Analysis of Water Quality Characteristics in Distribution Networks

International Journal of Civil, Mechanical and Energy Science, 8(1)
Jan-Feb, 2022
Available: https://aipublications.com/ijcmes/
Peer-Reviewed Journal
ISSN: 2455-5304
https://dx.doi.org/10.22161/ijcmes.81.1 1
Analysis of Water Quality Characteristics in Distribution
Networks
Raber Hamad1
, Mehmet İshak Yüce2
Department of Civil Engineering, Gaziantep University, Gaziantep, Turkey
1
Email: rabar_hussen@hotmail.com
2
Email : yuce@gantep.edu.tr
Received: 03 Jan 2022; Received in revised form: 11 Feb 2021; Accepted: 20 Feb 2022; Available online: 03 Mar 2022
©2022 The Author(s). Published by Infogain Publications. This is an open access article under the CC BY license
(https://creativecommons.org/licenses/by/4.0/)
Abstract— In this study, a model was developed by Epanet2.0 software to analyze water quality for
parameters of hydraulic and water quality model (chlorine concentration and water age model) for a
segment of Erbil city WDS by using observed and documented data. Controlling free residual chlorine
properly is important to ensure meeting regulatory requirements and satisfying customer needs. For the
calibration process and collecting field data digital pressure loggers for recording pressure in a WDS was
installed. For discharge measurements, ultrasonic flow meters were used. To assure the reliability of the
model a calibration process was carried out for extended period analysis and several alternatives had been
studied as a solution to overcome negative pressure zones by the calculated Hazen William C-factor. This
kind of study can be used to predict so many infrastructure projects.
Keywords— Water Quality, Water Distribution system, Calibration, Chlorine Concentration, Water age.
I. INTRODUCTION
drinking water utilities face the challenge of supplying
drinking water to their users despite the many factors that
can result in the retrogression of water quality before it is
delivered to the user’s tap. Frequently, raw water is
derived from groundwater sources that may be subject to
naturally occurring or accidental contamination (ILSI
1999, Gullick et al. 2003). The first Proposition of using
mathematical models to analyze water distribution systems
was in 1930 s by Hardy Cross (1936). water quality
models have reached operational status, but research and
development continue to further the understanding of the
processes taking place in the distribution system and to
translate this understanding into usable tools (i.e.
Epanet2.0 program).In studying water distribution system
the most important thing which has apriority to determine
the type of model that is most applicable because so many
factors have an impact on the degree of temporal (
overtime ) for example, steady-state modeling represents
external forces as constant in time (static) and determines
solutions that would occur if the system were allowed to
reach equilibrium (Wood 1980a). In dynamic modeling,
demands and supplies are allowed to vary with time and
the resulting temporal solution is determined (Clark et al.
1988a, Clark et al. 1988b). After the hard efforts of
researchers to computerize both water quality and
Hydraulic models finally in 1993 EPANET2.0 was
initially developed in the united states by the
Environmental protection agency as a distribution system
hydraulic-water quality model to support research efforts
at EPA (Rossman et al., 1994). The development of the
EPANET software has also convinced the requirement for
an acquisitive public-sector model and has served as the
hydraulic and water quality “engine” for many scientific
and commercial water quality models. Water age is
considered as a time to convey particles of water or time to
travel water particles from the source (after the treatment
process) to the consumer’s tap. There are two reasons
behind studying water age, firstly to ensure the contact
time with chlorine secondly, to avoid quality deterioration
with time. Water age varies according to the fluctuation of
demand in a model. To estimate water age in a water
distribution system a mathematical model that represents
the hydraulic behavior of the movement of water has been
used. (Clark and Grayman, 1998). In the mid-1980s a
Single period analysis travel time models were first

