Control System Design

PID TUNING

Diwakar Bista
 Control System Design – PID Tuning
• Introduction - PID
• A proportional–integral–derivative controller (PID controller or three-term controller) is a control loop mechanism
employing feedback that is widely used in industrial control systems and a variety of other applications requiring
continuously modulated control.

• The output of a PID controller, which is equal to the control input to the plant, is calculated in the time domain from
the feedback error as follows

t
de(t )
u (t ) =K p e(t ) + K d + K i ∫ e(τ )dτ
dt 0

Jan 2020 PID TUNING 2

 Control System Design – PID Tuning
• Introduction - PID
• The main objective of designing PID controller is to obtain,

• Fast rise time

• Minimal overshoot
• Zero steady-state error
• Open-Loop Step Response

• Designing of PID controller involves determining the values of proportional, derivative and integral gain / time
constants.
Type Rise time Overshoot Settling Time Steady state error
Kp Decreases Increases Small change Decreases
Ki Decreases Increases Increases Eliminates
Kd Small change Decreases Decreases Small change

Jan 2020 PID TUNING 3

 Control System Design – PID Tuning
• PID constant and time constant

Control action of PID

t
de(t )
u (t ) =K p e(t ) + K d + K i ∫ e(τ )dτ
dt 0

Taking Laplace transform on both sides of the equitation

1
U ( S ) =K p E ( S ) + SK d E ( S ) + K i E ( S )
S
 K d 1 Ki 
U ( S ) = K p 1 + S +  E (S )
 K p S K p 

Where,
 1  Kp
U ( S ) = K p 1 + STd +  E (S )
 STi  K d= Td × K p and K=
i ........................(i )
Ti
These are derivative time constant and integral
time constant respectively

Jan 2020 PID TUNING 4

 Control System Design – PID Tuning
• Tuning of PID controller
• Adjustment of controller parameters to obtain a specified close
loop response
• Value of tuning parameters depend on the desired closed loop
response and on the dynamic characteristics of the process to be
controlled.
• A set of tuning parameters can produce the desired response at
only one operating point.
• Tuning of controller is a repeated tedious process of continuous
trial and error.
• There are various methods for tuning PID controller.
Method Advantages Disadvantages
Manual tuning No math required Requires experienced personnel.
Ziegler–Nichols Proven method Process upset, some trial-and-error, very aggressive tuning.
Tyreus Luyben Proven method Process upset, some trial-and-error, very aggressive tuning
Consistent tuning; online or offline - can employ computer-automated control
Software tools system design (CAutoD) techniques; may include valve and sensor analysis; allows Some cost or training involved.
simulation before downloading; can support non-steady-state (NSS) tuning.
Cohen–Coon Good process models. Some math; offline; only good for first-order processes.
Can be used for auto tuning; amplitude is minimum, so this method has lowest
Åström-Hägglund The process itself is inherently oscillatory.
process upset

Jan 2020 PID TUNING 5

 Control System Design – PID Tuning
• Ziegler–Nichols PID tuning
•The Ziegler and Nichols method provide a useful empirical method for tuning of PID controllers.
•These tuning rules have been derived to provide step response for the closed loop system.

•Two method
• Continuous Cyclic Method
• Process Reaction Method

Kp
Ziegler–Nichols Ti
p Ki
Method =Td
Kd
and
= Ti
K
Kp Ki
Kd
Td

Jan 2020 PID TUNING 6

 Control System Design – PID Tuning
• Ziegler–Nichols PID tuning - Continuous Cyclic Method
• It is known as method of stability margin
• Proportional controller is used for tuning while D and I are kept zero
• Proportional gain is increase until system becomes marginally stable (neutral stable) and the value of
proportional gain for natural stability is taken for determining values of derivative and integral time
constant.
• This value of proportional gain is call called critical gain Kc .
•Time period of oscillation of output at critical gain is termed as critical time Tc
Controlling Parameters
Controller Type
Kp Ti Td
P Kc/2 ∞ 0
PI Kc/2.2 Tc/1.2 0
PID Kc/1.7 Tc/2 Tc/8

Jan 2020 PID TUNING 7

 Control System Design – PID Tuning
• Ziegler–Nichols PID tuning - Continuous Cyclic Method
• Steps for continuous cyclic method
• The controller is switched to pure proportional action
• Proportional gain is continuously increased until the closed
loop system shows permanent oscillation
• This value of proportional gin is denoted as the critical gain of
the controller Kc
• The length of period of the permanent oscillation is measured
and is denoted as critical period Tc
Controlling Parameters
• From these values of critical gain and critical period the Controller Type
controller tuning values of proportional gain, integral time Kp Ti Td

constant and derivative time constant are determined using P Kc/2 ∞ 0

the empirical relations as in table. PI Kc/2.2 Tc/1.2 0

• With reference to equation (i) the value of integral constant PID Kc/1.7 Tc/2 Tc/8

and derivative constant is determined.

Kp
K d= Td × K p and K=
i ........................(i )
Ti

Jan 2020 PID TUNING 8

 Control System Design – PID Tuning
• Ziegler–Nichols PID tuning - Example
• System
G (s) = 1
s3 +3s2 +2s
System: Go Step Response
2 Time (seconds): 24.7
Amplitude: 1.9

System: Go
Time (seconds): 20.3
1.8 Amplitude: 1.9

1.6

1.4

1.2
Amplitude

0.8

0.6

0.4

0.2

0
0 20 40 60 80 100 120 140
Time (seconds)

Jan 2020 PID TUNING 9

 Control System Design – PID Tuning
• Ziegler–Nichols PID tuning – Process Reaction Method
• It is also known as method of step response
• It assumes that the open loop step response of the process has and S-Shape, called as the process reaction
curve.
• The process reaction curve may be approximated to a time delay Td and rise time Tr.
• The empirical formulas are based directly on the rise time of the tangent at the turning point and on the
delay time of the step response.
• It is to be noted that the measurement of the step response needs to be taken only at the tuning point of
the process reaction curve.
Controlling Parameters
Controller Type
Kp Ti Td
• Not to be confused with
P Tr/Td ∞ 0
Time delay and Derivative
Time Constant PI 0.9 Tr/Td Td/0.3 0

PID 1.2 Tr/Td 2Td 0.5Td

Jan 2020 PID TUNING 10