APPLICATIONS OF

MATLAB IN ENGINEERING
Yan-Fu Kuo Fall 2015
Dept. of Bio-industrial Mechatronics Engineering
National Taiwan University

Today:
• Symbolic approach
• Numeric root solvers
• Recursive functions
Applications of MATLAB in Engineering Y.-F. Kuo 2

Problem Statement
• Suppose you have a mathematical function 𝑓(𝑥)
and you want to find 𝑥0 such that 𝑓 𝑥0 = 0, e.g.
𝑓 𝑥 = 𝑥 2 − 2𝑥 − 8 = 0
• How do you solve the 100

problem using MATLAB?

80

• Analytical Solutions
60

f(x)
• Graphical Illustration
40

• Numerical Solutions
20

-10 -5 0 5 10
x
Applications of MATLAB in Engineering Y.-F. Kuo 3

Symbolic Root Finding Approach

• Performing mathematics on symbols, NOT numbers
• The symbols math are performed using “symbolic
variables”
• Use sym or syms to create symbolic variables

syms x x=sym('x');
x + x + x x + x + x
(x + x + x)/4 (x + x + x)/4

• Define: 𝑦 = 𝑥 2 − 2𝑥 − 8
Applications of MATLAB in Engineering Y.-F. Kuo 4

Symbolic Root Finding: solve()

• Function solve finds roots for equations
𝑦 = 𝑥 ∙ sin(𝑥) − 𝑥 = 0
syms x syms x
solve('x*sin(x)-x', x) y = x*sin(x)-x;
solve(y, x)

• Find the roots for:

cos(𝑥)2 − sin(𝑥)2 = 0 and cos(𝑥)2 + sin(𝑥)2 = 0
Applications of MATLAB in Engineering Y.-F. Kuo 5

Solving Multiple Equations

• Solve this equation using symbolic approach:
𝑥 − 2𝑦 = 5
𝑥+𝑦 =6

syms x y
eq1 = x - 2*y - 5;
eq2 = x + y - 6;
A = solve(eq1,eq2,x,y)
Applications of MATLAB in Engineering Y.-F. Kuo 6

Solving Equations Expressed in Symbols

• What if we are given a function expressed in
symbols?
𝑎𝑥 2 − 𝑏 = 0
syms x a b
solve('a*x^2-b')

• 𝑥 is always the first choice to be solved

• What if one wants to express 𝑏 in terms of 𝑎 and 𝑥?

syms x a b
solve('a*x^2-b', 'b')
Applications of MATLAB in Engineering Y.-F. Kuo 7

Exercise
• Solve this equation for 𝑥 using symbolic approach
2 2
𝑥−𝑎 + 𝑦−𝑏 = 𝑟2

• Find the matrix inverse using symbolic approach

𝑎 𝑏
𝑐 𝑑
Applications of MATLAB in Engineering Y.-F. Kuo 8

Symbolic Differentiation: diff()

• Calculate the derivative of a symbolic function:
𝑦 = 4𝑥 5
syms x
y = 4*x^5;
yprime = diff(y)

• Exercise:
𝑥 2
𝑒 𝑑𝑓
𝑓 𝑥 = , =?
−𝑥+3 𝑥3 𝑑𝑥
𝑥 2 + 𝑥𝑦 − 1 𝜕𝑓
𝑔 𝑥 = 3 , =?
𝑦 +𝑥+3 𝜕𝑥
Applications of MATLAB in Engineering Y.-F. Kuo 9

Symbolic Integration:
• Calculate the integral of a symbolic function:

𝑧= 𝑦𝑑𝑥 = 𝑥 2 𝑒 𝑥 𝑑𝑥 , 𝑧(0) = 0

syms x; y = x^2*exp(x);
z = int(y); z = z-subs(z, x, 0)

• Exercise:
10
𝑥2 − 𝑥 + 1
𝑑𝑥
0 𝑥+3
 Applications of MATLAB in Engineering Y.-F. Kuo 10

Symbolic vs. Numeric

Advantages Disadvantages
Symbolic • Analytical solutions • Sometimes can't be solved
• Lets you intuit things about • Can be overly complicated
solution form
Numeric • Always get a solution • Hard to extract a deeper
• Can make solutions understanding
accurate
• Easy to code
Applications of MATLAB in Engineering Y.-F. Kuo 11

Review of Function Handles (@)

• A handle is a pointer to a function
• Can be used to pass functions to other functions
• For example, the input of the following function is
another function:
function [y] = xy_plot(input,x)
% xy_plot receives the handle of a function and plots that
% function of x
y = input(x); plot(x,y,'r--');
xlabel('x'); ylabel('function(x)');
end

• Try: xy_plot(@sin,0:0.01:2*pi);
 Applications of MATLAB in Engineering Y.-F. Kuo 12

Using Function Handles

xy_plot(@sin,0:0.01:2*pi); xy_plot(@atan,0:0.01:2*pi);

1 1.5

0.8

0.6

0.4
1
0.2
function(x)

function(x)
0

-0.2
0.5
-0.4

-0.6

-0.8

-1 0
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
x x
Applications of MATLAB in Engineering Y.-F. Kuo 13

fsolve()
• A numeric root solver
• For example, solve this equation:
𝑓 𝑥 = 1.2𝑥 + 0.3 + 𝑥 ∙ sin(𝑥)

f2 = @(x) (1.2*x+0.3+x*sin(x));
fsolve(f2,0)

