Modelo de red neuronal para la predicción del consumo de energía eléctrica en Iquitos

• Autores: Pinedo Amias, Jheferson Romulo, Lozano Gomez, Miller
• Formato: tesis de grado
• Fecha de Publicación: 2022
• Institución: Universidad Nacional De La Amazonía Peruana
• Repositorio: UNAPIquitos-Institucional
• Lenguaje: español
• OAI Identifier: oai:repositorio.unapiquitos.edu.pe:20.500.12737/8945
• Enlace del recurso: https://hdl.handle.net/20.500.12737/8945
• Nivel de acceso: acceso abierto
• Materia: Redes neuronales (informática); Programa de aplicación; Inteligencia artificial; Energía eléctrica; https://purl.org/pe-repo/ocde/ford#2.02.04

The main objective of this study was to evaluate the implementation of an artificial neural network model to improve the forecasting capacity of electricity consumption in the city of Iquitos. The methodology used in this research included the collection of historical data on electricity consumption in Iquitos, the implementation of an artificial neural network model, and the evaluation of the forecasting capacity of the model using statistical metrics such as the correlation coefficient and the error. mean square. The hypothesis of this research maintains that the implementation of an artificial neural network model will improve the forecasting capacity of electricity consumption in Iquitos, which will contribute to a more efficient management of electricity in the city. The results of this research indicate that the implementation of the artificial neural network model allowed reaching a correlation coefficient higher than 0.95 and reducing the Mean Square Error in the prediction of electricity consumption by 5%. In conclusion, the implementation of an artificial neural network model can improve the forecasting capacity of electricity consumption in Iquitos, which contributes to a more efficient management of electricity in the city. These results are important for planning and decision-making on the supply and demand of electricity in Iquitos and can serve as a basis for future studies on the use of artificial neural networks in the management and forecasting of electricity consumption in other regions of the world.