基于RBF-ARX模型的高速列车预测控制器设计  

Design on high-speed train predictive controller based on RBF-ARX model

作　　者：刘伯鸿 连文博 李婉婉 LIU Bohong;LIAN Wenbo;LI Wanwan(College of Automation&Electrical Engineering,Lanzhou Jiaotong Uninversity,Lanzhou 730070,China)

机构地区：[1]兰州交通大学自动化与电气工程学院

出　　处：《北京交通大学学报》2019年第5期73-79,共7页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

基　　金：国家自然科学基金(61164010)~~

摘　　要：针对高速列车在复杂环境运行时,传统预测控制器出现的动力学模型失配、在线辨识实时性不强等问题,提出一种基于RBF-ARX模型的列车预测控制方法.首先,构建RBF-ARX模型作为列车控制器的预测模型,使用SNPOM算法处理列车运行中的大量历史数据,得到预测模型的模型阶次及优化参数,并验证了模型的准确性.然后使用该模型构造预测控制器对列车目标曲线进行在线跟踪仿真.结果表明:基于RBF-ARX模型的高速列车预测控制器不仅解决了动力学模型失配的问题,并且在控制精度及舒适度等方面都优于传统预测控制器.When high-speed train are operating in the complex area,its tradition predictive controller will mismatch its dynamic model and not identify its model on-line.In order to solve this,a train predictive control method based on RBF-ARX model is proposed.Firstly,RBF-ARX model is constructed as the predictive model of train controllers,and SNPOM algorithm is used to process a large number of historical data in train operation to obtain the model order and parameters are optimized,and the accuracy of the model is verified.Then the model is used to construct apredictive controller to track the train target curve online.The results show that the predictive controller based on RBF-ARX model not only solves the problem of dynamic model mismatch,but also is superior to the traditional predictive controller in control accuracy and comfort index.

关 键 词：高速列车 RBF-ARX模型 系统辨识 预测控制 

分 类 号：U284.48[交通运输工程—交通信息工程及控制]