Separation control of flow over a backward-facing ramp

Kevin Maki, University of Michigan

0000-0001-7286-7402

ACCESS Allocation Request ASC160041

Past CoPI: Eric Johnsen University of Michigan
Abstract: Flow separation in wall bounded turbulent flows is known to have adverse effects in many engineering applications like renewable power generation in wind farms, reducing lift on aircraft wings, increasing drag on automobiles. In order the mitigate the adverse effects of flow separation it is important to understand the flow physics of boundary layer separation and the flow characteristics in the separation region. Separation of flow over a backward-facing ramp provides us with a canonical geometry to study the separation flow physics. The objective of this work is to gain insight into the mechanism that leads to the reattachment of the flow after it separates over a backward-facing ramp in the presence of wall-mounted cubes used as passive vortex generators (VGs). Wall resolved large-eddy simulation (LES) technique is utilized to perform high-fidelity numerical simulations of flow over a backward-facing ramp. The Reynolds number based on the freestream velocity (U0) and thickness of the boundary layer (δ) prescribed at the inlet is approximately 19,600. Two sets of studies have been conducted so far: 1) Single cube studies to identify the dependence of flow modulation on the VG configuration, namely its height relative to the boundary layer thickness (h/δ), and its upstream position with respect to the boundary layer thickness (xvg/h); and 2) The optimal configuration form the single VG case is used to configure a line array of equally-spaced cubes and the effect of spanwise inter-cube spacing (Lz) on the flow modulation is investigated. We now plan to investigate the effect of multiple line arrays of cubes on the flow modulation. Deriving from the previous studies, optimal configuration of a line array will be used to setup two types of flow problems, staggered and aligned arrays. For each case, the streamwise spacing between each line array will be varied and its effect on the flow modulation will be investigated. Previous studies of turbulent flow over spanwise heterogeneous roughness elements on a flat plate have shown the formation of secondary and tertiary turbulent structures, which result in high and low-speed pathways. It is expected that such structures will alter the flow behavior of vortical structures formed in the inlet section before the ramp, and thus, will affect the modulation of flow. To verify this hypothesis and achieve other aforementioned objectives we request renewal for our research allocation with an additional 425,000 SU’s on XSEDE Stampede2.

Allocations:

2020 SDSC Expanse CPU 2,000,000.0 Core-hours
2020 SDSC Expanse Projects Storage 1,000.0 GB
2020 TACC Dell/Intel Knights Landing, Skylake System (Stampede2) 36,000.0 Node Hours
2020 TACC Long-term tape Archival Storage (Ranch) 1,000.0 GB
The estimated value of these awarded resources is $18,245.60. The allocation of these resources represents a considerable investment by the NSF in advanced computing infrastructure for the U.S. The dollar value of the allocation is estimated from the NSF awards supporting the allocated resources.
2018 TACC Dell/Intel Knights Landing, Skylake System (Stampede2) 149,000.0 Node Hours
2018 TACC Long-term tape Archival Storage (Ranch) 1,000.0 GB
The estimated value of these awarded resources is $38,730.40. The allocation of these resources represents a considerable investment by the NSF in advanced computing infrastructure for the U.S. The dollar value of the allocation is estimated from the NSF awards supporting the allocated resources.
Click to show/hide prior allocations »
2016 TACC Dell PowerEdge C8220 Cluster with Intel Xeon Phi coprocessors (Stampede) 742,684.487 Core-hours
2016 TACC Dell/Intel Knights Landing, Skylake System (Stampede2) 69,161.0 Node Hours
2016 TACC HP/NVIDIA Interactive Visualization and Data Analytics System (Maverick) 29,709.0 SUs
2016 TACC Long-term tape Archival Storage (Ranch) 500.0 GB
The estimated value of these awarded resources is $30,308.44. The allocation of these resources represents a considerable investment by the NSF in advanced computing infrastructure for the U.S. The dollar value of the allocation is estimated from the NSF awards supporting the allocated resources.

Other Titles:

There are no prior titles for this project.