Accepted Manuscript

Leak Detection in Low-Pressure Gas Distribution Networks by Probabilistic Methods

Payal Gupta, Thaw Tar Thein Zan, Mengmeng Wang, Justin Dauwels, Abhisek Ukil

PII: S1875-5100(18)30323-8
DOI: 10.1016/j.jngse.2018.07.012
Reference: JNGSE 2657

To appear in: Journal of Natural Gas Science and Engineering

Received Date: 4 April 2018

Revised Date: 11 July 2018
Accepted Date: 15 July 2018

Please cite this article as: Gupta, P., Thein Zan, T.T., Wang, M., Dauwels, J., Ukil, A., Leak Detection
in Low-Pressure Gas Distribution Networks by Probabilistic Methods, Journal of Natural Gas Science &
Engineering (2018), doi: 10.1016/j.jngse.2018.07.012.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT
Leak Detection in Low-Pressure Gas Distribution Networks by Probabilistic Methods I

Payal Guptaa,∗, Thaw Tar Thein Zana , Mengmeng Wanga , Justin Dauwelsb , Abhisek Ukilc ,

a Energy Research Institute(ERI@N), Nanyang Technological University, Singapore

b School of Electrical and Electronics Engineering, Nanyang Technological University, Singapore
c Electrical & Computer Engineering, University of Auckland, New Zealand

Abstract

PT
The presence of leaks is a prevalent issue for aging gas distribution systems across the globe. These events, if not detected in time,
may bring about environmental and health hazards, besides economic losses. Therefore, the development of efficient detection,
quantification, and localization methods is crucial to all gas companies worldwide. In this paper, we present a leak monitoring
system, called Leak Analytics System (LAS) using a probabilistic approach to determine the location and the rate (severity) of

RI
leakage in low-pressure gas distribution networks. This work aims to develop a robust, cost-effective, and real-time online moni-
toring system for low-pressure gas distribution networks. The leakage events are estimated using pressure and flow data obtained
from steady-state modeling of the gas network. The robustness of the methodology is illustrated by analyzing gas networks in
the presence of measurement errors, which account for unavoidable sensor noise in flow and pressure data. The feasibility of the

SC
proposed method is demonstrated on a small artificial gas network. Moreover, the method is applied to a section of the Singapore
gas distribution network for a single as well as multiple leak scenarios. It is also experimentally shown that the severity of the leak
and the location for a single leak scenario can be determined within an accuracy of 95% and 80% respectively, even in the presence
of strong noise.

U
Keywords: Leak detection, Localization, Severity, Gas distribution network, Pressure, Flow.
AN
Nomenclature distribution networks. Transporting networks operate at very
high pressure (>1.6MPag) which transport gas from coastal
n̂` Number of nodes in leak zone supplied to regional areas. On the other hand, distribution net-
A Demand junction link incidence matrix works operate at medium (50kPag-0.3MPag) to low-pressure
M
B Link loop incidence matrix (0-50kPag) and distribute gas from regional areas to end cus-
D Inner diameter of the pipe (mm) tomers.
g Specific gravity of gas
L Pipe length (m) The failure of any of the pipeline network system leads to
D

m Number of branches (pipes) not only the loss of natural resources but also causes serious
n Number of nodes environmental impact and threat to public safety. Therefore,
n` Number of dummy nodes (zero demand nodes) it is essential to manage pipeline risk, prevent and mitigate
TE

P∗ Pressure at pressure regulator node the consequences of any failures (Muhlbauer, 2004) such as
Pi Pressure at the node i leaks. Several approaches have been applied to identify pos-
q Flow demand vector at the outlet nodes sible pipeline failure initiators and subsequently perform risk
Qj Gas flow inside the pipe j assessment (Veritas, 2010; Hopkins et al., 2009). Risk assess-
ment analysis for pipeline failures are based on qualitative and
EP

ri Residual at monitoring location i

S Node sensitivity matrix quantitative methods (Han and Weng, 2011; Jo and Ahn, 2005)
such as probability and statistics methods, analytic hierarchy
methods, fault tree analysis, event tree analysis, etc. Accord-
1. Introduction ing to (Pennenergy, 2016), most systems that constitute these
C

Pipeline networks are one of the vital and economically ef- pipeline networks across the globe are between 30 and 100
fective modes of transportation of natural gas. These networks years old. Hence, aging and lack of maintenance have increased
the risk of failures such as leaks in the pipes (especially at the
AC

are complex system and continuously expanding with hundreds

or thousands of kilometers of pipelines every year. Gas pipeline
networks are mainly classified into transporting networks and
I This work was supported by the energy innovation programme office
joints), which if not detected in time can cause substantial fi-
nancial loss and can lead to major accidents. These pipes are
therefore propelling the need for more reliable leak detection
across the pipeline industries worldwide.

(EIPO) through the national research foundation and Singapore energy market
authority. Project LA/Contract No.: NRF2014EWT-EIRP003-002.
∗ Corresponding author
Email addresses: pgupta@ntu.edu.sg (Payal Gupta),
tttzan@ntu.edu.sg (Thaw Tar Thein Zan), mengmeng@ntu.edu.sg
(Mengmeng Wang), jdauwles@ntu.edu.sg (Justin Dauwels),
a.ukil@auckland.ac.nz (Abhisek Ukil)
Preprint submitted to Journal of Natural Gas Science and Engineering
Due to maturity of transporting networks, several leak detec-
tion techniques have been developed for its continuous moni-
toring (Billmann and Isermann, 1987; Zhang, 1997). (Ma et al.,
2010) applied the negative pressure wave method that exploits
the pressure waves generated by leaks while (Rocha, 1989) ex-
plored the appearance of acoustic pressure waves caused by
August 7, 2018
 ACCEPTED MANUSCRIPT

Table 1: Current leak detection approaches and their applications.

Model Approach Application Advantages Disadvantages
Negative Pres- Pressure waves generated by Transporting networks, Easy to implement as no Leaks need to be quick and abrupt for
sure wave leaks are propagated as wave medium pressure net- mathematical model is re- detection, not applicable on long-range
and can be recorded using pres- works quired pipelines and slow leaks
sure transducers
Pressure point Presence of leak identified Transporting networks, Easy maintenance, Cost- Application too narrow as it is difficult
analysis when mean value of the pres- medium pressure net- effective, detect small to localize leak points. Not able to iden-
sure decreases under prede- works leaks tify multiple leaks
fined threshold
Acoustic sensor Adopts sound generated by Transporting networks, High detection and local- High sensor cost as many sensors re-
based analysis leaks as signal source for detec- distribution networks ization accuracy, easy to quired for long distance detection

PT
tion and localization implement as no mathe-
matical model is required
Mass/Volume Mass balance equation Transporting networks, easy principle Sensitive to arbitrary disturbances and
balance presents systematic devia- medium pressure net- dynamics of the pipeline leading to high
tion when leak occurs works false alarms
Transient based Conservation of mass, momen- Transporting networks, Can detect small leaks Sensor cost is high and model is com-

