Module - 1

CHE3001 Computational Methods in
Process Engineering
Instructor: Dr. P. Monash

School of Chemical Engineering (SCHEME)
VIT University Vellore
Date: 13 July 2020
Text book
Steven C. Chapra and Raymond P. canale “Numerical
Methods for Engineers” 7th edition, Mc Graw Hill Education,
2005.
Reference book
Dorfman K.D., Daoutidis P, Numerical Methods with Chemical
Engineering Applications, 1st ed., Cambridge University Press,
USA, 2017
Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 2

Course Objectives:
• Formulate problems for roots of a function, solution of
simultaneous equations, optimized value of a given
function, numerical integration and differentiation, ODE and
PDE
• Solve roots of a function, simultaneous equations,

optimization, numerical integration, ODE and PDE
• Develop MATLAB code for finding the roots of a function,

solution of a simultaneous equations, optimization,
numerical integration, ODE and PDE

Course Outcomes (CO)
• Formulate engineering problem as mathematical model for an
appropriate solution using numerical methods
• Determine roots of a single equation and simultaneous equations
• Solve optimization, regression and numerical integration using

different methods
• Evaluate ordinary differential equation involving initial value and

boundary value problems
• Estimate the solution for partial differential equation involving

elliptical and parabolic equation
• Create MATLAB program for roots finding, simultaneous

equations, optimization, regression and curve fitting, numerical
integration, ODE and PDE
Syllabus
Module – 1:
Roots of equations
Solve f(x) = 0 for x.
Module – 2:
Solution to simultaneous
linear algebraic equations
Given the a’s and the c’s, solve

a11x1 + a12x2 = c1
a21x1 + a22x2 = c2
for the x’s.
Module – 3:
Optimization
Determine x that gives optimum

f(x).
Module – 4:
Curve fitting and Interpolation

Module – 5:
Integration
I = ∫ f (x) dx
Find the area under the curve.
Module – 6:
Ordinary differential equations
Given
solve for y as a function of t.

yi + 1 = yi + f (ti , yi ) Δt
Module – 7:
Partial differential equations
Given
solve for u as a function of x and y

Assessment Methods
Assessment Date Max. Weightage Remarks
type Marks
Quiz 1 Almost every day 40 10 Five minute

Tests during
class till
CAT – I
Quiz 2 Almost every day 40 10 Five minute
Tests during
class till
CAT – II
Digital Submission before 31 50 10 Question and
Assignment October 2020 rubrics given
separately
CAT – I As per the announcement by the 15 Schedule will be
CAT – II University 15 announced by
the University
FAT 40

Introduction
• Numerical methods are techniques by which mathematical
problems are formulated so that they can be solved with
arithmetic operations.
• All numerical methods have one common characteristic -

invariably involve large numbers of tedious arithmetic
calculations.

Approaches used for solving Engineering
Problems
• Analytical or Exact solution
 Limited class of problems
• Graphical solutions
 Results are not precise
 Tedious and awkward to implement
• Solutions using calculators or slide rulers
 Slow and tedious
 Possibilities of manual mistakes

The sizes of the boxes indicate the level of emphasis directed toward each phase.
Simple mathematical
model
• Mathematical model can be
broadly defined as a
formulation or equation that
expresses the essential
features of a physical
system or process in
mathematical terms.

Example – Model formulation
(Falling Parachutist)
• Newton’s second law can be used to
determine the terminal velocity of a
free falling body near the earth’s
surface.

Finding the velocity of a parachutist
• Use the graphical approach to determine the drag

coefficient, c, needed for a parachutist of mass m = 68 kg to
have a velocity of 40 m/s after free falling for time t = 10 s.
Note: The acceleration due to gravity is 9.81 m/s2.

c f(C)
1 51.19385
2 44.96048
3 39.2952
4 34.13976
5 29.4423
6 25.15657
7 21.24136
8 17.65987
9 14.37924
10 11.37013
11 8.606282
12 6.064206
13 3.722867
14 1.563414
15 -0.43106
16 -2.27573
17 -3.98424
18 -5.56883
19 -7.04053
20 -8.40928
Initial Guesses:
C_lower= 2 f(C) = 44.96048 +VE
C_upper= 18 f(C) = -5.56883 -VE
Hence, at least one root between the limits
Iteration 1: [2, 18]

