Desarrollo de un modelo predictivo basado en inteligencia artificial para la optimización de la gestión energética en sistemas solares híbridos basados en energía solar

Abstract

Abstract: This work presents the development of an artificial intelligence based predictive model aimed at optimizing energy management in a grid-connected hybrid solar system. The proposed approach is applied as a case study to the hybrid photovoltaic installation located in Building 9 of the Universidad Técnica Particular de Loja. In this research, machine learning techniques are integrated to forecast photovoltaic generation and electrical demand in the short term, using a multi-horizon Random Forest model with a 5-minute resolution. Subsequently, the PV and demand predictions are used as inputs for an optimal dispatch algorithm formulated through linear programming, focused on reducing energy purchase costs from the public grid while maximizing the utilization of solar generation and battery storage. The results demonstrate a significant improvement compared to the baseline model and highlight economic benefits achieved through different operational strategies, such as Time of-Use (TOU) tariffs. Overall, this study contributes to the advancement of more efficient and sustainable energy systems aligned with Industry 4.0 principles.