Parameter identification based on linear model for buck converters

Parameter identification is an effective method to monitor the health condition of multiple components in power electronic converters. The hybrid model-based method has been proved to be a universal approach. However, the existing hybrid model of buck converter ignores the winding resistance of filter inductor and non-ideal factors, which will cause errors in the identification results. To solve these problems, leg midpoint voltage is used to replace the input voltage and switching signals, and a linear model is proposed, which can be implemented by fewer sensors. In order to verify the superiority of the linear model, this paper proposes a simple and practical parameter identification method for passive components. Compared with the hybrid model-based method, the proposed method does not need a sensor to measure driving signals and can obtain higher accuracy under non-ideal situations. Simulation and experimental results prove the proposed method to be effective and accurate.