OPTIMAL LOCATION OF

BOOSTER CHLORINATION STATIONS IN

WATER DISTRIBUTION NETWORKS







Tauseef Ahmed
Sumeet Yeotkar
Arpan Deshmukh





Contents
- Introduction
- Criteria for location selection
- Problem Formulation
- Multi-objective optimization for the location of booster
chlorination stations
- Methodology
- Case study 1
- Case study 2













Introduction
Restoring bacterial control within a distribution system often presents major difficulties for
water utilities. A commonly used indicator of water quality is the amount of residual chlorine in
a water distribution system

Chlorine booster stations are often utilized to maintain acceptable levels of residual chlorine
throughout the network. Many researchers have explored different methodologies for
optimally locating booster stations in the network for daily operations.

A common method to address perceived changes to water quality is to increase the amount of
the disinfectant in the water distribution system. Typically, disinfectants are applied at the
water treatment plant. Unfortunately, based on residence times associated with storage and
transport of water in a network, disinfectants applied at the treatment plant could take a long
time to neutralize a contamination event. Additionally, the reaction dynamics of disinfectants
make it difficult to maintain adequate residuals at critical locations without excessive residuals
elsewhere. Booster stations address these concerns through the reapplication of disinfectant at
strategic locations throughout a water distribution system.

Although booster disinfection is commonly practiced, a standardized procedure for the location
and operation of booster stations has not been adopted in the water utility community. Thus,
booster stations are often located near areas with low levels of disinfectant residual, and they
are operated with regard to the local goals of increased residual which often ignores the
system-level interactions.

Booster disinfection has been shown to minimize the total disinfectant required to maintain
adequate and uniform levels of residual when compared to adding disinfectant only at the
source of the distribution system (Boccelli et al. 1998). The location and operation of chlorine
booster stations is a problem which has been studied numerous times (Munavalli et al. 2003;
Ostfeld et al. 2006; Prasad et al. 2004; Propato et al. 2004a, 2004b; Tryby et al. 2002).

Criteria for location selection
Following criteria play an important role in proper selection of chlorine booster stations.
- Location should be such that a relatively large volume of water can be disinfected.
- The water to be treated travels in one direction.
- The chlorine residual in water has begun to decrease but has not totally dissipated.
- Chlorine can be applied uniformly in water.
- Location is acceptable by neighbours and easily accessible for chemical delivery vehicles
with room for chemical storage and feed equipment.
- Power is readily available.
- Communication systems are readily available for the SCADA system.
- Flow and/or residual pacing can be used.
- Safety concerns can be addressed.
- For a common inlet/outlet line, chlorine should be injected as the storage facility is
filling, although mixing of chlorine throughout the contents may be difficult.

Problem Formulation
The objective is to locate a given number of booster stations to support the activation of a
booster disinfection protocol that hyper-chlorinates water in the distribution system in order to
neutralize a contaminant that has been introduced into a system.
In this approach, several general assumptions must be made. First, it is assumed that water
quality sensors are used to support the automatic detection of contaminants in the distribution
system. This ensures that the booster stations can be activated quickly to minimize the impact
of the contamination incident that triggered them. Second, it is assumed that the booster
stations are being located to minimize the expected impact over an ensemble of contamination
incidents.
The sensors determine the time of detection, and the booster stations reflect where
chlorinated water enters the distribution system. In this approach, it is assumed that sensors
detect without error and that booster stations begin chlorinating immediately, or after a
suitable delay.
Finally, it is assumed that all booster stations are started simultaneously, and that they are on
throughout the duration of the contamination incident (i.e., until the end of the time-horizon
for modeling the contamination incident).

Three potential problems of conventional method of disinfection with chlorine in a WDS are
(1) high dosages of chlorine residuals near water sources;
(2) lack of chlorine residuals in the remote points in relation to water sources and
(3) formation of some potentially carcinogenic disinfection by-products (DBPs) at a level higher
than maximum contaminant level (MCL) regulated by environmental agencies.
The first two problems have almost been addressed directly by various researchers through
introducing optimal location and scheduling of booster disinfection within the system [5, 6].
Injection of disinfectant at optimally-located booster stations, in addition to the source, may
reduce the total disinfectant dose while keeping residuals within specified limits. This reduction
can be attributed to more uniform distribution of disinfectant residuals in space and time, and
less contact time with the water.

The objectives of such a booster facility location problem are
(1) to minimize the total disinfectant dose;
(2) to minimize the total cost of booster stations including the reduction of total number of
booster stations and operational cost;
(3) to maximize the volume of water supplied to consumers with chlorine residuals within
specified limits; and
(4) to maximize the volume of water supplied to consumers with DBP levels less than MCL.


