Hybrid Topology Optimisation of Permanent Magnet Motors Using Boolean Geometry Projection and Gaussian‐Based ON‐OFF Method

ABSTRACT This paper proposes a topology optimisation method for permanent magnet motors that hybridises the Boolean geometry projection (BGP) method and the ON‐OFF method based on Gaussian basis functions. This method represents the topology of a permanent magnet based on the BGP method in addition to generating that of a magnetic core, including flux barriers, using the Gaussian basis functions to obtain a novel but manufacturable machine structure. In contrast to existing optimisation methods such as parameter optimisation, the pre‐setting of the search region for geometric parameters is not required. This leads to the design of machine structures that is completely independent of the designer's experience and knowledge. Optimisation results demonstrate that the proposed method generates diverse and realistic rotor structures, such as U‐, V‐ and double V‐shaped configurations, and achieves higher average torque performance compared with existing parametric and hybrid optimisation methods. The findings confirm that the proposed BGP‐based hybrid framework provides a practical and effective design methodology for PM motors, and it can be further extended to a full motor design that simultaneously considers both mechanical and electromagnetic properties in future work.