There are several sources of uncertainties which need to be considered in a probabilistic reliability and lifetime assessment of safety-relevant components. In addition to the probabilistic distribution of material properties, the size and properties of flaws present in a component contribute to uncertainties in the lifetime analysis. In a current reactor safety research project, a methodology for a probabilistic fracture mechanics assessment of reliability for components with austenitic and dissimilar metal welds will be developed using the combined results from statistical evaluation of material properties and ultrasonic inspection (UT). Dissimilar metal welds present a particular challenge for ultrasonic testing due to the effects of the material anisotropy on ultrasonic propagation and scattering. Evaluation of inspection results is typically done using image-based techniques. As there is no simple relationship between UT response and flaw size, criteria for image evaluation and flaw sizing have to be defined taking into account the influence of various factors such as material and flaw properties on the UT response. In this paper, we present results from a study of the influence of grain structure on the ultrasonic inspection results and discuss the challenges of extracting data for probability of detection (POD) analyses.

This content is only available via PDF.
You do not currently have access to this content.