Toward real-time indoor airflow simulations for immersive visualization using adaptive localization method

Traditional approaches to simulate airflow movements in buildings are computationally expensive and do not achieve real-time prediction of results. This paper discusses an Adaptive Localization Method (ALM) that significantly minimizes the simulation domain to achieve close to real-time predictions. As the user interacts with the space by modifying boundary conditions (opening a window, etc), while being immersed in an Augmented Reality (AR) environment, the ALM detects the changes and narrows down the simulation space significantly for re-simulation instantly. This localized space is simulated and the newly generated airflow data is updated to corresponding spatial nodes for interactive, immersive AR visualization. The ALM is developed based on a series of simulations conducted to identify critical variables that alter the rate of change of velocity, magnitude and temperature of air due to changes in the boundary conditions. ALM based real-time AR model will aid in studying “what if” scenarios for buildings, particularly for applications such as remodeling and refurbishment to improve conditions, etc.