Journal of Automated Methods & Management in Chemistry, 2005 (2005), no.

2, 50–54

c 2005 Hindawi Publishing Corporation

Mathematical Modeling of Perfect Decoupled

Control System and Its Application: A Reverse
Osmosis Desalination Industrial-Scale Unit

C. Riverol1 and V. Pilipovik2

1 Chemical Engineering Department, University of West Indies, St. Augustine, Trinidad and Tobago
2 Project Engineering Department, J. C. Engineering, Los Palos Grandes, Caracas, Venezuela

Received 5 February 2004; Accepted 20 February 2004

This short paper outlines the computer simulation using real data of a decoupled control system for a desalination unit. The
control strategy incorporated a perfect decoupled controller for the control of the fresh water flow and conductivity. The model
was estimated using real data and empirical tools instead of mass balances. The success is demonstrated in the reduction of wide
fluctuations in the variables of the process and decreasing of the sensibility to the changes of pressure and/or pH and allows
predicting problems of quality of water and waste of energy in the future.

1. INTRODUCTION water shortages across the globe [1]. Desalination costs have
been continuously decreasing over the years as a result of ad-
When we try to separate pure water and a salt solution
vances in system design and operating experience, and the
through a semipermeable membrane, the pure water diffuses
associated reductions in specific unit size and specific power
through the membrane and dilutes the salt solution. The
consumption.
membrane rejects most of the dissolved salts, while allowing
The most widely used desalination processes are mem-
the water to permeate.
brane separation via reverse osmosis (RO), and three types
This phenomenon is known as natural osmosis. As wa-
of thermal separation: multistage flash desalination (MSF),
ter passes through the membrane, the pressure on the di-
multiple-effect evaporation, with thermal vapor compres-
lute side drops, and the pressure of the concentrated solu-
sion (MEE-TVC) and without (MEE), and mechanical va-
tion rises. The osmotic flux continues until an equilibrium
por compression (MVC). The MSF and RO processes dom-
is reached, where the net water flux through the membrane
inate the market for both brackish water and seawater de-
becomes zero at equilibrium; the liquid level in the saline wa-
salination, with a total share of more than 90%. All three
ter will be higher than that on the waterside. The amount of
types of thermal desalination systems are equipped with con-
water passing in either direction will be equal. The hydro-
denser tube bundles. The MEE and MVC systems are divided
static pressure difference achieved is equal to the effective
into evaporating effects, while MSF systems are divided into
driving force causing the flow, called osmotic pressure. This
flashing stages. All of the systems employ a number of large
pressure is a strong function of the solute concentration and
pumping units, including pumps for seawater intake, dis-
the temperature, and depends on the type of ionic species
tillate product, brine blowdown, and chemical dosing. The
present. Applying a pressure in excess of the osmotic pres-
MSF and MEE systems have additional pumps for the cool-
sure to the saline water section slows down the osmotic flow,
ing seawater. In addition, MSF has pumps for brine recycle.
and forces the water to flow from the salt solution into the
In MSF and MEE, steam extracted from low- and medium-
waterside. Therefore, the direction of flow is reversed, and
pressure turbine lines provides the heat necessary for flash-
that is why this separation process is called reverse osmosis
ing or evaporation. In MSF, the heating steam is routed to
(Figure 1).
the brine heater; in MEE, the heating steam is routed to the
Continuous progress in desalination technology makes
first evaporating effect. The MSF process operates with a top
it a prime, if not the only, candidate for alleviating severe
brine temperature in the range of 90–110◦ C. The MEE and
MVC processes are operated with lower top brine tempera-
Correspondence and reprint requests to C. Riverol Chemical Engineer-
tures in the range of 64–70◦ C. MVC is distinguished from
ing Department, University of West Indies, St. Augustine, Trinidad and To- the other processes by the presence of a mechanical vapor
bago; E-mail: c.riverol@eng.uwi.tt. compressor, which compresses the vapor formed within the
Controlling a Desalination Plant 51

