The integration of Direct Current (DC) microgrids marks an important development in energy systems, especially for incorporating renewable energy sources. However, protecting these microgrids presents unique challenges, particularly in detecting and classifying short-circuit faults. This paper presents a consensus-based methodology that employs the Adaptive Neuro-Fuzzy Inference System (ANFIS) to enhance the detection of fault types in direct current (DC) microgrids. The core aim of this methodology is to integrate data from all grid nodes to achieve a consensus, ensuring high precision in fault detection across a wide range of fault resistance values. By extracting features from voltage and current signals, the system efficiently classifies fault types shortly after a fault occurs. The use of consensus techniques significantly enhances fault identification accuracy by combining the assessments of multiple measurement devices, thus reducing the impact of errors from individual devices. The results show that this consensus-driven approach not only improves reliability but also performs well across different fault resistance scenarios, contributing to the overall safety and dependability of DC microgrid operations.

Keywords