C_New = (C_lower + C_upper)/2
cnew = 10 f(C) = 11.37013 +VE
Since, the root lies between positive and negative function

values
The new interval becomes [10,18]
Error Definitions

Roots of a equation
o Roots Cannot be obtained by direct methods
 e.g. f(x)=e-x-x
o Alternative methods
 Graphical methods
Roots of the equation may be
either real or complex
 Trial and error methods
o Type of functions in root finding

 Algebraic
 e.g. Polynomials are simple class of algebraic functions
 Transcendental – nonalgebraic functions, which includes trigonometric, exponential,

logarithmic and less familiar functions
 e.g. ,

Finding roots of quadratic equation
Solve x2 – 3x – 4 = 0 Solve x2 – 4 = 0
For a Quadratic equation of ax2 + bx + c, the solution or the roots of the equation
is

Graphical solutions
Graphical solution of x2 – 3x – 4 = 0 Graphical solution of x2 – 4 = 0
Trial and error Method: 0, +1, -1 +2, -2, +3, -3, +4, -4, +5, -5……

Example – Quadratic equation
A brick is thrown from a top of a building at a speed of 25 m/s and the building is
150 m height. Derive an equation to find the time at which the brick reaches the
ground
Solution: The distance covered by the brick is given by
So the above expression becomes
150
This is of the form

150
ax2 + bx + c

Bisection Method
• Bracketing - Two initial guesses are required
• Bisection Method (also named as interval halving, binary
chopping, Bolzano’s Method)
Theorem f (x)
If function f(x) is a real and

continuous between the interval xl
and xu
 If f(xl) and f(xu) have opposite

sign i.e. if f(xl) f(xu) < 0 - there is
at least one real root between xl xℓ
x
and xu xu

f (x)
Theorem
If function f(x) is a real and continuous
between the interval xl and xu
 If f(xl) and f(xu) does not change sign

between two points i.e. if f(xl) f(xu) > 0 ,
roots of the equation f(x) = 0 may still x
exist between the two points xℓ xu
Theorem
 If f(xl) and f(xu) does not change sign

between two points i.e. if f(xl) f(xu) > 0,
there may not be any roots for the
equation f(x) = 0 between the two points

Theorem
 If f(xl) and f(xu) changes sign between

two points i.e. if f(xl) f(xu) < 0, more
than one root for the equation f(x) = 0
may exist between the two points
Solve
Please check
(1) x3 + 2x2 + 3 x - 4 = 0 with in the limits [0,1]
what will happen
if the Limits of
(2) x3 - 2.7 x2 + 4.5 x – 6 = 0 with in the limits
the problem (2)
[1,3]
becomes [1.5, 4]

Advantages and drawbacks BS Method
Advantages
• Always convergent
• Error can be controlled - guaranteed
Drawbacks
• convergence is slow, if
one of the initial guesses
is close to the root
• e.g. f(x) = sin(x)
For case 1: Take the limits [2, 4]
For case 2: Take the limits [3, 4]
Check which one requires more

iteration to converge

Drawbacks BS Method
o If a function f(x) is such that it just

touches the x-axis it will be unable
to find the lower and upper guesses
e.g. f(x) =x2
o Function changes sign but root

does not exist i.e. root may be a
singularity point
e.g. f(x)=1/x

Algorithm for the bisection method
Step - 1
Choose x and xu as two guesses for the root such that f ( x  ) f ( xu )  0 , or in other words,
f (x ) changes sign between x and xu
Step - 2
Estimate the root, x m , of the equation f ( x )  0 as the mid-point between x and xu as
Step - 3
Now check the following
a) If f ( x  ) f ( x m )  0 , then the root lies between x and x m ; then x  x and
xu  x m .
b) If f ( x ) f ( x m )  0 , then the root lies between x m and xu ; then x  x m and
x u  xu .
c) If f ( x  ) f ( x m )  0 ; then the root is x m . Stop the algorithm if this is true.

