With the growing demand for accurate power forecasting in the energy sector, advanced machine learning models, and data analytics techniques are increasingly being employed to predict energy production and consumption patterns. This paper compares traditional regression models, such as Decision Tree, Multiple Linear, and Random Forest, with deep learning approaches, including Artificial Neural Networks (ANN) and Convolutional Neural Network (CNN) in forecasting the electrical power output of a Combined Cycle Power Plant (CCPP). A publicly available dataset with four features of ambient temperature (AT), relative humidity (RH), exhaust vacuum (V), and ambient pressure(AP) was used for training and testing the models. After preprocessing and feature scaling, the CNN model demonstrated strong performance with an RMSE of 3.16 and an R-squared value of 96.56%. Meanwhile, the ANN model achieved superior results with an RMSE of 2.83 and an R-squared value of 97.24%. The findings suggest that deep learning models, particularly ANN and CNN, outperform traditional machine learning approaches in capturing the complex patterns that influence power output, providing an effective solution for optimizing energy production in CCPPs.