Dynamics, Simulation, and Control of a Batch Distillation Column using Labview

Abstract

In this work, the dynamic behaviors of batch distillation introduced by using a step changes in the reflux and find the response of this test on the concentration of distillate methanol. And it has been recognized that first order plus dead time (FOPTD) represent the process dynamics of batch distillation. The operating strategies for batch distillation were studied methanol / water mixture, 30% mole of methanol and 70% mole of water. Batch distillation column with eight trays was operated and controlled in real time by using LabVIEW program; giving the temperatures data and finding the concentration of distillated methanol using an empirical temperature-composition relationship model. Zieglar-Nichols tuning rules were applied to determine the parameters of the implemented PID controller. It was found that 𝑘𝑝=1.93, Ti =2760 sec, 𝑎𝑛𝑑 Td = 690 sec. The comparisons of the designed PID controller with experimental work are included (according to rise time, percentage overshoot and settling time). MATLAB Simulink was used to simulate the system using PID controller.

Full text

International Journal of Current Engineering and Technology E-ISSN 2277 – 4106, P-ISSN 2347 – 5161
©2016 INPRESSCO®, All Rights Reserved Available at http://inpressco.com/category/ijcet
Research Article
303| International Journal of Current Engineering and Technology, Vol.6, No.1 (Feb 2016)
Dynamics, Simulation, and Control of a Batch Distillation Column using
Labview
Naseer A. Habobi† and Sura M. Yaseen†*
†Department of Chemical Engineering, University of Al-Nahrin, Iraq
Accepted 15 Feb 2016, Available online 22 Feb 2016, Vol.6, No.1 (Feb 2016)
Abstract
In this work, the dynamic behaviors of batch distillation introduced by using a step changes in the reflux and find the
response of this test on the concentration of distillate methanol. And it has been recognized that first order plus dead
time (FOPTD) represent the process dynamics of batch distillation. The operating strategies for batch distillation
were studied methanol / water mixture, 30% mole of methanol and 70% mole of water. Batch distillation column
with eight trays was operated and controlled in real time by using LabVIEW program; giving the temperatures data
and finding the concentration of distillated methanol using an empirical temperature-composition relationship
model. Zieglar-Nichols tuning rules were applied to determine the parameters of the implemented PID controller. It
was found that

=1.93, Ti =2760 sec,

Td = 690 sec. The comparisons of the designed PID controller with
experimental work are included (according to rise time, percentage overshoot and settling time). MATLAB Simulink
was used to simulate the system using PID controller.
Keywords: Batch distillation, LabView, Real time control system
1. Introduction
1
Batch distillation is one of the oldest liquid separation
units that are used for producing high purity products
and valuable specialized chemicals in small quantities.
One of the principal advantages of using batch
distillation is the flexibility. It means that one column
can be designed to separate different mixtures with
different compositions and specifications for final
purity (H. Galindez, et al 1988; A. Klingberg, 2000). The
other advantage of batch distillation is the lower
capital cost and relative simplicity compared with
larger continuous distillation columns. (S. Skogestad et
al 1997)
Although batch distillation is known to be less
energy efficient than its continuous counterpart, it has
received renewed interests in recent years due to the
flexibility it offers. In general to control on batch
distillation process one of following progress can be
applied:
1) Constant reflux: Reflux is set at a predetermined
value that is maintained for the run. Since pot
liquid composition is changing, instantaneous
composition of the distillate also changes.
2) Varying reflux with constant overhead
composition, the amount of flow rate of liquid
*Corresponding author: Sura M. Yaseen
returned to the column is constantly increased to
maintain a constant distillate composition.
3) optimize the reflux in order to achieve the desired
separation in a minimum of time or with a
maximum thermodynamic efficiency. Complex
operations may involve withdrawal of side
streams, provision for inter condensers, addition
of feeds to Another control method which is
possible to trays and periodic charge addition to
the pot. (Maira Mendes et al 2009).
Frattini et. al. 1997, developed a self-adjusting gain
scheduling PI control method and used it to the control
of a binary batch distillation column. This method
reached the control goal by continuously adjusting the
gain to higher value. Li and Wozny introduced an
adaptive control strategy to follow the pre-defined
optimal reflux-ratio profiles.
Monroy and Alvarez use a robust nonlinear control
method to compensate modeling errors, and adjust the
reflux ratio to control the composition of a batch
distillation process. Up to now, although researchers
around the world have put forward many advanced
control strategies for the control of batch distillation
process, most of these control strategy research only
carried out theoretical investigation or simulation
study, and few of them are applied to practical industry
batch distillation process or tested by experiment.
(ZhiyunZou et al, 2006)
Model predictive control (MPC) is also used in
control strategy in the process industry.

