Zheyuan Chen
University of York
Co-hort year: 2025
I completed my integrated Master’s degree in Mathematical and Theoretical Physics with distinction at Oxford, shortly before joining the Fusion CDT programme. During my studies, I developed strong interests in plasma physics and computational physics and came to appreciate the importance and potential of fusion energy. My project, supervised by Christopher Ridgers and Jiannan Yang, will focus on using machine learning to build models of relativistic particle acceleration relevant to high gain inertial confinement fusion (ICF). I will be based at York.
High-intensity lasers can ionise matter and create a state of matter called plasmas, containing strong electric and magnetic fields. These fields accelerate electrons and ions to extremely high energies, causing them to emit x-rays and gamma-rays. Such energetic particles and radiation are important for ICF, especially under the novel fast ignition schemes. However, modelling these laser–plasma interactions typically requires particle-in-cell (PIC) simulations, which are computationally expensive to run. My research focuses on using data from PIC simulations together with Gaussian Process Regression (GPR) algorithms to train and improve surrogate models (a surrogate model mimics the simulation behaviour in a black-box way). Surrogate models are much faster and computationally cheaper to run while still providing reliable predictions with well-quantified uncertainties. These models should prove valuable for diagnostics and for the design of targets for higher gain.
Supervisors
