The document presents a study on forecasting the energy output from a combined cycle thermal power plant (CCPP) using deep learning models, particularly artificial neural networks (ANN) and deep neural networks (DNN). Various operating thermal parameters such as ambient pressure, vacuum, relative humidity, and temperature were modeled, resulting in acceptable prediction accuracy, with R-squared values reaching up to 0.94. The study concludes that a single-layer ANN is recommended for its computational efficiency, while the models successfully predict the plant's energy output under varying atmospheric conditions.