Comparison of Cogging Torque Compensation Methods for a Flux-Switching Permanent Magnet Motor by Harmonic Current Injection and Iterative Learning Control

Due to high air-gap flux density generated by permanent magnets and the special double salient structure, the cogging torque of flux-switching permanent magnet (FSPM) motors is unfavorably considerable. Different algorithms based on harmonic current injection (HCI) and iterative learning control (ILC) have been recently proposed to compensate cogging torque from the perspective of control methods, and have proved their effects on torque ripple suppression. However, a thorough study to investigate the pros and cons of these two methods have not been conducted, which is exactly the aim of this paper. The two cogging torque compensation approaches, namely, HCI and ILC, are implemented on a prototyped three-phase 12/10 FSPM machine. For simplicity of the control structure, model predictive current control is employed to realize the control of the motor drive. The basic theories and implementations of two compensation methods are introduced in detail. Simulated and experimental results show that both schemes can compensate cogging torque and suppress torque ripple effectively. Comparatively, ILC is more general while HCI is slightly superior in torque pulsation reduction.