Post on 30-Mar-2015
transcript
ERT 210Process Control & dynamics
Anis Atikah binti Ahmad
CHAPTER 8Feedback Controllers
Ch
apte
r 8
Control objective: to keep the tank exit composition, x, at the desired value (set point) by adjusting the flow rate, w2, via the control valve
Figure 8.1: Schematic diagram for a stirred-tank blending system
Chap
ter 8
Chap
ter 8
Figure 8.2: Flow control system
Ch
apte
r 8
In feedback control, the objective is to reduce the error signal to zero where
(8-1)sp me t y t y t
and
error signal
set point
measured value of the controlled variable
(or equivalent signal from the sensor/transmitter)
sp
m
e t
y t
y t
Chap
ter 8
Basic Control ModesThree basic control modes:
Ch
apte
r 8
Chap
ter 8
requires very intensive energy
costly
highly endothermic reaction
Derivative control
Proportional control
Integral control
Proportional ControlC
hap
ter
8Ch
apte
r 8
For proportional control, the controller output is proportional to the error signal,
(8-2)cp t p K e t
where:
controller output
bias (steady-state) value
controller gain (usually dimensionless)c
p t
p
K
•At time t = 25 min, e(25) = 60–56 = 4•At time t = 40 min, e(40) = 60–62 = –2
Ch
apte
r 8
Chap
ter 8
Ch
apte
r 8
The key concepts behind proportional control are the following:
1. the controller gain (Kc) can be adjusted to make the controller output changes as sensitive as desired to deviations between set point and controlled variable;
2. the sign of Kc can be chosen to make the controller output increase (or decrease) as the error signal increases.
Some controllers have a proportional band setting instead of a controller gain. The proportional band PB (in %) is defined as
Large PB correspond to a small value of Kc and vice versa
%1001
cK
PB ≜
Ch
apte
r 8
In order to derive the transfer function for an ideal proportional controller (without saturation limits), define a deviation variable
as p t
Then Eq. 8-2 can be written as
(8-5)cp t K e t
The transfer function for proportional-only control:
(8-6)cP s
KE s
An inherent disadvantage of proportional-only control is that a steady-state error (or offset) occurs after a set-point change or a sustained disturbance.
Chap
ter 8
)48( )( )( -ptptp ≜
Ch
apte
r 8
For integral control action, the controller output depends on the integral of the error signal over time,
0
1* * (8-7)
τ
t
I
p t p e t dt where ,= integral time/reset timeτI
• Integral action eliminates steady-state error (i.e., offset) Why??? e 0 p is changing with time until e = 0, where p reaches steady state.
Integral Control
The corresponding transfer function for the PI controller in Eq. 8-8 is given by
τ 111 (8-9)
τ τI
c cI I
P s sK K
E s s s
Integral control action is normally used in conjunction with proportional control as the proportional-integral (PI) controller:
0
1* * (8-8)
τ
tc
I
p t p K e t e t dt
Proportional-Integral (PI) Control
The integral mode causes the controller output to change as long as e(t*) ≠ 0 in Eq. 8-8
•Can produce p(t) that causes the final control element (FCE) to saturate.
•That is, the controller drives the FCE (e.g. valve, pump, compressor) to its physical limit of fully open/on/maximum or fully closed/off/minimum.
If an error is large enough and/or persists long enough
•the integral term continue growing, •the controller command the FCE to move to 110%, then 120% and
more•this command has no physical meaning (no impact on the process)
If this extreme value is still not sufficient to eliminate the error
Ch
apte
r 8
*Antireset windup reduce the windup by temporarily halting the integral control action whenever the controller output saturates and resumes when the output is no longer saturates.
Disadvantage of Integral Action: Reset Windup
Can grow very largeCan grow very large
0
1* * (8-8)
τ
tc
I
p t p K e t e t dt
ysp Ch
apte
r 8
• Integral action eliminates steady-state error (i.e., offset) Why??? e 0 p is changing with time until e = 0, where p reaches steady state.
s
11K
E(s)
(s)P
Ic
• Transfer function for PI control
Chap
ter 8
The function of derivative control action is to anticipate the future behavior of the error signal by considering its rate of change.
Controller output is proportional to the rate of change of the error signal or the controlled variable.
Thus, for ideal derivative action,
τ (8-10)D
de tp t p
dt
τDwhere , the derivative time, has units of time.
Ch
apte
r 8
Chap
ter 8
Derivative Control
Ch
apte
r 8
Chap
ter 8
Advantages of derivative action
Ch
apte
r 8
For example, an “ideal” PD controller has the transfer function:
1 τ (8-11)c D
P sK s
E s
• Derivative action always used in conjunction with proportional or proportional-integral control.
• Unfortunately, the ideal proportional-derivative control algorithm in Eq. 8-11 is physically unrealizable because it cannot be implemented exactly.
Chap
ter 8
Physically unrealizable
Physically unrealizable
Ch
apte
r 8
• For “real” PD controller, the transfer function in (8-11) can be approximated by
τ1 (8-12)
ατ 1D
cD
P s sK
E s s
where the constant α typically has a value between 0.05 and 0.2, with 0.1 being a common choice.
• In Eq. 8-12 the derivative term includes a derivative mode filter (also called a derivative filter) that reduces the sensitivity of the control calculations to high-frequency noise in the measurement.
