Free Academic Seminars And Projects Reports

Full Version: APPLICATION OF CFD IN MECHANICALENGINEERING
You're currently viewing a stripped down version of our content. View the full version with proper formatting.

APPLICATION OF CFD IN MECHANICALENGINEERING
PRESENTED BY

Binu k s
99218
S8 mechanical

DEPARTMENT OF MECHANICAL ENGINEERING

GOVERNMENT ENGINEERING COLLEGE


THRISSUR


[attachment=8116]


Abstract


The steady improvement in the speed of computers and the memory size since the 1950 s has led to the emergence of computational fluid dynamics (CFD). This branch of fluid dynamics, complements experimental and theoretical fluid dynamics by providing an alternative cost-effective means of simulating real flows. This new methodology of solving a flow problem using a computer is given the name CFD. There are three methods for analyzing fluid flow problems. Experimental, Theoretical, Computational (CFD)

This paper deals with application of computational fluid dynamics in mechanical engineering
Computational Fluid Dynamics or CFD can be described, as the use of computers to analyse systems involving fluid flow, heat transfer and associated phenomena such as chemical reactions based on numerical approach. In this numerical approach the equations (usually partial differential in form) govern a process of interest are solved numerically. The technique is very powerful and spans over a wide range of industrial and non-industrial application areas.
CFD embraces a variety of technologies including mathematics, computer science, engineering and physics and these disciplines have to be brought together to provide the means of modelling fluid flows. Such modelling is used in many fields of science and engineering but, if it is to be useful, the results that it yields must be a realistic simulation of a fluid in motion. At present this depends on the problem being simulated, the software being used and the skill of the user.
Until recently the user of CFD has been a specialist, probably trained to doctoral level, working in a research and development department. Now, however, the technology is more widely available both in industry and academia and so it is being used to provide insights into many aspects of fluid motion.

Some important application areas for CFD are:
Hydrodynamics of ships
aerodynamics of aircraft and vehicles (Lift and Drag).
Flow over missiles (Lift, Drag and side force data)
Power plant, Combustion in IC engines and gas turbines Turbo-machinery: Flows inside rotating passages, diffusers etc.
Electrical and electronic engineering: cooling of equipment including microcircuits.
This paper also deals with an application in detail.

computational Fluid Dynamics (CFD)

1.0 Introduction:
The steady improvement in the speed of computers and the memory size since the 1950 s has led to the emergence of computational fluid dynamics (CFD). This branch of fluid dynamics, complements experimental and theoretical fluid dynamics by providing an alternative cost-effective means of simulating real flows. This new methodology of solving a flow problem using a computer is given the name CFD.

How can a fluid flow problem be analysed?
There are three methods.
(i) Experimental: Experimental fluid dynamics has played an important role in validating and delineating the limits of the various approximations to the governing equations.
(ii) Theoretical: This approach is generally clean, and normally available in formula form mostly applicable to linear problems.
(ii) Computational (CFD)

2. 0 What is CFD?
Computational Fluid Dynamics or CFD can be described, as the use of computers to analyse systems involving fluid flow, heat transfer and associated phenomena such as chemical reactions based on numerical approach. In this numerical approach the equations (usually partial differential in form) govern a process of interest are solved numerically. The technique is very powerful and spans over a wide range of industrial and non-industrial application areas.
CFD embraces a variety of technologies including mathematics, computer science, engineering and physics and these disciplines have to be brought together to provide the means of modelling fluid flows. Such modelling is used in many fields of science and engineering but, if it is to be useful, the results that it yields must be a realistic simulation of a fluid in motion. At present this depends on the problem being simulated, the software being used and the skill of the user.
Until recently the user of CFD has been a specialist, probably trained to doctoral level, working in a research and development department. Now, however, the technology is more widely available both in industry and academia and so it is being used to provide insights into many aspects of fluid motion. This increasing use has come about as there are now numerous commercial CFD software packages on the market and so it is not necessary for users to have to write their own programs in order to obtain flow results. Whilst the software is widely available, the means of learning about CFD and how to produce simulations with it tend to be restricted to post-experience courses in universities and polytechnics, where the level of assumed knowledge can be too great, or to courses run by software suppliers where users are shown how to run a particular software product. Also, there are several technical texts that describe the detailed mathematics of the modelling process, but these are often far too technical for the user of the software. Consequently, as the variety of users increases there is a need for a general text that is an introductory guide to the analysis of flow problems using CFD and describes the various stages of an analysis that must be undertaken if the user is to obtain sensible results.

