UNIVERSITI TEKNOLOGI MARA

TECHNICAL REPORT

(SOLVING ONE DIMENSIONAL BRATU TYPE EQUATION

USING RUNGE-KUTTA METHOD AND ADOMIAN
DECOMPOSITION METHOD)

(NUR ASYIKIN BINTI HASHIMI) – (2017565717)

(NUR IZZATI BINTI ABD JAMAL NASIR) - (2017508945)
(NIK PUTERI ZAKIRAH BINTI NIK OTHMAN) - ( 2017503817)

Report submitted in partial fulfillment of the requirement

for the degree of
Bachelor of Science (Hons.) (Mathematics)
Faculty of Computer and Mathematical Sciences

DECEMBER 2017
ACKNOWLEDGEMENTS

IN THE NAME OF ALLAH, THE MOST GRACIOUS, THE MOST MERCIFUL

Firstly, all praises to Allah S.W.T and with his grace, we managed to complete
our Final Year Project (FYP) successfully.

We would like to express our sincere gratitude to our supervisor, Modam Rosha
Mohamed and our lecterur, Dr. Nor Azni Shahari for their time, patience and guidance in
ensuring the successfulness of this project. Also, special thanks to in helping us
accomplishing this project during our supervisor maternity leave.

Next, a heartfelt thanks to our classmates, for their kindness and guidance that we
were able to complete this project. We would also like to thank the members of Fakulti
Sains Komputer dan Matematik (FSKM) and Universiti Teknologi Mara(UiTM) for
giving us the approval to carry out this project.

Lastly, not forget to mention, our sincere thanks to parents and family for their
support, encouragement and understanding as we managed to make this project a success.

ii
TABLE OF CONTENTS

ACKNOWLEDGEMENTS................................................................................................ii
TABLE OF CONTENTS...................................................................................................iii
LIST OF TABLES...............................................................................................................v
LIST OF FIGURES.............................................................................................................v
ABSTRACT.......................................................................................................................vi
1.0 INTRODUCTION.........................................................................................................1
In spite of this progress in explicit and exact solution for solving nonlinear differential
equations, the general equation to Bratu equation is impossible to find in the literature
explicitly and exactly by usual way. A simple way to do was to transform the Bratu
nonlinear differential equation into a linear second order differential equation with well
known propertiesThe usual way to solve a boundary or initial value problem consists of
computing the shooting method to Runge-kutta equation. Shooting method is a method
for solving nonlinear boundary value problem of differential equation by reducing
boundary value problem into an initial value problem with initial parameter established.
In order to give initial value of parameter, initial value problem can be solved in terms of
Runge-Kutta method where the solutions satisfies the boundary conditions......................1
1.1 Problem Statement...................................................................................................2
1.2 Research Objective.................................................................................................2
1.3 Scope and Limitations of Study...............................................................................2
1.4 Significance of Study and Research Benefit...........................................................3
1.5 Definitions of Terms and Concept...........................................................................3
2.0 BACKGROUND THEORY AND LITERATURE REVIEW.......................................4
2.1 Background Theory...........................................................................................4
2.2 Literature Review/ Related Research......................................................................5
3.0 METHODOLOGY AND IMPLEMENTATION...........................................................6
3.1 Flowchart Implementation of Bratu Equation.........................................................6
3.2 The Steps To Obtain Result.....................................................................................7
3.2 The Steps To Obtain Result.....................................................................................7
3.1 Citations...............................................................................................................12
3.2 Implementations/ Numerical Examples...............................................................12
1. RESULTS AND DISCUSSION..........................................................................13
2. CONCLUSIONS AND RECOMMENDATIONS...............................................14
REFERENCES..................................................................................................................15
APPENDIX A....................................................................................................................16
APPENDIX B....................................................................................................................18

iii
iv
LIST OF TABLES

Table 1: BIMB Rate for GIA Product...............................................................................10

Table 2: Rate of Profit for GIA.........................................................................................11

LIST OF FIGURES

Figure 1: Islamic Banking in Malaysia: Concept..............................................................11

v
ABSTRACT

The purpose is to give the reader a quick review regarding your research. It should be
brief, containing the information of what you have done in the research. The abstract of a
scientific   paper   or   report   is   capable   of   being   published   separately   and   hence   it   is
independent.  Therefore it should have the (1) background /general statements, (2)
problem statements (3) research objectives, (4) brief description of methodology, (5) the
results/findings, (6) significance of findings, recommendations and future directions. For
this reason, the heading of abstract section is not numbered. Numbering starts with the
introduction.

