Systems Analysis and Control
Matthew M. Peet
Arizona State University
Lecture 6: Calculating the Transfer Function
Introduction
In this Lecture, you will learn: Transfer Functions
• Transfer Function Representation of a System
• State-Space to Transfer Function
• Direct Calculation of Transfer Functions
Block Diagram Algebra
• Modeling in the Frequency Domain
• Reducing Block Diagrams
M. Peet Lecture 6: Control Systems 2 / 23
Previously:
The Laplace Transform of a Signal
Definition: We defined the Laplace transform of a Signal.
• Input, û = L(u).
• Output, ŷ = L(y)
Theorem 1.
Any bounded, linear, causal, time-invariant system, G, has a Transfer
Function, Ĝ, so that if y = Gu, then
ŷ(s) = Ĝ(s)û(s)
There are several ways of finding the Transfer Function.
M. Peet Lecture 6: Control Systems 3 / 23
Transfer Functions
Example: Simple System
State-Space:
ẋ(t) = −x(t) + u(t)
y(t) = x(t) − .5u(t) x(0) = 0
Apply the Laplace transform to the first equation:
L ẋ(t) = −x(t) + u(t) which gives sx̂(s) = −x̂(s) + û(s).
Solving for x̂(s), we get
1
(s + 1)x̂(s) = û(s) and so x̂(s) = û(s).
s+1
Similarly, the second equation yields:
1 1 − .5(s + 1) 1s−1
ŷ(s) = x̂(s) − .5û(s) = û(s) − .5û(s) = û(s) = û(s)
s+1 s+1 2s+1
Thus we have the Transfer Function:
1s−1
Ĝ(s) =
2s+1
M. Peet Lecture 6: Control Systems 4 / 23
Transfer Functions
Example: Step Response
The Transfer Function provides a convenient way to find the response to inputs.
1
Step Input Response: û(s) = s
1s−11 1 s−1
ŷ(s) = Ĝ(s)û(s) = =
2s+1s 2 s2 + s
1 2 1
= −
2 s+1 s
Consulting our table of Laplace
Transforms, Step Response
0.5
0.4
1 −1 2 1 1
y(t) = L − L−1 0.3
2 s+1 2 s 0.2
Amplitude
0.1
−t 1
= e − 1(t) 0
2 −0.1
−0.2
−0.3
−0.4
−0.5
0 1 2 3 4 5 6
Time (sec)
M. Peet Lecture 6: Control Systems 5 / 23
Transfer Functions
Example: Sinusoid Response
input
1
1
Sine Function: û(s) = s2 +1
0.8
0.6
0.4
1s−1 1 0.2
ŷ(s) = Ĝ(s)û(s) =
2 s + 1 s2 + 1
u(t)
0
−0.2
1 s−1 −0.4
=
2 s3 + s2 + s + 1 −0.6
−0.8
1 s 1 −1
= − 0 2 4
t
6 8 10
2 s2 + 1 s + 1
output
0.6
Consulting our table of Laplace Transforms, 0.4
0.2
1 1
y(t) = cos t − e−t 0
2 2
y(t)
−0.2
−0.4
−0.6
−0.8
0 2 4 6 8 10
t
Note that this is the same answer we got by integration in Lecture 4.
M. Peet Lecture 6: Control Systems 6 / 23
Inverted Pendulum Example
Return to the pendulum.
Dynamics:
M gl 1
θ̈(t) = θ(t) + T (t)
2J J
y(t) = θ(t)
For the first equation,
M gl 1
s2 θ̂(s) = θ̂(s) + T̂ (s)
2J J
Solve for θ̂(s):
1 1
θ̂(s) = T̂ (s)
J s2 − M2Jgl
Second Equation: ŷ(s) = θ̂(s)
Transfer Function:
ŷ(s) 1 1
Ĝ(s) = =
T̂ (s) J s2 − M2Jgl
M. Peet Lecture 6: Control Systems 7 / 23
Inverted Pendulum Example: Impulse Response
5 Impulse Response
x 10
18
Impulse Input: û(s) = 1 16
14
1 1 12
Amplitude
ŷ(s) = Ĝ(s)û(s) = 10
J s − M2Jgl
2 8
1 1
= q q 2
J (s − M gl )(s + M gl ) 0
0 5
Time (sec)
10 15
2J 2J
Figure: Impulse Response with
s
1 2J 1 1 g = l = J = 1, M = 2
= q − q
J M gl s − M gl s+ M gl
2J 2J
In time-domain:
2J √ M gl t √ M gl
s
1
y(t) = e 2J − e− 2J t
J M gl
Pendulum Accelerates to infinity!
