Design Optimization of a MG-PMSM based on Torque Balance and Trust Region Boundary Surface

The magnetic-geared permanent-magnet synchronous machine (MG-PMSM) has attracted substantial attention owing to its contactless speed modulation capability, three-port operation, and potential to replace mechanical gearboxes. Nevertheless, particular issues arise during the optimization of MG-PMSM. The torque balance of the inner rotor between the magnetic gear (MG) and permanent-magnet synchronous machine (PMSM) must be strictly maintained, forming a partially discrete set of combinations that satisfy the torque balance condition. Secondly, the parameter boundaries by torque balance and efficiency constraints invalidate certain sections of the parameter optimization range. The partially discrete parameters, partially invalid parameter ranges, complex structure, and nonlinear parameter coupling of MG-PMSMs pose significant challenges to their optimization. Therefore, this paper proposes an optimized design method for MG-PMSM based on torque balance and trust-region boundary surface. First, an accurate subdomain model (ASM) is established to characterize the magnetic field distribution within the MG-PMSM rigorously. Subsequently, dimensionality reduction conditioned on torque balance is performed. Thereafter, a parametric trust region boundary surface is established with efficiency and torque balance as the constraints. Torque density optimization is performed based on trust-region and combined with the multi-island genetic algorithm (MIGA). Finally, the accuracy of the proposed optimization method is validated through finite element analysis (FEA) and prototype testing.

© 2026 IEEE. All rights reserved, including rights for text and data mining and training of artificial intelligence and similar technologies. Personal use is permitted, but republication/redistribution requires IEEE permission.