AI-driven optimization of indoor environmental quality and energy consumption in smart buildings: a bio-inspired algorithmic approach

Abstract

This work presents an in-depth analysis of five bio-inspired optimization algorithms, namely Puma Optimizer (PO), Walrus Optimizer (WO), Flying Fox Optimization Algorithm (FFO), Waterwheel Plant Algorithm (WWPA), and Energy Valley Optimizer (EVO), which optimize the energy consumption while keeping the occupants’ comfort intact within smart building environments. It emulates operational challenges such algorithms might face in the real world for testing, such as dynamic energy demand, fluctuating occupancy, and time-varying weather, to meet the perfect balance between energy efficiency and indoor environmental quality. Key findings indicate that the algorithms achieve significant energy savings and maintain stable temperature and humidity levels across different zones. The comparison provides insight into each algorithm’s strengths in various scenarios and, potentially, in real-time smart building management systems applications. Further, integrations of multidimensional visualization techniques enhance the trade-off interpretations between energy consumption and occupants’ comfort. Thus, it is a valuable reference for sustainable building design.

Keywords

• bio-inspired algorithms
• smart buildings
• energy optimization
• occupant comfort
• machine learning