Knowledge-based support in Non-Destructive Testing for health monitoring of aircraft structures

Abstract

Maintenance manuals include general methods and procedures for industrial maintenance and they contain information about principles of maintenance methods. Particularly, Non-Destructive Testing (NDT) methods are important for the detection of aeronautical defects and they can be used for various kinds of material and in different environments. Conventional non-destructive evaluation inspections are done at periodic maintenance checks. Usually, the list of tools used in a maintenance program is simply located in the introduction of manuals, without any precision as regards to their characteristics, except for a short description of the manufacturer and tasks in which they are employed. Improving the identification concepts of the maintenance tools is needed to manage the set of equipments and establish a system of equivalence: it is necessary to have a consistent maintenance conceptualization, flexible enough to fit all current equipment, but also all those likely to be added/used in the future. Our contribution is related to the formal specification of the system of functional equivalences that can facilitate the maintenance activities with means to determine whether a tool can be substituted for another by observing their key parameters in the identified characteristics. Reasoning mechanisms of conceptual graphs constitute the baseline elements to measure the fit or unfit between an equipment model and a maintenance activity model. Graph operations are used for processing answers to a query and this graph-based approach to the search method is in-line with the logical view of information retrieval. The methodology described supports knowledge formalization and capitalization of experienced NDT practitioners. As a result, it enables the selection of a NDT technique and outlines its capabilities with acceptable alternatives.