Scribd
Upload
80 views

Linear Systems: Prof Sung-Ho Hur

The document is a lecture on mathematical models of linear systems. It discusses transfer functions, which represent the output over input of a linear system in the Laplace domain. Transfer functions can be derived from differential equations. It also covers block diagram models, where systems are represented as blocks and the overall transfer function is determined by how the blocks are connected. Block diagrams can be reduced by combining blocks in series or parallel.

Uploaded by

Yousef Bilbeisi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
80 views

Linear Systems: Prof Sung-Ho Hur

The document is a lecture on mathematical models of linear systems. It discusses transfer functions, which represent the output over input of a linear system in the Laplace domain. Transfer functions can be derived from differential equations. It also covers block diagram models, where systems are represented as blocks and the overall transfer function is determined by how the blocks are connected. Block diagrams can be reduced by combining blocks in series or parallel.

Uploaded by

Yousef Bilbeisi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 30

Linear Systems

Prof Sung-ho Hur


Kyungpook National University

Lecture W2-2: Mathematical Models of Systems

S. Hur Automatic Control 1 / 30


Introduction

In the previous lecture, we studied:


• Differential Equations of Physical System
• Linearisation
• The Laplace Trasform

In this lecture, we will study:


• The Transfer Function of Linear Systems
• Block Diagram Models

S. Hur Automatic Control 2 / 30


Outline

1 Introduction

2 The Transfer Function of Linear Systems

3 Block Diagram Models

4 Summary

S. Hur Automatic Control 3 / 30


The Transfer Function of Linear Systems
A Transfer Function is the ratio of the output of a system to the input of a system
in the Laplace or frequency domain assuming zero initial conditions.
A non-stationary system, also known as time-varying system, has one or more
time-varying parameters, and the Laplace transformation cannot be utilised.
From the following spring-mass-damper system:

d 2 y (t) dy (t)
M 2
+b + ky (t) =r (t)
dt dt
we can derive

Ms 2 Y (s) + bsY (s) + kY (s) =R(s)

The corresponding transfer function is

Output Y (s) 1
= G (s) = = 2
Input R(s) Ms + bs + k

S. Hur Automatic Control 4 / 30


The Transfer Function of Linear Systems

The transfer function of the RC network


shown in the figure is obtained by writing
the Kirchhoff voltage equation, yielding
 
1
V1 (s) = R + I (s) (1)
Cs

and the output voltage is


 
1
V2 (s) = I (s) (2)
Cs

Solving (1) for I (s) and substituting in (2), we have

(1/Cs)V1 (s)
V2 (s) = (3)
R + 1/Cs

S. Hur Automatic Control 5 / 30


The Transfer Function of Linear Systems

Then the transfer function is obtained as the ratio V2 (s)/V1 (s), which is

V2 (s) 1 1
G (s) = = =
V1 (s) RCs + 1 τs + 1

where τ is the time constant and the single pole is −1/τ .

S. Hur Automatic Control 6 / 30


The Transfer Function of Linear Systems

Considering the following second order dynamic system:

p(s) p1 s + p0
Y (s) = G (s)R(s) = R(s) = 2
q(s) s + q1 s + q0

p(s) p(s) n(s)


Y (s) = R(s) = = Y1 (s) + Y2 (s)
q(s) q(s) d(s)

where Y1 (s) is the partial fraction expansion of the terms involving factors of q(s)
and Y2 (s) is the partial fraction expansion of the terms involving factors of d(s).

S. Hur Automatic Control 7 / 30


The Transfer Function of Linear Systems

Taking the inverse Laplace transform yields

y (t) = y1 (t) + y2 (y )

Example 1: Consider the following differential equation:

d 2y dy
+4 + 3y = 2r (t)
dt 2 dt
with r (t) = 1 and assuming zero initial conditions.
Taking the Laplace transform

s 2 Y (s) + 4sY (s) + 3Y (s) = 2R(s)

S. Hur Automatic Control 8 / 30


The Transfer Function of Linear Systems
We obtain
2
Y (s) = R(s)
(s 2 + 4s + 3)

where R(s) = 1/s and q(s) = s 2 + 4s + 3 = (s + 1)(s + 3) = 0 is the


characteristic equation.
Then the partial fraction expansion yields
−1 1/3 2/3
Y (s) = + +
s +1 s +3 s
Now taking the inverse Laplace transform using the Laplace transform table, we
have
1 2
y (t) = −1e −t + e −3t +
3 3
The steady state response according to the final value theorem is:
2
lim y (t) =
t→∞ 3
S. Hur Automatic Control 9 / 30
The Transfer Function of Linear Systems

S. Hur Automatic Control 10 / 30


The Transfer Function of Linear Systems

Figure 2: Two-node electric circuit analog.