Multi-objective optimization for the location of booster chlorination
stations

Methodology
In the present model, a multi-objective optimization problem is defined to determine optimal
location and scheduling of booster disinfection. The objective functions are evaluated utilizing
the principle of linear superposition for the dynamic water quality simulation. It is assumed that
the network is fully calibrated and follows first-order kinetics for disinfectant decay, and that
the hydraulic solution is periodic. Results are shown as trade-off curves between the defined
objectives. Here, four objectives of booster disinfection systems are subsequently included in
the optimization problem within two phases which will be described in the following sections.

First phase
In the first phase, a two-objective optimization problem is defined as the objectives are the
minimization of the total disinfectant dose and maximization of the volumetric demand (or
percentage of safe drinking water supplied) within specified residual limits. The decision
variables are the locations of these boosters and the injection rate.

The objective of minimizing the total cost of booster stations is replaced with a substitute
objective of minimizing the total number of booster stations. This objective function is also
considered in this phase by frequently solving the optimization problem each time with
different specified numbers of booster stations. Finally, trade-off curves between the total
disinfectant dose and the volumetric percentage of the water with residuals within a specified
range for different numbers of booster stations are obtained. Given a number of booster
stations n
b
, the mathematical formulation for the two-objective optimization problem is

= =
=
b k
n
i
n
k
k
i
M f
1 1
1
Minimize

100 Maximaze
1
1
2
=

+
= =
V
V
f
h m
n
m
n
j
m
j

where

s s A
=
otherwise 0
when
max min
j
m
j j h
m
j m
j
c c c t Q
V

where
f
1
=total disinfectant dose;
f
2
=percentage of the total volume of water supplied during a hydraulic cycle with residual
within specified limits;
M
i
k
=disinfectant mass (mg) added at booster station i in injection period k;
V=total volume of demand over a hydraulic cycle;
Q
j
m
=demand at node j in monitoring period m;
At
h
=hydraulic (monitoring) time step;
c
j
m
=disinfectant concentration (mg/L) at monitoring node j and during monitoring interval
m; c
j
min
and c
j
max
=lower and upper bounds on disinfectant concentrations (mg/L) at
monitoring nodes;
=start of monitoring time;
n
k
=number of time steps in dosage cycle;
n
h
=number of time steps in hydraulic cycle (number of monitoring time steps); and
n
m
=number of monitoring nodes in which residual chlorine concentrations are controlled.

The aforementioned optimization problem is subject to the following constraints:
1 2
C f >
k b
k
i
n k n i M ,..., 1 ; ,..., 1 ; 0 = = >
k
i
n
i
n
k
km
ij
m
j
x c
b k

= =
=
1 1
o
1 ,..., ; ,..., 1
max
+ = = s
h m j
m
j
n m n j c c

where
C
1
=specified value representing Pareto optimal front for a range of f
2
greater than C
1
;
k
i
x =multiplier of dosage rate at Booster i and during Injection Period k;
km
ij
o =composite response of concentration at Node j and Monitoring Time m due to dosage
rate at Booster i and during Injection Period k.


Second phase
The most frequently selected locations for installing booster stations are chosen for the second
phase in which another optimization problem is defined for booster disinfection systems. The
decision variables are the amount of disinfection dose for each booster station installed in the
chosen locations.

The objectives of this optimization problem are the maximization of the volumetric demand
within specified residual limits and the maximization of the volume of water supplied with
produced THM concentration less than the MCL.

Mathematical formulation of the objective functions in this optimization model is almost the
same as the one in the first phase. The only difference is the first objective function which is
100 Maximaze
1
1
1
=

+
= =
V
W
f
h m
n
m
n
j
m
j

(12)
where

s A
=
otherwise 0
when
max
j
m
j h
m
j m
j
THM THM t Q
W (13)
where
f
1
=percentage of the total volume of water supplied during a hydraulic cycle with produced
THM concentration lower than the MCL;
THM
j
m
=produced THM concentration at Monitoring Node j and during Monitoring Time m;
THM
j
max
=maximum contaminant level (MCL) of the THM concentration formed at Monitoring
Node j.

Another constraint related to the first objective function is also added to the set of previous
constraints in this model:
2 1
C f >
where C
2
=specified value representing Pareto optimal front for a range of f
1
greater than C
2
.

