Water Supply and Distribution Sample Paper

Design of a water supply distribution network using EPANET 2.0: A case

study of Barangay Santa Rosa, Mulanay, Quezon

Abstract: In rural areas where there is no sufficient water supply system, a demand-based approach that balances
what consumers want and are willing to pay for is vital in determining design and service levels. The increasing
demand for water supply and rapid urbanization are some of the challenges that require water distribution systems
(WDS) to be increasingly efficient, reliable, and resilient. Based on the provincial statistics, almost 83% of the
households in the municipality of Mulanay are using non-potable or doubtful drinking water. Based on the annual
growth rate, the projected population for the year 2030 was found to be 4,325 with 865 households. This population
projection was used to compute for the value of total water demand which is 763,235 liters. Contour maps show that
Mulanay province including Barangay Santa Rosa is complicated in terms of topography. The water distribution
system of the case study was analyzed to show and represent the capabilities of the proposed system using EPANET
2.0 software. It is therefore recommended to further explore other areas of optimization to further reduce water losses
and inefficiency in water distribution system. Actual rates of water demands may also be considered to produce a
more reliable analysis. In the design of water distribution system, complexities exist due to hydraulic considerations,
water quality, topography, cost efficiency, and system resiliency. The use of EPANET 2.0 software was found to be
efficient in the analysis especially in the design of rural water distribution systems.

Keywords: Water distribution network, EPANET

1 GENERAL INFORMATION consumers want and are willing to pay for was adopted
in determining design and service levels. This approach
1.1 Introduction departs from the traditional mode of estimating water
demand based on purely engineering considerations,
Water is undoubtedly one of the most essential and is more attuned to preferences of consumers and to
commodities every living creature requires for their ability and willingness to pay. The traditional
survival. Its purpose in commercial, farming or mode often lead to costly, over-designed systems that
irrigation, navigation, recreation and domestic were unsustainable. For example, consumers in small
purposes makes it a valuable commodity with no other towns have daily per capita consumption of 80 to 100
alternative. Despite its abundance, water is unequally liters, compared to standard engineering design
distributed on the earth and its availability at any place parameters of 120 to 150 lpcd. The demand-based
varies greatly with time and size. The total supplies of approach lowers the cost of investment and translates
fresh water on earth far exceed human demand. About into more affordable tariffs and sustainable operations.
2.8% of the total amount of water on earth is fresh
water of which 0.6% is groundwater. The rest is Small water utilities are primarily concerned with
available in the seas and oceans while a negligible water for potable use, which is basically for the home.
amount as soil moisture. Out of the 2.2%, 2.15% is Aside from drinking, other domestic uses include
found in the form of ice sheets and glaciers while the washing, bathing, cooking and cleaning. Other
remaining 0.05% is available as surface runoff household needs might include tending and watering of
(Mustafa and Yusuf, 2012). home gardens and the upkeep of domestic animals.
Basic household water requirements have been
Starting in the 1970s, the Philippine Government estimated to average around 40 liters per person per
introduced certain developmental practices and day. The standard used for drinking water supplied by
concepts to strengthen the water sector and expand its Level II and Level III utilities is potability, or water that
coverage of the population. These led to the improved can be consumed directly by drinking without risk of
overall sustainability of water utilities, the immediate or long-term harmful effects.
establishment of more small water systems, the
institutionalization of support for all water service Meanwhile, the increasing demand for water supply
levels, and the increase in commitments of and rapid urbanization are some of the challenges that
development funds to maintain the positive impetus require water distribution systems (WDS) to be
that had been created (RWSVol1). increasingly efficient and resilient (Guppy et. al, 2018).
WDSs are systems which usually contains interrelating
In rural areas where no water supply system is in place, modules such as pumps, pipes, valves, pumps,
a demand-based approach that considers what reservoirs and tanks (Mala-Jetmarova et. al, 2017). The

