Thesis work on the implementation of the 'filtered mass density function' in a discontinuous spectral element method and the simulation of a scramjet.
With a concentration on heat transfer and fluid dynamics.
Lead researcher on the implemention of supersonic-turbulent combustion models into an h/p element method code for the simulation of hypersonic vehicles. Work is performed under the administration of the Air Force Research Laboratory.
Perform research on discontinuous spectral element methods under the supervision of Dr. Farzad Mashayek. Maintain two Linux cluster-computers, one with 40 cores, the other with 116 cores. Maintain lab resources and perform reporting for grantors.
Teaching assistant for a Numerical Methods (ME428) course in the department of Mechanical and Industrial Engineering. I was responsible for writing tests, assigning homework, grading student material, proctoring tests, and maintaining a gradebook. Assitantship was performed under the supervision of Dr. Suresh Aggarwal.
In this work, the filtered mass density function (FMDF) methodology is implemented in the discontinuous spectral element method (DSEM) for simulation of reacting flows in complex geometries. DSEM solves the compressible Navier-Stokes equations on an unstructured grid with high orders of accuracy. The FMDF is a probability density function method that has been developed for large eddy simulation (LES) of variable-density chemically reacting turbulent flows. The solution to the modeled FMDF transport equation is obtained by solving an equivalent set of stochastic differential equations (SDEs) which yield statistically equivalent results. The set of SDEs is solved by means of a Lagrangian Monte Carlo scheme. The validity of the scalar FMDF is appraised by determining the consistency against DSEM in a temporally developing shear layer. Additional results are shown for the DSEM-LES/FMDF simulation of a hydrogen-air reaction in a temporally-developing shear layer.
This paper is on the application of compressible filtered mass density function (FMDF) model to high speed turbulent combustion. The FMDF is a subgrid-scale probability density function model for large eddy simulation (LES) of turbulent flows and is obtained by the solution of a set of stochastic differential equations with a Lagrangian Monte Carlo method. The validity and consistency of LES/FMDF are established by simulating a premixed hydrogen-air reaction in a supersonic cavity flow configuration with detailed chemical kinetics models.
While using high order methods is desirable in order to accurately capture the small scale mixing effects in reacting flows, the challenge is to develop and implement such methods for complex geometries. In this work, a high-order Discontinuous Spectral Element Method (DSEM) code, which solves for the Navier-Stokes equations, has been modified by adding the appropriate components to solve for scalar transport equations in order to simulate the chemical reaction. Dealing with discontinuous solution at element interfaces is a challenge that is met by patching the fluxes at mortars thus making them continuous on interfaces. The patching is performed using the Lax-Fredrichs numerical flux for scalars, whereas a generalized Riemann solver is used for the Navier-Stokes equations. Direct numerical simulation is conducted in a temporally developing mixing layer to validate the method for a single step reaction (F+rO→[1+r]P). Next, the method is implemented to simulate a subsonic reacting flow in a slanted cavity combustor with gaseous fuel injectors to demonstrate the capability of the method to handle complex geometries. The results will be used for physical understanding of mixing and reaction in these types of combustors.
This paper describes the implementation of an Entropy Viscosity (EV) method in a Discontinuous Spectral Element Method (DSEM) for simulation of high-speed flows with shock and contact discontinuities. The underlying concept in the EV method is to use entropy production in the flow as an indicator to introduce additional numerical viscosity only in the regions near discontinuities. The combined DSEM-EV method is implemented in 1D and 2D and several model problems are simulated for validation. The results of the simulations have been compared with the exact solutions, where available, and demonstrate a proof-of-concept for implementation of EV method in conjunction with the DSEM.
The electrostatic charge injection method is applied at high hydrodynamic pressures, up to 40 bar, to evaluate the electrical and atomization performance of a point-plane type charge injection atomizer using Diesel fuel. The main focus of this paper is to investigate the effect of electrostatic charging on higher pressure injection systems in order to form finer sprays. Laser Diffraction Spectrometry (LDS) measurements and imaging studies are performed to investigate the effect of the electrostatic charge injection technique on drop size distribution as a function of orifice size, applied electrode voltage, inter-electrode positioning, hydraulic pressure and corresponding axial tip velocity at high Reynolds numbers. It is observed that spray dispersion is enhanced and drop size is reduced with the increase in specific charge. The various stages experienced by the electrostatic spray as the electrode voltage is increased from zero are described in detail on the basis of an imaging study. LDS measurements are conducted to show how drop size profiles vary as a function of downstream position from the nozzle tip. These measurements are used to understand the aerodynamic effects of the surrounding air on the secondary break-up of the dense spray in virtue of more air entrainment to fuel spray. Finally, these findings are discussed and several advantages of the electrostatic charge injection method for modern diesel engines are outlined.