Académique Documents
Professionnel Documents
Culture Documents
Automatic Control
Presentation Part 1
Dr. Clarence W. de Silva, P.Eng.
Professor of Mechanical Engineering
The University of British Columbia
e-mail: desilva@mech.ubc.ca
http://www.researchcentre.apsc.ubc.ca/MECH466/
C.W.
de Silva
http://www.researchcentre.apsc.ubc.ca/MECH466/
Each lecture will be repeated (given twice) for the benefit of those
with time conflicts.
Tutorial Sessions
A set of tutorials have been organized to assist the students in
their homework assignments. These tutorials will commence one
week after announcing the first assignment. Please see the
following web site for the dates, times, and the locations of these
tutorials:
http://www.researchcentre.apsc.ubc.ca/MECH466/
Each tutorial session will be repeated (given twice) for the benefit
of those with time conflicts.
Week Starts
Topics
Sections 12.1
Sep 18 Analogies;
7
8
Section 12.10
10%
20%
70%_
100%
Student Help: Office Hours: M, W, 3:00-4:00 p.m. Also, you may send me e-mail with your questions. Other face-to-face
meetings can be scheduled if needed.
CONTROL ENGINEERING
Objective of Control:
Make a dynamic system behave in a desired manner, according
some performance specifications.
Note:
Complex system (many inputs and many outputs, dynamic
coupling, nonlinear, etc.)
Unknown excitations
Unknown dynamics
Body
Response
Aerodynamic
Aerodynamic
Engine
Excitations
Control Surface
Excitations
TERMINOLOGY
Plant or Process: System to be controlled
Inputs: Excitations (known, unknown) to the system
Outputs: Responses of the system
Sensors: They measure system variables (excitations, responses, etc.)
Actuators: They drive various parts of the system.
Controller: Device that generates control signal
Control Law: Relation or scheme according to which the control signal is
generated
System
Typical Input
Typical Outputs
Human body
Neuroelectric
Muscle contraction,
pulses
body movements
Information
Decisions,
Company
finished products
Power plant
Fuel rate
Electric power,
pollution rate
Automobile
Robot
Steering wheel
movement
direction of heading
Voltage
motor
to
Controller
Actuator
Valve
Process
(Plant)
Controller
Actuator
Valve
Feedback
Signal
Transducer
or
Transmitter
Process
(Plant)
Controlled
Variable
(Output)
Controlled
Variable
(Output)
Temperature,
Pressure, etc.
Transmitter
Signal
Reference
Input Signal
Signal
Conditioning
Signal
Conditioning
Analog
Control
Hardware
Actuator
Analog SensorsTransducers
Plant
Output
(Response)
Reference Input
ADC
Digital
Control
Processor
Analog
Multiplexing
Digital
Multiplexer
DAC
Drive
System
Signal
Conditioning
Plant
Analog
Sensors/
Transducers
Digital
Sensors/
Transducers
Output
Measurement
for
Feedforward
Reference
Input
Controller
Plant
Measurement
for
Feedback
Output
Description
Sampled data
Digital data
Note: Sampling errors (aliasing) of sampled data and quantization errors (finite word length) of digital data.
Differential equations
Deterministic Systems
Stochastic (Random) Systems
Transfer-Function Domain
Laplace transfer functions or Fourier frequency
response functions
CONTROL TECHNIQUES
Broad Division:
1. Time domain techniques (Differential equations wrt time t; can be nonlinear)
2. Frequency domain techniques (Transfer functions
algebraic wrt frequency ; typically linear)
Servo Control: To track a specified trajectory
(Commonly uses proportional-integral-derivative or PID control; typically linear)
Compensators: Hardware/software modules that help the controller to achieve the required
system performance (Lead, Lag, Lead-Lag)
Modern Control Techniques:
(Use state-space representation; not so modern)
1. Linear quadratic regulator (LQR): Minimize a cost function (maximize a
performance index Optimal control).
2. Pole placement: Locate system poles (eigenvalues) to modify the modes (i.e.,
fundamental free natural responses) with respect to stability, speed of response, etc.
modal control.
INTELLIGENT CONTROL
(Knowledge-based Control)
Knowledge base (non-analytic); e.g., a set of rules, is
available regarding system behavior (through operating
experience, heuristics, control expertise, etc.).
Suitable for large-scale and complex systems (analytical
modeling difficult; important input-output signals not
available for measurement)
Popular Approaches:
Fuzzy Logic Control
Neural-Network Control
Fuzzy-Neural Control
ANALYTICAL MODELS
Dynamic System: Response variables are functions of time,
with non-negligible rates of changes.
