An Online Fault Detection and Classification Monitoring scheme for Photovoltaic Plants

The majority of the recent trends in photovoltaic (PV) energy utilization can be attributed to major global legislation intended to reduce the use of fossil fuels. However, the performance of these solar PV system gets affects by various faults that must be identified. In this regard, an effective and highly accurate solar PV fault detection method is proposed wherein Artificial Neural network (ANN) and Honey Badger Algorithm (HBA) have been used. The main motive of proposed HBA-ANN model is to enhance the accuracy of PV fault detection while lowering the complexity of model. We used a PV fault dataset from GitHub, which was later balanced and impartial, to achieve this goal. Also, during the pre-processing stage, the input and target variables are isolated. The next stage, in which the ANN is initialized and weights are determined. An HBA optimization procedure is then used to optimize or tune the value of these weights. Furthermore, by contrasting the suggested HBA-ANN model's performance with that of more established models like the Tree, K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Artificial Neural Network, the model's effectiveness is evaluated and validated. The simulated results were obtained for both phases, i.e. the training phase as well as the testing phase in terms of accuracy, precision, recall, and Fscore. The results of the simulations showed that the suggested HBA-ANN model outperformed all other comparable models in terms of every factor, demonstrating its superiority.