1
Water Supply and Distribution-Title 21 April 2020

reliability of WDS is dependent on system 2 METHODOLOGY

configuration, design, pressure and flow through the
systems elements. Currently, Water Distribution 2.1 The Study Area
Systems are still being operated using simplifying Santa Rosa is a barangay in the municipality of
assumptions and rigid-static analysis. As a result, the Mulanay, in the province of Quezon. Its population as
systems are not tested under various conditions, which determined by the 2015 Census was 3,697. This
can be encountered during its operational life, and represented 6.96% of the total population of Mulanay.
under different water demand scenarios (Savić et. Al, The population of Santa Rosa grew from 2,002 in 1990
2018). This usually generates alarming failures in to 3,697 in 2015, an increase of 1,695 people. The latest
meeting the actual demand during the operation of the census figures in 2015 denote a positive growth rate of
system in real life situations. 0.90%, or an increase of 170 people, from the previous
population of 3,527 in 2010. Based on
Continuous efforts should be made to drastically
MWSSU/NWRB 2014, out of 11,384 households in the
reduce waste/abuse, contamination (Fayomi et.al,
municipality 19 are using shallow well, 10 are using
2017, & Longe et. al, 2010) and systems operational
deep well, and 17 are using improved spring.
cost through optimization, analysis and design. Several
Accordingly, Brgy. Sta. Rosa is not included in the list
popular optimization and assessment techniques can be
of barangays with households utilizing Level 1 water
used to analyze a water distribution system. These
supply system.
methods may include several stochastic and
metaheuristic methods (Longe et. al, 2010). Figure 1. Map of Brgy. Santa Rosa, Mulanay, Quezon
Furthermore, inadequacy of water distribution systems
to satisfy demand and pressure usually is mainly caused
by rapid urbanization. Another problem is pipeline
leakage and disruption due to bursts. High pressure and
flow frequently often lead to burst, water loss,
reductions in water quality, high maintenance costs,
Poor systems design as a result of systems being
designed as a continuous system that relies on the
steady supply assumption (Ramulongo et.al, 2017).
Poor System design and configuration make WDSs
vulnerable to errors due to incorrect assumptions,
insufficient statistics and input calculation errors. The
associated high levels of contamination are a serious
problem caused by the intermittent supplies, bursts and
leakages. This is done in systems where supply
interruption periods are prolonged due to insignificant
or null pressures on a system and frequent system
disruption (Samani et. al, 2006, Jung et. al, 2007, &
Lavaei et. al, 2012)

In general, one of the major challenges hampering the

establishment of an efficient water supply system is the
incapacity to provide sufficient amount of water to
meet the requirement of the ever growing population.
A rapid increase in population and technological
advancement all over the world compels a
corresponding rise in the demand for potable water. As
such, water policy makers rely on population
projections to assess future demands for this much
needed commodity. These projections are necessitated 2.2 WDS Challenges in Mulanay, Quezon
by the population change with time as a result of births
Based on the CBMS survey in 2014, almost 83% of the
and the arrival of migrants which add individuals to a
households in the municipality are using non-potable
population, whereas deaths and the departure of
or doubtful drinking water. Water quality test results
migrants subtract individuals from a population
finds the presence of Coliforms and Bacteria. A study
(Kaneda and Bremner, 2014).
of the DOST with their Program Community

2
Water Supply and Distribution Sample Paper

Empowerment thru Science and Technology reveals 2.3 Water Demand Estimation
that all barangay where there is an overwhelming The water distribution system in consideration is
majority of households using untreated Level I, II, and shown in Figure 3. Water demand fluctuates between
III water systems like dug wells, unprotected springs being below average daily demand in the early morning
and community water systems. About one-half of the hours (before sunrise) and above average daily demand
household uses an open dug well and shared tubed/pipe during mid-day hours. Based on the assumptions,
deep well with no cover while around one–fourth of the Maximum daily demand (MDD) which is defined as
households uses bottled or mineral water for their water the demand on the day of the year that uses the most
needs. 9.64% of the population has shared faucets for volume of water and the Maximum hourly demand
their water connection. Very few use a deep well of (MHD) which is defined as the demand during the hour
their own with no cover, a faucet of their own use, and that uses the most volume of water were computed.
a shared rain collector. The rest of the population get
water from deep wells, developed and undeveloped A maximum of 15% loss accounting to non-revenue
springs, Jetmatic, rain collector (own use), and from water (NRW) resulting to waste, pilferages, leaks and
rivers, streams, and lakes.
other unaccounted for losses was also assumed.
Level I water supply system includes shallow well,
deep well, and protected spring. Based on Table 1. Design Parameters used based on 10-year
MWSSU/NWRB 2014, out of 11,384 households in the projection
municipality 19 are using shallow well, 10 are using
deep well, and 17 are using improved spring. Brgy. Sta. Parameter Values
Rosa is not included in the list of barangays with
households utilizing Level 1 water supply system. Cluster 1 2

