Abstract
Coreless permanent magnet linear synchronous motors (PMLSMs) are renowned for the high overload capacity, lightweight design, and outstanding thrust acceleration. However, the nonuniform magnetic field distribution at the longitudinal ends of the motor and the transverse arc edges of the primary windings give rise to significant end effects. Traditional modeling approaches are often computationally intensive or are limited to considering merely a single type of end effect. This article proposes a generalized modeling methodology capable of simultaneously accounting for both transverse and longitudinal end effects. By employing the equivalent surface current method and the method of images, the approach constructs a spatial coordinate matrix for both the permanent magnets (PMs) and thrust coils, thereby efficiently addressing the equivalent mutual inductance. To further optimize motor performance, the modified nondominated sorting genetic algorithm III (NSGA-III), incorporating adaptive crossover and mutation operators, is utilized to enhance thrust density, reduce ripples, and minimize manufacturing deviations. The efficiency of the proposed method is validated through finite element simulations and prototype experiments, with results showing a close agreement with theoretical predictions.
Funding Information
  • National Natural Science Foundation of China (52425702)
  • National Natural Science Foundation of China (52377068)
  • Doctoral Innovation Fund Program of Southwest Jiaotong University (CX2025YB29)
  • Postdoctoral Fellowship Program and China Postdoctoral Science Foundation (BX20250418)

This publication has 25 references indexed in Scilit: