Bio
Emily is a doctoral student at the centre for Data Intensive Science at UCL, collaborating closely with the Culham Centre for Fusion Energy. Prior to joining the CDT she was a scientific computing developer at the Rutherford Appleton laboratory and completed her MEng in Nuclear Engineering at the University of Birmingham.
She is strongly motivated to develop nuclear fusion into a sustainable emission-free energy source and believes that machine learning methodologies are key tools in achieving this goal.
Research interests
Emily's doctoral project is primarily concerned with deep learning applied to plasma equilibria.
Calculating the plasma equilibrium inside a fusion reactor is an important first step in plasma simulation and control. However traditional methods do not perform with the speed required to enable real-time control, which is vital for reactor operation.
Using modern deep learning methods such as physics-informed neural networks, this operation can be accelerated to sub-millisecond speeds while also obeying the physical laws controlling the system.
Emily is additionally interested in Uncertainty Quantification, a key concern when modeling the safety critical processes inside a nuclear reactor.