Research on Structural Parameters Optimization of Reluctance Resolver Based on Multi-objective Particle Swarm Optimization

Variable reluctance (VR) resolvers are mainly used in the servo control system to obtain the rotor position and speed feedback information of the motor. In order to further improve the accuracy, this paper takes the reluctance resolver as the research object and utilizes Ansoft to establish its finite element (FE) analysis model. The optimized parameters are the slot-opening width of stator, the rotor sine coefficient, and the minimum air-gap length. The total harmonic distortion and zero-position voltages of the output signals are selected as objective functions. The mathematical model based on support vector regression machine is established by the simulation samples. The improved multi-objective particle swarm optimization algorithm is applied to optimize the model to get the optimal parameters. The final optimal design is selected from the Pareto front using the TOPSIS method. The simulation of the design and experimental results verify the feasibility of the method.