deep learning

Remaining Useful Life Estimation Under Uncertainty with Causal GraphNets

A proposal for large-scale processing for wind turbine monitoring.

Charilaos Mylonas, Eleni Chatzi

Remaining Useful Life Estimation Under Uncertainty with Causal GraphNets

Relational VAE: A Continuous Latent Variable Model for Graph Structured Data

Motivated by the need for edge uncertainty in modeling wind farm wakes, a graph latent variable model that allows for full graph conditioning and graph latent variables is proposed.

Charilaos Mylonas, Imad Abdallah, Eleni Chatzi

Relational VAE: A Continuous Latent Variable Model for Graph Structured Data

Structural identification with physics-informed neural ordinary differential equations

Using NeuralODEs to deal with non-linear discrepancies in system identification.

Zhilu Lai, Charilaos Mylonas, Satish Nagarajaia, Eleni Chatzi

Bayesian graph networks for strain-based crack localization

Implemented Graph Networks with weight uncertainty for crack localization while using arbitrarily positioned strain sensors with good results.

Charilaos Mylonas, George Tsialiamanis, Keith Worden, Eleni Chatzi

Bayesian graph networks for strain-based crack localization

Conditional variational autoencoders for probabilistic wind turbine blade fatigue estimation using Supervisory, Control, and Data Acquisition data

Using conditional latent variable models for wind farm monitoring data.

Charilaos Mylonas, Imad Abdallah, Eleni Chatzi

Conditional variational autoencoders for probabilistic wind turbine blade fatigue estimation using Supervisory, Control, and Data Acquisition data

Deep unsupervised learning for condition monitoring and prediction of high dimensional data with application on windfarm SCADA data

Using conditional latent variable models for wind farm monitoring data.

Charilaos Mylonas, Imad Abdallah, Eleni Chatzi

Deep unsupervised learning for condition monitoring and prediction of high dimensional data with application on windfarm SCADA data