Step - 4
Find the new estimate of the root
x  xu
xm = 
2
Step - 5
Find the absolute relative approximate error as
x mnew - x mold
a = new
 100
xm
where
xmnew = estimated root from present iteration
xmold = estimated root from previous iteration
Compare the absolute relative approximate error a with the pre-specified relative error tolerance s
If a s , then go to Step 3, else stop the algorithm.

Finding roots using simple commands
o Use “fzero” command in Matlab o Use “Goal Seek” option in Excel
>> x0=[1 3];
>> x = fzero(@(x)x^3-2.7*x^2+4.5*x-6,x0);
One can also use
“solver” option in
MS - Excel
o Use “root” option in Mathcad

One can also use “fsolve”
option in Matlab to find the
root
Analyze the difference
between fzero and fsolve
command
Hint: use Help command in
Matlab
e.g. >>help fsolve

Goal Seek option in MS-Excel
32
Solution
x f(x)
1.96419 6.95E-05

Regula falsi or false position method
• Also named as method of false position or linear
interpolation method
Concerns in Bisection method
Convergence process in the bisection method is very slow
It depends only on the choice of end points of the interval [a,b].
The function f(x) does not have any role in finding the new bound value xm or xr
Condition of flase position method
 Function f(x) is a real and continuous between the interval xl and xu
 The functions f(xl) and f(xu) have opposite sign

i.e. if f(xl) f(xu) < 0

False position method - Basics
• A better approximation of xm or xr can be obtained by
taking the straight line or chord joining the points (xl, f(xl)
and xu, f(xu)) intersecting the x-axis.
Properties of similar triangle
D C
B C
E
A E
Two triangles ABC and EDC B C
AB BC f ( xl ) xm  xl D
=   A
ED DC f ( xu ) xm  xu

From the two similar
triangles, we get
0  f ( xl ) 0  f ( xu ) Triangle 1 Triangle 2

xm  xl xm  xu
Rearranging the above equation

 xm  xl  f  xu    xm  xu  f  xl 
xu f  xl   xl f  xu   xm  f  xl   f  xu 
xu f  xu   xl f  xu 
xm 
f  xl   f  xu 
(or)
f  xu  ( xl  xu )
xm  xU 
f  xl   f  xu 
Algorithm for false position method
Step - 1
Choose x L and xU as two guesses for the root such that f x L  f xU   0 , or in other
words, f  x  changes sign between x L and xU .
Step - 2
Estimate the new root, xr or xm , of the equation f x   0 from the derived

expression for false position method
One less function
f  xU  ( xL  xU ) evaluation than
xm (or) xr  xU  the other
f  xL   f  xU  expression
Step - 3
Check the condition
If f  x L  f  x m   0 , then the root lies between x L and x m ; then x L  x L and xU  x m
If f  x L  f  x m   0 , then the root lies between x m and xU ; then x L  x m and xU  xU
If f  x L  f  x m   0 , then the root is x m , Hence stop the algorithm

Step - 4
Find the new estimate of the root
f  xU  ( xL  xU ) This becomes xmold for next iteration and
xm  xU 
f  xL   f  xU  the newly calculated value becomes x mnew
Step - 5
Find the absolute relative approximate error

where
new old
x
m x m x mnew = estimated root from present iteration
a  new
100
x m x mold = estimated root from previous iteration
Compare the absolute relative approximate error a with the pre-specified relative error tolerance s
If a s , then go to Step 3, else stop the algorithm.
Note: Algorithmn in similar to bisection method except the formula to find the new
estimate.
CHE3001 Computational Methods in Process Engineering 37
Dr. Monash Purushothaman
Pitfalls of false position method
Suppose if the function

f(x) = x10-1
One-sided nature of
the expressions
i.e. one of the bound is
struck
One-sided nature of the expressions results in one of

Solution for the above situations
the bound to struck. In such cases the function value
Modified false position Method of the stagnant bound is divided in to half – to fasten
the convergence

When to use bisection or Regula falsi method
e.g. f(x) = xn –2
When n is small - false position method is often significantly faster than the
bisection method
When n becomes large - false position often becomes the slowest method
Solve the following example and analyse with method converges fast- why?
Example: f(x) = x2 – 2 with in the limits [1,2] for an accuracy of 0.001 %

