Developing sustainable transport system is considered both a necessity and a significant drive for engineering innovations nowadays. As the research problems become increasingly complex – for instance, the next-generation aerial vehicles often comprise multi-rotor configurations constantly interacting with its operating environment – researchers are constantly looking for new toolsets to help better explore the design space and improve efficiency of these vehicles. Among them, machine learning is a promising enabler to such objectives. In this seminar, I would like to share some preliminary developments on how we could potentially predict the flow and noise of a vehicle using transformer-based machine learning algorithms as well as to discuss how we may work towards gaining some physical insights on the key physical parameters.
|
 Nick Zang is currently a Senior Lecturer in Aerospace Engineering at Queen Mary University of London (QMUL). Graduated from NTU, he mainly works in the intersection between fluid flow and aeroacoustics, that is to understand, predict and control aerodynamic noise and its accompanying environmental impact based on insights from the physical flow mechanisms. Prior to joining QMUL, he was a Lecturer at University of Bristol, where he has been pivotal to securing national funding for an electric propulsor testbed as well as a large boundary layer facility for environmental flows. His research is funded by InnovateUK, EPSRC and Horizon Europe.
|