Chap
ter 8
Ch
apte
r 8
Proportional-Integral-Derivative (PID) ControlCh
apte
r 8
requires very intensive energy
costly
highly endothermic reaction
Expanded form
Parallel form
Series form
3 common PID control forms are:
Parallel Form of PID Control
The parallel form of the PID control algorithm (without a derivative filter) @ “Ideal” PID control is given by
0
1* * τ (8-13)
τ
tc D
I
de tp t p K e t e t dt
dt
The corresponding transfer function for “Ideal” PID control is:
11 τ (8-14)
τc DI
P sK s
E s s
Chap
ter 8
Ch
apte
r 8
The corresponding transfer function for “Real” PID control (parallel), with derivative filter is:
1
11
)(
)(
1 S
S
S D
DcK
sE
sP
Figure 8.8 Block diagram of the parallel form of PID control (without a derivative filter)
Chap
ter 8
Series Form of PID ControlConstructed by having PI element and a PD element operated in series.Commercial versions of the series-form controller have a derivative filter as indicated in Eq 8-15.
τ 1 τ 1(8-15)
τ ατ 1I D
cI D
P s s sK
E s s s
Ch
apte
r 8
Figure 8.9 Block diagram of the series form of PID control (without a derivative filter)
Chap
ter 8
PIPI PDPD
sI1
1
Ch
apte
r 8
Expanded Form of PID Control
In addition to the well-known series and parallel forms, the expanded form of PID control in Eq. 8-16 is sometimes used:
0
* * (8-16)t
c I Dde t
p t p K e t K e t dt Kdt
Chap
ter 8
Used in MATLABUsed in MATLAB
PID - Most complicated to tune (Kc, I, D) .- Better performance than PI- No offset- Derivative action may be affected by noise
PI - More complicated to tune (Kc, I) .- Better performance than P- No offset- Most popular FB controller
P - Simplest controller to tune (Kc).- Offset with sustained disturbance or setpoint change.
Controller Comparison
Ch
apte
r 8
Chap
ter 8
Ch
apte
r 8
• This sudden change is undesirable and can be avoided by basing the derivative action on the measurement, ym, rather than on the error signal, e.
• Replacing de/dt by –dym/dt gives
0
1* * τ (8-17)
τ
t mc D
I
dy tp t p K e t e t dt
dt
• One disadvantage of the previous PID controllers :
Derivative and Proportional KickFeatures of PID Controllers
Chap
ter 8
derivative kick : when there is a sudden change in set point (and hence the error, e) that will cause the derivative term to become very large.
From a parallel form of PID control in Eq. 8-13From a parallel form of PID control in Eq. 8-13
Ch
apte
r 8
(8-22)c sp mp t p K y t y t
• The controller gain can be made either negative or positive.
Reverse or Direct ActionCh
apte
r 8
Direct (Kc < 0) Reverse (Kc > 0)
Controller output p(t) increases as the input signal
ym(t) increases
Controller output p(t) increases as its input signal
ym(t) decreases
Ch
apte
r 8
Chap
ter 8
Reverse acting (Kc > 0)e(t)↑, p(t) ↑ym(t)↓, p(t) ↑
Reverse acting (Kc > 0)e(t)↑, p(t) ↑ym(t)↓, p(t) ↑
(8-22)c sp mp t p K y t y t
Direct acting (Kc < 0)e(t) ↓, p(t) ↑ym(t)↑, p(t) ↑
Direct acting (Kc < 0)e(t) ↓, p(t) ↑ym(t)↑, p(t) ↑
On-Off Controllers
Synonyms:“two-position” or “bang-bang” controllers.
Controller output has two possible values.
Ch
apte
r 8
Chap
ter 8
ym(t)↓ym(t)↓
ym(t)↑ym(t)↑
eg: thermostat in home heating system.-if the temperature is too high, the thermostat turns the heater OFF. -If the temperature is too low, the thermostat turns the heater ON.
eg: thermostat in home heating system.-if the temperature is too high, the thermostat turns the heater OFF. -If the temperature is too low, the thermostat turns the heater ON.
On-Off Controllers (continued)C
hap
ter
8Ch
apte
r 8
Typical Response of Feedback Control SystemsConsider response of a controlled system after a sustained disturbance occurs (e.g., step change in the disturbance variable)
Ch
apte
r 8
Figure 8.12. Typical process responses with feedback control.
No control: the process slowly reaches a new steady state
P – speed up the process response & reduces the offset
PI – eliminate offset & the response more oscillatory
PID – reduces degree of oscillation and the response time
Chap
ter 8
Figure 8.13. Proportional control: effect of controller gain.Ch
apte
r 8
-Increasing Kc tends to make the process response less sluggish (faster)
-Too large of Kc, results in undesirable degree of oscillation or even become unstable
-Intermediate value of Kc usually results in the best control.
Chap
ter 8
Ch
apte
r 8
-Increasing τI tends to make the process response more sluggish (slower)
-Too large of τI, the controlled variable will return to the set point very slowly after a disturbance change @ set-point change occurs.
Figure 8.14. PI control: (a) effect of reset time (b) effect of controller gain.Chap
ter 8
Ch
apte
r 8
Figure 8.15. PID control: effect of derivative time.
-Increasing τD tends to improve the process response by reducing the maximum deviation, response time and degree of oscillation.
-Too large of τD: measurement noise is amplified and process response more oscillatory.
-The intermediate value of τD is desirable.
Chap
ter 8