基于遗传算法与蝠鲼觅食算法的信控优化策略  

Optimization Strategy of Information Control Based on Genetic Algorithm and Manta Ray Foraging Algorithm

作　　者：姜圣 丁闪闪 华禹凯 JIANG Sheng;DING Shanshan;HUA Yukai(China Academy of Information and Communications Technology,Nanjing,210024;China Design Group Co.,Ltd.,Nanjing,210014)

机构地区：[1]中国信息通信研究院,南京210024 [2]华设设计集团股份有限公司,南京210014

出　　处：《现代交通技术》2024年第5期68-71,共4页Modern Transportation Technology

基　　金：江苏省交通运输科技与成果转化项目(KY2019067)。

摘　　要：主要研究智能信控优化策略,以提高城市干线平面交叉口的通行效率,同时,创新提出一种信号灯智能调控算法,相应算法融合了遗传算法优化的模糊控制和改进的蝠鲼觅食算法的优点。该方法通过对车流量与排队长度进行分级,利用遗传算法对偏差量化因子进行寻优,并根据模糊规则确定周期。在对蝠鲼位置进行多次寻优后,取得最优值,实现信号灯的动态调控。此外,运用VISSIM软件在真实运行的道路交叉口对算法有效性进行仿真验证,测试结果表明该算法满足交通管理要求,策略具备可行性。The research focuses on an intelligent information control optimization strategy to improve the passing efficiency of urban trunk plane intersection.At the same time,an intelligent control algorithm of signal light is proposed,which integrates the advantages of GA-Fuzzy control and improved manta ray foraging algorithm.By classifying the trafic flow and queue length,the quantization factor of deviation is optimized using genetic algorithm,and the period is determined according to fuzzy rules.The position of manta ray is optimized for many times to obtain the optimal value,and dynamic control of signal light is realized.In addition,the VISSIM software was used to simulate and verify the effectiveness of the algorithm at the real running intersection,and the test results showed that the algorithm meets the requirements of traffic management and the strategy is feasible.

关 键 词：城市交通 遗传算法 信号灯控制 

分 类 号：U121[交通运输工程] U491.54