End Teeth Topology Optimization of PMLSM Using Normalized Gaussian Network

This paper proposes a topology optimization method using a normalized Gaussian network (NGnet) for the end teeth of the permanent magnet linear synchronous motor (PMLSM), which optimization objectives are maximum average thrust force and minimum thrust ripple. With the proposed method, the optimal end teeth with irregular shape are obtained, and then the irregular boundaries are smoothed to obtain final optimal topology in order to easy manufacture. Compared with initial structure, its average thrust force is increased by 1.27% and thrust ripple is reduced by 94.41%. The results show that the proposed optimization method is effective.