考虑参数灵敏度的交通仿真模型参数寻优研究  

Study on parameter optimization of traffic simulation model considering parameter sensitivity

作　　者：王雪芹 肖志豪 WANG Xueqin;XIAO Zhihao(College of Traffic and Transportation, Chongqing Jiaotong University, Chongqing 400074, China)

机构地区：[1]重庆交通大学交通运输学院,重庆400074

出　　处：《重庆理工大学学报（自然科学）》2022年第6期207-215,共9页Journal of Chongqing University of Technology：Natural Science

摘　　要：当前广泛使用的交通仿真软件大多是根据国外的道路交通条件设计,仿真结果无法反映我国道路交通运行的真实情况。为此,以VISSIM仿真软件中的车辆跟驰模型和车辆换道模型为具体研究对象,以城市快速路交织区为具体仿真场景,根据调查的交通流数据对交织区车流的期望速度分布进行调整;引入LH-OAT算法对模型参数进行灵敏度分析,确定影响模型输出结果的关键参数,应用遗传算法(GA)对关键参数进行优化取值;依据校验指标对默认参数仿真结果和标定参数仿真结果进行误差分析。结果表明:进行参数灵敏度分析可有效简化模型,提高仿真模型输出结果的可控性,降低模型的运行成本;对关键参数进行标定,仿真结果会更加贴近实测数据,更能反映真实的道路交通运行情况。At present,most of the widely used simulation software are designed according to foreign road traffic conditions,and the simulation results can not reflect the real situation of road traffic operation in China.Therefore,combined with the operation characteristics of road traffic in China,it is of great significance to correct the parameters of traffic simulation model.Taking the vehicle following model and vehicle lane changing model in VISSIM simulation software as the specific research object and the urban expressway weaving area as the specific simulation scene,firstly,the expected speed distribution of vehicle flow in the weaving area is adjusted according to the investigated traffic flow data;Then LH-OAT algorithm is introduced to analyze the sensitivity of model parameters,and then Genetic Algorithm(GA)is used to optimize the key parameters;Finally,the error between the simulation results of default parameters and calibration parameters is analyzed according to the verification index.The results show that the parameter sensitivity analysis can effectively simplify the model,improve the controllability of the output results of the simulation model and reduce the operation cost of the model;By calibrating the key parameters,the simulation results will be closer to the measured data and better reflect the real road traffic operation.

关 键 词：交通仿真模型 参数灵敏度分析 参数标定 LH-OAT算法 GA算法 

分 类 号：U491.2[交通运输工程—交通运输规划与管理]