NDTnet 1998 Aug, Vol.3 No.8

Abstract

There is a need for ultrasonic data inversion schemes capable of providing reliable results concerning the type of a defect and its geometrical characteristics. Concern has been expressed about the capabilities of performing Non Destructive Evaluation of the Outer Surface Defects (OSD's), i.e., defects located in the vicinity of the outer surface, in nuclear PWR's vessels. The OSD's are insonified by both a "direct" field that passes through the inner surface (water/steel) of the component containing the defect and a "secondary" field reflected from the outer surface. Consequently, the Bscan images containing the signature of such defects are complicated and their interpretation is a difficult task.

Ultrasonic equivalent defect sizing model-based methods may be used to size an OSD in a material by obtaining a best-fit simple equivalent shape that matches the ultrasonic observed data. These methods usually involve the solution of a nonlinear optimization problem, which can be ill-behaved and difficult to solve in practice. Moreover, they requires the availability of defect classification information (i.e., if the defect is volumetric or planar, e.g. a crack or a lack of fusion), which, generally, may be as difficult to obtain as the defect parameters themselves.

In this paper, we deal with global processing of the information-bearing Bscan images, in order to extract from them the maximum available information on the defects. Our main objectives are to determine the type of an OSD and its geometric parameters by means of a two-step inverse approach. So, we illustrate the application of a model-based method for equivalent OSD's sizing [1] from segmented Bscan images. Then, we describe an estimation technique for extracting relevant characteristics on the type of the defect (defect classification information) from some selected Ascan signals of the original Bscan image.

The proposed two-step inverse approach is based on two direct models : a system model developed at the French Atomic Energy Commission (the Mephisto simulator) [2] and a parametric model associated to the echoes formation mechanism. We show that this inverse approach is a very useful tool in non destructive evaluation and have valuable practical applications. Characterization results for both artificial and actual defects are given.

1. Faur, M., Roy, 0., Benoist, Ph., Morisseau, Ph., Rev. Prog. QNDE, 1996, 16: 67-74.
2. Calmon, P., Roy, 0., Rev. Prog. QNDE, 1994,13: 101-108.

Abstract Source:
Book of Abstracts, 7th European Conference on Non-Destructive Testing, 26-29 May 1998, ISBN: 87-986898-0-00

Full-Text Source:
Proceedings of the 7th European Conference on Non-Destructive Testing, 26-29 May 1998, ISBN:

Publication Contact:
7th ECNDT, Park Alle 345, DK-2605 Broendby, Denmark, Fax: +45 46 26 70 11, Email: 101373.3414@compuserve.com