This paper presents a novel switched-flux memory motor (SFMM) by artfully incorporating the flux-mnemonic concept into the conventional switched-flux permanent magnet machine. The magnetic susceptibility of AlNiCo PM provides the flexible online controllability of air-gap flux by imposing a transient current pulse. To uniformize magnetization levels of PMs, a time-divisional magnetization strategy (TDMS) is proposed. Due to the uniqueness of hysteresis nonlinearity and instability regarding AlNiCo PM operating point, the time-stepping finite element method (TSFEM) dynamically coupled with a nonlinearity-involved parallelogram hysteresis model (NIPHM) of AlNiCo PM is performed to investigate the electromagnetic performance of the proposed SFMM. The results derived from the combinative algorithm verifies the flux-adjustable capability of the proposed motor equipped with TDMS and the validity of the proposed NIPHM.

This paper investigates the losses of a high-speed permanent magnet motor. The iron losses are calculated by a model that can consider the skin effect and rotational loss. The rotor eddy current losses are estimated by a fast hybrid method that can consider the end effect. Pulse-width modulation (PWM) harmonics brought by the voltage source inverter (VSI) are considered in the loss calculations. Then the temperature distribution of the motor is evaluated by using the calculated loss results and computational fluid dynamic (CFD) modeling. Finally, based on the CFD results, the motor structure is optimized to achieve better rotor cooling. The outer slots are closed to force the cooling air flow through the rotor surface. Calculated temperature distributions and optimization results are verified by measurements.

This paper investigates the electromagnetic and thermal performances of the open-circuit air cooled high-speed permanent magnet motor (HSPMM) with Gramme ring windings. Take a 36 kr/min, 75 kW HSPMM as an example, the electromagnetic losses, in particular the rotor eddy current loss and the stator iron loss of the motor are calculated. The influences of switching frequency and air gap length on these losses are studied. Then, a computational fluid dynamics (CFD) model is composed to calculate its coolant flow parameters and the temperature distribution. Based on the CFD results, a lumped-parameter thermal network (LPTN) is created and verified. Then, a 24 kr/min, 140 kW prototype with the same structure and cooling configuration is designed using the LPTN. Experiments show that the calculated results are with reasonable accuracy. Finally, a sensitivity analysis is applied to the 140 kW prototype based on the LPTN to disclose the influences of different parameters on the rotor permanent magnet temperature rise.

This paper presents a detailed magnetic equivalent circuit (MEC) to improve the analysis accuracy and reduce computation time in modeling of axial flux permanent magnet machines. In the proposed MEC model, the magnetic saturation, armature reaction, leakage flux, and rotor rotation are considered to calculate the machine's static characteristics. The investigated machine structure and feature are first introduced. Then, the air-gap flux density distribution, back electromotive force waveform, and the average torque are calculated with the proposed method. Finally, the accuracy of the proposed method is verified by the 3-D finite-element method, and good agreements are achieved.