CFD Basics and ANSYS FLUENT Guide
Uploaded by
ScribdTranslationsCFD Basics and ANSYS FLUENT Guide
Uploaded by
ScribdTranslationsBasic concepts of CFD and the use of ANSYS FLUENT in applications
Engineering: A Brief Introduction
Fluid dynamics
The dynamics of fluids is a field that is both interesting and stimulating for study and research. In the world
In reality, fluid is almost everywhere present. Therefore, we need a fluid dynamics to describe or
model these fluid flows.
In short, any fluid flow can be resolved/described by three basic physical laws or by three equations.
The continuity equation - Mass is conserved.
(2) Momentum equation (widely known as the Navier-Stokes equation) - Second law
of Newton.
Energy equation - Energy is conserved.
Three ways to solve problems
Any problem/phenomenon can be analyzed in three ways.
• Analytical / theoretical approach - using laws / theories and associated equations, such as the use of laws
Newtonian viscosity to solve a fluid flow problem, these solutions are exact.
Experimental approach - conducting experiments and trying to understand the phenomena and the relationships between
various variables, such as wind tunnel experiments that help to design and optimize the external shape of
planes, ships, cars, etc.
• Numerical approach - Solve a fluid flow problem using numerical techniques. These solutions
are approximate, not exact.
What is CFD?
Computational fluid dynamics (CFD) is a set of numerical methods applied to obtain
approximate solutions to fluid dynamics and heat transfer problems.
Thus, CFD is not a science in itself, but a way to apply the methods of a
discipline (numerical analysis) to another (fluid transfer / mass transfer and heat transfer).
Why CFD
CFD solutions are not exact, so why not use the analytical approach ???
The experimental approach is more reliable, like wind tunnel experiments, why not use an approach
experimental rather than CFD???
CFD is used because many engineering problems cannot be solved by an analytical approach or
experimental, where it is difficult to use an analytical or experimental approach.
Difficulties in other approaches
• Theoretical approach: this approach provides an exact solution, which is a significant advantage. But the solutions
analytics are only possible for a limited number of problems, usually formulated artificially and
idealized.
• Experimental approach: These approaches are reliable and describe real situations. For example, in the
aerospace industries, wind tunnel tests are very reliable. But sometimes, they are very expensive and, sometimes,
they also pose technical problems (it sometimes takes several years before an experiment is
configured and that all technical problems are resolved).
Advantage of the CFD
- Detailed analysis of the flow.
- It is applicable to a wide variety of situations.
- Interaction of various types of flows.
- Allows for the highlighting of issues related to flows.
- Make possible, quickly and cheaply, the parametric study and optimization of a given problem.
- Simulation-based conception instead of 'build & test'
More profitable and faster than the EFD
CFD provides a high-fidelity database for flow field diagnostics.
- Simulation of physical flow phenomena that are difficult to experiment with
Large-scale simulations (e.g. ships, airplanes)
Environmental effects (wind, weather, etc.)
*Risques (ex. explosions, rayonnements, pollution)
Physics (e.g., planetary boundary layer, stellar evolution)
CFD Application
The CFD is a very powerful technique that covers a wide range of industrial and non-industrial applications.
Some examples are -
Aerodynamics of airplanes and vehicles: Drag & Lift
Power plant: Combustion in internal combustion engines and gas turbines.
Hydrodynamics of ships
Biomedical engineering: blood flow in arteries and veins
Environmental engineering: distribution of pollutants
In short, CFD is applied in almost all engineering disciplines. Since the 1960s, industries
aerospace has integrated CFD techniques into design, research and development as well as in
manufacturing of aircraft engines and jet engines.
Proprietary code vs Commercial CFD codes
CFD is the resolution of the basic equations of fluid flow and heat transfer by applying
digital techniques. There are two different approaches -
(1) Write your own CFD code to solve a specific problem or type of problem. For example, you
Can you write your own MATLAB program to solve an incompressible pipe flow.
(2) There are commercial CFD software. Some of them are very useful and popular. Example: ANSYS FLUENT,
STAR-CD, CFX, OpenFOAM and COMSOL.
Problem solving using commercial CFD packages
CFD codes are structured around numerical algorithms capable of solving flow problems.
fluids. In order to facilitate access to their resolution power, all commercial CFD packages include
sophisticated user interfaces to enter the problem parameters and examine the results. Therefore,
All codes contain three main elements: (i) a pre-processor, (ii) a solver, and (iii) a post-processor.
