International Journal of Mechanical Sciences 89 (2014) 177193

Contents lists available at ScienceDirect

International Journal of Mechanical Sciences

journal homepage: www.elsevier.com/locate/ijmecsci

Theoretical prediction and crashworthiness optimization of multi-cell

square tubes under oblique impact loading
TrongNhan Tran a,c, Shujuan Hou a,b,n, Xu Han a,b,n, NhatTan Nguyen b,d, MinhQuang Chau c
a

State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha, Hunan 410082, PR China
College of Mechanical and Vehicle Engineering, Hunan University, Changsha, Hunan 410082, PR China
c
Faculty of Mechanical Engineering, Industrial University of Ho Chi Minh City, Go Vap District, HCM City, Vietnam
d
Center for Mechanical Engineering, Hanoi University of Industry, Tu Liem District, Hanoi, Vietnam
b

art ic l e i nf o

a b s t r a c t

Article history:
Received 7 April 2014
Received in revised form
22 August 2014
Accepted 27 August 2014
Available online 20 September 2014

Multi-cell square tubes under dynamic oblique impact loading were studied in our work. The theoretical
predictions of mean crushing force, mean horizontal force, and mean bending moment were proposed
by dividing the prole into basic angle elements based on a Simplied Super Folding Element (SSFE)
theory. The formulas of an oblique impacting coefcient () with a load angle of 151 were proposed based
on the geometric parameters, the inertia effect and the oblique loading angle by taking the effect of
oblique loading and dynamic crushing into account for aluminum alloy tubes. A new method was
proposed to nd out a knee point from Pareto set with maximizing the reex angle. The optimal
congurations of multi-cell tubes were analyzed under axial and more than one oblique impact loadings.
The results showed that the FE numerical results agreed well with the theoretical predictions.
& 2014 Elsevier Ltd. All rights reserved.

Keywords:
Crashworthiness
Theoretical prediction
Multi-objective optimization
Multi-cell tube
Energy absorption
Oblique impact loading

1. Introduction
Thin-walled tubes were widely used as energy absorber during
the past two decades. Extensive efforts by Wierzbicki and Abramowicz [1], Abramowicz and Jones [2,3], Guillow et al. [4], DiPaolo et al.
[5], Krolak et al. [6], were conducted to investigate the crushing and
energy absorption characteristics of the thin-walled tubes subjected
to axial impact loading by using experimental, theoretical and
numerical methods. For multi-cell tube, Wierzbicki and Abramowicz
[1] concluded that the number of angle elements on cross-section
of tube, to a certain extent, decided the effectiveness of the energy
absorption. The quasi-static axial crushing of single-cell, double-cell
and triple-cell hollow tubes and corresponding foam-lled tubes
were examined by Chen and Wierzbicki [7]. The work of Chen and
Wierzbicki [7] showed that the multi-cell tube increased the specic
energy absorption SEA by approximately 15%, compared to hollow
tube. Therefore, it is necessary to design multi-cell thin-walled tubes
as weight-efcient energy absorption components. In order to get a
simplication to replace the kinematical admissible model of Super
Folding Element (SFE) theory, the Simplied Super Folding Element
n
Corresponding authors at: College of Mechanical and Vehicle Engineering,
Hunan University, Changsha, Hunan 410082, PR China.
E-mail addresses: shujuanhou@hnu.edu.cn (S. Hou),
hanxu@hnu.edu.cn (X. Han).

http://dx.doi.org/10.1016/j.ijmecsci.2014.08.027
0020-7403/& 2014 Elsevier Ltd. All rights reserved.

(SSFE) theory was proposed by comprising three extensional triangular elements and three stationary hinge lines [1]. Assuming that
each panel and angle element has the same role, the theoretical
prediction of the mean crush force was deduced by dividing the
cross-sectional tube into distinct panel section and basic angle
element. Kim [8], Jensen et al. [9], Karagiozova and Jone [10], Zhang
et al. [1115], Naja and Rais-Rohani [16] and other authors have
investigated the multi-cell thin-walled tubes under axial impact
loadings and made many valuable conclusions. The progressive
collapse of tubes under axial loadings was summarized by Karagiozova
and Alves [17]. Otherwise, the global bending was an undesirable
energy-dissipating mechanism. Alternatively, the desirable energydissipating mechanism was the stable and progressive wrinkle deformation. Kim and Reid [18] also proposed an approximate method to
predict the bending collapse, the crumpling moment and the energy
absorption for tubes subjected to pure bending. Recently, Tran et al.
[19] utilized the Simplied Super Folding Element (SSFE) theory to
estimate the energy dissipation of angle elements in the theoretical
predictions and crashworthiness optimization of multi-cell triangular tubes.
Nevertheless, multi-cell thin-walled tubes as an energy absorber
normally bear the oblique impact loading. At that time, the tubes
are subjected to both axial force and bending moment. In case the
tube experiences a global bending, the energy absorption would
be smaller [20]. Therefore, it is necessary to study the mechanical

178

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

Nomenclature
2H
a
t
L0
b
B
d
SEA
Eb, Em
EA

wavelength
side length
wall thickness
tube lengths
panel width
sum of side-length and internal web lengths
crushing displacement
specic energy absorption
bending and membrane energy
total strain energy
effective collapse coefcient
rotation angle at bending hinge line

property of tubes under oblique impact loadings. According to the

aspects of numerical simulation, Han and Park [21] studied the
oblique behavior of thin-walled square mild steel tube. Their study
showed that the axial progressive collapse should be transferred to
global bending collapse at the critical load angle. Reyes et al. [22,23]
carried out extensive numerical and experimental analyses on the
square tubes under quasi-static oblique loading. Their researches
concluded that the energy absorption dropped sharply when load
angle exceeded the critical value. Nagel and Thambiratnam [24] also
investigated square tubes under dynamic oblique impact loading.
Their results made a conclusion that the impact velocity had no
signicant effect on the critical load angle. Qi et al. [25] investigated
the crash behavior of the single-cell straight, the single-cell tapered,
the multi-cell straight, and the multi-cell tapered tubes by the
numerical method. Their work showed that the MCT tube had the
best crashworthiness performance under oblique loading. Yang and
Qi [26] optimized the crashworthiness of the empty and foam-lled
thin-walled square tubes under oblique impact loading. Song [27]
also studied the windowed square tubes subjected to oblique
impact loading by the numerical method. Song utilized the variables for the sake of the load angle, the geometrical parameters of
window, and the impact velocity. In addition, a multi-objective
optimization design (MOD) method was employed for the crashworthiness design of multi-cell thin-walled tubes [2830].
Most of the above studies just emphasized independently
numerical simulations, theoretical analysis or experiments. In this
paper, theoretical predictions, numerical analysis and optimization
design were combined together for multi-cell square tubes under

