LAUSR

Torque ripple modeling and minimization for interior permanent magnet synchronous machines (IPMSMs) requires accurate information of the inductances, which vary nonlinearly due to magnetic saturation. However, existing approaches fail to consider the magnetic saturation and, thus, their performance are limited under different load conditions. Therefore, this paper improves the torque ripple model by considering magnetic saturation, and employs this model for the optimal current design to improve the performance of torque ripple minimization for IPMSMs under different load conditions. At first, numerical studies are performed to analyze and understand how magnetic saturation affects the torque ripples in IPMSMs. Then, a novel torque ripple model for IPMSMs is developed, in which the inductance term is replaced by exploring the machine electrical model. This improved torque ripple model is computationally efficient and it can provide fast and accurate torque ripple prediction. Based on this model, a genetic algorithm (GA)-based optimal stator current design approach is proposed to minimize the torque ripple in IPMSMs. The proposed GA-based approach can adaptively optimize the stator current under different load conditions, which can guarantee the robust performance of torque ripple minimization under different saturation levels. The proposed approach is validated through an experimental test on a laboratory IPMSM drive system.