Faults classification In A Microgrid Using Decision Tree Technique And Support Vector Machine

Abstract

Introduction of distributed generators into the conventional power grid results, complexity in operation and control problem along with creating a challenge in identification of fault disturbances in electric power system. This paper describes a new technique for classification of fault disturbances like LLG under different operating conditions. The pattern recognition techniques namely support vector machines (SVM) and decision tree (DT) are used to classify faults disturbances. Based on the study of this paper, it is observed that SVM and DT provides the best possible accuracy as compared to other techniques, implying its robustness under different operating scenarios such as variation in load, solar insolation and presence of noise and harmonics in the system parameters.