Multiscale Interactions in Turbulent Flows: Experiments and Simulation

Johns Hopkins University, Baltimore MD

Investigators

Abstract

PROPOSAL NO.: CTS-0553314 PRINCIPAL INVESTIGATOR: C. MENEVEAU INSTITUTION: JOHNS HOPKINS UNIVERSITY MULTISCALE INTERACTIONS IN TURBULENT FLOWS: EXPERIMENTS & SIMULATIONS This research will address several critical aspects in understanding and numerical modeling of turbulent flows interacting with multiscale boundary conditions. Specifically, a new database of turbulent flow through fractal bluff bodies and grids will be generated using thermal and particle image velocimetry in a wind tunnel. Fractal objects display large scale-disparity and complexity while being amenable to simple and standardized description. Hence, they offer an elegant idealization of the actual interactions between turbulence and boundaries in practical applications where boundaries are characterized by multiple length-scales. The data will include flow measurements characterizing the turbulence as well as measurements of the drag forces acting on the fractal objects. A salient motivation for the proposed work is the need to validate and support further improvements of a new prediction tool, called Renormalized Numerical Simulation (RNS). This technique models forces on unresolved features using drag coefficients determined dynamically from the large scales. RNS will be applied to model flow across fractal bluff body arrays and grids, and predicted forces and flow features will be compared with the measurements. Another objective is to use the data to study scale interactions using spatial filtering and appropriate statistical techniques. Turbulence research and simulation are part of a larger push to understand and predict systems with multiscale complex behavior and many degrees of freedom. Large-scale numerical simulation of such systems is, today, at center stage of scientific discussions with important societal and political ramifications (global change, energy, etc.). Also, the development of properly validated RNS methodology applied to flows with fractal boundaries may yield broader impacts. Fractals have been used as a descriptive tool in many disciplines such as biology (branching blood network, pulmonary structures, corals), astrophysics (large-scale structure of the universe, intermittency of interplanetary magnetic fields), geosciences (fractal coastlines, vegetation, clouds). RNS extends the geometric idea of fractals to the dynamics. Educational impact of the proposed work will focus on graduate education and training that stresses the interplay between physical experimentation and simulation. Educational Outreach will take place in collaboration with the Engineering Schools Center for Educational Outreach (CEO). In the summer of 2006 and 2007, a high-school teacher will be invited to work on a project dealing with flow visualization of turbulent wakes behind fractal grids.