Energy consumption estimation using artificial neural networks: Gümüşhane

The significance of energy load forecasting is highlighted by playing a fundamental role in the efficient utilization of energy resources, prevention of interruptions, and the formulation of sustainable energy policies. Additionally, it is a critical factor in capacity planning for energy infrastructure and determining the necessity for new investments. In this context, artificial intelligence techniques, particularly artificial neural networks, and metaheuristic algorithms used in determining the parameters of these networks have gained prominence in recent years. This study focuses on the determination of artificial neural network parameters using the Ali Baba and Forty Thieves algorithm, aiming to predict energy load demands more accurately and reliably based on real data. The analysis in this study involves initially scrutinizing specific variables for input data in the selected region and energy consumption data for the output. Subsequently, the parameters of the artificial neural network are determined using the Ali Baba and Forty Thieves algorithm. The identified parameters are then used to run the artificial neural network, and the Mean Squared Error (MSE) value is examined. To conduct a performance analysis, the same procedures are executed using PSO and BSA for comparison. Emphasizing the importance of utilizing artificial intelligence-based technologies in energy demand forecasting, this study aims to contribute to future planning in the sector.