Determining of the magnetic characteristics of the E-core Transverse Flux Machine based on neural network

The E-core Transverse Flux Machine (ETFM) is combined of the principle of transverse flux machine and conventional Switched Reluctance Machine. The paper is focused about the modelling and the imlementation of the ETFM for an application. The magnetic characteristics of the machine determines its electrical and mechanical behaviors. To analyze and predict the ETFM´s performance, good knowledge of its electromagnetic characteristics is essential. This paper investigates the use of Artificial Neural Networks (ANNs) for the modelling of the magnetic nonlinearity of the ETFM. The proposed ANN based determination of the magnetic characteristics is put into the ETFM model instead of the traditional look-up table.