Hamad Analysis of Water Quality Characteristics in Distribution Networks
www.aipublications.com Page | 2
introduced (Males et al., 1985). These models were
subsequently extended to dynamic representations that
determined varying water age throughout the distribution
system (Grayman et al., 1988). To predict concentrations
of chlorine, DBPs, and other constituents in a distribution
system Both models can be used as a complement of each
other it means Water quality models can be used as
conjunctive models with hydraulic models (Vasconcelos et
al., 1996). Relationships proposed by (Rossman et al) can
be used to estimate the effect of the interaction of the
flowing water with the pipe wall. (1994). In the early
1990s a simplified model of water age in tanks and
reservoirs was developed (Grayman and Clark,
1993). Water quality can be deteriorating during
transmission and distribution after treating in a water
treatment plant. To minimize the potential of this
deterioration should add chlorination for the finished
water. The concentration of chlorine can vary
increase/decrease according to the fluctuation of water
demand while the water flows through the pipes. Treated
water in the water treatment plant is disinfected then it
enters the transmission system (Clark and Coyle,
1990). The expectation of bacterial contamination and
growth is high in the transmission and distribution system.
Therefore, the advantage of remaining a detectable
disinfectant residual in the medium according to standards
is to minimize the potential of waterborne diseases and
biofilm growth. (Clark et al). (1993) stated that the
chlorine residual by the effect of demand fluctuation can
virtually disappear at various times during the day. Maul et
al. (1985a) proved that lower levels of chlorine residuals
and extended retention time of the water in the network
causes the highest bacterial concentrations are referred.
After the water is treated in the water treatment plant then
transfers to the water networks temporal and spatial
consumption can occur. Chemical reactions of the chlorine
with water constituents and with both the biofilm and
nodes(tubercles) formed on the pipe wall, as well as
reaction with the pipe wall material itself causes This
temporal and spatial consumption of chlorine through the
water journey in water networks (Wable et al., 1991;
Zhang et al., 1992; Kie´ne´ et al., 1998). chlorine is
consumed in the water itself and through contact with the
interior wall of the pipe.
II. CASE STUDY
Case study lies in Erbil city; Erbil city is the fourth largest
governorate of Iraq. The latitude of Erbil is 36.206293,
and the longitude is 44.008870. Erbil is a city is located at
Iraq with the global coordinates of 36° 12' 22.6548'' N and
44° 0' 31.932'' E. In this study a segment of Erbil city
water distribution system was used as a case study network
which is consists of 700 mm ductile iron pipe as a major
pipeline and the branches (i.e. connections) that conveying
water to the 25 % of Erbil city consumers considered as a
main pipe lines which is made up of (HDPE and ductile
pipes) with different diameters. This segment of Erbil city
network is being supplied by a gravity concrete reservoir
with capacity (20,000) m3 of water that water had already
been treated in a treatment plant then pumped to the
reservoir .it means the system is a combination of gravity
system and pumping system. The modeled portion of an
Erbil city water network has an approximate length (10,
680) m. The reservoir Elevation of a concrete reservoir
tank is about 498 m above msl and the final junction is 430
m higher than msl.. Figure below elaborates the case study
WDN sources, connections diameter, major pipeline and
main pipelines.
III. MODELLING
The first step to make a hydraulic model for water
distribution network is to collect all observed data from all
necessary components in different places and components
of the network, then introduce assembled observed data

