Performance of support vector regression machines on determining the magnetic characteristics of the E-core transverse flux machine

Abstract

The E-core transverse flux machine (ETFM) has major advantages with its different and unique structure in conventional electrical machines. It is a combination of transverse flux and reluctance principle. In this work, support vector regression machines (SVRMs) are used to obtain the magnetic characteristic parameters of the ETFM for the first time and it is compared with its artificial neural network model. The data for the training and testing of the SVRMs are obtained from experimental measurements. It is proven that SVRMs can conveniently be used in the modeling of the magnetic behaviors of highly nonlinear ETFM with better accuracy and efficiency.