Unit-V Controllers
Un-controlled vs Controlled System
Blue response resembles an un-controlled system with un-desired
response
This response is oscillatory as well as it takes much longer to
settle down
Why: Inertia effect, friction, backlash etc
The red response is of a controlled system with desired response
This response contains no oscillations and it settles to equilibrium /
steady state in lesser time
Controllability
Before a controller is implemented it is necessary to determine
the system is controllable
Test the “Controllability” of the system
Controllability is the ability of the system to be controlled
provided an external disturbance is available.
Proportional Integral Derivative Control
+ ep P
Input ∑ PID Plant Output
Block Diagram of PID Controller
u=control input / control effort & e=error
e=SP-PV=Input-Output
where, SP=Set Point & PV=Process Variable/control variable
Proportional Integral Derivative Control
PID is a combination of three control modes:
Proportional
Integral
Derivative
Combining modes depends on performance desired
Options include: P, PI, PD, PID
It is a control algorithm that operated on closed loop / feedback
approach
Why is PID used
Despite significant development in control theory, PID is popular in
industry because of following reasons:
Simple structure
Easy to design and implement
Robust performance over wide range of variation in process /
measurement uncertainties
Proportional Control
+ ep% Proportiona P%
Input ∑ Plant Output
l (Kp)
_
Proportional Control
P u P t K P eP PO
In Proportional Control, the control signal, u, is directly
proportional to the error, e.
As the gain is increased the system responds faster to changes in
set-point but becomes progressively under damped and eventually
unstable.
Proportional Control Action
PB
P Control Signal
PB= (100/ Kp) %
Kp = Kc
Advantages of Proportional Control
Simple and easy to design and tune
Rapid Response
Reduces Rise Time
Reduces Steady State Error
Dis-advantages of Proportional Control
Disadvantages:
Not possible to eliminate Steady State Error / Offset
Could lead to instability
Rise in overshoot/ oscillations
Application of Proportional Control
Applications:
Float Valve, Thermostat etc
Integral Control
+ ep% P%
Input ∑ Integral (Ki) Plant Output
_
Integral Control
Rate of change of integral control signal is proportional to
error.
Control signal proportional to integral of error.
When the error is zero, the control signal is a constant value.
When the error is constant, the control signal varies at constant
rate.
Integral Control Action
I Control Signal
Advantages of Integral Control
Advantages:
Eliminates steady state error/offset
Decreases Rise Time
Dis-advantages of Integral Control
Disadvantages:
Leads to minor increase in overshoot
Could make the system less stable
Increases Settling time
Derivative Control
+ ep% Derivative P%
Input ∑ Plant Output
(KD)
_
Derivative Control
P uD t K D de dt P 0
Derivative control produces a control signal proportional to the
rate at which the error is changing. Also known as rate
controller.
While sudden/rapid change in error leads to a control signal of
larger magnitude, gradual change leads to small magnitude.
The derivative control will generate no signal if the error is
constant
Thus, not used alone; used with P control
Derivative Control Action
D Control Signal
Advantages of Derivative Control
Advantages:
Reduces Settling time
more fast
Reduces Overshoot
Adds more stability if use with proportional controller
Dis-advantages of Derivative Control
Disadvantages:
Not possible to eliminate Steady State Error / Offset
Not possible to use alone
For small error, the speed of system response is also slow
Amplifies Noise
Parallel Form: PI Control
H s K P K I
s
Where,
K P Proportional Gain, K I Integral Gain
u t u P t u I t K P e K I edt
Proportional Integral (PI) Control helps minimise rise time,
settling time as well as eliminate steady state error.
PI: Parallel / Non-Interacting Form
ud
Kds
Ki ui
s
+
ysp + e +
y
Kp up
+
u
plant
-
Derivative Action does not Interact with Integral Action
Parallel Form: PD Control
H s K P K D s
Where,
K P Proportional Gain, K D Derivative Gain
u t u P t u D t K P e K D de
dt
Proportional Derivative (PD) Control helps reduce rise time,
settling time as well as minimize overshoot.
PD: Parallel / Non-Interacting Form
ud
Kds
Ki ui
s
+
ysp + e +
y
Kp up
+
u
plant
-
Transfer Function of PID control
Transer Function :
H s K P I K D s
K
s
Where,
K P Proportional Gain, K I Integral Gain
K D Derivative Gain
Control Signal :
u t u P t u I t u D t
K P e K I edt K D de dt
Where,
e Error Differencebetween reference& measured signal
PID: Parallel / Non-Interacting Form
ud
Kds
Ki ui
s
+
ysp + e +
y
Kp up
+
u
plant
-
Time Domain Specifications
Unit Step Response of Second Order System
Effect of P, I & D on Transient Specifications
Action Rise Overshoot Settling SS
Time Time Error
KP Decrease Increase Small Decrease
Change
KI Decrease Increase Initially Eliminate
Decrease then
Increase
KD Small Decrease Decrease Small
Change Change
PID: Stepwise Procedure for Manual Tuning
1. Obtain an open-loop response and determine what needs to be
improved
2. Add a proportional control to improve the rise time
3. Add a derivative control to improve the overshoot
4. Add an integral control to eliminate the steady-state error
PID: Stepwise Procedure for Manual Tuning
5. Adjust each of P, I & D until you obtain a desired overall
response referring to the table shown previously to find out
which controller controls what characteristics.
6. It is not necessary to implement all three controllers (P, I & D)
into a single system. For example, if a PI controller gives a good
enough response, then you don't need to add D control to the
system. Simple is better.
Applications of PID Control
1. Regulation of Processes in Industry
1. Flow
2. Temperature
3. Pressure etc
2. Servo / DC motor Control
3. Linear Position Control
Numerical
Figure shows an error time graph. Sketch the PID controller
output w.r.t time. Assume kp = 10, ki = 2, kd = 0.5 and P0=0 i.e the
controller output is zero when the error is zero.
Solution
Substituting values for t, following table can be generated for
values of u
t ≥0 ≤1 ≥1 ≤3 ≥3
u 0.5 11.5 11 15 5