UNIVERSITY OF GUYANA

FACULTY OF ENGINEERING AND TECHNOLOGY

DEPARTMENT OF CIVIL ENGINEERING
GROUP 16

COURSE NAME: ENGINEERING MATHEMATICS (EMT 3200)

 LECTURER: MR RUEL ALLEYNE

TOPIC: FINITE DIFFERENCES: FORWARD, BACKWARD AND CENTRAL
DIFFERENCES 1
 NAMES USI TASK
ANURADHA 1028267 Introduction(History,
MOHAN Application and
Advantage and
Disadvantage of Finite
Differences); Question
4.
DERRICK 1035265 Finite Differences using
GARBUTT First Derivative;
Question 1
JUMIAH 1031858 Taylor’s Series using
WHITTINGTON Finite Differences;
Question 2
CHEVRAJ 1011015 Finite
BOODRAM (L) Operators(Forward,
Backward and Central
Differences); Question
3

2
PRESENTATION OUTLINE

• Introduction(History, Application and Advantage and

Disadvantage of Finite Differences)

• Finite Differences using First Derivative and Example

• Taylor’s Series using Finite Differences and Example

• Finite Operators(Forward, Backward and Central Differences)

and Examples

3
HISTORY TO FINITE DIFFERENCES

Finite difference approximations are one of the oldest and simplest methods
used to solve differential equations (Lu, 2009). L. Euler (1707-1783) ca. 1768
used finite approximations in one dimension of space, and was later extended
by C. Runge (1856-1927) ca. 1908 to two dimension (Lu, 2009). In the early
1950s, the dawn of finite difference techniques in numerical applications was
initiated. The development of the techniques was stimulated by the emergence
of computers that offered a convenient framework for dealing with complex
problems of science and technology (Lu, 2009). Finite difference method was
extended to 3D due to parallel computations by Frankel and Vidale in 1992 and
Osclen and Archuleta in 1996 (Lu, 2009).

4
APPLICATION OF FINITE DIFFERENCES

• Calculation of dynamic processes in long electricial lines.

The electrical circuit structure includes components with concentrated and distributed
parameters. This creates difficulties in analyzing the dynamic processes, which are
described by differential equations, including those with partial derivations.

• Application of the finite difference method in seismic source and wave diffraction
simulation

For seismic problems such as an investigation of the displacements caused by a point

or a line source, analytical solutions are available only in some simple cases.
5
• Application in mechanical engineering.

Finite difference method is used in several mechanical engineering disciplines (such

as automotive industries and so). Finite difference method helps tremendously in
producing stiffness and strength visualizations and also in minimizing weight,
materials and costs. In addition, it allows detailed visualization of where structures
bend or twist, and indicates the distribution of stresses and displacements.
 

6
ADVANTAGE AND DISADVANTAGE OF USING FINITE DIFFERENCES

• Advantage
The finite-difference method is defined dimension per dimension; this makes it
easy to increase the “element order” to get higher-order accuracy.

• Disadvantage
With the finite-difference method, you may easily run into problems handling
curved boundaries for the purpose of defining the boundary conditions.

7
FORWARD, BACKWARD, AND CENTRAL DIFFERENCE
FORMULAS FOR THE FIRST DERIVATIVE

Figure 1: Finite Difference Approximations of Derivative

8
FORWARD, BACKWARD, AND CENTRAL DIFFERENCE FORMULAS FOR
THE FIRST DERIVATIVE

  Forward difference is the slope of the line that connects points :

 Backward difference is the slope of the line that connects points :

9
  Central difference is the slope of the line that connects points :

10
EXAMPLE 1

Consider the function f(x)= .Calculate its first derivative at point x = 3

 

numerically with the forward, backward, and central finite difference

formulas and using: Numerical Differentiation (a) Points x = 2.75, x = 3,
and x = 3.25.

11
SOLUTION:
 
Analytical Differentation: The derivative of the function is f′ (x) =
()=27
Numerical Differentiation:
The Points Used for numerical differentiation are:

12
Using the derivatives using the forward, backward, and central finite
difference formulas are:
 Forward Finite Difference:

= error=
 
Backward Finite Difference:
= error=
 
Central Finite Difference:
= error=

13
FINITE DIFFRENCE FORMULAS USING TAYLOR SERIES
EXPNASION

Taylor series expansion can be used to construct the forward, backward,

and central difference formulae, as well as many additional finite difference
formulas for approximating derivatives. The formulas estimate the
derivative at a given point based on the values of nearby points.

14
 Finite Difference Formulas of First Derivative

Several formulas for approximating the first derivative at point based on the values of
the points near are derived by using the Taylor series expansion. All the formulas
derived in this section are for the case where the points are equally spaced.
Two-point forward difference formula for first derivative

The approximation in Eq.5 is the same as the forward difference formula in Eq. 1.1
Two-point backward difference formula for first derivative

The approximation in Eq.9 is the same as the forward difference formula in Eq. 1.2.

15
 • Two-point central difference formula for first derivative

The approximation in Eq.13 is the same as the central difference formula Eq.1.3 for
equally spaced intervals.

• Finite Difference Formulas for the Second Derivative  

The same approach used in above to develop finite difference formulas for the
first derivative can be used to develop expressions for higher-order derivatives. In
this section, expressions based on central differences, one-sided forward differences,
and one-sided backward differences are presented for approximating the second
derivative at a point .

