Characterization and Modeling of Soft Magnetic Materials for Improved Estimation of PWM-Induced Iron Loss

Abstract

This article investigates the iron loss properties of soft magnetic materials under conditions of high excitation frequency and varying premagnetized dc-bias fields (i.e., different magnetization states). Using measured iron loss data collected from the customized test equipment, a modified dynamic Jiles–Atherton model has been developed specifically for the purpose of achieving improved pulsewidth modulation (PWM)-induced iron loss estimation. The accuracy and scalability of the proposed prediction model for different conditions are evaluated including the flux ripple amplitude, excitation frequency, dc-bias field, and symmetry of the triangular waveform. Experimental results have confirmed that the proposed model can accurately predict the dynamic hysteresis loop and corresponding iron loss over a wide range of operating conditions. By introducing a companion iron loss model developed for fundamental and low-order harmonic fields, the total iron loss under inverter PWM voltage excitation is explored. The results show that model parameters derived from a limited number of tests can be used to predict the iron loss in a much broader operating range, which makes it a promising tool for PWM-induced iron loss estimation in new machine designs intended for demanding operating conditions.