Fault Diagnosis of Buck Converter under Variable Operating Conditions Based on Optimal Fractional Wavelet Transform

Abstract

Aiming at the fault of the buck converter caused by the parameter degradation of the electrolytic capacitor, a fault diagnosis method based on optimal fractional wavelet transform is proposed. Firstly, the output voltage of open-loop and closed-loop Buck converter under variable conditions is collected as fault signal. Then, particle swarm optimization is used to optimize the p value of fractional wavelet transform, and the fault signal is transformed into p-order fractional wavelet transform. The wavelet energy is extracted as fault feature to construct fault feature vector. Finally, SVM classifier is used for training and testing to realize fault diagnosis of buck converter. Experimental results show that the fault diagnosis method based on optimal fractional wavelet transform can effectively and accurately diagnose the faults of open-loop and closed-loop buck converters under variable operating conditions.