作　　者： ; ; ; ; ;

机构地区： 道路灾变防治及交通安全教育部工程研究中心

出　　处： 《系统仿真学报》 2013年第5期911-914,920,共5页

摘　　要： 事故再现仿真中包含很多不确定参数,将这些参数所包含的不确定性引入到仿真结果中去,可使结果更为客观与可信。借助拉丁超立方试验设计(LHS)与蒙特卡洛方法,提出一种基于LHS的事故再现仿真结果不确定性分析方法。该方法用LHS生成试验样本集,据此安排试验并获试验结果,再通过回归分析获得与原事故再现仿真模型近似的响应面模型,最后通过结合所得响应面模型与蒙特卡洛方法确定事故再现结果的取值范围。将此方法应用到一具有显式表达式的算例中,结果显示本方法所得结果与蒙特卡洛方法及不确定度评定基本方法所得非常接近,最后用一个真实案例演示了方法的实用性。 There are many uncertain factors in accident reconstruction, so it is necessary to study the influence of these factors to improve the objectivity and confidence of simulation results. Based on Latin Hypercube Sampling （LHS） and Monte Carlo methods, a method for analyzing uncertainty of simulation results in accident reconstruction was proposed. In this method.＂ firstly, the sample set was generated by LHS; and then simulation experiments were conducted according to the test table; after that, the response surface model of the accident reconstruction model was obtained through regression analysis; finally, the range of accident reconstruction results couM be calculated by combining the response surface model and Monte Carlo method. Finally the method was applied to an accident case with explicit expression and the results obtained are very close to those with basic uncertainty analysis method and Monte Carlo method, and then the practicability of the method was also demonstrated by a true accident case.

关 键 词： 交通工程 事故再现 试验设计 拉丁超立方 不确定性 仿真

领　　域： [交通运输工程] [交通运输工程]