M. Peet Lecture 6: Control Systems 8 / 23
Constructing the Transfer Function: Suspension System
x1
mc
x2
mw
Recall the dynamics:
K1 c K1 c
z̈1 (t) = − z1 (t) − ż1 (t) + z2 (t) + ż2 (t)
mc mc m mc
c
K1 c K1 K2 c K2
z̈4 (t) = z1 (t) + ż1 (t) − + z2 (t) − ż2 (t) − u(t)
mw mw mw mw mw mw
y(t) = z2 (t)
M. Peet Lecture 6: Control Systems 9 / 23
Constructing the Transfer Function: Suspension System
x1
mc
x2
mw
Apply the Laplace Transform to the dynamics:
K1 c K1 c
s2 ẑ1 (s) = − ẑ1 (s) − sẑ1 (s) + ẑ2 (s) + sẑ2 (s)
mc mc m mc
c
K1 c K1 K2 c K2
s2 ẑ2 (s) = ẑ1 (s) + sẑ1 (s) − + ẑ2 (s) − sẑ2 (s) − û(s)
mw mw mw mw mw mw
ŷ(s) = ẑ2 (s)
M. Peet Lecture 6: Control Systems 10 / 23
Constructing the Transfer Function: Suspension System
We isolate the z1 and z2 terms:
c K1 K1 c
s2 + s+ ẑ1 (s) = + s ẑ2 (s)
mc mc mc mc
c K1 K2 K1 c K2
s2 + s+ + ẑ2 (s) = + s ẑ1 (s) − û(s)
mw mw mw mw mw mw
ŷ(s) = ẑ2 (s)
Which yields
K1 c
mc + mc s
ẑ1 (s) = ẑ2 (s)
c K1
s2 + mc s + mc
K1 c K2
mw + mw s mw
ẑ2 (s) = c K1 K2
ẑ1 (s) − c K1 K2
û(s)
s2 + mw s + mw + mw s2 + mw s + mw + mw
M. Peet Lecture 6: Control Systems 11 / 23
Constructing the Transfer Function: Suspension System
Now we can plug in for ẑ1 and solve for ẑ2 :
ẑ2 (s) =
K2 (mc s2 + cs + K1 )
û(s)
mc mw s4 + c(mw + mc )s3 + (K1 mc + K1 mw + K2 mc )s2 + cK2 s + K1 K2
Compare to the State-Space Representation:
z1 0 1 0 0 z1
0
K1 c K1 c
d − mc − mc
z2 (t) = (t) + 0 u(t)
z2
mc mc
0 0 0 1 0
dt z3
z3
K c K K c K2
z4 mw
1
mw − mw + mw
1 2
− mw z4 −m w
z1
1 0 0 0 z2 (t) + 0 u(t)
y(t) =
0 0 1 0 z3 0
z4
M. Peet Lecture 6: Control Systems 12 / 23
Block Diagram Algebra
Series (Cascade) Interconnection
The interconnection of systems can be represent by block diagrams.
u y1 y
G H
Cascade of Systems: Suppose we have two systems: G and H.
Definition 2.
The Cascade or Series interconnection of two systems is
y1 = Gu y = Hy1
or
y = H(G(u))
M. Peet Lecture 6: Control Systems 13 / 23
Block Diagram Algebra
Series Connection (Cascade)
u(s) y(s)
u(s) y1(s) y(s) H(s)G(s)
G(s) H(s)
Series Interconnection:
• The output of G is the input to H.
• Let Ĝ(s) and Ĥ(s) be the transfer functions for G and H.
• Then
ŷ1 (s) = Ĝ1 (s)û(s) ŷ(s) = Ĥ(s)ŷ1 (s) = Ĥ(s)Ĝ(s)û(s)
• The Transfer Function, T̂ (s) for the combination of G and H is
T̂ (s) = Ĥ(s)Ĝ(s)
Note: The order of the Ĝ and Ĥ!
M. Peet Lecture 6: Control Systems 14 / 23
Block Diagrams
Parallel Connection
G
+
u y
+
H
Parallel Interconnection: Suppose we have two systems: G and H.
Definition 3.