RI
methods tum and energy are utilized for medium pressure net- plex and require trained users
leak detection works
Distributed Presence of leaks causes a tem- Transporting networks, Fast response and high ac- Cost of implementation very high, diffi-
temperature perature change which is de- distribution networks curacy for leak location cult of retrofitting to existing pipes

sensing tected by fibre optic cables
Probabilistic Bayesian probabilistic ap-
based method proach is used for leak
detection and localization
SC
distribution networks High accuracy and small High computation cost, not able to iden-
sample data is required. tify multiple leaks
Cost-effective

leaks. Pressure point analysis (PPA) (Bin Akib et al., 2011),
another popular and cost-effective approach, detects leaks by
comparing the current pipeline pressure values to earlier val-
ues. (Wan et al., 2011) performed the detection as well as the
localization of leaks via wavelet transform technology for sig-
nal pre-processing and average-weighted localization scheme
for determining the leak location. However, the focus of the im-
M
plementation is limited to linear pipelines, without considering
more complex pipeline networks. Other common approaches
include frequency response based methods (Mpesha et al.,
2001; Askari et al., 2016), mass/volume balance (Dos Santos
D
U
et al., 2008). However, such methods require intensive human
AN involvement and lack real-time monitoring. Therefore, recent
works on the low-pressure networks focus on real-time moni-
toring utilizing specialized sensors to assist the detection and
the localization of leaks. Typical sensors may include dis-
tributed temperature sensors (Tanimola and Hill, 2009; Cam-
panella et al., 2016; Ukil et al., 2012), acoustic sensors (Karku-
lali et al., 2016), and pressure/flow sensors (Reddy et al., 2016).
However, direct utilization of the sensor data for leak detec-
tion and localization may result in false alarm problems and
faulty diagnostics. This is because the sensors are affected

et al., 2011), steady-state or transient models (Hauge et al., by background noise and other environmental interference.
2007). In (Ebrahimi-Moghadam et al., 2016), a two-dimensional, tur-
bulent and compressible gas flow model is developed to inves-
TE

All the above existing leak detection methods are applica-

tigate leak in above as well as underground gas distribution
ble on transporting networks because the fluids are operated
pipeline. However, this work is limited to a single leak and
at the extreme conditions such as high-pressure, high-flowrate,
the gas pressure is assumed to be in medium-pressure range
etc. where the presence of leaks are easily detected. Further-
i.e., 3 - 5 bar. In (Rui et al., 2017), a mathematical model is
more, in order to reduce the complexity, most leak detection
EP

developed for detecting two leaks in a single pipeline. How-

problems are reduced to a single-pipe problem. However, due
ever, in practice, leaks can occur at more than two locations
to the complex, branched and/or looped structure of distribution
on the same pipeline, or at different pipelines within the net-
networks, it is generally not possible to limit the detection prob-
work. A probabilistic approach is proposed in (Gupta et al.,
lem to a single pipeline, as in the case of transporting pipelines.
2016) to identify faulty pipe segments and estimate the leakage
C

Therefore, distribution networks need to be modelled holisti-

rate in the gas distribution systems. The advantage of the pro-
cally, as entire systems. Moreover, the proximity of distribu-
posed Bayesian probabilistic approach is its high accuracy even
tion networks to the urban infrastructure increases the failure
AC

with small sample size (as limited leak data is available from
probability of distribution gas pipelines compared to transport-
network operators). Additionally, uncertainties such as sen-
ing pipelines, and makes distribution systems more vulnerable
sor noise, demand variations, etc., are easily handled by proba-
to accidents. Hence, there is a need to develop robust and cost-
bilistic approaches (Uusitalo, 2007). However, the localization
effective schemes for detecting and localizing gas leaks in low-
method is also limited to a single leak, and the author assumes
pressure gas distribution networks.
that the sensors are available at all nodes. The presence of a
Currently, most distribution networks operators follow the large number of exit points at domestic consumers makes it too
conventional methods for leak inspection. Conventional costly to install and deploy sensors at all exit points. Table 1
leak detection methods involve personnel patrolling along the summarizes all the above-mentioned approaches.
pipeline looking for specific sounds, visual effects or odors that
may be released during a leak (Murvay and Silea, 2012; Liu
2
 ACCEPTED MANUSCRIPT

1.1. Our Contribution The demand junction link incidence matrix Ai is defined as:
In the view of the issues above and the lack of robust monitor-



+1, if gas in branch j enters node i
ing techniques for localizing leaks particularly in low-pressure 
ai j = 

−1, if gas in branch j leaves node i

distribution networks, we propose a leak monitoring system, 


 0, if branch j is not connected to node i .

referred to as Leak Analytics System (LAS). The methodology
involves steady-state modeling of the gas distribution network, The pressure values need to fulfil the energy conservation
combined with a probabilistic approach for leak location and law. The latter can be expressed as the following matrix equal-
severity estimation. Our work can be seen as an extension of ity:
the study of (Gupta et al., 2016) from single to multiple leaks. B∆P = 0, (2)
Our proposed method is able to handle multiple leaks. It does where the link loop incidence matrix B is defined as:
so by a divide-and-conquer approach: the distribution network

PT

is divided in leak zones, which are analyzed separately. In this



+1, if branch j is in loop i and the direction is same
bi j = 

−1, if branch j is in loop i and the direction is opposite

way, the computational complexity can drastically be reduced. 

To divide the network into leak zones, we introduce leak iso-

 0, if branch j is not in loop i ,

lation step for cluster (leak zone) formation, comprising nodes

RI
and ∆P is the unknown node pressure vector in the network:
with high probability of having leaks. In our work, we assume
h iT
that only a few sensors are available, which is more realistic ∆P = ∆P1 ∆P2 ... ∆Pm . (3)
and practical. Also, to illustrate the reliability of our method-
ology, we test the approach in a real gas distribution network. For low-pressure (0-75kPag) gas networks, high-pressure flow

SC
We experimentally show that for a single-leak scenario, the leak equations (Coelho and Pinho, 2007) have been mathematically
locations and the severity can be estimated with an accuracy of reworked in order to ease up calculations with following as-
80% and 95% respectively, while for a multi-leak scenario the sumptions:
localization accuracy remains above 75% in the presence of low 1. steady and isothermal flow of gas,

