改进的动车组速度跟踪系统的无模型自适应控制  

Improved model-free adaptive control for EMUs velocity tracking system

作　　者：周靓 夏金凤 李中奇 ZHOU Liang;XIA Jin-feng;LI Zhong-qi(School of Electrical and Automation Engineering,East China Jiaotong University,Nanchang 330013,Jiangxi,China;State Key Laboratory of Performance Monitoring and Protecting of Rail Transit Infrastructure,East China Jiaotong University,Nanchang 330013,Jiangxi,China;CRRC Zhuzhou Electric Locomotive Co.,Ltd.,Zhuzhou 412o0l,Hunan,China)

机构地区：[1]华东交通大学电气与自动化工程学院,江西南昌330013 [2]华东交通大学轨道交通基础设施性能监测与保障国家重点实验室,江西南昌330013 [3]中车株洲电力机车有限公司,湖南株洲412001

出　　处：《交通运输工程学报》2024年第2期267-280,共14页Journal of Traffic and Transportation Engineering

基　　金：国家自然科学基金项目(52162048,61991404);江西省主要学科学术和技术带头人培养计划项目(20213BCJ22002)。

摘　　要：为了提高列控系统跟踪精度与平稳运行,提出了一种改进的多输入多输出(MIMO)无模型自适应控制(MFAC)方法;基于动态线性化技术,将系统各动力单元输入输出数据等效成更符合高速动车组实际运行特性的全格式动态线性化(FFDL)数据模型;通过在目标准则函数中加入输出误差率,并对输出误差和输出误差率进行加权融合,推导出新的带有输出误差率的无模型自适应控制(MFAC-OER)方案;通过对FFDL数据模型的外界扰动、参数误差等不确定项进行延时估计,进一步提升了算法的控制性能和对系统的等价描述程度;以实验室配备的CRH380A型动车组半实物试验平台对该方法进行仿真测试,使其跟踪济南—徐州的实际速度-位移曲线,并与传统算法进行对比。仿真结果表明:通过MFAC-OER方法得到的动车组各动力单元速度误差为[-0.151,0.136]km·h^(-1),控制力和加速度分别在[-48,42]kN和[-0.785,0.687]m·s^(-2)以内且变化平稳,控制性能优于比例积分微分方法和传统MFAC方法;整体仿真结果证明了MFAC-OER方法不仅能快速到达系统稳态并且具有良好的抗外界干扰特性,满足动车组跟踪精度与安全要求。To improve the tracking accuracy and stable operation of the train control system,an improved multiple-input multiple-output(MIMO)model-free adaptive control(MFAC)method was proposed.Based on the dynamic linearization technology,the input-output data of each power unit of the system were equivalently transformed into a full form dynamic linearization(FFDL)data model that better fitted the actual operation characteristics of high-speed electric multiple units(EMUs).By incorporating the output error rates into the objective criterion function and weighting the fusion of output errors and output error rates,a new model-free adaptive control scheme with output error rates(MFAC-OER)was derived.The control performance of the algorithm and the equivalent description degree of the system were further improved by delayed estimation of uncertainty factors,such as external disturbances and parameter errors in the FFDL data model.The proposed method was simulated and tested on a CRH380A high-speed EMUs semi-physical test platform equipped in the laboratory to track the actual speed-displacement curve from Jinan to Xuzhou and compare it with some traditional algorithms.Simulation results show that the speed errors of each power unit of EMUs obtained by the MFAC-OER method are within[-0.151,0.136]km·h^(-1),with the control force and acceleration smoothly varying in the ranges of[-48,42]kN and[-0.785,0.687]m·s^(-2),respectively.The proposed method outperforms the proportional-integral-derivative(PID)and traditional MFAC methods in the control performance.The overall simulation results show that the MFAC-OER method can not only quickly reach the steady state of the system but also possesses good resistance to external disturbances,meeting the tracking accuracy and safety requirements of the EMUs.3 tabs,12 figs,30 refs.

关 键 词：动车组 列车自动驾驶 无模型自适应控制 动态线性化 输出误差率 扰动估计 

分 类 号：U266.2[机械工程—车辆工程]