and documented data from water authorities into Epanet2.0
program so as to build the hydraulic model, after
completed the model in a software then began the
characterization phase of the nodes and links. In the fixed
piezo metric level nodes was introduced the value of its
pressure. To the junction nodes were calculated and
entered in the demand, elevation and consumption's
patterns. For the demand were identified the branches
within the DMIs and manually calculated the consumption
using the ultrasonic flow meters for measuring out let
major pipeline and branches. Hazen-Williams formula for
the calculation of unit load losses with a roughness
coefficient k was found in a field. the model can be
developing by calibration process. Generally, several
parameters have been taken into account in this study such
as (finding actual Hazen William C- factor Value, make
calibration process for whole network and make
alternatives as a solution). for finding C factor some field
information should be known such as Elevation of a
junction, pressure data loggers for recording pressure,
diameter of a pipe line segment and installing ultrasonic
flow meters to measure flow rate.in order for the results to
be more precise, this process should be repeated in many
places then can take an average value of C- Factor. After
the hydraulic model has been created and C- factor values
were updated in a computer model, there was a problem
occurred in the model which is appearing negative
pressure zones in some DMIs in a case study area specially
during peak hours. A set of alternatives has been suggested
to overcome the effect of negative pressure there are three
alternatives has been modeled:
1. Adding Elevated water tank at the beginning of
negative pressure zones: installing water tank
before negative pressure zones which is works with
such a system that the Elevated tank is filled during
night (no peak hour times) and it will empty during
peak hour to compensate the shortage of water.
Adding an elevated tank before the area of negative
pressure zones so that when there is water the tank
will be filled then the stored water can be used
during water shortages (i.e. peak hour) to overcome
negative pressure. That’s why this alternative was
chosen because it’s an easy way to execute, it has a
low cost compared to the other alternatives and it
considers as a reliable solution to improving the
case study water distribution system
2. By adding a parallel pipe: Because of the water
Distribution network is exist so there is no way to
increase the diameter of transition lines, therefore,
to deliver excessive water in a reservoir it is
preferred to add another (700 mm) pipe which is
20% of the total length of case study network. The
second alternative that is selected among
alternatives is putting a pipe that has the same
diameter as existing pipeline and is parallel to it as
shown in a figure. The philosophy behind this kind
of model is to increase the suppling discharge,
because of the existing pipe diameter not sufficient
to supply demand discharge in peak hours specially
summer times and there is enough treated water in
reservoir so it needs to increase the pipeline
diameter by replacing the existing pipeline not
economic and not reliable
3. By adding the pump station at the outlet of the
reservoir: installation of pump station needs
maintenance, electricity and for this study the unit
headless increases but it overcomes the negative
pressure zones in the system. This alternative is
developed by to set a pump station after water
storage. The philosophy behind this alternative is to
increase energy to supply sufficient water to study
area water distribution system. The main advantage
of this alternative is that water discharge is much
higher and ensuring that water is accessible to all
users.
Subsequent to the proper calibration of a hydraulic model,
another calibration of parameters in a water quality model
may be required. Such as chlorine concentration and water
age. both the field and laboratory experiments were used to
calibrate water quality parameters or in general, they
contribute to the calibration process in water quality.
laboratory tests can be used to find the bulk chlorine decay
coefficient (kb) however, the chlorine residual was
obtained from the field measurements. For finding the bulk
chlorine coefficient was obtained from three samples in
three places along the major pipeline. For calculating
chlorine decay calibration of the Water Distribution
networks many samples were taken in four different for
over 3 days with a splinter but in regular time intervals.
For introducing the collected field data to the model, the
data should have converted to text file format then the
model can calibrate the model by a process known as trial-
and-error which make up of comparison of both the
simulated and the observed chlorine residual data to adjust
the coefficient of wall decay (kw) in the software. Water
age is the most important parameter that has a great
influence on the determination of chlorine bulk
coefficients because it depends upon how much time is
available in the system for bulk reactions and how long the
treated water that is transported will remain in the pipe.
The longer it stays, the more it affects the pipe wall and
contact with it to take place the reactions. Also, there is no
such facility to measure water age in WDS only it can be
simulated. Generally, Epanet2.0 simulates the water age

