First and Second Order Dynamic Systems
First and Second Order Dynamic Systems
FIRST ORDER
DYNAMIC SYSTEMS
MAPUA INSTITUTE OF TECHNOLOGY
School of Chemical Engineering & Chemistry
4 BASIC COMPONENTS IN A
FEEDBACK CONTROL LOOP
Operator
3. Controller
Load
1. Process
Refining Process (Plant)
Variable
OUTLINE
• General Equation
• Transfer Function
• Physical Examples
• Step Response
• Sample Problems
GENERAL EQUATION
u (dy/dt) + y = K(u) + y0 y
where:
• u = input variable = time constant
• y = output variable • K = steady state gain DTF 1
• t = time • y0 = y intercept
TF
Steady State
us (dy/dt)
0 + ys = K(us) + y0 ys
us ys = K(us) + y0 ys
y
K
DTF 1
y0
u
TRANSFER FUNCTION
Transfer Function = ratio of the Laplace Transform of the output
variable to that of the input variable
(5-3)
Y(s) K
U(s) = Y(s)
U(s) (s + 1)
DTF 1
where: DTF 2
• U(s) = Laplace Transform of the DEVIATION VARIABLE U DTF 3
• Y(s) = Laplace Transform of the DEVIATION VARIABLE Y
DTF 4
Deviation Variable
• Difference between the actual value of a
certain variable at any point and its steady
state value
• U = u - us
DTF 1
• Y = y - ys DTF 2
DTF 3
TF DTF 4
Deriving the Transfer Function (1/4)
• Subtract the Steady State Equation from the
Unsteady State Equation:
[ (dy/dt) + y = K(u) + y ]
0
Unsteady State
- [ y = K(u ) + y ]
s s 0 Steady State
DTF 2
= (dy/dt) + (y - ys) = K(u - us) + 0 DTF 3
TF DTF 4
Deriving the Transfer Function (2/4)
• Introduce Deviation Variables U & Y where:
• U = u - us
• Y = y - ys
• dY/dt = dy/dt
Since: dY/dt = d(y - ys)/dt = [dy/dt - dys/dt]
DTF 1
but at Steady Steady: dys/dt = 0
DTF 3
TF DTF 4
Deriving the Transfer Function (3/4)
• After introducing the Deviation Variables U & Y:
{
L (dY/dt) + Y = K(U) }
{sY(s) + Y(0)} + Y(s) = K{U(s)}
but at Steady Steady: y = ys therefore Y(0) = 0
DTF 1
sY(s) + Y(s) = K{U(s)} DTF 2
TF DTF 4
Deriving the Transfer Function (4/4)
• Factoring out Y(s):
Y(s){s + 1} = K{U(s)}
• Rearranging:
Transfer Function
Y(s)/U(s) = K/{s + 1} for First Order
Systems
DTF 1
DTF 2
DTF 3
Step Change in U
u U(t) = M (5-4)
u = us + M
us
0 time, t
Y(t) = ?
Example 5.1
Step Response
(5-18)
y
Y(t) = MK(1 - e ) -t/
y = ys + MK
Y() = MK
Y = 63.2% of MK
ys
0 time, t
Example 5.1
Y()
• Final Value of Y
– value of Y as time t approaches infinity
– ultimate value of Y
Sample Problem 1
• Example 5.1 (Coughanowr, p.57)
A thermometer having a time constant of 0.1min is at
a steady state temperature of 90oF. At time t = 0, the
thermometer is placed in a temperature bath
maintained at 100oF. Determine the time needed for
the thermometer to read 98oF.
Sample Problem 2
A well-stirred, continuous mixer has a single input
and a single output stream with a variable solute
concentration but fixed flow rate. The volume of the
mixer is 1000L and the flow rate through it is
500L/min. Initially, the solute concentration is 0.1g/L
and it has been constant at this value for a long time.
If the inlet concentration were to suddenly increase
to 0.15 g/L, and to remain at this new value, how
long will it take for the exit concentration to reach
0.125 g/L? Determine the exit concentration 5
minutes after the inlet concentration changed.
Process Control & Instrumentation
SECOND ORDER
DYNAMIC SYSTEMS
MAPUA INSTITUTE OF TECHNOLOGY
School of Chemical Engineering & Chemistry
OUTLINE
• General Equation
• Transfer Function
• Step Response
• Underdamped 2nd Order Systems
• Sample Problems
GENERAL EQUATION
u y
2(d2y/dt2) + 2(dy/dt) + y = K(u) + y0
where:
• u = input variable = damping factor
• y = output variable • K = steady state gain DTF 1
us ys
2(d2y/dt
0 2) ++2(dy/dt)
0 ++yyss==K(u
K(uss))++yy00
us ys = K(us) + y0 ys
y
K
DTF 1
y0
u
TRANSFER FUNCTION
Transfer Function = ratio of the Laplace Transform of the output
variable to that of the input variable
• U = u - us
DTF 1
• Y = y - ys DTF 2
DTF 3
TF DTF 4
Deriving the Transfer Function (1/4)
• Subtract the Steady State Equation from the
Unsteady State Equation:
- [ y = K(u ) + y
s s 0 ] Steady State
DTF 2
= (d y/dt ) + 2(dy/dt) + (y - ys) = K(u - us)
2 2 2
DTF 3
TF DTF 4
Deriving the Transfer Function (2/4)
• Introduce Deviation Variables U & Y where:
• U = u - us
• Y = y - ys
• d2Y/dt2 = d2y/dt2 & dY/dt = dy/dt
Since: dY/dt = d(y - ys)/dt = [dy/dt - dys/dt]
DTF 1
but at Steady Steady: dys/dt = 0
DTF 3
TF DTF 4
Deriving the Transfer Function (3/4)
• After introducing the Deviation Variables U & Y:
L { }
2(d2Y/dt 2) + 2(dY/dt) + Y = K(U)
DTF 2
TF DTF 4
Deriving the Transfer Function (4/4)
• Factoring out Y(s):
Y(s){2s2 + 2s + 1} = K{U(s)}
• Rearranging:
Transfer Function
Y(s)/U(s) = K/ {2s2 + 2s + 1} for Second Order
Systems
DTF 1
DTF 2
DTF 3
Step Change in U
u
U(t) = M (5-4)
u = us + M
us
0 time, t
Y(t) = ?
Problem 8.1
Underdamped Step Response ( < 1)
[ { }]
1 (1 - 2)
Y(t) = MK 1- e-t/ sin (1 - 2)(t/) + tan-1
(1 - )
2
y (5-51)
Y() = MK
ys
0 time, t
Critically Damped Step Response
( = 1)
Y(t) = MK
[ {
1 - 1 +
t
} ]
e-t/ (5-50)
Y() = MK
ys
0 time, t
Overdamped Step Response
( > 1)
[ { } { } ]
1 2
Y(t) = MK 1 - e-t/1 - e-t/2 (5-48)
(1 - 2) (2 - 1)
y
=1
Y() = MK
ys
0 time, t
Overdamped Step Response
1 = { + (2 - 1) }
2 = { - (2 - 1) }
Underdamped Step Response ( < 1)
y
Response Time
P
B
C
Y() = MK
ys
0 tr time, t
Terms Used to Describe an
Underdamped 2nd Order Step Response
• Overshoot
• Decay Ratio
• Rise Time
• Response Time
• Period of Oscillation
Overshoot
• A measure of how much the response exceeds
the ultimate value following a step change.
Overshoot = B/Y()