3.0 Some important application areas for CFD are:
Aerodynamics of aircraft and vehicles (Lift and Drag): CFD is used in conjunction with wind tunnel tests to determine the performance of various configurations.
Hydrodynamics of ships
Flow over missiles (Lift, Drag and side force data): Simulation performance can be obtained by using CFD tool.
Power plant, Combustion in IC engines and gas turbines (Air flow inside I.C. engines).
Turbo-machinery: Flows inside rotating passages, diffusers etc.
Electrical and electronic engineering: cooling of equipment including microcircuits: In this problem, electrical devices, such as integrated circuits, produce heat. This heat must be dissipated if the equipment is not to become too hot. For example, the hot devices heat the air that surround them and this hot air rises, creating air currents that move the heat away from the sources of heat. If insufficient heat is moved away, then it may be necessary to add fans that will force air over the hot devices.
Jet flow inside nuclear reactor halls: Such problems involve the simulation of fault conditions (like failure of nuclear reactors), which is very difficult to perform in actual experiment. Hence, computation is the only way of trying to understand such flows.
Chemical process engineering: Mixing and separation, polymer moulding
Flames in burners. There is a need to understand the complex interactions between fluid flow and chemical reaction in flames. This can assist in the production of more efficient designs for burners in boilers, furnaces and other heating devices.
External and internal environment of buildings: wind loading and heating/ventilation
Marine engineering: loads on off-shore structures
Environmental engineering (Modelling of distribution of pollutants and effluents in Rivers & Oceans): Various pollutants are discharged into rivers and oceans, and computer programs can be used to predict where pollutants will travel in these naturally occurring flows and what the pollutant concentration will be at given positions in the river or sea.
Hydrology and oceanography: Flows in rivers, estuaries, oceans
Meteorology: weather prediction
Biomedical engineering: Blood flows through arteries and veins

Increasingly CFD is becoming vital component in the design of industrial products and processes. The reasons for its popularity are:
Dramatic increase in computer power
Commercial CFD codes at reasonable cost
Easy to use front-ends in CFD codes for rapid set-up of problems and efficient analysis of the results
Improvements in the physics and number of sub-models available to the user e.g. Thermal radiation, turbulence, soot, and pollution chemistry.
Increased validation studies by code vendors leading to increased confidence in the results

4.0 Scope of CFD:
Classification of Process Industry Applications:
It is divided into three broad categories:
Those involving complicated geometry and simple physics
Those involving simple geometry and complicated physics
Those involving complicated geometry and complicated physics

4.1 Complicated Geometry and Simple Physics:
Air flows in ducts: mixing of liquids using impellers; conjugate (single phase) heat transfer including radiation; cyclone separators
Reasonably accurate CFD solution can be expected in reasonable about time and other resources
Calibration of codes will make it a powerful design tool
Ripe for exploitation

4.2 Simple Geometry and Complicated Physics:
Combustion in burner systems; two-phase flow; boiling heat transfer in a steam generator tube; cavitation; critical flow in two-phase systems.
Successful computation of a solution not guaranteed and accuracy of results to be established.
Case-specific advances possible through research and development, validation and calibration
A consortium-type of approach needed to bring application specific results to the status of Category I

4.3 Complicated Geometry and Complicated Physics:
Erosion/corrosion in furnaces; fluidised bed combustion; distillation; phase-separation devices in gas-liquid flows; system-level simulations
A turnkey-approach to problem solving cannot be expected and the problem has to be broken up into a series of (interconnected) Category-I and Category-II types of simulations
The solution may reflect only general trends
A problem-specific modelling strategy necessary and the problems are best addressed through academic research

5.0 Physical Phenomena Requiring Modelling:
Navier-Stokes equations (+ continuity + energy) have to be supplemented with additional terms and models for a number of cases:
- Non-Newtonian fluid behaviour
- Flows through porous media
- Rotating/moving components
- Turbulence
- Flows involving chemical reactions
- Combustion and allied process, e.g. Natural gas combustion
- Radiative heat transfer
- Multiphase flows