U
levels of sensor noise. 2. compressibility factor of gas to be one,
The rest of the paper is organized as follows. Section 2 3. neglecting potential energy term,
presents the mathematical description of gas distribution net- 4. constant friction coefficient over the length of pipe which
AN
works at steady state. Section 3 and Section 4 illustrate the leak is given by:
localization and leak isolation stages respectively in the mathe- !
12
matical model for LAS. We present the experimental results and f = 0.0044 1 + . (4)
discussion based on a synthetic network and Singapore gas dis- 0.276D
M
tribution network in Section 5, followed by concluding remarks With all the above stated assumptions, Lacey’s flow equa-
in Section 6. tion (Segeler, 1968; Abeysekera et al., 2014) is derived for low-
pressure (0-75kPag) regime from high-pressure flow equation
and is given by:
2. Modeling Gas Distribution Networks
D

s
(P1 − P2 )D5
The gas distribution networks can be viewed as a combi- Q = 7.1 × 10 −3
, (5)
nation of nodes (pressure regulators, valves, customer outlets, gL
TE

etc.) and link components (gas pipes) with varying lengths and where the specific gravity of the gas g is 0.589. The nodal pres-
diameters, aimed at delivering gas from the supply stations to sure P is in mbar and the flow rate Q is in m3 /h. Eq. (5) can
the customers. Pressure regulators or alternatively known as also be re-written as:
district regulators reduce the pressure of the gas from high-
∆P = P1 − P2 = KQ2 , (6)
EP

pressure distribution network and maintain the preset outlet

pressures. Another element of the distribution networks is the where the pipe coefficient K is defined as:
valves and junctions where two or more pipelines interconnect.
K = 11.7 × 103 LD−5 . (7)
In this work, the valves and junctions are considered ideal, in
other words, they do not cause any additional pressure drop. The inlet node of the distribution network is the gas reduction
C

In order to analyze the state of gas distribution networks, we station or the pressure regulator node, which converts high-
conduct steady-state simulations on the corresponding graphs, pressure gas to low-pressure. The sole purpose of the pressure
AC

yielding the flow and pressure information. The flow and pres-
sure values within the network can be determined from the
principle of mass and energy conservation (Osiadacz, 1988).
The flow values need to fulfil mass conservation at the junction
nodes, which can be expressed in matrix form as follows:
Ai Q = q, (1)
where Q is the standard flowrate values for the m pipes and can
be written as:
h iT
Q = Q1 Q2 ... Qm .
3
reduction station is to keep the output pressure near a certain
pressure level P set , which in our case is 5kPag for artificial
network while 2kPag for Singapore network. The gas flow-
ing through the pressure regulator nodes (denoted by P∗ ) can
be modeled by the following relation:
P∗ = P set , P∗ > P set , (8)
The above network formulation given by eq. (1), (2), and (6)
is solved iteratively until convergence with (8) as the constraint
set to obtain the flow values inside pipes and pressure values at
the nodes (Krope et al., 2011).
 ACCEPTED MANUSCRIPT
3. Leak Localization System Design
3.1. Probabilistic Framework
Our approach to leak localization is rooted in identifying
the most probable leakage locations and estimating the leakage
rates. The probabilistic framework provides a rigorous method-
ology to quantify modeling uncertainty and perform system
identification (Beck, 2010; Beck and Katafygiotis, 1998). The
probability of a model M is a measure of its relative plausibil-
ity within a set of models denoted by M and parametrized by
parameter set θ. In our system, the parameter set θ is defined by
the location θL of the leaks, and gas leakage rate (severity) θF
of the leak. Hence, by definition θ = (θL , θF ).
Considering an existing monitoring system installed in the
network at N monitoring locations, pressure and flow data are
measured for T test cases. Let y = [yi j ]N×T denote the data
matrix measure at N monitoring locations and T test cases
where yi j is the vector comprising of pressure and flow val-
ues at ith monitoring location during jth instance. Similarly,
flow and pressure quantities, denoted by ỹ(θ) = [ỹi j (θ)], can be
obtained from solving the model M(θ) corresponding to a par-
ticular value of parameter θ. As a consequence, the deviation
between the model values ỹi j (θ) and actual data yi j is given by:
ei j (θ) = yi j − ỹi j (θ).
This deviation may be caused by modeling errors or the
unavoidable sensor noise while taking measurements. We
model these uncertainties by means of a Gaussian distribution
parametrized by its mean and standard deviation σ, where we
assume the mean to be zero. By Bayes theorem, the probability
distribution of the parameters θ and σ is obtained as:
p(y|θ, σ)π(θ, σ)
p(θ, σ|y) =
D
, (10)
p(y)
where the term 1/p(y) is a constant for given data. Quantifi-
TE
cation of the prior π(θ, σ) and the selection of parameter un-
certainty are done based on the past experiences. Here we as-
sume that the information regarding the prior distribution of the
model parameters is not provided and thus, we can opt for a
uniform distribution of initial PDF π(θ, σ) over a range of ac-
EP
ceptable values of [θ, σ].
The posterior pdf (10) provides the relative plausibility of
the models based on the observed data y. The prediction er-
ror ei j (σ) is assumed to be Gaussian white-noise with standard
C
deviation σ independent of the locations in the network.
With all the above stated assumptions, the likelihood
AC
p(y|θ, σ) in eq. (10) can be written as follows:
T Y
N
(yi j (θ) − yi j )2
" #
Y 1
p(y|θ, σ) = √ exp − . (11)
i=1 j=1 2πσ 2σ2
From eq. (10) and eq. (11), the probability distribution on leak
parameters given sensor data y is obtained by:
T
"
P
k(yi j (θ) − yi j )k2 #
1 i=1
p(θ, σ|y) ∝ √ exp −
( 2πσ)T N 2σ2
3.2. Parameter Estimation
To determine the most probable value of leakage parameters
from the distribution (12), the optimal parameter value (θ̂, σ̂)
is calculated. This is done by maximizing the posterior term
p(θ, σ|y) (12), or equivalently, by minimizing the negative log
likelihood function f (θ, σ):
f (θ, σ) = − ln p(θ, σ|y)
T
1 X
= k(yi j (θ) − yi j )k2 + T N ln σ + c. (13)
2σ2 i=1
PT
Minimizing f (θ, σ) w.r.t. σ for a given θ leads to:
T
1 X
σ̂ = J(θ̂) = min
2
k(yi j (θ) − yi j )k2 . (14)
RI
θ TN
i=1
Finally, invoking the total probability theorem, the uncer-
tainty in the model parameters (given the measured data y) is
SC
given by (Katafygiotis et al., 1998):
p(θ|y) = c[J(θ)]−(T N−1)/2 . (15)
4. Leak Isolation Problem
U
In this work, we assume that the potential leak points ac-
(9) count for the pipe junctions, faulty flanges, etc., which con-
AN stitute the nodes of the network with zero demand. We refer
these nodes as dummy nodes. This is a standard assumption
in model-based leak detection and location literature in large
scale networks (Ponce et al., 2014; Preis et al., 2010). This as-
sumption reduces the complexity of the leakage model without
M
affecting the final results. Further, the leaks are simulated in the
network by adding an additional consumption/ demand value at
assumed leak nodes which account for leakage rate (severity).
To demonstrate the reliability of the approach for all leak sizes,
we generate the leakage rates at random between 0.1%-10% of
the total gas flowing inside the network.
To identify the most probable locations of the leaks, the leak
localization problem needs to be solved for all possible com-
binations of leaks in the network. For a given network with
n` potential leak points,
  the number of possible combinations
of leak scenarios is nk` , where k denotes the number of leak
points. This number is significantly increased in the case of
multiple leaks (k > 1) compared to single leak case (k = 1).
n  
k = 2 .
P̀ n` n`
Hence, the complexity of this problem becomes
k=1
This leads to a significant increase in the computational cost and
makes the procedure impractical for large networks. To reduce
the complexity of this leak localization problem, we propose a
leak monitoring system, called Leak Analytics System (LAS);
the system can be separated into three stages: leak detection,
leak isolation, and leak localization.
The fault detection and isolation stages in LAS are achieved
through the evaluation of residuals between measurements and
model outputs (Nejjari et al., 2015). For a given set of M pres-
sure monitoring locations, the residual ri is the difference be-
tween the actual pressure measurement vector p and the esti-
mated pressure values p̂ in leak free case:
. (12) ri = pi − p̂i . (16)
4
 ACCEPTED MANUSCRIPT

PT
Fig. 1: System architecture for the three stage Leak analytics system for leak localization in gas distribution networks.

