Utility of feature mapping in ultrasonic non-destructive evaluation

Abstract

Feature mapping presents us with a novel philosophy in ultrasonic non-destructive evaluation that utilizes maximum information from different data acquisition and signal feature domains for possible complete material characterization and defect analysis in a variety of different materials and structures. Feature mapping is a powerful technique that covers a wide area of data acquisition and analysis techniques applying both physically and statistically based principles. This procedure, beyond standard normal beam applications, employs many seldom used but interesting data collection procedures such as critical angles, surface waves, plate waves and backscattering techniques. It forms these into an extremely versatile data acquisition protocol, followed by detailed analysis through a state of the technology signal processing, pattern recognition and expert system and/or artificial intelligence implementation practice. This Paper explains the various existing possibilities for clever physically based data collection, and the types of feature domains and features available for anomaly representation in materials. Typical results are discussed to illustrate the capabilities and utilities of the feature mapping techniques, including a normal beam longitudinal wave high frequency area ratio feature mapping (F-map) system, oblique incidence shear wave amplitude F-maps, and plate leaky wave amplitude F-map results for interfacial weakness in an adhesive bond structure. Also discussed are the results from a normal beam longitudinal wave amplitude and frequency bandwidth F-map for crack detection in a composite material, as well as a backscattered energy F-map for defect edge detection in composite materials.