Ultrasound Defect Detection Using Fuzzy Logic
L.V.Guimarães and A.W.Elbern, Dep. de Engenharia Nuclear, UFRGS, Porto Alegre, RS, Brasil
ABSTRACT The automation of industrial ultrasonic inspection has been subject of large research in the last years. This paper presents a technique to detecting defects using threshold to distingue significant defects from others like slag inclusions or porosity for example. This technique is one of the most used, but it can easily lead to errors of interpretation, like detecting excessive defect points, most of them corresponding to false defects, or detecting only few points without detecting the existing real defects. Beside the existence of other digital techniques, we chose to use Fuzzy Logic ( FL ) to perform an automatic threshold detection system. This work we based the FL on the amplitude of the pulse signal and the distance between the last two significant adjacent A-scan signals. This information is then fuzzyfied and submitted to rules to decide if on the analyzed point exists or not a real defect. The used technique and our developed software show results that recommend the use of FL by its efficiency and reliability to be detect points of defects in ultrasonic signals.
Publication Source: NDTISS '99 - International Symposium on NDT Contribution to the Infrastructure Safety Systems, Nov 1999 in Tores Brazil.
Publisher: Center of Tecnology, Federal University of Santa Maria (UFSM) - [Homepage]
© NDTnet
