FREE PhD study and funding virtual fair REGISTER NOW FREE PhD study and funding virtual fair REGISTER NOW

Accelerator characterisation and optimisation using machine learning


   Cockcroft Institute

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
  Prof A Wolski  Applications accepted all year round  Funded PhD Project (Students Worldwide)

About the Project

Effective operation of a high-energy particle accelerator depends on detailed knowledge of the beam properties. Many useful techniques have been developed over the years to characterise beams, but as accelerators become increasingly sophisticated, more advanced techniques are often needed to achieve a deeper insight into the beam and machine behaviour. The challenge with many novel diagnostics and analysis methods is that they often require the processing of very large volumes of data, which can require significant computational resources and, because of the time taken to complete the analysis, often limits the applicability of a given technique. The goal of this PhD project is to develop approaches to characterising accelerator and beam behaviour based on artificial intelligence and machine learning, which offers the possibility of greatly improving the efficiency of accelerator tuning and operation.

Initial studies on CLARA, the Compact Linear Accelerator for Research and Applications at Daresbury Laboratory, have indicated the feasibility of applying machine learning for detailed measurement of the charge distribution within bunches of high-energy electrons. Knowledge of such beam properties is invaluable for understanding the overall behaviour of advanced accelerators. However, machine learning can also be applied in a range of other ways in accelerator R&D and operation. For example, after characterising the beam properties, the next step is generally to tune the accelerator to achieve specified performance goals. While traditional tuning methods based on machine modelling are often effective, in some cases the application of existing techniques can be difficult and time-consuming, and may not always achieve the desired results in a reliable fashion. Machine learning offers the possibility of developing tuning and optimisation tools based on neural networks trained using data from detailed, computationally intensive simulations performed off-line, but capable of rapidly generating results in the control room. As well as looking at diagnostics techniques, this PhD project will also explore the potential of machine learning to facilitate optimisation of accelerator performance.

Essential components of the project will include validation of new techniques in experimental tests, evaluation of the potential for practical implementation of methods based on machine learning in the control room, and assessment of the possible benefits (and drawbacks) of machine learning compared with conventional approaches. The studies will focus on CLARA, but will consider also applications to a wide variety of different types of accelerator.

The project is available from October 2022.  The student will be based at the Cockcroft Institute (CI) at Daresbury Laboratory, Warrington, UK, and will work closely with researchers across the CI, including academic staff from the partner universities and staff based at Daresbury Laboratory itself.  Research will involve theoretical, computational and experimental studies of beam dynamics in a particle accelerator. The student should gain a practical understanding of accelerators and their various subsystems and components by participating in studies on CLARA.

Potential applicants are encouraged to contact Prof. Andy Wolski (), or Dr David Dunning () for more information. This position will remain open until filled.

Funding and eligibility: Candidates should have a first or upper second class degree in physics, or a comparable qualification. A good understanding of electromagnetic theory will be required, and strong computational skills (including the ability to use standard scientific computing tools, and to develop customised software using an appropriate programming language). There is a possibility of funding for the project provided by the Science and Technology Facilities Council for 3.5 years; UK and international citizens are eligible to apply.  A full package of training and support will be provided by the Cockcroft Institute, and the student will take part in a vibrant accelerator research and education community of over 150 people. An IELTS score of at least 6.5 is required.

You can find out more about being a PhD student at the Cockcroft Institute at the CI website (https://www.cockcroft.ac.uk/), where you can download an application form and find out about the other PhD projects available at the Cockcroft. To apply for this project, please complete the application form available on the CI website and email it with your CV to .

For further details on the application process, please visit https://www.cockcroft.ac.uk/education-and-training/phd-information

Anticipated Start Date: October 2022 for 3.5 Years

Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs