China
Africa
Explore chapters and articles related to this topic
Design tools
Published in G.J.C.M. Hoffmans, H.J. Verheij, Scour Manual, 2021
G.J.C.M. Hoffmans, H.J. Verheij
LES is a mathematical model for turbulence used in computational fluid mechanics and is currently applied in a wide variety of engineering applications. The simulation of turbulent flows by numerically solving the Navier–Stokes equations requires resolving a very wide range of time and length scales, all of which affect the flow field.
Simulation of three-dimensional coherent and turbulent motions around a rectangular pile in an open channel flow
Published in Zhao-Yin Wang, Shi-Xiong Hu, Stochastic Hydraulics 2000, 2020
Compared with the k-ε model, the Large Eddy Simulation (LES) model is distinct in the way it handles the turbulence averaging process. In the last decade, it has been receiving increasing attention from researchers all over the world due to its capability of providing large scale chaotic flow pattern directly with moderate computational cost. For example, Shah and Ferziger (1997) used a LES model to study wind passing a cubic obstacle. Rodi (1997) made comparisons between the LES models and other turbulence models for the same problem. It is noted that no free surface is present in the above studies. On the other hand, Hodges and Street (1999) proposed a LES model for turbulent free surface flows. Li and Wang (1999) also developed a LES model to study the turbulence characteristics of a free surface shallow water flow.
On Computational Heat Transfer Procedures for Heat Exchangers in Single-Phase Flow Operation
Published in W. J. Minkowycz, E. M. Sparrow, J. P. Abraham, J. M. Gorman, Advances in Numerical Heat Transfer, 2017
In the LES model, time-dependent flow equations are solved for the mean flow and the largest eddies while the effects of the smaller eddies are modeled. The LES model is expected to emerge as the future model for industrial applications but it still limited to relatively low Reynolds numbers and simple geometries. Handling wall-bounded flows focusing on near-wall phenomena such as heat and mass transfer and shear at high Reynolds numbers present a problem due to near-wall resolution requirements. Complex wall topologies also present problems for LES.
Modelling marine turbine arrays in tidal flows
Published in Journal of Hydraulic Research, 2022
The accuracy of the predictions with LES relies on the adopted turbulent inflow boundary conditions, including turbulence length scale, turbulence intensity and vertical distribution of mean velocities. These can be hard to measure in the field with acoustic Doppler current profilers (ADCPs) due to the required high spatial and temporal resolution and logistical constraints. Proper uncertainty quantification can be of significant value in the prescription of turbulence at the inflow boundary conditions to provide more robust estimates of turbine array energy yield.
A Priori Sub-grid Modelling Using Artificial Neural Networks
Published in International Journal of Computational Fluid Dynamics, 2020
Alvaro Prat, Theophile Sautory, S. Navarro-Martinez
One of the most common approaches to describe turbulent flows is Large Eddy Simulations (LES), where the large scales are solved directly and the contribution of the unresolved small scales (the sub-grid or sub-filter scales) is modelled. LES is used widely in research environments, and it is extending its application to industrial environments (Löhner 2019). LES requires (at least) closures for the sub-grid stress tensor. There has been a large effort developing sub-grid models over the last three decades; from the classic Smagorinsky model (1963), to dynamic calibration of the model (Germano et al. 1991), scale-similarity models (Bardina and Ferziger Reynolds 1983) among others. The performance of the models depends on the type flow; in simple flows the Smagorinsky models tend to dissipate too much energy and do not allow energy back transfer from the small to the large scales. The dynamic model behaves well if the grid is resolved enough and local homogeneous directions can be found. These models are based on local equilibrium of sub-grid production and dissipation of sub-grid kinetic energy. Non-equilibrium modifications exist (one equation models) as well as a myriad of modifications (Piomelli 2014). Alternatively MILES approaches (Boris et al. 1992) use the dissipation of the numerical scheme as an intrinsic sub-grid model. The inherent numerical dissipation of certain schemes, can mask the behaviour of the sub-grid model (Garnier et al. 1999 Aug.). Similarly, high-order non-dissipative schemes require explicit filtering to maintain stability (by removing high frequency numerical noise), which can also interfered with the sub-grid model (Dairay et al. 2017). Sub-grid stress models developed in incompressible flows are often translated other flows, such as reacting flows, MHD, etc. However, the validity of these closures (and assumptions behind it) in other flows is often overlooked (Klein et al. 2015).
Experimental study and large Eddy simulation of a coaxial jet with perforated obstacles to control thermal mixing characteristics
Published in Experimental Heat Transfer, 2018
Besir Kok, Yasin Varol, Hüseyin Ayhan, Hakan F. Oztop
LES is one of the most widely used turbulence model in computational fluid dynamics applications. The main idea of LES model is to reduce computational cost by reducing the range of time and length–scales that are being solved for filtered Navier–Stokes equations. Filtering operation removes small-scale information from the numerical solution.