FEEDBACK CONTROL

SYSTEMS
SOLUTION TO LCCDE
LINEAR DIFFERENTIAL
EQUATIONS WITH
CONSTANT-
COEFFICIENTS (LCCDE)
BY DIRECT METHOD

• Homogeneous Solution
• Particular Solution
• Total Solution
LINEAR CONSTANT-COEFFICIENT DIFFERENTIAL
EQUATION

n
d y i
 ai dti  f (t)
i0

where:
ai  constants

LCCDE is also called Linear Time-Invariant (LTI) Differential Equation

2
SOLUTIONS TO LCCDE
• Classical Solution/Direct Method
y(t)  yc (t)  y p (t)
Roots of Method of
Characteristic Undetermined
Polynomial Coefficients

• Laplace Transform
LCCDE Laplace Y(s) Inverse y(t)
Laplace
Transform
Transform
 CLASSICAL SOLUTION

Solution to
n
diy
 ai dti  f (t)
i0
is of the form

y(t)  yc (t)  y p (t)

where:
yc (t)  Characteristic solution (Homogeneous)
y p (t)  Particular solution
4
HOMOGENEOUS SOLUTION

• Steps in obtaining the Homogeneous Solution

1. Equate f(t) to zero.
2. Create an auxiliary (characteristic) equation from the differential
equation.
3. Solve for the roots of the auxiliary equation.
4. Write the homogeneous solution from the corresponding roots.

6
Example:
CASE I: Real & Distinct Roots
Determine the Homogeneous Solution of the Differential Equation

d3 y d 2 y dy
4   6 y  3e 4t

dt 3 dt 2 dt

Solution: Equating the right side to zero and writing the

auxiliary equation
m  4m  m  6  0
3 2
7
Example:

CASE I: Real & Distinct Roots

By using Synthetic Division or other root-finding method

m   1, 2 ,3
The Homogeneous Solution is of the form

yc (t)  c1et  c2 e2t  c3e3t

8
Example:
• CASE II: Real & Repeated Roots
Determine the Homogeneous Solution of the Differential Equation

d4y d3y d2y

2  2  4t
dt 4
dt 3
dt
Solution: Equating the right side to zero and writing the auxiliary
equation

m 4  2m3 m 2  0
9
Example:
CASE II: Real & Repeated Root
By using Synthetic Division or other root-
• finding method
• m  0,0,1,1
The Homogeneous Solution is of the form
y (t)  c e (0)t  c te(0)t  c e t  c tet
c 1 2 3 4

or
y (t)  c  c t  c et  c tet
c 1 2 3 4 9
CASE III: Complex Conjugate Roots

For complex-conjugate roots m = a ± jb, the Homogeneous

Solution may be written as in Case I

yc (t)  c1e (a jb)t

 c2 e (a jb)t

or
y (t)  eat (c cos bt c sin bt)
c 3 4
where: c3  c1  c 2
c4  j(c1  c2 )
10
Example:
CASE III: Complex Conjugate Roots
Determine the Homogeneous Solution of the Differential Equation

d2y dy
 4 13y  e t
dt 2
dt
Solution: Equating the right side to zero and writing the
auxiliary equation

m2  4m 13  0
11
Example:
• CASE III: Complex Conjugate Roots
By using Quadratic Formula

• m  2 j3
The Homogeneous Solution is of the form
yc (t)  c1 e (2 j3)t  c2e (2 j 3)t
or
y (t)  e2t (c cos 3t c sin 3t)
c 3 4
12
 MATLAB IMPLEMENTATION
Using MatLab we can find the roots of a polynomial given its
coefficients using
roots(C)

where: C = coefficient vector

Example: Find the roots of m3  4m2  m  6  0

Solution: >>roots([1 -4 1 6])
ans =
3.0000
2.0000
13
-1.0000
PARTICULAR SOLUTION

Steps in obtaining the Particular Solution

1. Determine the form of f(t).
2. Write the particular solution corresponding to the
given f(t).
3. Using Method of Undetermined Coefficients, solve for the
constants by evaluating the left side of the differential
equation.

15
PARTICULAR SOLUTIONS OF SOME FUNCTIONS

f (t) y p (t)
A K
At K1t  K 2
At 2
K1t 2  K 2t  K 3
Acos t K1 cost  K 2 sin t
Asin t K1 cost  K 2 sin t
Aebt Kebt
16
Example:
Find the Particular Solution of the Differential Equation
d 2 y dy
  2 y  2t  40 cos 2t
dt 2 dt
Solution: From the Table, the Particular solution corresponding to
f (t)  2t  40 cos 2t

y p (t)  K1t  K 2  K3 cos 2t  K 4 sin 2t

17
Example:
In order to evaluate the left-side of the D.E., we will need the
first and second derivatives of yp

y p (t)  K1t  K 2  K3 cos 2t  K 4 sin 2t

dy p (t)
 K1  2K 3 sin 2t  2K 4 cos 2t
dt
d 2 y p (t)
2
 4K3 cos 2t  4K 4 sin 2t
dt 17
Example:
The given D.E. can be written as
(2K1)t (K1 2K2)(6K3 2K4)cos2t
(2K3 6K4 )sin2t  2t 40cos2t

Equating coefficients we can have the following relationships to solve for K1,K2,K3 and K4

