Modeling magnetic core loss for sinusoidal waveforms

Abstract

Among the challenging unsolved technical problems that have plagued the minds of scientist and engineers throughout the 20th and 21st century is the development of a quantifiable model to accurately estimate or explain Core Power Losses (CPL). Theoretical advances in magnets led to many model proposals, but as these models were experimentally examined, they quickly lost their validation. Many of the current models use manufacturer's material estimates to form limited curve fitted equations. These equations are only valid for a specific waveform over a specified range. Unless the designers use the same conditions used to determine the manufacturer's fitted equations, the models quickly lose their precision. The scope of this thesis is to explain and compare several of the current models and evaluate them using experimental data. The validity of some of the term components used in many of these models will also be investigated.