Abstract
Energy loss in shielding soft magnetic materials at low frequencies (1-100 Hz) can cause fluctuations in the material's magnetic field, and the resulting magnetic noise can interfere with the measurement accuracy and basic precision physics of biomagnetic signals. This places higher demands on the credibility and accuracy of loss separation predictions. The current statistical loss theory (STL) method tends to ignore the high impact of the excitation dependence of quasi-static loss in the low-frequency band on the prediction accuracy. STL simultaneously fits and predicts multiple unknown quantities, causing its results to occasionally fall into the value boundary, and the credibility is low in the low-frequency band and with less data. This paper proposes a progressive loss decomposition (PLD) method. Through multi-step progressive predictions, the hysteresis loss simulation coefficients are first determined. The experimental data of the test ring verifies the credibility of PLD's prediction of the two hysteresis coefficients, improving the inapplicability of the STL method. In addition, we use the proposed method to obtain the prediction results of the low-frequency characteristics of the loss of a variety of typical soft magnetic materials, providing a reference for analyzing the loss characteristics of materials.