Naseer A. Habobi et al Dynamics, Simulation, and Control of a Batch Distillation Column using Labview
304| International Journal of Current Engineering and Technology, Vol.6, No.1 (Feb 2016)
Fig.1Process diagram and unit elements allocation of Temperature Sensor
Fig.2 Computer Controlled Batch Distillation Unit
It’s used for years in advanced industrial applications it
is because accuracy also suitability for on-line control
must be taken into account. First of all, the model must
be accurate enough, capable of precise long-range
prediction. An inaccurate model gives erroneous
predictions which are likely to result in unacceptable
control. Although classical linear MPC algorithms
which use linear models give good or satisfactory
results quite frequently, the majority of processes are
nonlinear by nature. For such processes nonlinear MPC
algorithms based on nonlinear models must be used
(Maciej Lawry, 2011)
2. System Description
- The unit is basically composed by a boiler, a reflux
system and a tank for the distillation.
- Binary system in which methanol/water mixture is
used (feed = 1 liter).
- The steam of MVC goes to the head of the column is
sent to a total condenser.
- Sieve Plates Column with 8 plates with one
temperature taking and sample. The cooling water
flow that crosses the condenser is regulated and
indicated in a flow sensor.
- The pressure loss in the column can be measured
with a pressure manometer.
- The temperatures of the system are measured by
sensors placed in each stage measuring
temperature of vapor liquid equilibrium as shown
in Fig. 1.
- The system connected to computer controlled unit
which is includes:
The unit + a Control Interface Box + an 2560 Mega
Arduino board which was used as a Data
Acquisition Board interfacewith 16 analog inputs
and 10 Pulse Width Modulation (PWM) outputs to
control on:

Naseer A. Habobi et al Dynamics, Simulation, and Control of a Batch Distillation Column using Labview
305| International Journal of Current Engineering and Technology, Vol.6, No.1 (Feb 2016)
- The temperature of the heating element
(boiler heater).
- The solenoid valve (reflux ratio).
- 2560 Mega Arduino was used because it is more
accurate in connections between the temperature
sensors and PC + Computer Control, Data
Management Software Packages, for controlling
the process and all parameters involved in the
process.
- The temperature sensors (type J) of column is
connected to Arduino Mega 2560 which is placed
inside the Control Interface Box, as shown in Fig. 2.
- I/O USB single cable between the control interface
box and computer
- Computer process was built using LABView
including the front panel and block diagram as
shown in Fig. 3
- Real time curves representation about system
responses.
- Recording of all the process data and
- results in a file such as (Microsoft Word, xls, TXT)
- Graphic representation, in real time, of all the
process/system responses.
- Both the boiler heater and solenoid valve values
can be changed at any time allowing the analysis
about curves and responses of the whole process.
- All the outputs relays controlling both boiler
heater and solenoid valve and sensors values and
their responses are displayed on only one screen
in the computer.
Fig.3 Front Panel & Block Diagram of LABView
program
3. Experimental Work
All the feed of methanol-water is pumped into the still
pot of the reboiler in one time before distillation. The
reboiler is heated up and the flow-rate of steam is
controlled by switched on relay by DAQ. The methanol
vapor at the top of the column is cooled down by
flowing through a cooling water condenser. The flow
direction of the condensed methanol liquid is time-
sharing controlled by an electric solenoid valve. Basic
information about the experiment is summarized in
Table 1.
4. Results and Discussion
The reflux ratio R remains constant in the whole
process until the concentration of overhead goes down
with time.
From Table 2, it is clear that the experiments which
made at higher value of reflux ratio R= 100, permit to
obtain the maximum recovery of pure methanol, Next
figure also illustrates the relation of concentration of
distillate methanol with all reflux ratio ranges.
Table 2 the experimental results for batch distillation
constant reflux ratio
R%
x concentration
20
0.15333
40
0.2389
60
0.48708
80
0.6706
100
0.9477
4.1 Variable Reflux
The constructed controller made by LabVIEW program
is implemented with the batch distillation system to
control the distillate concentration. This is achieved by
changing the reflux ratio as manipulating variable to
control on temperature which is represent a
(concentration) of the distillate (top plate).
4.1.1 Dynamic behavior for implemented model
It is necessary to identify model parameters from
experimental data. The simplest approach involves
Introducing a step test into the process and recording
the response of the process. Then step changes of
variable reflux ratio are applied 20-40, 40-60, 60-80,
and 80-100. The response of the top distillate
concentration is shown in Fig. 5.
Fig.4 Experimental results with a constant reflux ratio
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
30 40 50 60 70
Methanol Concentration x
Time (min)
R=80
R=60
R=40
R=20
R=100

