This paper focuses on an approach to predict the temperature of a Photovoltaic panel under varying irradiation conditions in Cordoba, Colombia. The thermal model developed considers a heat transfer analysis in order to estimate the performance of a photovoltaic solar system due to local temperature variation. The heat transfer model analyzes the photovoltaic cell as a system exposed to radiation and natural convection by carrying out a first law energy balance which takes into account the radiation energy from the sun that hits the panel and the energy lost from the photovoltaic cell through natural convection and radiation. To determine the natural convection heat transfer coefficient, the Grashof number was employed along with Nusselt and Rayleigh number in a dimensionless form. The model has been implemented in the Matlab-Simulink platform that allows to establish a specific empirical correlation among the Nusselt number and Rayleigh for PV statics panels operating under natural convection condition. This experimental process consists in an iterative adjust of the theoretical equations of natural convection with experimental data gathered from a real PV module operation. The variables measured were the surface temperature, the environmental temperature and the solar irradiation provided by a pyranometer. It is found a good agreement between the radiation behavior and the predicted temperature. The higher values of the irradiation and environmental temperature coincides with predicted and observed PV surface temperatures and the thermal performance of the panel. The mean absolute error of the model was 3.09 K and the root mean square deviation 3.47 K.