Model: A representation of a system.
Types of Models:
1. Physical Models (Prototypes)
2. Analytical Models
3. Computer (Numerical) Models
4. Experimental Models (using input/output experimental data)
Models for Physical Dynamic Systems:
lumped-parameter models
continuous-parameter models.
Example: Spring element (flexibility, inertia, damping)
dv
= f
dt
Cf
b = damping constant
(Hookes Law)
k = stiffness
Inductor
Resistor
dP
=Q
dt
Cf = fluid
capacitance
Ri = v
di
L =v
dt
R = resistance
L = inductance
None
Thermal Resistor
dT
=Q
dt
Ct = thermal
capacitance
Fluid Capacitor
f = bv
df
= kv
dt
Rt Q = T
Rt = thermal resistance
Fluid Inertor
dQ
If
=P
dt
If = inertance
Fluid Resistor
Rf Q = P
Rf = fluid resistance
Force-Current Analogy
System Type
System-response
Variables:
Through Variables
Across Variables
System
Parameters
Mechanical
Electrical
Force f
Velocity v
M
Current i
Voltage v
C
K
B
1/L
1/R
3. A finite force cannot cause an infinite acceleration. A finite instantaneous change (step)
in velocity will require an infinite force. Hence, for an inertia element, v is a natural
v
C
i
-
v
A M a s s E l e m e n t a n d a n A n a lo g o u s C a p a c i to r E le m e n t.
v
b
L
v
v
A S p r in g E le m e n t a n d a n A n a l o g o u s I n d u c to r E le m e n t.
R
v
f
b
f = bv
v
i=
V i s c o u s D a m p e r a n d A n a lo g o u s E le c tri c a l R e s i s to r.
1
v
R
System Order:
represent/solve the dynamic system; typically, number of independent energy storage elements
Example (Electrical)
The circuit shown in the figure consists of an inductor L, a capacitor C, and two resistors R and
Ro. The input is the voltage vi ( t ) and the output is the voltage vo across the resistor Ro.
vR
R
vL
iL
Node
A
io
iC
vC
+
vi(t)
+
vo
Ro
C
Loop 1
Loop 2
-
Example (Contd)
(c) Obtain expressions for undamped natural frequency and the damping
ratio of the system.
(d) The system starts at steady state with an input of 5 V (for all t < 0). Then
suddenly, the input is dropped to 1 V (for all t > 0), which corresponds to
a step input as shown below.
Input
vi(t)
5.0
1.0
0
Time t
Solution
(a)
State Variables:
Current through independent inductors ( iL)
Voltage across independent capacitors (vc )
Constitutive Equations:
vL = L
iC = C
diL
dt
dvC
dt
vR = RiL
vo = Rio
First two equations are for independent energy storage elements State-space shell.
Continuity Equation:
Node A (Kirchhoffs Current Law):
iL iC io = 0
Compatibility Equations:
Loop 1 (Kirchhoffs Voltage Law):
vi vR vL vC = 0
vC vo = 0
diL
L
= vL = vi vR vC = vi RiL vC
dt
dv
C C = iC = iL io
dt
v
v
= iL o = iL C
Ro
Ro
==> State equations:
diL 1
= [RiL vC + vi ]
dt L
(i)
dvC 1
v
= [iL C ]
dt C
Ro
(ii)
Output Equation:
vo = vc
Vector-Matrix Representation:
x = Ax + Bu
y=C x
where:
1/ L
R / L
System matrix A = 1/ C 1/( R C )
o
1/ L
B=
Input = u = [vi ]
Output = y = [vo ]
(b)
From (ii):
dvC vC
iL = C
+
dt
Ro
Substitute in (i) for iL:
L
dv
v
dv
v
d
(C C + C ) = R (C C + C ) vC + vi
dt
dt
Ro
dt
Ro
d 2 vo
dvo
L
R
LC
+ ( + RC )
+ ( + 1) vo = vi
2
dt
Ro
dt
Ro
(iii)
(c)
The input-output differential equation is of the form
d 2 vo
dvo
1
2
vi
+ 2 n
+ n vo =
2
dt
dt
LC
Hence,
Natural frequency
Damping ratio
1 R
( + 1)
LC Ro
n =
1
2 LC (
R
+ 1)
Ro
L
+ RC )
Ro
(iv)
(v)
1
Note: LC has units of (frequency)2.