Naseer A. Habobi et al Dynamics, Simulation, and Control of a Batch Distillation Column using Labview
306| International Journal of Current Engineering and Technology, Vol.6, No.1 (Feb 2016)
It has been recognized that a first order may in general
represent process dynamics of batch distillation. The
calculations are carried further to estimate the steady
state in and the process time constant and time delay.
Fig. 5 The response of the top distillate temperature to
step change for manipulating reflux ratio 20-40
Fig. 6 shows the open-loop test for the four step
changes in manipulating variable reflux ratio from 20
to 100.
Fig.6 Open-loop test for the four step changes in
manipulating variable reflux ratio from 20-40, 40-60,
60-80, and 80-100
Using the experimental data in Fig. 6, transfer function
and PID controller are designed to maintain
composition specifications for the distillate product
taking the reflux flow rate as manipulated variable.
(Dale E.Seborg, 2004; Katsuhiko Ogata, 2010)
4.1.2 First order model
The response for the outlet temperature to step
change showing in Fig. 5 can recognize that a first
order plus time delay (FOPDT) may in general
represent process dynamic of the batch distillation and
by drawing a tangent line to a process variable PV
(controlled temperature) and using controller output
CV can calculate the gain, time constant and time delay
(Naseer A. Habobi, 2010) the process parameter we get
it as following.
    
   (1)
   
  (2)
and
Time delay L= 23 min (1380 sec)
Time constant τ =ymax – (∆y×0.63) (3)
τ=78-(5.5 ×0.63) = 75.035
this value represented on Fig. 5.2 and can read time
constant value which is =37 min
τ=37 min (2220 sec)
Now the transfer function of the first order plus dead
time (FOPDT) can be deduced as follow:
  
  (4)
  
  (5)
4.1.3 Initial Estimation for PID Controller parameters
Ziegler Nickols tuning method can be considered as
one of the earliest closed loop tuning method. This
method requires the value of the plant parameters
obtained to be keyed into the formula. Ziegler and
Nickols have developed PID tuning methods back in
the early forties based on open loop tests and also
based on closed loop test, which may be their most
widely known achievement.
The open loop method allows calculating PID
parameters from the process parameters. The
procedure as follows:
 Step 1: Make an open loop plant test(e.g a step
test)
 Step 2: Determine the process parameters: process
gain, dead time, time constant. (As illustrate
above).
 Step 3: Calculate the parameters using table 3
(Ahmad Athif, 2008)
72
73
74
75
76
77
78
79
35 40 45 50 55 60 65
Controlled Temp.°C
0
5
10
15
20
25
30
35
40
45
35 40 45 50 55 60 65
Reflux Ratio
Time (min)
60
65
70
75
80
35 45 55 65 75 85 95 105 115
Controlled Temp. °C

