Development of Intelligent Models for Building Energy Management

This study explores the application of advanced deep learning algorithms for accurate prediction and management of building energy consumption, particularly in high-occupancy settings. Leveraging real-world data from a building equipped with fan coil and gas boilers, the research focuses on training and testing Long Short-Term Memory networks to forecast hourly electricity usage. Additionally, an investigation into temperature control algorithms for rooms with fan coil systems is conducted. The LSTM model demonstrates superior performance, achieving high accuracy and robustness in electricity consumption prediction. The findings not only contribute to optimizing energy management strategies but also pave the way for the development of scalable data-driven algorithms crucial for enhancing building energy efficiency, facilitating fault diagnosis, and promoting the integration of renewable energy sources in smart buildings.