PCF
P(x)
P
Pa
Pm
Ph
M
M0

0
y, u

initial peak crushing force

instantaneous crushing force
mean crushing force at load angle
axial crushing force
mean crushing force
horizontal force
bending moment
fully plastic bending moment
oblique impacting coefcient
reex angle
angle formed by internal panels
load angle
ow stress of material
yield strength and ultimate strength of material

oblique collapses. Based on the SSFE theory [19], theoretical

expressions for mean crushing force of tubes under oblique
loading were derived by dividing the prole of tubes into the
basic angle elements (the right corner, 3-, T-shape, criss-cross,
4-, 5- and 6-panel angle elements). The theoretical solutions for
tube I, II and III were proposed for the calculation of the bending
resulted by the mean horizontal force and the mean bending
moment. Dynamic nite element analyses were performed by
ANSYS/LS-DYNA. A new method was proposed to get a knee
point from a Pareto set with maximizing the reex angle .

2. Theoretics
2.1. Theoretical prediction of multi-cell square tubes
For the predictions of collapse of thin-walled multi-cell square
tubes, the SSFE theory was applied to calculate the mean crushing
force [31]. In this theory, the variation of wavelength 2H and wall
thickness for different lobes was ignored, which was assumed to
be constant respectively. Each panel (ange) and angle element
had the same role during the collapse. The proles of tubes were
divided into seven basic elements (right-corner, 3-, T-shape, Crisscross, 4-, 5- and 6-panel angle element) to evaluate dissipated
bending and membrane energy during the collapse of a fold, as
shown in Fig. 1.
Regarding tubes under an oblique impact loading, the equilibrium of the element is expressed via the principle of virtual work,

Fig. 1. Cross-sectional geometry of multi-cell square tubes and typical angle elements.

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

179

which is E_ ext E_ int . Therefore, the external work from compression

to create a complete collapse of a single fold must be equal to the
sum of energy dissipation in bending and membrane of all panels.
Therefore,

hinge lines (Figs. 2 and 3). According to the assumed kinematics

of the SSFE theory, the dissipated bending energy of each panel
can be calculated by summing up the energy dissipation at the
bending hinge line, that is

1
P m 2H Eb Em

Efb M 0 bi 2M 0 b

where Pm, 2H, Eb and Em are, respectively, the mean crushing force,
the wavelength, the bending energy and the membrane energy.
is the effective collapse coefcient. In fact, the panel of folding
element after deformation is not entirely attened as displayed in
Fig. 2. Hence, the available collapse displacement is smaller than
2H. In this study, the value of is taken as 0.75, that lies in the
range of 0.70.75 [1].

i1

i1

where 2 is the rotation angle at the bending hinge line.

2.1.1. The bending energy of tube structure

In this study, the SSFE theory was applied to calculate the
dissipated energy in bending of each panel. The basic folding
element (BFE) includes the triangular elements and the bending

Bending hinge line

= 2
Fig. 2. Bending hinge line and rotation angle on basic folding.

Fig. 5. Collapse mode and extensional elements of criss-cross angle element.

Fig. 3. Basic folding element: (a) Asymmetric mode [7], and (b) Symmetric (extensional) mode.

Fig. 4. (a) Collapse mode of T-shape element, (b) Extensional element of T-shape element and (c) 3-panel angle element.

180

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

Since each panel (web) contributes the similar function and

multi-cell tubes are created by n panels (Fig. 1), the bending
energy of multi-cell tubes derived from Eq. (2) is
Etube
b

2M 0 nb 2M 0 B

wavelength crushing, was

E3panel
4M 0
m



H2 
1 2 tan =2
t

where b and B are, respectively, the panel width and the sum of
side-length and internal web lengths. M0 0t2/4 is the fully plastic
bending moment.
2.1.2. The membrane energy of Angle element
2.1.2.1. The membrane energy of right-corner, 3-panel, T-shape and
criss-cross angle element. In order to get the membrane energy of
right-corner in the SSFE theory, the basic folding element (BFE) was
formed by using the triangular elements and the bending hinge lines
(Fig. 3). Two possible collapse modes (asymmetric and symmetric)
were proposed in the BFE. The asymmetric and symmetric modes
can be referred as the quasi-inextensional and extensional modes,
respectively. In regards to the asymmetric mode and the symmetric
mode, the three triangular elements and two triangular elements
were both developed for each panel (web) after the deformation.
Simultaneously, it was assumed that the role of each panel of multicell tubes was the same. Accordingly, the dissipated energy in
membrane Em of right-corner element in the case of asymmetric
mode, during one wavelength crushing, could be calculated by
integrating the area of triangular elements (the shaded areas as
shown in Fig. 3(a)). Then, we get
Z
H2
4
Easym
0 tds 4M 0
m_rc
t
s
Then, the energy dissipation in membrane of right corner
element in the case of symmetric mode (Fig. 3(b)) was estimated
as
Z
H2
5
Esym
0 tds 8M 0
m_rc
t
s
The dissipation in membrane energy of 3-panel angle element
(Fig. 4c) was estimated by Zhang et al. [32]. According to this work,
the membrane energy of 3-panel angle element, during one

The structure of T-shape element was formed by a combination

of one right-corner element and one additional panel, as shown in
Fig. 4(a) and (b). Therefore, the dissipated energy in membrane of
T-shape element can be calculated by summing up the rightcorner element's membrane energy and one additional panel's
membrane energy [31]. On the other hand, each panel often
deforms through symmetric mode, and has the similar role in
the structural deformation. Consequently, the dissipated energy in
membrane of T-shape element, during one wavelength crushing, is
the sum of membrane energy absorbed by all three panels. Then
we get
ETshape
12M 0
m

H2
t

Being a symmetric structure and formed by four panels, the

energy dissipation in membrane of a criss-cross element can be
calculated by summing up the membrane energy of all four panels
(Fig. 5). Nevertheless, the criss-cross element is also created by
two right-corner elements that deform in a symmetric mode.
Moreover, each angle element contributes the similar role in the
structural deformation. Accordingly, the membrane energy of
criss-cross element, during one wavelength crushing, can be
determined by summing up the membrane energy of two rightcorner elements in the case of symmetric mode, that is
Ecc
m 16M 0

H2
t

2.1.2.2. The membrane energy of 4-, 6- and 5- panel angle

element. The membrane energy of 4-panel angle element (Fig. 6)
was analyzed by Tran et al. [31]. The structure of this angle element
was constructed by two additional panels and one right-corner angle
element. Thus, energy dissipation in membrane of 4-panel angle
element was calculated by summing up the membrane energy of

Fig. 6. (a) Collapse mode of 4-panel angle element, and (b) Extensional elements.

