Damage classification based on stiffness reduction in cross-ply laminates with convolution neural networks

Sai Badabagni; R. Talreja

doi:10.1117/12.2612092

18 April 2022 Damage classification based on stiffness reduction in cross-ply laminates with convolution neural networks

Sai Badabagni, R. Talreja

Author Affiliations +

Proceedings Volume 12046, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2022; 120460K (2022) https://doi.org/10.1117/12.2612092
Event: SPIE Smart Structures + Nondestructive Evaluation, 2022, Long Beach, California, United States

Abstract

A classification strategy with a deep learning model to categorize stiffness reduction in cross-ply laminates from transverse cracks is presented. Deep learning models are very successful in image-based classification which can be conveniently used on structural health monitoring data. In the current study, the transverse cracks in 90-degree pliés are modeled with a finite element model in Abaqus; and are solved for fundamental Lamb wave modes (symmetric S0 and antisymmetric A0) independently. The data from the simulation models are grouped into three different classes of stiffness ranges depending on the associated crack density in the 90-degree pliés. The time-domain data for each case is then post-processed with continuous wavelet transforms, and Hilbert transforms to obtain a featured information representation with visual images. The digital image from the database is represented as a matrix encoded in red-green-blue channels (RGB matrix) and is used as an input to train a convolution neural network (CNN) model to perform a categorical classification. The results from the test and validation data obtained for identical configuration and excitation frequency show that the CNN model can classify an image into respective stiffness categories accurately. The proposed model is applied successfully on cross-ply laminates of different configurations delivering a satisfactory performance with excellent validation and test accuracy, demonstrating the potential for developing a deep learning-based tool to classify and quantify damage in composite structures.

Conference Presentation

Citation Download Citation

Sai Badabagni and R. Talreja "Damage classification based on stiffness reduction in cross-ply laminates with convolution neural networks", Proc. SPIE 12046, Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems 2022, 120460K (18 April 2022); https://doi.org/10.1117/12.2612092

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available