ArticlePDF Available

Abstract and Figures

This paper presents an economic analysis of pressure control solutions for leakage and pipe burst reduction. In detail, it explores the operating conditions under which the installation of conventional mechanical pressure reducing valves (PRVs) or remotely real-time controlled (RTC) valves are cost effective compared to a scenario with no control. For a range of system sizes, hydraulic extended period simulations and empirical formulas were used to estimate leakage rates and pipe bursts, respectively, in numerous operational scenarios, including different precontrol leakage levels and demand patterns, the absence of pressure control, and the installation of a PRVor RTC valve. The total cost of the controlled system, including the installation cost of the control device, the flow-dependent operation and maintenance (O&M) cost, and the pipe burst repair cost over the planning horizon, was compared with the water-related O&M and pipe burst repair costs of the uncontrolled system. The results pointed out that no pressure controls are needed if leakage and the variable O&M cost of water are low. When these variables are high, remote RTC is attractive, especially when the demand pattern is peaked and the system is large. For more moderate cost and leakage, a conventional PRV may be better than RTC, especially in small systems and for relatively smooth demand patterns.
Content may be subject to copyright.
WR3072
Economic Analysis of Pressure Control for Leakage and Pipe Burst Reduction
Enrico Creaco
1
and Thomas Walski, F.ASCE
2
Abstract
This paper presents an economic analysis of pressure control solutions for leakage and pipe burst
reduction. In detail, it explores the operating conditions under which the installation of
conventional mechanical pressure reducing valves (PRVs) or remotely real time controlled
valves (RTC valves) are cost-effective compared to a scenario with no control. For a range of
system sizes, hydraulic extended period simulations and empirical formulas were used to
estimate leakage rates and pipe bursts, respectively, in numerous operational scenarios, including
different pre-control leakage levels and demand patterns, the absence of pressure control and the
installation of a PRV or of an RTC valve. The total cost of the controlled system, including the
installation cost of the control device, the flow dependent operation and maintenance (O&M)
cost and the pipe burst repair cost over the planning horizon, was compared with the water-
related O&M and pipe burst repair costs of the uncontrolled system. The results pointed out that
no pressure controls are needed if leakage and the variable O&M cost of water are low. When
these variables are high, remote RTC is attractive, especially when the demand pattern is peaked
and the system is large. For more moderate cost and leakage, a conventional PRV may be better
than RTC, especially in small systems and for relatively smooth demand patterns.
1
Assistant Professor, Dipartimento di Ingegneria Civile e Architettura, Univ. of Pavia, Via Ferrata 3, 27100 Pavia,
Italy. Adjunct Senior Lecturer in the School of Civil, Environmental and Mining Engineering - University of
Adelaide. Honorary Senior Research Fellow of the College of Engineering, Mathematics and Physical Sciences -
University of Exeter E-mail: creaco@unipv.it
2
Bentley Fellow, Bentley Systems, Incorporated, 3 Brian’s Place, Nanticoke, PA 18634 (corresponding author). E-
mail: Tom.Walski@bentley.com
Keywords
Pressure reducing valves, real time control, pressure control, economic analysis, leakage
reduction, water distribution
Introduction
Water distribution systems lose a significant amount of water due to leakage (Farley and Trow,
2003). This number can vary from a few percent of water product to nearly half depending on the
system, but worldwide, non-revenue water is estimated at 27% of water production (Wu, 2011).
Leakage is generally a major component of non-revenue water.
Leak detection and repair is generally the best way to reduce leakage but some leaks are
undetectable or very costly to repair. Leakage rates are a function of pressure in the distribution
system and therefore pressure control can be used to reduce leakage. An important added benefit
of pressure control lies in the reduction in pipe bursts (Thornton and Lambert, 2006; Lambert et
al, 2013). In fact, repairing pipe bursts is costly and causes external diseconomies (Pigou, 1932),
such as traffic slow down or disruption in the interested area.
In pressure zones with elevated (floating) storage, pressure is controlled by water levels in the
tanks and there is not much that can be done to modify pressure other than adjusting pressure
zone boundaries. In pressure zones without storage, pressure can be controlled by adjusting
pump controls in zones fed by pumps or installing control valves and adjusting pressure settings
in zones that can be served through some type of control valve. This paper addresses pressure
control for leak and pipe burst reduction in pressure zones which are or can be fed through
control valves.
In the recent years, significant research has been focused on the optimal placement of these
devices to reduce leakage. To this end, both single-objective (e.g., Vairavamoorthy and Lumbers,
1998; Araujo et al., 2006; Liberatore and Sechi, 2009; Ali, 2015) and multi-objective (e.g.,
Nicolini and Zovatto, 2009; Creaco and Pezzinga, 2015a; 2015b) methodologies have been used.
Approaches to pressure control can be grouped into two general categories (Puust et al., 2010;
Vicente et al., 2016):
- Conventional pressure reducing valves (PRVs).
- Remote real time control (RTC).
PRVs are mechanical devices that automatically reduce a higher inlet pressure to a steady lower
downstream pressure, regardless of changing flow rate and/or varying inlet pressure. They have
been used for many years with the primary purpose to reduce service pressure to acceptable
level. Though their setting can be changed to accommodate demand variations in the system, it is
unlikely that it will be adjusted continuously in real-time. Furthermore, without performing
measurements in the network, it is difficult to adjust the valve setting while guaranteeing that the
pressure head is larger than or equal to the minimum desired value at each node and for each
value of network demand.
To improve service pressure regulation by taking account of measurements in the network, RTC
can be adopted. To adjust the setting of control valves, RTC can be carried out in two different
modes, that is local RTC and remote RTC. An example of local RTC is that of a control valve
regulated as a function of the water discharge measured upstream. In detail, it is performed
thanks to a programmable logic controller (PLC), which can use the water discharge (or pressure
head) measurement, received from an electromagnetic flowmeter (or a manometer) in proximity
to the valve, to calculate a new valve setting. Though benefitting from the valve setting adjusted
in real time, the local RTC does not have full information about pressure in the network, which is
the variable of interest to be minimized under service constraints. In fact, in light of the
stochastic nature of demand, a single valve setting/water discharge relationship implemented
inside the PLC does not suffice to ensure a safe regulation in terms of service pressure at all
times. To overcome this problem, remote RTC (e.g., see Campisano et al., 2010; 2012; Creaco
and Franchini, 2013; Berardi et al., 2015; Campisano et al., 2016; Creaco et al., 2016; 2017) can
be performed. In detail, this approach monitors pressure heads at remote point(s) in the system,
usually the critical node(s) with the lowest pressure head (Campisano et al., 2010), and signals to
the control valve the reading so that a programmable logic controller (PLC) can set a valve
position at the source to maintain the minimum desired pressure at the remote point. Due to its
superior performance, which comes at quite similar installation costs compared to local RTC,
remote RTC is the kind of RTC considered in this work and the one referred to, when using the
word “RTC” hereinafter.
The results of some work (e.g., see Creaco et al., 2016) have shown that RTC is more effective at
regulating pressure and reducing leakage than the conventional PRVs. However, the better
performance of RTC comes at a larger installation cost, though this aspect has often been
overlooked in recent studies. Even at the Battle of Background Leakage Assessment for Water
Networks (Berardi and Giustolisi, 2016), which took place at WDSA2014, the same installation
cost was contemplated, for the sake of simplicity, for locally and remotely controlled valves, that
is PRVs and RTC valves.
Therefore, an economic analysis (Sullivan et al., 2015) is required to explore the cases in which
RTC is cost effective in the long run, compared to the adoption of conventional PRVs. In fact,
the selection of the more suitable control method should be based on a life cycle cost analysis
comparing the value of lost water and pipe break repairs with the cost of control measures. This
analysis should essentially consider three options for leakage control: no control, conventional
PRV, and RTC.
The “no control” case will be used as the basis with which the PRV and RTC controls will be
compared. The PRV control requires installation of a vault (valve chamber) equipped with the
control valve and with isolation valves to enable valve maintenance. RTC can more precisely
control pressure but it requires two vaults (one for the pressure sensor and one for the control
valve), plus communication and control equipment.
