基于SSI-PSO的汽车碰撞试验时序数据处理与分类方法

doi:10.3969/j.issn.1674-8484.2022.02.005

汽车安全与节能学报 ›› 2022, Vol. 13 ›› Issue (2): 259-268.DOI: 10.3969/j.issn.1674-8484.2022.02.005

基于SSI-PSO的汽车碰撞试验时序数据处理与分类方法

李晗¹(), 刘钊², 朱平¹()

1.上海交通大学机械与动力工程学院,上海 200240,中国
2.上海交通大学设计学院,上海 200240,中国

收稿日期:2021-12-23 修回日期:2022-03-06 出版日期:2022-06-30 发布日期:2022-07-01
通讯作者: 朱平
作者简介:*朱平,教授。E-mail： pzhu@sjtu.edu.cn。
李晗（1992—）,男（汉）,山东,博士研究生。E-mail： sjtulihan@sjtu.edu.cn。
基金资助:
国家自然科学基金(U1864211);国家自然科学基金(11772191);上海市自然科学基金(21ZR1431500)

Automobile crash test time-series data processing and classification method based on SSI-PSO algorithm

LI Han¹(), LIU Zhao², ZHU Ping¹()

1. School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China
2. School of Design, Shanghai Jiao Tong University, Shanghai 200240, China

Received:2021-12-23 Revised:2022-03-06 Online:2022-06-30 Published:2022-07-01
Contact: ZHU Ping

摘要/Abstract

摘要：

为实现汽车碰撞试验假人响应曲线数据集的类别辨识,研究了面向智能优化算法的问题转换与构造方法。针对假人曲线数据的特征处理与分类过程,提出了一种基于社会蜘蛛粒子群优化算法（SSI-PSO）的碰撞试验多变量时序数据特征选择与分类方法;利用汽车碰撞试验采集的假人曲线数据,测试和验证了该方法。结果表明：本文方法可获得面向假人曲线数据分类的最佳特征组合方式与较小规模的神经网络结构;该方法的假人曲线分类模型性能提升17.5%、分类精度达到96.5%。因而,实现了对碰撞试验假人响应曲线标注信息的有效分类。

关键词: 汽车碰撞, 安全数据集, 多变量时序数据, 社会蜘蛛粒子群优化（SSI-PSO）算法, 特征工程, 监督学习, 启发式优化算法

Abstract:

This paper investigated an optimization problem transformation and construction method for heuristic optimization algorithm to realize the category identification of dummy curve dataset from automobile crash test. A method of feature selection and classification was proposed for multi-variable time-series data in crash test based on a social spider inspired particle swarm optimization (SSI-PSO) for the feature processing and for the classification process of dummy curve data. The proposed method was tested and validated by using the dummy curve data collected from automobile crash test. The result shows that the optimal feature combination and the small-scale neural network for dummy curve classification are obtained by the proposed method. The performance of dummy curve classification model improves by 17.5% and classification accuracy reaches 96.5% based on the proposed method. Therefore, the labeling information of dummy response curve from crash test is classified effectively.

Key words: automobile crash, safety data, multi-variable time series data, social spider inspired particle swarm optimization (SSI-PSO) algorithm, feature engineering, supervised learning, heuristic optimization algorithm

中图分类号:

李晗, 刘钊, 朱平. 基于SSI-PSO的汽车碰撞试验时序数据处理与分类方法[J]. 汽车安全与节能学报, 2022, 13(2): 259-268.

LI Han, LIU Zhao, ZHU Ping. Automobile crash test time-series data processing and classification method based on SSI-PSO algorithm[J]. Journal of Automotive Safety and Energy, 2022, 13(2): 259-268.

图/表 11

图1 汽车安全试验假人头部碰撞响应曲线数据（部分）

曲线名称	曲线编号	计数
Driver Head Acceleration X	11HEAD0000H3ACXP	236
Driver Head Acceleration X	13HEAD0000H3ACXP	20
Driver Head Acceleration Y	11HEAD0000H3ACYP	235
Driver Head Acceleration Y	11HEAD0000H3ACZP	29
Driver Chest Displacement X	11CHST0000H3DSXP	235
Driver Chest Displacement X	13CHST0000H3DSXP	20

