PART 1-6
1. Introduction
2. Basis of design and modelling
3. Materials and geometry
4. Ultimate limit states in steel shells
5. Stress resultants and stresses in shells
6. Plastic limit state (LS1)
7. Cyclic plasticity limit state (LS2)
8. Buckling limit state (LS3)
9. Fatigue limit state (LS4)
ANNEX A (normative) - Membrane theory stresses in shells
ANNEX B (normative) - Additional expressions for plastic collapse resistances
ANNEX C (normative) - Expressions for linear elastic membrane and bending
stresses
ANNEX D [normative] - Expressions for buckling stress design

27
APPLICATION TO WEC TOWERS

• Generally in in a WEC tower, the Buckling limit state (LS3) is the

governing (compared to the Plastic limit state (LS1))
• Cyclic plasticity limit state (LS2)
• The nature of the loads arising from operation of the turbine are highly
cyclic.
• Low cycle fatigue is rarely present in steel tubular towers for wind
energy converters.
• Fatigue limit state (LS4)
• Welded joints in the shell (longitudinal and transversal) and in
connections have low design class: 71 – 80 according to EC3 Part 1-9.
• The fatigue endurance is most often the governing one for the design.
• This gives limitation of use of High Strength Steels for the cylindrical
shell.

28
TYPES OF ANALYSES IN EC3 PART 1-6

• Several types of analysis can be conducted following the structure

complexity and regarding the failure mode that is considered:

• Linear elastic Analysis (LA)

• Linear elastic Bifurcation Analysis (LBA)

• Materially Nonlinear Analysis (MNA)

• Geometrically and Materially Nonlinear Analysis (GMNA)

• Geometrically and Materially Nonlinear Analysis with

Imperfections included (GMNIA)

29
IMPERFECTIONS
• Imperfections are depend on Fabrication
Tolerance Quality Class.
• Three types of geometrical imperfections:
• Out-of-roundness
• deviation from circularity
• Eccencitries
• deviations from a continious middle
surface
• Local dimples
• local normal deviations from
nominal middle surface

30
PRODUCTION, ASSEMBLING, TRANSPORT
AND EXECUTION

tmax = 50mm

31 31
PLASTIC LIMIT STATE (LS1)

• Strength of the structure when stability is not important.

• Two potential failure modes:
• Tensile rupture or compressive yield,
• Plastic collapse mechanism involving bending.
• Where holes for fasteners are present additional check need to be
carried out taking into account the net cross section.

32
PLASTIC LIMIT STATE (LS1)

• Types of analyses / calculation of stresses:

• Equations – Annex A
• Linear Analysis of stresses (LA)
• elastic bending theory
N Ed M Ed
σ x , Ed = +
A Wel
• underestimated resistance for bending
• Materially Nonlinear Analysis (MNA)
• plastic resistance
• better estimation of true resistance to bending loads

33
PLASTIC LIMIT STATE (LS1)

• In wind towers the most important are the membrane stresses.

• Three components of membrane forces/stresses:
• Meridional (longitudinal) - x
Meridional
direction

Circumferential
• Circumferential (tangential) - θ direction

• Shear (in plane)

• Surface stresses may be important in the connection zone!

34
PLASTIC LIMIT STATE (LS1)
• Interaction – von Mises equivalent stress:

• In towers the circumferential and shear stresses are practically

negligible, but need to be checked!
• Bending moments in the tower are dominant compared to shear
and torsion.
• Influence of local wind pressure on the tower wall is small
compared to load due to operation of the turbine.
• Verification:

35
 PLASTIC LIMIT STATE (LS1)
• Materially Nonlinear Analysis (MNA)
• idealised elastic-plastic material behaviour
• load amplification factor for design loads is obtained rRMNA
pl

• MNA offers higher resistance: rRMNA

pl
≥ r LA
R pl

MATERIAL MODEL

36
BUCKLING LIMIT STATE (LS3)
• Verification of local buckling resistance of the
Application in
shell in steel tubular tower. WEC towers:
• Three design approaches:
• Stress limitation
Verification of
• analogy to uniform member in compression the shell
in EC3 Part 1-1.
• MNA+LBA
• analogy to the general method in EC3 Part
1-1 Verification of
• GMNIA door oppenings

• Geometrically and Materially Nonlinear

Analysis with Imperfections included

37
BUCKLING STRESS LIMITATION (LS3)
• Linear Analysis (LA) or Annex A is used to calculate the membrane
stresses in the shell, same as for the LS1.
• The design check is made with regards to design resistances
taking into account buckling strength:
• For individual stress components:

• Interaction of components:

38
DESIGN RESISTANCES (LS3)

• Design resistances for stress components are obtained using

buckling reduction factors c taking into account:
• Imperfections depending on the Fabrication Tolerance Quality Class
• Boundary Conditions of the cylindrical shell

• The partial safety factor may be defined in the National Annex. The
recommended value is:
• Yield strength depends on the steel grade, and on the thickness of
the shell as well.

