In this study, machine learning algorithms are trained and compared to identify and characterise impacts effects on typical aerospace panels with different geometries. Experiments are conducted to create a suitable impact dataset. Polynomial regression algorithms and shallow neural networks are applied to panels without stringers and optimised to test their ability to identify the impacts. The algorithms are then applied to panels reinforced with stringers, which represents a significant increase in complexity in terms of the dynamic characteristics of the system under test. The focus is not only on the detection of the impact position, but also on the severity of the event. The aim of the work is to demonstrate the validity of the application of machine learning to impact localization on realistic structures and to demonstrate the simplicity and efficiency of the computations despite the complexity of the test specimens.

Impact detection on thin structures via machine learning approaches

Dipietrangelo F.;Nicassio F.
;
Scarselli G.
2023-01-01

Abstract

In this study, machine learning algorithms are trained and compared to identify and characterise impacts effects on typical aerospace panels with different geometries. Experiments are conducted to create a suitable impact dataset. Polynomial regression algorithms and shallow neural networks are applied to panels without stringers and optimised to test their ability to identify the impacts. The algorithms are then applied to panels reinforced with stringers, which represents a significant increase in complexity in terms of the dynamic characteristics of the system under test. The focus is not only on the detection of the impact position, but also on the severity of the event. The aim of the work is to demonstrate the validity of the application of machine learning to impact localization on realistic structures and to demonstrate the simplicity and efficiency of the computations despite the complexity of the test specimens.
2023
978-1-6654-5690-6
File in questo prodotto:
File Dimensione Formato  
Impact detection on thin structures via machine learning approaches.pdf

solo utenti autorizzati

Descrizione: Atto di convegno
Tipologia: Versione editoriale
Licenza: NON PUBBLICO - Accesso privato/ristretto
Dimensione 2.33 MB
Formato Adobe PDF
2.33 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11587/502727
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact