微观交通仿真模型参数标定方法改善研究  

Research on Improvement of Parameters Calibration Method of Microscopic Traffic Simulation Model

作　　者：周晨静[1] 高亚聪 荣建[2] Zhou Chenjing;Gao Yacong;Rong Jian(Beijing University of Civil Engineering and Architecture,Beijing 102616,China;Beijing University of Technology,Beijing 100124,China)

机构地区：[1]北京建筑大学,北京102616 [2]北京工业大学,北京100124

出　　处：《系统仿真学报》2020年第11期2112-2120,共9页Journal of System Simulation

基　　金：国家自然科学基金青年基金(5170080357)。

摘　　要：微观交通仿真模型参数标定是微观交通仿真技术应用的前提,针对模型参数标定类型划分和标定结果取值方法缺乏精细化分析问题进行改善研究,将参数标定体系划分为全局参数和局部参数,提出综合工程实测与智能寻优相结合的微观交通仿真模型参数标定方法,并以信息熵作为参数标定结果取值的分析指标,提出聚类递归的参数取值方法。结合信号交叉口实际调查数据设计对照仿真实验和交叉验证实验,证实改善方法的有效性。Model parameter calibration is the precondition of application of micro traffic simulation technology. In view of the lack of refined analysis in the model parameter calibration classification and calibration result determination, the corresponding improvement is proposed. The parameter calibration system divides global parameters and local parameters, provides a micro-simulation model parameter calibration method that combines engineering measurement and intelligent optimization. Based on the information entropy as the analysis index of the parameter calibration results, a method of parameter recursion after clustering is proposed. On the basis of the actual survey data of signalized intersections, A simulation experiments and cross-validation experiments are designed to confirm the effectiveness of the improved method.

关 键 词：微观交通仿真 模型参数标定 信号交叉口 交通工程 

分 类 号：TP391[自动化与计算机技术—计算机应用技术] U491[自动化与计算机技术—计算机科学与技术]