One of the problems in “green” building design is that the current tools used to evaluate building’s thermal and airflow behaviors allow performance analysis of new and existing buildings extremely time-consuming. Owing to rigorous computational effort that is required for thermal and airflow behavioral studies, they fail to explore the design space in a timely manner that is conducive to architects and engineers.
To alleviate this problem, I developed a method using Artificial Intelligence (AI) techniques to rapidly solve airflow patterns as evidenced in the following publications, “Real-time Simulations Using Learning Algorithms for Immersive Data Visualization in Buildings” in International Journal of Architectural Computing; “Toward Real-Time Airflow Simulations for Immersive Visualization using Adaptive Localization Method,” published in the Tenth International Building Performance Simulation Association Conference held in Beijing, China; and “Adaptive Localization Method: An Approach to Real-Time Airflow Simulation and Immersive Visualization,” published in the Seventeenth International Conference on Computer Graphics and Vision (GRAPHICON) held in Moscow, Russia.
For the paper titled, “Reinforcement Learning and Real-time Thermal Performance Visualization in Buildings,” and significant accomplishment, I won the Young CAADRIA Award presented by the Association for Computer-Aided Architectural Design Research in Asia. The importance of the method I developed is that it allows an architect or project teams worldwide to explore different building configuration and see the effects of changes on the building’s thermal- and airflow and, thereby, optimize building energy use.