See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/331681375

Leak Detection in Water-Distribution Pipe System

Conference Paper · June 2018

DOI: 10.1109/ICCONS.2018.8663193

CITATIONS READS

0 846

5 authors, including:

Nikhil Binoy C
NSS College of Engineering
9 PUBLICATIONS   3 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Real Time Control System View project

All content following this page was uploaded by Nikhil Binoy C on 16 October 2019.

The user has requested enhancement of the downloaded file.

 Proceedings of the Second International Conference on Intelligent Computing and Control Systems (ICICCS 2018)
IEEE Xplore Compliant Part Number: CFP18K74-ART; ISBN:978-1-5386-2842-3

Leak Detection in Water-Distribution Pipe System

Dvajasvie.G, Farisha Banu PK, Sachin N Babu, Saheen K P, Nikhil Binoy C
Instrumentation and Control Engineering
NSS College of Engineering
Palakkad, India.
dvajasvie@gmail.com, farishabanu.ei@gmailcom, sachin.n.babu@gmail.com, saheen4success@gmail.com,
nikhilbinoyc@nssce.ac.in

Abstract— A reliable, low cost system for leak detection in
urban water distribution pipes is proposed in this paper.
Our proposed robotic system is capable of detecting leaks
of small leakage rates. This autonomous in-line pipe leak
detection system uses the presence of a radial pressure
gradient in the leak vicinity. The pressure gradient is
translated to a force, which in turn rotates a disc from
neutral position. This rotation is detected using gyroscope
and accelerometer sensors. The use of this method does not
require expert help. Leak signals are sent wirelessly to
computer for inspection. The method can be extended to
provide accurate location of leak.
Keywords— leak detection, inexpensive, automation,
remote sensing, water distribution system.
I. INTRODUCTION
Fresh water is one of the basic essential for all life forms.
Fresh water resources are getting scarcer each day. Water
distribution systems deliver quality water to area of water
shortage most of which are underground. Due to aging
pipelines, the pipes are prone to leaks. Leaks constitute a
major portion of supplying water. A recent study in Delhi,
suggested up to 25 percent of supply water is lost due to leaks.
Major population of India is dependent on water distribution
system for quality water provision. Thus significant loss of
water can cause serious socio-economic problems. The current
leak detection methods are expensive and requires expert
knowledge to use and also neither reliable nor robust. Most of
the systems also fail to provide accurate location of leak.
Shantanu Datta et al [1] Xiao-Jian Wang et al [2] summarizes
the various methods used for leak detection. Water audits,
document the difference in amount of water produced to the
total amount of water of recorded by usage meters. The
difference gives an estimate of the leak. This method is time
consuming and also does not localize the leak site. Sniffer gas
technique uses coloured or odor gas. The gas is filled in the
system. The gas leaks are then monitored using specially
designed sensors. This method is subject to weather
conditions, expensive and not reliable. Acoustic methods
depend on the acoustic signal produced due to interaction of
gas or liquid flows past a crack or hole [3],[4],[5]. This
method not only detects leaks, but also gives the leak location.
The main disadvantage is that the method is slow and it
requires experience. Peter et al [6] shows the use Infrared (IR)
thermography to detect leak in buried pipeline. The basic
principle is the difference in thermal conductivity of wet and
dry soil. An infrared camera is used to measure thermal
conductivity. J. H. Goh [7] et al describes a method in which
Electromagnetic Wave sensor is used. Changes on pipe wall
due to corrosion, cracks or leaks alter the electromagnetic
waveguide pattern and can be used to detect leaks. Main
drawback is that it can detect only one leak at a time.
Ultrasonic pig methods can detect leaks and cracks accurately.
The main disadvantage is, it cannot be used for plastic pipes.
Pressure based methods are discussed by A Abdulshaheed et
al [8]. The report shows that pressure based techniques are
considerably more sensitive to leak detection and can be used
for water distribution system. The radial pressure gradient has
already been used to detect leaks [9],[10],[11]. The MIT Leak
detector has two force sensors that detect leaks and gives
information about leak location and magnitude. The main
disadvantage is the detection is dependent on material
characteristics and this requires calibration for every new
pipeline. Sutej et al [12] proposes a prototype using leavers
that can detect the existence of a radial pressure gradient and
thus leak. The drawback is sensitivity is not same in all
directions.
II. SENSOR DESIGN
There exists a pressure difference of 30 PSI minimum across
the water-distribution pipe walls. At the leak, site there exists
a pressure drop due to fluid flow generating a suction force.
Sensor designed consists of a disc suspended by two springs.
These springs force the disc to maintain the neutral position
once it has passed the leak site. The perimeter of the disc
moves close to the pipe wall. The disc covers the entire
circumference of the pipe as shown in Fig.1. The disc has an
accelerometer and a gyroscope sensor mounted on it. The
periphery of the disc is mounted with long elastic membranes
of very small thickness. As the membrane moves near the
leak, the rapid change in pressure, forces the membrane
towards the circumference of the pipe. The membrane gets
attached and covers the leak site. This creates a drag on the
disc and the disc tilts. This tilting is detected by the
accelerometer and gyroscope sensors. Gyroscope and
accelerometer data are converted to angle measurement
detecting movement of the disc from neutral position. This
method works irrespective of material properties of the pipe.

