A prediction model with wavelet neural network optimized by the chicken swarm optimization for on-ramps metering of the urban expressway

Abstract

Urban expressway, which plays a significant role in medium-and-long distance express travel, influences transport efficiency of an area in a city. The operation efficiency of an urban expressway system can be promoted by on-ramp metering (ORM). Based on ALINEA (asservissement linéaire d'entrée autoroutière) algorithm, this paper proposed an improved ALINEA method with a wavelet neural network (WNN) optimized by chicken swarm optimization (CSO). The algorithm integrates K-means algorithm to select key-point for dynamic multiple on-ramps coordinated control. The amended ALINEA method mainly aimed at solving the problems of the in-flow of the next control period and the mainline multi-lanes condition. Simulation results demonstrated that the coordinated control algorithm proposed can increase the traffic efficiency of the urban expressway.