Algorithm for false position method
Step - 1
Choose x L and xU as two guesses for the root such that f x L  f xU   0 , or in other
words, f  x  changes sign between x L and xU .
Step - 2
Estimate the new root, xr or xm , of the equation f x   0 from the derived

expression for false position method
One less function
f  xU  ( xL  xU ) evaluation than
xm (or) xr  xU  the other
f  xL   f  xU  expression
Step - 3
Check the condition
If f  x L  f  x m   0 , then the root lies between x L and x m ; then x L  x L and xU  x m
If f  x L  f  x m   0 , then the root lies between x m and xU ; then x L  x m and xU  xU
If f  x L  f  x m   0 , then the root is x m , Hence stop the algorithm

Finding roots of equation – Open methods
Bisection method False position method
xl  xu xu f  xu   xl f  xu 
xr  xr 
2 f  xl   f  xu 
f(xu)
Next estimate, xr Next estimate, xr f(xu)
xl
xl
xu
f(xl) xu
f(xl)
o Open methods Characteristics

 Fixed point iteration method Initial estimates need not bracket the root
 single starting value
 Two curve graphical method
 two starting values –need not bracket the root
 Newton-Raphson method
Generally converge faster
 Secant method NOT guaranteed to converge

Newton-Raphson method
o Very efficient method for real
functions
o Quadratic convergence: the

number of correct digits doubles
every iteration
o Possibility to not converge
o Requires the calculation of the

function's derivative:
First described by Issac Newton in 1669
Joseph Raphson published a simplified

version in 1690

Newton-Raphson method - Geometrical Derivation
o Slope of tangent at xi is
f ( xi )  0
f '( xi ) 
xi  xi 1 A
o Solve for xi+1

)θ
f ( xi )
xi 1  xi  C
f ( xi ) B
AB f ( xi ) f ( xi )
tan(    f '( xi )  xi 1  xi 
AC xi  xi 1 f ( xi )

Taylor series expansion
f ''  xi  2
f  xi 1   f  xi   f '  xi   xi 1  xi    i 1 i  
x  x
2!
n
f '''  xi  3 f  xi  n
 i 1 i 
x  x  ........   i 1 i   Rn
x  x
3! n!
Truncate the expansion for first order Taylor series expansion, we get
f  xi 1   f  xi   f '  xi   xi 1  xi 
Rearrange the above expression, we get
f ( xi )
xi 1  xi 
f ( xi )

f(x0)
The Newton-Raphson method
requires the calculation of the
derivative of a function, which
tangent at f(x0) is not always easy.
When the step is too large or

the value is oscillating, other
more conservative methods
should take over the case.
The
initial
guess
f(x1)

Pitfalls of Newton-Raphson method
Inflection point in the neighbor of a root Oscilation between maximum or minimum
Jumps in functions with several roots Existence of a null derivative

Secant method
False position method Assumptions
2 Two initial points xi and xi 1 such
3
that f ( xi )  f ( xi 1 ) , then the new
4
estimate obtained by backward finite
difference method
f  xi 1   f  xi 
f '( xi ) 
xi 1  xi
1
Rearrangement of the Secant method

approximation yields the xi 2
3
required iterative
expression
f  xi 1  xi 
xi 1  xi  4
f ( xi 1 )  f ( xi )
1x
i 1

AB DC f ( xi ) f ( xi 1 ) f ( xi )( xi  xi 1 )
  xi 1  xi 
AE DE xi  xi 1 xi 1  xi 1 f ( xi )  f ( xi 1 )

Secant method – advantage and disadvantage
o Advantage of the secant method
 It can converge even faster and it doesn’t need to bracket the root
o Disadvantage of the secant method
 It is not guaranteed to converge

 It may diverge
Modified secant method

( xi 1  xi )
Fractional
perturbation
xi f ( xi )
xi 1  xi 
f ( xi  xi )  f ( xi )

Algorithm
Step - 1
Evaluate f  x  symbolically Only for Newton-
Raphson method
Step - 2
Estimate the new value of the root, xi 1 , as