The objective of this study is to determine which control measure produces the lowest life cycle
costs. There is not a single universal solution but different measures will be best depending on
site specific conditions such as
- Value of water.
- Costs for equipment and installation.
- Extent of initial leakage.
- Size of the system.
- Extent to which pressure control can reduce pipe breaks.
- Nature of demands and their patterns.
This sensitivity analysis is based on a realistic model of a single source pressure zone in a water
distribution system. The driving parameters are varied over a wide range to determine their
influence. Based on those calculations, an attempt is made to generalize the findings and provide
practical guidance.
Cost-Benefit Analysis
The results of the cost-benefit analysis are dependent on the cost data used. Due to local
conditions and design decisions, the costs and benefits can vary widely. There are two sides to
the total cost–benefit evaluation: costs to install and operate any leak and pipe burst control
measures and benefits of control due to leak and pipe burst reduction. Hereinafter, each element
of cost and benefit is described.
Total cost connected to water production
The total cost C
wp
(€) connected to water production is calculated as:
totbursttotvarinstwp
CCCC
,,
+
+
=
, (1)
where C
inst
(€) is the cost sustained by the water utility to install a pressure control solution,
which is either PRV or RTC, as is explained in the following sub-section. In the absence of
control, C
inst
is equal to 0. C
var,tot
(€) and C
burst,tot
(€) are the water-related total variable operating
and maintenance (O&M) cost and the total pipe burst repair cost, respectively, described in the
homonymous sub-sections.
The analysis must be applied to one pressure zone at a time.
Installation Costs
Cost estimates were performed for the two types of pressure control PRV and RTC. In this
work, numerous assumptions were made in developing these estimates including:
- The valves would be placed in underground vaults.
- In each vault, there would be one control valve (either PRV or RTC valve) with isolation
valves at either valve end.
- In each vault, a by-pass equipped with an isolation valve is present, to be used during
valve maintenance.
- Communication cost is included for the RTC alternative.
- Fixed O&M costs associated with the various elements present inside vaults and
evaluated over the cycle life of the control systems (e.g., 40 years) is small compared
with initial cost.
- There are no land acquisition or paving costs associated with the vaults.
Variable O&M cost
The price of water to consumers is a poor indicator of the cost savings from pressure control
since it includes the cost of debt reduction for capital facilities, fixed O&M costs and other costs
such as operator labor, meter reading, and other management functions which do not vary with
water production. (The variable O&M cost is referred to as the unit cost in this paper.) The
variable O&M cost of water, instead, is impacted by leak reduction. At the generic year i, the
variable O&M cost C
var,i
(€) primarily includes the cost of pumping and of treatment energy and
chemicals and is obtained by multiplying the water unit cost by the supplied water volume in the
year. To obtain the present worth variable O&M costs C
var,tot
to be used in eq. (1), the following
formula is used, derived from Sullivan et al. (2015):
( )
=
+
=
T
i
i
ivar
totvar
r
C
C
1
,
,
1
, (2)
where T (years) and r (-) are the reference period (life cycle) for total cost evaluation and the
present worth factor to transfer costs to the initial year, respectively.
As derived from surveys of the expenses sustained by water utilities in Italy, the water variable
unit costs can vary by orders of magnitude between water utilities. In this study, a range from
0.001 €/1000 L to 1.0 €/1000L is used. The lower value would be the case for a utility that
obtains its water from a high-quality mountain source and does minimal pumping and chemical
addition. The larger value, instead, would correspond to a water utility that may need to pump
water from a deep well or over a long distance and may require a high level of treatment such as
a membrane process or desalinization.
Pipe burst repair cost
The total pipe burst repair cost C
burst,tot
is also impacted by pressure control. In fact, the reduction
in the service pressure causes a decrease in the number n
b
of yearly pipe breaks. Hence decreases
the yearly pipe burst repair cost C
burst
(€), calculated as n
b
×c
burst
with c
burst
(€/break) being the
cost of the single repair.
In detail, under either uncontrolled or controlled conditions, C
burst,tot
for the life cycle can be
calculated as:
( )
=
+
=
T
i
i
iburst
totburst
r
C
C
1
,
,
1
, (3)
where C
burst,i
is the pipe burst repair cost at the generic year i.
The yearly number n
b
(-) of pipe breaks in the absence of pressure control can be evaluated as:
Ln
b
λ
=
, (4)
where
λ
(breaks/year/Km) and L (Km) are the pipe failure likelihood, to be derived from in-situ
observations, and the network length, respectively.
In order to assess the effect of pressure reduction on the number of pipe bursts, the following
relationship can be used (Lambert et al., 2013), which relates the reduction in n
b
to the reduction
in the maximum daily pressure head h
max
:
( )
=
3
,
11
max
max,p
b
b
bpb
h
h
C
n
nn
, (5)
where n
b,p
and h
max,p
are the values of n
b
and h
max
, respectively, following service pressure
reduction. Finally, C
b
(-) represent the fraction of pressure independent pipe bursts. The variation
range of C
b
is [0, 1]. As shown by Lambert et al. (2013), the lower values of
C
b
are associated
with old networks and high values of h
max
; the higher values, instead, are associated with new
networks and low values of h
max
.
Benefits
The implementation of a pressure control strategy impacts C
wp
in eq. (1). In fact, it will produce
an installation cost C
inst
and a decrease in C
var,tot
and
C
burst,tot
, because of the reduction in leakage
and supplied water volume, and because of the attenuation of pipe bursts. Therefore, the benefits
of PRV and RTC, compared to the no control scenario, can be evaluated in terms of percentage
reduction in C
wp
:
w,p
w,p
C
C
=100(%)Redution Cost
, (6)
where C
w,p
is the decrease in C
w,p
obtained through the generic pressure control measure (either
PRV or RTC).
In this context, it is assumed that pressure does not affect the outflow to the network users’, as is
the case with indoor consumption. Therefore, the water utility’s income from bills is not
significantly reduced because of pressure regulation.
Applications
Case study
The pressure zones considered for the analyses were derived from a network in Northern Italy
(Farina et al., 2014), which has already been used for research in the field of pressure control
(Creaco and Franchini, 2013). The layout of the network is reported in Figure 1.
The first zone analyzed, hereinafter indicated as “large system”, is the skeletonized model of the
network, made up of 26 demanding nodes, a single source node and 32 pipes. In this system,
which serves about 30,000 inhabitants, the source node has a ground elevation of 35 m a.s.l. and
constant total head of 40 m. All the other nodes are assumed to have an elevation of 0 m a.s.l..
The average nodal demands and the pipe features (in terms of length and diameter) are reported
in Tables 1 and 2 respectively. All the pipes are assumed to feature a Manning roughness
coefficient of 0.01 s/m
1/3
, typical value for PVC. The hydraulic grade upstream of the pressure
control is quite constant in the day.
The second zone analyzed, hereinafter indicated as “small system”, is supposed to serve a 10
time smaller urban area. This system was obtained by scaling down the large system. In detail,
nodal demands (Table 1) and pipe lengths (Table 2) were scaled down by factors of 10 and 10,
respectively. The factor of 10 for pipe lengths is actually obtained as the square root of the
scaling down factor of 10 used for the urban area. In this system, which serves about 3,000
inhabitants, the pipe diameters were re-designed. In this re-design process, they were slightly
oversized (no diameter smaller than 100 mm was adopted see Table 2) compared to users’
demands, in such a way as to account for reliability aspects, such as those related to fire
protection operating conditions. In fact, for small systems, hydrant demands can be much larger
than users’ demands.
At this stage, a caveat has to be made concerning the use of the terms “large system” and “small
system”, which only enable distinguishing the two systems analyzed based on the layout length.
Two different patterns were used for the hourly demand multiplier (Figure 2), to represent the
daily variation in the users’ demand in both the systems. The former pattern is quite smooth with
multiplier values ranging from 0.5 to 1.34 whereas the latter is more peaked with multiplier
values ranging from 0.1 to 3.0.
