MMC-RPC的多维泰勒网优化控制策略  

Multi-dimensional Taylor network optimal control strategy of MMC-RPC

作　　者：宋平岗 魏文龙 陈紫君 陈子豪 SONG Pinggang;WEI Wenlong;CHEN Zijun;CHEN Zihao(School of Electrical and Automation Engineering,East China Jiaotong University,Nanchang,Jiangxi 330013,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

出　　处：《机车电传动》2023年第2期123-130,共8页Electric Drive for Locomotives

基　　金：国家自然科学基金项目(51367008)。

摘　　要：为进一步解决铁路牵引供电系统中无功功率、负序电流和谐波失真率高等对电能质量的影响,将多维泰勒网(Multi-dimensional Taylor Network,MTN)优化控制应用于与模块化多电平换流器结合的铁路功率调节器(Railway Power Conditioner with Modular Multilevel Converter,MMC-RPC)。相比传统双闭环PI控制器,多维泰勒网控制器的优点在于抗扰动能力更强、鲁棒性更好和谐波失真率更低等。针对参数整定问题,将dq轴电流参考值与实际值之间误差平方的积分作为输出性能指标,并运用粒子群算法对MTN当中的参数进行优化。在MATLAB/Simulink软件中分别搭建了MMC-RPC采用MTN控制器和PI控制器的仿真模型,对电气化铁路中常见的一侧供电臂有机车负载和负载突变2种常见工况进行对比,验证了MTN控制器在MMC-RPC中的可行性和优越性。In order to solve the influence of high reactive power,negative sequence current and harmonic distortion rate on power quality in railway traction power supply system,multi-dimensional Taylor network optimization control was applied to railway power conditioner with modular multilevel converter(MMC-RPC).Compared with traditional double closed-loop PI controller,multi-dimensional Taylor network(MTN)controller has advantages of strong anti-disturbance ability,better robustness and lower harmonic distortion rate.Aiming at the parameter setting problem,the integral of the error square between the reference value and the actual value of the dq axis current was used as the output performance index,and the particle swarm optimization algorithm was used to optimize the parameters in MTN.In MATLAB/Simulink,the simulation models of MMC-RPC using MTN controller and PI controller were built respectively,and the two common working conditions of locomotive load and load mutation were compared in electrified railway,which verified the effectiveness and superiority of MTN controller in MMC-RPC.

关 键 词：铁路功率调节器 模块化多电平换流器 多维泰勒网控制器 粒子群算法 

分 类 号：U223.5[交通运输工程—道路与铁道工程]