Transmission Line Fault Detection Based on CEEMD-AOA-SVM

To accurately extract the characteristics of transmission line short-circuit fault, a fault detection method based on comprehensive ensemble empirical mode decomposition (CEEMD), arithmetic optimization algorithm (AOA) and support vector machine (SVM) is proposed. Taking the three-phase voltage of the line after fault as the feature vector, CEEMD method is used to decompose it, and a series of modal components are obtained. Calculate the multi-scale sample entropy (MSE) of each mode and the MSE of the obtained component is composed of the fault feature set. Because the penalty parameters and kernel function of SVM have a direct impact on the classification accuracy in the classification process, the parameters are determined by arithmetic optimization algorithm (AOA), and the obtained fault feature set is used for fault detection using AOA-SVM model. The simulation results indicate that the fault detection rate of CEEMD-AOA-SVM is 98.28%, which is significantly higher than that of EMD-SVM and CEEMD-SVM.