Study On Traffic Control Based On Improved Particle Swarm Algorithm

With the constant expansion of city scale and the sharp increase of car number, the key and difficult point in current traffic control field is how to maximize the capacity of existing urban transport network in the limited traffic resource to ensure the reasonable traffic flow.Firstly, for the problem of local optimum in particle swarm algorithm, the crossover and mutation operation of genetic algorithm are introduced PSO to improve it, and the steps of improved PSO are elaborated. That the improved particle swarm algorithm is superior to genetic algorithm and particle swarm optimization in terms of convergence speed and stability.Secondly, in order to improve the accuracy of the traffic flow control and optimization, chaos theory, applied to PSO is used to get the kernel parameter and Penalty coefficient of LS-SVM. Then ECLS-SVM is proposed to predict traffic. Comparing the predicted results of the steps based on ECLS-SVM algorithm and the predicted time and mean square error of algorithms of different models, ECLS-SVM algorithm can effectively improve the accuracy and efficiency of traffic flow forecasting. It has important theoretical significance and practical value for rational allocation and planning of traffic network resources.Based on the traffic flow prediction, the PSO is utilized for optimal control of the traffic lights in single and dual intersection for the sake of eases the pressure of urban traffic congestion and enhances the traffic efficiency. Mathematical models of traffic control in single and double intersection are established in connection with the specific example of traffic signal control.Thirdly, the standard PSO is improved by applying genetic algorithm in solving the local optimization and constraints problems of standard PSO. Then the improved GA-PSO is applied to traffic control. On the basis of single intersection model combining with previous traffic control model, the GA-PSO performs better in optimizing traffic control algorithm in contrast with PSO. Consequently, the double intersection traffic control is studied and traffic coordination optimization model is built. The GA-PSO is utilized for optimizing the model and then compare with standard PSO.Last, by comparing traffic control algorithm of standard PSO with improved PSO algorithm, the GA-PSO algorithm can increase the global search capability and can avoid falling into local optima. The Improved PSO algorithm is better in solving problem of traffic control, and it completely avoid the problems of standardized errors, incomplete statistics, local convergence and other issues, it can ensure a good traffic control.