Electric Motor Topology Optimization via Rotated Filter Projection and Adjoint Sensitivities

We present a unified projection-based framework for topology optimization of electric motor that concurrently handles rotor motion and density filtering on a static finite element mesh. The approach employs a projection operator to map the rotor design onto rotated configurations of the background mesh, thereby eliminating the need for mesh warping, remeshing, or sliding interfaces typically required for simulating rotor motion. The projection further ensures consistent density filtering across all rotation angles. The method’s details and implementation are discussed. As a concrete example, we apply this method to a representative synchronous reluctance motor (SynRM) rotor topology optimization problem. A multi-stage optimization strategy is used to handle the nonlinear constraints, resulting in an efficient high-torque rotor design with maximum torque deviation within 1% of the average torque. The results highlight the effectiveness and computational scalability of the proposed framework for high-fidelity electric motor topology optimization.