Use initial guess (xi) for NR Method Use initial guesses as (xi and xi-1) for
secant method
f ( xi )
xi 1  xi  f ( xi )( xi  xi 1 )
f ( xi ) xi 1  xi 
f ( xi )  f ( xi 1 )
Step - 3
Find the absolute relative approximate error a as
xi 1  xi
a  100
xi 1
If a > s , then go to Step 2, else stop the algorithm
Comparison of the root finding methods
Method Pros Cons
Bisection - Easy, Reliable, Convergent - Slow

- One function evaluation per iteration - Needs an interval [a,b] containing
- No knowledge of derivative is needed the root, i.e., f(a)f(b)<0
False - Easy, Reliable, Convergent - Faster than bisection method

position - One function evaluation per iteration - Needs an interval [a,b] containing
- No knowledge of derivative is needed the root, i.e., f(a)f(b)<0
Newton - Fast (if near the root) - May diverge

- Two function evaluations per iteration - Needs derivative and an initial guess
x0 such that f’(x0) is nonzero
Secant - Fast (slower than Newton) - May diverge

- One function evaluation per iteration - Needs two initial points guess x0, x1
- No knowledge of derivative is needed such that f(x0)- f(x1) is nonzero

Comparison – Example
Example: x-cos(x)=0, for13 correct digits requires
o 4 iterations of Newton [x0=0.8]
o 43 iterations of Bisection method (initial interval [0.6, 0.8]
o 5 iterations of Secant method [0.6, 0.8]
x-cos(x)=0

Case study - 1
Carbon dioxide gas (1 mole) is at 373 K and 50 atm. Find the
volume of the gas using Van der Waals equation.
 a 
 P  2  (v  b)  RT
 V 
Data - Van der Waals constants for carbon dioxide:
a = 3.61 L2 atm mol-2; b = 0.0428 L mol-1. R = 0.080206 L
atm/ mol K.
Specified error, εs = 0.001%.
Find the true percentage error for the obtained volume.

Case study
Carbon dioxide gas (1 mole) is at 300 K and 1 atm. Find the
volume of the gas using Van der Waals equation.
 a 
 P  2  (v  b)  RT
 V 
Data - Van der Waals constants for carbon dioxide:
a = 3.592 Pa m6/kg2; b = 0.04267 m2/kg. R = 0.080206 J/
kg.K.
Specified error, εs = 0.001%.

Case study
Determine the molar volume of saturated water in the liquid
and vapor phases at temperature of 523 K using Newton-
Raphson method. The molar volume is given by Vander
Waals equation of state.
 a 
 P  2  (v  b)  RT
 V 
The vapor pressure of water at 523 K at 40 bar (1 bar = 1×105

N/m2). Here, a = 0.5534 Nm4/mol2 and b = 3.049×10-5 m3/mol.
Take R= 8.314 Nm/mol.K. Stopping criterion, εs = 0.001%.

Case Study - 2
A 6 ft diameter spherical storage tank containing oil is attached
with a dipstick of 8 ft long as shown in the Figure.
The equation that gives the height, h, of the liquid in the

spherical tank for the given volume and radius is given by
3 2
f (h)  h  9h  3.8197  0
Use numerical method to find the height, h, to which the
dipstick is wet with oil. Specified error, εs = 0.5%.
Case Study
A 6 ft diameter spherical storage tank containing oil is attached
with a dipstick of 8 ft long as shown in the Figure.
General Equation
 3r  h 
2
V  πh 
 3 
Develop an equation that gives the height, h, of the liquid in the

spherical tank for the given volume and radius.
If r = 5 ft and volume = 400 ft3, find the height, h, to which the
dipstick is wet with oil. Specified error, εs = 0.5%.

Case study - 3
Suppose you are asked to cut a rectangular sheet with one of
its sides 1.25 m longer than the other and the area being
0.875 m2 from a thin iron sheet of 5 m2 area. What will be the
length of the ‘smallest side’?
Use bisection method to find the smallest side and compare
the result with analytical solution.

