Environmental Engineering Reference
In-Depth Information
6
Environmental Applications
of Computational Fluid
Dynamics
N.G. Wright
1
and D.M. Hargreaves
2
1
School of Civil Engineering, University of Leeds, UK
2
Faculty of Engineering, University of Nottingham, UK
•
Computational fluid-dynamic simulations can be
carried out in situations where a real-life simulation is
impossible, such as dam failures or the release of toxic
substances into the natural environment.
6.1 Introduction
Computational fluid dynamics (CFD) has been in use in
various fields of science and engineering for over 40 years.
Aeronautics and aerospace were initially themain fields of
application, but it has much to offer in other fields if due
consideration is given to their particular requirements. In
particular, CFD offers:
As with any engineering analysis tool, the advantages
must be considered alongside caution in its application.
Any CFD study must consider the following:
•
The results output by a CFD package are not necessarily
a valid representation for a particular fluid-flow prob-
lem. Knowledge and experience of fluid mechanics
is needed to evaluate the results critically. Further-
more, a detailed understanding of how a CFD code
works is needed to evaluate whether it has been applied
appropriately in a given situation.
•
There are several guidelines available on the validation
of CFD simulations (ASME, 1993; AIAA, 1998; Casey
and Wintergerste, 2000) in an engineering context
and these should be carefully studied (especially in
view of the comments on skills above). Flows in the
natural environment have additional complexity due to
uncertainties over process representation across widely
varying temporal and spatial scales and uncertainties in
the input parameters to many models. In some cases
there may be no data available to validate or calibrate
the CFD model and in this situation care must be
•
Full-scale simulation as opposed to the model scale
of many physical simulations such as flumes or wind
tunnels. In environmental applications this difference
can be of vital importance as the domain of interest
maybeofconsiderablesize.
•
Interpolation between measured data. Measured data
are often both temporally and spatially sparse. As long
as sufficient data are available, they can be used to
calibrate a CFD model that provides data at many time
and space points.
•
Excellent visualization through computer graphics.
•
Repeatability: a CFD simulation can be run again and
again with the same parameters or with many varia-
tions in parameters. As computer power continues to
develop, it opens the route to automated optimization
and parameter-uncertainty studies for more complex
problems.
Search WWH ::
Custom Search