基于粒子群算法下的交叉口信号配时优化

为了合理地优化交叉口信号配时,通过分析交叉口的评价指标,建立了以车辆的平均延误、停车次数最小、交叉口整体通行能力最大、各相位有效绿灯时间和交叉口周期时长作为约束条件的数学模型。并通过改进前人研究基础上的动态加权系数,将多目标的非线性优化问题转化为单一目标的非线性规划问题,为了得出更科学稳定的解,提出了改善粒子群算法系统稳定性的2种方法,并将其与粒子群算法结合起来。然后以Matlab为求解工具,结合临邑市某一交叉口实例进行求解分析。最后的结果表明,在使用改进后的粒子群算法进行优化后交叉口通行能力较之现状提升了9%,延误下降了28%,停车次数下降了9%,且各项优化结果均优于Webster,改进后的算法在程序中运行300代,到216代才开始收敛,而未改进的算法稳定性较差,优化结果和收敛曲线则随着实验次数的变化而变化,最后的结论证明了该算法和模型的可靠性。

Optimization of Intersection Singal Timing Based on Particle Swarm Optimization

In order to reasonably optimize the signal timing of intersections,by analyzing the evaluation index of the intersection,a mathematical model is established which takes the average delay of vehicles,the minimum number of stops,the maximum overall capacity of the intersection as the objective function,the effective green time of each phase and the period length of the intersection as the constraint conditions.The multi-objective nonlinear optimization problem is transformed into a single-objective nonlinear programming problem according to the dynamic weighting coefficient improved on the basis of previous studies.In addition,in order to obtain a more scientific solution,two methods to improve the stability of particle swarm optimization system were proposed,and combines them with particle swarm optimization algorithm.Then matlab is used as a solution tool to analyze the solution with an intersection in Linyi city.The final results show that the traffic capacity of intersections is increased by 9%,delay decreased by 28%,parking times decreased by 9% after optimization using the improved particle swarm optimization algorithm,and all optimization results are better than Webster.The improved algorithm does not converge until it runs for 300 generations and 216 generations in the program,the unimproved algorithm has poor stability,and the optimization result and convergence curve change with the number of experiments.The final conclusion proves the reliability of the algorithm and the model.