Spherical Harmonic Identification Using a Priori Information About an Electrical Device

This paper deals with a method using the spherical harmonic expansion to identify a magnetic equivalent model to a given electrical device. The most significant issue of this identification is the number of required magnetic sensors to obtain a valid model. Therefore, our method is based on a probabilistic approach to solve inverse problems using prior information of the source like periodicities or symmetries. Thus, the required number of sensors has strongly decreased. The method is illustrated by a finite-element method simulated permanent magnet motor