装甲车轮毂电机自适应预测控制  

Adaptive Predictive Control of ArmoredWheel Load Motor

作　　者：周春明 ZHOU Chun-ming(School of Mechanical and Electronic Engineering,Liaodong College,Dandong 118000,China)

机构地区：[1]辽东学院机械电子工程学院,辽宁丹东118000

出　　处：《火力与指挥控制》2021年第3期119-124,130,共7页Fire Control & Command Control

基　　金：辽宁省教育厅科学研究基金资助项目(2019JH 8/10100048)。

摘　　要：针对经典有限集模型预测控制(FCS-MPC)算法在无位置传感器控制应用场合依赖电机参数匹配精度等问题,提出一种参数自适应设置改进的FCS-MPC预测控制算法,算法将系统干扰等未建模因素折算为改进模型的等效电压总误差,并依据Popov超稳定性定理推导出改进模型参数的自适应律,实现等效总误差和总电感在线辨识估计,进而推导出电机转位置观测器的辨识表达式,实现改进算法的自适应预测控制,结合补偿后低能滤波和均值修正有效消除直流偏置量,进一步提高位置观测精度。仿真实验结果表明,改进算法在电机不同工况和负载下能准确估计转子的位置,验证了算法的有效性。An improved adaptive finite set model predictive control(FCS-MPC)control algorithm with no parameter dependence is proposed to solve the problem that the classical FCS-MPC algorithm relies on motor parameters in the case of position sensorless control.All unmodeled interference factors are converted into equivalent total errors to improve the classical FCS-MPC prediction model,and based on Popov’s superstability theorem,the adaptive law of the parameters of the improved model is derived,realizing the online identification and estimation of equivalent total error and total inductance,and then the identification expression of the motor rotation position observer is derived,and the adaptive predictive control of the improved algorithm is realized.Finally,combined with low energy filtering and mean value correction after compensation,the DC offset is effectively eliminated to further improve the position observation accuracy.The simulation results show that the proposed algorithm can accurately estimate the position under different working conditions and loads of the motor,and the effectiveness of the algorithm is verified.

关 键 词：永磁同步电机 无位置传感器 有限集模型预测控制 无参数依赖 位置观测器 

分 类 号：TJ01[兵器科学与技术—兵器发射理论与技术] TM301[电气工程—电机]