Signal Energy based Probabilistic Inspection of Defect in Composites

Ultrasonic guided wave based inspection is a well-known technique for structural health monitoring in composite sheets. Among various guided wave methods, Reconstruction Algorithm for Probabilistic Inspection of Damage (RAPID) is a popular and well accepted technique for defect detection and localization. Very recently a baseline free version of RAPID is developed by utilizing non-linear acoustic where two different sets of measurements are taken on a defected sample, one with very low signal excitation (~5 Volt) and another with high signal excitation (~100 Volt). However, in many situations very low excitation may not yield any signal for very highly attenuative materials in relatively large inspection areas. Here we propose a novel feature termed as scaled energy ratio and used it in the standard RAPID algorithm to successfully localize defects with single set of measurements (with a high voltage excitation only). This technique in one hand offers baseline free detection of defects and in the other hand requires only one set of measurement at a high excitation level. We report very encouraging results while implementing this new feature based RAPID algorithm in defect detection in glass fiber composite sheet for different sensor array locations.