High Reynolds number flow over smooth curved surfaces: separation, new wall modeling physics, and LES
Abstract
Accurate modeling of flow physics and turbulence near solid surfaces remains an open problem in computational fluid dynamics (CFD).In particular, there is strong interest in predicting high Reynolds number flows over smooth, curved bodies because of design, control, and maneuverability requirements of underwater vehicles. In order to advance the state-of-the-art, a novel wall modeling approach for Large Eddy Simulation (LES), the multi-timescale (MTS) wall model will be extended to include effects of curvature for applicability to high Reynolds number turbulent flow over smooth walls including curvature and separation. The MTS approach enables modelers to rigorously separate various time-scale dynamics that govern the near wall region of relevance for wall modeled Large Eddy Simulations (WMLES). In particular, quasi-equilibrium, and non-equilibrium laminar and turbulent physics that affect the relationship between wall stress and the LES-resolved flow away from the wall can be represented more faithfully than standard existing equilibriumwall models or ad-hoc non-equilibrium wall models. The quasi-equilibrium part of the MTS model takes the form of a temporal relaxation equation while the rapid laminar part is based on a relationship between wall stress and backward temporal convolution derived analytically from a generalized viscous Stokes problem. The turbulent non-equilibrium portion of the model is based on the attached eddy-hypothesis that enables to relate the fluctuating portion of the LES velocity to fluctuations in wall stress due to wall attached eddies. The currently available MTS model is applicable to flat smooth-wall topologies. In this proposed project it will be extended to include effects of wall curvature explicitly by casting the boundary layer equations in a streamline coordinate system. The resulting extended MTS wall model will be tested on flows with separation (channel flow with suction and blowing), flows with significant wall curvature (turbulent flow over a periodic two-dimensional hills), and as a capstone application, the model will be applied to flow over a 6:1 prolate spheroid at various Reynolds numbers and angles of attack. The latter represents an application to a complex 3D flow over a smooth-surface object in which wall curvature and separation play crucial roles in the dynamics. This project will significantly advance wall modeling for LES, a pace-setting open problem in CFD along a new promising direction. The improvementswill directly benefit modeling accuracy for flows over smooth surfaces including curvature and separation, as typically occur for underwater vehicles, manned and unmanned.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 12, 2023
- Source ID
- N000142312185
Entities
People
- Charles Meneveau
Organizations
- Johns Hopkins University
- Office of Naval Research
- United States Navy