A Unified Electromagnetic Inverse Problem Algorithm for the Identification of the Magnetic Material Characteristics of Electromagnetic Devices Including Uncertainty Analysis: A Review and Application

Abstract

Magnetic properties of the electromagnetic devices core materials are reconstructed by solving a coupled experimental-numerical electromagnetic inverse problem. However, the measurement noise as well as uncertainties of the forward model parameters and structure may result in dramatic recovery errors in the recovered values of the material parameters. In this paper, we review the use of the electromagnetic inverse problem for the identification of the magnetic material characteristics. The inverse algorithm is combined with a generic stochastic uncertainty analysis for a priori qualitative error estimation, and a quantitative error reduction. The complete inverse methodology is applied onto the identification of the magnetizing B-H curve of the magnetic material of a commercial asynchronous machine. Both numerical and experimental results validate the inverse approach, showing a good capability for magnetic material identification in electromagnetic devices. The proposed technique is general and can be applied into a wide range of applications in electromagnetic community.