A Radial Basis Function Neural Network-Based Fault Detection Scheme for a Microgrid System

A radial basis function neural network (RBFNN) based fault detection and classification scheme of a Microgrid (MG) system is presented in this paper. The proposed scheme processes eight inputs, including three-phase current and voltage, zero-sequence voltage, and current for the identification of faults in the Microgrid. The considered RBFNN is trained under various conditions such as line length, fault types, fault resistance, and ground resistance and provides four binary outputs, indicating different types of faults in the system with zero representing the absence and one denoting the presence of fault in each phase. MATLAB 2024a is used to train the RBFNN, distinguishing the symmetrical (LLL, LLLG) and unsymmetrical (LL, LLG, LG) faults in the MG system. The effectiveness of the proposed scheme is evaluated by comparing with conventional algorithms in the detection and classification of various types of faults. The result demonstrates the reliability and superiority of the proposed fault detection scheme.