DOC-DICAM: Domain Aware One Class Defect Identification in Composite Aerostructure Material

Yunker, Austin; Kettimuthu, Rajkumar; Kral, Zachary

doi:10.1109/icmla61862.2024.00157

DOC-DICAM: Domain Aware One Class Defect Identification in Composite Aerostructure Material

Conference · Wed Dec 18 04:00:00 EST 2024 · 2024 International Conference on Machine Learning and Applications (ICMLA)

DOI:https://doi.org/10.1109/icmla61862.2024.00157· OSTI ID:2562930

Yunker, Austin ^[1]; Kettimuthu, Rajkumar ^[2]; Kral, Zachary ^[3]

Argonne National Laboratory (ANL), Argonne, IL (United States); Spirit AeroSystems Inc
Argonne National Laboratory (ANL), Argonne, IL (United States)
Spirit AeroSystems, Inc., Wichita, KS (United States)

Fiber-reinforced composites are a common material used in the design of aircraft structures due to their good tensile strength and resistance to compression. During the manufacturing process, these structures are thoroughly inspected for flaws and defects to ensure structural integrity during commercial use. Non-destructive testing (NDT) is a collection of inspection methods that allow inspectors to evaluate material without altering it. Due to the high safety standards in aerospace manufacturing, the NDT process is done manually and can be a significant bottleneck in the development workflow. In this paper, we develop an AI-based assistance tool to drastically reduce inspection time. Typical AI workflows require large amounts of annotated data, but defects rarely occur resulting in strong class imbalance. To overcome this, we formulate the problem of defect identification as an anomaly detection task in which our primary focus is learning non-defect characteristics. To do this, we develop a multi-task self-supervised learning framework that embeds problem specific domain knowledge into the deep learning model. We verify our method using fuselage data generated in a production environment. As a result, we show that our method can effectively identify defects and requires minimal training and inference time.

Research Organization:: Spirit AeroSystems, Inc., Wichita, KS (United States)

Sponsoring Organization:: USDOE Office of Energy Efficiency and Renewable Energy (EERE)

DOE Contract Number:: EE0009397; AC02-06CH11357

OSTI ID:: 2562930

Conference Information:: Journal Name: 2024 International Conference on Machine Learning and Applications (ICMLA)

Country of Publication:: United States

Language:: English

References (10)

Nondestructive evaluation of thick-section composites and sandwich structures: A review Ibrahim, M. E. Composites Part A: Applied Science and Manufacturing, Vol. 64 https://doi.org/10.1016/j.compositesa.2014.04.010	journal	September 2014
Automatic defects detection in CFRP thermograms, using convolutional neural networks and transfer learning Saeed, Numan; King, Nelson; Said, Zafar Infrared Physics & Technology, Vol. 102 https://doi.org/10.1016/j.infrared.2019.103048	journal	November 2019
Sizing of flaws using ultrasonic bulk wave testing: A review Felice, Maria V.; Fan, Zheng Ultrasonics, Vol. 88 https://doi.org/10.1016/j.ultras.2018.03.003	journal	August 2018
Deep learning-based anomaly detection from ultrasonic images Posilović, Luka; Medak, Duje; Milković, Fran Ultrasonics, Vol. 124 https://doi.org/10.1016/j.ultras.2022.106737	journal	August 2022
Automatic anomaly detection from X-ray images based on autoencoders Presenti, Alice; Liang, Zhihua; Alves Pereira, Luis F. Nondestructive Testing and Evaluation, Vol. 37, Issue 5 https://doi.org/10.1080/10589759.2022.2074415	journal	June 2022
CutPaste: Self-Supervised Learning for Anomaly Detection and Localization Li, Chun-Liang; Sohn, Kihyuk; Yoon, Jinsung 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) https://doi.org/10.1109/CVPR46437.2021.00954	conference	June 2021
EISATC-Fusion: Inception Self-Attention Temporal Convolutional Network Fusion for Motor Imagery EEG Decoding Liang, Guangjin; Cao, Dianguo; Wang, Jinqiang IEEE Transactions on Neural Systems and Rehabilitation Engineering, Vol. 32 https://doi.org/10.1109/tnsre.2024.3382226	journal	January 2024
Evaluating Model Robustness for Defect Identification and Classification in a Composite Aerostructure Material Yunker, Austin; Lake, Rami; Kettimuthu, Rajkumar Journal of Nondestructive Evaluation, Diagnostics and Prognostics of Engineering Systems, Vol. 8, Issue 1 https://doi.org/10.1115/1.4065474	journal	July 2024
OC-DICAM: One Class Defect Identification in Composite Aerostructure Material Yunker, Austin; Kettimuthu, Rajkumar; Kral, Zachary ASME 2024 Aerospace Structures, Structural Dynamics, and Materials Conference https://doi.org/10.1115/ssdm2024-121466	conference	April 2024
Artificial neural networks for structural damage detection and classification Ferregut, Carlos M.; Osegueda, Roberto A.; Ortiz, Jaime SPIE Proceedings https://doi.org/10.1117/12.207718	conference	April 1995

Similar Records

Comparative Study on Deep Learning Methods for Defect Identification and Classification in Composite Aerostructure Material

Conference · Thu Jul 27 00:00:00 EDT 2023 · 2023 50th Annual Review of Progress in Quantitative Nondestructive Evaluation · OSTI ID:2562929

Effective Defect Detection Using Instance Segmentation for NDI

Conference · Thu Jan 23 23:00:00 EST 2025 · OSTI ID:2563178

Evaluating Model Robustness for Defect Identification and Classification in a Composite Aerostructure Material

Journal Article · Thu Jul 25 20:00:00 EDT 2024 · Journal of Nondestructive Evaluation, Diagnostics and Prognostics of Engineering Systems · OSTI ID:2563184

Related Subjects

Non-destructive testing (NDT)
anomaly detection
composites
deep learning
defects
ultrasonic testing (UT)

DOC-DICAM: Domain Aware One Class Defect Identification in Composite Aerostructure Material

Citation Formats

References (10)

Similar Records

Related Subjects