表1 汽车安全试验假人碰撞响应曲线数据统计信息（部分）

图2 SSI-PSO算法流程

图3 SSI-PSO算法与其他9种优化算法对比

图4 基于混合交叉验证分类精度构造目标函数

图5 本文方法流程图

图6 不同采样点的基础分类器混合交叉验证精度

测试序号	P_acc	R_sam / %
1	0.914 ± 0.007	10.0 ± 0.0
2	0.913 ± 0.005	10.1 ± 1.2
3	0.928 ± 0.007	10.9 ± 1.1

表2 特征优化问题的计算结果

图7 不同分类器的分类精度

图8 优化过程的历史全局最优解

图9 基于最优采样和加权特征处理的模型分类精度

参考文献 24

[1]	Alam S, Dobbie G, Rehman S. Analysis of particle swarm optimization based hierarchical data clustering approaches[J]. Swarm and Evolutionary Computation, 2015, 25: 36-51. doi: 10.1016/j.swevo.2015.10.003 URL
[2]	LU Huijuan, DU Bangjun, LIU Jinyong, et al. A kernel extreme learning machine algorithm based on improved particle swam optimization[J]. Memetic Computing, 2017, 9: 121-128. doi: 10.1007/s12293-016-0182-5 URL
[3]	赵东. 基于群智能优化的机器学习方法研究及应用[D]. 长春: 吉林大学, 2017.
	ZHAO Dong. Research and application of machine learning method based on swarm intelligence optimization[D]. Changchun: Jilin University, 2017. (in Chinese)
[4]	周志华, 俞扬, 钱超. 演化学习:理论与算法进展[M]. 北京: 人民邮电出版社, 2021: 1-6.
	ZHOU Zhihua, YU Yang, QIAN Chao. Evolutionary Learning: Advances in Theories and Algorithms[M]. Beijing: Posts Telecom Press, 2021: 1-6. (in Chinese)
[5]	ZHANG Yudong, WANG Shuihua, JI Genlin. A comprehensive survey on particle swarm optimization algorithm and its applications[J]. Math Probl Engineering, 2015, ID 931256: 1-38.
[6]	中国汽车技术研究中心有限公司. C-NCAP 管理规则(2018年版)[EB/OL]. [2021-11-09]. Http://www.c-ncap.org.cn/cms/picture/319947309459771392.pdf.
	China Automotive Technology Research Center Co., Ltd. C-NCAP Management Rules (2018 Edition) [EB/OL]. [2021-11-09]. Http://www.c-ncap.org.cn/cms/picture/319947309459771392.pdf. (in Chinese)
[7]	Formato R A. Central force optimization: a new metaheuristic with applications in applied electromagnetics[J]. Prog Electromagnetics Res, 2007, 77: 425-491. doi: 10.2528/PIER07082403 URL
[8]	Rashedi E, Nezamabadi-Pour H, Saryazdi S. GSA: A gravitational search algorithm[J]. Information Sciences, 2009, 179(13): 2232-2248. doi: 10.1016/j.ins.2009.03.004 URL
[9]	Kennedy J, Eberhart R. Particle swarm optimization[C]// IEEE Int’l Conf Neural Networks Proc, 1995, 48: 1942-1948.
[10]	Bahriye B, Dervis K. An artificial bee colony (ABC) algorithm for numeric function optimization[J]. IEEE Swarm Intell Symp, 2006, 8(1): 687-697.
[11]	Mirjalili S. Dragonfly algorithm: a new meta-heuristic optimization technique for solving single-objective, discrete, and multi-objective problems[J]. Neural Compu Appl, 2016, 27(4): 1053-1073. doi: 10.1007/s00521-015-1920-1 URL
[12]	Goldberg D E. Genetic Algorithms in Search, Optimization and Machine Learning[M]. Boston, US: Addison-Wesley Pub. Co., 1989: 1-26.
[13]	Storn R, Price K. Differential evolution: A simple and efficient heuristic for global optimization over continuous spaces[J]. J Global Opti, 1997, 11: 341-359.
[14]	GEEM Zong-Woo, KIM Joong-Hoon, Loganathan G V. A new heuristic optimization algorithm: Harmony search[J]. Simulation, 2001, 2(2): 60-68.
[15]	LIU Zhao, LI Han, ZHU Ping. Diversity-enhanced particle swarm optimization algorithm based on the group behaviour of social spiders [J]. Engineering Opti, 2021, 53(5): 811-829.
[16]	Salomon M, Sponarski C C, Larocque A, et al. Social organization of the colonial spider Leucauge sp. in the Neotropics: Vertical stratification within colonies[J]. J Arach, 2010, 38(3): 446-451.
[17]	Yip E C, Powers K S, Aviles L. Cooperative capture of large prey solves scaling challenge faced by spider societies[J]. Proc National Acad Sci USA, 2008, 105(33): 11818-11822. doi: 10.1073/pnas.0710603105 URL
[18]	Cuevas E, Cienfuegos M. A new algorithm inspired in the behavior of the social-spider for constrained optimization[J]. Expert Syst Appl, 2014, 41(2): 412-425. doi: 10.1016/j.eswa.2013.07.067 URL
[19]	XUE Bing, ZHANG Mengjie, Browne W N, et al. A survey on evolutionary computation approaches to feature selection[J]. IEEE Trans Evol Compu, 2016, 20(4): 606-626.
[20]	王翔, 胡学钢. 高维小样本分类问题中特征选择研究综述[J]. 计算机应用, 2017, 37(9): 2433-2438, 2448.
	WANG Xiang, HU Xuegang. Overview on feature selection in high-dimensional and small-sample-size classification[J]. J Compu Appl, 2017, 37(9): 2433-2438, 2448. (in Chinese)
[21]	Bishop C M. Pattern Recognition and Machine Learning: Information Science and Statistics[M]. New York, US: Springer-Verlag, Inc, 2006: 179-224.
[22]	Pedregosa F, Varoquaux G, Gramfort A, et al. Scikit-learn: machine learning in Python[J]. J Mach Learn Res, 2012, 12(10): 2825-2830.
[23]	Bishop C M. Neural networks for pattern recognition[M]. Oxford, UK: Oxford University Press. 1995: 116-163.
[24]	Esther-Lydia S, Rafael P, Manuel L, et al. Missing value imputation on missing completely at random data using multilayer perceptrons[J]. Neural Networks, 2011, 24(1): 121-129. doi: 10.1016/j.neunet.2010.09.008 URL

