针对目前交叉口信号配时优化目标单一、综合运行效率不高的问题，提出一种交叉口信号配时多目标优化方法。考虑绿灯时间、周期长度和饱和度等约束条件，通过加权系数法定义代价函数，使交叉口的延误时间、停车次数和通行能力在某种程度上达到最优。基于免疫克隆算法在处理多目标问题中具有最优解分布宽广性、均匀性好等特点，引入环境变异算子，提出环境变异免疫克隆算法对模型求解，增强了算法的全局搜索能力，提高了解的质量。仿真结果表明，与传统配时方法和改进粒子群算法相比，该文方法能有效减少信号交叉口的延误时间和停车次数，提高交叉口的运行效率。%To solve the problems of single optimization objective and low efficiency in signal timing,a novel multi⁃objective optimization method for signal timing of intersections is proposed. In consideration of constraint conditions of green⁃light dura⁃tion,cycle length and saturability,the cost function is defined by weighted coefficient method to make delay time,parking times and traffic capacity of intersections optimum to some extent. Based on the characteristics of uniformity and wide⁃distribu⁃tion of optimal solution while immune clone algorithm(ICA)is used to solve multi⁃objective problems,a new environment muta⁃tion operator was introduced,and the environment mutation ICA was proposed for model solution,which enhanced the global searching ability of the algorithm and improved the solution quality. The simulation results reveal that,compared with the tradi⁃tional timing method and inertia weight particle swarm algorithm,the presented method can efficiently reduce average delay, parking times of intersections,and improve the efficiency of traffic signal control.
展开▼