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

two additional panels and one right-corner angle element for a

symmetric mode. Consequently, the membrane energy Em of 4-panel
angle element, during one wavelength crushing, was evaluated by
integrating the triangular areas as shown in Fig. 6(b). Then we get


H2
1
rc
9
1
E4panel
Em
2Eapanel
8M 0
m
m
cos
t
Because of the symmetric structure and similar role of each
panel in the structure, the energy dissipation in membrane of
6-panel angle element can be calculated by summing up the
membrane energy of all six panels. However, the 6-panel angle
element was formed by a combination of two right corner
elements and two additional panels, as shown in Fig. 7. In
addition, it was assumed that each angle element had the similar
role in structure. Thus, the dissipated membrane energy can be
calculated by summing up the membrane energy of two right
corner elements acting on two additional panels and two right
corner elements. The deformation mode of right corner element in
6-panel angle element was symmetric. Regarding the additional
panel, it is too difcult to give a precise calculation for membrane
energy. Therefore, a simplied deformation model of additional
panels was assumed and the SSFE theory was used to solve this

181

problem. Represented in Fig. 7(b), the areas of ILP and IFP were
dened to be the extensional elements of two additional panels.
Hence, the dissipated energy in membrane of one additional panel
of 6-panel angle element, during one wavelength crushing, was
estimated by integrating the extensional areas. This was
Z
H2
H2
n
Eapanel
8M 0
10
0 tds 0 t
m
cos
t cos
s
Accordingly, the membrane energy of 6-panel angle element
[19] was


H2
2
sym
apaneln
11
2

2E

2E

8M
E6panel
0
m
m
m_rc
cos
t
As mentioned above, each angle element contributes the same
role in the structural deformation. Thus, the 5-panel angle element
was formed by a combination of one T-shape and one right corner
element (Fig. 8). In this case, the deformation mode of right corner
element in 5-panel angle element was a symmetric mode.
Furthermore, the membrane energy of T-shape and right-corner
ones in the case of symmetric mode can be determined. Hence, the
dissipated membrane energy of 5-panel one was estimated by

Fig. 7. (a) 6-panel angle element, and (b) Extensional elements of additional panel [19].

Fig. 8. 5-panel angle element.

182

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

Fig. 9. Flowchart of the MOPSO method.

summing up the membrane energy of one T-shape element and

one right-corner element. Then we get
E5panel
ETshape
Esym
m
m
m_rc 20M 0

H2
t

12

2.1.3. The mean crushing force

The theoretical expressions were developed for the mean
crushing forces of multi-cell tubes under oblique impact loading.
The prole of tube type I was formed by a combination of four
right-corner elements in a asymmetric mode, two 4-panel
angle elements, and two 5-panel ones, as shown in Fig. 1(a).
To substitute Eqs. (3), (4), (9) and (12) into Eq. (1), the theoretical
expression for mean crushing force of tube type I can be determined as


4panel
P mI 2H Etube
4Easym
2E5panel
b
m_rc 2E m
m

Fig. 10. Max reex angle based on denition of a knee point.

2M 0 B 16M 0



H2
H2
1
H2
40M 0
16M 0
1
cos
t
t
t

13

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

From Eq. (13), we can get another transformation, that is



P mI B H
8
B H
F
14
36
M0
H t
cos
H t
Applying the stationary condition of the mean crushing force
Pm/H, the half-wavelength can be determined as
s
B F
Bt
)H
15
0  2
t
F
H
To substitute the term H in Eq. (15) into Eq. (14), the proposed
expression of mean crushing force of tube type I is
p
M 0 B M 0 H
F
P mI

F 0:5 0 t 1:5 B0:5

16
H
t
2
where F 36 8= cos .
As indicated in Fig. 1(b), the prole of tube type II was created
by a combination of two right-corner angle elements in an
asymmetric mode, two 3-panel angle elements, two criss-cross
angle elements and four 4-panel ones. To substitute terms in Eqs.
(3), (4), (6), (8) and (9) into Eq. (1), the theoretical expression of
mean crushing force of tube type II is


3panel
4panel
P mII 2H Etube
2Easym
2Ecc
b
m_rc 2E m
m 4E m




H2
16
17
40 8 tan =2
2M 0 B 2M 0
cos
t
From Eq. (17), we can get another transformation, that is




P mII B H
16
B H
40 8 tan =2
G;
18

M0
H t
cos
H t
The half-wavelength can be received by using the stationary
condition of the mean crushing force Pm/H, that is
s
B G;
Bt
)H
0  2
19
t
G;
H
To substitute the place of H in Eq. (19) back into Eq. (18), the
theoretical expression of mean crushing force of tube type II is
p
M 0 L M 0 H
G;

G; 0:5 0 t 1:5 B0:5

20
P mII
H
t
2
where G; 40 8 tan =2 16= cos .

183

The prole of tube type III was constituted by a combination of

four 3-panel angle elements, two T-shape angle elements and one
6-panel one. To substitute terms in Eqs. (3), (6), (7), (11) into
Eq. (1), the theoretical expression for the mean crushing force of
tube type III is


P mIII 2H Etube
4E3panel
2ETshape
E6panel
b
m
m
m
2M 0 B 2M 0





H2
8
28 16 tan =2
cos
t

Then, an alternative form of Eq. (21) is





P mIII B H
8
B H
28 16 tan =2
Q ;

M0
H t
cos
H t

21

22

Postulating the stationary condition of the mean crushing force

Pm/H, the half-wavelength can be expressed as
s
B Q ;
Bt
)H
23
0  2
t
Q ;
H
From Eqs. (23) and (22), the theoretical solution of the mean
crushing force for tube type III is
p
Q ;
M 0 L M 0 H
24
P mIII

Q ; 0:5 0 t 1:5 B0:5

2
H
t
where Q ; 28 16 tan ; 2 8= cos .

2.2. Optimization design methodology

Among all the crashworthiness indicators, the energyabsorption is a vital analytical objective. It is expected that the
tube can absorb as much energy as possible with a structural total
weight m [28]. Therefore, the specic energy absorption SEA is
dened as
SEA

EA
m

25

In Eq. (25), EA denotes the total strain energy during crushing,

that can be determined from the crushing forcedisplacement

Fig. 11. Schematic of the computational model.

