Physical-Based Reduced-Order Model for Buildings Energy Efficiency

This research aims to investigate the feasibility and possible benefits of applying Reduced-Order Models (ROMs) to the Architecture, Engineering and Construction sector. The reduced-order model is a methodology for designing complex systems that exploit the synergy of interacting phenomena. Although it can significantly speed up design and run dynamic simulations, this technique is still not widely applied in the AEC field. Other industries often use them to optimize systems performance, create digital twins, and streamline multidisciplinary optimization methodology. The first section of this paper describes the proposed methodology, followed by the study of the advantages and limitations compared to traditional design systems, including an analysis of current research and applications. Instead of a prescriptive design based on the worst-case scenarios, this kind of simulation enables performance-based design with partial loads and dynamic behavior. This innovative approach makes it possible to support the designer during the conceptual phase and deal with the design problem more consciously to realize more sustainable and comfortable solutions. ROMs require low computational capacity and thus can handle real-time sensor data during operation and maintenance. A drawback is the difficulty of understand and model the physics underlying a specific system. The last section presents an operational workflow and the early stage of a test application that explores how physical-based ROM can support the designer in defining energy interventions on an existing building’s classroom. Thermal and energy behavior, heating, ventilation, and air conditioning systems and their interaction have been modeled. The final step consists in incorporating subsystems representing potential active and passive energy improvement interventions to understand how the system improves and minimizes energy consumption improving indoor environmental quality.