Algorithm 1 Leak analytics system for leak localization
Input: Current sensor measurements from the network (y) and leak-free measurements ŷ;
Network model.
Output: A near-optimal leakage rates and locations
RI
the correlation index. To form this cluster, the node with the
highest correlation value is chosen as the centroid node, and all
its neighboring nodes within a threshold distance form the leak
prone zone (cluster). The threshold distance depends on the size

SC
1. Residual ← y - ŷ //cf. eq. (16)
2. while Residual > Threshold do of the network or the number of nodes in the network. For our
centroid node ← max (corr(S i , y))//cf. eq. (18) analysis, we consider this distance to be twice the average pipe
1≤i≤NL
Leak zone ← all dummy nodes within threshold distance from centroid node. distance of the network.
for θL = 1, 2, . . . nodes in leak zone do Pressure sensor placement
102
restrict ← SetRestrictions (0 < θF < 10% total flow) Leak location

U
100 P1 101 P2
Golden section/ Nelder-Mead based search. Flow sensor placement
P9
while An optimization criterion is not reached do 120 P3
P8 P6
θF ← getConfig() 123 P10
105 P4
J(θ) ← eval J(θ) //cf. eq. (14)
end while
AN P13 122 P12 121
103

P32
P7 104
P5
106

P11 P31
Find (θkL , θkF ) such that J(θk ) = minθF J(θ) P15 P30 135 136 P51 P36
end for 124 134 P33
P14 108 P35 109
Find (θ̂L , θ̂kF ) such that J(θ̂) = minθL J(θk ) 107 P34
Correct leaks in the network and take measurement data y again. P16 P20
126 P37
3. end while P17 P18 P43
M
125 127 P46 118
P29 115
110 P44 113 P42
P21 P19 P25 P38
P22
The aim of these steps is to detect the presence of leaks and iso- 128
P24 130 P45
P41
129 117
P39
late the leak-free nodes of the network respectively and form a P47
D

P23 P26 P50 116

114 P40
subset of possible leak nodes n̂` ∈ n` before carrying out leak P49
P28
133 119
localization. We implement a threshold-based test to cope with 132 137 P27 Inlet
131 111 P48 112 5kPa
noise and model uncertainty effects. In the absence of faults,
TE

all residuals remain below the given thresholds while in the

presence of leaks, some residuals will exceed the correspond-
ing thresholds, signalling the occurrence of a leak. Once the
presence of leaks is identified, we determine the effect of leaks
EP
Fig. 2: Synthetic gas distribution network layout with 50 pipes and gas inlet
source.
After the leak isolation stage, we apply the leak localization

on every available pressure sensor measurement by means of
sensitivity analysis. For a network with n` possible faults (leaks
in nodes), we define the sensitivity matrix S as:
p̂i j − p̂i
si j = , i = 1, . . . , M; j = 1, . . . , n` ,
C
method (discussed in Section 3) to obtain the fault locations in
the leak zone of the network. The leak localization stage in-
volves inferring two parameters: leak location θL and leakage
rate θF . By means of the probabilistic framework, we estimate
the most probable leakage rate θ̂(i)
L of the parameter set θ for
(i)
(17)

fj each possible leakage scenario i. This forms a series of contin-

where p̂i j is the pressure estimate at node i due to leak value f j uous optimization problems which involve solving a non-linear
AC

occurring at node j and p̂i is the pressure estimate at node i in
the leak-free scenario. Leak isolation is performed by analyz-
ing the correlation between residual vector (16) and sensitivity
matrix (17) (Pérez et al., 2011). The same analysis may also
be preformed by means of different metrics. For each set of
possible leaks, the correlation is obtained as:
s jT r
Γj = , j = 1, . . . , n` .
ks j kkrk
The objective of this analysis is to form a leak prone zone
by clustering the highly probable leak nodes n̂` ∈ n` based on
system of pipe network equations at steady-state for all possi-
ble leak positions in the network. We solve the above system in
MATLAB by golden section search and parabolic interpolation
based optimization methods (Brent, 2013; Forsythe et al., 1977)
for single leak scenario. For the case of multiple leaks, we use
Nelder-Mead based optimization approach. Next, we calculate
the probability of occurrence pi for each possible leakage event
(18) i with the corresponding estimated leak severity θˆiL :
pi = Kc p(θ(i) , θ̂(i)
L |y), (19)
where Kc is a normalization constant. The leak combination
5
 ACCEPTED MANUSCRIPT

with the highest probability pi constitutes the most probable
leak locations, and their corresponding severity values are the
rate of gas outflow from the leaks. Algorithm 1 and Fig. 1 il-
lustrate all the stages of the proposed methodology. We repeat
these steps until the residual values (R) become below threshold
and no more leaks are detected in the network.
5. Experimental Results and Discussion
To test the leak localization method, we first apply it to a syn-
respectively. We assume a total of 12 sensors (6 pressure and 6
flow meters) to be placed for monitoring the network (shown in
Fig. 2). We selected the location of the sensors randomly. We
provide the pressure and flow values at the monitoring loca-
tions in case of no leak in the network in Table Appendix A.1.
As mentioned, we assume that leak points occur at the pipe
junctions, faulty flanges, etc., which constitute the nodes of the
network with zero demand. We indicate these nodes by aster-
isk in Table Appendix A.2. We assume the remaining nodes as
customer outlet points with gas demand 30m3 /h.

thetic network (see Fig. 2), and next to a section of the gas dis-
tribution network in Singapore (see Fig. 3a). First, we consider
a single leakage scenario followed by the occurrence of multi-
ple leaks. Leakage rates vary between 0.1% - 10% of the total
gas supply. We generate synthetic measurements yi j according
PT
5.1.2. Singapore Gas Distribution Network
Next, we consider an actual gas pipeline network, currently
operating in Singapore to distribute town gas from supply sta-
tions to consumers. The gas distribution network in Singapore

to the model described in Section 3. Specifically, the measure-
ment vector yi j is a vector set of pressure and flow values at the
sensor locations obtained from the network monitoring system.
RI
provides town gas to residential and commercial consumers,
and constitutes more than 3400 km of pipelines. The en-
tire network is mainly underground and operates at three pres-
sure regimes: high-pressure (28–40bar), medium-pressure (3–

