Machine learning enabled damage classification in composite laminated beams using mode conversion quantification

Vivek T. Rathod; Subrata Mukherjee; Yiming Deng

doi:10.1117/12.2559677

Machine learning enabled damage classification in composite laminated beams using mode conversion quantification

Vivek T. Rathod
, Subrata Mukherjee
, Yiming Deng

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

We propose a model assisted method to identify damage types and severity based on mode converted wave strength. Machine learning techniques are employed to develop classification models complemented by the finite element simulation models. Finite element simulation models provide the training data for various cases of damage and severity involving common types of damages in composites. Damage classification models are based on mode conversion strength versus frequency curves of participating four wave modes. For damage recognition and classification, a multi-layer Convoluted Neural Network (CNN) has been trained using the back-propagation paradigm on the generated dataset.

Original language	English
Title of host publication	Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV
Editors	Tzu-Yang Yu, H. Felix Wu, Peter J. Shull, Andrew L. Gyekenyesi
Publisher	SPIE
ISBN (Electronic)	9781510635371
DOIs	https://doi.org/10.1117/12.2559677
State	Published - 2020
Externally published	Yes
Event	Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV 2020 - None, United States Duration: Apr 27 2020 → May 8 2020

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	11380
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Conference

Conference	Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV 2020
Country/Territory	United States
City	None
Period	04/27/20 → 05/8/20

Keywords

Composite laminated beam
Damage classification
Embedded sensors
Guided wave mode conversion
K nearest neighbor
Machine learning
Naive Bayes
Random Forest
Support Vector Machine

Access to Document

10.1117/12.2559677

Cite this

Rathod, V. T., Mukherjee, S., & Deng, Y. (2020). Machine learning enabled damage classification in composite laminated beams using mode conversion quantification. In T.-Y. Yu, H. F. Wu, P. J. Shull, & A. L. Gyekenyesi (Eds.), Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV Article 113800B (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11380). SPIE. https://doi.org/10.1117/12.2559677

Rathod, Vivek T. ; Mukherjee, Subrata ; Deng, Yiming. / Machine learning enabled damage classification in composite laminated beams using mode conversion quantification. Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV. editor / Tzu-Yang Yu ; H. Felix Wu ; Peter J. Shull ; Andrew L. Gyekenyesi. SPIE, 2020. (Proceedings of SPIE - The International Society for Optical Engineering).

@inproceedings{63eb8018e4b240de9399b08404dddebf,

title = "Machine learning enabled damage classification in composite laminated beams using mode conversion quantification",

abstract = "We propose a model assisted method to identify damage types and severity based on mode converted wave strength. Machine learning techniques are employed to develop classification models complemented by the finite element simulation models. Finite element simulation models provide the training data for various cases of damage and severity involving common types of damages in composites. Damage classification models are based on mode conversion strength versus frequency curves of participating four wave modes. For damage recognition and classification, a multi-layer Convoluted Neural Network (CNN) has been trained using the back-propagation paradigm on the generated dataset.",

keywords = "Composite laminated beam, Damage classification, Embedded sensors, Guided wave mode conversion, K nearest neighbor, Machine learning, Naive Bayes, Random Forest, Support Vector Machine",

author = "Rathod, \{Vivek T.\} and Subrata Mukherjee and Yiming Deng",

note = "Publisher Copyright: {\textcopyright} COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.; Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV 2020 ; Conference date: 27-04-2020 Through 08-05-2020",

year = "2020",

doi = "10.1117/12.2559677",

language = "English",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "SPIE",

editor = "Tzu-Yang Yu and Wu, \{H. Felix\} and Shull, \{Peter J.\} and Gyekenyesi, \{Andrew L.\}",

booktitle = "Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV",

}

Rathod, VT, Mukherjee, S & Deng, Y 2020, Machine learning enabled damage classification in composite laminated beams using mode conversion quantification. in T-Y Yu, HF Wu, PJ Shull & AL Gyekenyesi (eds), Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV., 113800B, Proceedings of SPIE - The International Society for Optical Engineering, vol. 11380, SPIE, Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV 2020, None, United States, 04/27/20. https://doi.org/10.1117/12.2559677

Machine learning enabled damage classification in composite laminated beams using mode conversion quantification. / Rathod, Vivek T.; Mukherjee, Subrata; Deng, Yiming.
Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV. ed. / Tzu-Yang Yu; H. Felix Wu; Peter J. Shull; Andrew L. Gyekenyesi. SPIE, 2020. 113800B (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11380).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Machine learning enabled damage classification in composite laminated beams using mode conversion quantification

AU - Rathod, Vivek T.

AU - Mukherjee, Subrata

AU - Deng, Yiming

N1 - Publisher Copyright: © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

PY - 2020

Y1 - 2020

N2 - We propose a model assisted method to identify damage types and severity based on mode converted wave strength. Machine learning techniques are employed to develop classification models complemented by the finite element simulation models. Finite element simulation models provide the training data for various cases of damage and severity involving common types of damages in composites. Damage classification models are based on mode conversion strength versus frequency curves of participating four wave modes. For damage recognition and classification, a multi-layer Convoluted Neural Network (CNN) has been trained using the back-propagation paradigm on the generated dataset.

AB - We propose a model assisted method to identify damage types and severity based on mode converted wave strength. Machine learning techniques are employed to develop classification models complemented by the finite element simulation models. Finite element simulation models provide the training data for various cases of damage and severity involving common types of damages in composites. Damage classification models are based on mode conversion strength versus frequency curves of participating four wave modes. For damage recognition and classification, a multi-layer Convoluted Neural Network (CNN) has been trained using the back-propagation paradigm on the generated dataset.

KW - Composite laminated beam

KW - Damage classification

KW - Embedded sensors

KW - Guided wave mode conversion

KW - K nearest neighbor

KW - Machine learning

KW - Naive Bayes

KW - Random Forest

KW - Support Vector Machine

UR - https://www.scopus.com/pages/publications/85085739346

U2 - 10.1117/12.2559677

DO - 10.1117/12.2559677

M3 - Conference contribution

AN - SCOPUS:85085739346

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV

A2 - Yu, Tzu-Yang

A2 - Wu, H. Felix

A2 - Shull, Peter J.

A2 - Gyekenyesi, Andrew L.

PB - SPIE

T2 - Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV 2020

Y2 - 27 April 2020 through 8 May 2020

ER -

Rathod VT, Mukherjee S, Deng Y. Machine learning enabled damage classification in composite laminated beams using mode conversion quantification. In Yu TY, Wu HF, Shull PJ, Gyekenyesi AL, editors, Nondestructive Characterization and Monitoring of Advanced Materials, Aerospace, Civil Infrastructure, and Transportation XIV. SPIE. 2020. 113800B. (Proceedings of SPIE - The International Society for Optical Engineering). doi: 10.1117/12.2559677

Machine learning enabled damage classification in composite laminated beams using mode conversion quantification

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this