16
 • Three-point central difference formula for the second derivative

Central difference formulas for the second derivative can be developed using any
number of points on either side of the point , where the second derivative is to be
evaluated. The formulas are derived by writing the Taylor series expansion with a
remainder at points on either side of. in terms of the value of the function and its
derivatives at point .

Three-point central difference formula for the second derivative

An estimate for the second derivative can be obtained by solving Eq.16 for while
neglecting the remainder terms. This introduces a truncation error of the order of .

17
The table below lists difference formulas, of various accuracy, that can
be used for numerical evaluation of first, second, third, and fourth
derivatives.

18
19
QUESTION 2

Approximate the first-order derivative of at x =

 

0.5 using finite differences and a step size of h =

0.25 with accuracy

20
SOLUTION:
The derivative can be calculated analytically as and its true
 

value at
The data that we need is

21
The forward difference of accuracy is computed as
 

And a percentage relative error

22
The forward difference of accuracy is computed as
 

And a percentage relative error

23
The backward difference of accuracy is computed as
 

and percentage relative error

24
The central difference of accuracy is computed as
 

25
The results show that the central finite difference formula gives a
more accurate approximation. This will be discussed further in the
next section. In addition, smaller separation between the points
gives a significantly more accurate approximation.
 

26
FINITE OPERATORS: FORWARD, BACKWARD AND CENTRAL
DIFFERENCES
 
•   FORWARD DIFFERNCES
 

Figure 2: Graph showing approximation of forward differences

27
 From the graph, we can approximate that:

If h > 0, say h = where if a finite positive number, then:

This is called a one side difference or forward difference approximation to the

derivative of f(x) and is represented as:
 

 
 
 
 

28
• BACKWARD DIFFERNCES

Figure 2: Graph showing approximation of backward differences

29
 A second version of this arises, when considering a part to the left, so we can
approximate that:
If h < 0, say h = where , then:
 
=
 
Is called the first order or backward difference approximation of f (x) is
represented as:
 

 
 

30
• CENTRAL DIFFERENCES

Figure 4: Graph showing approximation of central differences

31
 By combining different Taylor series expansions, we can obtain approximations
of f’ (x) of various orders. For instance, substracting the two expansions:

 
Where is a number between

 
Where is a number between
 
The result is:

32
 So that:

Hence is an approximation of f’(x) whose error is proportional to

 
It is called the second order or centred difference approximation of f’(x) and is
represented by:

 
 

33
EXAMPLES……

34
 
QUESTION 3:
Determine the forward, backward, and
central differences using the forward,
backward, and central methods respectively,
to solve:
f (x) = 2
Where;
X=3
Δ x= 0.1 35
SOLUTION:

Forward method -
f‘ ( x ) f ( x + Δ x ) – f ( x ) ]/ Δ x

f‘ ( x ) = - =

= 291.36
  36
Backward method
 

f‘ ( x ) [ f ( x ) – f ( x - Δ x ) ] / Δ x

f‘ ( x ) = - =

= 250.73

37
Central method
 

f‘ ( x ) f ( x + Δ x ) – f ( x + Δ x ) ] / 2(Δ x)

f‘ ( x ) = -

= = 271.05

38
 
Testing the solution
True error = True value – Δv
=2
Now, recall that,
= a.
And so we get; (1.5)2 where x= 3
= (1.5)2
= 3 = 270.05 True Value
 
39
Now:
ET = TV-ΔV
Forward method
270.5-291.36= -21.31
Backward method
270.5-250.73= 19.32
Central method
270.5-271.05= -1

Note: the central method is the most accurate of the

three methods in evaluating these derivatives and
40
hence, such was proven in this example.
QUESTION 4:
Let f(x) = ln(x) and a = 3. Using both a
forward difference and a central
difference, and working to 8 decimal
places, approximate f 0 (a) using h = 0.1
and h = 0.01(HELM (2008)).
SOLUTION:

From the problem we know that:

f(x) = ln(x)
a= 3,
h=0.1, 0.01

42
 
Solution:
Using h as 0.1, we find the forward difference as

Forward: f’ (a) = =
=

= 0.32789823
 

43
Using h as 0.01, we find the forward difference as
 

Forward: f’ (a) = =

= 0.33277901

44
Using h as 0.1, we find the central difference as
 

Central: f’ (a) = =

= 0.33345687
 

45
Using h as 0.01, we find the central difference as
 

Central: f’ (a) = =
=

= 0.33333457

The accurate answer is, of course, 0.33333333

 

46
 REFERENCES
Berzan, V. (2019, December 23). Retrieved from APPLICATION OF THE FINITE
DIFFERENCE METHOD TO CALCULATION OF DYNAMIC PROCESSES IN LONG
ELECTRICAL LINES: https://zenodo.org/record/3591586#.YMAV6_lKjIU

HELM. ((2008)). Numerical Methods of Approximation.

Hong, M. (1986, April 21). Retrieved from Application of the finite difference method in
seismic: https://watermark.silverchair.com/

https://watermark.silverchair.com/87-3-731.pdf?token=AQ. (1986, April 21). Retrieved

from Application of the finite difference method in seismic.

Mustafa, M. M. (n.d.). Chapter1:Numerical Differentiation.

Sjodin, B. (2016, April 18). Retrieved from What’s The Difference Between FEM, FDM,
and FVM?
47