电工技术学报2025，Vol.40Issue(4)：1034-1045,12.DOI:10.19595/j.cnki.1000-6753.tces.240201
一种改进型永磁同步电机无模型预测电流控制策略
An Improved Model-Free Predictive Current Control Strategy for Permanent Magnet Synchronous Motors
摘要
Abstract
Permanent magnet synchronous motor(PMSM)has been widely used in industrial production,marine propulsion,and electric vehicle drive systems.To realize the high-performance control of PMSM,the finite control set model predictive current control(FCS-MPCC)algorithm has been applied to the speed control of PMSM by its single control mode and excellent dynamic performance.However,the prediction accuracy of this method largely depends on the accuracy of the mathematical model of the motor.In the actual control system,factors such as the magnetic saturation effect and temperature rise will cause parameter changes in motors,leading to the decline of model predictive control accuracy. The ultralocal model has been proposed as a model-free predictive control algorithm instead of the motor mathematical model to solve the parameter sensitivity problem in the FCS-MPCC algorithm.In this case,the current prediction can be realized by identifying the lumped parameters in the ultralocal model.The current sampling values of three consecutive sampling periods are subtracted successively to obtain the current error values for representing the lumped parameters in the ultralocal model. However,a stagnation problem exists in the current update.The specific phenomenon of this problem is that the phase current waveform appears to have an obvious peak,and the switching state of the inverter is no longer updated in several periods.This paper analyzes the mechanism of this phenomenon and then proposes a suppression strategy based on power factor angle calculation.The phenomenon of the current updating stagnation mainly occurs in the period with small phase current variation amplitude.During the initial several sampling cycles in this period,the current error value is unchanged even if the switching state of the inverter is different,which causes the lumped parameter expression to cross zero.The ultralocal model cannot accurately predict the current.The switching state of the inverter remains unchanged without updating under the inaccuracy prediction. Therefore,this paper adopts two steps to suppress such a phenomenon.The current updating stagnation state is first judged,and then the calculated switching state is updated on the inverter to make the system exit the stagnation state.When the system enters the current updating stagnation state,its main feature is that the switching state of the inverter remains unchanged for several periods.Once the continuous period value of the switching state of the inverter reaches the set value,the inverter state is directly forced to update.Then,to get the appropriate switching state,it is necessary to determine the sector of the reference voltage vector.Thus,a method based on power factor angle calculation is proposed.In this method,the sampled current and voltage of the current dq axis are calculated to obtain the value of the power factor angle,and the position of the reference voltage vector is calculated. Finally,the effectiveness of the proposed method is verified by experiments on a PMSM pair-tow platform.Under low speed and light load conditions,the THD value of the current after the suppression is reduced from 36.68% to 14.66% .Compared with the traditional FCS-MPCC algorithm,the THD value of phase current is reduced by 8.36%,which proves that the proposed method can reduce the sensitivity of motor parameters.关键词
永磁同步电机/预测控制/超局部模型Key words
Permanent magnet synchronous motor/predictive control/ultralocal model分类
信息技术与安全科学引用本文复制引用
罗力岩,樊启高..一种改进型永磁同步电机无模型预测电流控制策略[J].电工技术学报,2025,40(4):1034-1045,12.基金项目
国家自然科学基金资助项目(62373168). （62373168）