2K1  2 (eq.1)  6K 3  2K 4  40 (eq.3)

K1 2K2 0 (eq.2)  2K3  6K 4  0 (eq.4)

18
Example:
We can find K1 and K2 from eq.1 and eq.2
2 K1 1
K1   1 K2  
2 2 2
We can find K3 and K4 from eq.3 and eq.4.
using Cramer’s Rule
 6 2  K 3    40 

 2
  6 K 4   0 
 

K3  6 K 4  2
20
Example:

By substituting the constants, the particular

solution will be

1
y p (t)  t   6 cos 2t  2sin 2t
2

21
 MATLAB IMPLEMENTATION
Using MatLab we can find the Solutions to a System of Linear
Equations Ax=C given the Coefficients and Constants

x=inv(A)*C
where
C=constant vector
A=coefficient matrix
*=matrix multiplication
21
MATLAB IMPLEMENTATION

Example: Find the values of x1, x2 , x3

23
MATLAB IMPLEMENTATION
Solution:

>>A=[4 -2 1;-3 -1 4; 1 -1 3];

>>C=[15; 8; 13];
>> x=inv(A)*C
x =
2.0000 -2.0000 3.0000

24
EXCEL IMPLEMENTATION
Create an array containing the Coefficients
and Constants

Coefficient Matrix Constants

Variables 24
EXCEL IMPLEMENTATION

Take the inverse of the coefficient matrix

Inverse of Coefficient Matrix

25
 EXCEL IMPLEMENTATION
How to take the inverse of a matrix?
1. Put the cursor where the first row, first column element should be placed. For
example C7.
2. Select MINVERSE from the Insert Function menu.
3. Select the coefficient matrix or type the upper-leftmost: lower right-most elements of
the coefficient matrix. For example B2:D4.
4. The inverse of a 3x3 matrix is a 3x3 matrix but after executing Procedure (3) only
C7 is filled up. To show all the elements of the inverse, select a 3x3 range
starting from C7. For example C7:E9. Press F2 then press CTRL+SHIFT+ENTER .

26
EXCEL IMPLEMENTATION
Take the Product of the Inverse of Coefficient Matrix and the
Constants

Product of the Inverse

and Constants
(Values of X1,X2,X3)

27
 EXCEL IMPLEMENTATION
How to take the product of 2 matrices?
1. Put the cursor where the first row, first column element should be placed. For
example C11.
2. Select MMULT from the Insert Function menu.
3. Select the inverse of the coefficient matrix or type the upper-leftmost: lower right-
most elements of the inverse of the coefficient matrix for Array 1. For example
C7:E9. For Array 2 select the constants. For example F2:F4.
4. The product of a 3x3 matrix and a 3x1 matrix is a 3x1 matrix but after executing
Procedure (3) only C11 is filled up. To show all the elements of the product, select a
3x1 range starting from C11. For example C11:C13. Press
F2 then press CTRL+SHIFT+ENTER.
28
TOTAL SOLUTION
After solving for the
Homogeneous and Particular Solutions

y(t)  yc (t)  y p (t)

The final task will be to determine the constants from the
Homogeneous Solution by evaluating the Total Solution at
given Initial Conditions

30
INITIAL CONDITIONS
dy(t) d 2 y(t) d m y(t)
• Values of y(t), , 2
... at t=0.
dt dt dt m

We will assume that all initial conditions are zero.

dy(t) d 2 y(t) d m y(t)
y(t)   2
... m
0
dt dt dt
Example:
Find the Total Solution of the Differential Equation (Assume all
initial conditions=0)
d 2 y dy
  2 y  2t  40 cos 2t
dt 2 dt
Solution: First, we will solve for the Homogeneous Solution by
getting the roots of the auxiliary polynomial

m2  m  2  0
32
Example:

By using Synthetic Division or other root-finding method

m 1,2
Case I: Real and Repeated Roots.
The Homogeneous Solution is of the form

y (t)  c et  c e2t
c 1 2
Example:
After determining the Homogeneous Solution, we will find the
Particular Solution, which is
1
y p (t)  t   6 cos 2t  2 sin 2t
2
(from previous example)

The Total Solution will be

2t 1
y(t)  c1e  c2e
t
 t   6 cos 2t  2sin 2t
2
34
Example:
To determine c1 and c2, we will evaluate y(t) at t=0
1
y(0)  c1e0  c2 e0  0   6 cos(0)  sin(0)  0
2
11
c1  c2   (eq.1)
2

and dy(t) at t=0

dy(t) dt2t
 c1e  2c2e 1 2sin 2t  2 cos 2t
t

dt
dy(0)
 c1e 0  2c2e0 1 2sin(0)  2 cos(0)  0
dt

c1  2c2  5 (eq.2) 34
Example:
Using eq.1 and eq.2 we can solve for c1 and c2

7
c1  2 c2  
2
The Total Solution will be
7 1
y(t)  2et  e 2t  t   6 cos 2t  2sin 2t
2 2
REFERENCES:

• Elementary Differential Equations,

7thedition,by Rainville E.D.,and Bedient P.E.
• Schaum’s Outline Series “Feedback and Control
Systems”, 2nd edition, by DiStefano III, J.J.,
Stubberud A.R., and Williams I.J.
• Control Systems Engineering, 3rd edition, by Nise
N.S.

37