Professional Documents
Culture Documents
Thin-Walled Structures
journal homepage: www.elsevier.com/locate/tws
State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082, Hunan, PR China
College of Mechanical and Vehicle Engineering, Hunan University, Changsha 410082, Hunan, 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, Ha Noi, Vietnam
b
art ic l e i nf o
a b s t r a c t
Article history:
Received 2 January 2014
Received in revised form
13 March 2014
Accepted 26 March 2014
Available online 17 May 2014
The triangular tubes with multi-cell were rst studied on the aspects of theoretical prediction and
crashworthiness optimization design under the impact loading. The tubes' proles were divided into
2-, 3-, T-shapes, 4-, and 6-panel angle elements. The Simplied Super Folding Element theory was
utilized to estimate the energy dissipation of angle elements. Based on the estimation, theoretical
expressions of the mean crushing force were developed for three types of tubes under dynamic loading.
When taking the inertia effects into account, the dynamic enhancement coefcient was also considered.
In the process of multiobjective crashworthiness optimization, Deb and Gupta method was utilized to
nd out the knee points from the Pareto solutions space. Finally, the theoretical prediction showed an
excellent coincidence with the numerical optimal results, and also validated the efciency of the
crashworthiness optimization design method based on surrogate models.
Crown Copyright & 2014 Published by Elsevier Ltd. All rights reserved.
Keywords:
Crashworthiness
Multiobjective optimization
Triangular tube
Multi-cell
Energy absorption
Impact loading
1. Introduction
Thin-walled extrusions have been extensively applied in vehicle
crashworthiness components to absorb impact energy in the past
three decades. The tests and the theoretical expressions of square
and circular tubes under axial quasi-static and dynamic loading cases
were rst described by Wierzbicki and Abramowicz [1] and Abramowicz and Jones [2]. From then on, DiPaolo et al. [3,4], Guillow et al.
[5], Ullah [6] Zhang and Zhang [7], Alavi Nia and Parsapour [8] also
did many researches on these aspects. Beside square and circular
tubes, several other proles were also studied on their quasi-static or
dynamic responses, such as triangular tubes [912], hexagonal tubes
[13], etc. The structural collapse modes of triangular and square tubes
are different from those of circular tubes. Nevertheless, the crushing
curves of forcedisplacement of triangular and square tubes are
similar to those of circular tubes. The crushing curves of force
displacement of all the proles show that the crushing force rst
reaches an initial peak, then drops down and then uctuates around
a value of the mean crushing force. The extensional deformation has
n
Corresponding author 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.tws.2014.03.019
0263-8231/Crown Copyright & 2014 Published by Elsevier Ltd. All rights reserved.
more dominant effect on the crushing responses while the quasiinextensional mode occurs normally [14].
According to studies by Wierzbicki and Abramowicz [1], the
number of angle elements on cross-section of tube decided,
to a certain extent, the effectiveness of energy absorption. As
a matter of fact, it is necessary to design thin-walled multi-cell
tubes for weight-efcient energy absorption. Chen and Wierzbicki
[15] examined the axial crushing resistance of single-cell, doublecell and triple-cell hollow tubes, and the respective foam-lled
tubes under the quasi-static axial loading. The Simplied Super
Folding Element (SSFE) theory was applied to simplify SFE theory,
and three extensional triangular elements and three stationary
hinge lines were comprised instead of the kinematically admissible
model of SFE [1]. The average folding wavelength and the theoretical expression of the mean crushing force were deduced by
dividing the cross-sectional tube into distinct panel section and
angle element, assuming that the roles of each panel and of angle
element were at the same level. The work of Chen and Wierzbicki
[15] showed that the multi-cell tube could increase the specic
energy absorption SEA by approximately 15%, compared to the
respective hollow tube. Kim [16] used Chen and Wierzbicki's model
[15] to study multi-cell tubes with four square elements at the
corner. The SEA of new multi-cell tube was reported to increase by
190%, compared to conventional square tube. Zhang et al. [17] also
applied SSFE theory to derive a theoretical expression of the mean
184
Efb M 0 i b
i1
2. Theoretics
2.1. Theoretical prediction of multi-cell triangular tube
The SSFE theory was applied to solve the axial collapse of
triangular multi-cell thin-walled tubes. In the SSFE theory, the wall
thickness was assumed to be constant and the variation of
Fig. 1. Cross-sectional geometry of triangular multi-cell tube and typical angle element. (a) Tube type I (b) Tube type II and (c) Tube type III.
