Fault detection and classification in hybrid energy-based multi-area grid-connected microgrid clusters using discrete wavelet transform with deep neural networks

Abstract

Microgrid control and operation depend on fault detection and classification because it allows quick fault separation and recovery. Due to their reliance on sizable fault currents, classic fault detection techniques are no longer suitable for microgrids that employ inverter-interfaced distributed generation. Nowadays, deep learning algorithms are essential for ensuring the reliable, safe, and efficient operation of these complex energy systems. They enable quick responses to faults, reduce downtime, enhance energy efficiency, and contribute to the overall sustainability and resilience of microgrids. With this intent, this work proposes a “Discrete Wavelet Transform with Deep Neural Network (DWT-DNN)” for detecting and classifying the various faults that occurred in hybrid energy-based multi-area grid-connected microgrid clusters. The proposed DWT-DNN first extracts the input features from the point of common coupling of the cluster system using DWT, and then, these decomposed features are applied as input variables to train the DNN for the detection and classification of various faults. All the investigations are performed in the “MATLAB/Simulink 2022a” environment. To validate the effectiveness of the proposed DWT-DNN, the results are compared with wavelet packet transforms (WPT) in terms of accuracy in detecting and classifying the faults. From the simulation findings and observations, it is evident that the proposed DNN produced fruitful results.