粒子群优化广义预测控制在列车制动系统中的应用

Application of Particle Swarm Predictive Control in EMU Braking System

动车组列车制动系统是列车自动驾驶系统ATO的关键环节。针对动车组列车制动系统模型存在较大误差导致的列车在制动阶段控制效果较差这一问题,提出将动车组列车制动模型分为静态子系统和动态子系统两部分,根据列车制动系统的性能和要求,设计了CPSO(混沌粒子群算法)优化GPC广义预测控制器。该控制器由CPSO辨识动态子系统纯延时环节和外界干扰造成的GPC模型误差,并计算动车组列车所需的控制量。以CRH2型动车组为仿真对象,从仿真结果看出,CPSO-GPC控制器在遇到未知干扰时能够满足动车组列车对给定速度和位移的高精度跟踪要求。

The EMU braking system is the key of the automatic train operation(ATO).Aimed at the problem that the control effect of the EMU braking system is poor due to the large error of the EMU braking system model,the train braking model is divided into static subsystem and dynamic subsystem.According to the performance and requirements of the EMU braking system,CPSO(chaotic particle swarm optimization)is designed to optimize GPC generalized predictive controller.The controller identifies the GPC model error caused by the pure delay link and external disturbance of the dynamic subsystem by CPSO,and calculates the amount of control required.Taking CRH2 as the simulation object,the simulation results show that the CPSO-GPC controller can satisfy the high precision tracking requirements of the given speed and displacement of the trains when encountering unknown disturbances.