Case study 1
The case study used here is Mahalat WDS located in the central part of Iran. The WDS covers
approximately 46 km
2
, with a population of around 160,000. Model demands are
predominantly domestic with some commercial users. To reduce the high pressure head
induced by steep slope of the city, six pressure reduced vales (PRVs) are used to decrease
pressure heads to a fixed pre-specified values. The WDS is supplied by gravity from three wells
and two service tanks (reservoirs) around the city. The average water demand is 158.9 L/S. The
water is pumped into the system with a constant rate. The reservoirs store and balance the
fluctuations of water daily consumption.



Fig. 1. Mahalat WDS model with + indicating the potential location for installing booster station
1
2
3
4
5
6
7
8
9
10
11
12
13
14 15
16
17
18
19
20
160
21
Reservoir 1
Reservoir 2
Results and discussion
The multi-objective optimization problem of booster location and injection scheduling is
applied to Mahalat WDS model. Application of the model is carried out to assess

(1) the trade-off between the disinfectant dose and the percentage of safe drinking water
(SDW) within specified residual limits;
(2) the trade-off between the percentage of SDW within specified residual limits and the
number of booster stations for a specified amount of total disinfectant dose; and
(3) the trace of THM concentration within WDS nodes with respect to optimal booster locations
and scheduling.

Two existing stations for chlorine injection are located at the water treatment plants
(reservoirs). If all of the nodes of the network are to be considered as potential locations, the
network setup would require a very high computational time and a large memory to store
cumulative response coefficients. However, not all locations would be suitable due to
prohibitive costs, network hydraulics, and existing infrastructure. Therefore, 20 potential
locations including two existing locations and 18 potential boosters were assumed to be
available for installing booster station. These locations are shown in Fig. 1. Some of these
potential locations are located near the reservoirs, and the others are spread at critical points
throughout the network.

The value of C
1
is taken to be 75%. The hydraulics and booster injections are assumed to be
periodic with a period of 24 h. The water quality simulation duration is set to be 144 h and the
final 24-h results are used in the calculation of composite response coefficients. Optimal trade-
off analyses were carried out using flow proportional type boosters due to their performance
rather than constant mass type boosters. A comparison is also made among the solutions with
varying numbers of booster locations.

Phase#1
In the first phase, the multi-objective optimization model is applied to find the trade-off
between the disinfectant mass and the percentage of SDW with the number of booster stations
as a third parameter. For flow proportional boosters, constant concentrations added at the
booster nodes are decision variables.

The maximum value of these variables is equal to 4.0 mg/L as the upper limit on residual
concentrations. It was observed that for n
b
< 4, 99.9% SDW could not be achieved. These curves
become almost flat after 99% SDW for n
b
4. The optimal trade-off curves indicate that the
SDW increases significantly with a small increase in the total dosage rate up to 95%. Then, the
marginal improvement in SDW with the increase in the dosage rate diminishes. Furthermore,
the improvement of SDW versus dosage rate is neglected for n
b
> 7

When limiting the budget of total disinfectant dose, Fig. 2(b) can be applied which shows the
trade-off between the variation of SDW and the number of boosters for different amount of
total disinfectant dose. As it can be observed from this Fig., a significant increase in SDW can be
achieved for n
b
7.

It could be concluded from Fig. 2 that the optimal number of booster stations can be chosen
between four and seven, and the most efficiency can be achieved from seven optimal booster
stations. Although further considerations such as operational and budgetary limitations may
influence the final number chosen, seven booster stations seem to be the most efficient
number of booster stations from Figs. 2(a) and 2(b).




(a)

(b)
Fig. 2. (a) Pareto fronts of total dosage versus percentage of SDW for different values of n
b
; (2)
Trade-off between percentage of SDW and number of boosters for different amounts of total
dosage






Table 2. Relative frequency of optimal boosters for each Pareto front of specified number of
boosters
Boost
er no.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
N
u
m
b
e
r

o
f

b
o
o
s
t
e
r

s
t
a
t
i
o
n
s

(
n
b
)

