Thema: neuronales Netz, Ultraschalldaten, Rißgröße
Autor/Institution: Zgonc, K.;Achenbach, J.D.;Lee, Y.-C. /Northwestern Univ., Evanston, USA

Title: Crack sizing using a neural network classifier trained with data obtained from finite element models
Titel : Rißgrößenbestimmung mit einem neuronalen Netzwerk gespeist mit Daten aus Finite-Element-Modellen

Quelle : Review of Progress in Quantitative Nondestructive Evaluation. Proceedings of the Twenty-First Symposium, Snowmass Village, USA, 31.7.-5.8.1994 13 14A 1995 ; New York: Plenum Press ; ISBN 0-306-45062-3 ; S.779-786 ; 8S,9B,2T,4Q ; TIB-RO141(14,1);BAM-Z-0906/95 ; E

Abstract:

An ultrasonic data processing system has been developed and applied to the detection and sizing of EDM notches emanating from rivet holes. A shift- invariant neural network is successfully used to classify R/T curves for different EDM notches that are used to model cracks. A set of training data for the neural network (NN) ist obtained by 2-D FE models. The results show that numerical results can be applied to train the NN classifiers and that the ultrasonic data processing system gives a good prediction of the EDM notch sizes. To obtain even better generalization capabilities of the classifier would require a larger number of numerically obtained data. The data processing system is general and can be applied also to other nondestructive testing problems. In practice, plate waves are used to perform inspection around riveted joints. Therefore, future work will concentrate on 3-D FE modeling which will more accurately simulate real applications. F33,B13,H13,A25

Database Keywords: :Ultraschallprüfung, Signalverarbeitung, Finite Element Methode, :Fehlergröße, Riß, Testfehler 23 :Neuronales Netzwerk, Funkenerodierter Schlitz 24 Z:A


File Producer: FWD Förderverein Werkstoffdokumentation e.V. (DGZfP, BAM, FIZ Technik)
Brief Information about FWD: Virtual Library about FWD
| Frontpage | | Abstracts List | |Virtual Library: Archives and Databases|

Rolf Diederichs 1. Dec 1996, info@ndt.net