车载超级电容系统的RBF神经网络自适应鲁棒滑模控制方法  

RBF neural network adaptive robust sliding mode control method for on-board supercapacitor system

作　　者：王晓侃[1,2] 王琼 WANG Xiaokan;WANG Qiong(School of Electronic Information Engineering,Beijing Jiaotong University,Beijing 100044,China;Henan Mechanical and Electrical Vocational College,Xinzheng 451191,China)

机构地区：[1]北京交通大学电子信息工程学院,北京100044 [2]河南机电职业学院,新郑451191

出　　处：《中南民族大学学报（自然科学版）》2022年第1期59-63,共5页Journal of South-Central University for Nationalities：Natural Science Edition

基　　金：河南省重点科技攻关资助项目(172102210124,202102210269)。

摘　　要：针对城市轨道交通加速和制动过程中再生制动能量利用率不高、引起电网电压波动等问题,提出一种车载超级电容RBF神经网络自适应鲁棒滑模控制方法 .设计基于超级电容的车载储能控制系统,利用状态空间平均法建立电路模型,进行精确反馈线性化,构建了神经网络自适应滑模控制器.通过仿真验证:该方法可以有效抑制牵引电网电压波动,最大限度地回收和利用列车的再生制动能量.Urban rail train has the problems of low utilization of regenerative braking energy and voltage fluctuation during acceleration and braking,an on-board supercapacitor system based on RBF neural network adaptive robust sliding mode control is proposed to address these problems. The mathematics model is established by using state space averaging method,the accurate feedback linearization is carried out on the model,and an adaptive sliding mode controller based on neural network is designed.The simulation results show that proposed method can effectively restrain the voltage fluctuation of traction power grid,and maximize the recovery and utilization of regenerative braking energy of trains.

关 键 词：能量控制 车载超级电容 神经滑模控制 城轨列车 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]