Structural parameters optimization of permanent magnet spherical motor based on BP neural network model

In order to obtain a larger torque for a permanent magnet spherical motor, the method of optimizing the structural parameters of the permanent magnet spherical motor is studied. Based on the premise of volume minimization, this paper proposes a set of nonlinear data fitting by BP neural network, using genetic algorithm and particle swarm algorithm to calculate the maximum torque and the corresponding structural parameters. According to the sample data of the structural parameters of spherical motor and torque, BP neural network is trained to fit the sample space, and then parameters of the optimal algorithm are found in combination with the BP neural network. Due to its accuracy and feasibility, the finite element analysis is used to verify the optimization results. Finally the particle swarm algorithm determines structure parameters optimization for permanent magnet spherical motor.