The hydraulic simulation of the systems was carried out using the software WaterGEMS
(Bentley, 2016). Leakage was modelled through nodal emitters with exponent equal to 1, which
is a typical value used for leakage analysis (Ferrante, Meniconi and Brunone, 2014; Van Zyl and
Cassa, 2014; Schwaller and van Zyl, 2015). Emitter coefficients were calibrated to obtain three
different leakage levels under no control conditions: low (with leakage rates of 3%-5%), medium
(with leakage rates of 15%-16%) and high (with leakage rates of 30%-31%). In the simulations,
the PRV and RTC control valves were installed in lieu of pipe 31 (see Figure 1). The control
valves were sized in order to produce similar head losses under fully open valve conditions. In
detail, in the case of the large system, they were installed with DN350 and DN300 sizes,
respectively. In the case of the small system, instead, they were installed with DN125 and
DN100, respectively.
In this work, the installation costs (see Tables 3 and 4) of the control systems were evaluated
thanks to the price list of an Italian valve manufacturer (T.i.S. Group). In detail, for the large
system, the total installation costs C
inst
for the PRV and RTC controls were equal to 17,579 € and
33,937 €, respectively. For the small system, instead, they were equal to 7,770 and 26,851 €,
respectively. In this context, it has to be noted that moving from the large to the small system,
the PRV cost is almost halved. The RTC cost, instead, decreases only by 20%. This smaller cost
reduction is due to the presence of more fixed costs in RTC, independent of the system size, such
as those related to communications and power to site, remote sensor, additional vault, and PLC
programming.
Overall, 18 WaterGEMS simulations were carried out on each system (see Tables 5 and 6), to
consider scenarios with two different kinds of demand pattern (smooth and peaked), three levels
of leakage (low, medium and high) and three kinds of control (no control, PRV and RTC).
In the PRV scenarios, the single setting of the valve, in terms of pressure head at the downstream
node (i.e., node 20 – see Figure 1), was set to the lowest value guaranteeing pressure heads equal
to or higher than the desired value of 25 m at all the network nodes during the whole system
simulation. In the RTC scenarios, the head loss coefficient of the control valve was regulated at
each time step in such a way to always obtain a pressure head equal to the desired value of 25 m
at the critical node (i.e., node 1 – see Figure 1).
Results
The results of the WaterGEMS simulations were reported in Figures 3 and 4 and in Tables 5 and
6.
Large system pressure. Figure 3 concerns the scenarios with high level of leakage for the large
system. The upper portion of the figure, related to the smooth demand pattern, shows that the
pressure head at the critical node is largely reduced by introducing the PRV, in comparison with
the no control scenario. Only a minor additional reduction is obtainable through the variable
setting of RTC. The lower portion of the figure, related to the peaked demand pattern, shows that
the pressure head reduction obtainable through PRV is very small. In fact, the PRV head loss had
to be cautiously kept small to guarantee the minimum desired pressure head downstream at all
the hours of the day. Therefore, PRV can help in achieving 25 m at the controlled node only at
the demand peak of hour 8:00 in the morning. In the case of the peaked demand pattern, a
significant reduction can be obtained only through the variable setting of RTC.
Large system leakage. The results of Figure 3 are confirmed by Table 5, which reports the
leakage rates under the various control conditions. In fact, this table shows that, in the case of the
smooth demand pattern, the large leakage reduction takes place thanks to PRV whereas the
additional contribution of RTC to leakage reduction is marginal. Instead, when the demand
pattern is peaked, due to the single daily setting, PRV contributes to leakage reduction much less
than RTC, which benefits from the variable setting.
Difference between small and large systems in terms of pressure and leakage. Figure 4 and Table
6 enable comparing the results of the small system with those of the large system in Figure 3 and
Table 5. Due to the small system being oversized, the pressure head is quite uniform in the
network downstream of pipe 31, in which the valves are installed. As a result of this, a single
daily PRV setting is sufficient to cause a large reduction in pressure and leakage, even in the case
of peaked demand pattern. Therefore, in a similar way, the additional contribution of RTC to
pressure and leakage reduction is small, even in the case of peaked demand.
Pipe burst repairs. For the large and small systems respectively, Tables 7 and 8 report the
expected yearly number n
b
of bursts in the various scenarios. In detail, n
b
was set to 4 for the
large system in the no control scenarios, as derived from
λ
=0.031 breaks/year/Km and L = 131
Km for the not skeletonized network. In the small system, n
b
was set to 2 to take account of a
shorter length (= 41.4 Km = 100 Km ×10) and of the presence of smaller pipe diameters
(
λ
=0.049 breaks/year/Km). To assess n
b,p
in the controlled scenarios, in eq. (5) C
b
was set to 0.8
(new network and low value of h
max
, which is always around 40 m in the no control scenarios).
Furthermore, the maximum daily pressure head at the controlled node was used to assess h
max
.
Tables 7 and 8 also report h
max
and the yearly cost C
burst
of pipe burst repairs, obtained
considering a repair cost c
burst
= 900 for the single burst. In fact, preliminary surveys showed
that it is a reasonable value for repairing only the broken section of a pipe or using a repair
clamp. Overall, similar remarks to leakage reduction can be made as for the reduction in pipe
bursts. In detail, for the large system, in the case of smooth demand pattern, the larger reduction
in the number of bursts takes place thanks to PRV whereas the additional contribution of RTC is
marginal. Instead, when the demand pattern is peaked, due to the single daily setting, PRV
contributes to the reduction less than RTC, which benefits from the variable setting. For the
small system, the additional contribution of RTC is always small, even in the case of peaked
demand, because PRV is already sufficient to bring the service pressure head close to the desired
value of 25 m.
Economic comparison. Besides nodal pressure heads and leakage rates, the WaterGEMS
simulations enabled deriving the daily supplied volume (users’ demand + leakage) in the 18
scenarios for both the large (Table 5) and small (Table 6) systems. By setting T and r to 40 years
and 0.03, typical values for the cycle life of the control systems and for the present worth factor
in water works respectively, eq. (1,2,3) enabled evaluating C
inst
, C
burst,tot
, C
var,tot
, and C
wp
, in the
absence of control and in the presence of PRV or RTC, as a function of the water unit costs
ranging from 0.005 €/1000 L to 1.0 €/1000L in each of the 18 scenarios for either system. As an
example of the results, Figure 5 shows the cost breakdown, for the no control, PRV and RTC
scenarios for the large system in the case of smooth demand pattern and of water unit cost equal
to 0.01 €/1000 L. The figure shows that, for the no control scenario, C
inst
= 0 €, C
burst,tot
= 83,213
€ and C
var,tot
= 434,002 €, leading to C
wp
= 517,215 €. For the PRV scenario, the installation cost
C
inst
= 17,579 € is paid back, thanks to lower values of C
burst,tot
(= 72,031 €) and C
var,tot
(=413,357
€). In fact, in the PRV scenario, C
wp
= 502,966 €, with a 2.75 % cost reduction compared to the
no control scenario. In the RTC scenario, instead, the larger installation cost C
inst
= 33,937 is
hardly paid back, though RTC is able to yield the lowest values of C
burst,tot
(= 70,934 €) and
C
var,tot
(=411,897 €). In fact, RTC (with C
wp
= 516,768 €) produces almost no C
wp
reduction
compared to the no control scenario. When the water unit cost changes, C
wp
changes in the no
control, PRV and RTC scenarios. In detail, the only sensitive component inside C
wp
to variations
in the water unit cost is C
var,tot
, whereas C
inst
and C
burst,tot
are unaffected.
Subsequently, for either kind of demand and for each leakage level, the benefits of PRV and
RTC in comparison with the no-control scenario were evaluated in terms of percentage cost C
wp
reductions. The graphs in Figures 6 and 7 show the percentage cost reductions of PRV and RTC
for the large system and for the small system, respectively. The graphs show the cost savings due
to pressure control. When all the lines are horizontal at zero, pressure control is not justifiable (as
long as pressures are reasonable). When the solid line is above the dashed line and the horizontal
axis, a PRV is the most economical solution. When the dashed line is above the solid line and the
horizontal axis, RTC is the most economical.
