Intelligent Modelling of Self‑Organised Edge Turbulence in Tokamak Plasmas – Plasma Strand Project

Motivation. Achieving reliable, high-performance fusion requires a predictive understanding of turbulence and instability dynamics at the tokamak edge, where fluctuations spread nonlocally and strongly influence pedestal structure, confinement, and the approach to disruptive events. These processes are difficult to diagnose experimentally and too computationally intensive to model in real time with first-principles simulations, motivating new approaches that blend physics-based modelling with modern AI.

Aims. This PhD will develop a physics-informed AI framework to understand and predict turbulence at the plasma edge. The student will combine open-source simulations (GENE or its GPU-accelerated version GX, and BOUT++), GBS datasets provided by collaborators, and experimental data from KSTAR to build fast surrogate models that emulate complex plasma behaviour. The project integrates new simulations with existing datasets, ensuring both breadth and feasibility.

The key goals are to: i) develop rapid AI surrogate models that reproduce turbulence at a fraction of the computational cost of full simulations; ii) identify the dominant mechanisms driving turbulence evolution and regime transitions using interpretable, data-driven methods; iii) validate predictions against high-resolution diagnostics from the superconducting KSTAR tokamak to ensure real-world relevance.

By the end of the PhD, the project will deliver intelligent models that bridge heavy simulations and real-time plasma control, advancing predictive capability for next-generation fusion devices.

Training & Skills. The student will receive strong training in plasma physics, edge turbulence, and computational modelling under the supervision of Prof. E Kim. They will work with state-of-the-art open-source simulation tools (GENE/GX, BOUT++), supported by additional GBS datasets from NTU collaborators. A collaborator at Deakin University will provide general guidance on AI methods. The student will also analyse experimental data from the superconducting KSTAR tokamak, gaining experience with high-resolution diagnostics and ensuring that models are grounded in real plasma behaviour. This combination of physics, computation, AI, and experimental analysis will provide a multidisciplinary skill set for future fusion research.

The student, supported by Fusion CDT community studentship, is expected to take the training at York in their first year. They will obtain training in ethics, literature review, presentation skills, and scientific article writing provided by the Doctoral College or the Centre for Academic Writing at Coventry University. The student is expected to disseminate research outcomes through journal paper publications and conference presentations.

Plasma strand students are based at University of York for the initial six months of the PhD, for the taught modules. During that first six months students will typically travel to undertake taught modules at all of the Fusion CDT partner universities. After the taught programme, for the remainder of the PhD this project will be based at Coventry University.

The project will be mainly based in Coventry with foreseeable visits to NTU and fusion laboratories (e.g., KSTAR at KFE) for modelling/data analysis. The student will have the opportunity to attend international conferences to present the results of the work.

This project may be compatible with part time study, please contact the project supervisors if you are interested in exploring this.

This project is being offered by Coventry University as part of the Fusion CDT Community Studentship scheme. For further information and details of how to apply please contact Professor Eun-jin Kim (ad3116@coventry.ac.uk).