Structural Health Monitoring
Research at the intersection of Scientific Machine Learning and structural dynamics. PhD work at Sapienza University of Rome.
Start Here
New to SHM? Begin with the Neural Networks for Dynamic Systems article, then explore the Paper Notes for specific methods. Each note follows a consistent 7-section template for easy comparison.
Paper Notes
Structured summaries with method snapshots and reproduction notes.
Physics-Informed Neural Networks for Structural Damage Detection
PINNs enforce governing equations as soft constraints during training, enabling damage detection with minimal labeled data.
Read note →Neural ODEs for Structural Damage Identification
Neural ODEs offer a principled way to embed physical priors into damage identification pipelines for nonlinear structures.
Read note →Bayesian Inference for Damage Localization Under Uncertainty
Bayesian model updating provides a principled framework for damage localization that naturally quantifies uncertainty in identified structural parameters.
Read note →Simulations
Interactive experiments applying SHM methods to physical systems.
Nonlinear Damage Simulator
Simulate damage progression in a nonlinear structural system using Neural ODEs.
Coming soonArticles
Long-form writing connecting SHM theory to practice.
Computational Fluid Dynamics with Machine Learning
Exploring how machine learning is revolutionizing computational fluid dynamics simulations.
Read article →Neural Networks for Dynamic Systems
An exploration of how neural networks can model and predict complex dynamical systems in engineering applications.
Read article →