Research on residual current classification of low voltage distribution network based on NEVO-SVM

So as to effectively improve the accuracy of residual current classification in low-voltage distribution networks, this paper proposes a classification model based on the support vector machine optimized by the improved energy valley optimization algorithm (NEVO-SVM). Taking the highest accuracy as the criterion, the hyperparameters of SVM are optimized and selected based on the improved energy valley optimization algorithm, and then the SVM is used to classify and predict the types of residual current in the low-voltage distribution network. Experimental results show that the improved energy valley optimization algorithm is used to optimize the SVM hyperparameters, and the obtained NEVO-SVM classification model can accurately distinguish the collected residual current signals This research provides a theoretical basis for the development of new residual current protection devices.