Large system economic analysis. The analysis of Figure 6 for the large system shows that for
low water unit costs, implementing PRV or RTC has no benefits. In the case of smooth demand
pattern, when the water unit cost increases, some benefits arise associated with pressure control
and PRV is the more beneficial solution. Finally, RTC slightly surpasses PRV for the large
values of the water unit cost. The transition between PRV and RTC as better solution takes place
for lower water unit costs as leakage increases. The higher benefits of RTC are because leakage
volume savings pay back the larger installation cost when the water unit cost is high.
In the case of peaked demand pattern, when there are benefits from pressure control, RTC is the
more cost-effective solution. This is the result of the superior performance of RTC at reducing
pressure heads and leakage when the demand is very variable, as highlighted from Figure 3 and
Table 5.
Difference between small system and large in terms of economic analysis. The analysis of Figure
7 shows that, for the small system, the benefits associated with pressure control are shifted
rightwards in the water unit cost axis. This is because in the small system, due to the low leakage
volume savings obtainable and to the proportionally large installation cost of PRV and RTC, the
pressure control system is paid back only for much larger values of the water unit cost than in the
large system. In the case of smooth demand pattern, PRV is always the more beneficial solution
for pressure control. The large benefits of RTC remarked for the large system under peaked
demand pattern conditions tend to vanish as well, in the case of the small system. This is due
both to the large installation cost of RTC and to its reduced superiority in reducing pressure and
leakage in the small system, as highlighted from Figure 4 and Table 6.
Other Considerations
While this analysis accounted for many of the factors affecting decision making for pressure
control including: pipe burst rate, leakage rate, value of water, size of system, and shape of
demand pattern, there are quite a few other factors that must be taken into consideration when
deciding if to perform any pressure control and the type of pressure control that should be used:
- The choice of the set-point pressure head is a critical choice. If this variable is very high
relative to source pressure, there is little room for pressure reduction.
- Customers tend to like higher pressure than minimal pressure as long as high pressures
are not destructive.
- Water for fire protection is significantly better when pressures are normally high.
- Fire sprinkler systems work better and are less costly when pressures are higher.
- RTC can more effectively respond to very large flows such as fires.
- In-building pumping is more likely to be required when pressures are low.
- The outlet pressure of a PRV may not be an optimal indicator of system pressure in large
zone with significant elevation differences, in addition to considering the stochastic
nature of demand in space and time.
- The location of a critical customer in terms of distance from the control valve may
influence the use of RTC. In this analysis, it was at a remote location. If the crucial
customer is near the control valve, then the benefits become much smaller while the cost
remain greater than the PRV control case.
- The effects of leak reduction do not significantly impact the capital plant. If major capital
investment is needed in response to leakage, more aggressive leak control may be needed
to defer the construction.
- The exponent on the pressure dependent leakage may differ from the value used in this
study. Higher values correspond to greater impact of pressure management while values
approaching 0.5 correspond to less sensitivity of leakage to pressure. The effect of this
variable will be investigated in further studies.
- Establishing new pressure zones in an existing system may require closing “back door”
feeds into that new zone, which can impact the reliability of supply to that zone in case of
a failure.
- When flows are very larger and available head is high, then the use of turbines for energy
recovery becomes attractive.
- The value of the incoming pressure that would need to be dissipated through the control
valve also impacts the need for control. For example, when there are 100 m (140 psi) of
excess head above what is needed, pressure control is important while if there are 2 m (3
psi), no pressure control is justified.
- Pipe bursts may be much more frequent and related pressure reduction effects may be
much larger under different conditions from those considered in the paper (such as old
networks with higher pressure heads in the no control scenario).
The variables at stake in the analysis are very numerous (pipe diameters, roughness, network
layout and so on) and every network is different from the others. Whereas it is expected that a
change in each variable could affect the results by changing the size of the head losses in
between valve and controlled node, the analysis has already provided an insight into the different
behavior of normally designed and oversized networks. In fact, it showed that RTC tends to be
more cost-effective in networks with high head loss. In networks where head loss across the
network is small, the marginal benefits of RTC compared to PRV tend to vanish.
The calculations were performed under the assumption that there would not be a significant
change in leakage and in the number of pipe bursts over time. If leakage and the yearly number
of pipe busts accelerate over time, the benefits of real time control increase. On the other hand, if
major pipe replacement work is performed, the need for real time control diminishes. However,
the best way to reduce leakage is with an active leak detection and repair program, which could
reduce leakage rates and thus, reduce the need for pressure control, especially RTC.
If a major capital cost can be delayed or downsized through leak reduction, the benefits of leak
reduction can be significantly greater, whether through pressure management or leak repair.
The study did not investigate retrofitting an existing PRV valve for use with RTC.
For zones fed by pumps, installing control valves directly downstream of the pumps is wasteful
in energy. Other methods for pressure control can be used for pumped zones such as right-sizing
pumps, trimming impellers or using variable speed pumps.
With regard to economic factors, a high interest rate will decrease the present worth of the water-
related O&M and pipe burst repair costs, thus reducing the economic benefits of pressure control
in the life cycle of the control systems. A lower interest rate, instead, will increase these benefits.
Installation of a vault with control equipment provides an opportunity to install flow metering
and thus establish the zone as a district metered area (DMA), besides providing water discharge
data that can be used inside RTC to improve its performance (Creaco and Franchini, 2013). The
use of DMAs can improve water accountability and lead to quicker leak detection.
Conclusions
While detecting and repairing leaks is the preferred approach to leak and pipe burst reduction,
pressure control can also be successfully used in pressure zones with no floating storage tanks.
Depending on the amount of leakage, the variable unit cost of water, the local equipment and
construction costs, demand patterns and interest rates, different pressure control measures can
result in lowest life cycle costs.
The results of the analysis showed that no pressure control is needed when the leakage rate is
low and the variable water unit cost is also low, as long as the pressure is not so high as to result
in excessive pipe breakage. RTC is most beneficial when the leakage rate is high and the variable
water unit cost is also high, especially in the case of peaked demand pattern and in large systems.
In between those two extremes conventional PRVs are best for pressure control.
Acknowledgments
The Authors are grateful to the valve manufacturer T.i.S. group, and in particular to the Product
Manager Mr. Andrea Boccuzzi, for sharing information about the cost and performance of
control valves.
References
Ali, M. (2015). “Knowledge-based optimization model for control valve locations in water
distribution networks.” J. Water Resour. Plann. Manage., 141(1), 04014048.
Araujo, L. S., Ramos, H., and Coelho, S. T. (2006). “Pressure control for leakage minimisation
in water distribution systems management.” Water Resour. Manage., 20(1), 133–149.
Berardi, L. and Giustolisi, O. (2016). "Special Issue on the Battle of Background Leakage
Assessment for Water Networks." J. Water Resour. Plann. Manage., 142(5), C2016001.
Bentley (2016) WaterGEMS, Bentley Systems, Exton, Pa.
Berardi, L., Laucelli, D., Ugarelli, R., Giustolisi, O. (2015). “Leakage Management: Planning
Remote Real Time Controlled Pressure Reduction in Oppegård Municipality.” Procedia
Engineering 119(1), 72-81, doi: 10.1016/j.proeng.2015.08.855.
Campisano, A., Creaco, E. and Modica, C. (2010). “RTC of valves for leakage reduction in
water supply networks.” J. Water Resour. Plng. Mgmt., 136 (1), 138–141.
Campisano, A., Modica, C., Reitano, S., Ugarelli, R., and Bagherian, S. (2016). “Field-Oriented
Methodology for Real-Time Pressure Control to Reduce Leakage in Water Distribution
Networks.” J. Water Resour. Plng. Mgmt., 10.1061/(ASCE)WR.1943-5452.0000697, 04016057.
