Model-based interpretation of experimental eddy current signals obtained from steam generator tubes by bobbin probe
| Autor: | Y. H. Kim, E.-L. Kim, Y. H. Choi, S.-J. Song |
|---|---|
| Rok vydání: | 2003 |
| Předmět: |
Engineering
Artificial neural network Bobbin business.industry Mechanical Engineering Acoustics Metals and Alloys Rotational symmetry Boiler (power generation) Estimator Finite element method law.invention Probabilistic neural network Mechanics of Materials law Materials Chemistry Eddy current Electronic engineering business |
| Zdroj: | Insight - Non-Destructive Testing and Condition Monitoring. 45:337-343 |
| ISSN: | 1354-2575 |
| DOI: | 10.1784/insi.45.5.337.52878 |
| Popis: | Model-based interpretation tools for eddy current signals have been developed by the novel combination of neural networks and finite element modelling for quantitative flaw characterization in steam generator tubes. In the present work, interpretation of experimental eddy current signals was carried out in order to verify the developed inversion tools. A database was constructed using synthetic eddy current signals generated by the finite element models. The hybrid neural networks of a probabilistic neural network (PNN) classifier and back propagation neural network (BPNN) size estimators were trained using the synthetic signals. Experimental eddy current signals were then obtained from axisymmetric artificial flaws fabricated in steam generator tubes. Estimation of the flaw parameters was carried out by feeding experimental signals into the neural networks. The excellent performance obtained in the present work demonstrates the high potential of the developed inversion tools as a practical interpretation of experimental eddy current signals. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|