Postgraduate research project

Characterisation of cast austenitic stainless steels using ultrasonic backscatter and artificial intelligence

Funding
Competition funded View fees and funding
Type of degree
Doctor of Philosophy
Entry requirements
2:1 honours degree View full entry requirements
Faculty graduate school
Faculty of Engineering and Physical Sciences
Closing date

About the project

If you enjoy numerical modelling and are you interested in the application of artificial intelligence (AI) to engineering problems, this may be the right project for you.

Our main aim is to combine the power of finite element modelling with artificial intelligence to develop ultrasonic characterisation of castings in-situ. Whilst helping to improve the way the industry manages safety-critical assets, you'll develop a transferrable skillset comprising:

  • numerical modelling
  • large data analysis 
  • AI (which would prove useful on many potential future career paths)

Stainless steel castings possess outstanding corrosion resistance and mechanical properties. For these reasons, they underpin numerous safety-critical installations and are ubiquitous in, e.g. nuclear power plants. However, their excellent performance comes at a cost - they are challenging to inspect for damage in situ.

Ultrasound - the preferred inspection method - suffers from the coarse-grained microstructure, which scatters and attenuates injected acoustic energy making reflections from defects illegible. Having some knowledge about the microstructure before testing offers game-changing possibilities.

First, one could tailor the inspection protocol to minimise the detrimental effect of noise. Second, it would help select the correct procedure for a specific defect which is a crucial maintenance challenge. Unfortunately, such information is currently only available from destructive tests.

Your PhD will pave the way towards in-situ microstructure characterisation using ultrasound and help transform common inspection practice. You'll employ artificial intelligence to support developing the physical understanding of how microstructure reveals itself in ultrasonic backscatter.

Using GPU-powered ultrasound simulations, you will run a large set of virtual experiments for numerous synthetically created microstructures with carefully controlled parameters. Learning from data, including choosing a suitable AI methodology, and proposing in-situ characterisation methods is the promise of this project.

You will use Southampton's High-Performance Computing facility, as well as advanced material characterisation techniques in the School of Engineering. The true potential of the methods you propose will be showcased using ultrasonic measurements on samples of industrial relevance.

As a PhD student, you will be a part of the Dynamics Group within the University's Institute of Sound and Vibration Research. You will contribute to developing the exciting capability in AI-powered NDT, working alongside researchers using acoustics to interrogate and characterise structures and materials of different scales and complexities.