基于事件触发预测控制的高速列车轨迹跟踪  

High-speed train trajectory tracking based on event-triggered predictive control

作　　者：金文斌 张佳辉 姚柏伶 孙鹏飞[2] 王青元[2] JIN Wenbin;ZHANG Jiahui;YAO Bailing;SUN Pengfei;WANG Qingyuan(CRRC Changchun Railway Vehicles Co.,Ltd.,Changchun,Jilin 130062,China;School of Electrical Engineering,Southwest Jiaotong University,Chengdu,Sichuan 610031,China)

机构地区：[1]中车长春轨道客车股份有限公司,吉林长春130062 [2]西南交通大学电气工程学院,四川成都610031

出　　处：《机车电传动》2024年第4期148-154,共7页Electric Drive for Locomotives

基　　金：中国中车科研计划项目重点课题(2023CCB71)。

摘　　要：高速列车运行环境的开放性导致列车容易受到外部阻力的影响,而该阻力难以通过建模的形式进行描述,为此文章研究了以“安全、平稳、准点”为跟踪目标的列车模型预测跟踪控制策略。首先,搭建了高速列车带约束、多目标预测控制模型,计算系统的Tube不变集,对列车状态和控制力进行紧束;在此基础上,为降低在线求解最优控制问题的计算量,设计了动态事件触发Tube模型预测控制策略,仅在满足触发条件时求解最优控制序列;最后,基于上海市域列车实际线路开展仿真验证,结果表明文章提出的控制策略可在保证较好跟踪性能的同时,显著减小模型预测控制的触发次数,降低在线计算量。The open operating environment of high-speed trains exposes them to disturbances from external resistances,which are difficult to characterize through modeling.To address this issue,this paper investigates a train model predictive tracking control(TMPC)strategy,with safety,ride comfort and punctuality as the tracking objectives.Firstly,a multi-objective predictive control model with constraints was constructed for high-speed trains,and the Tube invariant set of the system was calculated to tightly constrain train states and control forces.On this basis,in order to reduce the computational workload for solving the optimal control online,a dynamic event-triggered Tube model predictive control(DETMPC)strategy was designed,enabling the optimal control sequence to be solved only when the triggering conditions were satisfied.Furthermore,simulation verification was carried out using Shanghai urban trains running on real tracks.The results show that the control strategy proposed in this paper not only guarantee good tracking performance,but also reduce the frequencies of triggering the model predictive control(MPC),lowering the workload of online calculations significantly.

关 键 词：高速列车 列车自动驾驶 轨迹跟踪 模型预测控制 事件触发控制 

分 类 号：U292.914[交通运输工程—交通运输规划与管理]