for the whole model for an extended period of 72 hours or
more.
IV. RESULT AND DISSCUSSION
Table1: computed pressure results of all alternatives
Pressure in m
Node ID
pump
station
Parallel
pipes
Normal
condition
Add
tank
J2indust 68.94 29.34 26.53 28.26
J3 63.53 24.02 22.12 22.97
J4Gasha 58.54 30.49 24.08 25.85
J5 39.61 19.51 5.15 18.6
J6bajgr1 56.12 32.92 23.49 26.31
J7 57.1 34.62 24.47 28.19
J8bajgr2 55.08 34.89 23.63 27.09
J9 57.92 38.01 26.47 30.28
J10tawfer 51.98 36.89 22.58 27.11
J11 46.38 31.29 16.99 21.51
J12havala
n
52.55 30.88 23.73 28.57
J13 50.2 30.84 21.38 29.09
J14hasaro
k
52.18 31.08 25.18 29.75
J15 34.68 15.13 7.68 14.17
J16hanaci
ty
52.9 32.1 25.89 30.84
J17 31.66 16.39 4.65 16.49
J18qalat 51.86 32.73 24.85 31.89
J19 25.2 12.94 -1.8 14.53
J20minara 50.06 31.36 23.05 31.62
J21 45.15 28.1 18.15 27.76
J22 45.82 28.13 18.82 27.64
J23 41.82 24.18 14.81 23.7
J24kamar
ani
45.79 28.11 18.78 27.61
J25 44.2 28.43 18.05 27.6
J26badaw
a1
43.29 26.2 16.29 25.86
J27 26.44 14.08 -0.56 14.89
J28badaw
a2
43.25 26.18 16.25 25.83
J29 24.42 19.02 -2.58 16.06
J30Aso 44.61 27.69 17.61 27.38
J31 49.19 33.08 22.19 32.97
J32farmn
baran
45.6 28.68 18.6 28.37
J33 39.08 22.52 12.08 22.3
J34zanko
99
46.37 29.46 19.38 29.15
J35 42.53 28.52 15.54 28.92
J36 52.14 35.33 25.14 35.04
J37 44.37 28.37 17.37 28.28
J38 55.09 38.3 28.11 38.01
J39 56.62 40.02 29.63 39.78
J40 59.03 42.25 32.05 41.97
J41 61.03 44.41 32.05 44.17
J42 61.01 44.24 32.44 43.96
J43 61.67 45.18 32.7 44.97
J44 69.63 52.86 33.66 52.58
After developing a model for both hydraulic and water
quality for three alternatives as a solution to compensate
water for negative pressure zones during periods of low
pressure or critical times. Epanet2.0 can display analysis
results for a network for each alternative. To make a
comparison of all three alternative analysis results for a
network includes display the difference between these
three alternative analysis results for both node and pipe
values as discussed in detail in this study such as node
pressure, hydraulic total head, water quality parameters
after calibration, flow, velocity, and hydraulic head loss
while the demand, demand pattern observed chlorine
concentration is fixed in analysis for all alternatives. The
analysis results can be used to compare three analyses
according to the following criteria.
1. Velocity: the existing system hasn’t an economic
velocity, this term is true for the third alternative
and sometimes to the second alternative but the
first alternative has an economic velocity.
2. Pressure: The third alternative have a higher
pressure compared with the rest alternatives.
However, the second alternative has the lowest.
3. Flow: The third alternative has the highest
discharge, and the third alternative is alternative
two.

4. Unit head loss: The third alternative has the
highest unit head loss however the other
alternatives have a lower unit head loss.
5. Hydraulic calibration: although all alternatives
have matchmaking with the existing system it
means they are well fitted to the working system
but the first and second alternatives have a better
conforming with the system.
6. Water quality calibration: first and second
alternatives have more response to the existing
system.
7. Water age: The first alternative has the highest
water age while the third alternative has the
lowest due to effect of pump station.
8. Chlorine concentration calibration: both first and
second alternative calibration were more
compatible with the existing system than the
Third alternative.
9. cost: The third alternative needs the highest cost
while the first alternative needs a lower cost. The
third alternative requires the cost of supply
boosters, construction of booster station,
monitoring, power supply, and maintenance while
alternative one has an only cost of construction.
10. Long-term solution: third alternative can work for
longer term than the rest alternatives even number
of populations in case study area increased if
water treatment plant has a capacity to provide
sufficient water.
11. Reliability: the similarity of alternatives is that all
alternatives are reliable but the first alternative
and second alternative have more reliability
compared with the third alternative.
12. Constancy: First alternative and second
alternative can work under different conditions.
13. Confidence: during the system works, the third
alternative has more confidence for water to reach
every point.
14. Inviolability: Third Alternative can’t work if one
of its parts fails such as the sudden failure of
pumps or power supply.
15. Communicability: both the first and second
alternatives are easy to treat and system problems
are understandable.
16. Simplicity: alternative one considers as the easiest
to carry out, while alternative three is complex
needs cost and experience.
17. Compatibility: all alternatives are conforming
with existing norms and procedures of the case
study area.
18. Reversibility: alternative three if fails the model
cannot return to the prior state, however, there are
no such criteria in alternatives one and two.
19. Wholesomeness: Alternative three has a
capability of success in different future states and
alternative two seems to be a solution for the long
term if alternative three and alternative two
integrated may have more success however
alternative one seems to be a solution for a shorter
period especially if case study area had more
population density in near future.
V. CONCLUSION
1. A conclusion Calibration of the hydraulic model
explained a reduced C-factor for 700mm from
140 according to standards to 92 which is very
low compared with normal conditions. This could
have two reasons may because of the high age of
installed pipes in the network and the velocity in
the pipe, not economic velocity causes high unit
head loss.
2. Major pipeline not sufficient for future demands.
3. From the obtained results of all alternatives, the
pressure is quite enough to serve all nodes.
4. the First alternative is applicable when the
solution is required for short terms, the second
alternative is applicable when the solution is for
long terms, the third alternative is applicable
when the solution requirement is for long terms in
addition to availability experience and higher
cost.
ACKNOWLEDGEMENTS
I would like to express my special thanks of gratitude to
my supervisor Prof. Dr. Mehmet İshak YÜCE for his
constant support to complete my thesis.
REFERENCES
[1] American Water Works Association (1989). “Distribution
Network Analysis for Water Utilities.” AWWA Manual M-
32, Denver, Colorado
[2] American Water Works Association (1996). “Ductile Iron
Pipe and Fitting.” AWWA Manual M-41, Denver, Colorado.
[3] American Water Works Association (1998). “Distribution
System Requirements for Fire Protection.” AWWA Manual
M-31, Denver, Colorado.
[4] Basford, C., and Sevier, C. (1995). “Automating the
Maintenance of Hydraulic Network Model Demand
Database Utilizing GIS and Customer Billing Records.”

