Digital Engineering for Fusion: Uncertainty Quantification and Validation Frameworks

About the Project

Project Overview

The development of reliable predictive models and through - life engineering strategies for future fusion power plants presents one of the most significant challenges in modern engineering. Fusion systems — particularly those based on tokamak concepts — operate under extreme electromagnetic, thermal, and structural conditions. Central to these systems are magnet components responsible for plasma confinement and stability, which are subject to complex interactions between electrical currents, magnetic fields, and mechanical stresses.

As fusion technologies transition from experimental devices to commercial power plants, there is an increasing reliance on simulation - driven design. However, the credibility of these models is limited by uncertainties in material properties, boundary conditions, and operational parameters, as well as the scarcity of representative experimental data from full - scale fusion environments. This creates a critical need for robust methodologies that integrate experimental evidence with computational models to improve predictive accuracy and reliability.

The project will investigate how uncertainty in multi - physics simulation models can be reduced through the integration of experimental measurements, statistical inference techniques, and digital twin architectures. In doing so, it will contribute to overcoming one of the major bottlenecks identified by national and international fusion programmes, including those led by the UK Atomic Energy Authority.

The objectives of the PhD are:

• develop inverse uncertainty quantification methods to refine uncertain model inputs using experimental data
• establish a hierarchical validation framework for multi - physics fusion systems
• define benchmark problems and datasets for validation of fusion simulations
• integrate uncertainty quantification and validation within a digital twin architecture
• demonstrate improved predictive capability for fusion magnet systems through data - driven model calibration

Research Approach

The PhD will adopt a multi - disciplinary methodology combining experimental, computational, and data - driven techniques. Experimental test rigs will be used to generate validation datasets capturing electromagnetic and structural interactions. These will be complemented by multi - physics simulations to model system behaviour under different operating conditions.

A central focus will be the development of inverse uncertainty quantification techniques, where experimental observations are used to refine uncertain model parameters using approaches such as Bayesian inference and Gaussian process modelling. This will enable systematic reduction of uncertainty and improved confidence in simulation outputs.

The project will also develop a hierarchical validation framework, allowing models to be tested across multiple levels of complexity — from simplified physical experiments to coupled multi - physics systems. These methodologies will be embedded within a digital twin framework, enabling continuous model updating and supporting predictive maintenance and operational decision - making.

Training and Research Environment

This project is highly interdisciplinary and will be conducted at the University of York within a leading research environment in digital engineering, AI, and complex systems.

The student will benefit from expertise in:

• digital twins and data - centric engineering
• multi - physics modelling and simulation
• machine learning and uncertainty quantification
• complex engineering systems and predictive maintenance

The project is aligned with UK fusion initiatives and offers opportunities to collaborate with national laboratories, industry partners, and international research programmes. Primarily at UoY.

Person Specification

We are seeking highly motivated candidates with:

• a first - class or upper second - class (2.1) degree (or equivalent) in Engineering, Physics, Mathematics, or a related STEM discipline
• strong interest in fusion energy, digital twins, or data - driven engineering
• knowledge of (or willingness to develop skills in):
  • computational modelling
  • statistics / machine learning
  • experimental methods

Experience in uncertainty quantification, multi - physics simulation, or data analysis is desirable but not essential.

How to Apply

For informal enquiries, please contact: Suresh.perin@york.ac.uk

Applicants must apply via the University’s online application system at https://www.york.ac.uk/study/postgraduate-research/apply/. Please read the application guidance first so that you understand the various steps in the application process. To apply, please select the PhD in Engineering for September 2026 entry. Please specify in your PhD application that you would like to be considered for this studentship.

This project is aligned with the UK’s national fusion strategy and offers a unique opportunity to lead the way in sustainable energy technology.