Sensorless Control of Five-Phase Flux-Intensifying Permanent Magnet Motor Based on Gradient Descent Adaptive Filter SMO

This paper presents a sensorless control strategy of five-phase flux-intensifying interior permanent magnet (FPFT-IPM) motor based on sliding mode observer (SMO). Due to the back-EMF harmonics caused by inverter nonlinearity and motor secondary saliency, the position estimation errors will be increased. Thus, an adaptive filter (AF) based on gradient descent (GD) algorithm is proposed for improving estimated accuracy of rotor position. The feasibility of the algorithm is verified by simulation and experiment.