Electromagnetic Nondestructive Evaluation of Tubes using Data Mining Procedure

International Conference on Innovative Research - ICIR Euroinvent 2017

Savin, A.^1 ; Iftimie, N.^1 ; Vizureanu, P.^2 ; Steigmann, R.^1 ; Dobrescu, G.S.^1

National Institute of Research and Development for Technical Physics, Nondestructive Testing Department, Iasi, Romania^1

Gheorghe Asachi Technical University, Faculty of Materials Science and Engineering, Iasi, Romania^2

关键词: Differential sensors;    Electromagnetic nondestructive evaluation;    Features extraction;    Forward problem;    Non destructive evaluation;    Simulated model;    Standard samples;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/209/1/012005/pdf
DOI  :  10.1088/1757-899X/209/1/012005

来源: IOP

【 摘 要 】

The fundamental issues in nondestructive evaluation consists in the identification of events corresponding to the flaws which can appear in the examined object and their extraction from noises. This is usually done by comparison with pre-established thresholds, experimentally determined by using standard samples or in the basis of the solution of the forward problem and simulations. This paper presents the features extraction using data mining procedure in the case of tubes from steam generators having different flaws. The data mining is carried on using simulated models in CIVA 9 and experimental data gathered using an inner differential sensor developed in this purpose.