We are delighted to welcome Dr Peter Green, University of Liverpool, as our next guest speaker.

Dr Peter Green became a lecturer in Uncertainty and Engineering at the UoL in 2015. His background is in structural dynamics, but he now develops uncertainty quantification and machine learning methods for engineering disciplines.

Peter obtained a MEng degree at the University of Sheffield before undertaking a PhD in nonlinear structural dynamics (2009-2012). In 2012, he won an EPSRC/University of Sheffield 1-year fellowship to study probabilistic modelling methods. He was then employed on the EPSRC Programme Grant ‘Engineering Nonlinearity’ (EP/K003836). Here, his research grew to encompass machine learning methods for engineering applications.

His current research sits between Big Data analytics, Machine Learning and multiple engineering disciplines. Application projects include span: data-based control strategies for Additive Manufacturing, machine-learnt rotorcraft dynamics models for deployment in flight simulators, robust optimisation of ship scheduling problems under uncertain weather conditions, characterising the risk of disproportionate collapse for cable-stayed bridges and analysing the robustness of structures subjected to blasts. Fundamental research addresses decision-making from large datasets and the validation of models in situations where data is sparse.

AI Approaches for Automatic Defect Detection in Laser Powder Bed Fusion Builds

In this talk we will discuss recent advancements in the automatic data-based detection of porosity in Laser Powder Bed Fusion (L-PBF) builds. Specifically, we will look at ‘AI’ approaches that, from photodiode measurements, can detect the onset of porosity in real-time. In the spirit of transparency, we will highlight how the proposed approach appears to work, but for almost exactly the opposite reasons to what was originally hypothesised! Finally, we will use the case study to discuss the general challenges that are associated with applying AI approaches in real industrial settings, and what steps we might need to take to avoid another ‘AI winter’ within a manufacturing context.

This event will be held via MS Teams - please register here.