Artificial Neural Network Approach to Determine Elastic Modulus of Carbon Fiber-Reinforced Laminates

Xianbo Xu, Nikhil Gupta

Research output: Contribution to journalArticle

Abstract

Recognized as a prominent material for engineering applications, carbon fiber-reinforced laminated composites require significant effort to characterize due to their anisotropic structure and viscoelastic nature. Dynamic mechanical analysis has been used to accelerate the testing process by transforming the measured viscoelastic properties to elastic modulus. To expand the transformation to anisotropic materials, artificial neural network approach is used to build the master relationship of the storage modulus in three in-plane directions. Using rotation transformation, the stiffness tensor can be calculated to extrapolate the frequency domain viscoelastic properties in any orientation with respect to the fiber direction. The viscoelastic properties are transformed to time domain relaxation function using the linear relationship of viscoelasticity. Stress response with a certain strain history is predicted and the elastic modulus is extracted. Compared to the experimental flexural test results, the artificial neural network-based method achieved an error of less than 7.3%. The results show that the transformation can predict the anisotropic material behavior at a wide range of temperatures and strain rates.

Original languageEnglish (US)
JournalJOM
DOIs
StatePublished - Jan 1 2019

Fingerprint

Carbon fibers
Laminates
Elastic moduli
Neural networks
Viscoelasticity
Laminated composites
Dynamic mechanical analysis
Tensors
Strain rate
Stiffness
Fibers
Testing
carbon fiber
Temperature
Direction compound

ASJC Scopus subject areas

  • Materials Science(all)
  • Engineering(all)

Cite this

Artificial Neural Network Approach to Determine Elastic Modulus of Carbon Fiber-Reinforced Laminates. / Xu, Xianbo; Gupta, Nikhil.

In: JOM, 01.01.2019.

Research output: Contribution to journalArticle

@article{3f0979de00c540ea998e4653c5c730af,
title = "Artificial Neural Network Approach to Determine Elastic Modulus of Carbon Fiber-Reinforced Laminates",
abstract = "Recognized as a prominent material for engineering applications, carbon fiber-reinforced laminated composites require significant effort to characterize due to their anisotropic structure and viscoelastic nature. Dynamic mechanical analysis has been used to accelerate the testing process by transforming the measured viscoelastic properties to elastic modulus. To expand the transformation to anisotropic materials, artificial neural network approach is used to build the master relationship of the storage modulus in three in-plane directions. Using rotation transformation, the stiffness tensor can be calculated to extrapolate the frequency domain viscoelastic properties in any orientation with respect to the fiber direction. The viscoelastic properties are transformed to time domain relaxation function using the linear relationship of viscoelasticity. Stress response with a certain strain history is predicted and the elastic modulus is extracted. Compared to the experimental flexural test results, the artificial neural network-based method achieved an error of less than 7.3{\%}. The results show that the transformation can predict the anisotropic material behavior at a wide range of temperatures and strain rates.",
author = "Xianbo Xu and Nikhil Gupta",
year = "2019",
month = "1",
day = "1",
doi = "10.1007/s11837-019-03666-7",
language = "English (US)",
journal = "JOM",
issn = "1047-4838",
publisher = "Minerals, Metals and Materials Society",

}

TY - JOUR

T1 - Artificial Neural Network Approach to Determine Elastic Modulus of Carbon Fiber-Reinforced Laminates

AU - Xu, Xianbo

AU - Gupta, Nikhil

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Recognized as a prominent material for engineering applications, carbon fiber-reinforced laminated composites require significant effort to characterize due to their anisotropic structure and viscoelastic nature. Dynamic mechanical analysis has been used to accelerate the testing process by transforming the measured viscoelastic properties to elastic modulus. To expand the transformation to anisotropic materials, artificial neural network approach is used to build the master relationship of the storage modulus in three in-plane directions. Using rotation transformation, the stiffness tensor can be calculated to extrapolate the frequency domain viscoelastic properties in any orientation with respect to the fiber direction. The viscoelastic properties are transformed to time domain relaxation function using the linear relationship of viscoelasticity. Stress response with a certain strain history is predicted and the elastic modulus is extracted. Compared to the experimental flexural test results, the artificial neural network-based method achieved an error of less than 7.3%. The results show that the transformation can predict the anisotropic material behavior at a wide range of temperatures and strain rates.

AB - Recognized as a prominent material for engineering applications, carbon fiber-reinforced laminated composites require significant effort to characterize due to their anisotropic structure and viscoelastic nature. Dynamic mechanical analysis has been used to accelerate the testing process by transforming the measured viscoelastic properties to elastic modulus. To expand the transformation to anisotropic materials, artificial neural network approach is used to build the master relationship of the storage modulus in three in-plane directions. Using rotation transformation, the stiffness tensor can be calculated to extrapolate the frequency domain viscoelastic properties in any orientation with respect to the fiber direction. The viscoelastic properties are transformed to time domain relaxation function using the linear relationship of viscoelasticity. Stress response with a certain strain history is predicted and the elastic modulus is extracted. Compared to the experimental flexural test results, the artificial neural network-based method achieved an error of less than 7.3%. The results show that the transformation can predict the anisotropic material behavior at a wide range of temperatures and strain rates.

UR - http://www.scopus.com/inward/record.url?scp=85069681987&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069681987&partnerID=8YFLogxK

U2 - 10.1007/s11837-019-03666-7

DO - 10.1007/s11837-019-03666-7

M3 - Article

JO - JOM

JF - JOM

SN - 1047-4838

ER -