高性能永磁同步电机显式模型预测控制算法研究  

Research on High-Performance Explicit Model Predictive Control Algorithm for Permanent Magnet Synchronous Motors

作　　者：刘忠永 范涛 何国林[2] 温旭辉[1,2] Liu Zhongyong;Fan Tao;He Guolin;Wen Xuhui(University of Chinese Academy of Sciences,Beijing 100049 China;Institute of Electrical Engineering Chinese Academy of Sciences,Beijing 100190 China)

机构地区：[1]中国科学院大学,北京100049 [2]中国科学院电工研究所,北京100190

出　　处：《电工技术学报》2023年第22期6039-6058,共20页Transactions of China Electrotechnical Society

基　　金：国家重点研发计划资助项目(2021YFB2500600)。

摘　　要：该文提出一种应用于永磁同步电机的级联式高性能显式模型预测控制算法,基于多参数规划思想建立系统参数化模型,离线求解有约束条件下的最优解并以状态量的分段仿射函数形式保存,解决了连续控制集模型预测控制算法在线求解的算力需求问题;全面介绍了显式模型预测控制的应用思想及设计流程,分析了在永磁同步电机控制中模型失配、死区效应、数字延时等非理想因素带来的影响并给出了应对措施;与传统PI控制算法进行对比,通过仿真与实验验证了显式模型预测控制算法的有效性及优越性。Since the 1970s,model predictive control(MPC)algorithms have proven to be effective control strategies for multi-input and multi-output nonlinear dynamic systems with complex constraints.However,due to the significant computational demands associated with optimization problems,applying MPC to control objects with fast response time requirements,such as motor systems,presents challenges in achieving iterative solutions for constrained problems within short control cycles.This paper proposes a high-performance control strategy for permanent magnet synchronous motors based on the explicit model predictive control(EMPC)concept.The strategy involves the establishment of linearized models for current and speed control using multiparametric programming to eliminate coupling effects and nonlinear influences between states.The offline solution yields optimal control actions under effective constraint conditions,stored as piecewise-affine functions.During operation,the corresponding optimal control actions are obtained by querying the partition containing the current state combination.The paper comprehensively analyzes model mismatch,digital delay,and dead-zone effects in EMPC applications and provides corresponding solutions.A fully automated dyno test platform verifies the proposed algorithm's efficacy.The paper introduces the theoretical foundations of multiparametric programming and presents a comprehensive design process for explicit model predictive control.It establishes linearized models for controlling current and speed in permanent magnet synchronous motors while linearizing voltage and current constraints.The effects of parameter mismatch,digital delay,and inverter nonlinearity on EMPC are analyzed,and respective compensatory measures are proposed.The EMPC problem definition and optimal control law are solved using the MPT3 toolbox,displaying critical partitions and optimal cost functions under different state combinations.The simulation verifies the algorithm and the preceding analysis of various non-ideal fact

关 键 词：永磁同步电机(PMSM) 显式模型预测控制 最优化控制 转速预测控制 电流预测控制 

分 类 号：TM341[电气工程—电机]