Cost Optimization in Power Generation: Leveraging Renewable Energy with Advanced Neural Networks

This research examines the incorporation of renewable energy sources-specifically wind, hydro, and solar-into a conventional power generation system consisting of ten thermal units, focusing on their effects on fuel and emission expenses. Artificial neural networks (ANNs) were employed to create synthetic data for renewable energy generation based on environmental variables like temperature, humidity, and historical records. Two scenarios were evaluated: the first involved solely traditional power generation, while the second included forecasts of renewable energy output. The ANN models were trained using the generated datasets, enabling precise predictions of renewable contributions. Additionally, power generation optimization was carried out using the Sine Cosine Algorithm in MATLAB. The findings revealed notable decreases in both total fuel costs and emissions with the use of renewable energy. In the first scenario, total costs represented the operational expenses of conventional generators, while the second scenario showcased the economic benefits of integrating renewable resources, emphasizing their potential to reduce environmental impacts. This study offers important insights into the economic viability of integrating renewable energy, aiding the transition to more sustainable energy systems.