Title :
Neural Network Approach for Modelling Hysteretic Magnetic Materials Under Distorted Excitations
Author :
Fulginei, Francesco Riganti ; Salvini, Alessandro
Author_Institution :
Appl. Electron. Dept., Roma Tre Univ., Rome, Italy
Abstract :
A Neural Network (NN) approach for modelling dynamic hysteresis is presented. The modelling of the dynamic behavior of hysteretic materials and devices must take into account magnetodynamic effects. In the present paper these tasks are simultaneously modelled by means of an ad-hoc Neural System (NS) based on an array of 3-input 1-output Feed Forward NNs. Each NN is dedicated to a particular typology of the excitation field (prediction of flux density from a known waveform of the magnetic field strength or vice-versa) and it manages just a fixed portion of the dynamic hysteresis loop. The whole hysteretic path is reconstructed by the union of the evaluations made by different NNs of the NS. The NS is able to perform the simulation of any kind of dynamic loop (saturated and non-saturated, symmetric or asymmetric) generated by any assigned arbitrarily distorted excitations into a fixed range of frequencies. Numerical validations are presented.
Keywords :
eddy current losses; feedforward neural nets; magnetic hysteresis; magnetic leakage; physics computing; 3-input 1-output feed forward neural network array; arbitrarily distorted excitations; dynamic behavior; dynamic hysteresis loop; excitation field typology; hysteretic devices; hysteretic magnetic materials; hysteretic path; magnetodynamic effects; neural network approach; neural system; Arrays; Artificial neural networks; Magnetic hysteresis; Materials; Mathematical model; Saturation magnetization; Training; Magnetic hysteresis; magnetic losses; magnetodynamic; neural networks;
Journal_Title :
Magnetics, IEEE Transactions on
DOI :
10.1109/TMAG.2011.2176106