Power plants producing energy through solar fields use a heat transfer fluid that lends itself to be influenced and changed by different variables. In solar power plants, a heat transfer fluid (HTF) is used to transfer the thermal energy of solar radiation through parabolic collectors to a water vapor Rankine cycle. In this way, a turbine is driven that produces electricity when coupled to an electric generator. These plants have a heat transfer system that converts the solar radiation into heat through a HTF, and transfers that thermal energy to the water vapor heat exchangers. The best possible performance in the Rankine cycle, and therefore in the thermal plant, is obtained when the HTF reaches its maximum temperature when leaving the solar field (SF). In addition, it is necessary that the HTF does not exceed its own maximum operating temperature, above which it degrades. The optimum temperature of the HTF is difficult to obtain, since the working conditions of the plant can change abruptly from moment to moment. Guaranteeing that this HTF operates at its optimal temperature to produce electricity through a Rankine cycle is a priority. The oil flowing through the solar field has the disadvantage of having a thermal limit. Therefore, this research focuses on trying to make sure that this fluid comes out of the solar field with the highest possible temperature. Modeling using data mining is revealed as an important tool for forecasting the performance of this kind of power plant. The purpose of this document is to provide a model that can be used to optimize the temperature control of the fluid without interfering with the normal operation of the plant. The results obtained with this model should be necessarily contrasted with those obtained in a real plant. Initially, we compare the PID (proportional–integral–derivative) models used in previous studies for the optimization of this type of plant with modeling using the multivariate adaptive regression splines (MARS) model.