CFD is at the core of the design process for aerospace and automotive as it is used to reduce drag and hence CO2 emissions.
However, CFD prediction of industrial applications require large amounts of computational resource and is both time-consuming and expensive. Acceleration of this part of the design process would allow designs to be produced more quickly and with greater exploration of the design space.
This project will develop a technique using AI/ML and Reduced Order Models to predict flows based on the vehicle prior to carrying out full CFD calculations. This will involve developing techniques to encode geometrical information and evaluating the best Artificial Neural Networks or other techniques such as Reduced Order Models to reproduce the flow field and important performance parameters.
By limiting the scope to automotive shapes, the geometry and flows will have similar structures and can be relatively easily parameterised/characterised.
Primary supervisor: Prof. Gary Page
Secondary supervisor: Dr. Eve Zhang and Dr Miguel Martínez García
Entry requirements for United Kingdom
A good first (undergraduate) degree in Engineering, Maths, Physics or Computer Science (or equivalent to UK upper second class).
English language requirements
Applicants must meet the minimum English language requirements. Further details are available on the International website.
Find out more about research degree funding
How to apply
All applications should be made online and must include a research proposal. Under the programme name, select 'Aeronautical and Automotive Engineering'. Please quote the advertised reference number AACME-23-029 in your application.
To avoid delays in processing your application, please ensure that you submit the minimum supporting documents.