184

curve
Z
EA

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

d
0

P xdx

26

where P(x) is the instantaneous crushing force.

In addition, the initial peak crushing force (PCF) of multi-cell
square tubes can, to a certain extent, evaluate the impact characteristics. Meanwhile, the mean crushing force P at load angle is
also dened to be another crashworthiness indicator [29]. Mean
crushing force P can be formulated as
Z
E
1 d
P A
P xdx
27
d d 0
where d is the crushing displacement of tubes at a specic time.
For the nonlinear contact-impact mechanics problems, it is
very difcult to obtain the analytical expressions for the crashworthy indicators SEA and PCF. Therefore, the surrogate models

based on a response surface method (RSM) [2830] are an

effective mathematical regression in the multi-objective optimization design of this kind of issues.

2.2.1. Multi-objective optimization design (MOD) based on surrogate

models
For the safety of automobile industry, the smaller the PCF is, the
lower the deceleration is. Nonetheless, an increase in SEA often
leads the increase in PCF. On the other hand, SEA can interact with
PCF. As a result, It is too difcult for us to have two concurrently
optimized objectives. We have to impose the optimum on a Pareto
set. The multi-objective optimal problem of both maximizing SEA
and minimizing PCF was dened by the multi-objective particle
swarm optimization MOPSO methods [30]. A owchart showing
the procedure of MOPSO is provided in Fig. 9.

Fig. 12. Deformation process of three tubes.

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

185

The multi-objective optimal problem can be formulated as

8


1=SEA; PCF
>
< Minimize
s:t: 1 r t r2 mm
28
>
:
80 r a r 100 mm

point, such as a minimum distance selection method to get a

knee point from Pareto set [33]; the solution to nd a knee
point with maximum bend-angle [34]; or a modied multiobjective evolutionary algorithm [35]. However, if there is a great

where t and a are the cross-sectional wall thickness and side

length.

Table 1
Design matrix of thin-walled structures for crashworthiness.
n

2.2.2. Knee point

In many cases, the designer has to get a better solution (termed
as knee point) from the optimal solutions set based on their
requirements. There are several methods which can get a knee

Fig. 13. The crushing forcedisplacement curve of (a) tube I, (b) tube II and
(c) tube III.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

t (mm)

1
1.25
1.5
1.75
2
1
1.25
1.5
1.75
2
1
1.25
1.5
1.75
2
1
1.25
1.5
1.75
2
1
1.25
1.5
1.75
2

a (mm)

80
80
80
80
80
85
85
85
85
85
90
90
90
90
90
95
95
95
95
95
100
100
100
100
100

Tube type I

Tube type II

Tube type III

SEA
(kJ/kg)

PCF
(kN)

SEA
(kJ/kg)

PCF
(kN)

SEA
(kJ/kg)

PCF
(kN)

12.914
14.174
15.024
16.169
17.035
12.374
14.04
14.814
15.938
16.416
11.272
13.27
14.25
14.538
15.492
11.003
12.662
13.424
14.238
14.918
10.412
11.755
13.003
13.519
13.681

69.331
88.068
107.114
125.74
143.022
74.315
94.116
114.467
134.261
152.533
78.93
99.949
121.595
142.777
162.284
83.773
106.08
129.035
151.289
171.756
88.422
111.748
136.175
159.917
181.906

14.135
15.806
17.04
18.355
19.372
14.481
15.805
16.928
18.094
18.371
13.214
14.392
15.219
16.223
17.18
12.638
13.619
14.915
16.008
16.752
12.337
13.554
14.821
15.649
16.114

73.385
93.419
113.662
133.311
151.463
78.791
99.808
121.127
141.625
161.44
83.498
105.921
128.896
151.251
171.731
88.596
112.382
136.743
160.211
181.684
93.609
118.335
143.939
168.569
191.301

12.758
13.719
14.954
16.054
16.632
11.871
13.383
13.476
14.731
15.232
11.232
12.35
13.47
14.004
14.539
10.719
11.487
12.735
13.844
14.602
10.108
11.307
12.25
13.278
14.146

70.25
89.238
108.402
126.94
144.084
75.252
95.524
116.217
136.207
155.091
79.826
101.498
123.625
145.184
165.009
84.553
107.517
130.822
153.278
173.872
89.312
112.9
137.456
161.126
182.973

Fig. 14. Response surface of (a) PCF; and (b) SEA.

186

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

deviation among the orders of magnitude of different objectives,

these methods seem ineffective. In our study, a new method was
developed to determine a knee point by maximizing the reex
angle , as shown in Fig. 10. Mathematically, the formula is given
as
Maximize 3601 


a1 a2 b1 b2
s:t: cos qq
2
2
a21 b1 a22 b2

29

where is the internal angle formed by two lines f1 and f2.

3. Numerical simulation and crashworthiness optimization

3.1. Numerical simulation
For the multi-objective optimization design, the effects of
geometric parameters (the cross-sectional wall thickness t and side
length a) were studied by numerical simulations. Thus, FE model
was performed by using ANSYS/LS-DYNA to simulate the multi-cell

square tubes under the oblique impact loading. The structures were
thin-walled square tubes with 6, 8 and 6 cells, as shown in Fig. 1.
The cross-sectional side length a (80 mmrar100 mm) and wall
thickness t (1 mmrtr2 mm) were dened to be design variables.
The tube lengths L0 were dened to be a constant of 250 mm.
The tube was modelled by using Belytschko-Tsay four-node
shell elements with ve integration points in the element plane.
The material AA6060 T4 of tube was modelled by using material
model #24 (Mat_Piecewise_Linear_Plasticity) with mechanical
properties as: Young's modulus E 68,200 MPa, initial yield stress
y 80 MPa, the ultimate stress u 173 MPa, poisson's ration
0.3 and the power law exponent n 0.23 [36]. Since aluminum
is insensitive to the strain rate effect, this effect is neglected in the
nite element analysis. An automatic single surface contact was
used for the self-contact among the shell elements to avoid
interpenetration of folding generated during the axial collapse.
Simultaneously, an automatic node to surface contact between the
thin-walled column and rigid-wall was dened to simulate the
real contact. A friction coefcient of 0.3 is utilized for all contact
interaction. To generate enough kinetic energy as applied in
vehicle crashing, a lumped mass of 500 kg was attached to the

Fig. 15. Relative error curves for all the response surfaces at all the design sampling points.

