A model for h24 simulation of photovoltaic solar panels temperatures at any environmental condition

Abstract

The paper proposes a model enabling the prediction of the temperatures of photovoltaic solar panels PV by means of the simulation of their thermal behaviour extended to the entire 24 h of a day. To study the continuous interaction between a PV module and its surrounding environment, the analysis was therefore not limited to the daytime of power generation as done by current models, but was extended to any operating setting under widely varying climatic conditions.

To estimate the junction, frontsheet and backsheet temperatures, an original reformulation of the relationships for the convective heat exchange between a PV panel and the surroundings is also proposed.

To assess the model, a dataset from the experimental station of the University of Rome’ Tor Vergata’ was used, extended up to 14 months of continuous operation of a commercial PV module with a time step of 1 min. The model’s adherence to the available data was assessed both on single days, and globally, demonstrating a higher accuracy (with approximately RMSE 1.5 °C and MAE 1 °C) than models available in the literature. The possibility of working on a net size of 510,701 observations also made it possible to highlight and discuss in detail incidental criticalities resulting from the effect of rain, of moisture, and from the complexity of convection when local turbulences are generated by the wind. The results show the proposed model, requiring very few weather input parameters, reliable and robust enough to fulfill the intended purposes.