A REVIEW OF ARTIFICIAL-INTELLIGENCE APPLIED TO ULTRASONIC DEFECT EVALUATION

The main objective in ultrasonic defect evaluation is to locate and cl assify suspect flaw indications quickly and accurately. Since the volu me of data to be assessed can be very large, traditional forms of defe ct evaluation involving a skilled human interpreter are often unsuitab le. The progress in the automated evaluation of ultrasonic data has be en considerable in recent years and this paper outlines some of the ap proaches adopted in this area. Traditional pattern recognition techniq ues and the currently popular neural network approaches have been wide ly employed to process feature sets, extracted from A-scan signals. Kn owledge-based system techniques, although not so widespread, are also considered. A number of authors have taken the approach that such AI t echniques should be embedded in an integrated software framework for d efect evaluation, and this is also discussed.