Background: Permanent Magnet Synchronous Generators (PMSGs) face design challenges due to multi-physics coupling and conflicting optimization goals.
Objective: Develop a multi-objective optimization model for PMSGs to enhance efficiency, minimize torque ripple, and ensure structural feasibility.
Methods: We derived key equations for slot filling ratio, generator length, efficiency, and THD in the Optimization Model. Optimization variables were reduced via geometric relationships and empirical equations. A hybrid workflow combining 2D-FEA and evolutionary algorithms (EA) identified Pareto-optimal solutions.
Results: Compared with the initial design, the optimized PMSG prototype has significantly improved in terms of maximum stator partial flux density, torque ripple coefficient, total harmonic distortion(THD), power generation efficiency, slot fill factor, and temperature rise. The bench test verified the accuracy of the optimized design, and the voltage error was less than 3.3% compared with the simulation results.
Conclusions: This study proposes a systematic methodology to balance technical performance and engineering constraints, offering a scalable solution for PMSG design in energy systems.