Researchers from The University of Western Australia, the University of Cambridge and The Alan Turing Institute have radically redesigned and improved techniques for making predictions in the engineering and physical sciences.

The research, published in the prestigious Proceedings of the National Academy of Sciences (PNAS), builds on the well-known Finite Element Method (FEM), which has been used as a predictive tool in engineering and physical sciences for more than 70 years.

Connor Duffin, PhD student from UWA’s School of Physics, Mathematics and Computing and lead researcher on the project, said it was that the first time a key missing ingredient was included in observed data.

“It's a new combination of data and mathematical models that enhances predictions in an extremely powerful way."

Connor Duffin, PhD student

“The research demonstrates the method in the context of internal ocean waves, which regularly occur on Australia’s North West Shelf," Mr Duffin said.

"These waves have a significant impact on the engineering design, safety and operations of Australia’s offshore energy industry and improved methods of prediction provides significant benefit.”

Co-author Professor Mark Girolami (Royal Academy Chair in Data Centric Engineering at the University of Cambridge and Programme Director for Data-Centric Engineering at The Alan Turing Institute) said the research had commercial interest.

“Digital Twins – the pairing of the physical and virtual world – is of significant current interest to the broader engineering community. By integrating data with FEMs, this new work provides the foundation and methodology by which these Digital Twins can be realised,”Professor Girolami said.

“It also serves as an ideal springboard into the forthcoming ARC Industrial Transformation Research Hub for Transforming Energy Infrastructure through Digital Engineering, hosted at UWA and directed by Shell Professor Phil Watson, with which The Alan Turing Institute is very excited to collaborate.”