2 M 0 b
Since the role of structural each panel is similar and the multicell tube is constituted by m panels (as shown in Fig. 1), the energy
dissipation for bending of multi-cell tube is inferred as
2 M 0 mb 2 M 0 B
Etube
b
f
Easym
m_r c 2E m 4M 0
185
sym
Esym
m_r c 2E m_f 8M 0
H2
t
H2
cos
t
Fig. 3. Basic folding element: (a) asymmetric mode [15] and (b) symmetric (extensional) mode.
Fig. 4. (a) Relationship between the membrane energy of right corner and of 2-panel angle element and (b) 3-panel angle element.
186
H2
1 2 tan =2
t
10
H2
t
11
16M
13
0
m
t
cos
Fig. 5. (a) Collapse mode of T-shape element and (b) extensional elements.
Fig. 6. (a) 2-panel angle element and (b) 4-panel angle element.
187
Fig. 7. (a) Collapse mode of 4-panel angle element and (b) extensional elements.
Fig. 8. (a) Collapse mode of 6-panel angle element and (b) extensional elements.
Rigid wall
Lumped Mass
v = 10 m/s
L0 = 250 mm
P m I 2H
panel
Etube
3Eind:2
E4m panel
b
m
2 M 0 B 2M 0
H2
1
6 cos 12 1
:
t
cos
14
Substituting Eq. (16) back into Eq. (15), the mean crushing force in
quasi-static loading of tube type I is obtained as
p
6 cos 12 1
F ;
H t
H t
M0
cos
Pm I
15
B F ;
H
)H
s
Bt
F ;
17
where
The half-wavelength was obtained under the stationary condition of the mean crushing force P m =H 0, then
0
F ;
M0 B M0 H
16
F ; 6 cos 12 1
1
:
cos
188
H2
8
32 12 tan =2
t
cos
18
32 12 tan =2
G;
H t
H t
M0
cos
19
Substituting Eq. (20) back into Eq. (19), the equation of the mean
crushing force for tube type II under the quasi-static loading is
obtained as
p
M0 B M0 H
G;
P m II
8
:
cos
The structure of the tube type III (Fig. 1c) was formed by three
independent 2-panel angle elements, six 4-panel angle elements,
189
and one 6-panel angle element. Substituting the terms of Eqs. (4),
(9), (12) and (13) into Eq. (1), the general theoretical solution to
the mean crushing force of tube type III is obtained as
P m III 2H Etube
3E2m panel 6E4m panel E6m panel
b
H2
32
32 6 cos
2 M 0 B 2M 0
t
cos
22
32 6 cos
Q ;
H t
H t
M0
cos
23
32
:
cos
EA
m
26
Fig. 11. The crushing foredisplacement 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
1
1.3
1.6
1.9
2.2
1
1.3
1.6
1.9
2.2
1
1.3
1.6
1.9
2.2
1
1.3
1.6
1.9
2.2
1
1.3
1.6
1.9
2.2
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 II
SEA
PCF (kN) SEA
(kJ/kg)
(kJ/kg)
PCF (kN)
18.262 39.362
20.226 52.274
22.774
65.287
23.854 77.82
24.732 88.936
17.635 42.151
19.491 55.941
21.079 69.78
22.231 82.904
23.891 94.376
16.689 48.183
18.63
63.406
19.883 80.385
21.425 97.058
22.916 111.899
16.092 50.984
17.793 67.207
19.279 85.237
20.865 102.754
22.276 117.81
15.569 50.155
17.083 66.443
19.26
83.029
20.628 98.954
21.053 113.035
49.572
65.965
82.467
98.125
111.611
53.133
70.384
88.192
104.633
118.899
60.979
69.095
99.84
120.552
138.568
65.978
83.194
105.95
128.907
148.295
63.148
83.857
104.85
124.628
142.093
52.592
69.599
85.801
101.135
114.541
56.038
74.202
92.462
109.979
125.572
64.544
85.478
107.641
128.512
146.48
68.196
90.665
114.112
135.856
153.825
66.649
87.852
110.016
130.951
157.552
17.356
19.876
21.86
22.003
23.107
16.842
19.272
20.924
21.602
22.693
16.559
18.9
20.273
20.679
21.524
15.253
16.954
18.229
19.146
19.71
14.398
16.139
17.748
18.55
18.715
23.339
25.227
28.199
29.092
29.644
22.753
24.273
26.388
27.564
28.107
21.338
24.05
25.536
26.272
26.42
20.932
22.371
24.579
25.306
25.948
19.337
20.011
22.979
23.92
24.152
190
Fig. 12. The response surface of (a) peak crushing force and (b) SEA.
