On Integrating Numerical and Laboratory Turbulent Flow Experiments

Abstract

Critical issues of numerical turbulent flow experiments, relate to the treatment of the unresolved (subgrid scale) flow features and required boundary conditions (BC’s) ‐ denoted as supergrid modeling. Because actual BC choices select flow solutions, emulating particular flow realizations demands precise characterization of their initial, inflow, and other relevant conditions at open-outflow or at solid and other facility boundaries. The inherently intrusive nature of both, computed and laboratory observations of turbulent phenomena is noted in this context. The flow characterization issue becomes a very challenging one in collaborative numerical/laboratory efforts, where potential sources of discrepancies need to be clearly evaluated and controlled. Achieving closure of the supergrid model based on laboratory data requires identifying appropriate data acquisition and its suitable post-processing (reduction) for effective use in the simulations. An overview of the subgrid and supergrid issues typically involved is presented, and relevant discretization aspects are noted. Turbulent inflow BC issues encountered when integrating computational and laboratory experiments in a particular complex flow problem of practical interest are discussed. Background URBULENT flows are of considerable importance in many applications in engineering, geophysics, meteorology, and astrophysics. Laminar (organized) flows transition to (chaotic) turbulent regimes, and these become normal states of fluid motion, often also involving complex multiphase physics, compressibility, and stratification. Availability of effective predictive design tools is a crucial aspect in the applications. In particular, there is a crucial interest in recognizing and understanding the local nature of the instabilities and their global nonlinear development in space and time, to achieve flow control in the applications. Laboratory studies typically demonstrate the end outcome of complex non-linear three-dimensional physical processes with many unexplained detailed features and mechanisms. Time-dependent flow experiments based on numerical simulations with precise control of initial and boundary conditions (IC’s and BC’s), are ideally suited to supplement the experiments carried out in the laboratory, providing insights into the underlying flow dynamics and topology leading to the laboratory observations. Numerical experiments can be used to isolate suspected fundamental mechanisms from others which might confuse issues, and extensive space/time diagnostics available based on the simulation database can be exploited to develop an analytical and conceptual basis for improved modeling of the phenomena of interest. In this collaborative context, a crucial aspect is that of adequately characterizing the conditions in the laboratory and numerical turbulent flow experiments, so that potential sources of discrepancies can be clearly evaluated and controlled. Capturing the dynamics of all relevant scales of motion, based on the numerical solution of the Navier‐Stokes (NS) equations, constitutes direct numerical simulation (DNS), which is prohibitively expensive in the foreseeable future for practical flows of interest at moderate-to-high Reynolds number. On the other end of computer simulation possibilities, the Reynolds-averaged Navier‐Stokes (RANS) approach, with averaging typically carried out over time, homogeneous directions, or across an ensemble of equivalent flows, is typically employed for turbulent flows of industrial complexity. Large eddy simulation (LES) has become the effective intermediate approach between DNS and RANS, capable of simulating flow features that cannot be handled with RANS, such as significant flow unsteadiness and strong vortex-acoustic couplings, and providing higher accuracy than RANS at reasonable cost (e.g., [1]). LES is based on the expectation that the physically meaningful scales of turbulence can be split into two groups: one consisting of the resolved geometry and regime specific scales (so-called energy containing scales), and the other associated with the unresolved smallest eddies, for which the presumably more-universal flow dynamics is represented with subgrid