Investigation into the Role of Initial Conditions on Rayleigh-Taylor Instabilities by Low Atwood Experiments and Simulations

Mueschke, N ; Andrews, M ; Schilling, O

Lawrence Livermore National Laboratory

关键词: Diffusion;    Simulation;    Texas;    Kinetic Energy;    Randomness;   

DOI  :  10.2172/15004813
RP-ID  :  UCRL-ID-154412
RP-ID  :  W-7405-ENG-48
RP-ID  :  15004813

美国|英语

来源: UNT Digital Library

【 摘 要 】

The primary goal of the research being conducted this summer is to investigate the role of initial conditions in the development of a two fluid mix driven by Rayleigh-Taylor instability. The effects of initial conditions will be studied through the use of experimental facilities located at the Buoyancy-Driven Mixing Lab at Texas A&M University and through high resolution direct numerical simulations of the experiment by the MIRANDA code developed at Lawrence Livermore National Lab. The Experimental Objectives are: (1) Analyze the early time development of a two fluid Rayleigh-Taylor driven mix between two miscible fluids at low Atwood numbers. (2) Quantify the initial conditions of the unstably stratified fluids by means of statistical mixing parameters and spectral analysis of the centerline density fluctuations. (3) Capture PLIF images of initial development of the flow for use in simulation setup. (Wayne Kraft) (4) Determine exactly what component of the experimental mixing data (position downstream from the splitter plate) most accurately represents the initial conditions of the experiment. The Simulation Objectives are: (1) Perform two dimensional and three dimensional simulations of the experimental setup. Analyze the results of these simulations for comparison to the experimental results. (2) Various methods of implementing the initial conditions in the simulations are to be investigated. Some of those methods are: (a) Various simplified density profile assumptions will also be investigated, such as repeating saw-teeth patterns, etc. There is also a concern to add some degree of randomness to these simplified perturbation profile assumptions. (b) Convert portions of raw PLIF data to a set of parameterized surfaces that can be directly input as both two dimensional and three dimensional surfaces. (c) Determine and implement a method for directly converting the initial density spectral data into a density profile that can be implemented in two and three dimensional simulations. (3) Quantify the dynamical quantities associated with the evolution equations of density, kinetic energy, and enstrophy. The Modeling Objectives are: (1) Perform a similar set of simulations using the artificial diffusion equations proposed by Oleg Schilling to validate their use. Results are to be compared to the experimental and DNS simulations. (2) Perform comparisons between DNS simulations of experiment and the proposed EZTurbMix models under development by Oleg Schilling.