vi
1.0 INTRODUCTION
(Left, Cap, Times New Roman, Size 14, Bold)

The Bratu problem is one of the most investigated boundary value problems in
mathematics. This problem derives its importance from combustion theory where it has
been used for several applications such as fuel ignition model of the speculation of
thermal combustion, the thermal reaction method model, the Chandrasekhar model of the
growth of the universe, reactive heat transfer and theory of chemical action. The exact
solutions has been applied to test the accuracy and efficiency of many approximate
methods such as Adomian Decomposition method[7] and Runge-Kutta method. Bratu
equation exhibits a bifurcation pattern which only characterizes nonlinear differential
equations. The one-dimensional Bratu boundary value problem may be written [7]

where

and is a constant. The Bratu type initial value problems have been examined by

numbers of authors[5-7].This research is to investigate the explicit and exact general

solution between Adomian Decomposition Method and Runge-Kutta Method. Hence,
after computing the solution between two method stated above, this research wanted to
compare the absolute error of two methods.

In spite of this progress in explicit and exact solution for solving nonlinear differential
equations, the general equation to Bratu equation is impossible to find in the literature
explicitly and exactly by usual way. A simple way to do was to transform the Bratu
nonlinear differential equation into a linear second order differential equation with well
known propertiesThe usual way to solve a boundary or initial value problem consists of
computing the shooting method to Runge-kutta equation. Shooting method is a method
for solving nonlinear boundary value problem of differential equation by reducing
boundary value problem into an initial value problem with initial parameter established.
In order to give initial value of parameter, initial value problem can be solved in terms of
Runge-Kutta method where the solutions satisfies the boundary conditions.

1
1.1 Problem Statement

1.2 Research Objective

The objective of our project:

I. To compute the actual value of Bratu type equations.

II. To discover the numerical analysis of second derivative of Bratu type equations.
III. To compare numerical solution of Bratu equation with the exact solution.
IV. To determine the effective numerical method to solve Bratu type equations.

1.3 Scope and Limitations of Study

The range of research focuses to one–dimensional boundary value problem in

Bratu equation. The one-dimensional boundary value problem may be written

where
u

and is constant.

The limitation of this study is there is a constraint in boundary value problem

which is fixed . Moreover, another limitation of this study is highest derivative

in differential equation which is second order differential equation. Besides that, the
limitation for this research is there is limited access to data since calculation is based on
boundary value problem. The restriction of research is time constraint where the time
required to solve Bratu problem and to compare absolute error between exact and
numerical solution is constrained by deadline of the proposal which are two semester.

2
1.4 Significance of Study and Research Benefit

The significance of this research is to evaluate the general solution by using an

explicit and exact method for solving nonlinear differential equations such that various
initial and boundary may be applied. The Bratu nonlinear differential equation is
transformed into linear second derivative in differential equation in regards of the
generalized Sundman linearization theory developed by Saravi et. Al (2013) is well used.
It is possible to compute the exact solution which examined by Wazwaz(2005) to Bratu
type initial and boundary value problem by using Adomian Decomposition method. The
method applied in this study, linear harmonic oscillator differential equation which also
known as exact analytical solution is directly related to Bratu nonlinear differential
equation.
The benefit of this study is the explicit and exact general solution has been
resolved by using generalized Sundman transformation where it is closely related to Bratu
differential equation. Therefore, it shows simplicity to the Bratu equation where it is a
special case of general solution so it could compute the explicit and exact solution.
Through these findings, the researcher found that the best method between Runge-Kutta
and Adomian Decomposition method is by having the lowest absolute error compared to
others.

1.5 Definitions of Terms and Concept

Keyword Meaning
Bratu equation An equation to solve nonlinear equation with given
boundary conditions.
Initial Value Problem Initial condition of unknown function with given point of
the solution.
Boundary Value Problem A data value that corresponds to minimum or maximum
value for specified equation.
Absolute Error The difference between measured value and true value.
Exact Solution A first order differential equation that does not contains
higher derivative of unknown functions.
Numerical Equation An equation where coefficients and constants are numbers.

