ACCURACY IMPROVEMENT OF SMALL DEFECT DETECTION FOR ULTRASONIC INSPECTION BY USING SCIENTIFIC VISUAL ANALYSIS

When we diagnose the structural integrity of a steel member in service and evaluate its remaining life time, we need to improve analyzing method to get the accurate information of a internal defect of a member. And by the recent demand for high level quality control of welding of steel joint the accuracy improvement of defect detection in the image display became essential in nondestructive evaluation (NDE)[l][2][3]. So far the defect informaUon obtained by an ultrasonic test is displayed in several ways and A-scan and B-scan display's are commonly used in a field inspection and C-scan display is used in laboratory test of the steel structural member [ 4][5]. In this study we made the different types of artificial small defects by drilled hole in a steel plate specimen to observe by an ultrasonic scanning pulse echo test[6]. The artificial drilled holes with diameter 1mrn and 2mrn are made with different lengths and inclinations and the graphic images are compared among these cases. 3D images of these defect are obtained by a graphic software[7] on a workstation by using the maximum amplitude data of boundary and bottom echo and time of flight data by which we find from where the incident wave returned back. And in order to apply the C-scope immersion method of ultrasonic test to a field inspection, the effects of surface conditions of specimen are an81yzed. The accurate defect size obtained by the ratio of the maximum amplitudes between the boundary echo and bottom echo is shown in the detail waveform analysis of the reflection wave by using Scientific Visual Analysis(SVA)