A Novel Fault Detection Technique for Single Open Circuit in a Packed E-Cell Inverter

Recently, fault detection has played an essential role in ensuring the safety and reliability of inverter operation. This paper illustrates the efficiency of employing the random forest decision tree approach for fault detection of single open circuit switch fault in a Packed E-Cell inverter. First, Discrete Wavelet Analysis is utilized to analyze the output voltage, using the first level of detailed coefficients. Then, three statistical features are extracted from this analysis to serve as input to the classifier during training. The output of the classifier, which indicates the fault status, is then trained accordingly. Finally, a fault detection strategy is implemented, and the results obtained using MATLAB/Simulink affirm the effectiveness of the proposed algorithm highlighting its high accuracy and robustness.