POD analysis of defect in radiographic testing of metal additive parts
In view of vulnerability to cracks and pore defects due to lacking of defect detection probability data in radiographic examination of additive manufacturing,the linear defects and pore defects of GH3625 superalloy additive parts were researched.The CIVA2020 simulation platform was used to simulate X-ray inspection and obtain the probability of detection(POD)curve of defects,research the influence of different size changes of the two types of defects on the POD of defects.The detectable size of defects and POD under different influencing factors were determined,and the POD curve equation of linear defect affected by depth and the POD curve equation of pore defect affected by radius were fitted by Sgompertz function,the defect detection probability model of additive manufacturing line defects and pore defects was established respectively.The detectable length size of linear defects was 0.211 mm,the detectable width size was 0.213 mm,the detectable depth size was 0.178 mm at a probability of 90％under 95％confidence level,the detectable diameter size of pore defects was 0.188 mm,and the detectable height size was 0.190 mm.The simulation results were compared and verified by micro-focus radiography and film radiography of actual specimens.The research results showed that the defect detection probability model is more accurate and can provide a basis for reliability analysis of crack and pore defect detection in additive manufacturing.
additive manufacturingradiographic testingprobability of detectionlinear defectpore defect
NETL
NSTL
万方数据
