Location

La Jolla, CA

Start Date

1979 12:00 AM

Description

The objective of this work has been to demonstrate the feasibility of estimating automatically the size and orientation of subsurface defects in metals. The approach has been to (1) obtain computer-generated spectra from various elastic scattering theories, (2) use these spectra to train empirical nonlinear Adaptive Learning Network (ALN) models, and (3) evaluate the theoretically trained ALN's on eight physically recorded defect specimens via a blind test. The results demonstrate that very good defect characterization is possible and that a fully automatic and general purpose NDE system can be implemented. An average orientation error of 10.2 degrees has been achieved and the defect average volume error is 17.5 percent.

Book Title

Proceedings of the ARPA/AFML Review of Progress in Quantitative NDE

Chapter

10. Inversion of Data Based on Elastic Wave Scattering Theory

Pages

341-367

Language

en

File Format

application/pdf

Share

COinS
 
Jan 1st, 12:00 AM

Inversion of Physically Recorded Ultrasonic Waveforms Using Adaptive Learning Network Models Trained on Theoretical Data

La Jolla, CA

The objective of this work has been to demonstrate the feasibility of estimating automatically the size and orientation of subsurface defects in metals. The approach has been to (1) obtain computer-generated spectra from various elastic scattering theories, (2) use these spectra to train empirical nonlinear Adaptive Learning Network (ALN) models, and (3) evaluate the theoretically trained ALN's on eight physically recorded defect specimens via a blind test. The results demonstrate that very good defect characterization is possible and that a fully automatic and general purpose NDE system can be implemented. An average orientation error of 10.2 degrees has been achieved and the defect average volume error is 17.5 percent.