An Improved Core Loss Prediction Model Using Interval-Based Nonlinear Fitting for High-Frequency Transformers

Accurate prediction of core losses in soft magnetic materials under high-frequency and high-flux-density conditions remains a significant challenge. This paper addresses this challenge by introducing an improved core loss separation model (INF -Bertotti). The core innovation of this research lies in the development of an interval-based nonlinear fitting (INF) method. This method enables the dynamic prediction of variable coefficients within the Bertotti loss separation framework, effectively capturing the complex nonlinear behavior of loss components that conventional constant-coefficient models fail to characterize. Comparative evaluations demonstrate that the proposed INF-Bertotti model achieves superior prediction accuracy. Furthermore, the practical impact of the model is validated through its successful application in the design optimization of a high-frequency transformer (HFT), highlighting its potential as a reliable and versatile solution for core loss prediction in high-frequency power applications.