Train Movement Simulation And ATO Control Strategy Optimization In Urban Rail Transit

In order to improve the precision of train movement simulation, the paper constructs the exact single-particle model based on train traction calculation and adopts the hybrid control strategy which has better performance at present in vehicle control, thus the software can precisely calculate the train running information of speed, locations and such on. Main design idea is as follows:the software builds the circuit diagram and related equipments by reading packaged text data, uses interactive interface to set the parameter of circuit, equipment and train, adopts the hybrid control strategy, designs the modules of train running route planning, the timing of speed adjustment, the control of speed range, the point of stopping break for optimizing the train movement simulation. The software will have ability to simulate the movement in various conditions and use the speed-distance curve to show the process.The results of the train movement process simulation in the software show some defects of hybrid control strategy, such as the frequent shift of working conditions, the far lower speed compared to the limit. The performance of the strategy is also uncontrollable. This paper introduces a particle swarm optimization based on the reference of previous studies. The optimization decomposes the control strategy into the sequence of working conditions and effective distance and the distance sequence is the data of particle. It adopts the areas above and under the speed limit as the evaluation indicators to calculate the best sequence of working conditions by greedy algorithm. It uses the PSO based on adaptive dynamic neighborhood topology and generalized learning(ADPSO) to optimize the distance sequence. The simulation shows the optimization is valid by contrasting of other intelligent algorithms. According to the test, the performance of the optimization result is also better than the target curve and the curve of hybrid control.