Machine Learning for Physical Systems
Machine learning for physical systems integrates machine learning with engineering mathematics, to optimise design, operation, and maintenance. Research at the University of Glasgow is developing statistical tools to fuse data-driven models with process understanding, to better represent systems in operation, from bridges to wind turbines.
- Structure-constrained or informed machine learning
- Experimental design and active learning
- Condition and performance monitoring
Researchers
Publications
- Towards Multilevel Modelling of Train Passing Events on the Staffordshire Bridge, ArXiv (2024).
- Data-Centric Monitoring of Wind Farms, In Data-Centric Monitoring of Wind Farms (2023).
- Encoding Domain Expertise into Multilevel Models for Source Location, ArXiv (2023).
- Hierarchical Bayesian modelling for knowledge transfer across engineering fleets via multitask learning, Computer‐Aided Civil and Infrastructure Engineering, 38(7) (2022).
- A sampling-based approach for information-theoretic inspection management, Proceedings of the Royal Society 478(2262) (2022).
Links