1 and 2, which employs membranes, has a simple layout,

and is compact and modular. Existing units can be expanded
to handle larger capacities. However, RO membranes are
more sensitive to the conditions of the feed seawater, scaling,
fouling, and pH than thermal processes. Furthermore, un-
like thermal processes, RO membranes do not provide high-
purity water. On average, the permeate salinity varies over a
range of 30–150 ppm. The actual value depends on the pro-
cess recovery, which is defined as the amount of product per
unit mass of feed water. Depending on the intended use of
the water, a second RO pass may be needed to reduce the
salinity to an acceptable level.
Fresh water Saline water The action control described in this paper is based on de-
coupled controllers applied to parallel transfer function pro-
Semipermeable cesses. This method shows sensitivity to disturbance and tun-
membrane ing of inner loops, however the conditions were studied and
(a) the sensibility was reduced.

2. DESIGN OF THE CONTROL SYSTEM

Osmotic BASED ON EMPIRICAL MODEL
pressure
Four system parameters can be controlled in an RO unit: feed
temperature, pH, conductivity, and pressure. The present
study discusses two parameters which should be monitored
and controlled for proper RO: conductivity (C) and inlet
fresh water flow (F), and does not include the effect of the
temperature over the behavior of the system. At industrial
scale, the pressure affects the behavior of the system more sig-
nificantly than the temperture. Additional parameters were
(b) not considered, however several literature can be consulted
[2, 3] with respect other models. The manipulated variables
Applied pressure are pH and feed pressure (P). The model was got empirically
using parameter estimation. The zero/poles discrete model
was obtained using real data and least square analysis and
later it was transformed in continuous transfer function (s-
transform). The result is similar to that in [2] although the
value of the parameters is different. The system is




F G11 G12 P
= × , (1)
C G21 G22 pH

(c) where
0.0045(0.104s + 1)
G11 = , G12 = zero,
Figure 1: Principles of reverse osmosis: (a) natural osmosis, (b) os- 0.012s2 + s + 1
motic equilibrium, (c) reverse osmosis. (2)
(−0.12s + 0.22) 10(−3s + 1)
G21 = , G22 = 2 .
0.1s2 + 0.3s + 1 s + 5s + 1
evaporator to the desired pressure and temperature. The sys- The system has to work well in a range of P = 800–1000 kPa,
tem also includes plate heat exchangers for preheating the F = 33000–54000 m3 /d and pH = 6–7.2.
feed using heat recovered from the brine blowdown stream As in the desalination plant, a manipulated input affects
and the distillate product. more than one controlled output. One approach to handling
All thermal processes produce a high-purity distillate this problem is known as decoupling [4]. The idea is to de-
product, with a salinity of less than 10 ppm. This is achieved velop “synthetic” manipulated inputs that affect only one
by a wire-mesh mist eliminator, which removes entrained process output each. This approach is illustrated in Figure 3
brine droplets formed in the distillate stream. The average where a multivariable process is controlled with a perfect de-
conventional sizes are 3 000 m3 /d for MVC, 33 000 m3 /d for coupler and single-loop controllers with r1 and r2 the set
MSF, and 12 000 m3 /d for MEE. The RO process, see Figures points, respectively.
52 Journal of Automated Methods & Management in Chemistry

Membrane
module

Distillate
Feed
seawater
Feed Product
treatment Energy treatment
recovery

Reject brine

Figure 2: Desalination plant.

r1 − u1 u1
Gc1 (s) G11 (s) y1
G12 (s)
−
G11 (s) G21 (s)

G21 (s) G12 (s)

−
G22 (s)
Gc2 (s) y2
G22 (s)
r2 − u2 u2

Figure 3: Control system.