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

end of the tube, whereas another end impacts on an oblique rigid

wall with an initial velocity of 10 m/s. The load angle is of 151 in
this study [25]. The three multi-cell sections are not axialsymmetric and therefore the oblique loading from different directions should be different. The schematic of the computational
model is shown in Fig. 11. In this paper, we just predicted the
theoretical solution for the loading condition as shown in Fig. 11.
As shown in Fig. 1, all tubes are multi-cell structures. Having
the same length, same side-lengh, and same wall thickness, tube I
and II are the same weight, while tube III is the heaviest one.
Fig. 12 shows the deformation process of three tubes at different
periods of time. There is no transition from progressive to global
bending collapse mode and no global bending collapse mode
appearing. Fig. 13 shows three corresponding crushing force
displacement curves of tube types I, II and III under oblique impact
loading. The crushing force, after touching the initial peak, falls

187

Table 2
Optimal results.
Type of
cross-section

Terms

Optimal design
variables (mm)

SEA
(kJ/kN)

PCF
(kN)

Type I

Approximate value
FE analysis value
RE
Approximate value
FE analysis value
RE
Approximate value
FE analysis value
RE

t 1.4, a 80

14.378
14.279
0.693
16.746
16.607
0.837
14.095
14.151
 0.396

98.523
98.779
 0.259
107.631
107.093
0.502
106.681
106.532
0.140

Type II

Type III

t 1.43, a 80

t 1.48, a 80

sharply and then uctuates periodically around the mean values

of crushing force corresponding to the formation. Finally, it
completes the collapse of folds one by one. In addition, a decrease
of mean crushing force in crushing forcedisplacement curve does
not appear. This can explain that there is no evidence of bending
phenomenon in these multi-cell thin-walled tubes under oblique
impact loading, as shown in Fig. 12. Fig. 13 also shows that the
exact value of the effective crushing distance on the crushing
forcedisplacement curve is not unique.
3.2. Crashworthiness optimization
To construct the quartic tting models for SEA and PCF, a series
of 25 design sampling points were selected in the design space, as
shown in Table 1. The crashworthiness analyses were performed
on those models to get response surfaces of the SEA and PCF.
Accordingly, the response surfaces of SEA and PCF for tubes I, II
and III are plotted in Fig. 14, where the response surfaces of SEA
and PCF behaved monotonically over the design domain. SEA and
PCF were inclined to a smaller side-length and a thinner wall
thickness. Of all the six RS functions, the maximum tting RE
interval is (  2.66%; 2.68%), which is considered acceptable. Fig. 15
shows relative error curves for all the response surfaces at all the
design sampling points.
The Pareto frontiers of 1/SEA vs. PCF for three types of tubes are
plotted in Fig. 16. At the same time, the Pareto frontiers show
evenly distributed solution points in the Pareto space. In fact, any
point on Pareto frontiers can be an optimum, and a range of
optimal solutions are supplied for the decision-maker. That is why
several solutions are suggested to detect the best result or a Knee
point, which have a large trade-off value in comparison with other
Pareto-optimal points. Accordingly, a method for identifying Knee
point with maximizing the reex angle was proposed as shown
in Fig. 10. The results of Eq. (29) show that Knee points with
maximizing reex angle for tube types I, II and III are 183.2581,
185.7081 and 184.5371, respectively as shown in Fig. 16. Deriving
from the results of Eq. (32), the optimal design variables of multicell square sections for tube types I, II and III under oblique loading
are presented in Table 2. The relative errors REs between FE
numerical results and RS tting values are also summarized in
Table 2. The REs show that RS tting has a high precision.

4. Validation and discussion

Fig. 16. Pareto fronts of 1/SEA vs. Peak crushing force for (a) tube I, (b) tube II and
(c) tube III.

Fig. 17 shows the schematic of force acting on tube at load angle

of . In this gure, the crushing force P under oblique impact
loading composes of the axial crushing force Pa and the horizontal
force Ph. The Pa and Ph cause the progressive collapse and the
bending of tubes, respectively. Consequently, the relationship

188

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

Fig. 17. Schematic of forces acting on tube.

for tube type I;

between P and Ph is
P h P sin

30

and the bending moment is generated by the horizontal force




a t
31
M P h x P h L0 
2 tan
In Section 2.1.3, the theoretical expressions of mean crushing
force in Eqs. (16), (20), and (24) have been created for tubes I,
II and III. Nevertheless, these expressions are applied under quasistatic loading cases, where the effects of the dynamic crushing and
oblique loading angle are not taken into account. With respect to a
dynamic loading, dynamic amplication effects including inertia
and strain rate effects should also be taken into account. In reality,
the aluminum alloy is not sensitive to strain rate; thus, strain rate
effect can be ignored. A dynamic enhancing coefcient was
brought in and inertia effect were taken into account [37,38]. This
coefcient is a variable for different geometric parameters of tube
structure as described by Langseth and Hopperstad [39] and
Hanssen et al. [40]. According to their works, the value of this
coefcient is in the range of 1.31.6 for AA6060 T4 extruded tubes
under the axial dynamic loading. Regarding the oblique impact
loading, the inuence of load angle must be taken into consideration. Thus, an oblique impacting coefcient was proposed to
express the relationship among the inertia effect, geometric
parameters and oblique loading angle under the oblique impact
loading. Obviously, it is too difcult to determine the precise value
for oblique impacting coefcient . In case of load angles exceeding
101, value of for tubes under oblique loading is relatively lower.
Basically, the larger a load angle is, the smaller is. Based on the
geometric parameters, the inertia effect and the oblique loading
angle, the following formulas for at load angle of 151 are given as
I 10nt=a 1:24n cos

32

II 10nt=a 1:17n cos

33

for tube type II; and

III 10nt=a 1:1n cos

34

for tube type III.

