A Deep Learning Framework for Acoustic Emission Sources Localization and Characterization in Complex Aerospace Panels
Abstract
This paper presents a data-driven approach based on deep stacked autoencoders for the localization and characterization of acoustic emission sources in complex aerospace panels. The approach leverages the multimodal and dispersive reverberations of acoustic emissions. The approach is validated by Hsu-Nielsen pencil lead break tests on a fuselage section of a Boeing 777 instrumented with a single piezoelectric sensor.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- Apr 01, 2021
- Source ID
- 10.32548/2021.me-04179
Entities
People
- Arvin Ebrahimkhanlou
- Brennan Dubuc
- Melanie Schneider
- Salvatore Salamone
Organizations
- Office of Naval Research Global