Support vector regression based method for the inverse problem of fuzzy parameters in the simulation model of thin-walled energy-absorbing tubes installed in the locomotive front end

PDF(1491 KB)

Journal of Vibration and Shock ›› 2024, Vol. 43 ›› Issue (18) : 28-35.

XU Ping1,2，HUANG Qi1,2，XING Jie1,2，HE Jiaxing1,2，XU Kai1,2，XU Tuo1,2

Author information +

History +

Abstract

In order to obtain accurate model parameters influencing the finite element calculation accuracy of crashworthiness structures and improve the precision of impact simulation, a method based on Support Vector Regression (SVR) for parameter inverse problem solving is proposed. A finite element model is established for the preloaded thin-walled energy-absorbing circular tube in the locomotive front anticlimbing structure. The accuracy of the model is validated through tests. A Latin hypercube experimental design is employed to drive the finite element model with a small number of calculations to obtain a dataset. Fuzzy parameters in the finite element model are used as input variables, the difference between the calculated and experimental loads is taken as the target response. The SVR method is applied to construct a mapping relationship, and the Strengthen elitist GA(SEGA) is utilized to optimize hyperparameters. The optimal SVR model is then used again with SEGA optimization for inverse problem solving, obtaining the best combination of fuzzy parameters and using them to configure the finite element model, the simulation results show an improvement in the matching degree of crashworthiness indicators and load curves compared to the initial calculations. The study provides a new approach for accurately setting fuzzy control parameters in finite element models and improving the accuracy of collision simulations.

Key words

Crashworthiness / Thin-walled circular tube / Finite element model / Fuzzy parameter inversion / Support Vector Regression / Genetic Algorithm.

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

XU Ping1, 2, HUANG Qi1, 2, XING Jie1, 2, HE Jiaxing1, 2, XU Kai1, 2, XU Tuo1, 2. Support vector regression based method for the inverse problem of fuzzy parameters in the simulation model of thin-walled energy-absorbing tubes installed in the locomotive front end[J]. Journal of Vibration and Shock, 2024, 43(18): 28-35

References

[1] GAO Guangjun, ZHUO Tianyu, GUAN Weiyuan. Recent research development of energy-absorption structure and application for railway vehicles[J]. Journal of Central South University, 2020, 27(4): 1012-1038.
[2] BAROUTAJI A, SAJJIA M, OLABI A G. On the crashworthiness performance of thin-walled energy absorbers: Recent advances and future developments[J]. Thin-Walled Structures, 2017, 118: 137-163.
[3] 姚如洋, 赵振宇, 尹冠生, 等. 薄壁开孔圆管在轴向荷载作用下的理论研究[J]. 振动与冲击, 2020, 39(2): 141-147.
YAO Ruyang,ZHAO Zhenyu,YIN Guansheng, et al. Theoretical analysis on thin-walled holed circular tubes under axial loading[J]. Journal of vibration and shock, 2020, 39(2): 141-147.
[4] JIN X, LU J, GUAN W. Crashworthiness analysis and multiobjective robust optimization of two-stage variable thickness expansion tube under impact loading[J]. Structural and Multidisciplinary Optimization, 2022, 65.
[5] WEI L, ZHANG L, TONG X, et al. Crashworthiness study of a subway vehicle collision accident based on finite-element methods[J]. International Journal of Crashworthiness, 2021, 26: 159-170.
[6] CLAUSEN A H, TRYLAND T, REMSETH S. An investigation of material properties and geometrical dimensions of aluminium extrusions[J]. Materials & Design, 2001, 22(4): 267-275.
[7] 王琼, 刘桂萍. 汽车车身耐撞性设计参数的区间稳健性优化[J]. 振动与冲击, 2020, 39(5): 37-44.
WANG Qiong, LIU Guiping. Robust optimization design mode for vehicle body crashworthiness based on parametric interval analysis[J]. Journal of vibration and shock, 2020, 39(5): 37-44.
[8] 彭云枫. 面向非线性材料的不确定性参数识别混合数值方法研究[D]. 湖南大学, 2022.
PENG Yunfeng. Research on hybrid numerical method for uncertain parameter identification of nonlinear materials[D]. Hunan University, 2022.
[9] XU X, ZHU Y, TIAN K, et al. Study on an Integral Algorithm of Load Identification Based on Displacement Response[J]. Sensors (Basel, Switzerland), 2021, 21(19): 6403.
[10] 秦兴元, 朱平, 刘钊, 等. 焊核区域材料塑性参数的反演识别[J]. 机械设计与研究, 2017, 33(4): 72-75.
QIN Xingyuan1, ZHU Ping, LIU Zhao, et al. An Inverse FE Modeling Methodology to Extract Plastic Properties of Nugget Materials[J]. Machine Design and Research, 2017, 33(4): 72-75.
[11] 谭飞, 韩旭. 基于代理模型的功能梯度梁的材料特性参数反求[J]. 复合材料学报, 2008(5): 175-180.
TAN Fei, HAN Xu. Surrogate-based inverse technique for material characterization of functionally graded beam[J]. Acta Materiae Compositae Sinica, 2008(5): 175-180.
[12] 乔良, 宋小欣, 谢延敏, 等. 基于PSO-RBF代理模型的板料成形本构参数反求优化研究[J]. 中国机械工程, 2014, 25(19): 2680-2685.
QIAO Liang, SONG Xiaoxin, XIE Yanmin, et al. Reverse and Optimization of Sheet Forming Parameters Based on PSO-RBF Surrogate Model[J]. China Mechanical Engineering, 2014, 25(19): 2680-2685.
[13] 刘方园, 王水花, 张煜东. 支持向量机模型与应用综述[J]. 计算机系统应用, 2018, 27(4): 1-9.
LIU Fangyuan, WANG Shuihua, ZHANG Yudong. Overview on Models and Applications of Support Vector Machine[J]. Computer Systems & Applications, 2018, 27(4): 1-9.
[14] 李恩颖, 王琥, 李光耀. 基于支持向量机回归的材料参数反求方法[J]. 机械工程学报, 2012, 48(6): 90-95.
LI Enying, WANG Hu, LI Guangyao. Material Parameter Inverse Technique Based on Support Vector Regression[J]. Journal of Mechanical Engineering, 2012, 48(6): 90-95.
[15] 唐维, 谢延敏, 黄仁勇, 等. 基于自适应SVR-ELM混合近似模型的镁合金差温成形本构参数反求[J]. 工程设计学报, 2017, 24(5): 536-544.
TANG Wei, XIE Yanmin, HUANG Renyong, et al. Shape optimization of blank holder structure in nonisothermal stamping of magnesium alloy based on ensemble surrogate[J]. Journal of Plasticity Engineering, 2018, 25(2): 22-29.
[16] KONG S, CUI H, WU G, et al. Full-scale identification of ice load on ship hull by least square support vector machine method[J]. Applied Ocean Research, 2021, 106: 102439.
[17] LIU Y, WANG L, GU K. A support vector regression (SVR)-based method for dynamic load identification using heterogeneous responses under interval uncertainties[J]. Applied Soft Computing, 2021, 110: 107599.
[18] YANG H, JIANG J, CHEN G, et al. Dynamic load identification based on deep convolution neural network[J]. Mechanical Systems and Signal Processing, 2023, 185: 109757.