Figure 1: Two-mass mechanical system.

Example 2: Consider the mechanical system shown in Figure 1 and its electrical
circuit analog shown in Figure 2.

S. Hur Automatic Control 11 / 30


The Transfer Function of Linear Systems
The electrical circuit analog is a force-current analog; that is, the velocities v1 (t)
and v2 (t) of the mechanical system are directly analogous to the node voltages
v1 (t) and v2 (t) of the electrical circuit.
The simultaneous equations, assuming that the initial conditions are zero, are

M1 sV1 (s) + (b1 + b2 )V1 (s) − b1 V2 (s) = R(s)


V2 (s)
M2 sV2 (s) + b1 (V2 (s) − V1 (s)) + k =0
s
These equations are obtained using the force equations for the mechanical system
of Figure 1.
Rearranging the equations, we obtain

(M1 s + (b1 + b2 ))V1 (s) + (−b1 )V2 (s) = R(s)


 
k
(−b1 )V1 (s) + M2 s + b1 + V2 (s) = 0
s

or in the matrix form:


S. Hur Automatic Control 12 / 30
The Transfer Function of Linear Systems
" #" # " #
M1 s + b1 + b2 −b1 V1 (s) R(s)
k
=
−b1 M2 s + b1 + s
V2 (s) 0
Assuming that the velocity of M1 is the output variable, we solve for V1 (s) by
Cramer’s rule to obtain
(M2 s + b1 + k/s)R(s)
V1 (s) =
(M1 s + b1 + b2 )(M2 s + b1 + k/s) − b12
Then the transfer function of the mechanical (or electrical) system is
V1 (s) (M2 s + b1 + k/s)
G (s) = =
R(s) (M1 s + b1 + b2 )(M2 s + b1 + k/s) − b12
(M2 s 2 + b1 s + k)
=
(M1 s + b1 + b2 )(M2 s 2 + b1 s + k) − b12 s
If the transfer function in terms of the position x1 (t) is desired, then we have
X1 (s) V1 (s) G (S)
= =
R(s) sR(s) s
S. Hur Automatic Control 13 / 30
Outline

1 Introduction

2 The Transfer Function of Linear Systems

3 Block Diagram Models

4 Summary

S. Hur Automatic Control 14 / 30


Block Diagram Models

S. Hur Automatic Control 15 / 30


Block Diagram Models

S. Hur Automatic Control 16 / 30


Block Diagram Models

S. Hur Automatic Control 17 / 30


Block Diagram Models

S. Hur Automatic Control 18 / 30


Block Diagram Models

S. Hur Automatic Control 19 / 30


Block Diagram Models

S. Hur Automatic Control 20 / 30


Block Diagram Models

forward path
closed-loop TF =
1 + open-loop TF

Forward path: Gc (s)Ga (s)G (s).


Open-loop transfer function (TF): Gc (s)Ga (s)G (s)H(s).

Y (s) Gc (s)Ga (s)G (s)


=
R(s) 1 + Gc (s)Ga (s)G (s)H(s)

S. Hur Automatic Control 21 / 30


Block Diagram Models

S. Hur Automatic Control 22 / 30


Block Diagram Models

S. Hur Automatic Control 23 / 30


Block Diagram Models

Example 3: Reduce the block diagram in the figure.

S. Hur Automatic Control 24 / 30


Block Diagram Models

S. Hur Automatic Control 25 / 30


Block Diagram Models

Example 4: Reduce the block diagram in the figure.

S. Hur Automatic Control 26 / 30


Block Diagram Models

S. Hur Automatic Control 27 / 30


Block Diagram Models

G G
1 2
 
Y (s) G6 G1 G2 G5 + G1 G6
T (s) = = 1−G 1 G3
G1 G2 G4
G5 + =
R(s) 1 + 1−G G2 1 − G1 G3 + G1 G2 G4
1 G3

S. Hur Automatic Control 28 / 30


Outline

1 Introduction

2 The Transfer Function of Linear Systems

3 Block Diagram Models

4 Summary

S. Hur Automatic Control 29 / 30


Summary

What have we learnt today?


The Transfer Function of Linear Systems
• How to derive the transfer function from a differential equation.

Block Diagram Models


• Series Interconnection
• Parallel Interconnection
• Feedback
• Block Diagram Reduction

Next week, we will study


• State Variable Models

S. Hur Automatic Control 30 / 30

You might also like