A public turbulence database cluster and applications to study Lagrangian evolution of velocity increments in turbulence

Abstract

The JHU turbulence database [1] can be used with a state of the art visualisation tool [2] to generate high quality fluid dynamics videos. In this work we investigate the classical idea that smaller structures in turbulent flows, while engaged in their own internal dynamics, are advected by the larger structures. They are not advected undistorted, however. We see instead that the small scale structures are sheared and twisted by the larger scales. This illuminates the basic mechanisms of the turbulent cascade. THE JHU TURBULENCE DATABASEIn [1] a database containing a solution of the 3D incompressible Navier-Stokes (NS) equations is presented. The equations were solved numerically with a standard pseudo-spectral simulation in a periodic domain, using a real space grid of 1024 3 grid points. A large-scale body force drives a turbulent flow with a Taylor microscale based Reynolds number R λ = 433. Out of this solution, 1024 snapshots were stored, spread out evenly over a large eddy turnover time. More on the simulation and on accessing the data can be found at http://turbulence.pha.jhu.edu. In practical terms, we have easy access to the turbulent velocity field and pressure at every point in space and time. VORTICES WITHIN VORTICESOne usual way of visualising a turbulent velocity field is to plot vorticity isosurfaces -see for instance the plots from [3]. The resulting pictures are usually very "crowded", in the sense that there are many intertwined thin vortex tubes, generating an extremely complex structure. In fact, the picture of the entire dataset from [3] looks extremely noisy and it is arguably not very informative about the turbulent dynamics.In this work, we follow a different approach. First of all, we use the alternate quantityfirst introduced in [4]. Secondly, the tool being used has the option of displaying data only inside clearly defined domains of 3D space. We can exploit this facility to investigate the multiscale character of the turbulent cascade. Because vorticity is dominated by the smallest available scales in the velocity, we can visualize vorticity at scale ℓ by the curl of the velocity box-filtered at scale ℓ. We follow a simple procedure:• we filter the velocity field, using a box filter of size ℓ 1 , and we generate semitransparent surfaces delimitating the domains D 1 where Q > q 1 ;• we filter the velocity field, using a box filter of size ℓ 2 < ℓ 1 , and we generate surfaces delimitating the domains D 2 where Q ≥ q 2 , but only if these domains are contained in one of the domains from D 1 ;and this procedure can be used iteratively with several scales (we use at most 3 scales, since the images become too complex for more levels). Additionally, we wish sometimes to keep track of the relative orientation of the vorticity vectors at the different scales. For this purpose we employ a special coloring scheme for the Q isosurfaces: for each point of the surface, we compute the cosine of the angle α between the ℓ 2 filtered vorticity and the ℓ 1 filtered vorticity: cos α = (∇ × u 1 ) · (∇ × u ...

Flight–crash events in turbulence

“…The turbulence Reynolds numbers were R λ = 115 and 170. Additionally, we used the flow field at R λ = 430 made available from the Johns Hopkins University database (33).…”

Anisotropic character of low-order turbulent flow descriptions through the proper orthogonal decomposition

“…The reader is referred to the documentation provided by JHU and summarized in Graham et al [32] (see also Refs. [33,34]). Through investigation of several sets of data, focus is placed on interpretation the physics presented through POD and anisotropy invariant analyses, rather than a detailed exploration of each turbulent flow.…”

New insights into the fine-scale structure of turbulence

“…By a sophisticated experimental set-up, they were able to obtain high-fidelity, volumetric and timeresolved measurements of the velocity gradient and pressure Hessian fields which give novel insights into the statistical signatures of the small-scale structure of turbulence. All of their results are furthermore carefully compared to publicly available direct numerical simulation (DNS) data from the Johns Hopkins turbulence database (Li et al 2008). Finally, they use their results to confirm a recent closure theory for the pressure Hessian introduced by Wilczek & Meneveau (2014) and furthermore make suggestions for future improvements.…”