Parameter Estimation of DC-DC Converters for Failure Detection Based on Linear Approximation Model

In recent years, failure detection method for DC-DC converters has been required since it is necessary for the stable and reliable operation in the field of power electronics. Due to the fact that DC-DC converters are consists of several electric components such as capacitor and inductor, it is effective that parameter fluctuation of those components caused from component deterioration can be determined. Therefore, a parameter estimation with observed data of the converter is one of the effective methods for the failure detection. In this study, we propose a parameter estimation method for failure detection of the DC-DC converter based on a discretized state space model that expresses dynamic characteristics of the converter. The proposed method estimates the parameter values such as capacitor, and inductor from the sampling data in start-up state in the converter operation. The proposed method is effective since it does not require any special signal or operation to estimate the parameters. To verify the proposed method, we perform a simulation study using MATLAB/Simulink and the results show that the results are useful for the purpose of failure detection.