Development of computational fluid dynamics based multiple linear and neural network metamodels for bioaerosol fate and transport in indoor environments

Abstract

Linear, quadratic, and artificial neural network (ANN) based metamodels were developed to predict bioaerosol fate and transport in indoor environments using computational fluid dynamic (CFD) simulations. The objective of the metamodels is to provide quick and relatively accurate information during emergencies. The response to the 2001 anthrax-release events indicate that decontamination of indoor spaces following the release of biological agents is challenging. The ability to efficiently and rapidly decontaminate rooms/buildings is limited by the lack of quantitative understanding of the behavior of bioaerosols and decontaminants in relation to building geometry, airflow pattern and surface properties. The work here addresses this challenge.The variables characterizing the system were identified and dimensionless groups were developed. Design of experiments was used to explore the design space and determine the multiple scenarios. The results derived from the CFD simulations of these scenarios were used to develop the metamodels. The CFD model comprised of multiple sub–models. Large eddy simulation (LES) with the Smagorinsky sub-grid scale model was applied to compute the airflow. Bioaerosols were modeled as a dispersed solid phase using the Lagrangian treatment. For inactivation studies disinfectant mass fraction was calculated by solving a mass transport equation. Kinetic decay constants were included for spontaneous decay of the disinfectant and for the reaction of the disinfectant with surfaces. An inactivation rate equation accounted for the reaction between the spores and the disinfectant. For the study of electrostatic forces the Poisson equation was solved to determine the electric field.The ANN based metamodels were most successful in predicting the number of viable bioaerosols remaining in an arbitrary enclosed space and their spatial heterogeneity. Sensitivity analysis showed that the ratio of the particle tracking time to residence time and the location of input and output with relation to the height of the room had the most impact. Mass fraction of the disinfectant, inactivation rate constant and contact time had the most influence on the inactivation of the spores. Screen voltage was the most significant variable influencing the dispersion of charged bioaerosols.To the best of the author’s knowledge this is the first attempt of developing userfriendly models or metamodels for fate and transport of bioaerosols from CFD simulations. The metamodels besides being useful during emergencies also gives the opportunity to better understand the system and therefore provide information for designing more secure and safer indoor environments.%%%%Ph.D., Environmental Engineering – Drexel University, 2010