Qatar University

College of Engineering
Numerical Methods (GENG 300)

Instructor: Prof. Usama Ebead

Email: uebead@qu.edu.qa
Phone: 44034189
Office: BCR-I205
 Prof. Usama Ebead

Instructor Information
• Prof. Usama Ebead
• Civil and Architectural Engineering
• College of Engineering
• Office: Building BCR- Room I205
• Phone:+974 4403 4189
• E-mail: uebead@qu.edu.qa

2
 Prof. Usama Ebead

Course Contents

1 Introduction 2 Roots of Equations

Linear Algebraic Ordinary Differential

3 Equations 4 Equations

Numerical
5 Curve Fitting 6 Differentiation and
Integration

3
 Prof. Usama Ebead

Course Objectives
• Tointroduce students to the most commonly used
numerical methods in the different engineering fields.
• The course is not theorem‐oriented one.
• The emphasis will be on the understandingof the
numerical methods concepts and on applying these
concepts for solving various problems.
• MATLAB and Microsoft Excel will be used as tools to
solve the problems using the different numerical methods.

4
 Prof. Usama Ebead

Tentative Grading
Assessment Percentage

Homework and Quizzes 15 %

Midterm(s) 20 %

Project 10 %

Lab 20 %

Final Exam 35 %

5
 Prof. Usama Ebead

Course Objectives

• What do we mean by Numerical Methods?

- These are the methods used to obtain numerical solution
of a mathematical problem that is usually iterative in
nature leading to approximated solutions.

• Why do we need them?

- In case that a closed formula analytical solution does not
exist.
- In case that an exact analytical/mathematical solution is
difficult or impractical to obtain.
6
 Prof. Usama Ebead

Mathematical Model/Equations
• It is an expression of a system in mathematical terms.

Dependent Variable = 𝑓(Independent variables)

E.g.: Newtons 2nd Law:

𝐹 =𝑚𝑎
𝐹 = 𝑛𝑒𝑡 𝑓𝑜𝑟𝑐𝑒 (𝑘𝑔. 𝑚/ 𝑠 2 )
𝑚 = 𝑚𝑎𝑠𝑠 𝑜𝑓 𝑡ℎ𝑒 𝑜𝑏𝑗𝑒𝑐𝑡 𝑘𝑔
𝑎 = 𝑎𝑐𝑐𝑒𝑙𝑒𝑟𝑎𝑡𝑖𝑜𝑛 (𝑚/𝑠 2 )

7
 Prof. Usama Ebead

Mathematical Model
• Complex example: The air resistance force
Model for falling parachute:

When the chute opens, air

resistance increases dramatically:
The air resistance force is much
greater than the weight force, so
the skydiver slows down.

The weight force

8
 Prof. Usama Ebead

Mathematical Model
• Complex example: 𝑑𝑣
=𝑔−
𝑐
Model for falling parachute: 𝑑𝑡 𝑚𝑣

v = terminal velocity (𝑚/𝑠)

t = time (𝑠)
g = gravitational constant (9.8 𝑚/𝑠2)
c = drag coefficient (𝑘𝑔/𝑠)
m = mass of the object (𝑘𝑔)
• Solution for differential equation cannot be solved
analytically using simple algebraic solution. The exact
solution for differential equation can be solved using
calculus or by approximation using numerical methods.

9
 Prof. Usama Ebead

Dimensional Homogeneity
Each of the terms of an equation must be
expressed in the same units.
s = v t + 1/2 a t2
s is the distance (position) in meters
v is velocity in m/s
a is acceleration in m/s2
t is time in seconds
m = m/s  s + m/s2  s2 = m
 Prof. Usama Ebead

Significant figures
A significant figure is any digit, including a zero, provided it is
not used to specify the location of the decimal point for the
number.

5604 has four significant figures

34.52 has also four significant figures
0.5 has one significant figure

We can express numbers using exponents:

413 m= 0.4(103)m (one significant figure)
413 m= 0.41(103)m (two significant figure)
413 m= 0.413(103)m (three significant figure)
 Prof. Usama Ebead

Error Analysis
• Why errors are of a concern?
- For many engineering problems, we cannot obtain exact
solutions.
- Numerical methods yield approximate results that are
close to the exact solution.
• How much error is present in our calculation and is it
tolerable?
- True Error: Numerical error equal to discrepancy
between the true value and approximated value, 𝐸𝑡
𝐸𝑡 = 𝑇𝑟𝑢𝑒 𝑣𝑎𝑙𝑢𝑒 − 𝑎𝑝𝑝𝑟𝑜𝑥𝑖𝑚𝑎𝑡𝑒𝑑 𝑣𝑎𝑙𝑢𝑒
- True percent relative, 𝜀𝑡
𝑡𝑟𝑢𝑒 𝑒𝑟𝑟𝑜𝑟
𝜀𝑡 = × 100%
𝑡𝑟𝑢𝑒 𝑣𝑎𝑙𝑢𝑒

