Electrothermal Modeling Based Digital Twin Method for Degradation Parameters Identification of DC-DC Converter

Abstract

For most digital twin (DT) method, it is challenge to achieve high accuracy parameters identification due to the digital model is not a perfect replica of physical model. In this paper, a novel DT approach based on electrothermal model, which features more realistic than ideal model, is firstly proposed to obtain accurate parameters identification. By calculating the power loss of the proposed digital model, the temperature of self-heating devices in the digital model, such as MOSFETs, diodes and capacitors, can be obtained, so that the values of the temperature-dependent parameters in these devices can be updated, and hence more accurate results can be guaranteed. The proposed method is validated on a 500W buck converter and the experimental results show that the maximum estimated error of the on-state resistances of MOSFET is 0.6%, which is hundreds of times higher accuracy than conventional DT methods.