摘要
交通流量预测一直是实时自适应交通控制的关键问题。以城市道路网络中典型的两相邻交叉口为研究对象,提出了基于粒子群优化的RBF神经网络的信号交叉口交通流量预测模型。该模型以RB F神经网络为基础,采用分组优化策略,用粒子群优化算法对基函数的中心、方差和RBF网络权值进行优化,从而提高了网络的预测精度。通过仿真,并与其他算法对比,表明了本文方法的有效性。
This paper develops a particle swarm optimization （PSO） based RBF neural network model to predict the traffic flows in two adjacent intersections of an urban street network, which has long been considered a key element in a real time adaptive control system. It is composed of a RBF NN, whose parameters including clustering centers, variances of Radial-Basis Function and weights are optimized by PSO algorithm. Therefore it has not only simplified the structure of NN, but also enhanced training speed and mapping accurate. The performance and effectiveness of the proposed model are evaluated by using traffic data and compared with other RBF NN.
出处
《公路交通科技》
CAS
CSCD
北大核心
2006年第7期116-119,共4页
Journal of Highway and Transportation Research and Development
基金
山东省中青年科学家发展基金资助项目(031BS147)
山东省科技攻关计划资助项目(031080112)
关键词
粒子群优化
交通流
RBF网络
预测模型
particle swarm optimization
traffic flow
RBF NN
forecasting model
