Proyección de la demanda de energía eléctrica a mediano plazo, mediante redes neuronales y regresión lineal múltiple dentro de la ciudad de Cuenca

Abstract

This thesis focuses on the design and simulation of a prediction model for medium-term (up to three years) electricity demand forecasting in the Yanuncay and Baños parishes of the city of Cuenca. The model is based on the application and comparison of multiple linear regression techniques and artificial neural networks using the student trial version of MATLAB® software LicenseNo: DEMO. This research project develops a more accurate and robust tool for the efficient planning and management of electrical resources in these sectors of Cuenca. Additionally, this methodology offers a practical alternative for assessing energy demand, considering economic, climatic, social, and technological factors. The document addresses the problem and objectives of the work, provides a literature review on the basic concepts necessary for understanding the requirements of electricity supply, and discusses the factors that may affect it. Furthermore, it explains different methods used for sizing electricity demand and describes the methodology employed in the research project, detailing the data used and the models developed, as well as the results obtained from the multiple linear regression (MLR) and artificial neural network (ANN) models for the considered parishes. Finally, the document presents the conclusions derived from the project and recommendations for future work based on the generated results.