Aldebrez1997 Real TimeSimulationOfLiquidProcessPlant

317 D-3
REAL-TIME SIMULATION OF LIQUID PROCESS PLANT

USING GENESIS SOFTWARE
Fareg Mohamed Aldebrez, Mortaza Bin Mohamed, and Shaharum Bin Sulaiman
Faculty of Electrical Engineering, Universiti Teknologi Malays,ia

Locked Bag 79 1, 80990 Johor Bahru, Malaysia
E-mail: debrez@fkeserv.fke.utm.my
Abstract
This paper presents the development of a mathematical model of a multivariable liquid process plant.
This multivariable liquid process rig incorporates two interacting loops; level control loop and
temperature control loop. Based on the developed model, an experiment was conducted on the actual
system for the purpose of real-time data collection and estimating the parameters of the model using a
real-time software called GENESIS. This software is a real-time, multitasking software used in this
work for running the actual plant and collecting the real data from the actual process. It is also used to
perform a computer simulation study on the developed model to verify its performance. The
simulation study was performed in terms of three case studies, that are, level conhrol loop, temperature
control loop and interaction loop respectively. The performance of each loop was investigated and the
extent of the interaction effect between the two single loops was also highlighted.
Keywords
Modeling, Simulation, GENESIS Real-Time Software, and Process Control.
1. INTRODUCTION
2. LIQUID PROCESS PLANT
As the industry develops steadily, the production scale
becomes larger and in the meantime the process In this Liquid Process Plant, there are two interacting
complexity grows. Especially, in some production loops, namely, level control loop and temperature
processes, the controlled variables are related to each control loop. Figure (1.0) shows the schematic diagram
other. Hence, in such a case it is inevitable that of the multivariable liqui'd process plant.
interaction should be taken into consideration when I
some variables are under control. Because the

interactions are mutual, so there must be some coupling
channels among the single variable control systems
Consequently, the control system being treated is no
longer a single variable system. This multivariable
theory has become a very important field in process
control theory and by depth of its contents, it has been
admitted in general, that it is one of the most difficult
fields in process control theory.
The objective of this paper is to develop a mathematical
model of a multivariable liquid process plant and then
conduct a computer simulation study to verify its
performance in comparison to the actual plant
performance.
Fig. (1.0): Multivariable Liquid Process Plant.
SICE '97 July 29-31, Tokushima PR0001-3/97/'0000-1321$4.00 01997 SICE

1321
In this section, we will describe the signal flow of the
process and how the two single loops are interacting
with each other. Firstly, the cold water is pumped from
its storage tank into the reaction tank. Likewise, the hot
water is also pumped from its storage tank to the
reaction tank where the reaction takes place. The hot Secondly, the transfer function of the temperature
water flows through a coil inside the reaction tank. control loop is stated in equation (2.0) as follows:
Therefore, the interaction between the two single loops
r
occurred only in terms of the water temperature.
However, the water level inside the reaction tank is only -in
affected by the inlet and outlet flow of the cold water.
After that, the hot water completes its loop and flows T=< (2.0)
back into its storage tank. However, the cold water with
its new temperature flows into tank 101 and considered
as a final product. The main objective of this paper is to
control the level and temperature of this product inside
tank 101. Moreover, it is also aimed to evaluate the Equations (1.0) and (2.0) can be written in the transfer
interaction between the two loops. function matrix form as follows:
3. PLANT EQUATIONS
Based on some mass and energy balance equations

assumptions, a mathematical model represents ') , the
actual process was developed. The transfer function
matrix was derived using two methods; block diagram (3.0)
simplification, and solving the mathematical equations
of the process. This model was used to construct the
strategy during the performance of the simulation study,
and its performance was also investigated in the
simulation study later. The block diagram of the
multivariable process is shown in Fig. (2.0) below:
4. COMPUTER SYSTEM & PLANT INTERFACE
4.1 GENESIS Software
GENESIS software used in this work is a real-time

software developed by ICONICS, Inc. It consists of
three modules; Strategy Builder, Graphics Builder and
Runtime module.
The GENESIS Runtime system is a real-time,
multitasking environment that allows us to run and
control the strategy created in the Strategy Builder using
U built in commands and a graphic interface I). It also
allows us to use the graphic displays created in the
Figure (2.0): B. D.of 2x2 Multivariable Control Loop Graphics Builder to assist our operators. It is used to run
the actual process for the purpose of real data collection
From the block diagram shown in Fig. (2.0), a closed from the actual rig and estimation of system parameters.
loop transfer function of the whole process was
obtained by solving the equations of each single loop
separately. Firstly, the transfer function of the level
control loop is obtained as in equation (1 .O):
- 1322 -
4.2 Plant Interface OPTOMUX is capable of using two types of message
protocols. The first protocol is 2-pass protocol which is
The actual process was connected with the host through used during normal operation. The other one is 4-pass
an intelligent digital and analog I/O device called protocol. This protocol is sometimes useful during
OPTOMUX I). It is a 4,s or 16 point rack that initial setup and installation because it allows the host to
accommodates photo-isolated analog or digital power examine and display the command massage that
I/O modules. Each OPTOMUX unit consists of a OPTOMUX received.
removal brain board and an I/O mounting rack.
OPTOMUX units communicate with the host computer
over an RS-422/485 communications link. It is possible 4.3 System Parameters Data Collection
to communicate with up to 256 individual units on a
single serial link for a total of 4096 digital and/or analog There are ten parameters involved in the whole liquid
I/O. The connections between the host computer and process plant. The level control loop has five gains, (k,,
OPTOMUX units is illustrated in Fig. (3.0). k2, k,, k,, and k5), while the temperature control loop
OPTOMUX units can be located where the actual involves four gains (k,, k,, k,, and k,,,), K, represents the
control is needed (in the process field). interaction between level control loop and temperature
control loop. The values of these parameters were
obtained by experimental study conducted on the actual
process. The final values of the system parameters are
ZOONTROI FIELD listed in table (1 ,O).
ROOM
OPTOMUX Cabinet
Table (1.0): List of the system parameters values
GENESIS was also used to perform the simulation

