Electric motors are the key elements in electric propulsion systems. The performance of Electric vehicles (EVs) significantly depends on the electric motors. Permanent magnet synchronous machines (PMSMs) with rare-earth magnets are widely used in EV applications because they fulfill most requirements of EV motors. However, low efficiency at high speed, limited resources and fluctuating prices of rare-earth permanent magnets (PMs) have forced industries to develop alternatives to rare-earth machine technologies. Recently, Variable-Flux PMSMs (VF-PMSMs) also known as memory motors have been introduced to overcome the drawbacks of PMSMs. This thesis focuses on the modeling, analysis and control of the Aluminum-Nickel-Cobalt (AlNiCo) magnet-based VF-PMSMs.
This thesis presents the effect of different magnetization pulse widths and methods on the magnetization level, back-EMF and no-load losses of the VF-PMSM. The injection of the magnetization or de-magnetization current pulse will change the magnet flux linkage and back-EMF harmonics. An adaptive nonlinear filter is used to estimate the back-EMF during the motoring mode. The harmonics present in the machine back-EMF due to different magnetization and de-magnetization current pulse widths and magnetization methods are analyzed. Besides, the quality of the back-EMF for different speeds and machine no-load losses are presented for different magnetization states (MSs).