Research Projects

Grid-connected PV systems are gaining more and more attention as viable energy sources, so having a solar inverter that is both reliable and stable is more important than ever. A fault in an inverter can significantly impact the whole system, potentially jeopardizing the grid's safety. Therefore, developing a Fault Diagnostic Mechanism that can accurately detect and categorize failure situations is crucial. Therefore, a Fault Diagnostic Mechanism is required to detect and categorize failure situations. The proposed work presents a comprehensive fault detection and classification method for grid-connected single-phase PV inverters. The proposed approach employs four machine learning algorithms: Logistic Regression, k-Nearest Neighbors (kNN), Decision Trees, and Random Forest. The above suggested algorithms meets the need for a reliable diagnostic predictor for PV inverters linked to the grid, making green energy systems more stable and reliable.