Photovoltaic Panel Temperature Monitoring and Prediction by Raman Distributed Temperature Sensor With Fuzzy Temperature Difference Threshold Method

Solar cells can operate at a lower efficiency after a certain temperature, which is caused by a negative thermal coefficient. Therefore, the temperature prediction of photovoltaic (PV) modules is critical to accurately evaluate the efficiency of photovoltaic devices. We propose and experimentally demonstrate a Fuzzy Temperature Difference Threshold Method (FTDTM) based on Raman Distributed Temperature Sensor (RDTS) system for the detection and prediction of PV module temperature. The FTDTM consists of four steps, i.e., division of the universe, establishment of fuzzy relationships, definition of relationship matrix and calculation of predicted temperature. The experimental results show that the proposed RDTS can detect and accurately predict temperature trends by using the FTDTM. When the system window base of the FTDTM is set to 2, the average absolute error of the predicted temperature is 1.08 °C, and the fluctuation range of prediction error is ±3.7 °C. In addition, the experiment studied some factors affecting temperature distribution characteristics of PV modules, including solar radiation intensity, surface dust and inclination angle. We provide a solution for large-scale PV module temperature detection and early warning through RDTS systems with FTDTM.