Thereby, the theoretical solution for tube type I under oblique
impact loading is,
p
F
P I I P mI I 0:5 0 t 1:5 B0:5
35
2
where F 36 8= cos .
From Eqs. (30), (31), (32) and (35), the mean horizontal force
and mean bending moment can be predicted respectively as
Pm
hI P I sin

36







at
at
m

P
L

sin

L

Mm
I
0
0
I
hI
2 tan
2 tan

37

The mean crushing force at load angle 151 for tube type II is
p
G;
38
P II II P mII II 0:5 0 t 1:5 B0:5
2
where G; 40 8 tan =2 16= cos .
Combining Eqs. (30), (31), (33), and (38), the mean horizontal
force and mean bending moment for tube type II are given as
Pm
hII P II sin

39







at
at
m

P
L

sin

L

Mm
II
0
0
II
hII
2 tan
2 tan

40

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

189

Table 3
Relative errors among FE numerical results and theoretical predictions for three types of tube.
n

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25

Tube type I

Tube type II

Tube type III

FE Num. result P
(kN)

Theo. prediction P
(kN)

Diff.
(%)

FE Num. result P
(kN)

Theo. prediction P
(kN)

Diff.
(%)

FE Num. result P
(kN)

Theo. prediction P
(kN)

Diff.
(%)

27.723
40.472
53.831
71.135
85.431
27.989
39.876
56.764
70.214
88.213
28.574
41.07
56.482
71.696
89.487
28.936
41.687
58.206
72.511
90.557
29.863
42.832
57.66
73.877
91.278

28.108
39.998
53.482
68.486
84.948
28.836
40.996
54.771
70.583
86.870
29.547
41.972
56.031
71.646
88.752
30.241
42.926
57.265
70.682
90.596
30.920
43.859
58.473
74.677
92.405

1.39
 1.17
 0.65
 3.72
 0.57
3.03
2.81
 3.51
0.53
 1.52
3.40
2.20
 0.80
 0.07
 0.82
4.51
2.97
 1.62
 2.52
0.04
3.54
2.40
1.41
1.08
1.23

32.978
47.958
64.049
82.66
103.756
34.142
48.253
65.957
85.15
106.289
34.297
49.362
66.316
87.045
108.125
35.819
50.528
69.054
89.107
110.863
36.307
51.663
70.244
89.982
111.905

33.450
47.670
64.831
82.852
101.666
34.306
48.840
65.538
83.716
103.905
35.141
49.984
66.814
85.842
106.099
35.958
51.103
68.259
87.832
108.251
36.756
52.199
69.675
90.087
110.363

1.43
 0.60
1.22
0.23
 2.01
0.48
1.22
 0.63
 1.68
 2.24
2.46
1.26
0.75
 1.38
 1.87
0.39
1.14
 1.15
 1.43
 2.36
1.24
1.04
 0.81
0.12
 1.38

26.316
40.005
51.56
68.516
84.141
27.078
40.448
52.349
69.16
85.262
27.825
40.974
52.924
67.845
86.014
28.497
41.518
54.137
69.988
86.657
28.968
42.515
55.068
72.652
88.164

27.363
39.090
52.474
67.463
84.017
28.049
40.025
53.673
68.939
85.779
28.720
40.940
54.848
70.386
87.509
29.375
41.836
56.001
71.807
89.210
30.017
42.714
57.131
73.203
90.882

3.98
 2.29
1.77
 1.54
 0.15
3.59
 1.05
2.53
 0.32
0.61
3.22
 0.08
3.64
3.75
1.74
3.08
0.77
3.44
2.60
2.95
3.62
0.47
3.75
0.76
3.08

For tube type III, the form of mean crushing force at load angle
151 is
p
Q ;
41
P III III P mIII 0:5 0 t 1:5 B0:5
2
where Q ; 28 16 tan =2 8= cos .
To substitute Eqs. (34) and (41) back into Eqs. (30) and (31), the
theoretical expressions of mean horizontal force and mean bending moment are given as
Pm
hIII P III sin

42







at
at
m
P III sin L0 
Mm
III P hIII L0 
2 tan
2 tan

43

The theoretical values of P for multi-cell tubes are determined

at the displacement of 50% in each specic case. These values are
dened as the equivalent constant force with a corresponding
amount of displacement. In Eqs. (35), (38) and (41), 0 is the ow
stress of material with power law hardening estimated by an
energy equivalent stress [41]:
r
y u
0:106GPa
44
0
1n
where y and u denote the yield strength and the ultimate
strength of the material, respectively; and n is the strain hardening
exponent.
Consequently, Eqs. (35), (38), and (41) are used to calculate
P for all cases at load angle 151. The relative errors between FE
numerical results and theoretical predictions for all cases of three
tubes are listed in Table 3. For tube I, the relative errors between
Eq. (35) and FE numerical results are ranging from  3.51% to
4.51%. For tube II, the relative errors between Eq. (38) and FE
numerical results are from  2.36% to 1.43%. For tube III, those are
from  2.29% to 3.98%. The relative errors are acceptable. Fig. 18
shows that there are a very close agreement between the theoretical predictions and the FE numerical results in these cases.

Fig. 18. Differences of FE numerical results and theoretical predictions: (a) tube I;
(b) tube II and (c) tube III.

190

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

Fig. 19. Final deformation of optimal tube type I at: (a) 01, (b) 151, (c) 301and (d) Crushing/mean crushing force-displacement curve.

Concerning oblique impact loading, the automobile industry

stipulates a load angle of 301 for an impact condition on bumper
system [22]. Thus, the tubes are subjected to both axial and
horizontal forces in an oblique impact condition. If the tube
experiences a global bending, the energy absorption will be lower.
Then, both axial and horizontal forces will be transferred to the
rest of the structure. It is necessary to understand the mechanism
of tubes under oblique impact loadings. To validate the theoretical
predictions and to consider inuences of load angle on SEA and P,
the optimal tubes were used for analysis under different loading
cases, that included axial and oblique dynamic loading. The load
angles are set to be 01, 151 and 301, respectively.
Based on Table 3, Eqs. (32)(43) were adopted to calculate the
mean crushing force, the mean horizontal force, and the mean
bending moment at load angle 151 for three optimal tubes. The
deformation results and crushing forcedisplacement curves of
three tubes at different load angles are shown in Figs. 1921. At
load angles 01 and 151, all three optimal tubes show progressive
wrinkle and no evidence of bending, as shown in Fig. 19(a) and (b),
Fig. 20(a) and (b) and Fig. 21(a) and (b). All three optimal tubes
were bent in the case of load angle 301, as shown in Figs. 19(c), 20
(c) and 21(c). Regarding the optimal tube type I with 6 cells, the
cross-sectional side-length and wall thickness are 80 mm and
1.4 mm, respectively. The mean crushing force obtained from FE
analysis is 46.815 kN. Obviously, the sum of side length and
internal web length of cross-section B is 603.831 mm. Substituting
items into Eqs. (32) and (35), the theoretical prediction of mean

crushing force at load angle 151 is

P I151 1:366  0:5  0:106  1:11:5  603:8310:5

6:878
2  0:75

47:902kN

45

For optimal tube type I, the difference between FE numerical

result and theoretical prediction is 2.32% obtained from Eq. (45).
The theoretical expression shows a good agreement with FE
numerical result. Then, the mean horizontal force and mean
bending moment for optimal tube type I under oblique impact
loading can be estimated from Eqs. (36), (37). Then we get
Pm
47:902  0:258 12:398kN
hI151