Meanwhile, as of 2014, there are 1,448 consumers No. of Houses 600 265
(both commercial and domestic) are using the level III
or the local waterworks system in the municipality. Persons/DU 5 5
Barangays of Poblacion, Buenavista, Butanyog,
Total Persons 3000 1325
Average Use/Person
150 150
(lpcd)
Daily Demand
529,412 233,824
(liters)
ADD (lps) 8 4

MDD (lps) 15 6

MHD (lps) 25 11

Two water–use patterns were also assumed for the

analysis. First, ADD Pattern 1 within a 24-hour periods
are characterized by demands that are 25% to 40%
(take 0.3 or 30%) of the ADD during the hours between
mid-night and 6 AM and 150% to 300% of the ADD
during morning or evening periods but the ADD must
turn out to be 100%. Secondly, For MDD Pattern 1,
Water –use patterns within a 24 hour periods are
characterized by demands that are 25% to 40% (take
Figure 2. Access to Safe Water in Mulanay, Quezon 0.3 or 30%) of the ADD during the hours between mid-
night and 6 AM and 150% to 325% of the ADD during
Latangan, and Santa Rosa are the barangays who are
morning or evening periods but the ADD must turn out
able to use the level III water system run by the LGU.
to be 180%.

3
Water Supply and Distribution-Title 21 April 2020

3 RESULTS AND DISCUSSION

Based on the annual growth rate, the projected

population for the year 2030 was found to be 4,325 with
865 households. This population projection was used to
compute for the value of total water demand which is
763,235 liters.

The EPANET software employs the use of the

numerous head loss formulae including Darcy-
Figure 3. Average Daily Demand Pattern 1
Weisbach and Hazen Williams. The former was chosen
for this analysis. Table 2 and 3 gives values for water
requirements and Pressure Heads at the links and nodes
of the distribution system, respectively.

The demand was distributed based on the projected

values per node. The average highest demand was 0.05
lps per node. Reservoirs 1 and 2 were strategically
located and used as the main water supply for cluster 1
and cluster, respectively. The entire barangay was
Figure 4. Maximum Daily Demand Pattern 1 divided into two clusters based on the given
assumptions and topography of the location. Cluster 1
corresponds to Junction 1 to 47 while Cluster 2
corresponds to Junction 48-70.
Furthermore, the Needed Fire Flow (NFF) which
corresponds to water demands associated with fighting 200mm and 100mm diameter PVC pipes were used in
fires is computed using the formula: the entire system. Figure 5 shows the EPANET
network map of Barangay Santa Rosa showing the pipe
NFFi = Ci Oi (X +P)I diameter and the pressure per node based on the
Where: simulation conducted. During the initial analysis,
system errors such as negative pressure in various
NFFi is the needed fire flow at a location i junctions were observed but after series of
configuration, the simulation was successful. Table 2
Ci is the construction factor based on the size of the shows the velocity of water passing through the pipes
building and its constructionOi is the occupancy over a time series of 24 h.
factor reflecting the kinds of materials stored in
buildings (0.75 to 1.25) Based on other studies using EPANET, some WDS
experiences negative flow of water which naturally due
(X+P)I is the sum of the exposure of other buildings ( to the high altitude of the area. Mulanay particularly the
values range from 1.0 to 1.75) Barangay Santa Rosa is complicated in terms of
topography. The elevations used in the simulation were
Ci is the portion of the NFF attributed to the size of based on the contour map of the area being studied.
the building and its construction

Ci =220F* SQR( Ai) liters per minute

Where:

Ci is in liters per minute

Ai (sq.m) is the effective floor area, typically equal to

the area of the largest floor in the building plus 50%
of the area of all other floors;

