NDTnet 1998 Aug, Vol.3 No.8

Abstract

The MISTRAL project is financed by the Brite Eurarn Programme of the European Commission. The consortium of the project gathers Electricite de France (France), the University of Cagliari (Italy), Steria (France), the University of Gent (Belgium), Siemens KWU (Germany), Tecnatorn SA (Spain), TNO Institute of Applied Physics (Netherlands), the Instituto de Soldadura e Qualidade (Portugal) and Quimigal Adubos (Portugal).

Introduction to the MISTRAL project
Chemical and Energy producers using Non Destructive Testing have a strong economical need for improving the accuracy of the detection and characterisation of defects in steel components without impacting onto the price of their production. The MISTRAL project proposes to design, develop and evaluate multi-sensor approaches on welded components shared by most of the companies in these fields.

The MISTRAL multi-sensor approach for NDT consists of dedicated multi-technique NDT probes, a set of new processing tools merging the information acquired and an efficient use of fracture mechanics code in conjunction with standard acceptance criteria. At any stage of the NDT process, MISTRAL aims at bringing some added-value coming from the complementary aspects of two different types of NDT data. MISTRAL demonstrates the efficiency of such a combined approach for radiography and ultrasound on the one side and for ultrasound and Eddy current on the other side. The fusion of various sources of information is to allow substantial improvements in detection and characterisation of defects whereas the use of a combined probe avoids double sequential acquisition, saving time as a consequence.

Acquisition and fusion of NDT information
The first result of the project concerns the development of a new Eddy Current (EQ and Ultrasonic (UT) sensor which are adapted to different types of components. The EC part is a new type of probe whereas the UT part is classically a pitch and catch probe. The fusion architecture retained so far is decentralised [Gau97] includes an adaptation of tools using the Dempster Shafer Theory [Gro94] for fusing information which provides promising results.

Reference
[Gro94] X Gros, Stracham, Lowden, Edwards, "NDT data fusion"pp355-360 ECNDT 94.
[Gau97] S. Gautier, B. Lavayssiere, E. Fleuet, J. Idier, "Using complementary type of Data for 3D flaw imaging", QNDE San Diego, July 1997 (to be published).

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