The compressible wall law and turbulence modeling
Iowa State University
Abstract
Turbulence models become inaccurate in high Mach number boundary layers. It hypothesized that this is due to their violating the law-of-the-wall. A wrinkle is that there may not, strictly, be a high Mach number law-of-the-wall, as will be discussed. However, approximate reasoning is enough to guide modeling. That reasoning, in the context of the k–ω turbulence model, will be reviewed.
The standard ω-equation has regular and singular solutions. The k–ω model requires that the singular solution be selected. That is not attractive for numerical discretization. Indeed, the singular solution violates the high Mach number wall scaling cited above. A procedure will be introduced to map a non-singular numerical solution of the ω-equation into the singular behavior. The mapping is consistent with the high Mach number wall law. It is shown that this procedure provides virtually the same results as the singular solution in low speed flow, and significantly better results in high Mach number boundary layers. For rough surfaces, the mapping is modified by introducing a displacement of the origin in the mapping function.
About the speaker
Dr. Paul Durbin is the Joseph C and Elizabeth A Anderlik professor in the Aerospace Engineering department at Iowa State University. His Ph.D. is from the department of Applied Math and Theoretical Physics at the University of Cambridge. He has held positions as a research scientist at NASA, as a senior fellow at the Center for Turbulence Research, and as a professor in Mechanical Engineering at Stanford University. His research is on turbulence and transition in fluid flows, including theory, modeling and computer simulation. Recent work is on Reynolds averaged closure models and hybrid RANS/LES models for turbulent flow and laminar-to-turbulent transition. He is the general editor of the recent series 6 books, Computation and Analysis of Turbulent Flows.