RC and
L
Ro
(d)
Initial Conditions:
For t < 0 (initial steady state):
diL
=0
dt
dvc
=0
dt
diL (0 )
1
0
[
Ri
(0
)
v
(0
)
v
(0
)]
=
=
+
L
C
i
(i):
dt
L
dvC (0 )
vC (0 )
1
(ii): dt = 0 = C [iL (0 ) R ]
o
iL (0 ) vC (0 ) = 0
( )
( ) = 2.5 V
iL 0 = 2.5 A , vc 0
Note: State variables cannot undergo step changes (because that violates the
corresponding physical laws constitutive equations). Specifically,
Inductor cannot have a step change in current (needs infinite voltage).
Capacitor cannot have a step change in voltage (needs infinite current).
Hence,
( ) ( )
v ( 0 ) = v ( 0 ) = 2.5 V
iL 0+ = iL 0 = 2.5 A
+
+
v
(0
Note: Since i ) = 1.0
diL (0 + )
+
+
i
(0
)
v
(0
) + 1.0 = 2.5 2.5 + 1.0 = 4.0 A/s 0
=
L
C
(i): dt
dvC (0+ )
+
+
=
i
(0
)
v
(0
) = 2.5 2.5 = 0.0 V/s
L
C
(ii):
dt
diL
=0
dt
dvc
=0
dt
and vi = 1.0
Substitute:
diL ()
(i): dt = 0 = iL () vC () + 1.0
dvC ()
(ii): dt = 0 = iL () vC ()
( ) = 0.5 V
Solution: iL ( ) = 0.5 A , vc
For the given parameter values,
(iii):
d 2 vo
dvo
+
+ 2vo = 1
2
dt 2
dt
vo = vc
Response
iL
2.5
Slope = 4 A/s
0.5
0
vc = vo
2.5
Slope = 0 V/s
0.5
0
Massless Casing
Instrument
Fixture
k2
Shaker
Table
Input u(t)
An Equivalent Model
m
Compressive
force f1
k1
v1
P
Compressive
force f2
k2
u(t)
State Variables:
1. Force through independent springs.
2. Velocity across independent masses.
Constitutive Equations:
For mass:
f1 = mv
For spring 1:
f1 = k1 (v1 v )
For spring 2:
f2 = k 2 (u v1 )
For damper:
f 1 f 2 = b(u v1 )
v =
1
f1
m
f = k v k1 f + k1 f + k u
1
1
1
2
1
b
b
f = k 2 f k 2 f
2
1
2
b
b
Output Equation:
v=v
Vector-Matrix Form:
State vector
Input vector
v
f
1
x = f (3rd order)
2
u = [u (t )] (a scalar)
Output vector y =
[v ]
k1
A=
0
0
k1
b
k , B=
2
b
1
m
k1
b
k2
b
(a scalar)
0
k
1
0 ,
C=
[1
0 0]
Fluid Elements
System (State) Variables:
Pressure (across variable) of each independent fluid capacitor (A-type element)
Volume flow rate (through variable) of each independent fluid inertor (T-type
element)
Fluid Capacitor
dP
Cf
=Q
dt
Note: Fluid capacitor stores potential energy (a fluid spring)
Mechanical A-type element (inertia) stores kinetic energy.
V
C
=
bulk
For a liquid control volume V of bulk modulus :
V
C
=
For an adiabatic (zero heat transfer, fast-process) gas: comp kP
Ratio of specific heats at constant pressure and constant volume
k=
cp
cv
Celastic
A2
=
k
Cbulk + Celastic .
C grav =
A
g
Fluid Inertor
I
f
dQ
=P
dt
If =
x
A
I f =
x
A
Fluid Resistor
In the approximate, linear case: P = R f Q
n
In the general, nonlinear case: P = K R Q
For viscous flow through a uniform pipe with circular cross-section of diameter d and
length l:
R f = 128 l
d 4
For a rectangular cross-section of height b which is much smaller than its width w:
R f = 12l
wb3
cd = discharge coefficient
2 j j
F ( s ) exp( st )ds
L f (t ) = F ( s)
f (t ) exp ( st ) dt
k1 f1 (t ) + k2 f 2 (t )
k1 F1 ( s) + k2 F2 ( s)
exp(at ) f (t )
F (s + a)
f(t-)
exp(-s)F(s)
( n)
(t ) =
d n f (t )
s n F ( s) s n 1 f (0 + ) s n 2 f 1 (0 + )
dt n
" f
F ( s)
+
s
Impulse function (t )
1
s
tn
n!
s n +1
exp(at )
1
s+ a
sin nt
n
s + n2
2
os nt
(0 + )
f (t )dt
n 1
s
s + n2
2
f (t )dt
s