A function handle Initial guess

Applications of MATLAB in Engineering Y.-F. Kuo 14

Exercise
• Find the root for this equation :
2𝑥 − 𝑦 − 𝑒 −𝑥
𝑓 𝑥, 𝑦 =
−𝑥 + 2𝑦 − 𝑒 −𝑦
using initial value 𝑥, 𝑦 = (−5, −5)
Applications of MATLAB in Engineering Y.-F. Kuo 15

fzero() 4

3.5
• Anther numeric root solver 3

2.5
• Find the zero if and only if
2
the function crosses the 1.5
x-axis 1

0.5
f=@(x)x.^2
0
fzero(f,0.1) -2 -1 0 1 2

fsolve(f,0)

Number of iterations Tolerance

• Options:
f=@(x)x.^2
options=optimset('MaxIter',1e3,'TolFun',1e-10);
fsolve(f,0.1,options)
fzero(f,0.1,options)
Applications of MATLAB in Engineering Y.-F. Kuo 16

Finding Roots of Polynomials: roots()

• Find the roots of this polynomial:
𝑓(𝑥) = 𝑥5 − 3.5𝑥4 + 2.75𝑥3 + 2.125𝑥2 − 3.875𝑥 + 1.25

roots([1 -3.5 2.75 2.125 -3.875 1.25])

• roots only works for polynomials

• Find the roots of the polynomial:

𝑓 𝑥 = 𝑥3 − 6𝑥2 − 12𝑥 + 81
Applications of MATLAB in Engineering Y.-F. Kuo 17

How Do These Solvers Find the Roots?

• Now we are going to introduce more details of
some numeric methods
Applications of MATLAB in Engineering Y.-F. Kuo 18

Numeric Root Finding Methods

• Two major types:
• Bracketing methods (e.g., bisection method)
Start with an interval that contains the root
• Open methods (e.g., Newton-Raphson method)
Start with one or more initial guess points

• Roots are found iteratively until some criteria are

satisfied:
• Accuracy
• Number of iteration
Applications of MATLAB in Engineering Y.-F. Kuo 19

Bisection Method (Bracketing)

Assumptions:
• 𝑓(𝑥) continuous on [𝑙, 𝑢]
• 𝑓(𝑙) ∙ 𝑓(𝑢) < 0

Algorithm:
Loop
1. 𝑟 = (𝑙 + 𝑢)/2
2. If 𝑓(𝑟) ∙ 𝑓(𝑢) < 0 then new interval [𝑟, 𝑢]
If 𝑓(𝑙) ∙ 𝑓(𝑟) < 0 then new interval [𝑙, 𝑟]
End
Applications of MATLAB in Engineering Y.-F. Kuo 20

Bisection Algorithm Flowchart

Start:
Given 𝑙, 𝑢

Calculate 𝑓(𝑙) 𝑓(𝑢)

Calculate
𝑙+𝑢 no
𝑟= 2
𝑓(𝑟)

yes
(𝑢 − 𝑙) < 𝜀 Stop
yes no
𝑓(𝑟) ∙ 𝑓(𝑢) < 0

𝑙=𝑟 𝑢=𝑟
Applications of MATLAB in Engineering Y.-F. Kuo 21

Newton-Raphson Method (Open)

Assumption:
f(x)
• 𝑓(𝑥) continuous y
• 𝑓′(𝑥) known

f(x00)
f(x11)

Algorithm: x22
Loop f(x22) x11 x00 x
𝑓 𝑥𝑛
𝑥𝑛+1 = 𝑥𝑛 −
𝑓′ 𝑥𝑛
End
Applications of MATLAB in Engineering Y.-F. Kuo 22

Newton-Raphson Algorithm Flowchart

Start:
Given 𝑥0

Calculate 𝑓(𝑥0 ), 𝑛 = 0

no
Calculate
𝑓 𝑥𝑛
𝑥𝑛+1 = 𝑥𝑛 − ′ yes
𝑓 𝑥𝑛 𝑓(𝑥0 ) < 𝜀 Stop
𝑛 =𝑛+1
Applications of MATLAB in Engineering Y.-F. Kuo 23

Bisection vs. Newton-Raphson

Bisection • Reliable
• No knowledge of derivative is needed
• Slow
• One function evaluation per iteration
• Needs an interval [a,b] containing the root, f(a)·f(b)<0
Newton • Fast but may diverge
• Needs derivative and an initial guess x0, f’(x0) is
nonzero
Applications of MATLAB in Engineering Y.-F. Kuo 24

Recursive Functions
• Functions that call themselves
• Example, factorial of an integer n
𝑛! = 1 × 2 × 3 × ⋯ × 𝑛
• A factorial can be defined in terms of another
factorial:
𝑛! = 𝑛 × (𝑛 − 1)!
= 𝑛 × (𝑛 − 1) × (𝑛 − 2)!
= 𝑛 × (𝑛 − 1) × (𝑛 − 2) × (𝑛 − 3)!
=𝑛× 𝑛−1 × 𝑛−2 ×⋯
Applications of MATLAB in Engineering Y.-F. Kuo 25

Factorial Recursive Function

• The function includes a recursive case and a base
case
• The function stops when it reaches the base case

function output = fact(n)

% fact recursively finds n!
if n==1
output = 1; Base case
else
output = n * fact(n-1); Recursive case
end
end
Applications of MATLAB in Engineering Y.-F. Kuo 26

End of Class