Topic background - There is about 130.000 kilometre of water distribution mains in the
Netherlands alone. Their lifetime is exceeding 50 years, and need gradually replacement.
Inspection methods to assess their status are available, but they are not sufficiently advanced to
detect small leaks, or to easily find defects in plastic materials. In order to advance in this field, a
new type of leak and defect detection method needs to be developed. The expected social impact
is that the water mains network remains in good operation as long and as good as possible and
that defects can be found before actual catastrophic failure of the piping occurs, leaving customers
with an interrupted and sometimes fouled water supply.
Research challenge - The proposed method relies on an electric connection through the pipe
which, when there are leaks in pipes, can give a signal. When the local electric field is measured,
this gives information about the location of the leak. Plastic pipes seem suitable to this method,
perhaps other types of pipe too, rubber couplings can probably be inspected as well. To be able to
do this, another electrical path is necessary, outside the area of that leak to the surrounding soil.
This could be done by capacitively coupling an AC signal through the pipe, or by a 'tether'
connection: a wire to outside. The change of the dielectric constant of the material can also be
used, indicating partial failures.
A number of research questions emerge:
1. Is the method suitable for existing water pipes, and which ones? The water pipes in the
Netherlands consist of PVC, Polyethylene, concrete and iron.
2. Which defects can be detected and how? One can think of cracks, holes, but also of a damage
that is not completely 'through', or for example local crack formation or inclusions.
3. Which parameters determine the sensitivity and in which way? Element size, wall thickness,
crack shape, the effect of the contra-connection (tethering or a capacitive method),
4. How can existing sensor data be merged to increase the resolution of detection? And how can
false positives be avoided?
The innovation opportunity will be the development of a novel inline inspection method for plastic
piping systems, able to detect defects and small leaks before an actual problem is onset. The
resulting knowledge will be used in actual inspection systems.
Objectives and methodology - The objective is to develop a method able to detect small defects
and leaks inside water mains by means of a small (AC) voltage applied to the water in the pipe
while being scanned from the inside. An advised start-up approach would be literature research
towards crack and failure occurrence and potential risks, combined with initial steps of doing
experimental work on the measurement principle and development of the measurement system.
Next, experimental work combined with theory and modelling of the electric field in water will guide
into the most promising direction. Data processing, combining other inspection data and control
theory can lead to optimal detection. Finally, there is an opportunity to build a pilot system to be
used in real life inspection equipment.
Students’ requirements: The candidate must hold a MSc degree in electrical engineering,
mechanical engineering or computer sciences. We are looking for a tech enthusiast, with skills in
the field of electronics, control theory, signal processing or similar, with experience in experimental
work. We offer a very social, challenging environment with lots of opportunities for self-
Keywords: Electronic leak detection, machine learning, inspection method, data processing.
Academic supervisors: prof. dr. ir. Jacquelien Scherpen and prof. dr. ir. M. (Ming) Cao (Faculty of
Science and Engineering, University of Groningen)
Wetsus supervisor: Doekle Yntema (Theme coordinator Smart Water Grids)
|Intitulé||PhD project - Inline electronic leak detection in water distribution systems|
|Job location||Oostergoweg 9, 8911 MA Leeuwarden|
|Publié||septembre 6, 2021|
|Date limite d'inscription||octobre 29, 2021|
|Types d'emploi||PhD  |
|Domaines de recherche :||Informatique,   Sciences de l'information,   Algorithmes,   Intelligence artificielle,   Réseau de neurones artificiels,   Informatique et société,   Architecture informatique,   Communications informatiques (réseaux),   Infographie,    and 21 more. Cyber Security,   Informatique dans les mathématiques, les sciences naturelles, l'ingénierie et la médecine,   Informatique dans les sciences sociales, les arts, les lettres et les humanités,   Extraction de données,   Structures de données,   Bases de données,   Informatique distribuée,   Interactions homme-machine,   Systèmes d'information (informatique de gestion),   Systèmes d'exploitation,   Informatique parallèle,   Langages de programmation,   Informatique quantique,   Génie logiciel,   Théorie de calcul,   Sciences informatiques,   Conception de jeux,   Big Data,   Apprentissage automatique,   Vision industrielle,   Vision par ordinateur  |