Turbulence computations with 3-D small-scale additive turbulent decomposition and data-fitting using chaotic map combinations

Thesis/Dissertation · Wed Jan 01 04:00:00 EST 1997

DOI:https://doi.org/10.2172/666048· OSTI ID:666048

Mukerji, Sudip ^[1]

---

1. Univ. of Kentucky, Lexington, KY (United States)

Although the equations governing turbulent fluid flow, the Navier-Stokes (N.S.) equations, have been known for well over a century and there is a clear technological necessity in obtaining solutions to these equations, turbulence remains one of the principal unsolved problems in physics today. It is still not possible to make accurate quantitative predictions about turbulent flows without relying heavily on empirical data. In principle, it is possible to obtain turbulent solutions from a direct numerical simulation (DNS) of the N.-S. equations. The author first provides a brief introduction to the dynamics of turbulent flows. The N.-S. equations which govern fluid flow, are described thereafter. Then he gives a brief overview of DNS calculations and where they stand at present. He next introduces the two most popular approaches for doing turbulent computations currently in use, namely, the Reynolds averaging of the N.-S. equations (RANS) and large-eddy simulation (LES). Approximations, often ad hoc ones, are present in these methods because use is made of heuristic models for turbulence quantities (the Reynolds stresses) which are otherwise unknown. They then introduce a new computational method called additive turbulent decomposition (ATD), the small-scale version of which is the topic of this research. The rest of the thesis is organized as follows. In Chapter 2 he describes the ATD procedure in greater detail; how dependent variables are split and the decomposition into large- and small-scale sets of equations. In Chapter 3 the spectral projection of the small-scale momentum equations are derived in detail. In Chapter 4 results of the computations with the small-scale ATD equations are presented. In Chapter 5 he describes the data-fitting procedure which can be used to directly specify the parameters of a chaotic-map turbulence model.

• Research Organization: Univ. of Kentucky, Lexington, KY (United States)
• Sponsoring Organization: USDOE Assistant Secretary for Fossil Energy, Washington, DC (United States)
• DOE Contract Number: FG22-93PC93210
• OSTI ID: 666048
• Report Number(s): DOE/PC/93210--11c; ON: DE98007435
• Country of Publication: United States
• Language: English

Related Subjects

42 ENGINEERING
97 MATHEMATICS AND COMPUTING
ALGORITHMS
CALCULATION METHODS
DATA PROCESSING
FLOW MODELS
NAVIER-STOKES EQUATIONS
PARALLEL PROCESSING
THREE-DIMENSIONAL CALCULATIONS
TURBULENT FLOW