39
BUCKLING REDUCTION FACTORS (LS3)

• For each stress component buckling reduction factors are

obtained separately.
• The buckling reduction factors are determined as a
function of the relative slenderness of the shell:

Annex D
40
RELATIVE SLENDERNESS OF THE SHELL
(LS3)
• The relative shell slenderness parameters for different stress
components:

• Meridional

• Circumferential

• Shear

• Elastic critical buckling stresses can be obtained using:

• appropriate expressions in Annex D,

• LBA under buckling relevant action.

41
CRITICAL MERIDIONAL BUCKLING STRESS–
ANNEX D
• The elastic critical meridional buckling stress:

• The parameter Cx depends on effect of boundary conditions and

dimensionless length parameter w:

Boundary conditions

dimensioneless length parameter

42
EFFECT OF BOUNDARY CONDITIONS –
ANNEX D
• The parameter Cxb depends on effect of boundary conditions:

43
EQUIVALENT LENGTH – ANNEX D

• Steel tubular towers for WEC are long cylinders with stepwise
variable wall thickness.
• Each cylindrical section j of length lj, for buckling in the
meridional direction should be treated as an:
• equivalent cylinder of overall length l = L,
• with uniform wall thickness t = tj
• Considering the connections in tower:
• The ring flange connection - L should be the length of the
assembling segment.
• The friction connection - L should be the length of the
whole tower.
• For long equivalent cylinders the parameter Cxb should be
conservatively taken as Cxb = 1!

44
MERIDIONAL BUCKLING PARAMETERS –
ANNEX D
• Meridional elastic imperfection reduction factor:

• Characteristic imperfection amplitude Dwk depends on the

fabrication quality parameter Q:

• Other parameters should be taken as:

45
 Equivalent length
parameter

Boundary
conditions

Critical meridional Realtive

buckling stress slenderness

Imperfections

FLOWCHART FOR
Reduction DESIGN CHECK
factor
USING BUCKLING
STRESS
LIMITATION
Design resistance

46
EQUIVALENT CYLINDER
• Transformation of steeped cylinders into equivalent cylinder EN
1993-1-6:2007

47
EQUIVALENT CYLINDER
• Assume that the thickest/thinnest plate ratio is smaller than 1,5 the
part (24,5 m long) is made of lengths with the constant thickness

Nomogram in
Annex D 2.3.1

48
EXAMPLE 5 RESISTANCE OF A TOWER SEGMENT
A WET is made of 4 cylindrical parts, each 24.5 m long.
Assume that an equivalent thickness of the tower part is 17 mm
and the radius 2096 mm. Steel grade used is S355.

At the most loaded cross section the axial force and

corresponding bending moment are:
N Ed = 4190kN
M Ed = 41815kNm

Check meridional and circumferential buckling stress state.

49
50
51
MERIDIONAL STRESS STATE

52
CIRCUMFERENTIAL STRESS STATE
• Transformation of steeped cylinders into equivalent cylinder EN
1993-1-6:2007

53
CRITICAL CIRCUMFERENTIAL STRESS STATE
• Critical cross-section

54
Assume: The non-uniform distribution of pressure resulting
from external wind loading on cylinders for the purpose of
shell buckling design, is substituted by an equivalent
uniform external pressure. The maximum design wind
pressure qw,max = 0.72 kN/m2 is obtained in the load
analysis for the wind speed 23.5m/s.

55
SHEAR STRESSES

f yk
43
3

56
SHEAR STRESSES

39 N/mm2

0.14 < 1.0

57
INTERACTION CHECK

58
QUESTIONS

• What are the four limit states covered by the EC3 Part 1-6?
• Which limit states are the most important for the wind turbine
tower?
• What does the GMNA stands for?
• What are the geometrical imperfections in towers dependent on?
• Which stress components are the most important in a wind turbine
tower?
• Which analysis type (design approach) is most conservative for the
plastic limit state?
• How does the boundary conditions of the shell reflect it’s critical
buckling stress?
• What is the level of magnitude of the design circumferential
buckling strenght?

59
QUESTIONS

• What are major differences in assessment of the critical load of

columns and shells, according Eurocode EN 1993-1-*?
• What are major differences in assessment of the resistance of
columns and shells, according Eurocode EN 1993-1-*?

• Make assessment of the following formula and compare it with the

critical stress according Eurocode EN 1993-1-* for assumed
parameters of your choice:

60
HOME EXERCISE 5
Calculate characteristic buckling resistance of a shell loaded by axial centric force.
Assume that fabrication quality of the shell is B class. Geometry, including the
boundary conditions, of the shell are given in table below. Assume any other data if
necessary.

1. Draw the flow chart

including relevant formulas for
buckling due to
circumferential and shear
stresses.
2. What is the most
dominating effect for the
buckling resistance of the
supporting tower for wind
turbines and what that
depends on?

61