Photovoltaic Hot-Spots Fault Detection Algorithm using Fuzzy Systems

Faults in photovoltaic (PV) modules, which might result in energy loss and reliability problems are often difficult to avoid, and certainty need to be detected. One of the major reliability problems affecting PV modules is hot-spotting, where a cell or group of cells heats up significantly compared to adjacent solar cells, hence decreasing the optimum power generated. In this article, we propose a fault detection of PV hot-spots based on the analysis of 2580 PV modules affected by different types of hot-spots, where these PV modules are operated under various environmental conditions, distributed across the U.K. The fault detection model comprises a fuzzy inference system (FIS) using Mamdani-type fuzzy controller including three input parameters, namely, percentage of power loss (PPL), short circuit current ( \text{I}_{\mathrm{ sc}} ), and open circuit voltage ( \text{V}_{\mathrm{ oc}} ). In order to test the effectiveness of the proposed algorithm, extensive simulation and experimental-based tests have been carried out; while the average obtained accuracy is equal to 96.7%.