摘要
针对无传感器表贴式永磁同步电机高速运行过程中电气参数摄动影响电流环性能和转子位置估计精度的问题,提出一种基于参数辨识的无传感器高速永磁电机无差拍电流预测控制方法。首先,为了提升电流环控制器的动态性能,结合永磁电机控制系统的特点,采用无差拍电流预测控制并进行模型参数敏感性分析。其次,针对多参数在线辨识存在的欠秩问题,提出在3种不同时间尺度下,采用基于神经元迭代求解的总体最小二乘法在线分步辨识电机定子电感、电阻和永磁体磁链。最后将辨识结果用于更新无差拍电流预测控制器及滑模观测器参数。实验结果表明,基于参数辨识的无传感器高速永磁电机无差拍电流预测控制方法能有效提高电流环控制器稳态性能及转子位置估计精度。
During the high-speed operation of sensorless surface-mounted permanent magnet synchronous motor(SPMSM),the perturbation of electrical parameters affects the performance of current loop and the accuracy of rotor position estimation.Therefore,a deadbeat predictive current control(DPCC)method for sensorless high speed permanent magnet motor based on parameter identification was proposed.Firstly,combined with the characteristics of permanent magnet motor control system,DPCC was adopted to improve the dynamic performance of the current loop controller.Besides,the parameter sensitivity of DPCC was analyzed.Secondly,in order to solve the rank deficient problem,a total least square(TLS)method based on neuron iterative solution was used to identify the inductance,resistance and permanent magnet flux linkage on-line and step by step.Finally,the identification results were used to update the parameters of deadbeat predictive current controller and sliding mode observer.The experimental results show that DPCC method of sensorless high-speed permanent magnet motor based on parameter identification can effectively improve the steady state performance of current loop controller and the accuracy of rotor position estimation.
作者
刘刚
张婧
郑世强
毛琨
LIU Gang;ZHANG Jing;ZHENG Shiqiang;MAO Kun(Science and Technology on Inertial Laboratory,Beihang University,Beijing 100191,China;Ningbo Innovation Research Institute,Beihang University,Ningbo 315800,China)
出处
《电机与控制学报》
EI
CSCD
北大核心
2023年第9期98-108,共11页
Electric Machines and Control
基金
国家自然科学基金(61822302)。
关键词
高速永磁同步电机
无差拍电流预测控制
无传感器控制
多参数在线辨识
总体最小二乘算法
神经元
high speed permanent magnet synchronous motor
deadbeat predictive current control
sensorless control
multi parameter online identification
total least squares algorithm
neuron