SC
5.5bar) and low-pressure (50 / 20 / 2kPag) (Singapore Power,
2016). Here we investigate a 3 km long section of low-pressure
(2kPag) pipeline network that covers the residential area in the
central region of Singapore. For this network, the average dis-

U
tance and the minimum distance between two adjacent nodes
AN are approximately 20m and 1m respectively. For security rea-

1E-10

1E-20

1E-30
Normalized Probability

1E-40
M
1E-50

120 121 122 123 124 125 126 127 134

(a)

(a) 1E-10

1E-20
D

1E-30

1E-40

1E-50
TE

1E-60

77 78 79 80 81 82 83 84 85 86 87 90

Node ID

(b)

Fig. 4: Results for probable leak location w.r.t. node IDs for (a) artificial net-
work and (b) Singapore network with zero noise. X-axis represents the node
IDs of the nodes inside the leak zone.
EP

(b)
1.0

Fig. 3: Gas distribution network in Singapore; (a) extracted network overlaid 0.8

on an aerial view (b) section of Singapore gas distribution network where pipes 0.6

are made of ductile iron.

C

0.4
Normalized Probability

0.2

5.1. Gas Networks 0.0

120 121 122 123 124 125 126 127 134

AC

5.1.1. Simulated Network (a)

1.0

To illustrate the applicability of our approach, first we con- 0.8

sider a small example network of 38 nodes, 50 pipes, and a gas 0.6

inlet source (see Fig. 2). (Osiadacz et al., 1995) and (Djebed- 0.4

jian et al., 2008) considered a similar network for testing op- 0.2

timization algorithms for gas distribution networks. The net- 0.0

77 78 79 80 81 82 83 84 85 86 87

work is assumed to operate in the low-pressure regime (0-70 Node ID

mbar). The pipe diameter is either 4, 6, 8 or 12 inches (1 inch = (b)

25.4 mm). We provide the network specifications in Table Ap- Fig. 5: Results for probable leak location w.r.t. node IDs for (a) artificial and
(b) Singapore network in the presence of noise. X-axis represents the node IDs
pendix A.2. The average distance and the minimum distance
of the nodes inside the leak zone.
between two adjacent nodes are approximately 100m and 50m
6
 ACCEPTED MANUSCRIPT
100

Average Distance
25

accuracy (%)
Leak location
90 100

Leak Location

Accuracy (%)
20

Error (m)
80 90

15
70 80

60 70 10

50 60 5

0.0 0.5 1.0 1.5 2.0 50 0

0.0 0.5 1.0 1.5 2.0

estimation error (%)

4
16

Estimation Error (%)

3
Leak rate

12

Leak Rate
8 2

4 1

0
0
0.0 0.5 1.0 1.5 2.0
0.0 0.5 1.0 1.5 2.0

PT
Sensor Noise (%)
Sensor Noise (%)

(a) (b)

Fig. 6: Accuracy of estimated leak location and mean absolute error of estimated leak severity w.r.t. noise value with 95% confidence interval for single leak scenario
in (a) artificial network and (b) Singapore network.

RI
sons, GIS data for the network was not provided. Therefore, in the gas networks. We add zero-mean Gaussian noise to
we extract the network skeleton from the available RGB im- the sensor measurements, independently at different locations.
age of the layout by applying a color based image segmentation For low-pressure distribution networks, the range of the pres-

method based on K-means clustering algorithm. Finally, we im-
pose an XY-coordinate system onto the network (see Fig. 3a).
We discretize the network into 150 nodes among which 50
nodes are the gas outlets carrying a demand value, and the re-
maining 100 nodes are the dummy nodes (without demand)
SC
sure readings is usually between 0-10kPag with an accuracy
of ±0.1% (ABB, 2016). To illustrate the performance of the
scheme, we select the noise standard deviation (denoted by ε)
within and above this range, i.e., 0.5%, 1% and 2% in magni-
tude of the pressure and flow data at the sensor locations. While

which may account for valves points, pipes junctions, etc. Note
that in Singapore’s gas network, approximately 70% of the
pipes are made of ductile iron (see Fig. 3b). Therefore, due
to corrosion, faulty flanges and pipe fitting, and other factors,
these dummy nodes have the highest risk for leaks. We assume
a total of 16 sensors (8 pressure and 8 flow meters) to be placed
in the network for monitoring. We arbitrarily select the loca-
M
tions and the number of sensors.
5.2. Detection of single leak without sensor noise
First, we assume that a single leak is present in each net-
D
U
generating the observed data y for both networks, the leak loca-
AN tions are the same as in Section 5.2.
The leakage events are predicted using the aforementioned
steps and the normalized probability distribution is obtained
(see Fig. 5) at different values of sensor noise standard devi-
ation. At the lower level of noise values, the normalized proba-
bility of the most probable leak node is clearly larger compared
to the probability values of other nodes. However, we can see
that when the sensor noise in the measurement data increases
from 0.5% to 2%, the probability of the most probable leak lo-
cation is reduced while the peak values of the nearby nodes are

increased considerably. It should be noted that the estimated

work. We generate the measurement data yi j assuming that the
leak location event still coincides with the actual leak position.
sensors are noise-free. For the artificial network, we consider
The estimated leakage rates are also computed with good ac-
TE

the leak location at node 122 (see Fig. 2) with a severity of 3.5
curacy. Concretely, the error remains below 12% for all 100
m3 /h, which corresponds to approx. 0.5% of the total gas sup-
test cases. Finally, to test the reliability of the leak localization
ply. For the Singapore network, we consider a single leak of 5
m3 /h at node 80 (see Fig. 3a). Fig. 4(a) and Fig. 4(b) show the
results for the normalized probability obtained for each possi-
EP

ble leak position inside leak zone in the artificial and Singapore
network respectively. We also provide the pressure and flow
values at the monitoring locations for the artificial network in
case of a single leak at node 80 in Table Appendix A.1. With
C

no sensor noise present in the measurement, the peak value of

posterior PDF is almost equal to one while the other values are
essentially zero. The proposed method identifies the leak po-
AC

sition correctly in both networks and the corresponding leak

severity, i.e., 3.5 m3 /h and 5 m3 /h are also accurately estimated.
In case of single leak scenario, the centroid node obtained by
leak localization method always coincides with the actual leak
position. However, such ideal scenario rarely occurs in the real
gas distribution system. Fig. 7: Results for distance of estimated leaks from the correct leak position in
Singapore network.
5.3. Detection of single leak with sensor noise for the single-leak scenarios, we consider 100 simulations with
Next, we consider the more realistic scenario in which sen- different noise realizations. In each trial, the leak node is se-
sor noise is taken into account while trying to locate the leaks lected uniformly at random while the leakage rate is sampled
7

