TABLE OF CONTENT

CONTENT

PAGE

Abstract

Introduction

Theory

Result

Discussion

12

Conclusion

14

References

15

Appendices

16

ABSTRACT
1

The experiment is about open loop test and closed loop test with load disturbance and set
point test was conducted. There are three types analysis which is Control Process (LIC 11), Flow
Control (FIC 21) and Liquid Flow Process (FIC 31). The result of curve obtain from using three
loop analysis. After that, the calculated values of Response Rate (RR), Dead Time (Td) and Time
constant (Tc) by using Tangent Method. Therefore, the optimum values which is P and I is then
calculated using Ziegler-Nichols. After that, the values of P and I obtain by using Ziegler-Nichols
was used for closed loop test with load disturbance and set point test. The values of Kc, I and PB
may be divided or times by 4 to stabilize the process. Finally, the result of closed loop test curves
obtain for each analysis as shown in result and appendices.

INTRODUCTION
2

Control is used in many aspect of engineering such as mechanical, electrical and

chemical. Process control is one of the most used chemical engineers that applied to chemical
reactors, heat exchanger and mass transfer equipment. The example was shows when the
operation of all equipment must be consider when designing because plant never operate in
steady state. The consideration must include the disturbance to make safe and profitable
chemical plant design. In process control is basically means regulate the set point value to
maintain the quantity of process substance. There are single variable processes in which only one
variable is to be controlled as well as multivariable processes may require regulation. The
example of the process variable are temperature, pressure and fluid flow rate and liquid level and
other properties of materials such as material stream composition, density, pH viscosity, and
speed.
The process control devices used in this experiment is Distributed Control System (DCS)
off Delta V Emmerson and Foxboro. There are two loops which is open loop and closed loop
test. The open loop test process can be used as single pulse and double pulse test. For open loop,
the output is moved directly meanwhile the closed loop, output is moved indirectly because of
changing in set point. The process movement larger if we able measure and factor the
disturbance and set point. That cause by disturbance.
The controller is set to automatic if the process changed into the closed loop system. The
controller set to manual if the process change into open loop system. If the process in automatic
mood, the process will stabilized to desired SP. The value of P and I are inserted by using
calculation values of Ziegler-Nichols. However, the poor combination of values P and I will
results unstable oscillation. Hence, change the values of I, PB and Kc by divide or times of 4 to
control the unstable oscillation to become more stable curves.

THEORY
3

When the controller in manual mode the process will change into open loop process. The
Manipulated Variable (MV) is increase to 5%. The response curves obtain to find the values of
RR, Td and Tc. All the values calculated using the tangent method by drawing the tangent line to
steepest point of the curve. Figure below shows the step input (MV) and the associated step
response curve with the tangent line.

Figure 1
RR is the response rate with the unit of 1/s. The value of t is the value of Time Constant. The
dead time period is from the initial/start of open loop test to the cross section between tangent
line and initial steady state baseline. The time constant is measured from the intersection of
initial steady state baseline and the tangent line to the intersection of final steady state baseline
and tangent line as shown in Figure 1 Then, the values of Tc and Td calculated as formula below:

Td (time) = Td (length) x b (time/length)

4

Tc (time) = Tc (length) x b (time/length)

The units of Td and Tc are in second. Then, the value of Td and RR obtain calculated by using
Ziegler-Nichols tuning rule to get the optimum controller settings.

For numerical technique, the diagram shows the tangent analysis of plotted step response data.

The data of numerical method was collected by using distributed control system (DCS),
supervisory control and data acquisition (SCADA) and paperless recorder. By using the 3-point
central difference formula, the slope at to is as formula below:

Where the PV is Process Value and to as Td. Then, the RR, Td and Tc are defined as:

