KOMPERDA-DISSERTATION-2021.pdf (109.41 MB)
Computational Particle Tracking in Turbulent Flow Applications
thesisposted on 2021-08-01, 00:00 authored by Jonathan Komperda
The present dissertation develops and applies particle-based numerical schemes to simulate complex real-world engineering problems, such as combustion and the dispersion of evaporating droplets. Particle-based numerical methods typically consist of a carrier fluid and a dispersed phase of particles. In these methods, particles move along trajectories, without regard for the underlying grid, based on interpolated values from the Eulerian solver. At each time-step, particle properties are calculated on the grid via ensemble averaging, resulting in a hybrid particle-mesh solution. In this thesis, a coupled discontinuous spectral element method (DSEM) and particle-based filtered mass density function (FMDF) solver is developed for the large eddy simulation (LES) of compressible reacting flows on unstructured grids. The hybrid DSEM-LES/FMDF scheme features high-order accuracy and h-P refinement with minimal computational overhead. The novel scheme is validated by simulating the Taylor-Green vortex and spherical explosion problems. The non-reacting simulations demonstrate the high accuracy of the solver, achieving a correlation coefficient exceeding r=0.996 for the first subgrid moment, and the reacting simulations validate the solver for flame compositional structures ranging from pure mixing to the infinitely fast reaction limit. The spherical explosion simulation validates the scheme on unstructured grids in the presence of shockwaves. Also in this thesis, the discrete phase model (DPM) is applied to the simulation of evaporating droplets in a real-world dentistry clinic that performs aerosol-generating procedures during the COVID-19 pandemic. The simulation tracks the residence time and travel distance of evaporating droplets, aerosols, and droplet nuclei to determine airborne aerosol concentration, droplet deposition, and droplet escape through ventilation. The results show that detached jets from the room's ventilation system entering patient treatment areas enhance droplet residence time and travel distance, exceeding the six-foot social distancing guideline. Additionally, droplet deposition is observed in regions away from the patient treatment area.
DepartmentMechanical and Industrial Engineering
Degree GrantorUniversity of Illinois at Chicago
Degree namePhD, Doctor of Philosophy
Committee MemberAggarwal, Suresh K. Brezinsky, Kenneth Jaberi, Farhad Jacobs, Gustaaf B. Megaridis, Constantine M.
Submitted dateAugust 2021
CategoriesNo categories selected
Fluid MechanicsComputational Fluid DynamicsTurbulent FlowCompressible FlowCombustionReacting FlowsSpectral Element MethodDiscontinuous Spectral Element MethodFiltered Density FunctionFiltered Mass Density FunctionParticle Mesh MethodHigh-Performance ComputingShock CapturingEvaporating DropletsDiscrete Phase ModelFluentCOVID-19SARS-CoV-2AerosolsLarge Eddy SimulationDirect Numerical SimulationReynolds Averaged Navier Stokes