uniformly at random in the range 0.1% - 10% of the total gas
flow in the network. In Fig. 6, we report the accuracy in leak
detection along with mean absolute percentage error (MAPE)
of the leak amount and its 95% confidence level. In addition,
we report the average distance error of the estimated leak loca-
tion from the actual leak location. We compute the accuracy in
leakage location to estimate the number of times the leak loca-
tion is correctly identified out of 100 random simulations, while
we compute MAPE to assess the estimates of the leakage rate.
It can be seen from the figures that sensor noise mostly affects
the detection of the leak location, and less the estimation of the
leak strength. Even for strong noise, the leakage rate can still
be accurately estimated.
Fig. 7 shows the test cases for Singapore network where the
estimated leak node does not coincide with the actual leak posi-
tion. We plot the shortest pipe distance from the actual leak po-
sition to the estimated leak location w.r.t. the normalized prob-
ability values. Localization errors below 5m, between 5m and
15m, and more than 15m are regarded as good, medium, and
low localization accuracy. We can see that almost all detections
are within the good and reasonable performance range.
100
identification accuracy (%)
80
Leak location
60
40
20
0
0 2 6 10 14
Number of Sensors
Fig. 8: Results for number of correct identification of leak location w.r.t. the
number of monitoring sensors in the network.
C EP
Fig. 9: Effect of monitoring locations on the localization of leaks.
AC
5.4. Effect of number and location of sensors
We also investigate how the number of sensors affects the
proposed LAS. To this end, we vary the number of sensors in
the network and conduct 20 simulations (with different noise
realization) for each configuration. For simplicity, we assume
that there is a single leak and this number in known in advance.
We select the leak location and the leakage rate using the same
approach as in Section 5.3. We report the results for the number
of correct identification of leaks with varying sensor configura-
tion in Fig. 8. It is obvious and also illustrated (see Fig. 8) that
the accuracy of leak localization increases with the number of
ACCEPTED MANUSCRIPT
monitoring locations. However, due to budget constraints, only
a limited number of sensors can be installed in the network and
therefore, it is important to optimize the location of these sen-
sors,
Placing sensors in a strategic manner is important, as we il-
lustrate in Fig. 9. The first configuration (see Fig. 9(a)) where
the sensors are distributed throughout the network, identifies the
leak location correctly, while the identification results are worse
in the second configuration (see Fig. 9(b)) where the sensors are
concentrated in one sub-region of the network. Consequently,
it is important to determine the optimal location of sensors in
PT
order to achieve the best possible leak localization for a given
budget.
5.5. Detection of multiple leaks with sensor noise
RI
After the successful localization of a single leak, we now in-
vestigate the scenario with more than one leak. The implemen-
tation of the overall localization methodology follows the steps
SC
in Algorithm 1. For each network, we evaluate two test cases at
different noise values (see Table 2). Threshold based leak detec-
tion stage confirms the presence of leaks in the networks. For
multi-leak scenario, we apply the leak isolation stage to iden-
tify the leaks zones for each network. For the first test case, we
U
show the leak zones for both the networks in Fig. 10. Finally,
we apply the leak localization step on the leak zone nodes to
AN estimate the leak locations and the leakage rates. The estimated
leak locations are corrected, and the above-mentioned steps are
repeated until residuals become below threshold. Table 2 lists
the estimated leak locations and the corresponding estimated
leakage rates along with the normalized probability value at the
M
18 22 26
estimated leak nodes combination (15). It can be seen from the
results that large leaks (>1% total flow) are always correctly
identified and the severity of the leakages are also estimated
correctly.
D
TE
Fig. 10: Zone that contains potential leaks in (a) artificial network and (b) Sin-
gapore network. Blue circles indicate the potential leak points and red markers
indicate the actual leak locations.
To test the reliability for multiple leak scenarios, we consider
20 simulations with different noise realizations for the Singa-
pore network, and report results for percentage accuracy in leak
location identification (see Fig. 11). The number of leaks are
varied between 1 to 10 and we consider a leak location estimate
to be accurate if more than 50% of the actual leak locations are
correctly identified. As mentioned earlier (see Section 5.4), the
accuracy of the leak location identification strongly depends on
the placement of the sensors. Therefore, the performance of
LAS may be further improved by optimizing the location of the
8
 ACCEPTED MANUSCRIPT

Table 2: Results for multiple leakage locations by varying noise values in Singapore network.
Artificial network Singapore Network
Actual Leak Nodes Actual Leak Rate (m3 /h) Actual Leak Nodes Actual Leak Rate (m3 /h)
(i) 122, 130 (ii) 105, 128 (i) 5.0, 4.0 (ii) 7.0, 3.0 (i) 78, 84 (ii) 62, 143 (i) 5.6, 9.0 (ii) 9.5, 3.7

Noise Estimated Leak Location Estimated Leakage Rate (m3 /h) Normalized Estimated Leak Location Estimated Leakage Rate (m3 /h) Normalized
Value Probability Probability
(%) Location 1 Location 2 Severity 1 Severity 2 Location 1 Location 2 Severity 1 Severity 2
122 130 5.0 4.0 1 78 84 5.3 8.8 0.89
ε=0
105 128 6.89 3.15 0.99 62 143 9.45 3.3 0.99

122 130 4.6 4.2 0.98 78 84 5.1 8.15 0.78

ε=1
105 128 7.43 3.14 0.99 62 143 9.35 3.8 0.85

PT
122 130 5.18 4.05 0.8 52 78 3.7 6.9 0.56
ε=2
105 129 6.66 3.15 0.66 62 143 9.4 3.93 0.65

sensors. An important challenge for leak detection in the low- 6. Conclusion

pressure gas distribution networks is dealing with the demand

uncertainties and noisy measurements. Here, our probabilis-
tic methodology is modified accordingly, which allows for the
explicit treatment of the uncertainties arising from the measure-
ment noise. We ran all the experiments on a 4-core Intel Core
RI
We proposed a leak localization method LAS for localizing
leakage events in gas distribution networks. We successfully
tested this approach on a small artificial network as well as on a
section of the low-pressure (2kPag) Singapore gas distribution

SC
network. We identified the location of leakages with 80% accu-
racy and estimated the leakage rate with 95% accuracy, even in
the presence of noise. Additionally, we applied the proposed
100
LAS to the scenario of multiple leaks. For detecting multi-

U
ple leaks, the computational cost of the naive probabilistic ap-
Accuracy (%)
Leak Location

80
proach is significantly higher. We proposed to reduce this cost
by carrying out leak isolation, before applying the localization
60
AN algorithm. The localization accuracy remains above 75% for
40 multiple leak case scenarios. We demonstrated the effective-
ness of the developed system by considering noisy sensor data.
20
0 0.5 1 1.5 2
Certainly, the performance of our localization methodology
M
Sensor Noise (%) is linked to the measurement locations in the network and the
threshold values of the data for leak detection. For measuring
Fig. 11: Accuracy of estimated leak locations w.r.t. noise value with 95% con- the threshold levels, we assume that there are no leaks present
fidence interval in Singapore network.
in the network while taking measurements from the sensors.
D