Case study - 4
In environmental engineering, the following equation can be
used to compute the oxygen level ‘c’ (mg/L) in a river
downstream from a sewage discharge:
c  10  20(e0.15x  e0.5x )
Determine the downstream distance, x, where the oxygen

level first falls to a reading of 5 mg/L using Secant method for
a specified error of 0.001%.

Case study
The concentration of pollutant bacteria c in a lake decreases
according to the following equation.
c  70e1.5t  25e0.075t
Determine the time required for the bacteria concentration to
be reduced to 9 using Newton-Raphson method with an initial
guess of t = 10 and a stopping criterion of 0.1%.

In chemical engineering, plug flow reactors are often used to convert reactants into
products. It has been determined that the efficiency of the conversion can
sometimes be improved by recycling a portion of the product stream so that it
returns to the entrance for an additional pass through the reactor.
The recycle rate is defined as R = volume of fluid returned to entrance/volume

leaving the system
Suppose that we are processing a chemical A to generate a product B. For the case
where A forms B according to an autocatalytic reaction (that is, in which one of the
products acts as a catalyst or stimulus for the reaction), it can be shown that an
optimal recycle rate must satisfy
where XAf = the fraction of reactant A that is converted to product B. The optimal
recycle rate corresponds to the minimum-sized reactor needed to attain the desired
level of conversion. Use a numerical method to determine the recycle ratios
needed to minimize reactor size for a fractional conversion of XAf = 0.9.

Case study
In a chemical engineering process, water vapor (H2O) is
heated to sufficiently high temperatures that a signifi cant
portion of the water dissociates, or splits apart, to form
oxygen (O2) and hydrogen (H2):
If it is assumed that this is the only reaction involved, the

mole fraction x of H2O that dissociates can be represented by
where K = the reaction equilibrium constant and pt = the total

pressure of the mixture.
If pt = 3 atm and K = 0.05, determine the value of x using
numerical method.
Other Methods
• Muller’s method
• Bairstow’s method
• Ridders’s method
• Brent method
• Lagurre’s method
• Jenkins-Traub method
• Ralston and Rabinowitz method

Module - 1

Hochgeladen von

Dokumentinformationen

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Module - 1

Hochgeladen von

Copyright:

Verfügbare Formate

CHE3001 Computational Methods in

Instructor: Dr. P. Monash

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 2

• Solve roots of a function, simultaneous equations,

• Develop MATLAB code for finding the roots of a function,

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 3

• Determine roots of a single equation and simultaneous equations

• Solve optimization, regression and numerical integration using

• Evaluate ordinary differential equation involving initial value and

• Estimate the solution for partial differential equation involving

• Create MATLAB program for roots finding, simultaneous

Solve f(x) = 0 for x.

Given the a’s and the c’s, solve

Determine x that gives optimum

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 6

solve for y as a function of t.

solve for u as a function of x and y

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 8

Quiz 1 Almost every day 40 10 Five minute

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 9

• All numerical methods have one common characteristic -

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 10

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 11

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 13

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 14

• Use the graphical approach to determine the drag

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 15

Iteration 1: [2, 18]

Since, the root lies between positive and negative function

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 18

o Type of functions in root finding

 e.g. Polynomials are simple class of algebraic functions

 Transcendental – nonalgebraic functions, which includes trigonometric, exponential,

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 19

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 20

Graphical solution of x2 – 3x – 4 = 0 Graphical solution of x2 – 4 = 0

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 21

Solution: The distance covered by the brick is given by

So the above expression becomes

This is of the form

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 22

If function f(x) is a real and

 If f(xl) and f(xu) have opposite

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 23

 If f(xl) and f(xu) does not change sign

 If f(xl) and f(xu) does not change sign

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 24

 If f(xl) and f(xu) changes sign between

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 25

Check which one requires more

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 26

o If a function f(x) is such that it just

e.g. f(x) =x2

o Function changes sign but root

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 27

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 28

If a s , then go to Step 3, else stop the algorithm.

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 29

>> x0=[1 3];

o Use “root” option in Mathcad

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 31

Dr. Monash Purushothaman CHE3001 Computational Methods in Process Engineering 32

Convergence process in the bisection method is very slow

It depends only on the choice of end points of the interval [a,b].