A new predictive model for a photovoltaic module's surface temperature

Abstract

The current study developed an analytical model to predict the PV module's operating temperature based on an experimental database, which considers cell temperature, local meteorological data (irradiance, ambient temperature, wind velocity, and humidity), voltage, and current generated by the photovoltaic system associated with the purely resistive load. Based on the analysis of the 172-day database, it was possible to compare the most used correlations in the literature with the analytical model developed in the current work. For all conditions, the model showed a better response to climate variation – with 100% of the data within an error band of ± 20% and an absolute mean percentage error of 3.1% – predicting well the PV module's operating temperature for both sky conditions (clear or cloudy) and demonstrated that the thermal capacity of the PV module to climatic variations should not be neglected. Moreover, the new model considered the PV module's thermal response capacity to include the variations in the incident solar irradiance caused by the presence of clouds (shading effect). By considering a global heat capacity as a mean value of the heat capacities of the layers of the PV module, the term transient – generally neglected in several works – is considered in the energy equation in the current work, which gives a better response to the variations in the incident radiation.