Fault detection and classification in DC microgrid clusters

Abstract

With the rising popularity of DC microgrids, clusters of such grids are beginning to emerge as a practical and economical option. Short circuit problems in a DC microgrid clusters can cause overcurrent damage to power electronic devices. Protecting DC lines from large fault currents is essential. This paper presents a novel localized fault detection and classification technique for the protection of DC microgrid clusters. In this paper, a variational mode decomposition (VMD) and artificial neural network (ANN) based technique is proposed for accurate and effective fault detection and classification. This research aims to train an ANN that can detect and classify faults in DC microgrid clusters with multiple sources and loads by applying VMD to extract features of current signals. Different types of short circuit faults such as Pole to Pole and Pole to ground faults are considered under various grid operating conditions. The proposed method is capable of real-time fault detection and diagnosis, which can help prevent system failures and minimize downtime. The results indicate that the proposed approach is efficient and effective in detecting/classifying faults in DC microgrid clusters improving the reliability and system safety. The performance evaluation is carried out through rigorous case studies in MATLAB/Simulink environment to prove the efficacy of the proposed method. The VMD-ANN approach is shown to outperform other traditional signal processing techniques in terms of accuracy and robustness. Moreover, the proposed method is applicable to a wide range of DC microgrid clusters, making it a versatile and valuable tool for future research and development.