初始化对交通元胞自动机模型稳定性的影响  

Influence of the Initialization Method on the Stability of Traffic Cellular Automata Model

作　　者：邓建华[1] 冯焕焕[1] 葛婷 DENG Jianhua;FENG Huanhuan;GE Ting(College of Civil Engineering,Suzhou University of Science and Technology,Suzhou 215011,China)

机构地区：[1]苏州科技大学土木工程学院,江苏苏州215011

出　　处：《复杂系统与复杂性科学》2023年第2期105-110,共6页Complex Systems and Complexity Science

基　　金：国家自然科学基金(51808370);苏州科技大学基金项目(341311108,XKQ201305)。

摘　　要：针对现有交通元胞自动机模型运行初始不稳定,数据输出存在较长时间的初始波动问题,基于Fisher-Yates算法原理,设计出一种新的交通流初始化方法。该方法可以确保车辆从进入元胞空间到随后的演化更新,其位置及更新时机的随机性。通过对采用新交通流初始化方法的模型进行演化实验,结果表明:任意空间占有率条件下交通流的初始波动区间都在50步以内;当演化更新总步数达到3600步时,模型剔除初始波动区间的输出数据已充分收敛,这时模型运行已足够稳定。Aiming at the initial instability of the existing traffic cellular automata model and the initial fluctuation of data output for a long time,a new traffic flow initialization method is designed based on the principle of Fisher-Yates algorithm.This method can ensure the randomness of the location and update timing of the vehicle from entering the cell space to the subsequent evolution and update.Through the evolution experiment of the model using the new traffic flow initialization method,the results show that the initial fluctuation range is within 50 steps under the condition of arbitrary space occupancy;When the evolution update reaches 3600 steps,the output data of the model after excluding the output of the initial fluctuation interval has converged enough,and the operation of the model is stable enough.

关 键 词：元胞自动机 交通流初始化 Fisher-Yates算法 初始波动区间 

分 类 号：U491.1[交通运输工程—交通运输规划与管理]