永磁同步电机双环模型预测控制研究

Research on Double Loop Model Predictive Control of PMSM

有限控制集模型预测电流控制(FCS-MPCC)算法对永磁同步电机(PMSM)的8个电压矢量进行预测计算,根据建立的成本函数选择出最优电压矢量并应用,计算量较大。针对该问题,提出一种基于先择扇区原则的改进型模型预测电流控制算法。该算法减少了计算的电压矢量数,缩减了计算时间,对微处理器的性能要求降低。将控制外环速度环的PI控制也改为模型预测控制设计了二阶扩展状态观测器,改善了速度控制的动态性能,电机的速度控制响应更快。以PMSM控制的摇臂装置为研究对象,建立仿真模型,并搭建相应的实验平台,仿真和实验验证了该算法的伺服动态响应性与稳定性。通过实验证明了双环MPC算法比双环PI算法对逆变器的损耗更小,温升更低。

The finite control set model predictive current control(FCS-MPCC)predicts and calculates the eight voltage vectors of permanent magnet synchronous motor(PMSM).According to the established cost function,the optimal voltage vector is selected and applied,which leads to excessive calculation.To solve the problem,an improved model predictive current control(SS-MPCC,select sector MPCC)was proposed based on the principle of selecting sectors first.The algorithm decreased the number of voltage vectors to be calculated,shortened the calculation time,and reduced the performance requirements of microprocessors.The PI control that controls the outer speed loop was also changed to model predictive control(MPC).A second-order extended state observer was designed to improve the dynamic performance of speed control and make the speed control response faster.A simulation model was established by use of a rocker arm device controlled by PMSM,and the corresponding experimental verification platform was built to carry out the verification experiment.The servo dynamic response and stability of the MPC algorithm were verified by simulation and experiment.Experiments showed that the double loop MPC algorithm had lower loss and temperature rise of inverter than the double loop PI algorithm.