Wind turbines are highly reliable machines that are typically operated in remote offshore areas. The deployment of wind turbines is growing massively as countries strive to decarbonise their energy systems. Increasing operational efficiency to reduce costs is critical to achieve net zero ambitions. The growth in this sector will continue over decades to come. Data mining offers methods for remotely monitoring these turbines through learning the difference between normal and faulty behaviour.
The challenge arises from the sparsity of faulty behaviour data in any single turbine model. This PhD seeks to investigate how graph learning would be able to inform failure prognostic models with the potential to generalise the model to different turbine models. The project will be undertaken in close collaboration with the Ørsted, the industrial sponsor. Ørsted have developed wind turbine graphical model, and are interested in developing machine learning tools that will be able to make use of this model. As an operator of a very large and varied fleet of wind turbines, they will also be providing rich operational and maintenance data to support the machine learning process.
You will join a group of other PhD and postdoctoral researchers working in various areas of renewable energy. As a PhD researcher on this project, you will have freedom to explore and develop machine learning tools along with developing the validation process for the analytic models. This will require working with both your academic supervisor team and the industrial partners, with different expectations from both in terms of reporting style and content. This project will include the opportunity to spend significant time embedded with the industrial sponsor in Copenhagen.
Environment
Durham University, one of the top 100 universities in the world (ranked 78th in QS World University Rankings 2024) is inviting applications for a fully funded PhD studentship to work within the Department of Engineering. The successful candidate will have full access to the facilities available in the Department of Engineering at Durham University. They will be also able to access training through Durham University’s Researcher Development Programme. Durham University provides a wide variety of training courses, ranging from paper writing support through to high performance computing courses. These run continuously and are highly valued by all researchers at the University.
This project presents an excellent opportunity for a candidate interested in an applied research career, within a 3.5 year funded study programme.
Durham supervisors: Dr Peter Matthews, Prof Simon Hogg
Ørsted supervisor(s): Pierre-Julien Trombe
Requirements:
Essential:
· Good Degree (2:1 or above) in Engineering or other highly numerate discipline (eg Physics, Computer Science, Mathematics); or equivalent work experience
· Excellent communication skills (written, spoken)
· Ability to work with multiple supervisors (academic and industrial)
· Strong programming skills
Desirable:
· An understanding of the technical and economic challenges facing wind energy systems
· Expertise in Python (ideally with appropriate modules, Numpy, SciKit, PyTorch or similar)
· Basic SQL experience
· Experience with machine learning and testing/validation of machine learning outputs
· Ability to work remotely
· Willingness and ability to undertake long periods (max 1 year) of independent work embedded with the industrial sponsor in Copenhagen
Funding:
Funding duration 3.5years (UK Home student)
This project will provide funding for UK Home students:
Home fees: £4,712 p.a.
Stipend: £18,622 p.a.
Note: This funding is for UK Students (overseas students are welcome to apply but would need to cover the difference between home students' fees and overseas fees, the difference being about £24k).
Application process:
In the first instance (or for informal discussion), contact Peter Matthews ([Email Address Removed]). Application can then be formally submitted through the online postgraduate (research) application system (please highlight that your application should be considered by Dr Peter Matthews): https://www.durham.ac.uk/study/postgraduate/research-degrees/how-to-apply/
We are looking to start in January 2024, but will consider starting sooner.
Continue with Facebook