F is the coefficient based on class of construction

4
Water Supply and Distribution Sample Paper

Figure 5. EPANET Network Map of Brgy. Santa Rosa, Mulanay, Quezon

Table 2. Flow and velocity of network links

Unit Friction
Link Length Diameter Roughness Flow Velocity Status
Headloss Factor
ID m mm LPS m/s m/km
Pipe L2 101.94 200 120 1.58 0.05 0.02 0.038 Open
Pipe L3 237.41 200 130 1.57 0.05 0.02 0.033 Open
Pipe L4 302.48 200 130 0.39 0.01 0 0.04 Open
Pipe L5 55.92 200 130 0.34 0.01 0 0.046 Open
Pipe L6 47.36 200 130 0.28 0.01 0 0.039 Open
Pipe L7 107.14 200 130 0.25 0.01 0 0.042 Open
Pipe L8 45.93 200 130 0.2 0.01 0 0.041 Open
Pipe L9 122.75 200 130 0.14 0 0 0.06 Open
Pipe L10 41.54 200 130 0.11 0 0 0 Open
Pipe L11 112.03 200 130 0.06 0 0 0 Open
Pipe L12 140.87 200 130 0.03 0 0 0 Open
Pipe L13 183.74 200 130 1.16 0.04 0.01 0.035 Open
Pipe L14 303.75 200 130 1.11 0.04 0.01 0.035 Open

5
Water Supply and Distribution-Title 21 April 2020

Pipe L15 145.88 200 130 0.99 0.03 0.01 0.036 Open
Pipe L16 200.49 200 130 0.87 0.03 0.01 0.036 Open
Pipe L17 161.6 200 130 0.75 0.02 0.01 0.037 Open
Pipe L18 211.03 200 130 0.71 0.02 0 0.037 Open
Pipe L19 105.53 200 130 0.67 0.02 0 0.037 Open
Pipe L20 69.71 200 130 0.59 0.02 0 0.039 Open
Pipe L21 64.47 200 130 0.47 0.02 0 0.04 Open
Pipe L22 95.64 200 130 0.35 0.01 0 0.043 Open
Pipe L23 142.45 200 130 0.31 0.01 0 0.043 Open
Pipe L24 258.42 200 130 0.27 0.01 0 0.043 Open
Pipe L25 437.13 200 130 0.19 0.01 0 0.047 Open
Pipe L27 240.81 200 130 -228.12 7.26 213.3 0.016 Open
Pipe L28 140.77 200 130 11.4 0.36 0.83 0.025 Open
Pipe L29 141.41 200 130 11.34 0.36 0.82 0.025 Open
Pipe L30 230.01 200 130 5.94 0.19 0.25 0.027 Open
Pipe L31 118.5 200 130 5.89 0.19 0.24 0.027 Open
Pipe L32 267.15 200 130 5.83 0.19 0.24 0.027 Open
Pipe L33 148.88 200 130 5.32 0.17 0.2 0.028 Open
Pipe L34 169.72 200 130 5.72 0.18 0.23 0.027 Open
Pipe L35 112.42 200 130 5.69 0.18 0.23 0.027 Open
Pipe L36 79.95 200 130 5.65 0.18 0.23 0.027 Open
Pipe L37 53 200 130 5.62 0.18 0.22 0.027 Open
Pipe L38 157.07 200 130 5.59 0.18 0.22 0.027 Open
Pipe L39 191.08 200 130 5.57 0.18 0.22 0.028 Open
Pipe L40 255.46 200 130 5.54 0.18 0.22 0.028 Open
Pipe L41 121.82 200 130 5.51 0.18 0.22 0.028 Open
Pipe L42 185.11 200 130 5.48 0.17 0.21 0.028 Open
Pipe L43 220.76 200 130 5.45 0.17 0.21 0.028 Open
Pipe L44 305.18 200 130 0.05 0 0 0.04 Open
Pipe L45 96.05 100 130 0.03 0 0 0.06 Open
Pipe L46 17.62 100 130 0.03 0 0 0 Open
Pipe L47 24.43 100 130 0.03 0 0 0 Open
Pipe L48 65.26 100 130 0 0 0 0 Open
Pipe L49 20.73 100 130 0.03 0 0 0 Open
Pipe L50 30.9 100 130 0.03 0 0 0.093 Open
Pipe L51 147.33 100 130 0.03 0 0 0.039 Open
Pipe L52 63.24 100 130 0.04 0.01 0 0.067 Open
Pipe L53 100.02 100 130 0.04 0.01 0 0.056 Open
Pipe L54 70.71 100 130 0.04 0.01 0 0.04 Open
Pipe L55 46.95 100 130 0.04 0.01 0 0.03 Open
Pipe L56 93.13 100 130 0.04 0.01 0 0.06 Open
Pipe L57 39.09 100 130 0.04 0.01 0 0.072 Open
Pipe L58 120.77 100 130 0.04 0.01 0 0.058 Open
Pipe L59 83.86 100 130 0.08 0.01 0 0.048 Open
Pipe L60 134.33 100 130 0.04 0.01 0 0.052 Open
Pipe L61 95.89 100 130 0.04 0.01 0 0.053 Open

