Implementation of Two-Dimensional Finite Element Analysis Method in Axial Flux Permanent Magnet Motor

Axial flux permanent magnet (AFPM) motor has been employed increasingly because of its excellent performance in high efficiency, structural compactness, and so on. Three-dimensional (3-D) finite element analysis (FEA) with large time consumption is necessary for AFPM motor accurate calculation for its particular space magnetic flux path, unfavorable to motor design in the previous stage. Alternatively, two-dimensional (2-D) FEA equivalent method is generally used for designing AFPM motor to save time; a new 2-D FEA approximately equivalent method combined with the Romberg integral formula is proposed in this paper aiming at procuring excellent effectiveness. Furthermore, with the assistance of the method, harmonic content of no-load back electromotive force (back-EMF) of AFPM motor is weaken by optimizing the design of permanent magnet. optimization method of multi-objective genetic algorithm (GA) combined with back propagation (BP) neural network and treatment way of skewing permanent magnet are adopted to realize the optimum design jointly. Finally, simulation analysis verifies that the 2-D FEA equivalent and optimization design methods are effect.