Fig. 13. (a) SEA vs structural weight and (b) Pm vs structural weight.
estimated as
Z d
Pxdx
EA
0
27
Fig. 14. Pareto spaces for multi-objective optimization: (a) tube type I, (b) tube
type II and (c) tube type III.
~ x
yx yx
i i
i1
29
30
191
mathematically given as
Maximize x; xL ; xR L R
33
31
Table 2
Difference of numeric result and theoretical prediction for three tubes.
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
Num. Pm (kN)
Theo. Pm (kN)
Diff. (%)
Num. Pm (kN)
Theo. Pm (kN)
Diff. (%)
Num. Pm (kN)
Theo. Pm (kN)
Diff. (%)
22.23
34.107
46.497
61.379
77.295
23.173
34.142
47.275
62.465
79.196
23.787
35.095
48.408
64.341
81.044
24.151
35.645
49.676
66.14
82.19
24.819
36.812
51.437
67.423
83.731
23.016
34.028
46.344
59.817
74.337
23.736
35.098
47.809
61.717
76.710
24.435
36.137
49.230
63.561
79.013
25.115
37.146
50.611
65.352
81.250
25.776
38.129
51.956
67.096
83.427
3.54
0.23
0.33
2.54
3.83
2.43
2.80
1.13
1.20
3.14
2.73
2.97
1.70
1.21
2.51
3.99
4.21
1.88
1.19
1.14
3.86
3.58
1.01
0.49
0.36
25.894
41.1
55.331
72.1
91.64
26.75
42.39
58.3
74.137
94.125
27.482
43.184
59.954
76.129
95.039
28.264
43.859
60.278
77.219
95.897
29.109
44.984
60.994
79.107
96.597
27.148
40.143
54.680
70.589
87.738
27.996
41.404
56.406
72.827
90.533
28.820
42.627
58.080
74.998
93.244
29.620
43.816
59.707
77.108
95.879
30.400
44.974
61.291
79.162
98.443
4.84
2.33
1.18
2.10
4.26
4.66
2.33
3.25
1.77
3.82
4.87
1.29
3.13
1.49
1.89
4.80
0.10
0.95
0.14
0.02
4.44
0.02
0.49
0.07
1.91
39.973
60.35
85.345
111.642
139.519
41.129
61.05
87.9
113.899
144.17
42.716
62.715
89.541
117.161
148.092
43.862
63.768
91.482
120.81
151.03
45.268
65.507
93.192
122.495
153.514
41.221
60.954
83.028
107.183
133.223
42.510
62.868
85.648
110.581
137.467
43.760
64.726
88.190
113.878
141.583
44.976
66.531
90.660
117.082
145.584
46.160
68.290
93.066
120.201
149.477
3.12
1.00
2.72
3.99
4.51
3.36
2.98
2.56
2.91
4.65
2.44
3.21
1.51
2.80
4.39
2.54
4.33
0.90
3.09
3.61
1.97
4.25
0.14
1.87
2.63
192
were different in weight. Tube I is the lightest one while tube III is
the heaviest. The axial crushing of multi-cell tubes was presented
with a displacement equal to about 70% of the initial length. Fig. 10
shows the deformation process of three tubes at different times.
Sometimes the exact value of the effective crushing distance on
the crushing forcedisplacement curve was not unique. The
corresponding crushing forcedisplacement curves of three tubes
are also shown in Fig. 11. After reaching the initial peak and before
rising steeply whenever the deformation capacity is exhausted at
the effective crushing distance, the crushing force fell sharply and
then uctuated periodically and around the values of the mean
crushing force in correspondence with the formation, and nally
completed the collapse of folds one by one.
p
F ;
2
where
F ; 6 cos 12 1
1
cos
p
G;
2
where
Fig. 15. Comparison between numerical prediction and theoretical prediction:
(a) tube type I, (b) tube type II and (c) tube type III.