Naseer A. Habobi et al Dynamics, Simulation, and Control of a Batch Distillation Column using Labview
307| International Journal of Current Engineering and Technology, Vol.6, No.1 (Feb 2016)
4.2 PID Controller Design
The design of a controller requires specification of the
parameters: proportional gain (Kp), integral time
constant (Ti), and derivative time constant (Td). There
is crucial problem to tune properly the gains of PID
controllers because many industrial plants are often
burdened with the characteristics such as high order,
time delays and nonlinearities. Ziegler-Nichols open
loop tuning method is used to find the optimal PID
parameters. The optimal parameters obtained for the
reflux ratios are applicable over the entire useful range
(20 to 100) which is represents 20% to 40% from the
controller output. (Katsuhiko Ogata, 2010)
Table 3 Ziegler–Nichols Tuning Rule Based on Step
Response of Plant
PID Controller
Kp
Ti
Td
1.2 × τ/L
2L
0.5 L
So the estimation of PID parameters became as follow:
Kp= 37/23 = 1.93
Ti=2 × 1380 = 2760 sec
Td= 0.5 × 1380 = 690 sec
To verify the PID parameters, MATLAB-SIMULINK is
used to simulate the system Fig. 7 Show the block
diagram, and Fig. 8 shows the scope curve. (Dale
E.Seborg, 2004)
Fig.7 Simulation of close loop PID control using
MATLAB- Simulink
Fig. 8 The Scope view by MATLAB- Simulink
4.2.1 Response Experimental Results with PID Controller
To study the performance of the controller The
experimental data which is correspond to the
manipulating variable reflux ratio changes 20-40, 40-
60, 60-80, and 80-100 are compares with the process
model of PID controller. Fig. 9 showed response for
close loop control process for 20-40 reflux ratio with
PID controller.
Fig. 9 Response for the open loop control temperature
for 20 to 40 reflux ratio with PID controller
The unit-step response of this system can be obtained
with MATLAB and the resulting unit-step response
curve is shown in Fig. 10.
Fig.10 Unit step response of the implemented system
with PID controller
From Fig. 10 and by using MATLAB following
parameters are obtained:
 Rise Time: 4.8774×103
 Settling Time: 8.6848×103
 Settling Min: 0.0497
 Settling Max: 0.0550
 Overshoot: 0
Fig.11 Response for the open loop control temperature
for 40 to 60 reflux ratio with PID
0
10
20
30
40
50
60
70
80
90
1000 2000 3000 4000 5000
Temperature °C
Time (sec)
Expiremental work
PID
0
10
20
30
40
50
60
70
80
3500 4500 5500 6500 7500
Tempreture °C
Time (sec)
Experimental work
PID

Naseer A. Habobi et al Dynamics, Simulation, and Control of a Batch Distillation Column using Labview
308| International Journal of Current Engineering and Technology, Vol.6, No.1 (Feb 2016)
Fig. 12 Response for the open loop control
temperature for 60 to 80 reflux ratio with PID
Fig. 13 Response for the open loop control
temperature for 80 to 100 reflux ratio with PID
Fig. 14 Response for the close loop control process for
20 – 100 Reflux ratio for experimental work with PID
controller
The data obtained from the MATLAB for PID controller
and experimental data is summarized in Table 4 to
compare the performance of each case. It is noted that
experimental results have a faster rise time compared
to PID and need less time to achieve the set point value.
Experimental results showed also less in settling time
compared with PID controller both cases give zero
percent overshoot
Table 4 The performance of PID and experimental
work
PID controller
Experimental
work
Rise time tr
4.8774×103sec
1140 sec
Settling time ts
8.6848×103sec
3300 sec
Over shoot %
0
0
Figures 11-13 showed, respectively, the experimental
data that correspond to reflux ratio changes 40-60, 60-
80 and 80-100 with PID controller. And final Fig. 14
represents the complete controlled range from 20 –
100 reflux ratio.
Conclusions
Batch Distillation is highly nonlinear process;
therefore modeled using two point method to identify
the process which is first order plus dead time
(FOPTD) and by using Ziegler-Nichols method to
estimate the controller parameter and tuned by
MATLAB Simulink to give PID controller, PID controller
has sensitive proportional gain. LABView was
programming on PC to control on distillation column
and compared with proposed PID controller; actually
the comparisons are made between the process
performance (rise time, percentage overshoot, and
settling time).
From the present work carried out to study the
optimal operation for batch distillation in different
control strategies the following conclusions are
obtained:
1- At constant reflux operation, it was found that
increasing reflux ratio will increase the distillate
concentration but it will be at the expense of the
experiment time and power consumption.
2- Studying the dynamic behavior by introducing a step
test into the process and recording the response. It has
been recognized that first order plus dead time
(FOPTD) may represent process dynamics of batch
distillation. The transfer function was found to be:
  
