An intelligent system approach to real-time ultrasonic flaw classificationin weldments

Various advanced ultrasonic flaw classification approaches have been propos ed for the determination of flaw types in weldments. Among them, ultrasonic pattern recognition approaches have seemed to be most promising, and thus extensive investigations have been carried out in the ultrasonic nondestruc tive evaluation community, In spite of these extensive endeavors, these app roaches have not been widely used in the current industrial field applicati ons due to some critical barriers in cost, time and reliability. To reduce such barriers, here we propose an intelligent system approach realized by t he novel combination of following four ingredients such as 1) development o f a PC-based real-time ultrasonic testing system, 2) construction of an abu ndant experimental database of ultrasonic flaw signals in weldments, 3) est ablishment of an invariant ultrasonic pattern recognition algorithm by use of newly proposed normalized features, and finally 4) implementation of thi s invariant algorithm in the form of intelligent flaw classification softwa re. In this paper we address in detail the proposed approach including its four ingredients and its capability of malting the industrial field applica tion of ultrasonic pattern recognition approaches low cost, rapid and relia ble in their performance. The performance of this approach has been investi gated with abundant ultrasonic signals captured from various flaws in weldm ents with variation in the operational variables of the ultrasonic testing. Especially, we have examined the capability of this approach for reducing the effect of the operational variables on the classification performance i n terms of the class-conditional probability density functions and correct accept rates. The reasonable and consistent performances obtained in this w ork demonstrate the high potential of this approach to serve as a convenien t and robust tool for many real world flaw classification problems in weldm ents.