Campisano, A., Modica, C. and Vetrano, L. (2012). “Calibration of proportional controllers for
the RTC of pressures to reduce leakage in water distribution networks.” J. Water Resour. Plng.
Mgmt., 138(4), 377–384.
Creaco E., Alvisi S., Franchini M. (2016). Multi-step approach for optimizing design and
operation of the C-Town pipe network model. Journal of Water Resources Planning and
Management, 142(5), C4015005.
Creaco E., Franchini M. (2013). “A new algorithm for the real time pressure control in water
distribution networks.” Water Science and Technology-Water Supply, 13(4), 875-882.
Creaco E., Pezzinga G. (2015a). “Multi-objective optimization of pipe replacements and control
valve installations for leakage attenuation in water distribution networks.” J. Water Resour.
Plann. Manage., 141(3), 04014059.
Creaco E., Pezzinga G. (2015b). “Embedding Linear Programming in Multi Objective Genetic
Algorithms for Reducing the Size of the Search Space with Application to Leakage
Minimization in Water Distribution Networks.” Environmental Modelling & Software, 69, 308-
318.
Creaco E., Campisano A., Franchini M., Modica C. (2017). “Unsteady flow modelling of
pressure real time control in water distribution networks.” Journal of Water Resources Planning
and Management, accepted.
Farina G., Creaco E., Franchini M. (2014). “Using EPANET for modelling water distribution
systems with users along the pipes.” Civil Engineering and Environmental systems, 31(1), 36-50.
Farley, M., and Trow, S. (2003). Losses in water distribution networks, IWA, London.
Ferrante, M., Meniconi, S., Brunone, B., 2014. Local and global leak laws. Water Resources
Management 28, 3761-3782. doi:10.1007/s11269-014-0708-x
Lambert, A., Fantozzi, M., and Thornton, J. (2013). “Practical approaches to modeling leakage.”
Conference on Computing and Control for the Water Industry CCWI2005, Exeter, UK.
Liberatore, S., and Sechi, G. M. (2009). “Location and calibration of valves in water distribution
networks using a scatter-search meta-heuristic approach.” Water Resour. Manage., 23(8), 1479–
1495.
Nicolini, M., and Zovatto, L. (2009). “Optimal location and control of pressure reducing valves
in water networks.” J. Water Resour. Plann. Manage., 135(3), 178–187.
Pigou, A.C. (1932). “The Economics of Welfare.” Macmillan and Co., London, 4th edition.
Puust, R., Kapelan, Z., Savic, D. A., and Koppel, T. (2010). “A review of methods for leakage
management in pipe networks.” Urban Water J., 7(1), 25–45.
T.i.S. Group [Valve manufacturer]. Company profile. Bergamo, IT.
Schwaller, J., van Zyl, J.E., 2015. Modeling the Pressure-Leakage Response of Water
Distribution Systems Based on Individual Leak Behavior. J Hydraul Eng-ASCE 141, 04014089-
8. doi:10.1061/(ASCE)HY.1943-7900.0000984
Sullivan, W.G., Wicks E.M., and Koelling C.P. (2015). “Engineering Economy, 16th Edition.”
Prentice Hall College Div, New Jersey, ISBN13: 9780133439274.
Thornton, J. and Lambert A. (2006). “Managing pressures to reduce new break frequencies, and
improve infrastructure management” Water 21, December 2006
Vairavamoorthy, K., and Lumbers, J. (1998). “Leakage reduction in water distribution systems:
Optimal valve control.” J. Hydraul. Eng., 124(11), 1146–1154.
Van Zyl, J. E., and Cassa, A. M. (2014). “Modeling elastically deforming leaks in water
distribution pipes.” 140(2), 182–189.
Vicente, D., Garrote, L., Sánchez, R., and Santillán, D. (2016). “Pressure Management in Water
Distribution Systems: Current Status, Proposals, and Future Trends.” J. Water Resour. Plann.
Manage., 142(2), 04015061.
Wu, Z. (2011). “Introduction to Water Loss Reduction,” in Water Loss Reduction, Ed. Wu, Z.,
Bentley Institute Press, Exton, PA.
Table 1. Average demands d at the various network nodes in the large and small
systems.
Node
d (L/s) in the d (L/s) in the
Large System Small System
1 1.38 0.14
2 0.18 0.02
3 0.92 0.09
4 2.57 0.26
5 2.83 0.28
6 8.36 0.84
7 3.42 0.34
8 5.33 0.53
9 1.52 0.15
10 0.86 0.09
11 2.27 0.23
12 1.31 0.13
13 8.52 0.85
14 0.88 0.09
15 0.22 0.02
16 0.39 0.04
17 0.45 0.04
18 2.82 0.28
19 0.03 0.00
20 0.31 0.03
21 2.60 0.26
22 1.03 0.10
23 1.00 0.10
24 0.53 0.05
25 0.76 0.08
26 0 0
27 0 0
Footnote: 1L/s = 15.8 GPM
Table 2. For network pipes, upstream and downstream nodes, length L and diameter
D in the large and small systems.
Pipe
Upstream Downstream
L (m) in the D (mm) in the L (m) in the D (mm) in the
Node Node Large System
Large System Small System Small System
1 1 2 10.1
100 3.2 100
2 2 3 2874.5
125 909.0 100
3 3 4 1732.8
150 548.0 100
4 1 16 2851.4
125 901.7 100
5 4 5 2648
200 837.4 100
6 5 7 144.5
200 45.7 100
7 5 6 364.9
200 115.4 100
8 7 10 817.4
150 258.5 100
9 6 13 1269.8
200 401.5 100
10 7 8 332.7
300 105.2 100
11 8 11 628.3
150 198.7 100
12 9 10 269.7
150 85.3 100
13 11 9 241.3
150 76.3 100
14 8 18 887.8
300 280.7 100
15 12 14 2055.9
150 650.1 100
16 13 12 130.9
250 41.4 100
17 21 13 991.1
250 313.4 100
18 14 15 6.8
200 2.2 100
19 15 16 607.2
150 192.0 100
20 15 17 1669.7
125 528.0 100
21 17 16 1046.8
150 331.0 100
22 18 21 132.1
300 41.8 100
23 18 19 392.5
450 124.1 100
24 19 20 154.5
450 48.9 100
25 22 23 2469.3
200 780.9 100
26 22 19 1593.6
250 503.9 100
27 24 23 2567
125 811.8 100
28 25 24 2337.7
100 739.2 100
29 17 25 2452.7
150 775.6 100
30 27 26 19.6
450 6.2 150
31 26 20 9.6
450 3.0 150
32 20 11 490.9
150 155.2 100
Footnote: 1 m = 3.25 ft
100 mm = 4 in.
Table 3. Purchase cost (€) of the various items for the PRV control in the large and
small systems, with DN350 and DN125 respectively (1 $ = 0.93 €).
Item Quanity large system
small system
Vault 5x10x6 ft 1
5,580
5,580
Gate Valves 3
856
180
PRV 1
9,431
1,649
Total
17,579
7,770
Table 4. Purchase cost (€) of the various items for the RTC control in the large and
small systems, with DN300 and DN100 (1 $ = 0.93 €).
Item Location Quanity
large system
small system
Vault 5x10x6 ft Control Node 1
5,580
5,580
Pressure Trasducer Control Node 1
800
800
Communications and Power to Site Control Node 1
1,500
1,500
Vault 8x14x6 ft Valve Site 1
7,440
7,440
Gate Valves Valve Site 3
702
113.22
Motorized Plunger Valve Valve Site 1
10,360
5,042
Communications and Power to Site Valve Site 1
1,500
1,500
RTU-PLC Valve Site 1
4,650
4,650
Total
33,937
26,851
Table 5. Daily supplied volume and leakage rates in the various scenarios considered
for the large system.
