Session: 11-13-01: Fluids and Public Health and Medicine / Industrial Flows

Paper Number: 165944

Digital Twins for Next-Generation HEPA Filters: Enhancing Air Flow Efficiency Through Alternative Computational Methods 

Digital twins are virtual replicas of physical objects that allow for detailed simulation and analysis. In the context of High Efficiency Particulate Air (HEPA) filters, a digital twin can be used to simulate the filter's performance under various conditions, providing a more accurate and comprehensive understanding of its efficiency. Creating digital twins involves advanced computational methods to replicate the filter's intricate geometry, enabling the estimation of filter efficiency using Computational Fluid Dynamics (CFD) simulations.  However, current digital twin models fail to accurately capture the complexity of a realistic filter.  CFD simulations provide a powerful tool for modeling fluid flow through filter media. However, the complexity and expense of modeling air flow around the full collection of individual fibers of a realistic filter is prohibitive. Thus, this study explores the feasibility of alternative CFD methods to efficiently predict air flow in realistic filtration media. Traditional CFD using Ansys Fluent will be used to compute air flow around a 10 x10 x10 mm cube of fibers that are explicitly geometrically modeled and meshed to high resolution. As an alternative CFD approach, Lattice-Boltzmann methods (LBM) to predict air flow through directly model fiber media will be investigated. For the LBM approach, individual fibers will be approximated by voxelized structures of flow blockage. A second alternative CFD approach using OpenFOAM will be investigated that approximates fibers by localized zones of porous media blockages. Using the traditional CFD model as a baseline, efficiencies and effectiveness of each CFD method will be compared and reported. Particulate matter represented by discrete particles of various diameters will be advected through the various CFD models, and results will be compared. The final goal is to develop an efficient digital twin of filtration that leverages the power of high-performance computing and advanced simulation techniques to offer a comprehensive solution for estimating HEPA filter efficiency. Studies have demonstrated the effectiveness of CFD in optimizing filter designs, showcasing the potential of this approach for HEPA filters.  The proposed method of using digital twins and CFD addresses several weaknesses of current approaches. It provides a more accurate representation of the filter's structure and performance, reduces the need for costly and time-consuming experiments, and allows for extensive parametric studies to optimize filter designs. By leveraging the power of high-performance computing and advanced simulation techniques, this method offers a comprehensive solution for estimating HEPA filter efficiency.  Therefore, future work will extend the most efficient approach for air modeling to include particle dynamics and multimodal capture mechanics to determine the viability and life cycle performance of new filter media types/configurations that meet HEPA standards. 

Presenting Author: Autumn Parker Institute for Clean Energy Technology Mississippi State University

Presenting Author Biography: Autumn is a graduate student in mechanical engineering at Mississippi State University. She has been working with Institute for Clean Energy Technology and CAVS since fall 2024. She is working under a DOE funded University Nuclear Leadership Program fellowship. After obtaining a PhD degree, she plans to pursue a career in the energy sector.

Authors:

Autumn Parker Institute for Clean Energy Technology Mississippi State University
Greg Burgreen Center for Advanced Vehicular Systems (CAVS) Mississippi State University
Omar Es Sahli Mechanical Engineering Department Mississippi State University
Shanti Bhushan Center for Advanced Vehicular Systems (CAVS) Mississippi State University
Alta Knizley Institute for Clean Energy Technology Mississippi State University