Nondestructive Estimation of Magnetization Distribution in Permanent Magnet Using Quasi-Newton Method Based on 2-D Fourier Series Expansion

A permanent magnet (PM) with an accurate magnetization distribution is required for the design of high-quality electrical machines. The nondestructive diagnosis of the magnetization distribution in PM is essential to facilitate the practical design of the synchronous machines. While some methods for magnetization evaluation have been proposed, an optimal method has not been established due to the indefiniteness of the magnetization distribution. Then, the method using nonlinear programming based on the 1-D Fourier series expansion has been proposed by the authors. In this article, the method based on the 2-D Fourier expansion is proposed. The performance of the proposed method was demonstrated in PM, in which orientations were set to parallel and polar isotropic. The performance of the proposed method is compared with the singular value decomposition method.