G; 32 12 tan =2
34
8
:
cos
35
p
Q ;
0:5
1:5 0:5
s
t
B
P dym:
III
m
III
III
0
m III
2
36
s0
where
Q ; 32 6 cos
32
:
cos
Table 3
Optimal results by using the method of Deb and Gupta (Knee point).
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 1.23, a 80
19.897
19.728
0.856
25.282
25.717
1.691
25.100
24.800
1.210
49.315
49.088
0.462
63.436
63.056
0.603
63.967
64.267
0.467
Type II
Type III
t 1.25, a 80
t 1.21, a 80
193
sy su
1n
0:106 GPa
37
Fig. 16. (a) Deformation result and (b) crushing forcedisplacement curve of tube I.
Fig. 17. (a) Deformation result and (b) crushing forcedisplacement curve of tube II.
194
Fig. 18. (a) Deformation result and (b) crushing forcedisplacement curve of tube III.
352:620:5
6:46
31:335 kN
2 0:7
38
With 6 cells, the optimal tube type II in Table 3 has the sidelength of 80 mm and the wall thickness of 1.25 mm (Fig. 17). The
mean crushing force received from FE analysis was 37.01 kN.
Simultaneously, for this optimal tube, the sum of side-length and
internal web length B is 439.06 mm. To substitute items into
Eq. (35), the theoretical prediction of mean crushing force is
0:5
P dym:
0:106 1:421:5
m II 1:3
439:060:5
7:41
37:864 kN
2 0:7
39
As listed in Table 3, optimal tube type III has 9 cells (Fig. 18).
The width and the wall thickness of this cross-section are
respectively of 80 mm and 1.21 mm. Then, the mean crushing
force in FE analysis was 53.67 kN. As a matter of course, the
parameter of prole of tube III is B 439.376 mm. Replacing items
into Eq. (36), the theoretical prediction of mean crushing force is
0:5
P dym:
0:106 1:361:5
m III 1:45
439:3760:5
10:089
54:903 kN
2 0:7
40
From the results above, Eq. (36) were adopted to calculate the
mean crushing force for three optimal tubes. Subsequently, the
differences between numerical predictions and theoretical solutions for optimal tube types I, II and III were respectively of 2.97%,
2.3% and 2.05%. These differences show that the proposed equations
are appropriate to the numerical predictions. In addition, the stable
and progressive folding deformation patterns that are developed in
all the three types of tube are the desirable energy-dissipating
mechanism.
5. Conclusions
The proles of three types of tubes were divided into the basic
elements: 2-, 3-, T-shape, 4- and 6-panel angle element. Based on
the Simplied Super Folding Element theory, theoretical expressions of the mean crushing force were proposed for the three
types of triangular multi-cell thin-walled tubes under the axial
crushing loading. Numerical simulations of tubes under the axial
dynamic impact loading were also carried out, and a dynamic
enhancement coefcient was introduced to account for the inertia
effects of aluminum alloy AA6060 T4. Numerical results showed
that tube types I and III were better than tube type II in the aspect
of energy absorption. Simultaneously, the stable and progressive
folding deformation patterns appeared for all the three types
of tubes.
The two RS models of PCF and SEA for each tube were constructed. Pareto sets were obtained by using the linear weighted
average methods (LWAM). In this paper, the Pareto solutions of three
types of tubes were identied to seek out the knee points. The
relative errors between RS approximate value and FE analysis value
at the Knee points were obtained and those were also acceptable.
Finally, the theoretical expressions excellently agreed with the
numerical results, and simultaneously validated the efciency of
the crashworthiness optimization design method based on the
surrogate models and the numerical analysis techniques.
Acknowledgments
The nancial supports from National Natural Science Foundation of China (Nos. 11232004 and 51175160), 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] 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.
[4] DiPaolo BP, Tom JG. Effects of ambient temperature on a quasi-static axialcrush conguration response of thin-wall, steel box components. Thin-Walled
Struct 2009;47:98497.
195