微观交通仿真模型参数标定发展与优化实证研究  

Empirical Study on Development and Optimization of MicroscopicTraffic Simulation Model Parameter Calibration

作　　者：单籽跃 周晨静[1,2] 韦昌永 高亚聪 SHAN Ziyue;ZHOU Chenjing;WEI Changyong;GAO Yacong(Beijing University of Civil Engineering and Architecture,General Aviation Technology Beijing Laboratory,Beijing 100044,China;Beijing Future Urban Design Advanced Innovation Center,Beijing 100044,China;Guangxi Zhuang Autonomous Rejion Expressway Development Center,Nanning 530028,China;Beijing University of Technology,Beijing 100124,China)

机构地区：[1]北京建筑大学通用航空技术北京实验室,北京100044 [2]北京未来城市设计高新尖创新中心,北京100044 [3]广西壮族自治区高速公路发展中心,南宁530028 [4]北京工业大学,北京100124

出　　处：《交通工程》2022年第4期46-53,共8页Journal of Transportation Engineering

基　　金：国家自然科学基金青年基金项目(5170080357);北京建筑大学研究生创新项目PG2022037;广西高速公路交通拥堵防控智能决策系统.

摘　　要：微观交通仿真模型参数标定结果可移植性差、解释性差是制约仿真技术应用的瓶颈,在总结梳理当前微观仿真模型参数标定发展历程的基础上,提出了精细分类标定和分类取值的思想与具体实施路径.结合以往方法形成4种交叉对比方案,以城市信号交叉口为模拟对象,通过仿真实验对该4种方案的标定结果进行比选分析,结果显示所提出优化方法在参数标定结果上更具有物理解释力,在标定实验结果上准确性更高,证实对多种类参数进行精细化区分是提高标定效果的有效途径.The poor portability and interpretability of the calibration results of microscopic traffic simulation model parameters are the bottleneck of restricting the application of simulation technology.On the basis of summarizing the development process of current micro simulation model parameter calibration,the idea and specific implementation measure of fine classification calibration and classification value are proposed.Taking urban signalized intersections as simulation objects,the calibration results of the four schemes are compared and analyzed through simulation experiments.The results show that the proposed optimization method has more physical explanatory power in parameter calibration results and higher accuracy in calibration experimental results,It is proved that the fine distinction of various parameters is an effective way to improve the calibration effect.

关 键 词：仿真模型参数标定 微观交通仿真 信号交叉口 分类标定 分类取值 

分 类 号：U491[交通运输工程—交通运输规划与管理]