Session: 10-03-02: CFD Applications - II

Paper Number: 95997

95997 - Aerosol Dispersion Modeling With a Low-Cost Flow Simulation Tool: Analysis of Performance and Boundary Condition Sensitivity 

Understanding the dispersion of potentially infectious aerosols released from breathing and speaking is critical to maintaining a healthy indoor environment. This is especially true in classrooms, which tend to be densely occupied.  Computational Fluid Dynamics software utilizing such approaches as large eddy simulation can be useful in modeling this aerosol dispersion, but require significant expertise and computing power; in addition, such models tend to be expensive and time-consuming to create.  This study used a low-cost commercial off-the-shelf flow simulation tool to model aerosol dispersion within a university classroom, and assessed its utility by comparing results to those from physical experiments. SolidWorks Flow Simulation is a computational fluid dynamics package embedded within SolidWorks that uses the finite volume method to predict gaseous flows and tracer dispersion given user-defined boundary conditions. A SolidWorks model of a university classroom was created using parameters from a series of full-scale carbon dioxide tracer gas release experiments conducted in the physical classroom in the spring of 2021 at both low (520 cfm) and high (900 cfm) supply flowrates.  Steady state concentrations at 13 monitoring points throughout the room were obtained from the flow simulation tool, and compared to the experimental results using four performance measures for dispersion model evaluation (Hanna and Chang 2012, https://doi.org/10.1007/s00703-011-0177-1).  The performance measures for the baseline model fell well within published acceptance criteria for Fractional Mean Bias (FB), Normalized-Mean Square Error (NMSE), Fraction of C_predicted within a factor of two of C_observed (FAC2), and Normalized Absolute Difference (NAD).  A sensitivity analysis was then conducted to determine the relative influence of each of four boundary conditions: vent flow pattern (vertical versus dispersed), student heat flux, window heat flux, and door pressure difference.  The most critical boundary condition for creating an optimum model in the low flow case was the vent flow pattern. Adding an accurate representation of the classroom vent flow (i.e., dispersion in four directions in accordance with the louvred design) for the low flow model improved the FB from 0.88 to -0.06, the NMSE from 1.07 to 0.04, the FAC 2 from 0.08 to 1.0, and the NAD from 0.44 to 0.08.  Differences from the other boundary conditions were insignificant.  In the high flow case, similar trends were observed, though the addition of the dispersed vent flow had slightly less significant results, but still acceptable parameter values.  Regardless, all parameters met minimum acceptance criteria in each run of the sensitivity analysis.  Given its ability to create an acceptable SolidWorks model at relatively low cost, this accessible tool could be used by a variety of interested parties to quickly predict the spread of aerosols, gases, and other passive scalars within rooms under a variety of ventilation and physical arrangement conditions. This software tool may already be available to many academic institutions and provides a reasonable expectation of accuracy.

Presenting Author: Philip Dacunto United States Military Academy

Presenting Author Biography: COL Phil Dacunto, PE is an associate professor of environmental engineering at the United States Military Academy at West Point, NY. He earned a PhD at Stanford University in 2013, and currently serves as Deputy Head of the Department of Geography and Environmental Engineering. He has served as a US Army officer since 1995, and his research interests include indoor air quality and aerosol dispersion.

Authors:

Mallory Hirn United States Military Academy
Andrew Rodriguez United States Military Academy
Mark Owkes Montana State University
Philip Dacunto United States Military Academy
Andrew Ng United States Military Academy
John Rogers United States Military Academy
Michael Benson United States Military Academy