汽车工程 ›› 2021, Vol. 43 ›› Issue (12): 1787-1792.doi: 10.19562/j.chinasae.qcgc.2021.12.007

• • 上一篇    下一篇

基于遗传算法的高精度事故重建与损伤分析

王成铭1,2,范颖3,邹冬华2,3,王金明2,李正东2,陈忆九2,仲梁维1()   

  1. 1.上海理工大学机械工程学院,上海 200093
    2.上海市法医学重点实验室司法鉴定科学研究院,上海 200063
    3.贵州医科大学法医学院,贵阳 550025
  • 收稿日期:2021-08-11 出版日期:2021-12-25 发布日期:2021-12-24
  • 通讯作者: 仲梁维 E-mail:zlvcad@126.com
  • 基金资助:
    国家自然科学基金(82171872);上海市自然科学基金面上项目(21ZR1464600);司法部司法鉴定重点实验室资助项目、上海市法医学重点实验室资助项目(17DZ2273200);上海市司法鉴定专业技术服务平台(19DZ2292700);中央级科研院所基本科研业务费(GY2020G-4)

High-accuracy Accident Reconstruction and Injury Analysis Based on Genetic Algorithm

Chengming Wang1,2,Ying Fan3,Donghua Zou2,3,Jinming Wang2,Zhengdong Li2,Yijiu Chen2,Liangwei Zhong1()   

  1. 1.School of Mechanical Engineering,University of Shanghai for Science and Technology,Shanghai 200093
    2.Shanghai Key Laboratory of Forensic Medicine,Academy of Forensic Science,Ministry of Justice,Shanghai 200063
    3.Guizhou Medical University,Guiyang 55002
  • Received:2021-08-11 Online:2021-12-25 Published:2021-12-24
  • Contact: Liangwei Zhong E-mail:zlvcad@126.com

摘要:

为提高交通事故的重建精度,对行人的颅脑损伤风险进行分析。基于3D激光扫描技术建立事故车辆多刚体模型,采用多目标遗传算法求取最优近似解,运用优化软件ISIGHT和多刚体动力学软件MADYMO建立耦合模型,精确重建人-车碰撞运动学响应,最后采用THUMS有限元假人模型对颅脑损伤风险进行预测分析。结果表明:事故车辆逆向重建效果良好,优化后的人车运动学行为与监控视频一致,假人的头部损伤部位和损伤情况与伤情报告完全吻合。

关键词: 事故重建, 3D激光扫描技术, 多刚体动力学, 遗传算法, 有限元, 颅脑损伤

Abstract:

In order to enhance the reconstruction accuracy of traffic accidents, the risk of pedestrian craniocerebral injury is analyzed,the multi-rigid-body model for accident vehicle is built based on 3D laser scanning technology, and the multi-objective genetic algorithm is used to obtain the optimal approximate solution. A coupled model is set up with optimization software ISIGHT and multi-rigid-body dynamics software MADYMO and the kinematic response of the human-vehicle collision is accurately reconstructed. Finally, finite element dummy model THUMS is adopted to predict the risk of craniocerebral injury. The results show that the reverse reconstruction of accident vehicle has a good result, the optimized human-vehicle kinematic behaviors are consistent with that in surveillance video, and the injury position and injury situation of dummy's head completely agree with the injury report.

Key words: accident reconstruction, 3D laser scanning technology, rigid multibody dynamics, genetic algorithm, finite element, craniocerebral injury