Study on Simulation Optimization of Dynamic Traffic Signal Based on Complex Networks

Abstract

Regional traffic signal optimization has been always a hot research field. The study of signal control in urban traffic network is a fundamental work to enhance traffic efficiency and depress traffic congestion. In this paper, the optimization goal which can be used to optimize the network performance index, includes average vehicle travel delay, queue lengths, intersection throughputs, etc. The improved mesoscopic traffic simulation model is presented to simulate the processes of traffic movement. And the SPSA(Simultaneous Perturbation Stochastic Approximation) algorithm is improved to the timing schemes of signal controllers in a network. And the first time the complex network theory is used to find the evolution of network topology interactive with optimal signal. The simulation results present that the optimal signal timings and optimal network topology can both alleviate traffic congestion and more effective on improve average network journey velocity.