Speaker: Ben Pellegrini, CEO, Intellegens

Location: AM4U stage (11.0-E72)

Timing: 18 November, 12:45-13:00

Can machine learning help to accelerate AM innovation? AM project teams want to improve product performance and ensure reliable, repeatable processes. Material properties, operating conditions, and process settings interact to impact these objectives in complex and subtle ways. Understanding these effects requires expensive testing and analysis. But can we use existing data about these factors to find key property / process relationships and design efficient experimental programs? Machine learning should be an ideal tool to do this. But machine learning methods struggle when they try to build useful models from real-world, sparse and noisy experimental data. Cambridge University spinout Intellegens has developed novel deep learning methods that train models on sparse, noisy data to extract value from AM project data. The technology has been validated including through materials design work at Rolls-Royce and GKN Aerospace, and the on-going Project MEDAL, working with the UK’s Advanced Manufacturing Research Centre (AMRC) and global aerospace giant Boeing. We’ll show how this technology can help AM teams to to find out which process parameters are really controlling target properties
, choose which experiments to do next so that results can be achieved with far fewer tests, get a quantitative understanding of uncertainties to focus on the most reliable routes, and
 predict the likely performance of candidate processes – avoiding costly experiments that are likely to fail.

• Alchemite™ for Additive Manufacturing
• An overview of the Alchemite™ Technology
• Case studies

Subscribe to our newsletter to get the latest news on our other upcoming events.