摘要
越来越多的交通事故依赖于仿真软件进行再现分析,事故仿真再现结果的不确定性问题亦越来越受到重视。介绍了可用于研究该问题的上下界法、差分法以及响应曲面法;接着通过分析给出了各方法所需仿真次数;然后借助三个算例验证了各方法的实用性与适用范围,并总结出各方法的优缺点,发现响应曲面法适应性最强;最后探讨了求解事故再现中大不确定性问题的响应曲面法。所得结论可用于指导研究事故仿真再现结果的不确定性问题。
More and more accidents are simulated by accident reconstruction software, so the uncertainty problem of reconstruction resuhs by using accident simulation data draws more and more attention. The upper and lower bound method, finite difference method and response surface method, which can be used in evaluating uncertainty of reconstruction results were introduced. And then the simulation time consumption needed for each method was analyzed. The feasibility of these methods was proved through three numerical examples, and the strength and weakness of each method were summarized. It is concluded that the response surface method is of the most adaptability. Finally, a method for analyzing the large-scale uncertainty problem in accident reconstruction was proposed. Conclusions obtained can be used to guide research on evaluating the uncertainty of reconstruction results.
出处
《振动与冲击》
EI
CSCD
北大核心
2013年第6期176-180,184,共6页
Journal of Vibration and Shock
基金
国家自然科学基金(51208065)
湖南省自然科学基金(12JJ3044)
湖南省高等学校科学研究项目(12C0017)
同济大学道路与交通工程教育部重点实验室开放基金(201203)
公安部应用创新计划项目(2011YYCXGDST077)
关键词
事故再现
不确定性分析
上下界法
差分法
响应曲面法
仿真
accident reconstruction
uncertainty analysis
upper and lower bound method
finite difference method
response surface method
simulation