The Parallel interconnection of two systems is
y1 = Gu y2 = Hu y = y1 + y2
or
y = H(u) + G(u)
M. Peet Lecture 6: Control Systems 15 / 23
Block Diagrams
Parallel Connection
y1(s)
G(s)
+ u(s) y(s)
u(s) y(s) H(s)+G(s)
+
H(s)
The Transfer function of a Parallel interconnection:
• Laplace transform:
ŷ(s) = ŷ1 (s) + ŷ2 (s) = Ĝ(s)û(s) + Ĥ(s)û(s) = Ĥ(s) + Ĝ(s) û(s)
• The Transfer Function, T̂ (s) for the parallel interconnection of G and H is
T̂ (s) = Ĥ(s) + Ĝ(s)
M. Peet Lecture 6: Control Systems 16 / 23
Block Diagrams
Lower Feedback Interconnection
-
K(s) G(s)
u(s) + y(s)
Feedback:
• Controller: z = K(u − y) Plant: y = Gz
In the Frequency Domain:
ẑ(s) = −K̂(s)ŷ(s) + K̂(s)û(s) ŷ(s) = Ĝ(s)ẑ(s)
so
ŷ(s) = Ĝ(s)ẑ(s) = −Ĝ(s)K̂(s)ŷ(s) + Ĝ(s)K̂(s)û(s)
Solving for ŷ(s),
Ĝ(s)K̂(s)
ŷ(s) = û(s)
1 + Ĝ(s)K̂(s)
M. Peet Lecture 6: Control Systems 17 / 23
Block Diagrams
Upper Feedback Interconnection (Regulators)
There is another Feedback
interconnection K(s)
• u is the input
• y is the output
ŷ(s) = Ĝ(s)ẑ(s) -
G(s)
ẑ(s) = u(s) − K̂(s)ŷ(s) u(s) + y(s)
Which yields
ŷ(s) = Ĝ(s) u(s) − K̂(s)ŷ(s) = Ĝ(s)û(s) − Ĝ(s)K̂(s)ŷ(s)
hence the Transfer Function is:
Ĝ(s)
ŷ(s) = û(s).
1 + Ĝ(s)K̂(s)
M. Peet Lecture 6: Control Systems 18 / 23
The Effect of Feedback: Impulse Response
Inverted Pendulum Model
Transfer Function 1
Ĝ(s) = M gl
Js2 − 2
Controller: Static Gain: K̂(s) = K
Input: Impulse: û(s) = 1.
Closed Loop: Lower Feedback
K
Ĝ(s)K̂(s) Js2 − M2gl K
ŷ(s) = û(s) = K
= M gl
1 + Ĝ(s)K̂(s) 1 + Js2 − M gl Js2 − 2 +K
2
First Case:
M gl
• If K > 2 ,then K − M2gl > 0, so 2.5
Impulse Response
2
K/J 1.5
ŷ(s) = 1
s + K/J − M2Jgl
2
Amplitude
0.5
r ! −0.5
K M gl −1
y(t) = q sin K/J − t −1.5
J K/J − M gl 2J −2
2J −2.5
0 2 4 6 8 10 12
Time (sec)
M. Peet Lecture 6: Control Systems 19 / 23
The Effect of Feedback: Impulse Response
Inverted Pendulum Model
6 Impulse Response
x 10
18
16
14
12
Amplitude
10
0
0 5 10 15 20 25
Time (sec)
Second Case:
• If K < M2gl , then K − M2gl < 0, so
K 1 1
ŷ(s) = q + q
J s − K/J − M gl
s+ K/J − M gl
2J 2J
K √K/J− M gl t √ M gl
y(t) =
e 2J + e− K/J− 2J t
J
Important: Value of K determines stability vs. instability
M. Peet Lecture 6: Control Systems 20 / 23
Block Diagrams
Reduction
Now lets look at how to reduce a more complicated interconnections
e(s) + + y(s)
K1 1/s 1/s
- u(s) -
Label
• The output from the inner loop z
• The input to the inner loop u
First Close the Inner Loop using the Lower Feedback Interconnection.
K1
s K1
ẑ(s) = K1
û(s) = û(s)
s +1 K1 + s
M. Peet Lecture 6: Control Systems 21 / 23
Block Diagrams
Reduction
We now have a reduced Block Diagram
e(s) + y(s)
K1/(K1+s) 1/s
- u(s)
Again, apply the Lower Feedback Interconnetion:
K1
s(K1 +s) K1
ŷ(s) = ê(s) = ê(s)
1 + s(KK11+s) s(K1 + s) + K1
K1
So the Transfer function is T̂ (s) =
s2 + K1 s + K1
M. Peet Lecture 6: Control Systems 22 / 23
Summary
What have we learned today?
Transfer Functions
• Transfer Function Representation of a System
• State-Space to Transfer Function
• Direct Calculation of Transfer Functions
Block Diagram Algebra
• Modeling in the Frequency Domain
• Reducing Block Diagrams
Next Lecture: Partial Fraction Expansion
M. Peet Lecture 6: Control Systems 23 / 23