6
Water Supply and Distribution Sample Paper

Pipe L62 101.68 100 130 0 0 0 0 Open

Pipe L63 62.57 100 130 0.04 0 0 0.058 Open
Pipe L64 90.25 100 130 0.04 0 0 0.054 Open
Pipe L65 85.45 100 130 0.04 0.01 0 0.049 Open
Pipe L66 94.86 100 130 0.04 0.01 0 0.045 Open
Pipe L67 248.71 100 130 0.11 0.01 0 0.045 Open
Pipe L68 106.52 100 130 0.05 0.01 0 0.051 Open
Pipe L69 121.41 100 130 0.03 0 0 0.059 Open
Pipe L70 109.13 100 130 0.48 0.06 0.07 0.036 Open
Pipe L71 205.8 100 130 -0.43 0.05 0.06 0.037 Open
Pipe L72 178.4 100 130 0.46 0.06 0.06 0.036 Open
Pipe L73 101.94 100 130 0.01 0 0 0.056 Open
Pipe L74 101.33 100 130 0.01 0 0 0.057 Open
Pipe PI1 50 12 100 -0.29 2.55 1333.32 0.048 Open
Pump PU2 #N/A #N/A #N/A 239.54 0 -51.37 0 Open
Pump PU1 #N/A #N/A #N/A 1.89 0 -66.67 0 Open

Table 2. Flow and velocity of network nodes

Node Elevation Base Demand Demand Head Pressure

Id m LPS LPS m m
Junc J1 130 0.07 0.01 216.67 86.67
Junc J2 120 0.07 0.01 216.66 96.66
Junc J3 55 0.07 0.01 216.66 161.66
Junc J4 30 0.14 0.03 216.66 186.66
Junc J5 30 0.14 0.03 216.66 186.66
Junc J6 25 0.14 0.03 216.66 191.66
Junc J7 25 0.14 0.03 216.66 191.66
Junc J8 20 0.14 0.03 216.66 196.66
Junc J9 20 0.14 0.03 216.66 196.66
Junc J10 80 0.14 0.03 216.66 136.66
Junc J11 20 0.14 0.03 216.66 196.66
Junc J12 20 0.14 0.03 216.66 196.66
Junc J13 20 0.14 0.03 216.66 196.66
Junc J14 20 0.14 0.03 216.66 196.66
Junc J15 20 0 0 216.66 196.66
Junc J16 20 0.14 0.03 216.66 196.66
Junc J17 20 0.14 0.03 216.66 196.66
Junc J18 20 0.14 0.03 216.66 196.66
Junc J19 20 0.14 0.03 216.66 196.66
Junc J20 20 0.14 0.03 216.66 196.66
Junc J21 20 0.2 0.04 216.65 196.65
Junc J22 20 0.2 0.04 216.65 196.65
Junc J23 20 0.2 0.04 216.65 196.65
Junc J24 20 0.2 0.04 216.65 196.65
Junc J25 20 0.2 0.04 216.65 196.65

