Neural network system identification in noise-amplifier flows: an experimental study using optical-flow PIV data
Résumé
A neural-network system identification method along with a standard proper orthogonal decomposition was implemented to identify dynamical systems of typical noise-amplifier flows: the transitional flat plate boundary layer in the presence of the Tollmien-Schlichting instability and the backward-facing step flow. The influence of the sensor nature and the number of snapshots on the success of the network training and validation is discussed.
Origine : Fichiers produits par l'(les) auteur(s)
Loading...