货运铁路会让列车调度指挥与运行控制协同优化研究

Research on Collaborative Optimization of Dispatch Command and Operation Control for Meeting and Passing Trains of Freight Railway

针对既有货运铁路调度集中系统和列车运行控制系统分立设置,列车会让质量高度依赖调度员和司机的经验水平,突发事件响应慢等问题,构建货运铁路协同优化仿真系统,实现列车会让过程中调度指挥与运行控制的有机统一。首先依据货运铁路列车会让过程,增加通过列车和对向列车的位置、速度,以及列车牵引制动特性、线路条件等约束,构建列车协同会让优化模型;然后采用双深度Q网络(DDQN)进行模型求解,在计算列车速度曲线和选择列车工况序列时,考虑对向列车速度和距会让点的距离,协同优化通过、会让列车的运行过程。最后以某地方铁路实际线路、列车数据为例,验证模型和算法的可行性和适应性。试验结果表明,该系统可辅助调度人员实时掌握列车运行状态,有效调整列车运行速度,提高列车会让运输效率。

Aiming at the problem that the centralized traffic control system and the train operation control system of the existing freight railway are set up separately,and the train meeting and passing quality highly relies on the experience level of dispatchers and drivers,and the response to emergencies is low,a freight railway collaborative optimization simulation system is constructed to achieve organic unity between dispatch command and train operation control during train meeting and passing process.Firstly,based on the train meeting and passing process of freight railway,a model for collaborative optimization of train meeting and passing was established by adding constraints such as positions and speeds of passing train and oncoming train,traction and braking characteristics and line conditions.Secondly,the Double Deep Q-Network(DDQN)was employed for model solving,when calculating the train speed curve and selecting the sequence of train operating conditions,the speeds of oncoming trains and the distance to the train meeting and passing point were considered to collaboratively optimize the process of train passing.Finally,taking actual railway lines and train data from a specific location as an example,the feasibility and adaptability of the model and algorithm were verified.Experimental results demonstrated that the system can assist dispatchers in real-time monitoring of train operation status,effectively adjust train speeds,and improve the transport efficiency of train meeting and passing.