Machine learning in metallurgy

24 May 2021 1pm – 2.30pm


Machine learning in metallurgy is increasingly becoming of interest by organisations to aid discovery and design of metallic materials. Featuring presentations from academia and industry, this webinar is focused on understanding the use of machine learning in metallurgy, it’s impact on industry and the development of machine learning in the UK.

Event Programme

The Modern Day Blacksmith

Delivered by Gareth Conduit, Chief Technology Officer at Intellegens, this presentation will focus on the latest developments in machine learning technologies and you will learn about how to tailor machine learning for materials discovery and its practical application to commercial materials design.

Feature extraction and stereological characterisation of directionality solidified

Delivered by Joel Strickland, PhD Student at Leicester University, this presentation is focused on the application of an automatic and standardised feature detection and comprehensive single crystal characterisation methodology. You will learn the process by which the solid-liquid isotherm shape can be deduced from the microstructure by quantifying microscopic changes (in this case, local primary spacing between adjacent dendrites) over macroscopic length scales.

Advancing digitalisation through Servitization - Creating value for our Customers

Delivered by Raj Balesar, New Products Commercialisation Manager at Tata Steel Europe, this presentation will highlight, through one example currently underway at TSE, what the barriers to servitization are and how we are dealing with these.

What can Machine Learning teach us about phase transitions?

Delivered by Cinzia Giannetti, Associate Professor at Swansea University, this presentation will focus on the identification of phase transitions with supervised Machine Learning methods and, taking the Ising model as a prototype, you will learn how to obtain a quantitative estimation of the critical temperature from Monte Carlo generated configurations without any previous knowledge of the system.