Influences of iron loss coefficients estimation on the prediction of iron losses for variable speed motors

• Published in: 2015 IEEE International Electric Machines & Drives Conference (IEMDC) pp. 1254 - 1259
• Main Authors: Juergens, Jonathan, Ponick, Bernd, Winter, Oliver, Fricasse, Antonio
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2015
• Subjects: Bertotti equation; curve fitting; eddy current; excess losses; Frequency estimation; hysteresis; Iron; iron losses; Jordan equation; least square method; loss coefficient; Loss measurement; Magnetic hysteresis; Mathematical model; Steel
• DOI: 10.1109/IEMDC.2015.7409222

This paper addresses the influence of the estimation of hysteresis, eddy current and excess loss coefficients on the prediction of the overall iron losses in traction motors. Two standard iron loss models (according to Bertotti and Jordan) are presented and it is examined how the determination of the corresponding loss coefficients is effecting the iron loss calculation for variable speeds. It is shown that the accuracy of the iron loss calculation does not mainly depend on the number of iron loss terms in the model but rather depends on the computation of the corresponding coefficients. To reduce the uncertainty of the iron loss calculation, common curve fitting methods have to be used carefully. Provided that constant coefficients and exponents of the loss component terms are used, the resulting coefficients estimation by means of curve fitting methods depends on the considered frequency range of the measured losses given by the electrical steel manufacturer. Three different FEM analysis tools are used to compare the iron loss calculations for a simplified electrical machine model and a permanent magnet assisted synchronous reluctance machine (PMaSyRM) model.