978-1-5386-2842-3/18/$31.00 ©2018 IEEE 1151

 Proceedings of the Second International Conference on Intelligent Computing and Control Systems (ICICCS 2018)
IEEE Xplore Compliant Part Number: CFP18K74-ART; ISBN:978-1-5386-2842-3
Further the radial pressure gradient exists only around the
leak. This increases the reliability of this method by
eliminating false alarms.
III. ROBOT DESIGN
A Robotic system was designed with the prime objective of
moving the sensor inside the pipe. Pipe of diameter “4 inch”
(110cm) is chosen. The movement of the robot is constrained
Fi 1
such that it does not disturb the sensor arrangement. The entire
setup is shown in Fig.2. The robot moves in a straight pipe.
Fig. 1. Sensor Position inside Pipe.
The robot is mounted with two microcontrollers. One
microcontroller controls the direction of motion and speed of
robots and other processes sensor data. Circular nature of pipe
wall causes less contact surface area between pipe wall and
robot wheel. Thus speed control is provided as a mechanism to
counteract slippage problems arising due to less contact
surface. The motor and rear wheels are coupled to a single DC
motor using a gear arrangement. This converts high speed, low
torque output of a DC motor to high torque, low speed motion
required to drive the heavy robot. The sensor designed is
mounted to the front of the robot. One “High watt” 9V battery
is used as power supply. The connection between the robotic
segment and an operator is established wirelessly using
Bluetooth. The robot moves the sensor at various speed across
the leak so as to determine the reliability of this method. The
robot was able to run for 20 min continuously.
978-1-5386-2842-3/18/$31.00 ©2018 IEEE
Fig. 2. In-line pipe leak detection Robot
IV. TEST APPARATUS DESIGN
The proposed system is tested in the laboratory using a control
loop as shown in Fig.3. The robot is placed in the pipe and
closed. The pipe is pressurized using a compressor. The
pressure output of the laboratory compressor is 100 PSI. The
control loop is manually adjusted to keep the pressure within
safe limits as the pipe chosen has a pressure rating of 56 PSI.
Pressure regulator can be adjusted to vary pressure inside the
pipe, thus serves as a method to measure the sensitivity of
designed sensor.
Fig. 3. Control Setup for simulating real time leak condition
V. DATA ACQUISITION AND CONTROL
A. Angle conversion
The sensor designed provides six raw values, three from
gyroscope and three from the accelerometer. Accelerometer
reads both tilt of the sensor disc and acceleration of the robot.
Static tilt of disc is of importance to leak detection and should
be separated from robot acceleration. Gyroscope integrates
angular velocity with respect to time to produce a tilt angle.
Thus the measurement is prone to drift due to bad initial
conditions. Thus a complimentary filter is designed. The filter
combines low frequency component of accelerometer and high
frequency component of gyroscope to provide reliable angle
measurement of disc tilt.
B. Robot Motion and Speed control
A LabVIEW software is created in the operator’s computer.
Serial Communication is established between computers and
on board microcontroller wirelessly using a HC-05 Bluetooth
module on robot side and laptops inbuilt Bluetooth at the
operator’s side. This helps in controlling both speed and
direction of the robot.
C. Leak Monitoring
A separate LabVIEW software is designed to aid leak
monitoring. Wireless connection is established between
1152
 Proceedings of the Second International Conference on Intelligent Computing and Control Systems (ICICCS 2018)
IEEE Xplore Compliant Part Number: CFP18K74-ART; ISBN:978-1-5386-2842-3
second on board microcontroller and leak monitoring
computer using a similar method as described above. The
LabVIEW program compares the current angle measurement
with neutral position angle and raises alarm in case of a leak.
A graph is plotted in terms of angle measurement. There exists
a sudden change in angle value whenever a leak exists. The
time difference between leaks can provide a method to
localize leak. The complete block diagram of the robotic
system is shown in Fig. 4.
VI. RESULTS AND OBSERVATION
The system was deployed in a pipe of “4 inch” (110cm)
diameter. Leak of size 2mm is accurately measured when the
pipe is pressurized to 25PSI gauge pressure. The sensor
designed was found satisfactory and provided reliable results.
The result of Leak test in a pipe with two 2mm holes separated
by a distance using LabVIEW is shown in Fig. 5.
Fig. 4. Block Diagram of Robotic setup
Fig. 5. Leak Monitoring in LabVIEW
Drawback and Remedies:
978-1-5386-2842-3/18/$31.00 ©2018 IEEE
x The sensor should be specifically designed for each