5.1 Non-Newtonian Fluid Behaviour:
For a Newtonian fluid,
- strain rate is directly proportional to stress and hence we have
For many process fluids, such direct proportionality does not hold good and a more complicated non-Newtonian Rheological behaviour is expected.
- One of the simplest non-Newtonian models is the power law fluid:
More complicated behaviour exhibited by visco-elastic fluids ( fluids with memory ) but these are not considered here.
5.2 Flow Through Porous Media:
Flow through sand, packed beds, filter media, tube banks etc.
Characterized by a large frictional pressure drop, negligible acceleration effect and a volume porosity ( )
Generally, pressure gradient velocity given by Darcy law
In the most general case, porosity may be directional as is the resistance coefficient
5.3 Rotating/Moving Components
This situation arises in a number of applications: pumps, turbines, internal combustion engines, mixing vessels etc.
Movement of grid modifications in the transient term and the advection term
- Transient term:
- Advection term:
- The right hand side of the equations remains unchanged

5.4 Special Cases of Grid Motion
Pure rotational motion: Equations solved in a rotating frame of reference:
Sliding mesh method
- Useful in turbines and mixing
- Two grids: one stationary and one rotating with the impeller
- The moving grid slides along the stationary grid along a surface of slip
- Special treatment for control volumes adjacent to the slip surface to compute the convective and diffusive fluxes across it
- Flow calculation transient even for steady flows but formulation requires no empirical input
5.5 Multiphase Flow
Some bulk phenomena and buoyancy effects can be predicted reasonably well
Predictions of fine detail cannot be taken for granted
Modelling of physical phenomena such as bubble formation, coalescence, interaction with wall, bubble-bubble interactions, bubble-induced turbulence enhancement/suppression, bubble dispersion by liquid phase turbulence etc. need to be addressed to capture the fine detail.
Similar attention to detail may be necessary for other flow regimes as well such as slug flow and annular flow
Numerically, obtaining a converged solution may not be straightforward or trivial

6.0 How does a CFD code works?
CFD codes are structured around the numerical algorithms. Nowadays all commercial CFD packages include sophisticated user interfaces to input problem parameters and examine the results. All code contains three elements.
1. A pre-processor
2. The solver
3. The post-processor
The functions of each of these elements within the context of a CFD code are described below:

6.1 Pre-processor:
Pre-processing consists of the input of a flow problem to a CFD program by means of a user-friendly interface and the subsequent transformation of this input into a form suitable for use by the solver. The user activities at the pre-processing stage involves the following steps:
Definition of the geometry of the region of interest: the computational domain
Grid generation - the sub-division of the domain into a number of smaller, non-overlapping sub-domains: a grid (or mesh) of cells (or control volumes or elements).
Selection of the physical and chemical phenomena that need to be modelled.
Definition of fluid properties:
- Normal shear stress or pressure;
- Viscosity, which enables us to find the tangential shear stress (the viscous
Shear stress);
- Density (constant or temperature dependent)
Specification of appropriate boundary conditions at cells, which coincide with or touch the domain boundary).

6.1.1 The Grid:
As already mentioned the grid is nothing but the subdivision of a domain. The grid can be Cartesian or polar, orthogonal or non-orthogonal, structured or unstructured. If you are able to generate the desired grid to your CFD problem 50% of the work is over. In order to maximize productivity of CFD personnel all major codes now include their own CAD-style interface. Otherwise facilities to import data from proprietary surface modelers and mesh generators such as PATRAN and I-DEAS are now available.

6.1.2 Boundary conditions for fluid flow problems:
When solving any system of partial differential equations it is the boundary conditions, together with the initial conditions, that determine the exact solution. The form of the boundary conditions that is required by any partial differential equation depends on the equation itself and the way that it has been discretised. Some common boundary conditions are, however, met when solving fluid flow problems with computers. These can be classified either in terms of the numerical values that have to be set or in terms of the physical type of the boundary condition. One has to specify boundary conditions for the following variables:
For the velocity components, which will affect the momentum equations. These conditions are usually given by specifying the velocity components and if this is not done then the derivatives of the velocity components normal to the boundary is usually zero.
For the pressure and possibly, mass flow, which will influence the continuity equation if a SIMPLE-like algorithm is being used. Usually, the fluid pressure needs to be specified at a minimum of one point in the flow.
For the turbulence variables such as the turbulence kinetic energy (k) and the rate of dissipation of k , i.e. .

