A Modified Fixed-Point Iteration Algorithm for Magnetic Field Computation With Hysteresis Models

Abstract

The fixed-point iteration method is widely used in electromagnetic field analysis involving hysteresis property due to its strong robustness, but it has the problem of low computational efficiency. In this article, a modified fixed-point iteration algorithm is proposed where the convergence factor is adaptively adjusted for different time steps according to the variation of residuals, instead of being globally static in traditional ways. Numerical analysis for a C-shaped iron core characterized by the inverse vector Preisach model is performed using the traditional method and the proposed method respectively. The efficiency, stability, and applicability of both methods are assessed, in which the proposed approach shows superior performance.