Calculation of Iron Loss in Electrical Generators Using Finite-Element Analysis

Abstract

The accurate calculation of iron loss from finite-element analysis in electrical machines is essential if optimal machines are to be designed. This paper conducts a holistic review of the extensive literature field before examining, in detail, several methods in order to recommend an optimum engineering solution. Both frequency domain and time domain methods are discussed including the use of different orthogonal components as well as the relative merits of using all, or some, of the eddy current, anomalous, and hysteresis loss components. A theoretical cubic meter of iron is simulated to quickly demonstrate the inaccuracies of Cartesian coordinate methods before calculations on several manufactured machines are undertaken showing the superior accuracies of major/minor loop calculation. Calculation undertaken using the radial tangential orthogonal plane is shown to have less than 1% average difference to that of the major/minor loop yet is over six times quicker. The peak percentage error in an individual element is shown to be less than 5%. Discussions are also made regarding the method of curve fitting to gain loss constants and any possible sources of inaccuracy particularly during manufacture.