Proceedings of the AWWA Computer Conference,
American Water Works Association, Norfolk, Virginia.
[5] Bowen, P. T., Harp, J., Baxter, J., and Shull, R. (1993).
Residential Water Use Patterns. AWWARF, Denver,
Colorado.
[6] Brainard, B. (1994). “Using Electronic Rate of Flow
Recorders.” Proceedings of the AWWA Distribution System
Symposium, American Water Works Association, Omaha,
Nebraska.
[7] Buchberger, S. G., and Wells, G. J. (1996). “Intensity,
Duration, and Frequency of Residential Water Demands.”
Journal of Water Resources Planning and Management,
ASCE, 122(1), 11.
[8] Buchberger, S. G., and Wu, L. (1995). “A Model for
Instantaneous Residential Water Demands.” Journal of
Hydraulic Engineering, ASCE, 121(3), 232.
[9] Cesario, L. 1995. Modeling, Analysis, and Design of Water
Distribution Systems. Denver, Colo.: AWWA.
[10]Clark, R.M. (2012). Modeling Water Quality in Drinking
Water Systems Second Edition. AWWA
[11]Clark, R.M., and D. L. Tippen. (1990). “Water Supply” in
Standard Handbook of Environmental Engineering. Edited
by Robert A. Corbitt, NY, McGraw-Hill, Chapter (5), 173-
220.
[12]Grayman, W. M., Clark, R. M., and Males, R. M. (1988).
“Modeling Distribution System Water Quality: Dynamic
Approach.” Journal of Water Resources Planning and
Management, ASCE, 114(3).
[13]Gullick, R.W., M.W. LeChevallier, R.C. Svindland, and M.
Friedman. 2004. Occurrence of Transient Low and Negative
Pressures in Distribution Systems. Journal AWWA,
96(11):52–66.
[14]Haestad Methods,Thomas M. Walski,Donald V. Chase,
Dragan A. Savic, Walter Grayman, Stephen Beckwith,
Edmundo Koelle.(2003). ADVANCED WATER
DISTRIBUTION MODELING AND MANAGEMENT.
First Edition, Second Printing.
[15]Males, R. M., Clark, R. M., Wehrman, P. J., and Gates, W.
E. (1985). “An Algorithm for Mixing Problems in Water
Systems.” Journal of Hydraulic Engineering, ASCE, 111(2).
[16]National Research Council (NRC). 2006. Drinking Water
Distribution Systems: Assessing and Reducing Risks.
Washington, D.C.: National Academies Press, 15–46.
[17]Panguluri, S., W.M. Grayman, and R.M. Clark. 2005.
Distribution System Water Quality Report: A Guide to the
Assessment and Management of Drinking Water Quality in
Distribution Systems. Cincinnati, Ohio: USEPA Office of
Research and Development.
[18]Santiago Arnalich, Maxim Fortin, Amelia Jimenez (2011).
Epanet and development (How to calculate water network
by computer), Arnalich, Printed by Amazon.

Analysis of Water Quality Characteristics in Distribution Networks

More Related Content

What's hot (18)

Similar to Analysis of Water Quality Characteristics in Distribution Networks (20)

More from AI Publications (20)

Recently uploaded (20)

Analysis of Water Quality Characteristics in Distribution Networks