Forecasting the temperature of a building-integrated photovoltaic panel equipped with phase change material using artificial neural network

This research aims to conduct a numerical investigation into the impact of incorporating phase change material on the thermal management of a building-integrated solar photovoltaic panel. The study compares the outcomes with those of a photovoltaic panel without the phase change material layer, utilizing sunlight intensity and ambient temperature data from the Meteorological Organization of Jiangsu, China on October 15, 2022. The remarkable effectiveness of phase change material in reducing the temperature of the photovoltaic panel is evident, with the system incorporating phase change material showing a temperature 1.28–29.15 °C lower compared to the system without phase change material. This observation results in a 0.10–2.17% increase in solar panel efficiency. Subsequently, artificial neural network and group method of data handling techniques were employed to establish an accurate relationship for predicting the hourly photovoltaic panel temperature in the photovoltaic-phase change material system. The performance evaluation of this model indicated R2, RMSE, and MSE values of 0.97602, 1.483, and 2.22, respectively.