   
3- Zieglar-Nichols tuning rules were applied to
determine the parameters of the implemented PID
controller. It was found that:
Kp= 1.93
Ti=2760 sec
Td= 690 sec
4- MATLAB Simulink was used to simulate the system
using PID controller, flowing data are obtain
Rise Time: 4.8774×103sec
Settling Time: 8.6848×103sec
Overshoot: 0
5-By comparing the experimental work result obtained
by control with LabView with Simulink of PID
controller result obtained in Table 4 showed that: the
experimental results have a faster rise time compared
to PID and need less time to achieve the set point value.
Experimental results showed also less in settling time
compared with PID controller. And it’s obvious that
rise time for experimental work less than PID
controller by 3737 sec.
Table 5 the experimental results for batch distillation
variable reflux ratio
R%
Temp.
x concentration
Duration
0-20
78
0.6694
16min
20 -40
72.5
0.8011
25 min
40- 60
69.58
0.8727
18 min
60- 80
67.62
0.92234
16 min
80-100
66.64
0.9478
7 min
0
10
20
30
40
50
60
70
80
4500 5500 6500 7500 8500
Tempreture °C
Time (sec)
Experimental work
PID
0
10
20
30
40
50
60
70
80
5500 6500 7500 8500 9500
Tempreture °C
Time (sec)
Experimental work
PID
0
10
20
30
40
50
60
70
80
90
1000 6000 11000
Tempreture °C
Time (sec)
Expiremental
work

Naseer A. Habobi et al Dynamics, Simulation, and Control of a Batch Distillation Column using Labview
309| International Journal of Current Engineering and Technology, Vol.6, No.1 (Feb 2016)
From above table we can obtain the average
concentration at 66 min approximate 1 hour and it was
equal to 0.8859
References
H. Galindez, and A. Fredenslund, (1988), Simulation of
Multicomponent Batch Distillation Processes, Comput.
Chem. Engng., 12, 281
Klingberg, Jan (2000), Modelling and optimization of batch
istillation, Lund Institute of Technology.
S. Skogestad, B. Wittgens, E. Sørensen, and R. Litto, (1997).
Multivessel Batch Distillation, AIChE J., 43, 971
Maira Mendes Lopes, TahWun Song 12 December. (2009).
Chemical Engineering and Processing 1298–1304 Batch
distillation: Better at constant or variable reflux,
Department of Chemical Engineering, Polytechnic School,
University of São Paulo, Av. Prof. Luciano Gualberto, trav. 3,
No. 380, 05508-970 São Paulo, SP, Brazil
A. M. Frattini-Fileti and J. A. F. Rocha-Pereira(1997),, The
development and experimental testing of two adaptive
control strategies for batch distillation, in Distillation and
absorption ’97, R. Darton Ed., Rugby, UK: IChemE, pp. 249-
259.
ZhiyunZou and Dehong Yu (2006), State Space Modeling and
Predictive Control of a Binary Batch Distillation Column.
School of Mechanical Engineering XianJiaotong University
Xian 710049, Proceedings of the 6th World Congress on
Intelligent Control and Automation, Dalian, China
Maciej _Lawryinczuk (2011), Predictive Control of a
Distillation Column Using a Control-Oriented Neural
Model. Institute of Control and Computation Engineering,
Warsaw University of Technology ul.Nowowiejska 15/19,
00-665 Warsaw, Poland
Dobnikar, U. Lotriˇc, and B. ˇSter (Eds.): ICANNGA 2011, Part
I, LNCS 6593, pp. 230–239.
Dale E.Seborg, (2004) University of California, Santa Barbara,
Thomas F,Edgar, University of Texas at Austin
Duncan A. Mellichamp, University of California, Santa
Barbara Process Dyanimc and Control/ Second Edition,
Katsuhiko Ogata (2010), Modern Control Engineering/
Handbook Fifth Edition
Naseer A. Habobi, (2010), PC-Based Controller for Shell and
Tube Heat Exchanger Iraqi Journal of Chemical and
Petroleum Engineering Chemical Engineering Department
- College of Engineering - University of Nahrin – Iraq.
Vol.11 No.1 47-53, ISSN: 1997-4884
Ahmad Athif Bin MohdFaudzi (2006), Implementation of
generalized Predictive Control (GPC) for a real time
process control using Labview Faculty of Electrical
Engineering / UniversitiTeknologi Malaysia Master of
engineering (Electrical Mechatronics and Automatic
Control)