3
2.0 BACKGROUND THEORY AND LITERATURE REVIEW

2.1 Background Theory

4
2.2 Literature Review/ Related Research

5
3.0 METHODOLOGY AND IMPLEMENTATION

3.1 Flowchart Implementation of Bratu Equation

Start

Define Bratu equation

Define boundary
value problem

Implementation of formula
using ADM and
Runge-Kutta Method

Obtain Result

Compare absolute error for each

method solution and exact solution,

End

6
 1. 3.2 The Steps To Obtain Result

The first Bratu-Type problem considered by Wazwaz(2005) may be written in the form

u " ( x ) + l eu ( x ) = 0 , 0 < x <1

u ( 0 ) = u ( 1) = 0

Implementation of Adomian Decomposition Method

Using Adomian Decomposition Method (ADM), Taylor expansion allows us to exchange

nonlinear terms. Toward this end, we regarded as

2 3

e ; 1 + u + u2! + u3!
u

ADM equation can be written as

� 2 3
�
u "( x ) + l � 1+ u + u + u �= 0, 0 < x <1
� 2! 3! �
� �
u ( 0 ) = u ( 1) = 0

The computed solution

u ( x) = x
2
0

1 1 6 1 8
u ( x) = 6 x
4

30 x 168 x
1
+ + ,

1 8 1 1
u 2 ( x ) = 168 x + 1350 x + 11088 x + ...,
10 12

1 1 323
u 3 ( x ) = 1800 x + 10395 x + 6879600 x + ...,
10 12 14

Thus, the general solution for Adomian Decomposition Method

7
8
Implementation of Fourth Runge-Kutta Method

where

9
10
Tables and the titles are presented as below. The title of the table is placed above the
table. Use only three lines for tables and unless there are really a lot of columns and rows
involved, then highlight all the lines. Place the tables in the center.

The title of the table is placed above the table. Number the tables as Table 1, Table 2,
Table 3,….

Table 1. BIMB rate for GIA products

Investment
in period
Source: BNM (2009) (month) Weight
1 0.76
3 0.82
6 0.88
9 0.94
12 1.00
15 1.02
18 1.04
24 1.06
36 1.12
48 1.18
≥60 1.24

Table 2. BIMB rate for GIA products

Investment
in period
Source: BNM (2009) (month) Weight
1 0.76
The source of tables or 3 0.82 figures is placed below
them respectively. Please 6 0.88 use the APA style, of
course. Figures and title 9 0.94 of figures are presented
as below. Place the 12 1.00 figures in the center and
15 1.02
18 1.04
24 1.06
36 1.12
48 11 1.18
≥60 1.24
the title of the figure is placed below the table. Number the figures as Figure 1, Figure 2,
Figure 3,….

Figure 1. Islamic banking in Malaysia: concept

Please write the information or give comment regarding the tables or figures. This habit
should be instilled after titles of sections or subsections as well. Lead the reader to the
information you are presenting and what they should know. Do not let them read your
mind from the tables, figures or titles. Table 2 is another example of a table

Table 3. Rate of profit for GIA

Rate of
profit
Investment period 1 Month 3 Month 6 Month 9 Month 12 Month
Weight 0.76 0.82 0.82 0.88 0.94
Date
16/05/05-15/06/05 2.880 3.107 3.107 3.335 3.562
16/06/05-15/07/05 1.320 1.424 1.424 1.528 1.633
16/07/05-15/08/05 2.000 2.158 2.158 2.316 2.474
16/08/05-15/09/05 2.300 2.482 2.482 2.663 2.845
16/09/05-15/10/05 2.110 2.277 2.277 2.443 2.610
16/10/05-15/11/05 1.640 1.769 1.769 1.899 2.028
16/11/05-15/12/05 2.330 2.514 2.514 2.698 2.882
16/12/05-15/01/06 1.680 1.813 1.813 1.945 2.078
16/01/06-15/02/06 1.170 1.262 1.262 1.355 1.447
16/02/06-15/03/06 2.000 2.158 2.158 2.316 2.474
16/03/06-15/04/06 1.980 2.136 2.136 2.293 2.449
16/04/06-15/05/06 1.620 1.748 1.748 1.876 2.004
16/05/06-15/06/06 1.730 1.867 1.867 2.003 2.140
16/06/06-15/07/06 1.640 1.769 1.769 1.899 2.028

12
 16/07/06-15/08/06 1.890 2.039 2.039 2.188 2.338
16/08/06-15/09/06 2.310 2.492 2.492 2.675 2.857
Explanation of the important values in tables and figures is vital. Elaboration of the
information will lead to a better understanding to the readers.

3.1 Citations

There are many ways and style of citations and references. For this technical report, we
shall use the American Psychological Association (APA) style. The information can be
served from this web-site http://owl.english.purdue.edu/owl/resource/560/01/. Arrange
the references in alphabetical order.

3.2 Implementations/ Numerical Examples

In this section, please explain how the project was conducted in detail (the technical part
of course)-how you get/use the data, show how you solve the equations, show how you
derive a formula, show how you analyse a result etc. The writing should include
everything such that when another person does the same method of project, then he will
get the same result. The information in this section should have a one-to-one mapping
with the information in methodology.