12
 Prof. Usama Ebead

Error Analysis
• What if we can not get the true value? How can we
calculate the error?
- We normalized the error to approximate the value.
- Numerical methods use iterative approach to compute
answers. A present approximation is made on the basis of
a previous approximation.
- Absolute Percent relative error, 𝜀𝑎

𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑎𝑝𝑝𝑟𝑜𝑥. −𝑝𝑟𝑒𝑣𝑖𝑜𝑢𝑠 𝑎𝑝𝑝𝑟𝑜𝑥.

𝜀𝑎 = × 100%
𝑐𝑢𝑟𝑟𝑒𝑛𝑡 𝑎𝑝𝑝𝑟𝑜𝑥.

13
 Prof. Usama Ebead

Accuracy vs. Precision

Accuracy: refers to the

closeness of a measured
value to a standard or Precision: refers to the
known value. closeness of two or more
measurements to each other.

Ex.
If the mass of a standard sample is 50 gram, if you took multiple
measurements of it and obtain values:
A) 47.5, 47.6, 47.5, and 47.7 grams your scale is precise,
but not that accurate;
B) 49.8, 50.5, 51.0, and 49.6 gram your scale is accurate
but not as as precise

14
 Prof. Usama Ebead

Accuracy vs. Precision

15
 Prof. Usama Ebead

Taylor Series
• What is the Taylor Series?
- A mathematical formulation that used widely in numerical
methods to predict a function value in approximate
fashion.
• Why it is called series?
- It is built term-by-term, starting with a zero-order
approximation. The higher the order the lower the error.
- Errors result from using an approximation in place of an
exact mathematical procedure.

16
 Prof. Usama Ebead

Taylor Series
Zero-order approximation:
𝑓(𝑥𝑖+1 ) ≅ 𝑓(𝑥𝑖 )
First-order approximation:
𝑓(𝑥𝑖+1 ) ≅ 𝑓(𝑥𝑖 ) + 𝑓′(𝑥𝑖 )(𝑥𝑖+1 − 𝑥𝑖 )
Let’s call 𝒙𝒊+𝟏 – 𝒙𝒊 = 𝒉
𝑓(𝑥𝑖+1 ) ≅ 𝑓(𝑥𝑖 ) + 𝑓 ′ 𝑥𝑖 ℎ
Second-order approximation:
𝑓 ′′ 𝑥
′ 𝑖
𝑓(𝑥𝑖+1 ) ≅ 𝑓(𝑥𝑖 ) + 𝑓 𝑥𝑖 ℎ + ℎ2
2!
17
 Prof. Usama Ebead

Taylor Series
• Higher nth-order approximation:
𝑓′′ 𝑥𝑖 2
𝑓 𝑥𝑖+1 = 𝑓 𝑥𝑖 + 𝑓′ 𝑥𝑖 ℎ + ℎ +
2!
𝑓 3 𝑥𝑖 𝑓 𝑛 𝑥𝑖 𝑛
ℎ3 + ⋯ … … . + ℎ + 𝑅𝑛
3! 𝑛!

Where 𝑅𝑛 is a remainder term.

18
 Prof. Usama Ebead

Taylor Series
• Example: Use Taylor Series to approximate 𝑓 𝑥 = cos(𝑥)
𝜋 𝜋 𝜋
to evaluate 𝑓 on the basic of value 𝑓 . 𝑥𝑖 = and
3 4 4
𝜋
𝑥𝑖 + 1 =
3

Order n 𝑓 (𝒏) (𝑥) 𝒇(𝝅/𝟑) |𝜺𝒕|%

0 cos 𝑥 0.707106781 41.4
1 − sin 𝑥 0.521986659 4.4
2 − cos 𝑥 0.497754491 0.449
3 sin 𝑥 0.499869147 0.0262
4 cos 𝑥 0.500007551 0.00151
5 − sin 𝑥 0.500000304 0.0000608
6 − cos 𝑥 0.499999988 0.00000244
𝜋 𝜋 𝜋 𝜋 𝜋
1st order: cos = cos − sin − 4 = 0.52198
3 4 4 3
𝜋 𝜋 𝜋 𝜋 2
2nd order: : cos = 0.52198 − cos − /(2!)=0.4977 19
3 4 3 4
 Prof. Usama Ebead

Taylor Series
graphically

20
 Prof. Usama Ebead

Commonly Used Taylor Series

21
 Prof. Usama Ebead

Thanks!

22