Colloquium Elizabeth Cross

A spectrum of physics-informed machine learning approaches for structural health monitoring

Structural health monitoring as a research field began by studying physical models for systems, comparing modelled and monitored response to understand observed behaviours. As sensing technology advanced and monitoring data became more readily available, many practitioners and researchers adopted a purely data-driven approach capitalising also on technology enhancement from the machine learning field. Today, where we would like to be able automatically assess the health of our structures across their operational envelope, we find that despite these advances, we often lack data that represent all behaviours of interest, thereby precluding an entirely data-driven approach.

This talk will present a number of ways of incorporating the physical insight an engineer will often have of the structure they are attempting to model or assess into a machine learning approach as a potential solution to this problem. The talk will demonstrate how grey-box models, that combine simple physics-based models with data-driven ones, can improve predictive capability for structural assessment and system identification tasks. A particular strength of the approaches demonstrated here is the capacity of the models to generalise, with enhanced predictive capability in different regimes, increasing applicability in light of the aforementioned challenge. 

Lizzy Cross is a Professor in the Dynamics Research Group at the University of Sheffield with a research focus on advanced data analysis and machine learning for structural health monitoring (SHM) and nonlinear system identification. She has just completed an EPSRC Innovation Fellowship pioneering physics-informed machine learning for structural dynamics. Lizzy is a co-director of the Laboratory for Verification and Validation, a state-of-the-art dynamic testing facility (lvv.ac.uk). She has published over 140 articles, including 45 journal papers, 6 invited book chapters. She was recently awarded the Achenbach medal which recognises an individual (within 10 years of PhD) who has made an outstanding contribution to the advancement of the field of SHM.