13
 2. RESULTS AND DISCUSSION

All results of the research in presented in this section. It is highly recommended that your
discussion is done after every presentation of results (tables, graphs, charts etc). You can
provide interpretation of results, comparisons with other findings or other shortcomings in
the project. The information gathered here may lead to certain conclusion as well. If the
results are unusual or it is not as expected as compared to other findings then you should
explain why you think it might be. Otherwise, the reader may think that you have made a
mistake.

14
 3. CONCLUSIONS AND RECOMMENDATIONS

The conclusion summarizes all the inferences that can be drawn from the results and
discussion of the research. It answers all the objectives of the project clearly. It should
also explain the meaning of the findings of the research.

This section should not introduce new material but present a report that relates to the
material, results, evidence, and discussion that appears in the research only.

Any recommendations should be based strictly on the materials discussed in your results
and conclusion. Recommendations tell the reader what to do; what the best decision or
solution is; what the next of action to be taken; what the alternatives are.

15
REFERENCES

Please make sure that all references cited in the text are listed in this reference section.
The referencing method use is APA Formatting and Style Guide. The information can be
served from this web-site http://owl.english.purdue.edu/owl/resource/560/01/. Arrange
the references in alphabetical order.

Altunkaynak, A., Özger, M., & Çakmakci, M. (2005). Water consumption prediction of
Istanbul city by using fuzzy logic approach. Water Resources Management, 19(5),
641-654
Arbués, F., Garcııa-Valiñas, M. a. Á., & Martıınez-Espiñeira, R. (2003). Estimation of
residential water demand: a state-of-the-art review. The Journal of Socio-
Economics, 32(1), 81-102
Azar, A. T., El-Said, S. A., & Hassanien, A. E. (2013). Fuzzy and hard clustering analysis
for thyroid disease. Computer Methods and Programs in Biomedicine, 111(1), 1-
16.
Bezdek, J. C. (1981). Pattern recognition with fuzzy objective function algorithms:
Kluwer Academic Publishers.
Corona-Nakamura, M. A., Ruelas, R., Ojeda-Magana, B., & Finch, D. W. C. (2009, 11-14
Oct. 2009). Identification of domestic water consumption in a house based on
fuzzy clustering algorithms. Paper presented at the IEEE International Conference
on Systems, Man and Cybernetics, 2009. SMC 2009.
Corona-Nakamura, M. A., Ruelas, R., Ojeda, m., x, a, B., & Andina, D. (2008, Sept. 28
2008-Oct. 2 2008). Classification of domestic water consumption using an ANFIS
model. Paper presented at the World Automation Congress, 2008. WAC 2008.
Ghosh, S., & Dubey, S. K. (2013). Comparative analysis of k-means and fuzzy c-means
algorithms. International Journal of Advanced Computer Science and
Applications (IJACSA), 4(4).

16
Information that is not essential but supports findings, validates conclusion, explain
conclusions can be placed in the appendices section. If you have information in the
appendices then please use Appendix A, Appendix B, Appendix C,…. Appendices exist if
you mention them in your text.

APPENDIX A
Conventional model

Solving
E =E + r kE 0 (1)
t t 1 t
where Q0 = 0

From (1)

E = E 0 + r1 kE0
1

= E 0 (1 + r1 k )

E = E + r kE
2 1 2 0

= E 0 (1 + r1k ) + r2 kE0
= E 0 (1 + r1k + r2 k )

E 3 = E 2 + r3 kE
0
= E 0 (1 + r1 k + r2 k ) + r3 kE0
= E 0 (1 + r1 k + r2 k + r3 k )
3
= E 0 (1 +  ri k )
i =1
3
= E 0 (1 + k  ri )
i =1

Here, we can guess that

n
E n = E 0 (1 + k  ri ) (2)
i =1
By using the principle of mathematical induction,

E1 = E 0 (1 + kr1 ) , E1 is true

17
If E a  1 is true, then
a 1
E a 1 = E 0 (1 + k  ri )
i =1
(3)

From (1)

E a = E a 1 + ra kE0

Substitute equation (2) , we get

= E 0 (1 + kr1 + kr2 + ... + kra 1 ) + ra kE0

= E0 (1 + kr1 + kr2 + ... + kra 1 + kra )
a
= E 0 (1 + k  ri )
i =1
n
Therefore, equation (2) is true, that is E n = E 0 (1 + k  ri ) for all n elements of natural
i =1
numbers.

18
APPENDIX B
u "( x ) + le
u (x)
=0

19