Improved Methodology for Predicting POD of Detecting Synthetic Hard Alpha Inclusions in Titanium

Abstract

This paper provides a progress report on the development of methodology to estimate Nondestructive Evaluation (NDE) capability. The methodology uses combinations of physical modeling of an inspection process, along with laboratory and production data, to estimate Nondestructive Evaluation (NDE) capability. The methodology is based on a physical/statistical prediction model and will be used to predict Probability of Detection (POD), Probability of False Alarm (PFA) and Receiver Operating Characteristic (ROC) function curves. These output functions are used to quantify the NDE capability. The physical model will explain and allow predictions for the effects of making changes to the inspection setup (e.g. probe properties and scan increment). The statistical/empirical model will quantify unexplained variability, adjust for model bias, and provide a means for obtaining corresponding uncertainty intervals. Previous work on this project was reported in Meeker et al. (1996). The particular focus of this work is on the use of ultrasonic methods for detecting hard-alpha and other subsurface flaws in titanium using gated peak detection. This is a uniquely challenging problem because the inspection must detect very complex subsurface flaws in the presence of significant “material” noise. The underlying framework of the methodology should, however, be general enough to apply to other NDE methods. This paper describes recent work based on application of the new methodology to the detection of synthetic hard alpha flaws in titanium alloys.