多约束永磁同步电机稳定模型预测控制策略

摘要

针对多变量约束条件的永磁同步电机(PMSM)调速问题,提出一种稳定模型预测控制(MPC)策略.采用更加简单、精确的PMSM线性化模型设计一个线性MPC控制器；在线性MPC策略中引入终端函数和终端多面体不变集概念,通过构造包含多变量约束条件的线性规划问题,使线性MPC控制器能够保证系统的闭环稳定性.与传统抗饱和(Anti - Windup)方法相对比,MPC显示出在多变量系统和多变量约束条件下的应用优势及潜力.仿真结果表明,稳定MPC控制器结构简单,计算量小,静、动态性能优异,鲁棒性强.%A stabilizing model predictive control ( MPC) strategy is proposed which can efficiently deal with the speed control of PMSM with tnulti-variable constraints. A single linear MPC controller is designed with a simpler and more accurate linearization model. Moreover, introducing the conceptions of terminal function and terminal polytope invariant set into the linear MPC algorithm, the linear program problem with multi-variable constraints is constructed so that the stability of closed-loop system can be guaranteed with the MPC controller designed. With contrast to anti-windup method, the application superiorities and potential of MPC for multi-variable system and multi-variable constraints system are presented. The simulation results verify that the stabilizing MPC has simple structure, small calculation demand, excellent static and dynamic performance and strong robustness.