Location

Brunswick, ME

Start Date

1-1-1992 12:00 AM

Description

A neural network approach has been developed to determine the depth of a surface breaking crack in a steel plate from ultrasonic backscattering data. The network is trained by the use of a feedforward three-layered network together with a back-propagation algorithm for error corrections[1,2]. The signal used for crack insonification is a mode converted 45° transverse wave. The plate containing a surface breaking crack is immersed in water and the crack is insonified from the opposite uncracked side of the plate. A numerical analysis of the backscattered field is carried out based on elastic wave theory, by the use of the boundary element method. The numerical data are calibrated by comparison with experimental data. The computed backscattered field provides synthetic data for the training of the network. The training data have been calculated for cracks with specified increments of the crack depth. The performance of the network has been tested on experimental data for cracks of different depths than used for network training.

Book Title

Review of Progress in Quantitative Nondestructive Evaluation

Volume

11A

Chapter

Chapter 3: Interpretive Signal Processing and Image Reconstruction

Section

Neural Networks

Pages

701-708

DOI

10.1007/978-1-4615-3344-3_90

Language

en

File Format

application/pdf

Share

COinS
 
Jan 1st, 12:00 AM

Neural Network for Crack-Depth Determination from Ultrasonic Backscattering Dat

Brunswick, ME

A neural network approach has been developed to determine the depth of a surface breaking crack in a steel plate from ultrasonic backscattering data. The network is trained by the use of a feedforward three-layered network together with a back-propagation algorithm for error corrections[1,2]. The signal used for crack insonification is a mode converted 45° transverse wave. The plate containing a surface breaking crack is immersed in water and the crack is insonified from the opposite uncracked side of the plate. A numerical analysis of the backscattered field is carried out based on elastic wave theory, by the use of the boundary element method. The numerical data are calibrated by comparison with experimental data. The computed backscattered field provides synthetic data for the training of the network. The training data have been calculated for cracks with specified increments of the crack depth. The performance of the network has been tested on experimental data for cracks of different depths than used for network training.