Demand
Pattern
Leakage
Level
Kind of
Control
Daily Supplied Volume
(ML/day)
Leakage Rate
(% of prod)
smooth low no control 4.52 3.3
smooth low PRV 4.47 2.3
smooth low RTC-TCV 4.46 2.2
smooth medium no control 5.14 15.1
smooth medium PRV 4.90 10.9
smooth medium RTC-TCV 4.88 10.6
smooth high no control 6.28 30.5
smooth high PRV 5.77 24.3
smooth high RTC-TCV 5.71 23.6
peaked low no control 4.51 3.2
peaked low PRV 4.49 2.8
peaked low RTC-TCV 4.46 2.1
peaked medium no control 5.13 15.0
peaked medium PRV 5.05 13.6
peaked medium RTC-TCV 4.88 10.6
peaked high no control 6.26 30.3
peaked high PRV 6.22 29.8
peaked high RTC-TCV 5.72 23.7
Footnote: 1 ML/day = 0.26 MGD
.
Table 6. Daily supplied volume and leakage rates in the various scenarios considered
for the small system.
Demand
Pattern
Leakage
Level
Kind of
Control
Daily Supplied Volume
(ML/day)
Leakage Rate
(% of production)
smooth low no control 0.4608 5.3
smooth low PRV 0.4519 3.4
smooth low RTC 0.4518 3.4
smooth medium no control 0.5209 16.2
smooth medium PRV 0.4903 11.0
smooth medium RTC-TCV 0.4899 10.9
smooth high no control 0.6297 30.7
smooth high PRV 0.5607 22.2
smooth high RTC 0.5594 22.0
peaked low no control 0.4605 5.2
peaked low PRV 0.4529 3.6
peaked low RTC 0.4516 3.4
peaked medium no control 0.5203 16.1
peaked medium PRV 0.4944 11.7
peaked medium RTC 0.4897 10.9
peaked high no control 0.6297 30.7
peaked high PRV 0.5710 23.6
peaked high RTC 0.5594 22.0
Footnote: 1 ML/day = 0.26 MGD
Table 7. Yearly number of bursts n
b
and yearly pipe repair cost C
burst
as a function of
maximum daily pressure head h
max
at the controlled node in the various scenarios
considered for the large system (1 €= 0.93 $).
Demand
Pattern
Leakage
Level
Kind of
Control
h
max
(m)
n
b
(-)
C
burst
(€)
smooth low no control 39.66
4.00 3,600
smooth low PRV 26.61
3.44 3,097
smooth low RTC 25.00
3.40 3,060
smooth medium no control 39.06
4.00 3,600
smooth medium PRV 26.94
3.46 3,116
smooth medium RTC-TCV 25.00
3.41 3,069
smooth high no control 37.12
4.00 3,600
smooth high PRV 27.48
3.52 3,172
smooth high RTC 25.00
3.44 3,100
peaked low no control 39.96
4.00 3,600
peaked low PRV 34.23
3.70 3,333
peaked low RTC 25.00
3.40 3,056
peaked medium no control 39.62
4.00 3,600
peaked medium PRV 35.44
3.77 3,395
peaked medium RTC 25.00
3.40 3,061
peaked high no control 38.09
4.00 3,600
peaked high PRV 37.29
3.95 3,556
peaked high RTC 25.00
3.43 3,084
Table 8. Yearly number of bursts nb and yearly pipe repair cost C
burst
as a function of
maximum daily pressure head h
max
at the controlled node in the various scenarios
considered for the small system (1 €= 0.93 $).
Demand
Pattern
Leakage
Level
Kind of
Control
h
max
(m)
n
b
(-)
C
burst
(€)
smooth low no control
39.91 2.00 1,800
smooth low PRV
25.46 1.70 1,533
smooth low RTC
25.00 1.70 1,528
smooth medium no control
39.85 2.00 1,800
smooth medium PRV
25.50 1.70 1,534
smooth medium RTC-TCV
25.00 1.70 1,529
smooth high no control
39.69 2.00 1,800
smooth high PRV
25.58 1.71 1,536
smooth high RTC
25.00 1.70 1,530
peaked low no control
39.99 2.00 1,800
peaked low PRV
27.62 1.73 1,559
peaked low RTC
25.00 1.70 1,528
peaked medium no control
39.97 2.00 1,800
peaked medium PRV
27.79 1.73 1,561
peaked medium RTC
25.00 1.70 1,528
peaked high no control
39.87 2.00 1,800
peaked high PRV
28.04 1.74 1,565
peaked high RTC
25.00 1.70 1,529
Figure 1. WDN layout.
0.0
1.0
2.0
3.0
4.0
0 4 8 12 16 20 24
Demand Multiplier (-)
t(h)
smooth peaked
Figure 2. Smooth and peaked demand patterns in terms of hourly demand multiplier.
Smooth Demand Pattern
20
25
30
35
40
0 4 8 12 16 20 24
h (m)
t(h)
no control PRV RTC
Peaked Demand Pattern
20
25
30
35
40
0 4 8 12 16 20 24
h (m)
t(h)
no control PRV RTC
Figure 3. Pressure head h [m] at the critical node in the large network in the case of
smooth and peaked demand patterns.
Smooth Demand Pattern
20
25
30
35
40
0 4 8 12 16 20 24
h (m)
t(h)
no control PRV RTC
Peaked Demand Pattern
20
25
30
35
40
0 4 8 12 16 20 24
h (m)
t(h)
no control PRV RTC
Figure 4. Pressure head h [m] at the critical node in the small network in the case of
smooth and peaked demand patterns.
Figure 6. Large system. Percentage cost reduction in the PRV and RTC scenarios
compared to the no valve scenario, as a function of the variable water unit cost.
Figure 7. Small system. Percentage cost reduction in the PRV and RTC scenarios
compared to the no valve scenario, as a function of the variable water unit cost.
... Konvansiyonel basınç kırıcılar ile uzaktan gerçek zamanlı kontrol edilen basınç kırıcılar kullanılarak şebekede uygulanmış ve sonuçları araştırılmıştır. Sızıntı seviyesi ile bakım ve işletim maliyetlerinin düşük olduğu bölgelerde aktif basınç kontrolüne ihtiyaç duyulmadığı, karmaşık ve büyük sistemlerde uzaktan kontrollü basınç kırıcıların kullanılmasının uygun olduğu ortaya konulmuştur [12]. Diğer bir çalışmada, basınç yönetiminin su kayıp azaltma yöntemleri içerisindeki en iyi uygulamalardan biri olduğu ortaya konulmuştur. ...
... Sızıntıların azaltılmasında bu faktörlerin kontrol altına alınması ve yönetilmesi oldukça önemlidir. Basıncın düzenlenmesi ve basınç dalgalanmasının azaltılması için uygulanan basınç kontrolünün mevcut arızalarda sızıntı hacminin ve yeni arıza oluşma riskinin azalması ve borunun ekonomik ömrünün uzaması gibi önemli katkılar sunmaktadır [5,12,21]. Doğrudan ölçümlere dayanarak farklı çapta borular ve koşullar için sızıntı ve basınç arasındaki ilişkiyi tanımlayan FAVAD yöntemini (Denklem (1)) önerilmiştir [3]. Pratik uygulamalar için FAVAD teorisinde, basınç ve sızıntı arasındaki ilişki N1 katsayısı ile tanımlanmıştır [3,22]. ...
Article
Operating conditions deteriorate depending on the density of failures in urban water management. The most important factor in the occurrence of failures is the high pressure in the system and the fluctuation in the pressure. Therefore, it is necessary to control the pressure and reduce the fluctuations in order to reduce the occurrence of new faults. However, pressure control is a costly process that requires extensive fieldwork. Therefore, before applying pressure management, the costs (equipment, field works, and installation) and benefits (potential water volume to be added to the system, potential reductions in failure) should be considered. The aim of this study is to develop a calculation tool to analyse the costs and benefits of applying pressure management in an isolated zone. For this purpose, cost components were defined according to field data. In addition, based on the approaches suggested in the literature, reductions in leakage and failure rates due to pressure control are determined. These fault and cost components are defined in the developed calculation tool. It is thought that this developed calculation tool will be a reference especially for practitioners. In this calculation tool, cost-benefit components are also defined for isolated site studies.
