IRON LOSSES PREDICTION USING NEURAL NETWORK BASED MAGNETIC MODEL

Marinel Temneanu, Traian Balan, Dana Balan, Octavian Postolache

Abstract

The possibility to predict the iron losses under nonsinusoidal waveform of magnetic flux density using neural network based magnetic models is analyzed in this paper. Two different types of model structures (predictors with and without feedback) have been used. The predictions are compared to the measured output for different magnetic materials and induction waveforms.

Keywords

iron losses, neural network model

Download

IMEKO-TC4-2001-112.pdf

DOI

-

IMEKO TC

TC4 - Measurement of Electrical Quantities

Back to the proceedings