Document Details

Use of Monte-Carlo methods to derive quantitive probability of defect detection data
Author:	Walker, Tony
Co-Author(s):

Safety is of paramount importance in the nuclear industry and accordingly the most stringent standards of manufacture, analysis and quality have to be applied to components such as the Reactor Pressure Vessel. A key discipline to achieving and demonstrating that these standards have been met is that of Non-Destructive Evaluation (NDE), and the reliability of the NDE methods and techniques themselves is of great importance. There is now a regulatory expectation that the NDE of major nuclear vessels requires some level of validation. Validation of inspection system performance has now been developed into a formal process termed Inspection Qualification. The concept of' probability of defect detection' or PoD of an inspection system is quite intuitive and has proven a very difficult quantity to determine satisfactorily. Attempts to establish empirical values of PoD have been largely unsatisfactory due to the fact that it is simply not practical to conduct a sufficient number of 'tests' to establish any values with a reasonable degree of statistical confidence. This paper describes an approach to the determination of PoD data developed within Rolls-Royce Submarines. The RR PoD methodology relies on a statistical model of inspection performance based upon the fundamental principles of the inspection system (the so called Essential Parameters) and circumvents much of the subjectivity of other alternative approaches.

ISBN:
Price:	£21.28 (£21.28 inc.)
Add to Basket