Multi-domain parallel computation of wake flows

Osawa, Yasuo; Kalro, V.; Tezduyar, Tayfun E.

doi:10.1016/s0045-7825(98)00305-3

Cited by 27 publications

(10 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Remark The data projection approach presented here is very similar to that described in. 79 However, in the present work, the step where the velocity is projected onto the inlet nodal space is not needed since the quadrature-point values are used directly in the enforcement of weak BCs.…”

Section: Domain Coupling and Data Transfermentioning

confidence: 99%

“…36,37,38,39,40,41 Both VMS formulations proved their accuracy and efficiency through numerous high-Reynolds-number flow computations for applications such as parachutes, 42,43,44,45,46,47,48,49 turbomachinery, 50,51,52,53,54 thermo-fluids, 55,56 flows with topology change, 57,58,59,60,61,62,63,64 wind turbines, 65,66,67,68,69 hydro turbines, 70 bioinspired propulsion, 71,72,73 cardiovascular hemodynamics, 74,75,76,77 and bridge aerodynamics. 21,78 To simulate the incident turbulence, a Multi-Domain Modeling (MDM) approach 79,80,81,82,68 is employed, where th...…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Modeling and simulation of bridge-section buffeting response in turbulent flow

Helgedagsrud

Bazilevs

Mathisen

et al. 2019

Math. Models Methods Appl. Sci.

View full text Add to dashboard Cite

Buffeting analysis plays an important role in the wind-resistant design of long-span bridges. While computational methods have been widely used in the study of self-excited forces on bridge sections, there is very little work on applying advanced simulation to buffeting analysis. In an effort to address this shortcoming, we developed a framework for the buffeting simulation of bridge sections subjected to turbulent flows. We carry out simulations of a rectangular bridge section with aspect ratio 10 and compute its aerodynamic admittance functions. The simulations show good agreement with airfoil theory and experimental observations. It was found that inflow turbulence plays an important role in obtaining accurate wind loads on the bridge sections. The proposed methodology is envisioned to have practical impact in wind engineering of structures in the future.

show abstract

Section: Domain Coupling and Data Transfermentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Modeling and simulation of bridge-section buffeting response in turbulent flow

Helgedagsrud

Bazilevs

Mathisen

et al. 2019

Math. Models Methods Appl. Sci.

View full text Add to dashboard Cite

show abstract

“…In fact computations over different subdomains can even be carried out on different computational platforms (essentially) in parallel, because the data transfer between the different platforms would just be a 2D data exchange and therefore would not add a significant communication cost. For further details on the MDM see [8].…”

Section: The Multi-domain Methods For Wake Computationsmentioning

confidence: 99%

“…The Multi-Domain Method (MDM) [7,8] developed recently was originally intended for parallel 3D computation of the unsteady wake flow in the far downstream of an object and the influence of this wake on secondary objects located far downstream. Also recently, a computational technique [9] largely based on the MDM was designed for simulation of unsteady wake flows behind a circular cylinder, including cases with computational domains extending far downstream.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

3D simulation and visualization of unsteady wake flow behind a cylinder

Osawa

Tezduyar

1999

J Vis

Self Cite

View full text Add to dashboard Cite

In this paper we focus on 3D simulation of unsteady wake flow behind a circular cylinder.We show that in addition to accurate formulations and sufficiently-refined meshes, efficient computing methods are essential components of an effective simulation strategy. We use the Multi-Domain Method (MDM) we developed recently in computation of two cases. At Reynolds number 300, we demonstrate how the MDM enables us to use highly-refined meshes to capture wake patterns which we otherwise cannot fully represent. At Reynolds number 140, we show that with the MDM we can extend our computations sufficiently downstream, and with sufficient accuracy, to successfully capture the second phase of the Karman vortex street, which has been observed in laboratory experiments, and which has double the spacing between the vortices compared to the first phase.

show abstract