These conditions have to be applied at a variety of boundaries such as the following:
Solid walls: Many boundaries within a fluid flow domain will be solid walls, and these can be either stationary or moving walls. If the flow is laminar then the velocity components can be set to be the velocity of the wall. When the flow is turbulent, however, the situation is more complex. This complexity is due to the velocity of a flow varying extremely rapidly near a wall if the flow is turbulent. To capture this rapid variation, which occurs in a direction away from the wall, many grid points would be required in this direction near the wall, and this increases the amount of computational effort required to produce a solution. One way of reducing the effort is to specify the velocity near a solid wall suing experimental data for boundary layers, which shows that the velocity variation should be logarithmic with the distance from the wall at points more than a known distance from the wall. This can be seen in Figure 3.10 where the velocity in the boundary layer is plotted against distance away from the wall. Both the velocity and distance have been transformed into non-dimensional quantities as shown. Looking at the diagram, three regions can be seen. Near the wall there is a viscous sub-layer where the effects of turbulence are damped out by the wall itself. Then there is a log-law region where the velocity is a logarithmic function of the distance from the wall. Finally, there is an outer layer, which is where the boundary layer and the external flow merge. If the mesh is built so that the first point where the velocity is calculated is in the log-law region, then the very rapid variation near the wall will not need to be modelled. Similar methods can be used to specify the values of both the turbulence variables k and .
Inlets: At an inlet, fluid enters the domain and so the fluid velocity might well be known for the problem being simulated. In some programs the pressure equation can only be solved if the mass flow at an inlet is known. Also, the fluid carries with it other quantities such as k and and so these must be specified as well. We say that variables are convected into the domain.
Outlets: Where the fluid leaves the domain is known as an outlet. Normally, the pressure is set to zero at an outlet and the velocity components and any turbulence variables are left to find their own values, which will have a zero spatial derivative in a direction normal to the boundary. If the flow is swirling through the outlet then a pressure gradient is required to provide the necessary centripetal force to the fluid and so a constant pressure boundary condition will be invalid To overcome this, iterative procedures are used which start by specifying a constant pressure at the outlet but then try to find the pressure that matches the velocity of the swirling flow.
Symmetry boundaries: When the flow is symmetrical about some plane there is no flow through the boundary and the derivatives of the variables normal to the boundary are zero.
Cyclic or periodic boundaries: These boundaries come in pairs and are used to specify that the flow have the same values of the variables at equivalent positions on both of the boundaries.

6.2 The Solver:
The following are the distinct streams of numerical solution techniques available.
i) Finite difference
ii) Finite element
ii) Spectral method
iv) Finite volume method
The major differences between the above separate streams are associated with the way in which the flow variables are approximated and with the discretization process.
Finite Difference Methods:
These methods describe the unknown of the flow problems by means of point samples at the node points of a grid of coordinate lines. A schematic of the finite difference method is shown below:



finite element methods:
These methods use simple piecewise functions (e.g. linear or quadratic) valid on elements to describe the local variables of unknown flow variables .

Spectral method:
These methods approximate the unknowns by means of truncated Fourier series or series of Chebyshev polynomials.

The finite volume method:
The finite volume method was originally developed as a special finite difference formulation.

Integration of governing equations of fluid flow, over all (finite) control volume is carried-out over the domain to obtain the solution. The control volume integration, distinguish the finite volume method from all other CFD techniques.

In outline the numerical methods that form the basis of the solver perform the following steps:

Approximation of the unknown flow variables by means of simple functions.
Discretization by substitution of the approximations into the governing flow equations and subsequent mathematical manipulations.
Solution of the algebraic equations.

6.3 The Post Processor:
As in pre-processing a huge amount of development work has recently taken place in the post-processing field. Owing to the increased popularity of engineering workstations, many of which have outstanding graphics capabilities, the leading CFD packages are now equipped with versatile data visualization tools. These include:

Domain geometry and grid display
Vector plots
Line and shaded contour plots
2D and 3D surface plots
Particle tracking
View manipulation (translation, rotation, scaling etc.)
Colour postscript output
More recently these facilities also include animation for dynamic result display and in addition to graphics all codes produce trusty alphanumeric output and have data export facilities for further manipulation external to the code. As in many other branches of CAE the graphics output capabilities of CFD codes have revolutionised the communication of ideas to the non-specialist.