pipe diameter. In practical conditions, pipe
dimensions vary. Research on elastic membrane can
be conducted and proper arrangement can achieved to
tackle the dimension variation problem to some
extent.
x The duty cycle of the robotic system is 20 minutes,
which is not satisfactory. This can be increased
drastically by making the sensor arrangement self-
propelling. This makes the system, lightweight and
eliminates the necessity of two microcontrollers and a
drive mechanism. This completely eliminates the
problem of wheel slip.
x The location of the leak is not indicated clearly.
Further, Bluetooth is not feasible in real time due to
its limited range. Adding a GPS to the system
extends the range satisfactorily and creating an
android map application will accurately pinpoint
location of leak in real time applications.
x The spring arrangement on which disc is suspended
is susceptible to vibrations. Further, research will
help in finding proper suspension arrangements.
VII. CONCLUSION
This research focuses on developing low cost, reliable sensor
that can be used to detect leaks in water-distribution pipes.
The prototype built is inexpensive and costs only ₹ 6000 in
making. The leak detection method is continuous and can
detect leaks irrespective of material properties, thus suitable
for both old metal pipes and new plastic pipes. Research for
suitable disc and elastic membrane can be conducted and
1153
 Proceedings of the Second International Conference on Intelligent Computing and Control Systems (ICICCS 2018)
IEEE Xplore Compliant Part Number: CFP18K74-ART; ISBN:978-1-5386-2842-3
system can be translated to water-leak detection. Proper
material selection can also make the system more sensitive
and thus detect smaller leak and at low flow conditions. The
sensor can be modified such that it provides minimal
disturbance to natural fluid flow. The system does not require
any expertise to operate. Once the GPS module is added to the
leak location can be accurately determined thus increasing
reliability and ease of use. Further, computational algorithms
can be developed to analyze the magnitude and nature of the
leak and extended to identify water theft.
REFERENCES
[1] Datta, Shantanu & Sarkar, Shibayan. (2016). A review on different
pipeline fault detection methods. Journal of Loss Prevention in the
Process Industries. 41. 10.1016/j.jlp.2016.03.010.
[2] Wang,X.J., Simpson,A.R., Lambert, M.F. and Vitokovsky, J.P.(2001).
“Leak Detection in pipeline system using hydraulic methods: a review”.
Conference on Hydraulics in Civil Engineering, The Institutuion of
Engineers, Australia,Hobart, 23-30 November, 391-400.
[3] P.-S. Murvay, I. Silea, ”A survey on gas leak detection and localization
techniques,” Journal of Loss Prevention in the Process Industries, vol.25,
pp. 966–973, 2012.
[4] Brodetsky, I., & Savic, M. (1993). Leak monitoring system for gas
pipelines. In Acoustics, speech, and signal processing, 1993. ICASSP-
93., 1993 IEEE International Conference on, Vol. 3 (pp. 17e20). IEEE.
978-1-5386-2842-3/18/$31.00 ©2018 IEEE
View publication stats
[5] Kim, M., & Lee, S. (2009). Detection of leak acoustic signal in buried
gas pipe based on the time-frequency analysis. Journal of Loss
Prevention in the Process Industries, 22(6), 990-994.
[6] Bach, P. M., & Kodikara, J. K. (2017). Reliability of Infrared
Thermography in Detecting Leaks in Buried Water Reticulation
Pipes. IEEE Journal of Selected Topics in Applied Earth Observations
and Remote Sensing, 10(9), 4210-4224.
[7] J.H.Goh, A.Shaw, J.D.Cullen, A.I Al-Shamma’a, “Water pipe leak
detection using electromagnetic wave sensor for the water industry”,
IEEE sumposium on Computers & Informatics,2011,290-295.
[8] Abdulshaheed, A., Mustapha, F., & Ghavamian, A. (2017). A pressure-
based method for monitoring leaks in a pipe distribution system: A
Review. Renewable and Sustainable Energy Reviews, 69, 902-911.
[9] Chatzigeorgiou, D., Wu, Y., Youcef-Toumi, K., & Ben-Mansour, R.
(2014, May). MIT Leak Detector: An in-pipe leak detection robot.
In Robotics and Automation (ICRA), 2014 IEEE International
Conference on (pp. 2091-2091). IEEE.
[10] Chatzigeorgiou, D., Youcef-Toumi, K., & Ben-Mansour, R. (2015).
Design of a novel in-pipe reliable leak detector. IEEE/ASME
Transactions on Mechatronics, 20(2), 824-833.
[11] Chatzigeorgiou D., Wu Y., Youcef-Toumi K. and Ben Mansour R.,
“Detection & Estimation Algorithms for In-Pipe Leak Detection,”
American Control Conference (ACC),June 4-6, 2014. Portland, Oregon,
USA.
[12] Sutej Pramod Kulgod, Srinidhi S, Satyesh Shanker Awasthi, “ Low cost
in-line pipe leak detection robot”, IEEE Distributed Computing, VLSI,
Electrical Circuits and Robotics, 2016,79-83.
1154