H.X. Wang, H.J. Zheng

Pages: 79-90

Abstract
In order to overcome the problems of long prediction time and low prediction accuracy existing in traditional congestion prediction methods, this paper designs a traffic congestion prediction method based on genetic algorithm. The congestion information of traffic jam section is sampled and combined with the optimal grid parameter distribution model of traffic flow distribution, the output characteristics of congestion are extracted. On this basis, V2X communication technology is used to obtain the real-time traffic flow data in the road section, and the congestion volume is predicted according to the coding, cross mutation and selection process of genetic algorithm. The experimental results show that the maximum time of generating congestion prediction results is only 4.76s, and the maximum fitting degree between the predicted results and the actual results can reach 98%, which proves that the proposed method can effectively achieve the design expectation.
Keywords: traffic jam road section; congestion volume prediction; information sampling; feature analysis; v2x communication technology; genetic algorithm