基于卡尔曼滤波滑模变结构转子位置观测器的PMSM无差拍控制

Deadbeat predictive current control based on a sliding mode observer with Kalman filter for PMSM speed and rotor position

针对永磁同步电机(表贴式永磁同步电机)无转速传感器矢量控制系统,设计了新的滑模观测器进行转子位置和转速的观测。在通常滑模电流观测器基础上,采用sigmoid函数代替常数切换函数在一定程度上减小了抖振,采用扩展卡尔曼滤波器代替常用低通滤波器,用锁相环技术完成了转速的提取。用李亚普诺夫稳定性定理证明了设计的滑模变结构观测器的渐进稳定性。提出在同步旋转轴系下的一种基于无差拍算法的永磁同步电机离散化电流预测控制方法以提高电机电流环的性能。通过实验验证了设计的滑模观测器和无差拍控制器有效性和可行性。

A sliding mode observer is presented for sensorless control of permanent magnet synchronous machines （ surface PMSM） . Compared to conventional sliding mode observers, sigmoid function was used to weaken the chattering problem,and Kalman filter was applied to substitute for conventional low-pass fil-ter. Phase-locked loop can improve the precision of the rotor speed estimation. The asymptotically stabili-ty was got by Lyapunov stability theory. A discrete predictive current control scheme for PMSM based on the dead-beat algorithm in the synchronous frame was presented to improve the performance of the current loop. Experimental results verify its feasibility and effectiveness.