The first step is to define the process transfer function order to cancel the y1 and y2 feedback and to cancel the de-
matrix: coupler when the constraints are not active. Inversion of the



 controllers at the inputs permits to reconstruct the manip-
y1 (s) G11 (s) G12 (s) u1 (s) ulated variables u1 and u2 which can then be used by two
= ×
y2 (s) G21 (s) G22 (s) u2 (s) other controllers to achieve control objectives with less pri-



 (3) ority than the constraints while respecting the allowable r1
F G11 G12 P and r2 set point values.
=⇒ = × .
C G21 G22 pH Pole-zero cancellation method led to the following con-
trollers for the parallel control method:
The perfect decoupler is built in the inverse form where each
   
branch of the decoupler is fed before the other branch pickup 0.012s2 + s + 1 10 s2 + 4.3s + 1
point. Its transfer function is Gc1 = , Gc2 = . (6)
(0.1s + 1) (−2.7s + 1)


G11 (s)G22 (s) −G12 (s)G22 (s) The primary objective of the control system was to keep the



u1 (s) −G11 (s)G21 (s) G11 (s)G22 (s) u (s) pressure at 800 psi and the conductivity at 450 µS/cm in the
= · 1 . (4)
u2 (s) G11 (s)G22 (s) − G12 (s)G21 (s) u2 (s) face of disturbances by the pressure or/and pH. The decou-
pled controller has the advantage that it does not need tun-
This decoupler is realizable only if the decoupler transfer ing as the classical PID; only if the model is changed the con-
function matrix is open-loop stable. It can be shown that the troller has to be modified. With respect to the behavior of the
transfer function matrix of the decoupler in series with the conductivity using real data, its value is practically constant
process is along the time using different disturbances in the pressure. A



 resume of the strategy is shown in Figure 4 with conductivity
y1 (s) G11 (s) 0 u1 (s) records collection at 5-minute intervals (22000 values) to a
= × . (5)
y2 (s) 0 G22 (s) u2 (s) 30% step.
The PID control originally gave very good results with
For the perfect decoupler in the inverse form, single-loop system response. This needed to be improved to remain com-
controllers are tuned on the single direct transfer functions petitive in a modern market, and has been achieved by the
G11 (s) and G22 (s). Controllers are then properly tuned even use of digital control. The decoupled control system offers
if one control loop is open. an easy solution to the new demand of the market and it
If the set points r1 and r2 are constrained variables, a con- is fast implementation. It is important to note that the pro-
troller structure can be used to modify the control strategy in cess has a delay time of 2.88 minutes as can be observed in
Controlling a Desalination Plant 53

0.6 1000
900
800
Deviation (set point)

0.4 700

Pressure (psi)
600
500
0.2 400
300
200
0 100
−60 −40 −20 0 20 40 60
Sample (%) 0
0 2 4 6 8 10 12 14 16 18 20
Time (min)
Figure 4: Behavior of the conductivity.

Figure 6: Behavior of the manipulated variable in real conditions.

1.4

1.2
Table 1: Resume of the results using a long-term control (ISE
1 stands for integral square error).
Flux (gpm)

0.8 Step over acid inlet flow (%) 10 20 30 40

0.6 Flow rate (ISE) 0.001 0.002 0.002 0.002
Conductivity (ISE) 0.001 0.003 0.007 0.007
0.4
Trans-memb pressure (ISE) 0.002 0.002 0.002 0.002
0.2

