The defect detection and non-destructive evaluation in weld zone of austenitic stainless steel 304 using neural network ultrasonic wave

In recent years, the importance of non-destructive evaluation has rapidly i ncreased due to the collapse of large structures and the shooting up of saf ety accidents. The ultrasonic method, which is often used as a major non-de structive testing (NDT) technique in many engineering fields, is praying a significant role as a volumetric test regarded highly for evaluating a mate rial's integrity. This paper is recommended for publication the detecting a ny defects of the weld zone in austenitic stainless steel type 304 using ul trasonic waves, employing neural network on the basis of the detected defec ts and evaluating them. In detecting defects, we drew a distance amplitude curve on a standard scan sensitivity and a preliminary scan sensitivity sho wn in the correlation between the ultrasonic probe, the instrument and the materials on a quantitative standard, and quantitative evaluation is used t o draw a distance amplitude curve. A total of 93.3% of the defect types was distinguished by testing 30 defects after organizing a neural network syst em based on the defects on the ultrasonic evaluation and learning the neura l network system. Thus the proposed ultrasonic wave-neural network in this work is useful for defects detection and Ultrasonic Non-Destructive Evaluat ion (UNDE) of the weld zone of austenitic stainless steel 304.