46




0:08  0:0011
1:305kN m
Mm
12:398 0:25 
I151
2  0:267

47

For the optimal tube type II, the cross-sectional side-length and
the wall thickness are 80 mm and 1.43 mm, respectively. Moreover,
the mean crushing force obtained from FE analysis is 58.815 kN.
As a matter of course, the sum of side length and internal web
length B is 635.47 mm. To substitute items into Eqs. (33), (38), the
theoretical prediction of mean crushing force for optimal tube type
II under oblique impact loading is
P II151 1:302  0:5  0:106  1:431:5  635:470:5
59:114kN

8:404
2  0:75
48

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

191

Fig. 20. Final deformation of optimal tube type II at: (a) 01, (b) 151, (c) 301 and (d) Crushing/mean crushing force-displacement curve.

The variance between FE numerical result and theoretical

expression is 0.509% for optimal tube type II. This variance reveals
the agreement between the proposed expression and the FE
simulation. Accordingly, the mean horizontal force and mean
bending moment for optimal tube type II under oblique impact
can be measured from Eqs. (39), (40). Then we get
Pm
59:114  0:258 15:3kN
hII151

49




0:08  0:0012
1:537kNm
15:3 0:25 
Mm
II151
2  0:267

50

The optimal tube type III has the side-length of 80 mm, and the
wall thickness of 1.48 mm. FE numerical result for the mean
crushing forces is 48.827 kN. Concurrently, for this optimal tube,
the sum of side length and internal web length B is of
608.269 mm. To substitute items into Eqs. (34) and (41), the
theoretical prediction of mean crushing force at load angle of 151 is
P III151 1:231 

0:5

608:2690:5

1:5

 0:106  1:48

7:43
48:946kN
2  0:75

51

The discrepancy between FE numerical results and theoretical

predictions for optimal tube type III is 1.36%. Consequently, this
discrepancy reveals the agreement between the proposed expression and the FE numerical results. The mean horizontal force and
mean bending moment of optimal tube type III under oblique

impact can be predicted from Eqs. (42), (43). Then we get

Pm
48:946  0:258 12:668kN
hIII151

52




0:08  0:0011
1:297kN m
Mm
12:668 0:25 
III151
2x0:267

53

The values of mean horizontal force and mean bending moment

for three optimal tubes under oblique loading are also obtained from
Eqs. (46), (47), (49), (50), (52), and (53). In addition, Fig. 22 (a) and
(b) shows an inuence of load angle on P and SEA. P and SEA
decrease slightly in the rang of 0151 load angle, then fall sharply
when the load angle exceeds 151. Consequently, the global bending
always dominates tubes under oblique impact at load angle of 301.

5. Conclusions
The SSFE theory was used to get the theoretical expressions of
forces and moment for thin-walled square multi-cell tubes under
oblique impact loading. The proles of tubes were divided into
several basic elements (the right-corner, 3-, T-shape, Criss-cross,
4-, 5- and 6-panel angle element). The oblique impacting coefcients () were proposed based on the geometric parameters, the
inertia effect and the oblique loading angle by considering the
effect of oblique loading and dynamic crushing for aluminum alloy
AA6060 T4. The theoretical expression of the mean horizontal
force and the mean bending moment was also created. Numerical
simulations were also implemented for tubes under dynamic

192

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

Fig. 21. Final deformation of optimal tube type III at: (a) 01, (b) 151, (c) 301 and (d) Crushing/mean crushing force-displacement curve.

Fig. 22. Effects of load angle on (a) P (b) SEA.

impact loadings. The theoretical predictions agreed well with the

FE numerical results. FE numerical results showed that the energy
absorption of tube types I and II were better than tube III.
Response surfaces of PCF and SEA were constructed for three
tubes. Pareto sets were obtained by the MOPSO method. A new
method was introduced to nd out a knee point from a Pareto

set with maximizing reex angle . The relative errors between RS

tting values and FE numerical results were acceptable. For the
optimal tube congurations, effects of load angle on P and SEA
were also considered under axial and oblique impact loading. For
the optimal congurations at the knee points, the theoretical
predictions agreed well with FE numerical results.

