Speaker
Description
Within smart cities, Artificial Intelligence (AI), machine learning, and Information and Communication Technologies (ICT) are being used in order to improve energy efficiency through energy forecasting, load balancing, and optimized control systems. These objectives are closely aligned with Sustainable Development Goal 7 (Affordable and Clean Energy), which emphasizes enhancing energy access, efficiency, and sustainability.
In semi-arid urban regions, buildings are subject to extreme energy demands, particularly for cooling. Office buildings have predictable occupancy and energy demands. This presents a unique opportunity for sustainable energy strategies that adapt dynamically to occupancy, weather, and behavioral variations, and therefore lies the need to forecast energy consumption in smart buildings using artificial intelligence (AI), with a focus on climate-adapted metrics and technologies.
A comprehensive classification of energy performance metrics and indicators, including Energy Use Intensity (EUI), Coefficient of Performance (COP), and Occupant-Centric Metrics, helps examine energy consumption and, therefore, select appropriate forecasting models and interpret energy behavior in context.
In the literature, Many approaches are used for short-term energy forecasting, like Hybrid deep learning models and Artificial Neural Networks. However, their success depends heavily on data quality and climate context, and the critical role of accurate occupancy. Plus, Integrating Energy forecasting into Building Management Systems can significantly improve energy resilience and user comfort in smart buildings. It is an essential foundation for advancing inclusive and adaptive energy strategies in the context of sustainable urban development in semi-arid environments.
Keywords: Energy forecasting, Energy Optimization, Smart buildings, Semi-arid climates
