微观交通仿真系统参数校正研究  

Study on Microscopic Traffic Simulation Model Systematic Parameter Calibration

作　　者：孙剑[1] 杨晓光[1] 刘好德[1] 

机构地区：[1]同济大学道路与交通工程教育部重点实验室,上海200092

出　　处：《系统仿真学报》2007年第1期48-50,159,共4页Journal of System Simulation

基　　金：国家自然科学基金项目(50578123)

摘　　要：微观交通仿真模型在交通系统管理、控制和优化中得到了广泛的应用,大多数仿真模型参数均是针对模型开发国家的交通情况确定的,并不一定适合我国国情。提出了微观仿真模型参数校正流程,然后以合肥市大东门区域VISSIM仿真系统模型为实例,建立了仿真模型参数校正的遗传模拟退火启发式算法,实现了对VISSIM的仿真参数的自动化校正,根据实测结果和仿真实验比较分析,验证了算法的有效性。Microscopic simulation models have been widely used in traffic system management, control and optimization; however, most of the simulation parameters are determined according to the foreign situations and not always suit for our china. Firsdy, the procedure for microscopic simulation model calibration was put forward. Then a VISSIM simulation example of Big-East-Door area in Hefei was used. The application of a Genetic Simulated Annealing （USA） algorithm as an optimization method for finding a suitable combination of micro-simulation parameter values was proposed. Finally, field experimental results and simulation analysis were compared, which have proven that the algorithm is proper.

关 键 词：交通仿真模型 参数校正 遗传模拟退火 VISSIM 

分 类 号：U491.13[交通运输工程—交通运输规划与管理]