A partnership between The University of Sheffield and Thames Water, as part of the Water Infrastructure and Resilience (WIRe) EPSRC Centre for Doctoral Training
Closing Date for Applications: Noon Monday 21st March 2022
Start Date: 26th September 2022 (duration 4 years)
Project Background
Water supply and sewer networks are large, complex engineered systems that have traditionally relied on limited model-based monitoring and operational decision making. However, increasing demand for robust state estimation and reliable network performance highlights the need for the adoption of data-driven approaches that can leverage real-time information acquired by a sensor network overlaying the water supply and sewer network to provide low-latency state information. The variety of sensors that comprise the sensing infrastructure and the significant cost of installing the sensors gives rise to challenging design problems, most notably producing sensor placement strategies and estimation algorithms that take into account the specific operational constraints of any network.
Recent information theoretic tools and machine learning methods open the door to the development of a quantitative analytical framework that will facilitate the design of sensor placement and state estimation techniques that combine data-driven approaches with the insight obtained from calibrated hydrodynamical models. This project aims to answer questions such as, ‘what is the optimal placement and sampling of sensors that enables best forecasting of network performance?’, as well as use the developed tools to take critical decisions on network/sensor installation and maintenance activities.
This project will involve collaborative working with Thames Water, who are committed to deploying large numbers of sensors and adopting new approaches to improve network performance.
Aims and Objectives
The main aim of this project is to develop data-driven sensor placement and state estimation methods that predict water system problems with strong theoretical guarantees of robustness and reliability. Specifically, the project is structured along the following objectives:
1. Develop optimal sensor selection and placement strategies that maximize the amount of information acquired by the sensing infrastructure.
2. Design and validate data-driven state estimation and faults detection algorithms that augment the information from the sensor network to the hydrodynamic models.
3. Design and validate machine learning forecasting tools that predict the state of the network and key performance metrics, such as the occurrence of fault events, such as blockages, overflows, leaks etc.
Environment
This PhD will be undertaken within the EPSRC Centre for Doctoral Training in Water Infrastructure and Resilience (CDT WIRe). CDT WIRe is a collaboration between the three leading UK Universities in water resilient infrastructure. Students will benefit from a bespoke training scheme delivered by world leading experts from academia and industry, access to world leading experimental and computational facilities as well as close and regular contact with industry and end user partners. CDT WIRe is committed to promoting a diverse and inclusive community, and offer a range of family friendly, inclusive employment policies. For further information on the CDT WIRe scheme visit the web site: https://cdtwire.com/
The project will be supervised by Dr Iñaki Esnaola, with the co-supervision of Prof George Panoutsos, Prof Vanessa Speight, and Prof Simon Tait at the University of Sheffield in collaboration with staff from Thames Water.
Eligibility Criteria
Home (usually established by residence) and International applicants (which now includes EU) are welcome to apply.
Selection Criteria
Applicants need to have a good first degree in mathematics, engineering, computer science, or physical sciences. Applicants with strong mathematical skills are particularly welcome. A strong background in calculus, linear algebra, and probability theory is indispensable for this project.
How to apply
Interested candidates should email a covering letter and their Curriculum Vitae to Miss Lindsay Hopcroft ([Email Address Removed]). For data protection purposes, please state in your covering letter that you give permission for your CV to be shared with industrial partners.
For information and informal enquiries please contact: Dr Iñaki Esnaola ([Email Address Removed]).
Continue with Facebook
