基于分布式模型的动车组预测控制方法

Predictive Control Using a Distributed Model for Electric Multiple Unit

针对动车组由若干动车/拖车组成的动力单元固定编组耦合构成,难以用集中式模型进行有效描述的问题,提出一种动车组运行过程的分布式描述与建模方法.基于动车组牵引/制动特性曲线和实际运行数据,采用子空间模型辨识方法建立了动车组各动力单元的分布式状态空间模型;提出基于分布式模型的动车组预测控制方法,给出了各动力单元牵引/制动力和运行速度同步跟踪控制算法.基于CRH380AL型动车组运行过程数据的对比仿真结果验证了本文方法的有效性.

A distributed power electric multiple unit （EMU） include coupled integral power units characterized by several motor vehicles and trailers, whose dynamic behaviors are difficult to describe with the typical single-point-mass modeling method. To handle these problems, this paper presents a distributed description and the corresponding modeling approach for the EMU operation process. Based on the traction/brake characteristic curve and the actual operation data, employ the subspace identification method to establish the distributed state-space model for each power unit. According to the principle of model predictive controller, then propose a synchronous tracking control algorithm for traction/braking forces and running speed of each power unit. With the actual data of the CRH EMU type-380AL, comparative simulation results prove the effectiveness of the proposed method.