This thrust focuses on scientific machine learning methods for understanding structural and dynamical systems from data.
The emphasis is on combining data-driven learning with mechanics, governing principles, physical constraints, and computational models so that machine learning methods remain meaningful for engineering analysis and infrastructure monitoring.
What We Study#
Scientific Machine Learning for Engineering Systems
Methods that combine machine learning with physical principles, numerical modelling, and engineering knowledge for structural and infrastructure applications.
Interpretable AI for Dynamical Systems
Learning approaches that aim to make predictions and representations more interpretable in terms of structural behaviour, physical variables, and engineering mechanisms.
Physics-Guided Learning
Machine learning methods that use physical constraints, structural mechanics, and engineering insight to improve reliability, robustness, and generalization.
Computational Modelling and Data
Approaches that connect simulation, measurement, and learning to support model interpretation, monitoring, and decision support in civil infrastructure systems.
Methods and Tools#
Research in this thrust may involve differential equations, structural dynamics, numerical methods, optimization, machine learning, scientific computing, and Python/MATLAB programming.
The emphasis is on developing AI/ML tools that are not only accurate, but also interpretable, physically meaningful, and suitable for engineering use.
Student Background#
Students interested in this thrust may benefit from background in differential equations, structural dynamics, numerical methods, machine learning, optimization, Python/MATLAB programming, or scientific computing.
It is not necessary to have expertise in all areas. Specific topics are shaped based on the student’s background, interests, and expected time commitment.
Interested Students#
Students interested in scientific machine learning, interpretable AI, structural dynamics, computational modelling, or AI for engineering systems are encouraged to read the broader Research page and contact me through the Join the Group page.
Specific project topics are discussed individually after understanding the student’s background, interests, and available research opportunities.