T. Tran et al. / International Journal of Mechanical Sciences 89 (2014) 177193

Acknowledgments
The nancial supports from National Natural Science Foundation
of China (Nos. 11232004, 11372106), New Century Excellent Talents
Program in University (NCET-12-0168) and Hunan Provincial Natural Science Foundation (12JJ7001) are gratefully acknowledged.
Moreover, Joint Center for Intelligent New Energy Vehicle is also
gratefully acknowledged.
References
[1] Wierzbicki T, Abramowicz W. On the crushing mechanics of thin-walled
structures. J Appl Mech 1983;50:72734.
[2] Abramowicz W, Jones N. Dynamic axial crushing of square tubes. Int J Impact
Eng 1984;2:179208.
[3] Abramowicz W, Jones N. Dynamic progressive buckling of circular and square
tubes. Int J Impact Eng 1986;4:24370.
[4] Guillow SR, Lu G, Grzebieta RH. Quasi-static axial compression of thin-walled
circular aluminium tubes. Int J Mech Sci 2001;43:210323.
[5] DiPaolo BP, Monteiro PJM, Gronsky R. Quasi-static axial crush response of a
thin-wall, stainless steel box component. Int J Solids Struct 2004;41:370733.
[6] Krolak M, Kowal-Michalska K, Mania R, Swiniarski J. Experimental tests of
stability and load carrying capacity of compressed thin-walled multi-cell
columns of triangular cross-section. Thin-Walled Struct 2007;45:8837.
[7] Chen W, Wierzbicki T. Relative merits of single-cell, multi-cell and foam-lled
thin-walled structures in energy absorption. Thin-Walled Struct 2001;39:
287306.
[8] Kim H-S. New extruded multi-cell aluminum prole for maximum crash
energy absorption and weight efciency. Thin-Walled Struct 2002;40:31127.
[9] Jensen , Langseth M, Hopperstad OS. Experimental investigations on the
behaviour of short to long square aluminium tubes subjected to axial loading.
Int J Impact Eng 2004;30:9731003.
[10] Karagiozova D, Jones N. Dynamic buckling of elasticplastic square tubes
under axial impactII: structural response. Int J Impact Eng 2004;30:16792.
[11] Zhang X, Cheng G, Zhang H. Theoretical prediction and numerical simulation
of multi-cell square thin-walled structures. Thin-Walled Struct 2006;44:
118591.
[12] Zhang X, Huh H. Crushing analysis of polygonal columns and angle elements.
Int J Impact Eng 2010;37:44151.
[13] Zhang X, Zhang H. Energy absorption of multi-cell stub columns under axial
compression. Thin-Walled Struct 2013;68:15663.
[14] Zhang X, Zhang H. Theoretical and numerical investigation on the crush
resistance of rhombic and kagome honeycombs. Compos Struct 2013;96:
14352.
[15] Zhang X, Zhang H. Energy absorption limit of plates in thin-walled structures
under compression. Int J Impact Eng 2013;57:8198.
[16] A. Naja, M. Rais-Rohani, Inuence of cross-sectional geometry on crush
characteristics of multi-cell prismatic columns, In: Proceedings of the 49th
AIAA/ASME/ASCE/AHS/ASC Structures, Structural Dynamics and Materials
Conference (2008).
[17] Karagiozova D, Alves M. Dynamic elastic-plastic buckling of structural elements: a review. Appl Mech Rev 2008;61 (040803-040803).
[18] Kim TH, Reid SR. Bending collapse of thin-walled rectangular section columns.
Comput Struct 2001;79:1897911.

193

[19] Tran T, Hou S, Han X, Tan W, Nguyen N. Theoretical prediction and

crashworthiness optimization of multi-cell triangular tubes. Thin-Walled
Struct 2014;82:18395.
[20] Tarlochan F, Samer F, Hamouda AMS, Ramesh S, Khalid K. Design of thin wall
structures for energy absorption applications: enhancement of crashworthiness due to axial and oblique impact forces. Thin-Walled Struct 2013;71:717.
[21] Han DC, Park SH. Collapse behavior of square thin-walled columns subjected
to oblique loads. Thin-Walled Struct 1999;35:16784.
[22] Reyes A, Langseth M, Hopperstad OS. Crashworthiness of aluminum extrusions subjected to oblique loading: experiments and numerical analyses. Int J
Mech Sci 2002;44:196584.
[23] Reyes A, Langseth M, Hopperstad OS. Square aluminum tubes subjected to
oblique loading. Int J Impact Eng 2003;28:1077106.
[24] Nagel GM, Thambiratnam DP. Dynamic simulation and energy absorption of
tapered thin-walled tubes under oblique impact loading. Int J Impact Eng
2006;32:1595620.
[25] Qi C, Yang S, Dong F. Crushing analysis and multiobjective crashworthiness
optimization of tapered square tubes under oblique impact loading. ThinWalled Struct 2012;59:10319.
[26] Yang S, Qi C. Multiobjective optimization for empty and foam-lled square
columns under oblique impact loading. Int J Impact Eng 2013;54:17791.
[27] Song J. Numerical simulation on windowed tubes subjected to oblique impact
loading and a new method for the design of obliquely loaded tubes. Int J
Impact Eng 2013;54:192205.
[28] Hou S, Li Q, Long S, Yang X, Li W. Design optimization of regular hexagonal
thin-walled columns with crashworthiness criteria. Finite Elem Anal Des
2007;43:55565.
[29] Hou S, Li Q, Long S, Yang X, Li W. Multiobjective optimization of multi-cell
sections for the crashworthiness design. Int J Impact Eng 2008;35:135567.
[30] Hou S, Han X, Sun G, Long S, Li W, Yang X, Li Q. Multiobjective optimization for
tapered circular tubes. Thin-Walled Struct 2011;49:85563.
[31] Tran T, Hou S, Han X, Chau M, Crushing Analysis and numerical optimization
of angle element structures under axial impact loading, Composite Structures,
http://dx.doi.org/10.1016/j.compstruct.2014.09.019, in press.
[32] Zhang X, Zhang H. Numerical and theoretical studies on energy absorption of
three-panel angle elements. Int J Impact Eng 2012;46:2340.
[33] Sun G, Li G, Zhou S, Li H, Hou S, Li Q. Crashworthiness design of vehicle by
using multiobjective robust optimization. Struct Multidiscip Optim 2011;44:
99110.
[34] Deb K, Gupta S. Understanding knee points in bicriteria problems and their
implications as preferred solution principles. Eng Optim 2011;43:1175204.
[35] Branke J, Deb K, Dierolf H, Osswald M. Finding knees in multi-objective
optimization. In: Yao X, Burke E, Lozano J, Smith J, Merelo-Guervs J, Bullinaria
J, Rowe J, Tio P, Kabn A, Schwefel H-P, editors. Parallel Problem Solving from
Nature - PPSN VIII. Berlin Heidelberg: Springer; 2004. p. 72231.
[36] Santosa SP, Wierzbicki T, Hanssen AG, Langseth M. Experimental and
numerical studies of foam-lled sections. Int J Impact Eng 2000;24:50934.
[37] Alghamdi AAA. Collapsible impact energy absorbers: an overview. ThinWalled Struct 2001;39:189213.
[38] S.S. Hsu, N. Jones, Dynamic axial crushing of aluminium alloy 6063 -T6 circular
tubes, 2004.
[39] Langseth M, Hopperstad OS. Static and dynamic axial crushing of square thinwalled aluminium extrusions. Int J Impact Eng 1996;18:94968.
[40] Hanssen AG, Langseth M, Hopperstad OS. Static and dynamic crushing of
circular aluminium extrusions with aluminium foam ller. Int J Impact Eng
2000;24:475507.
[41] Tam LL, Calladine CR. Inertia and strain-rate effects in a simple plate-structure
under impact loading. Int J Impact Eng 1991;11:34977.