7
Water Supply and Distribution-Title 21 April 2020

Junc J26 20 0.2 0.04 216.65 196.65

Junc J27 20 0.2 0.04 216.65 196.65
Junc J28 20 0.2 0.04 216.65 196.65
Junc J29 20 0.2 0.04 216.65 196.65
Junc J30 15 0.2 0.04 216.65 201.65
Junc J31 15 0.2 0.04 216.65 201.65
Junc J32 15 0.2 0.04 216.65 201.65
Junc J33 15 0.2 0.04 216.65 201.65
Junc J34 15 0.2 0.04 216.65 201.65
Junc J35 15 0.2 0.04 216.65 201.65
Junc J36 15 0.2 0.04 216.65 201.65
Junc J37 15 0.2 0.04 216.65 201.65
Junc J38 15 0.2 0.04 216.65 201.65
Junc J39 15 0.2 0.04 216.65 201.65
Junc J40 15 0.2 0.04 216.65 201.65
Junc J41 20 0.2 0.04 216.65 196.65
Junc J42 20 0.2 0.04 216.65 196.65
Junc J43 20 0.2 0.04 216.65 196.65
Junc J44 20 0.2 0.04 216.65 196.65
Junc J45 20 0.2 0.04 216.65 196.65
Junc J46 20 0.27 0.05 216.65 196.65
Junc J47 20 0.27 0.05 216.65 196.65
Junc J48 20 0.27 0.05 95.61 75.61
Junc J49 20 0.27 0.05 95.61 75.61
Junc J50 20 0.14 0.03 95.66 75.66
Junc J51 20 0.14 0.03 95.7 75.7
Junc J52 20 0.14 0.03 95.72 75.72
Junc J53 20 0.14 0.03 95.78 75.78
Junc J54 20 0.14 0.03 95.82 75.82
Junc J55 30 0.14 0.03 95.86 65.86
Junc J56 30 0.14 0.03 95.89 65.89
Junc J57 30 0.14 0.03 95.91 65.91
Junc J58 30 0.14 0.03 95.95 65.95
Junc J59 30 0.14 0.03 95.98 65.98
Junc J60 30 0.27 0.05 96.05 66.05
Junc J61 40 0.27 0.05 96.08 56.08
Junc J62 40 0.27 0.05 96.08 56.08
Junc J63 40 0.14 0.03 96.25 56.25
Junc J64 40 0.14 0.03 96.37 56.37
Junc J65 40 0.14 0.03 96.25 56.25
Junc J66 40 0.14 0.03 95.97 55.97
Junc J67 40 0.14 0.03 95.96 55.96
Junc J68 40 0.07 0.01 95.89 55.89
Junc J69 40 0.07 0.01 95.87 55.87
Junc J70 40 0.07 0.01 95.87 55.87
Resvr R2 45 #N/A -11.42 45 0

8
Water Supply and Distribution Sample Paper

Resvr R1 150 #N/A -1.6 150 0

L. Guppy, P. Uyttendaele, K.G. Villholth, V. Smakhin

4 CONCLUSIONS Groundwater and Sustainable Development Goals: Analysis
The water distribution system of the case study was of Interlinkages
analyzed to show and represent the capabilities of the UNU-INWEH Report Series (2018)
proposed system using EPANET 2.0 software.

It is therefore recommended to further explore other L. Ramulongo, N.S. Nethengwe, A. MusyokiThe nature of
areas of optimization to further reduce water losses and urban household water demand and consumption in
inefficiency in water distribution system. Actual rates makhado local municipality: a case study of makhado
of water demands may also be considered to produce a newtown
Procedia Enviro. Sci., 37 (2017), pp. 182-194
more reliable analysis.
Mulanay Municipal Ecological Profile of 2017
In the design of water distribution system, complexities
exist due to hydraulic considerations, water quality,
topography, cost efficiency, and system resiliency. The O.S.I. Fayomi, D.O. Olukanni, G.U. Fayomi, O.O. JosephIn
use of EPANET 2.0 software was found to be efficient situ assessment of degradable carbon effusion for industrial
in the analysis especially in the design of rural water waste water treatment
distribution systems. Cogent Engineering, 2017 (4) (2017), p. 1291151

Rural Water Supply Design Manual Volume 1

REFERENCES

B.S. Jung, B.W. Karney, P.F. Boulos, D.J. WoodThe need

for comprehen-sive transient analysis of distribution
systems
J. American Water Works Associa-tion, 99 (1) (2007),
pp. 112-123

D. Savić, H. Mala-jetmarova, N. SultanovaHistory of

Optimization in Water Distribution System Analysis
(2018)

E.O. Longe, D.O. Omole, I.K. Adewumi, S.A. OgbiyeWate

r resources use, abuse and regulations in Nigeria
J. Sustain. Dev. Afr., 12 (2) (2010), pp. 35-45

G. Tsakiris
Rational design of urban water supply and distribution
systems
Water Util J, 8 (2014), pp. 5-16

H. Mala-Jetmarova, N. Sultanova, D. Savic

Lost in optimisation of water distribution systems? A
literature review of system operation
Environ. Model. Software, 93 (2017), pp. 209-254

H.M. Samani, A. MottaghiOptimization of water

distribution networks using integer linear programming
J. Hydraul. Eng., 132 (5) (2006), pp. 501-509

J. Lavaei, S.H. LowZero duality gap in optimal power flow

problem
IEEE Trans. Power Syst., 27 (1) (2012), pp. 92-107