I undertake research in data-centric aeronautics: applying existing, and developing novel, data-driven algorithms in turbomachinery aerothermodynamics and machine learning for better aerodynamic inference and decision-making. My research interests vary from statistical theory and numerical linear algebra to turbomachinery applications. Explore my current projects below.
Bayesian spatial models from engine sensor measurements
Understanding aeroengine aerothermodynamics by generating spatial fields of temperature and pressure based on sensor measurements.
Dimension reduction in blade aerodynamics
Finding lower-dimensional representations of high-dimensional design spaces for the aerodynamic design and manufacturing of aeroengine blades.
Polynomial approximations for machine learning
Leveraging polynomial approximations via least squares, compressive sensing and numerical integration rules for learning.