面向永磁同步电机无传感器控制的二阶改进扩展状态观测器

A Second-Order Improved Extended States Observer for Permanent Magnet Synchronous Motor Sensorless Control

针对传统扩展状态观测器(extended states observer,ESO)为保证收敛性和提升系统动态性能而增大矫正项增益,进而加剧对于高频噪声敏感性的问题,提出了一种面向永磁同步电机(permanent magnet synchronous motor,PMSM)无位置传感器控制的改进扩展状态观测器(improved ESO,IESO)。结合PMSM扩展反电动势全阶模型,将扩展反电动势作为ESO扩展的状态;通过将低阶动态方程的输出作为高阶动态方程的输入的方法设计IESO,从而有效降低矫正项增益,进而减小系统对高频噪声的敏感性;在矫正项中引入饱和函数以限制高增益观测器可能引入的峰值效应,并进行了收敛性分析。数值仿真结果表明:相较于传统线性ESO,采用IESO后,PMSM无传感器控制系统在电机启动、调速及突加负载时动态性能均有提升,误差最大值减小,同时收敛速度加快约25%;在测量电流加入0.5 A随机高频噪声后,传统方法收敛时间增加了约25%,转速误差最大值增大了约40%,而IESO并没有显著增加,同时相较于传统方法,IESO的角度误差高频噪声减小了约60%,因而IESO能够显著抑制系统高频噪声。

In view of the problem that the conventional extended states observer(ESO)increases the gain of the correction term in order to ensure convergence and improve the dynamic performance of the system,and thus aggravates the sensitivity to high-frequency noise,an improved ESO(IESO)for sensorless control of permanent magnet synchronous motor(PMSM)is proposed in this paper.First,combined with PMSM extended back electromotive force model,extended back electromotive force is regarded as the extended state of ESO.Then,IESO is designed by using the output of the low-order dynamic equation as the input of the high-order dynamic function to effectively reduce the gain of the correction term,and thus reduce the sensitivity of the system to high-frequency noise.Then,the saturation function is introduced into the correction term to limit the peaking phenomenon that may be introduced by the high gain observer,and the convergence analysis is carried out.The numerical simulation results reveal that,compared with the conventional linear ESO,through the IESO proposed in this paper,the dynamic performance of the PMSM sensorless control system is improved when the motor is started,the speed is adjusted and the load is suddenly applied.The maximum value of error is reduced and the convergence speed is accelerated by about 25%.Besides,when 0.5 A random high-frequency noise is added to the measured current,the convergence time of the traditional method increases by about 25%,and the maximum speed error increases by about 40%,while those of IESO do not increase significantly.Compared with traditional methods,IESO’s angle error high-frequency noise is reduced by about 60%,so IESO can significantly suppress system high-frequency noise.