基于KH算法的高速列车ATO控制策略优化研究

摘要

In order to overcome the problems caused by the simplification of line parameters and the application of train traction calculation single-mass model, the EMU multi-mass model is established by including the power-off high-speed train coasting through a neutral section into the operating conditions.With the constraint of the fixed running time defined in the train schedule, an ATO control strategy optimization model for high-speed trains is established based on energy consumption, punctuality, accuracy of parking and comfort.The KH (Krill Herd) algorithm is used to optimize ATO control strategy for high-speed trains.Taking the data of a certain section of Lanzhou-Xinjiang high-speed railway as a simulation example, the results show that the KH algorithm can obtain a better ATO control strategy than the particle swarm optimization algorithm with fewer iterations and the coasting of the train through neutral sections can affect the optimization results.This validates the superiority of the proposed algorithm and the rationality of including the train coasting through a neutral section into the operating conditions in optimizing ATO control strategy of high-speed trains.%针对传统高速列车自动驾驶(ATO)控制策略优化时简化线路参数、列车牵引计算采用单质点模型等问题,将列车通过牵引供电分相区断电惰行纳入运行工况,建立动车组多质点模型.在满足列车运行图固定运行时间条件下,以能耗、准点性、停车准确性及舒适性为指标建立高速列车ATO控制策略优化模型.利用磷虾群(KH)算法对高速列车ATO控制策略进行优化.以兰新高速铁路某区间线路数据为例,仿真测试表明KH算法可以在较少的迭代次数下获得较粒子群算法更优的ATO控制策略,且列车过分相区断电惰行会对优化结果产生影响,验证了所提算法在优化高速列车ATO控制策略中的优越性及将列车过分相区断电惰行纳入运行工况的合理性.