A Multivariate Statistical Approach to Wrinkling Detection in Composites.

Area of Science:

• Materials Science
• Non-Destructive Testing
• Ultrasonic Imaging

Background:

• Nondestructive inspection of carbon-fiber reinforced polymers (CFRPs) using ultrasound is hindered by wave scattering from material plies.
• Existing methods like the total focusing method (TFM) struggle with high backscattering, obscuring flaw signals.
• Detecting internal wrinkling (out-of-plane fiber waviness) is particularly difficult as it doesn't produce additional scattering, only alters signal scattering patterns.

Purpose of the Study:

• To develop an automated method for detecting wrinkling defects in CFRPs using ultrasonic imaging.
• To address the limitations of current methods in identifying subtle structural variations like fiber waviness.
• To improve the reliability and sensitivity of ultrasonic inspection for internal defects in composite materials.

Main Methods:

• Proposed a novel image processing technique analyzing regional variations in ultrasonic image intensity.
• Characterized the 2-D spatial autocorrelation of image areas in pristine components to establish baseline intensity distributions.
• Identified wrinkling by detecting significant deviations from the established spatial autocorrelation distribution.

Main Results:

• The proposed method successfully detects wrinkling by identifying deviations in spatial autocorrelation.
• Demonstrated equivalent performance to existing methods for large wrinkling amplitudes.
• Showed superior performance compared to an alternative approach for detecting low wrinkling amplitudes.

Conclusions:

• The developed method offers a robust approach for automated wrinkling detection in CFRPs via ultrasonic imaging.
• Analyzing spatial autocorrelation of image intensity provides a sensitive indicator of ply layer displacement.
• This technique enhances the capability of ultrasonic non-destructive testing for identifying subtle defects in composite materials.