study on the developed strategy. For instance, in the
first case study, Algorithm Blocks that run level process
were created in the Strategy Builder. Each algorithm
block of the strategy performs a preprogrammed
function. After that, graphic displays were created in the
Graphics Builder to enteir and display information for
monitoring and controlling the level control strategy.
Then a system simulation was performed in the runtime
mode using the combination of strategy and graphics. A
similar procedure is done for constructing the other two
case studies using the corresponding strategy. For the
three case studies, the developed strategies of level,
temperature and 2x2 multivariable control loops are
shown in Fig. (4.0), (5.0) and (6.0) respectively.
B2- 16A
Fig. (3.0): Host Comp. and OPTOMUX connection.
Fig. (4.0): The Strategy of Level Control Loop.

The host computer issues instructions to OPTOMUX by
sending messages over the serial communications link.
1323 -
cold water inlet to the reaction tm.k, the level inside
tank10 1 increases, however, the temperature decreases.
This is considered as a clear indication of the interaction
between the two interacting loops.
Fig. (5.0): Temperature Control Loop Strategy.
Fig. (7.0): Response of the 2x2 Multivariable Loop.
Fig. (6.0): Multivariable Control Loop Strategy.

..... ..... ........... ........
5. PLANT SIMULATION AND RESULTS
Based on the mathematical model which was developed

earlier, a system simulation study has been performed to
verify the performance of that mathematical model.
Fig.(8.0): Response of the Multivariable Loop.
This simulation study was performed using GENESIS
software in terms of three case studies; two single loop
case studies, and one multivariable loop case study. In
system simulation study, the time scale was 30 min. : 3
hours. In every case study, a suitable strategy to run the
process was built in the Strategy Builder. After that, a
corresponding static and dynamic displays were created
in the Graphics Builder in order to enter and display
information for monitoring and controlling the process.
Then, the developed strategy was run and monitored in
the Runtime module. The responses of the two single
loops (level and temperature) are shown in Fig. (7.0)
and (8.0). The response of the multivariable control
loop is illustrated in Fig. (9.0). From the obtained
results it is noted that both level and temperature
responses were obtained and reached the desired values. Fig. (9.0):Response of the 2 x 2 Multivariable Loop.
Finally, the interaction effect between the two control
loops (level and temperature) is shown in Fig. (10.0).
From this figure, it can be shown that as we increase the
- 1324
I Si;! 58 , z ;+ I i 1s
I
1,~
I
1
4 . ... . ., .I.... ...... ..... .,.
Fig. (10): Interaction Effect on the Multivariable Loop.
6. CONCLUSION
In this paper, a mathematical model that represents the

actual multivariable liquid process rig was developed
and a closed loop transfer function matrix was obtained.
As a result of the conducted simulation study, the
developed strategy performed well and showed a
satisfactory results. The interaction between the two
interacting loops was also demonstrated as shown in
Fig. (1 0.0).
7. REFERENCES
[I] GENESIS Tutorial Guide, version 4.1, The

Strategy Builder, The Graphics Builder, and
Runtime, GENESIS Control Series, ICONICS,
Inc., 1994.
[2] Liu Chen Hui, General Decoupling Theory of
Multivariable Process Control Systems, Lecture
Notes in Control and Information Sciences, Edited
by A. V. Balakrishnan and M. Thoma, Springer-
Velrag Berlin Heidelberg, 1983.
[3] OPTOMUX Operations Manual, OPTOMUX BI
and B2 Digital and Analog Brain Boards
Operations hlanziaf, Cyrano, Inc., 1990.
[4] Pradeep B. Deshpande, and Raymond H. Ash,
Elements of Computer Process Control With
Advanced Control Application, Prentice-Hall, Inc.,
1983.
[5] Stephanopoulos George, Chemical Process
Control: An Introduction to Theory and Practice,
P T R Prentice Hall, Inc., 1984.
- 1325 -

Aldebrez1997 Real TimeSimulationOfLiquidProcessPlant

Uploaded by

Copyright:

Available Formats

Aldebrez1997 Real TimeSimulationOfLiquidProcessPlant

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Aldebrez1997 Real TimeSimulationOfLiquidProcessPlant

Uploaded by

Copyright:

Available Formats

317 D-3

REAL-TIME SIMULATION OF LIQUID PROCESS PLANT

Faculty of Electrical Engineering, Universiti Teknologi Malays,ia

Modeling, Simulation, GENESIS Real-Time Software, and Process Control.

some variables are under control. Because the

Fig. (1.0): Multivariable Liquid Process Plant.

SICE '97 July 29-31, Tokushima PR0001-3/97/'0000-1321$4.00 01997 SICE

Based on some mass and energy balance equations

4. COMPUTER SYSTEM & PLANT INTERFACE

4.1 GENESIS Software

GENESIS software used in this work is a real-time

GENESIS was also used to perform the simulation

Fig. (3.0): Host Comp. and OPTOMUX connection.

Fig. (4.0): The Strategy of Level Control Loop.

Fig. (5.0): Temperature Control Loop Strategy.

Fig. (7.0): Response of the 2x2 Multivariable Loop.

Fig. (6.0): Multivariable Control Loop Strategy.

5. PLANT SIMULATION AND RESULTS

Based on the mathematical model which was developed

Fig. (10): Interaction Effect on the Multivariable Loop.

In this paper, a mathematical model that represents the

[I] GENESIS Tutorial Guide, version 4.1, The

You might also like