The Genetic Algorithm Application Research Of Automatic Train Regulation On CBTC System

With the rapid development of urban railway traffic automatic train control technology, CBTC (Communication-Based Train Control) is the trend of urban railway traffic control system in the future. The moving block technology based on CBTC brings new challenge to automatic train regulation, because higher traffic density and shorter headway of trains result in more difficulty for train operation adjustment. So the correlation study has practical significance.In view of this, study algorithms of automatic train regulation based on GA (Genetic Algorithms, GA) considering combinatorial optimization of train operation adjustment and feature of CBTC system, then, corresponding simulation is carried out.First of all, model of automatic train regulation pertinently is put forward throng analyzing characteristics of urban railway traffic,moving block and train operation adjustment.Secondly, based on the model and characteristics of automatic train regulation of urban railway traffic, GA is selected to get solution of the model. Application process of GA is detailed description. Different from traditional coding method by train serial number, real coding is put forward according to time. GA is improved by using elitism in selection, two-point crossover, mutation operator with breeding device, and constrained conditions via parameter-free penalty functions method.At last, based on Batong line of Beijing metro in china, simulation environment is constructed. In the paper, the simulation is carried out under train serious delay and train group delay according to practical line data and operational situation of Batong line combined with CBTC system. Simulation results prove algorithms of automatic train regulation based on GA is effective and feasible for busy line.