基于SSI-PSO的汽车碰撞试验时序数据处理与分类方法

Automobile crash test time-series data processing and classification method based on SSI-PSO algorithm

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 11

参考文献 24

相关文章 11

编辑推荐

Metrics

本文评价

期刊信息

在线期刊

作者中心

审稿中心

联系我们

[1]	屈贤, 林繁国, 张金龙. 基于磁流变缓冲吸能装置的汽车碰撞仿真分析[J]. 汽车安全与节能学报, 2020, 11(4): 487-492.
[2]	陈现岭, 刘智超, 蒋斌庆, 刘珍海. 整车碰撞虚拟仿真中橡胶悬置的仿真方法 [J]. 汽车安全与节能学报, 2020, 11(2): 212-219.
[3]	许万，周大永. 三元锂离子动力电池挤压损伤容限的试验研究[J]. JASE, 2019, 10(1): 106-111.
[4]	解文娜，颜凌波，张伟锋，等. 汽车碰撞试验假人零部件质心测量方法及装置[J]. JASE, 2018, 9(3): 281-287.
[5]	王心怡，刘晓颖，陈聪，唐亮，王东林. 跌落工况下的人体腰椎损伤风险的几种影响因素[J]. JASE, 2018, 9(2): 178-185.
[6]	宋正超，章斯亮. 基于改进EGO 算法的汽车40% 偏置碰撞优化设计[J]. JASE, 2017, 08(03): 246-251.
[7]	刘洋, 赖兴华, 马春生. 汽车生产年份和乘员乘坐位置对侧面碰撞下乘员伤害的影响[J]. 汽车安全与节能学报, 2012, 3(4): 313-318.
[8]	邹猛, 魏灿刚, 许述财. 汽车碰撞中锥形晶胞薄壁金属管吸能特性的仿真研究[J]. 汽车安全与节能学报, 2012, 3(4): 326-331.
[9]	曹立波，肖慧青，江岸. 保护汽车碰撞中孕妇乘员安全的包裹式安全带[J]. 汽车安全与节能学报, 2012, 3(1): 12-18.
[10]	蒋小晴, 杨济匡, 肖志, 叶映台, 郭杰. 前纵梁碰撞模拟中的焊点建模方法比较[J]. 汽车安全与节能学报, 2010, 1(4): 327-331.
[11]	王方, 肖志, 万鑫铭, 杨济匡. 汽车低速碰撞中的人体颈部有限元模型验证[J]. 汽车安全与节能学报, 2010, 1(3): 214-218.