RESULT
CALCULATION OF OPTIMUM P AND I OF SELF REGULATING PROCESS USING
TANGENT METHOD ( FIC21)
6

PVi = 20.2%

= 78o

PVf = 26.1%

a = 10%/17mm

MV = (40-35)% = 5%

RR =

b = 15s/12mm

tan a
MV b
tan 78 10 /17 mm
5 15 s /12 mm

Td = 1mm x (15s/12mm

= 0.4428/s

= 1.25s

Tc = 2mm x (15s/12mm) = 2.5s

I = 3.33(Td)
= 3.33(1.25) = 4.163s

PB = 111.11 x RR x Td
= 111.11 x 0.4428/s x 1.25s = 61.50 %

KC =

100
PB

100
61.50

= 1.626

CALCULATION OF OPTIMUM P AND I OF NON SELF REGULATING PROCESS USING

TANGENT METHOD (LIC11)
7

PVi = 50.5 %

= 30o

PVf = None. Level is non self-regulating process

a = 10%/17mm

MV = 10%

RR =

b = 15s/12mm

tan a
MV b
tan 30 10 /17 mm
10 15 s /12 mm

Td = 2mm x (15s/12mm

= 0.0247/s

= 2.5s

I = 3.33(Td)
= 3.33 (2.5) = 8.325 s

PB = 111.11 x RR x Td
= 111.11 x 0.0247/s x 8.325s = 6.86 %

KC =

100
PB

100
6.86

= 14.58

CALCULATION OF OPTIMUM P AND I OF SELF REGULATING PROCESS USING

NUMERICAL METHOD (FIC31)

PV, m3 /h
0.723926
0.723926
0.793828
0.874512
0.916816
0.937793
0.948281
0.948281

Time, h:m:s
10:33:39
10:33:40
10:33:41
10:33:42
10:33:43
10:33:44
10:33:45
10:33:46

Time, s
0
1
2
3
4
5
6
7

PV, %
24.13
24.13
26.46
29.15
30.56
31.26
31.61
31.61

RR1 =

24.4624.13
( 2)(1)(5)

= 0.233

RR2 =

29.1524.13
(2)(1)(5)

= 0.502

RR3 =

30.5626.46
(2)(1)(5)

= 0.410

MV%
35
40
40
40
40
40
40
40

RR, 1/s
0.233
0.502
0.410
0.211
0.105

RRMAX = RR2 = 0.502/s

Td , s

Tc , s

2.98

Td = 3 2(1)

Tc = 2(1)

29.1523.13
29.1524.13

31.6124.13
29.1524.13

= 1s

= 2.98s

I = 3.33 Td
= 3.33 x 1s = 3.33 s

PB = 111.11 x RR x Td
= 111.11 x 0.502/s x 1s
= 55.78%

KC = 100/PB
= 100/55.78 = 1.626

1) FIC21 SELF-REGULATING GRAPHICAL ANALYSIS

2) LIC11 NON SELF-REGULATING GRAPHICAL ANALYSIS
10

3) FIC31 SELF-REGULATING NUMERICAL ANALYSIS

Table summary of parameters and optimum controller settings for PI controller based on ZieglerNichols
SYSTE

PROCES

LOO

PROCESS

M
Delta V

S
Flow

P
BEHAVIOUR
FIC21
Self-

Delta V

Level

LIC1

Foxboro

Flow

1
FIC31

PB

KC

I (s)

Td

TC

RR

(s)
2.5

(1/s)
0.442

(%)
61.5

1.62

4.162

(s)
1.2

Regulating
Non Self

0
6.86

6
14.5

5
8.325

5
2.5

0
-

8
0.024

Regulating
Self

55.7

8
1.79

3.33

2.9

7
0.502

Regulating

Table of Tuning Test, Set Point Test and Disturbance Test

*Controller Gain (Kc) , Integral Time (I) and Proportional Band (PB)
SYSTE

PROCES

LOO

Delta V
Delta V
Foxboro

Flow

FIC2

Level

1
LIC1

Flow

1
FIC3
1

TUNING TEST

SET POINT

DISTURBANC

TEST

E TEST

PB

KC

I (s)

OLD

NEW

(MV5%)
OLD
NEW

(%)
-

0.40

16.6

1.1m3/

1.5m3/

34.7

44.7

7
14.5

5
8.32

h
h
500mm 540mm 30.0

40.0

233.1

8
-

5
13.2

1m3/h

41.85

1.5m3/

31.85

DISCUSSION

The experiment is to determine the parameters of RR, Td and Tc from the curve using
Tangent Method. The open and closed loop test was conducted for FIC21, LIC11 and FIC31.

11

Then, the Ziegler-Nichols tuning rules was calculated by using parameters RR and Td calculated
before. For a fast process, PI formula is used in this calculation.

