15 November 2024 in IEEE Transactions on Magnetics
A segmented asymmetric V-type permanent magnet structure is proposed to address the issues of high output torque ripple and large cogging torque in traditional permanent magnet synchronous motors. Firstly, the relationship between the pole-span angles of the segmented asymmetric V-type permanent magnet structure and the output torque, torque ripple, and cogging torque is investigated. Then, the rationality of the segmented asymmetric V-type permanent magnet structure and the correctness of the theoretical derivation are verified by finite element simulation. Subsequently, with the optimization objectives of increasing the average output torque of the motor, reducing the torque ripple, and decreasing the cogging torque, the optimal pole-span angle combination of the motor is determined by using the Pareto frontier distribution and the weight analysis method. The effectiveness of the optimized design is finally verified by finite element simulation and prototype tests. The results reveal that the segmented asymmetric V-type permanent magnet structure can ensure ample output torque while having low cogging torque and torque ripple in output characteristics.
20 November 2023 in IEEE Transactions on Energy Conversion
This Convex-pole series-connected hybrid excitation generator has the advantages of small size and high power density, which is the main development trend of hybrid excitation generator for automobiles. The design is based on a hybrid excitation generator with built-in combined pole permanent magnet and convex pole electromagnet. The air gap magnetic density expression based on numerical analysis is derived, and a multi-objective optimization model with the rotor eccentricity and V-shaped permanent magnet clamping angle as design variables and the slot torque amplitude, no-load induced electric potential fundamental waveform amplitude and no-load induced electric potential waveform distortion rate as optimization objectives is established. The target optimization model is sampled using the Latin hypercube sample statistics, and the optimization results are screened by a combination of Pareto frontier statistics and defined parameter matching coefficients to determine the optimal rotor structure parameter combinations. Based on the optimized motor structure parameters, a prototype is fabricated and tested to verify the performance of the prototype and the correctness of the above theoretical analysis.