Section astrophysics, chemical manufacturing and power generators.

Section A: Introduction
to CFD

Existence
of modern technologies had eased the application of science for humans. With
progression of computational performance, Computational Fluid Dynamics (CFD)
software was introduced to the modern society to study on flow of fluid without
any physical subject to be tested on. This software was implemented to a wide
variety of industries which includes automotive, aerospace, astrophysics,
chemical manufacturing and power generators. Before the introduction of CFD,
researchers had to rely solely on their knowledge of mathematics and fluid
mechanics to estimate an outcome of fluid flows. While the software is of aid
to modern society, CFD would need a computer to process its simulation and data
retrieval of the desired simulation. Therefore, a substantial amount of
computational performance was needed to ensure the data was processed quickly
and accurately to meet current industries standard of efficiency.

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!


order now

CFD
software is a tool where knowledge of fluid mechanics, mathematics and computer
science were applied in simulating a fluid flow motion. These motions are
derived by mathematical equations and represented in computer programming
language only readable by computers. The results were then converted to quantifiable
data that is displayed in readable human language. During the simulation in
CFD, all progress was predicted using numerical data as from derivation from
all available mathematical equations in fluid dynamic flow.

While
fluid is solved in mathematical solutions, there are still fundamental steps
required to be obeyed. (Fawehinmi et al., 2005). There are three
methods in this problem solving through Experimental, Analytical and
Computational solving where data are validated. Back when there is no
computational aid such as CFD, Experimental solving uses a more tedious
technique which is to build a scaled down replica model representing the real
object of study and the flow properties are to be measured and recorded. This
technique is frequently used, where experimental data were all recorded, and
the results were compared to theoretical values derived from mathematical
equations from fluid dynamics. It is different from Analytical solving as
analytical does not require any physical model to obtain theoretical data. The
method involved for analytical is by comparing measurements of the desired object
of study in mathematical modelling to related equations from equations of fluid
dynamics. But Analytical solving applies to a limited number of simplified flow
geometries. There were recorded errors from the above methods of solving which
leads to search for improvement in error reduction while accurately obtain
results subjected to fluid dynamics.

Through
years of testing and perfecting, CFD software is now able to display results
accurately while having little to no error. It applies to any fluid flow, be it
simple or complex fluid flow. With such accuracy, computational solving method
had to be ideal choice of studying fluid flow problems as the results were more
reliable albeit the other methods, Experimental and Analytical solving approach
were still being used.

Fluid
flows are controlled and influenced by partial differential equations that is
represented by laws of Conservation of Mass, Momentum and Energy. CFD uses the
above governing equations and below represented by following Navier-Stokes equations
based from conservation laws:

a)     Conservation
of Mass (from continuity equation)

 

b)     Conservation
of Momentum (from Momentum equation)

 

Where;

 

i:    Local change of time

ii:   Momentum convection

iii:  Surface force

iv:  Molecular-dependent momentum exchange
(diffusion)

v:   Mass force

 

c)     Conservation
of Energy (from Energy equation)

 

 

Where;

i:    Local energy change with time

ii:   Convection term

iii:  Pressure work

iv:  Heat flux (diffusion)

v:   Irreversible transfer of mechanical energy into
heat

When all the
conservation equations were applied to Navier-Stokes equation, the following
simplified general form is formed:

 

Since
CFD was used due to its high efficiency and low risks, it also improves safety
while saving on production costs. An example would be the use of wind tunnel to
determine efficiency of a car model cutting through the air (fluid flow) will
increase costs in Research and Development sector of a company. By using CFD as
the base to study fluid flow on the desired model, less electricity would be
used on simulation compared to a physical wind tunnel. Costs of material will
also be saved since no models were needed and changes to the car model to
improve efficiency and performance can be done without wastage of any
materials. Besides that, CFD was also to be used in ensure minimal losses in
piping by simulating the pipe designs with different angles to reduce losses at
bends of pipe, attachment of valves, taper and other pipe attachments. (Gabryjonczyk, 2013).
Therefore, this CFD software can solve many fluid problems, preventing design
failure while ensuring safety of its design for prototype or production models.

 

Here
is the order of the process of CFD software based from the 5th slide
of (Choudhary,
2015):

 

Figure 1: Flow chart of
Computational Fluid Dynamics

 

The
cycle of CFD is not that complex to understand to begin with as it all begins
with the problem related to fluid where humans tend to solve from their
knowledge of Fluid Mechanics. This data is then brought into Navier-Stokes
Equations to determine the nature of the flow. This data is then translated
into Discretized Form as a computer only understood programming language. In
this form, the computer can analyse and compute the entered data, Grid by Grid
for each mesh on the model. After the meshing process, simulation is executed
to show every result obtained from fluid flow. The results are then converted
from programming language to human language for quantifying and displayed for
user to compare and analyse the data.