摘要
城市交通系统是一个随机性很强的、复杂的巨型系统。为了获得良好的通行效率,必须对城市区域交通协调控制信号进行整体优化,但是到目前为止还没有一个能较好完成此项任务的、实用的实时智能优化方法。在粒子群优化算法中引入灾变策略和模型,开发了灾变粒子群优化算法,解决了基本粒子群算法易陷入局部极小点的缺陷,并将其应用于城市区域交通协调控制信号配时优化。仿真结果表明:与基本粒子群算法(在陷入局部极小点时)、固定周期法和遗传算法等配时方法相比,采用所开发的灾变粒子群优化算法对区域交通协调控制信号进行智能优化配时,被控区域的车辆平均延误可以分别平均减少25.2%、41%和11.5%,并可以大大提高路口的通行效率。开发的灾变粒子群优化算法也可以应用于其他许多对象的优化。
Urban traffic system is a huge system in a random way,it is necessary to optimize the traffic control signals in an area,but there isn't still a much applied algorithm for intelligent optimization in time until today.A catastropheparticle swarm optimization algorithm(CPSO) has been developed through introducing a cusp catastrophic model in particle swarm optimization algorithm,the stability and parameters of which have been researched ,then the problem on getting in local best point easily has been solved.The algorithm has been effectively used in dealing with the optimization of signal timing to urban traffic.Simulation data shows that signal timing optimization to urban traffic based on catastrophe-particle swarm optimization algorithm could separately reduce 25.2% ,41% and 11.5% of the average delay per vehicle in area based on PSO,fix cycle means and genetic algorithm approach.And which has a much wider range of applications.
出处
《计算机工程与应用》
CSCD
北大核心
2005年第29期19-23,共5页
Computer Engineering and Applications
基金
广东省自然科学基金(编号:010486)
广东省教育厅高校自然科学研究项目(编号:Z03075)资助
关键词
灾变粒子群优化算法
尖点灾变模型
稳定性
城市交通控制
车辆延误
信号配时
catastrophe-particle swarm optimization algorithm,a cusp catastrophic model,stability,urban traffic control,average delay per vehicle,signal timing