Enhanced Hybrid Distributed Generation System Featuring Improved Islanding Detection and Power Quality Monitoring Using Wavelet Transform and Neural Networks

This study presents an innovative analysis of a hybrid distributed generation (DG) system, integrating solar and wind energy resources. The system is analyzed at the Point of Common Coupling (PCC), with a specific focus on addressing challenges related to islanding and power quality (PQ). The research introduces a distinctive method for disturbance detection and categorization, combining the strengths of wavelet transforms (WT) and recurrent neural networks (RNN). This hybrid technique distinguishes disturbances effectively by analyzing WT-derived indices from the PCC's negative sequence voltage. Features extracted from these indices are used to train a recurrent neural network, enabling precise disturbance identification. The proposed method was evaluated against traditional techniques, demonstrating superior accuracy and reliability in disturbance detection. Moreover, when integrated with machine learning classifiers, the method exhibits robust performance. The findings of this research offer a significant contribution to enhancing the operational efficiency and dependability of hybrid DG systems, particularly in managing islanding and PQ issues.