Title: Machine Learning-Based Models for Delamination Detection in a Composite Laminate

Authors: Junyan He, Linqi Zhuang, Adarsh Chaurasia, Ali Najafi

DOI: 10.33599/nasampe/s.24.0086

Abstract: In this study, a random-forest-based machine learning model that uses a set of five stress wave factors as input features was evaluated for delamination location prediction in composite plates with different layup combinations. Meanwhile, a comprehensive study was also conducted using the same training data to compare the performance of the random-forest-based machine learning model and a deep learning model. The deep learning model converts 1D time-domain data into a 2D time-frequency domain image through continuous wavelet transformation, employing it as input features for delamination location prediction. Our findings reveal the efficacy of the proposed stress wave factors as input features. The comparative study between the two machine learning models demonstrates their capability to accurately predict delamination locations for specific layups or composite shapes. However, the prediction accuracy diminishes for unseen cases, such as different layups and delamination shapes. To enhance prediction accuracy for these unseen scenarios, additional training data is required.

References: [1] Su Z, Ye L, Lu Y. Guided Lamb waves for identification of damage in composite structures: A review. Journal of sound and vibration. Elsevier, 2006. [2]Mitra M, Gopalakrishnan S. Guided wave based structural health monitoring: A review. Smart Materials and Structures. IOP, 2016. [3]He, J., Zhuang, L., Chaurasia, A., and Najafi, A., “Prediction of delamination location in composite structures with different ply orientations: a framework integrating finite element simulation and deep learning,” ASME International Mechanical Engineering Congress and Exposition, 2023. [4]He, J., Zhuang, L., Chaurasia, A., and Najafi, A., “Investigating the Application of Stress Wave Factors in Machine Learning for Delamination Location Prediction in a Composite Laminate,” AIAA SCITECH 2024 Forum, 2024. [5]Talreja, R., “Application of acousto-ultrasonics to quality control and damage assessment of composites,” Acousto-ultrasonics: theory and application, Springer, 1988. [6]Pillarisetti, L. S. S., and Talreja, R., “On quantifying damage severity in composite materials by an ultrasonic method,” Composite Structures, Elsevier, 2019. [7]Zhuang, L., Chaurasia, A., and Najafi, A., “Impact Damage Evaluations in a Composite Laminate Using Guided Wave-Based Simulation,” ASME International Mechanical Engineering Congress and Exposition, 2022. [8]“pyDPF-Core,” https://dpf.docs.pyansys.com/version/stable/, 2023. [9]Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., and Duchesnay, E., “Scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, 2011. [10]Zhuang, L., He, J., Chaurasia, A., and Najafi, A., “Integrating Deep Learning with Guided Wave-based Simulation to Predict Delamination Location in Composite Structures,” Proceedings of The American Society for Composites-Thirty-Eighth Technical Conference, 2023.

Conference: SAMPE 2024

Publication Date: 2024/05/20

SKU: TP24-0000000086

Pages: 10

Price: $20.00