Arc Fault Detection Based on Feature Extraction by Wavelet and Empirical Mode Decomposition

In order to achieve the improvement in the detection accuracy of electrical arc faults, a method combining window division with Empirical Mode Decomposition (EMD) and Wavelet Decomposition is proposed. The above method enables the extraction of current features in multiple dimensions. The arc fault is identified with machine learning classification models. Wavelet transform and EMD methods are firstly applied to decompose the current signals and calculate the fault features in various dimensions. The collected feature information is used as input to a classification algorithm for arc fault diagnosis. Experimental results show that the accuracy of the arc fault detection of the feature extraction method based on the Gradient Boosting Decision Tree (GBDT) is as high as 98%. This method effectively captures features of the arc fault and achieves high efficiency and accuracy in the detection of arc faults.