Dynamic prediction of a building integrated photovoltaic system thermal behaviour

Abstract A dynamic numerical thermal model has been developed for rooftop building integrated photovoltaic systems, considering a fully or partially integrated configuration, their integration structure and an insulated air gap at the underside. The two-dimensional mathematical model was validated using a test bench representing a residential partially integrated photovoltaic system. The accuracy of the model was studied by deriving the equivalent thermal resistance (or Ross coefficient). Values obtained with the developed model were compared to a nominal operating cell temperature thermal model based on manufacturer datasheet, and the measured data. The results were indicative of a well ventilated air gap and an appropriate choice of Nusselt number. The model was additionally tested for a fully integrated photovoltaic system to demonstrate its utility for different integration architectures. The mean absolute error of the model was evaluated to 2.71 °C for module temperature. It could, therefore, be useful for design studies requiring the prediction of thermal behaviour, as may become important for future regulations and business models such as self-consumption. Future work will consider facade photovoltaic systems, shading elements and coupling to an electrical model. Preliminary results indicate an accuracy of 4.7% in electrical energy production using a simplified electrical model.