3 1 1 0
0.0
7
0 0 0 0 0 0 0 0 0 0 0
0.1
1
0.3
1
0.5
1
0 0
4 1 1 0
0.8
1
0 0 0 0
0.1
9
0 0 0 0 0 0
0.3
2
0.2
8
0.2
8
0.0
8
0.0
4
5 1 1
0.0
4
0.6
6
0 0 0 0
0.3
4
0.0
5
0 0
0.4
3
0
0.2
2
0.0
3
0.2
3
0.4
3
0
0.5
7
6 1 1
0.0
4
0.0
3
0.1
2
0.0
2
0
0.1
7
0.9
7
0.0
3
0.8
3
0
0.2
7
0
0.0
5
0.0
4
0.5
5
0.3
8
0.0
4
0.4
6
7 1 1
0.3
9
0.1
6
0
0.0
2
0
0.4
5
0.8
5
0.1
5
0.8
3
0.1
3
0.1
5
0
0.0
4
0.5
3
0.3
0.1
3
0
0.8
7
1
0
1 1 0.5
0.4
9
0.6
7
0.1
3
0.0
6
0.5
2
0.9
5
0.5
7
1.0
3
0.0
4
0.0
3
0.2
4
0.6
9
0.0
2
0.3
6
0.5
9
0.1
1.0
1
1
5
1 1
0.7
4
1.1
5
1.4
2
0.2
1
0.3
7
0.7
7
1.0
5
1.0
8
1.1
3
0.4
8
0.2
9
0.4
4
0.7
0.2
7
0.5
8
0.1
1
1.2
5
0.9
6


Location of booster stations:
A further analysis is made on the solutions obtained in Fig. 2 in order to choose the appropriate
locations of booster stations. Relative frequencies of 20 potential nodes for installing booster
station were analyzed in Table 2 for each Pareto front with specified number of boosters.

The location of booster numbers of 1 and 2 are water treatment plants (reservoirs) which are
the existing nodes for all solutions with the relative frequency of 1. As discussed in the previous
section, the number of seven solutions is favorite. Therefore, relative frequency of seven
boosters is considered in more details. The most frequently selected nodes for seven boosters
are Nodes 1, 2, 8, 9, 11, 16 and 20. As it can be observed in Table 2, these locations have
dominantly been selected in the Pareto front compared to other potential locations in the WDS.







Case Study 2

Optimization for the location of chlorine booster stations in Al-Khobar water distribution
system, located on the eastern part of the Kingdom of Saudi Arabia, to determine the optimum
schedule of chlorine injection criteria for disinfectants was adopted, according to which the
minimum residual chlorine should not be less than 0.2 mg/l and greater than 4.0 mg/l.

Currently, in Al-Khobar water distribution system, there are three existing locations for the
injection of chlorine. For the purpose of selecting booster disinfection locations, thirty nodes in
the network are considered as potential booster locations. These nodes are selected based on
water quality simulation results. Some of the nodes are near the sources of water. Others are
selected based on the spatial behavior of these nodes. Some are selected in the areas having
the problem of low residual chlorine. The monitoring constraints should be satisfied at all the
consumer nodes.

There are some nodes that are neglected for monitoring constraints. These nodes are either the
groundwater reservoir or the nodes having zero demands. So, there is no need to fulfill the
condition of monitoring constraints at these locations. The model setup is the same as for the
existing design. The duration of water quality simulations is taken as 360 hours (15 days).

Results
The optimal results (location/scheduling) are obtained when the number of injection periods of
chlorine is 1 (1-24 h). The duration of the injection period is 24 hours. A constant concentration
of chlorine will be injected continuously at all the optimal booster stations for 24 hours
duration. The monitoring constraints for the residual chlorine are bounded between l = 0.2 mg/l
and u = 4.0 mg/l.

Table 1 and Figure 1 summarize the results of the study. They show the effect of the number of
booster stations on the total mass dosage rate. Figure 1 shows that the mass dosage rate
decreases as the number of booster stations increases. Then, at certain number of booster
stations, the amount of decrease in the mass dosage rate is not significant. The results indicate
that there is a considerable decrease in dosage rate up to 10 booster stations. After this, the
decrease in the dosage rate is insignificant. The mass dosage rate for the existing design is
89.272 kg/d.

The table shows the percentage decrease in the total mass dosage rate for the optimal results
when compared to the existing design. It can be observed from the table that increasing the
number of booster stations increases the percentage reduction. It shows that when the number
of booster stations is 6, the percentage reduction in the total mass dosage rate is 25%. For 7, 8,
9 and 10-station designs, the percentage reductions are 31%, 40%, 46% and 51%, respectively.
In greater than 10-station design, there is insignificant increase in the percentage reduction.




Conclusion

The results illustrate that a booster disinfection strategy can reduce considerably the amount of
chlorine required to maintain acceptable chlorine residual throughout the distribution system.
The important conclusions derived from this study are given below:

The optimization results indicated that the booster disinfection efficiency can be maximized
by increasing the number of booster stations and the dosage duration. It can be concluded that
dosage reduction is a function of the number of booster stations and the number of scheduling
intervals.

The results also showed that the considerable reduction in dosage rate, by increasing the
number of booster stations, is up to a certain limit; after this, the reduction is insignificant.