多种群分层联合优化的城轨列车ATO研究

Study on Urban Rail Train ATO Based on Unified Optimization of Multi-swarm Hierarchical Structure

针对城轨列车控制系统运行模式曲线的设计需求,在满足安全、精确停车及各种约束条件下,以运行时间和能耗为目标,建立列车运行的多目标优化模型。将粒子群优化PSO算法与布谷鸟搜索CS相结合,即多种群分层PSO-CS联合优化算法。在底层,该方法将整个种群分成若干个小种群,小种群使用PSO算法寻优,再将寻优得到的精英粒子送往高层使用CS算法深度优化,高层优化后的粒子再返回到底层各自的小种群中去。将该方法与多目标粒子群MOPSO分别用于列车运行过程的优化,仿真实验表明,所提出算法得到的Pareto前沿解的收敛性和多样性更好。将该算法用于城轨列车运行曲线的优化设计中,不仅能够获得更优的列车运行控制策略,还能为设计者提供更多选择方案。

To meet the needs of operational mode curve design of urban rail transit train control system,the study established a multi-objective optimization model of urban rail transit train on the running time and energy consumption,under the conditions of safety,accurate stopping and various constraints.The particle swarm optimization（PSO）algorithm was combined with cuckoo search（CS）to form the multi-swarm hierarchical PSOCS joint optimization algorithm.In the bottom layer,the method divided the whole swarm into several small populations,and each small population was optimized with the PSO algorithm,from which the elite particles from each small population were selected and sent to the top layer,where they were deeply optimized by the CS algorithm,and returned to the bottom layer of original small population.Both this method and multi-objective particle swarm optimization（MOPSO）algorithm were put into use in the train operation process optimization,respectively.The simulation results show better convergence and diversity of the Pareto front solution obtained by the proposed algorithm.Therefore,when applied to the optimization design of urban rail train operation curve,the algorithm not only provides better train control strategies,but also offers more options for designers.