Additionally, we selected the sensor locations randomly on the

i7-6700HQ system with parallel computing enabled. The total synthetic and Singapore pipeline network. Therefore, it is im-
number of distinct possible leakage events that can occur in a portant to optimize the number and the location of the sensors
TE

gas distribution network with n` possible leaking nodes and k to obtain more reliable leak detection. We will address these
leakage locations is: topics in the future. We generated the pressure and flow data
n` utilized in this study through a gas network model and will re-
Nk = . (20) place by real sensor data from the Singapore gas network in
k!(n` − k)!
future work. Currently, our proposed system assumes a con-
EP

The solution scheme to solve the optimization problem for
leak localization with k leakages involves an exhaustive search
over the discrete parameter subspace. Specifically, the most
C
stant gas demand (30 m3 /h) which is a required input into the
network flow model. We will replace this demand value either
by gas consumption model (Van Westering, 2013) or by actual
measured gas demand at the customer outlets in future work.

P
probable leakage event requires the solution of Nk optimiza-
k
tion problems, considering that each function evaluation of y
Refrences
AC

involved in (15) requires the solution of a non-linear algebraic

system of equations governing the steady-state flow in the gas
network. Our algorithm complexity involves the solution of
n̂  
k = N̂k optimization problems where N̂k < Nk and
P̀ n̂`
only
k=1
n̂` denotes the leak nodes inside the leak zone. Consequently,
the computational effort becomes much smaller compared to
exhaustive search (Gupta et al., 2016). In practice n̂` is much
smaller than n` , since we assume the threshold distance to be
twice the average pipe length, which is much smaller than the
whole network size.
ABB, 2016. 2600T series guage pressure transmitters 266GSH. ABB, data
sheet DS/266GSH/ASH-EN Rev. I.
Abeysekera, M., Rees, M., Wu, J., 2014. Simulation and analysis of low pres-
sure gas networks with decentralized fuel injection. Energy Procedia 61,
402–406.
Askari, S., Montazerin, N., Zarandi, M. F., 2016. High-frequency modeling of
natural gas networks from low-frequency nodal meter readings using time-
series disaggregation. IEEE Trans. on Ind. Informat. 12 (1), 136–147.
Beck, J. L., 2010. Bayesian system identification based on probability logic.
Structural Control and Health Monitoring 17 (7), 825–847.
Beck, J. L., Katafygiotis, L. S., 1998. Updating models and their uncertainties.
i: Bayesian statistical framework. J. of Eng. Mechanics 124 (4), 455–461.

9
 ACCEPTED MANUSCRIPT

Billmann, L., Isermann, R., 1987. Leak detection methods for pipelines. Auto-
matica 23 (3), 381–385.
Bin Akib, A., bin Saad, N., Asirvadam, V., 2011. Pressure point analysis for
early detection system. In: Signal Processing and its Applications (CSPA),
2011 IEEE 7th International Colloquium on. IEEE, pp. 103–107.
Brent, R. P., 2013. Algorithms for minimization without derivatives. Courier
Corporation.
Campanella, C. E., Ai, G., Ukil, A., Mar 2016. Distributed fiber optics tech-
niques for gas network monitoring. In: IEEE Int. Conf. on Ind. Technology
(ICIT). Taipei, Taiwan, pp. 646–651.
Coelho, P. M., Pinho, C., 2007. Considerations about equations for steady state
flow in natural gas pipelines. J. of the Brazilian Soc. of Mech. Sciences and
Eng. 29 (3), 262–273.
distribution networks. Control Eng. Practice 19 (10), 1157–1167.
Ponce, M. V. C., Castañón, L. E. G., Cayuela, V. P., 2014. Model-based
leak detection and location in water distribution networks considering an
extended-horizon analysis of pressure sensitivities. Journal of Hydroinfor-
matics 16 (3), 649–670.
Preis, A., Whittle, A. J., Ostfeld, A., Perelman, L., 2010. Efficient hydraulic
state estimation technique using reduced models of urban water networks.
Journal of Water Resources Planning and Management 137 (4), 343–351.
Reddy, R. S., Payal, G., Karkulali, P., Himanshu, M., Ukil, A., Dauwels, J.,
March 2016. Pressure and flow variation in gas distribution pipeline for leak
detection. In: IEEE Int. Conf. on Ind. Technology (ICIT). Taipei, Taiwan,
pp. 679–683.
Rocha, M., 1989. Acoustic monitoring of pipeline leaks. ISA, Paper (89-0333).

Djebedjian, B., Shahin, I., El-Naggar, M., 2008. Gas distribution network op-
timization by genetic algorithm. In: Ninth Int. Congr. of Fluid Dynamics &
Propulsion, Alexandria.
Dos Santos, P. L., Azevedo-Perdicoulis, T.-P., Ramos, J. A., de Carvalho, J. M.,
Jank, G., Milhinhos, J., 2011. An lpv modeling and identification approach
to leakage detection in high pressure natural gas transportation networks.
PT
Rui, Z., Han, G., Zhang, H., Wang, S., Pu, H., Ling, K., 2017. A new model
to evaluate two leak points in a gas pipeline. Journal of Natural Gas Science
and Engineering 46, 491–497.
Segeler, G. C., 1968. Gas Engineers Handbook (E-Book), Segeler. Ind. Press.
Singapore Power, 2016. Singapore power. http://www.singaporepower.
com.sg, [Online; accessed 18-September-2016].

IEEE Trans. on Control Syst. Technology 19 (1), 77–92.
Ebrahimi-Moghadam, A., Farzaneh-Gord, M., Deymi-Dashtebayaz, M., 2016.
Correlations for estimating natural gas leakage from above-ground and
buried urban distribution pipelines. Journal of Natural Gas Science and En-
RI
Tanimola, F., Hill, D., 2009. Distributed fibre optic sensors for pipeline protec-
tion. Journal of Natural Gas Science and Engineering 1 (4-5), 134–143.
Ukil, A., Braendle, H., Krippner, P., 2012. Distributed temperature sensing:
review of technology and applicat. Sensors J., IEEE 12 (5), 885–892.

gineering 34, 185–196.
Forsythe, G. E., Moler, C. B., Malcolm, M. A., 1977. Computer methods for
math. computations.
Gupta, P., Goyal, A., Dauwels, J., Ukil, A., 2016. Bayesian detection of leaks
in gas distribution networks. In: Ind. Electronics Soc., IECON 2016-42nd
Annual Conf. of the IEEE. IEEE, pp. 855–860.
SC
Uusitalo, L., 2007. Advantages and challenges of bayesian networks in envi-
ronmental modelling. Ecological modelling 203 (3), 312–318.
Van Westering, W., 2013. Gas distribution network modelling and optimization.
Veritas, D. N., 2010. Risk assessment of pipeline protection. RECOM-
MENDED PRACTICE DNVRPF107.
Wan, J., Yu, Y., Wu, Y., Feng, R., Yu, N., 2011. Hierarchical leak detection

U
Han, Z., Weng, W., 2011. Comparison study on qualitative and quantitative and localization method in natural gas pipeline monitoring sensor networks.
risk assessment methods for urban natural gas pipeline network. Journal of Sensors 12 (1), 189–214.
hazardous materials 189 (1-2), 509–518. Zhang, J., 1997. Designing a cost-effective and reliable pipeline leak-detection
Hauge, E., Aamo, O. M., Godhavn, J.-M., 2007. Model based pipeline moni-
AN
toring with leak detection. In: Proc. 7th IFAC Symp., On Nonlinear Control
system. Pipes and Pipelines Int. 42 (1), 20–26.