Pre-processor: This step includes the creation of geometries and meshes.
Solver: This part consists of numerically solving the fluid flow equations in the domain.
computer science.
Post-processor: In this step, the result of the simulation is analyzed or represented in a useful form.
In ANSYS WORKBENCH, Design Modeler & Meshing acts as a pre-processor, FLUENT is the solver, and CFD-
post is the post-processor.
There are three basic methods for solving the CFD problem.
Method of finite differences: Approximation of derivatives using Taylor series expansion...
(2) Finite element method: The physical problem is divided into a finite set of elements. On each
element is solved the <<weak>> or <<variational>> formulation of the NS equations. From the solutions on
For each element, we reconstruct the general solution.
(3) The finite volume method is a particular case of the finite difference method. It is a very
popular: The physical problem is subdivided into a finite set of volumes (called control volume). Then on
each VC, the equations (in their integral form) are solved. From each solution on the VC, one
rebuild the solution to the initial problem.
ANSYS FLUENT
ANSYS, Inc. is a developer of technical simulation software (computer-aided engineering or CAE) whose
The headquarters is located south of Pittsburgh, in the Southpointe business park, in Cecil Township, Pennsylvania.
in the United States. Ansys CFD, a proprietary computational fluid dynamics (CFD) program, is one of its
most important products (Source-Wikipedia).
ANSYS FLUENT is one of the most popular CFD software. We will use ANSYS FLUENT in this course to
solve engineering problems.
ANSYS FLUENT CFD Solver is based on the finite volume method
The domain is discretized into a finite number of control volumes.
General conservation (transport) equations for mass, momentum, energy, species, etc. are solved.
on this set of control volumes
Problem solving steps by ANSYS FLUENT
To solve engineering problems using ANSYS FLUENT, the necessary steps are:
(1) pre-analysis
(2) geometry
stitch
physical configuration
(5) Numerical solution
verification and validation
ANSYS WORKBENCH
ANSYS WORKBENCH is a graphical user interface that allows the user to use all the tools/software.
required from a single location. You can access the preprocessor, solver, and post-processor tools.
from ANSYS WORKBENCH. You will control your workflow between different tools through Workbench.
Steps for your simulation
Pre-analysis: you want to numerically solve a real technical problem using ANSYS FLUENT. Then,
you will need boundary conditions. You will observe the actual physical situation or you will obtain
data to be able to represent the real case. You will also obtain theoretical or experimental results
correspondents to compare your simulation. These steps are important, especially to achieve the condition
of the correct limit. For example, if you want to simulate airflow over the wing of a commercial aircraft, you
you should know the speed range of this wing, you will also need to look for the experimental results or
previous digital ones, as well as relevant theories.
Steps for your simulation
• Geometry: you need to do the geometry. You can use the ANSYS software, which you can use from
ANSYS WORKBENCH. You can also use any other CAD software you like, such as AutoCAD,
Solidworks, CATIA, Autocad Inventor, etc.
• Meshing: meshing is one of the most important steps in your simulation. The results of the simulation
depend on the quality of the mesh. Poor quality can produce a poor simulation result, or even a
divergence.
These steps are pre-possess. In this course, you no longer have to manage Geometry & Mesh. These will be provided.
so that you can proceed to the next steps.
Steps for your simulation
• Physical configuration: This is done in the ANSYS solver. Your focus will be on understanding and performing.
the physical configuration, the numerical result, and the verification and validation.
In the physical installation phase, you provide data on the accuracy of the solution, the boundary conditions,
the physics involved, the materials involved, the properties of the parts involved, etc. In summary, you are describing
numerically simulate the actual situation.
I noticed well:
MODELING
Numerical modeling is the transcription of the physical phenomenon into computer language.
D I S C R E T I Z AT I O N
The numerical model of the structure being composed of an infinite number of points, it is impossible to
solve the partial differential equations, even numerically. The structure is therefore
cut into a finite number of points which translates into a decomposition of geometry into shape
Simple. The set of these nodes and elements constitutes the mesh of the model.
NUMERICALSIMULATION
The code solver used approximates the equations at each node of the mesh.
by respecting the fundamental principles of physics (conservation of mass and energy).