A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Classification of acoustic emission signatures using a self-organization neural network
Acoustic emission (AE) testing is a promising technique for use in structural health monitoring. A critical factor for the successful implementation of this technique in both laboratory and field is the ability to achieve reliable source identification, which can be best achieved through adoption of pattern recognition techniques. In particular, there is a need for discrimination of AE signals from fatigue crack growth and extraneous AE in order to quantitatively evaluate fatigue crack growth. This paper presents a waveform descriptor-based classification of AE signals during laboratory fatigue testing of a full-scale steel bridge girder using a self-organization neural network. The pitfalls of AE signal classification based on automatically extracted waveform features are demonstrated through the consideration of typical signals from individual sources and the complications during the AE signal acquisition and waveform feature extraction process. Careful interpretation of the classification results is emphasized in order to identify the likely origin of AE data contained within each class.
Classification of acoustic emission signatures using a self-organization neural network
Acoustic emission (AE) testing is a promising technique for use in structural health monitoring. A critical factor for the successful implementation of this technique in both laboratory and field is the ability to achieve reliable source identification, which can be best achieved through adoption of pattern recognition techniques. In particular, there is a need for discrimination of AE signals from fatigue crack growth and extraneous AE in order to quantitatively evaluate fatigue crack growth. This paper presents a waveform descriptor-based classification of AE signals during laboratory fatigue testing of a full-scale steel bridge girder using a self-organization neural network. The pitfalls of AE signal classification based on automatically extracted waveform features are demonstrated through the consideration of typical signals from individual sources and the complications during the AE signal acquisition and waveform feature extraction process. Careful interpretation of the classification results is emphasized in order to identify the likely origin of AE data contained within each class.
Classification of acoustic emission signatures using a self-organization neural network
Klassifizierung von Schallemissionsmerkmalen mittels selbst organisierten neuronalen Netzen
Yan, Tinghu (author) / Holford, K. (author) / Carter, D. (author) / Brandon, J. (author)
Journal of Acoustic Emission ; 17 ; 49-59
1999
11 Seiten, 21 Quellen
Article (Journal)
English
Schallemissionsprüfung , akustische Signalverarbeitung , Rissprüfung , Ermüdungsriss , Merkmalextraktionsverfahren , Bildklassifikation , Muster , Cluster-Bildung , rechnerunterstützte Physik , Wellenformanalyse , Bilderkennung , neuronales Netz , selbstorganisierendes System , Ermüdungsrisswachstum
Classification of Acoustic Emission Signatures Using a Self-organization Neural Network
| British Library Online Contents | 1999
Pattern Recognition of Acoustic Signatures Using ART2-A Neural Network
| British Library Online Contents | 1996
Analysis of Artificial Acoustic Emission Waveforms Using a Neural Network
| British Library Online Contents
Neural Network Approach to Acoustic Emission Source Location
| British Library Online Contents | 1996
Acoustic emission source location based on artificial neural network
| British Library Conference Proceedings | 2022