Syst. Citeseer. Appendix A. Case study Network data

Hopkins, P., Goodfellow, G., Ellis, R., Haswell, J., Jackson, N., Grid, N., 2009.
Pipeline risk assessment: new guidelines. In: WTIA/APIA Welded Pipeline Table Appendix A.1: Pressure and flow data at sensor locations
Symposium. pp. 0–22. Pressure (mbar) Flowrate (m3 /h)
M
Jo, Y.-D., Ahn, B. J., 2005. A method of quantitative risk assessment for trans-
Node ID No leak leak Branch No leak Leak
mission pipeline carrying natural gas. Journal of hazardous materials 123 (1-
ID
3), 1–12.
102 41.09 40.89 P1 160.82 163.86
Karkulali, P., Mishra, H., Ukil, A., Dauwels, J., 2016. Leak detection in gas
103 39.13 38.87 P16 80.80 78.71
distribution pipelines using acoustic impact monitoring. In: Ind. Electronics
110 39.26 39.00 P28 30.0 30
D

Soc., IECON 2016-42nd Annual Conf. of the IEEE. IEEE, pp. 412–416.
115 42.11 41.93 P29 35.26 35.95
Katafygiotis, L. S., Papadimitriou, C., Lam, H.-F., 1998. A probabilistic ap-
129 35.61 33.17 P35 70.01 70.70
proach to structural model updating. Soil Dynamics and Earthquake Eng.
134 36.65 35.26 P45 30.0 30
17 (7), 495–507.
TE

Krope, J., Trop, P., Goričanec, D., 2011. Flow-pressure anal. of loop gas net-
works. Int. J. of Syst. Applicat., Eng. & development 5 (4), 477–484.
Liu, J., Yao, J., Gallaher, M., Coburn, J., Fernandez, R., 2008. Study on
methane emission reduction potential in chinas oil and natural gas ind. Tech.
rep., Tech. rep., April.
EP

Ma, C., Yu, S., Huo, J., 2010. Negative pressure wave-flow testing gas pipeline
leak based on wavelet transform. In: Computer, Mechatronics, Control and
Electronic Engineering (CMCE), 2010 International Conference on. Vol. 5.
IEEE, pp. 306–308.
Mpesha, W., Gassman, S. L., Chaudhry, M. H., 2001. Leak detection in pipes
by frequency response method. J. of Hydraulic Eng. 127 (2), 134–147.
C

Muhlbauer, W. K., 2004. Pipeline risk management manual: ideas, techniques,

and resources. Elsevier.
Murvay, P.-S., Silea, I., 2012. A survey on gas leak detection and localization
AC

techniques. J. of Loss Prevention in the Process Ind. 25 (6), 966–973.

Nejjari, F., Sarrate, R., Blesa, J., 2015. Optimal pressure sensor placement in
water distribution networks minimizing leak location uncertainty. Procedia
Eng. 119, 953–962.
Osiadacz, A., 1988. Method of steady-state simulation of a gas network. Int. J.
of Syst. Sci. 19 (11), 2395–2405.
Osiadacz, A. J., Gorecki, M., et al., 1995. Optimization of pipe sizes for distri-
bution gas network design. In: PSIG Annual Meeting. Pipeline Simulation
Interest Group.
Pennenergy, 2016. Pennenergy oil and gas. http://www.pennenergy.com,
[Online; accessed 23-October-2016].
Pérez, R., Puig, V., Pascual, J., Quevedo, J., Landeros, E., Peralta, A., 2011.
Methodology for leakage isolation using pressure sensitivity anal. in water

10
 ACCEPTED MANUSCRIPT

PT
RI
SC
Table Appendix A.2: Synthetic gas network specifications
Branch Start End Length Diameter
ID Node Node (m) (inch)
1 101 100 216.1 6
2 102 101 199.9 6

U
3 106 102 98.1 8
4 106 105* 78.9 3
5 105* 104 139.9 3
6
7
104
104
101
103
92.0
199.9
4
3
AN
8 103 100 81.1 8
9 100 120 135.9 6
10 121 120 20.1 4
11 134* 121 93.9 4
M
12 121 122* 75.9 4
13 122* 123* 78.0 6
14 126 122* 153.0 4
15 124* 123* 70.1 6
16 125 124* 63.1 6
D

17 126* 125 110.0 3

18 127 126* 60.0 3
19 129 126* 96.0 4
20 127 134* 96.9 4
TE

21 128 125 100.9 4

22 129 128 73.2 4
23 131 128 89.9 3
24 130* 129 67.1 4
C EP
AC

11
 ACCEPTED MANUSCRIPT

Branch Start End Length Diameter

ID Node Node (m) (inch)
25 130* 127 92.0 4
26 137 130* 95.1 4
27 133 137 67.1 3
28 137 132 98.1 3
29 133 135 267.9 3
30 135 134* 60.0 3
31 136* 135 70.1 3
32 136* 103 49.8 8
33 107 136* 53.0 8
34 108 107 223.1 4
35 109 108 221.9 4
36 109 106* 93.0 8

PT
37 118* 109 15.8 6
38 117 118* 111.9 6
39 117 116 88.1 6
40 119 117 84.1 6
41 116 115* 70.1 6

RI
42 115* 113 150.0 6
43 113 108 103.9 4
44 113 110* 252.1 6
45 114 113 89.9 4
46 110* 107 118.9 8

SC
47 111 110* 152.1 12
48 111 112 56.1 12

U
AN
M
D
TE
C EP
AC

12
 ACCEPTED MANUSCRIPT

HIGHLIGHTS

Manuscript No.: JNGSE-D-18-00569

Title: Leak Detection in Low-Pressure Gas Distribution Networks by Probabilistic Methods
Journal: Journal of Natural Gas Science & Engineering
Article Type: Full Length Article

PT
RI
• Leak Analytics System based on probabilistic and sensitivity approach.

• Leak Monitoring system for low-pressure urban gas distribution networks.

SC
• Three stages of the system: leak detection, leak isolation, and leak localization.

• Robust and accurate model for detecting single and multiple leaks.

U
AN
M
D
TE
C EP
AC