7.0 Conceptual Relationship between Consistency, Stability and convergence:

8.0 Note of caution and limitation of CFD:
In solving fluid flow problems we need to be aware of the two things:
(i) The results produced by a CFD code are at best as good as the physics (and chemistry) embedded in it and,
(ii) At worst as good as its operator

The user of a CFD code must have following skills:
(i) One must have the skills of identification of the physical and chemical phenomena that need to be considered.
(ii) To decide whether problem is 2D or 3D in nature.
(ii) Understanding of physical sub-models like turbulence and chemical kinetics and make the right choices of selecting good sub models.

7.1 Limitations of CFD:
A number of approximations are made in modelling physical phenomena.
For industrial flows, tests for goodness of prediction not practical in many cases
- Too large flow domains for systematic grid refinement study
- Too long calculation times for systematic parametric study
- No reliable data for systematic validation
Many geometrical details are not modelled
Idealities are imposed in formulation

9.0 Case study for Electronic chip cooling using a CFD code:

9.1 Problem Description:
The problem to be considered is shown schematically in Figure below. The configuration consists of a series of side-by-side electronic chips, or modules, mounted on a circuit board. Airflow, confined between the circuit board and an upper wall, cools the modules. To take advantage of the symmetry present in the problem, the model will extend from the middle of one module to the plane of symmetry between it and the next module. As shown in the figure, each half-module is assumed to generate 2.0 Watts and to have a bulk conductivity of 1.0 W/m2-K. The circuit board conductivity is assumed to be one order of magnitude lower: 0.1 W/m2-K. The airflow enters the system at 298 K with a velocity of 1 m/s. The Reynolds number of the flow, based on the module height, is about 600. The flow is therefore treated as laminar.

Application of cfd in analysing Flow over missiles

REFERENCES:
1) COMPUTER SIMULATION OF FLOW AND HEAT TRANSFER BY P.S. GOSHDASTIDAR

2) FLIUD MECHANICS BY STEPHEN M RICHARDSON

3) COMPUTER TECHNIQUES FOR FLUID DYNAMICS VOL:1 & 2 BY C.A.J FLETCHER

4) NUMERICAL HEAT TRANSFER AND FLUID FLOW BY SUHAS V & PATANKER

5) COMPUTATIONAL FLUID DYNAMICS FOR ENGINEERS VOL:1 & 2 BY KLAUS A HOFFMAN & STEVE T CHIANG

6) COMPUTATIONAL FLUID MECHANICS AND HEAT TRANSFER BY DALE A ANDERSON & JOHN C TAMEHILL

7) DATA VALIDATION OF C.F.D CODES BY D GOLDSTIEN,D HUGHES, R JOHNSON

8) WW.CFDONLINE.COM

9) WW.HOWSTUFFSWORK.COM

10) WW.MEDIA.MIT.EDU

11) WW.NAS.NASA.GOV
COMPUTATIONAL FLUID DYNAMICS
PRESENTED BY

BINU.K.S 99218
mech-A
DEPARTMENT OF MECHANICAL ENGINEERING

[attachment=8118]

WHAT IS CFD
COMPUTATIONAL FLUID DYNAMICS OR CFD IS
THE ANALYSIS OF SYSTEMS INVOLVING FLUID
FLOW , HEAT TRANSFER AND ASSOCIATED
PHENOMINA SUCH AS CHEMICAL REACTIONS
BY MEANS OF COMPUTER BASED NUMERICAL
APPROACH .

IN THIS CFD APPROACH , THE EQUATIONS
( USUALLY PARTIAL DIFFERENTIAL EQUATIONS )
THAT GOVERN A PROCESS OF INTEREST ARE
SOLVED NUMERICALLY.

TYPES OF FLUID FLOWS
VISCOUS OR INVISCID FLOWS

INCOMPRESSIBLE OR COMPRESSIBLE FLOWS IN PIPES OR OPEN CHANNELS

FLOWS IN PIPES AND TURBINES

WATER WAVES

HOW CAN A FLUID FLOW
PROBLEM BE ANALYSED?

EXPERIMENTAL

THEORETICAL

COMPUTATIONAL