The values of P, I and Kc are used for closed loop. The values is set up and process run in
Auto mode. The values of Kc, P and I is enter into the system control to get the result. For
example, the curve of FIC21 began to oscillate actively so the value of Kc is divided by 4 and
value of I is times by 4. The process turn into stable after the values changes and is closed to set
point. Kc and I are considered as the control variables to gain the optimum controller setting.
Therefore, the value of process to achieve optimum value is undefined. Hence, the graph line set
point and PV are the optimum value under steady state condition.

For FIC21, from open loop test, parameters such as RR, Td and Tc is calculated by using
Tangent Method. The values is 0.4428/s , 1.25s and 2.5s respectively. Then, by using tuning
rules, the values of PB is 61.50%, Kc is 1.626 change into 0.407 and I is 4.1625s into 16.65s for
tuning test in closed loop test. The curve is fluctuation and not stable. Hence, by dividing and
times the values of Kc and I turn the process to stable and slower. For set point test, the values of
set point is increases from 1.1m3/h to 1.5m3/h. For load disturbance test, the value of MV
increases from 34.7% to 44.7% which is 10% differences.

For LIC11, same method is used by using Tangent method and Ziegler-Nichols tuning
rules to obtain the parameters as above. However, for the closed loop test the process is oscillate
and stable. Hence, no change in the value of PB, I and Kc. The values of RR and Td is 0.0247/s
and 2.5s. There is no value of Tc because the process in non self-regulating. The values of PB,
Kc and I is 6.86%, 14.58 and 8.325s respectively. The set point increase in set point test from
500mm to 540mm. The value of MV also increase by 10% in load disturbance test.

And lastly, for FIC31 the Numerical Tangent Method was used to obtain the result of
parameters. The values of RR, Td and Tc is 0.502/s, 1s and 2.98s respectively. After that, for
12

closed loop the process is oscillate and not stable. Then, the values of PB is times to 4 from
55.78% to 223.11% and I is times with 4 too from 0.056min to 0.222min. Then the process
become stable and closed to set point. The values form set point test is 1m3/h to 1.5m3/h. The
result of load disturbance test is 31.85% to 41.85%. Through the process, we can see if the
process need to change the control variable is the process is not stable or too slow.

CONCLUSION

The aims of this experiment is to analyze open loop process response which consists of
value of response rate (RR), dead time (Td) and time constant (Tc), to determine the effect of
13

tuning rules Zieglar-Nichols and observe the effect change of load disturbance and set point
during the process. For open loop test, the value of Response Rate(RR), dead time (Td) and time
constant (Tc) is determine by Tangent Method through graphical and numerical analysis.
Graphical analysis is used for FIC21 and LIC11 and numerical analysis is for FIC31. These
values of RR and Td are then substituted into the Tuning Rules by Ziegler-Nichols to be used for
the closed loop test in order to achieve optimum control value in the process. Once the process
characteristics have been identified, the P+I mode is selected because it is used for fast process
response.

For closed loop test, the values of Kc and I then are inserted into the Gain and Reset
section respectively. If the process oscillates and not stable, the value of Kc will reduce by divide
to 4 and increase value of I by times to 4. These value are applied for loop FIC21 and FIC31
which is not stable. After that, the process become stable and near to set point. Next move on to
set point test. The set point test changes for set point about 10% from the original set point. Then
the load disturbance test. The test is changes of MV about 10% from the current MV until the
process become stable.

REFERENCES
1. Abdul Aziz Ishak, & Anuar Ahmad. (n.d.). Application of Numerical Technique in
Tangent Analysis of a Discrete Step Response Data. Retrieved from

14

http://aabi.uitm.edu.my/controlnotes/PID%20Tuning/numerical%20analysis%20of
%20tangent%20method.pdf
2. King Saud University. (2002). INTRODUCTION TO PROCESS CONTROL.
Retrieved from http://home.manhattan.edu/~gennaro.maffia/Day%2010%20King
%20Saud.pdf
3. Abdul Aziz Ishak. (2013). PID TUNING (FUNDAMENTAL CONCEPTS AND
APPLICATIONS) (3rd ed.). KUALA LUMPUR, 56000: UITM PRESS.

APPENDICES

15

OPEN LOOP TEST

16

CLOSED LOOP TEST (TUNING TEST, LOAD

DISTURBANCE TEST AND SET POINT TEST)

17