0
0 2 4 6 8 10 12 14 16 18 20 the constraints while respecting the allowable r1 and r2 set
Time (min) point values.
When saturation in not active, the variables ûi differ from
DC ui only by additive constants equal to the value of the con-
Actual system troller integrators. Therefore, the variable ûi can be used as
manipulated variables by another controller to regulate ad-
Figure 5: Comparison of the behavior of the water flow using both ditional variables.
systems. Finally, a brief demonstration of that the decoupled con-
trol system was a success is depicted in Table 1 where the be-
havior of the controlled variables was evaluated using differ-
Figure 5, compared to Figure 6 that shows the behavior of ent disturbances in the pH (acid flow) in the treatment area.
one of the manipulated variables (pressure). The response is It is easy to observe that as the conductivity is affected di-
similar; however, in the classical system the response tried to rectly with the changes in the pH, however the water flow
become unstable or took a long time period to settle. The is not affected for the pH, such that the conductivity may
classical system is very sensible to any change in the manipu- be controlled with the pH and the fresh water flow with the
lated variables; the use of the decoupled variables can reduce pressure.
the sensibility because it tries to become the MIMO system
in several SISO systems [5].
3. CONCLUSION
As an extension of our reasoning, if the set points r1
and r2 are constrained variables, a cascade controller struc- A decoupled control system was developed for an RO desali-
ture can be used to modify the control strategy in order to nation unit. Testing showed that the control system could
cancel the y1 and y2 feedback and to cancel the decoupler reasonably be used for evaluating the performance of this
when the constraints are not active. The decoupling can be unit. The performance of the control system was demon-
achieved at the controller input with a suitable modification strated through evaluation of some short- and long-term
to the decoupler transfer functions. Inversion of the con- control strategies. Some of the interesting results obtained
trollers at the inputs permits to reconstruct the manipulated from this evaluation are the following: (i) there is a delay
variables u1 and u2 which can then be used by two other con- time around 2 minutes in the unit that must be considered
trollers to achieve control objectives with less priority than in the design of any model or control system; (ii) in future
54 Journal of Automated Methods & Management in Chemistry

developments for the evaluation of the long-term control

strategies, benchmarks need to be able to assess settlers’ per-
formance.

REFERENCES
[1] H. H. Ettouney, H. T. El-Dessouky, R. S. Faibish, and P. J.
Gowin, “Evaluating the economics of desalination,” Chemi-
cal Engineering Progress, vol. 98, no. 12, pp. 32–40, December
2002.
[2] I. Alatiqi, A. ghabris, and S. Ebrahim, “Measurement and con-
trol in reverse osmosis desalination,” Desalination, vol. 75, pp.
119–140, 1989.
[3] A. Mindler and A. Epstein, “System identification and control
of reverse osmosis desalination,” Desalination, vol. 59, pp. 343–
379, 1986.
[4] W. Ray, “Some recent applications of distributed parameter
system theory—A survey,” Automatica, vol. 14, pp. 281–299,
1978.
[5] J. A. Mandler, “Modeling for control analysis and design
in complex industrial separation and liquefaction processes,”
Journal of Process Control, vol. 10, no. 2, pp. 167–175, 2000.
 Photoenergy
International Journal of
International Journal of Organic Chemistry International Journal of Advances in
Medicinal Chemistry
Hindawi Publishing Corporation
International
Hindawi Publishing Corporation Hindawi Publishing Corporation
Analytical Chemistry
Hindawi Publishing Corporation
Physical Chemistry
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

International Journal of

Carbohydrate Journal of
Chemistry
Hindawi Publishing Corporation
Quantum Chemistry
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Submit your manuscripts at

http://www.hindawi.com

Journal of
The Scientific Analytical Methods
World Journal
Hindawi Publishing Corporation
in Chemistry
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of International Journal of International Journal of Journal of Bioinorganic Chemistry

Spectroscopy
Hindawi Publishing Corporation
Inorganic Chemistry
Hindawi Publishing Corporation
Electrochemistry
Hindawi Publishing Corporation
Applied Chemistry
Hindawi Publishing Corporation
and Applications
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014

Journal of  Chromatography   Journal of Journal of International Journal of

Theoretical Chemistry
Hindawi Publishing Corporation
Research International
Hindawi Publishing Corporation
Catalysts
Hindawi Publishing Corporation
Chemistry
Hindawi Publishing Corporation
Spectroscopy
Hindawi Publishing Corporation
http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014 http://www.hindawi.com Volume 2014