Characterization of Out-of-Plane wrinkles in woven CFRP Laminate: Development of a novel algorithm utilizing ultrasonic scan data

• Published in: Composites. Part A, Applied science and manufacturing Vol. 190; p. 108644
• Main Authors: Admay Amif, Md, Khan, Irrtisum, Jack, David A.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.03.2025
• Subjects: algorithms; Defect Tracking and Imaging; electric potential difference; Out-of-plane wrinkle; Ultrasonic testing; ultrasonics; Woven CFRP; Woven CFRP; Ultrasonic testing; Defect Tracking and Imaging; Out-of-plane wrinkle
• ISSN: 1359-835X
• DOI: 10.1016/j.compositesa.2024.108644

•Development of a novel algorithm for characterizing out-of-plane wrinkles in woven CFRP laminates.•Precise quantification and spatial analysis through three-dimensional lamina by lamina extraction.•Two-dimensional wrinkle attributes validated against microscopy from sectioned samples.•Improvement in analyzing conventional ultrasound data to enhance non-destructive testing capabilities. Out-of-plane wrinkles in laminated composites reduce the structural integrity. This study presents an algorithm to characterize wrinkle height and intensity of each lamina within a plain weave laminated composite, utilizing full waveform ultrasonic scan data. This algorithm enables the extraction of individual laminae in three dimensions. An immersion tank ultrasound testing system is employed for scanning coupled with a 7.5 MHz spherically focused transducer. Surface construction from the waveform entails a spatial Gaussian averaging followed by a tracking of voltage peaks in time of individual A-scans, from which we extract individual lamina interfaces. Four samples are presented with various wrinkle amplitudes encompassing a wide range of industrially relevant scenarios. The comparison of the ultrasonically characterized wrinkle heights with microscopic images of sectioned samples demonstrates consistency between the two methods, with a maximum deviation of 0.06 mm wrinkle amplitude across all samples considered in this study.