As the global demand for energy-efficient buildings grows, traditional simulation tools are becoming too slow to keep up with the complexity of sustainable design.
This research presents a comprehensive review of Artificial Neural Networks (ANNs) as a high-speed solution for Building Performance Simulation (BPS). By acting as “metamodels” (or digital proxies) ANNs can predict a building’s energy consumption and comfort levels almost instantaneously, allowing architects to test thousands of design variations in seconds.
The study explicitly details the entire lifecycle of creating these AI models, from data pre-processing to final testing.
While acknowledging that ANNs require significant initial data to “learn,” the authors demonstrate that the trade-off is worth it: the resulting models are powerful enough to guide both the design of new structures and the retrofitting of old ones.
For the engineering community, this paper serves as a technical manual for integrating AI into the heart of sustainable urban development.
Learn more about this study here: https://doi.org/10.1016/j.enbuild.2020.109972
Reference
Roman, N. D., Bre, F., Fachinotti, V. D., & Lamberts, R. (2020). Application and characterization of metamodels based on artificial neural networks for building performance simulation: A systematic review. Energy and Buildings, 217, 109972