基于多模型SGPC的高速列车速度及位置跟踪控制

研究了列车自动驾驶系统(ATO)的多模型阶梯式广义预测控制(SGPC)方法。首先针对高速列车运行过程的非线性特性,采用多模型策略,建立一个线性自适应模型和一个神经网络非线性模型并行辨识系统的动态特性。然后在每个采样时刻根据性能指标切换到最优的局部模型作为系统的当前模型。设计SGPC控制器,从而实现高精度的列车速度和位置跟踪控制。最后模拟列车运行中参数突变的影响,将该方法与传统的单模型线性广义预测控制方法进行对比研究。仿真结果表明:其控制效果明显优于单模型的线性广义预测控制器。

Multiple-model Stair-like Generalized Predictive Control of Positioning and Velocity Tracking in High-speed Train

The Multiple-model Stair-like Generalized Predictive Control (SGPC) method of the Automatic Train Operation (ATO) system is researched. Firstly, on the basis of the nonlinear characteristics of the high-speed train operation, a linear robust adaptive model and a neural network nonliear adaptive model are established by the multiple-model method to identify dynamic characteristic in parallel. Secondly, the best sub-model is selected as global model according to the switching index, and a stair-like generalized predictive controller is designed to ensure high precision tracking in both position and velocity. Consequently, the full-range operation is achieved. Finally, simulation results showed that the control effect by the proposed method is superior to that of normal linear generalized predictive controller, and can readily accommodate itself to the train operation parameters vareations.