... Again, a tradeoff was discovered between cost and leakage water loss. Creaco and Walski [46] investigated the economics of leakage reduction through PRVs and RTCs. It was also found that a tradeoff exists. ...
Article
Full-text available
Since water distribution systems are so important to public health and many are leaking in unknown locations, a modeling study was performed to investigate the feasibility of installing pressure-reducing valves (PRVs) in various locations throughout several systems. A PRV was tried in each pipe, one by one, and the total cost (energy costs plus opportunity costs of losing water that could have been sold) was calculated. It was found that installing a PRV in the upstream pipes reduced costs the most and that putting a PRV in some pipes actually lost money due to the high cost of the PRV and associated fittings. Also, a PRV on the upstream portion of a large branch saved water leakage. Energy is saved when a PRV is placed near a pump for systems with high energy consumption.
... We expect that the relative importance of water-related features to system resilience will diminish if adequate leak detection and automatic isolation mechanisms are implemented. However, such mechanisms require high investments [43] and are seldom deployed in real-world distribution water systems. ...
Article
Full-text available
Climate change is increasing the frequency and the intensity of weather events, leading to large-scale disruptions to critical infrastructure systems. The high level of interdependence among these systems further aggravates the extent of disruptions. To mitigate these impacts, models and methods are needed to support rapid decision-making for optimal resource allocation in the aftermath of a disruption and to substantiate investment decisions for the structural reconfiguration of these systems. In this paper, we leverage infrastructure simulation models and Machine Learning (ML) algorithms to develop resilience prediction models. First, we employ an interdependent infrastructure simulation model to generate infrastructure disruption and recovery scenarios and compute the resilience value for each scenario. The infrastructure-, disruption-, and recovery-related attributes are recorded for each scenario and ML algorithms are employed on the synthetic dataset to develop accurate resilience prediction models. The results of the prediction models are analyzed and possible design strategies suggested based on the resilience enhancement attributes. The proposed methodology can support infrastructure agencies in the resource-allocation process for pre- and post-disaster interventions.
... Thornton and Lambert (2008) stated that pressure zones should be created and the most suitable PRVs should be selected for PM. Creaco and Walski (2017) stated that there is no need for PM in areas with low leakage levels and operating costs. In PM, the room construction, device and equipment selection, and placement and automation systems for monitoring data create significant costs. ...
Article
Full-text available
Methods and tools used to reduce leakage in distribution systems are often time consuming and costly and require special requirements. Therefore, cost-benefit analysis is very important for basic reduction methods applied in water loss management. In this study, cost and benefit analysis standards were developed for basic methods such as pressure management, number of teams, and pipe rehabilitation and active leakage control, in managing leakages. Moreover, a new cost algorithmic structure was developed and the economically recoverable water amount was determined by applying calculation tool developed to make detailed analyzes systematically and accurately. The most important advantage of this study is the development of an economic analysis model and algorithmic structure for basic reduction methods according to field data. It is thought that the cost analysis and algorithmic structures developed will make a significant contribution to the economic leakage level analysis and serve as a reference for sustainable water loss management.
... The net present value (NPV) analysis can be used to calculate the costs and savings of each leakage reduction option. The option having the lowest NPV is the least cost strategy, and the corresponding leakage defines the LR-ELL (Creaco & Walski, 2017;Fanner et al., 2007;. Moreover, the sustainable economic level of leakage (SELL) can be estimated by adding social and environmental costs and benefits (externalities) to the internal costs of the utility (Ashton & Hope, 2001;Howarth, 1998;Lambert et al., 2015;Smout, Kayaga, & Munoz-Trochez, 2010). ...
Article
Population growth, increased consumption, and climate change are impacting water production, and water shortage is already a reality in many countries. In addition, a portion of the water produced is lost throughout distribution. Pressure reduction can reduce losses, the frequency of bursts, and water consumption. Consumption reduction is generally considered beneficial to improve water availability, but it is rarely quantified. This work introduces a methodology that applies a statistical paired t-test to consumption data to evaluate the reduction. We applied it to eight district metering areas (DMAs) in the municipality of Palmas, Brazil. Four DMAs had the pressure reduced, while four did not. Two years of consumption data and 8,973 connections were used altogether. As a result, three of four DMAs showed a statistically significant consumption reduction after 1 year of pressure reduction. Among the DMAs where pressure remained unchanged, none showed a consumption reduction. We estimated a consumption reduction of 24,500 m3 for three sectors for the first year after pressure reduction.
Article
Full-text available
Water losses in water distribution systems (WDSs) cause inefficiency of water resources and increase operating costs. Water loss management (WLM) methods generally have high initial investment and operating costs. However, the budget planned within the scope of WLM in administrations is generally limited. Therefore, the most appropriate method should be determined by considering the budget conditions and cost–benefit analysis. The aim of this study is to propose a new economic water loss level (EWLL) model in WDSs with the different budget constraints of 5, 15 and 20% of the annual income. This EWLL model was developed by the discrete stochastic optimization algorithm. The EWLL and economically recoverable leakage volumes were determined by considering the budget constraints. Moreover, the most appropriate methods were determined to reach the EWLL values defined according to the budget constraints. The EWLL was calculated as 8.62% by the optimization model without budget constraints. Moreover, the EWLL values with budget constraints of 5, 15 and 20% of the annual income were determined as 56.82, 21.14 and 18.02%, respectively. This EWLL model will make a significant contribution to the annual planning in WLM depending on the budget constraints of the administrations.
Article
The inclusion of new strategies is crucial to achieve the different targets of the sustainable development goals for the guarantee of supply in the different cities and reduction the consumption of non-renewable resources. The development of these strategies implies the improvement of the sustainability indicators and green rating systems of the city. This research proposes a decarbonisation strategy, which includes different optimization procedures based on a self-calibration process according to recorded flow values over time. These stages are integrated into one tool to define the best making decision in the management of the supply system, analysing whether self-consumption of energy is feasible. It was applied on the Bahamas. The application of the strategy enabled the decrease of the annual consumption of energy equal to 32%. The self-consumption could represent 30% of the consumed energy of the pump station. The making decision to define the best operation strategy, establishing a Levelized Cost of Energy around 0.12 €/kWh when the feasibility of using photovoltaic systems combined with micro hydropower was done. It implies the reduction of 40% of the tCO2 emission, getting a cost of carbon abatement values around 400 €/tCO2 for different discount rates and scenarios.
Article
Full-text available
One of the major problems with urban water supply networks is the large volume of water wastage due to various incidents that cause damage. These damages are mainly irreversible and costly; thus, it is increasingly important to understand the causes of such incidents to prevent and reduce water wastage. This research aims to estimate the failure rate of main urban water supply pipes. In terms of nature, this research is applied, while in terms of administration, it is a combination of descriptive-analytical methods based on library sources. In addition to interviewing experts in the study area, the research uses findings of other research and data provided by Tehran’s Water and Wastewater Organization, which pertain to District 1. The research identifies major criteria affecting failure rates of main pipes of various types. Then, it uses Gene Expression Programming (GEP) software to perform computations and provide the model. Findings conclude that as the working pressure, age, and length of the pipe and the number of service lines increase, the number of incidents increases also. On the other hand, with the increase in the diameter of the pipes and the depth at which they are installed, their failure rates decrease. The type of soil also buried directly impacts the gradual loss of the pipes. Because the proposed models in this research have no limited input parameters and take into account all the effective components, they are highly useful and reliable tools to predict failure rates of the main urban water supply pipes.
Article
Full-text available
The potential of unsteady flow modelling for the simulation of remote Real Time Control (RTC) of pressure in water distribution networks is explored in this paper. The developed model combines the unsteady flow simulation solver with specific modules for generation of pulsed nodal demands and dynamic adjustment of pressure control valves in the network. The application to the skeletonized model of a real network highlighted improved capability of the unsteady flow simulation of RTC compared to the typical extended period simulation (EPS) models. The results show that the unsteady flow model provides sounder description of the amplitude of the pressure head variations at the controlled node. Furthermore, it enables identification of the suitable control time step to be adopted for obtaining a prompt and effective regulation. Nevertheless, EPS–based models allow consistent estimates of leakage reduction as well as proper indications for valve setting under network pressure RTC at a much smaller computational cost.
Article
Full-text available
Pressure management has been used for more than 3 decades to reduce leakage from water distribution systems. While few of these studies have been published, information on the ranges of field study leakage exponents is available. In contrast, several studies on the pressure-leakage relationship of individual leaks have been published, and verified models have been developed for predicting the response of elastically deforming leaks. The main aim of this paper was to determine whether researchers’ current understanding of the pressure-leakage response of individual leaks can be reconciled with the observed pressure-leakage response of district metered areas containing many leaks. To investigate this, a model of the distribution of individual leaks and their parameters was developed based on available literature and expert advice. A repeatability study showed that such a model can indeed produce typical distributions of leakage exponents found in field studies. A sensitivity analysis of the various parameters showed that the average system pressure and condition of the system had the greatest influence on the system leakage exponent. Finally, it was shown that a small fraction of leaks with high leakage numbers (i.e., in highly flexible material or closing under zero pressure conditions) can explain the high background leakage exponents often found in systems with only background leakage.
Article
In this paper, a novel field-oriented methodology to setup real-time control (RTC) for leakage reduction by pressure control valves in water distribution networks is presented. The paper introduces modalities to address the selection of proper RTC system architecture based on the network connectivity at the valve sites. Criteria for target node identification and RTC strategy selection in case of single-control (one valve–one target node) and multiple-control (multiple valve–one target node) architectures are developed. The impact on the control performance of controller calibration and communication protocol selection procedures, and of background noise in pressure signals is also explored. Then, developed criteria and procedures are applied to a Norwegian water distribution network in which a future field-pilot RTC system will be installed. Benefits in terms of pressure control effectiveness and water leakage reduction are evaluated by simulation under different control scenarios as a basic step to assess installation potentiality.
Article
Optimal management of water and energy resources worldwide is a basis for environmental and socioeconomic sustainability in urban areas, which has become even more relevant with the advent of the smart and water sensitive city paradigm. In water distribution networks (WDNs) water resource management is concerned with increased efficiency, which is primarily related to the reduction of leakages, whereas energy management refers to optimal pump, valve, and source scheduling strategies considering the hydraulic system requirements. These management goals require planning of asset renewal and improvement works in the short time (operational) and medium time (tactical) horizons, considering the financial sustainability of relevant actions. The battle of background leakage assessment for water networks (BBLAWN) was designed as a competition held at the 16th Water Distribution Systems Analysis Conference, in Bari (Italy) in 2014 (WDSA), to address the aforementioned management goals. The teams taking part in the BBLAWN were asked to develop a methodology for both reducing real water losses and saving energy in a real WDN considering the possibility of asset renewal and strengthening. Fourteen teams from academia, research centers, and industry presented their solutions at a special session of the WDSA 2014 conference. This paper briefly describes the BBLAWN and presents one of the solutions provided by the organizers to illustrate the ideas and challenges embedded in the posed problem. The overview of the solutions provided by the participants shows that management decisions need to be supported by engineering judgment and with tools that combine computationally effective multiobjective optimization and hydraulic models capable of assessing pressure-dependent background leakages.
Article
The advantages of pressure management (PM) on new break frequencies in water distribution systems and some important issues for infrastructure and energy management are discussed. Pressure management can be defined as the practice of managing system pressures to an optimum level of service ensuring sufficient and efficient supply to legitimate uses and consumers, while eliminating or reducing pressure transients and variations. There are many tools that can be used when implementing pressure management, including pump controls, altitude controls, and installation of pressure reducing and sustaining valves. The benefits achieved by the pressure management includes, reduction in the repair backlog, fewer emergency repairs, more planned works, and reduced inconvenience to customers. The range of operating pressures is moved away from the threshold pressure ad the break frequency is immediately reduced by reducing any pressure surges.
Article
Pressure management (PM) is commonly used in water distribution systems (WDSs). In the last decade, a strategic objective in the field has been the development of new scientific and technical methods for its implementation. However, due to a lack of systematic analysis of the results obtained in practical cases, progress has not always been reflected in practical actions. To address this problem, this paper provides a comprehensive analysis of the most innovative issues related to PM. The methodology proposed is based on a case-study comparison of qualitative concepts that involves published work from 140 sources. The results include a qualitative analysis covering four aspects: (1) the objectives yielded by PM; (2) types of regulation, including advanced control systems through electronic controllers; (3) new methods for designing districts; and (4) development of optimization models associated with PM. The evolution of the aforementioned four aspects is examined and discussed. Conclusions regarding the current status of each factor are drawn and proposals for future research outlined.
Article
A multiobjective approach is used here to optimize design and operation of the C-Town pipe network, searching for trade-off solutions between (1) installation cost, (2) operational cost, and (3) cost of the pressure-reducing valves. Due to the large number of decisional variables and to the complexity of the constraints considered, the optimization problem was tackled in five steps: (1) identification of some feasible (on the basis of the many constraints) first attempt solutions; (2) application of a multiobjective genetic algorithm to the 2D optimization problem with objective functions 1 and 2, in order to obtain optimal trade-off solutions between the installation cost and operational cost, without considering the installation of pressure-reducing valves; (3) application of the multiobjective genetic algorithm to the optimization problem with objective functions 2 and 3 for each of the solution selected at the end of Step 2, in order to assess how the operational cost can decrease thanks to the installation and operation of pressure-reducing valves; (4) derivation of the 3D Pareto surface by grouping the solutions found at the end of Steps (2) and (3). A solution was extracted from the 3D Pareto surface of optimal solutions following some specific criteria. This solution was then further refined (Step 5) in order to allow for variable settings of the pressure-reducing valves installed and to make it compliant with the battle guidelines concerning leakage modeling.
Article
Pressure control entails the cheapest technical solution to achieve leakage reduction in water distribution networks in shortmedium time horizon. This work reports the planning of remote real time controlled (RRTC) pressure reducing valves (PRV) for the Oppegård (Norway) hydraulic network. It was achieved by using an advanced hydraulic model which integrates the pressure-dependent background leakage model with the simulation of PRVs based on remote control nodes. Results demonstrate that RRTC PRVs instead of existing locally controlled ones permits the reduction of the background leakages of about 35%. This rate increases over 40% if few additional RRTC PCV are installed.
Article
This paper presents a knowledge-based optimization model for minimizing leakage in water distribution systems through the most effective location and setting of control valves. The optimization model is based on the use of a genetic algorithm (GA), which is known for its robustness for handling the optimization of water distributions networks. Knowledge of the water distribution system has been incorporated into the optimization model to reduce the search space and to enhance the model efficiency in finding the optimal solutions. A single optimization model to search for both the optimal location of valves and their settings has been used. These two improvements over other models have enhanced the efficiency of the model and improved the search for the optimum solution. The model has been used to test leakage minimization in a benchmark water network studied by other researchers and has proven to be very efficient and robust.
Article
This paper shows how pipe replacements and control valve installations can be optimized in water distribution networks to reduce leakage, under minimum nodal pressure constraints. To this end, a hybrid multiobjective algorithm, which has pipe diameters and valve positions and settings as decisional variables, was set up. The algorithm also enables identification of the isolation valves that have to be closed in order to improve effectiveness of the control valves installed. The algorithm is initially applied to the optimal valve location problem, where it explores the trade-off between the number of installed control valves and the daily leakage volume. In this context, the analysis of the results proves the new algorithm more effective than a multiobjective genetic algorithm widely adopted in the scientific literature. Furthermore, it shows that if some isolation valves identified ad hoc are closed in the network, the installation of control valves determines larger leakage volume reductions. In a second application of the algorithm, pipe replacements and control valve installations are simultaneously performed. In this case, a Pareto front of trade-off solutions between installation costs and daily leakage volume is obtained. For the choice of the final solution within the front, an economic criterion based on the long-term convenience analysis is also illustrated.