Optimizing input data for training an artificial neural network used for evaluating defect depth in infrared thermographic nondestructive testing

A. O. Chulkov, D. A. Nesteruk, V. P. Vavilov, A. I. Moskovchenko, N. Saeed, M. Omar

Research output: Contribution to journalArticle

Abstract

Ten different sets of input data have been used for training and verification of the neural network intended for determining defect depth in infrared thermographic nondestructive testing. The input data sets included raw temperature data, polynomial fitting, principle component analysis, Fourier transform and others. A minimum error (up 0.02 mm for defects in CFRP at depths from 0.5 to 2.5 mm) has been achieved by using polynomial fitting in logarithmic coordinates with further computation of the first temperature derivatives (the TSR technique), and close results have been obtained by processing raw data with the PCA technique. Both techniques require no use of reference points.

Original languageEnglish
Article number103047
JournalInfrared Physics and Technology
Volume102
DOIs
Publication statusPublished - 1 Nov 2019

Fingerprint

Nondestructive examination
polynomials
education
Polynomials
Infrared radiation
Neural networks
carbon fiber reinforced plastics
Defects
Carbon fiber reinforced plastics
defects
Fourier analysis
Fourier transforms
Derivatives
Temperature
temperature
Processing
carbon fiber reinforced plastic

Keywords

  • Composite material
  • Data processing
  • Defect depth
  • Infrared thermographic testing
  • Neural network

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Condensed Matter Physics

Cite this

@article{63ba1479ab8644bba670e1f77a36c18f,
title = "Optimizing input data for training an artificial neural network used for evaluating defect depth in infrared thermographic nondestructive testing",
abstract = "Ten different sets of input data have been used for training and verification of the neural network intended for determining defect depth in infrared thermographic nondestructive testing. The input data sets included raw temperature data, polynomial fitting, principle component analysis, Fourier transform and others. A minimum error (up 0.02 mm for defects in CFRP at depths from 0.5 to 2.5 mm) has been achieved by using polynomial fitting in logarithmic coordinates with further computation of the first temperature derivatives (the TSR technique), and close results have been obtained by processing raw data with the PCA technique. Both techniques require no use of reference points.",
keywords = "Composite material, Data processing, Defect depth, Infrared thermographic testing, Neural network",
author = "Chulkov, {A. O.} and Nesteruk, {D. A.} and Vavilov, {V. P.} and Moskovchenko, {A. I.} and N. Saeed and M. Omar",
year = "2019",
month = "11",
day = "1",
doi = "10.1016/j.infrared.2019.103047",
language = "English",
volume = "102",
journal = "Infrared Physics and Technology",
issn = "1350-4495",
publisher = "Elsevier",

}

TY - JOUR

T1 - Optimizing input data for training an artificial neural network used for evaluating defect depth in infrared thermographic nondestructive testing

AU - Chulkov, A. O.

AU - Nesteruk, D. A.

AU - Vavilov, V. P.

AU - Moskovchenko, A. I.

AU - Saeed, N.

AU - Omar, M.

PY - 2019/11/1

Y1 - 2019/11/1

N2 - Ten different sets of input data have been used for training and verification of the neural network intended for determining defect depth in infrared thermographic nondestructive testing. The input data sets included raw temperature data, polynomial fitting, principle component analysis, Fourier transform and others. A minimum error (up 0.02 mm for defects in CFRP at depths from 0.5 to 2.5 mm) has been achieved by using polynomial fitting in logarithmic coordinates with further computation of the first temperature derivatives (the TSR technique), and close results have been obtained by processing raw data with the PCA technique. Both techniques require no use of reference points.

AB - Ten different sets of input data have been used for training and verification of the neural network intended for determining defect depth in infrared thermographic nondestructive testing. The input data sets included raw temperature data, polynomial fitting, principle component analysis, Fourier transform and others. A minimum error (up 0.02 mm for defects in CFRP at depths from 0.5 to 2.5 mm) has been achieved by using polynomial fitting in logarithmic coordinates with further computation of the first temperature derivatives (the TSR technique), and close results have been obtained by processing raw data with the PCA technique. Both techniques require no use of reference points.

KW - Composite material

KW - Data processing

KW - Defect depth

KW - Infrared thermographic testing

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=85072609756&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072609756&partnerID=8YFLogxK

U2 - 10.1016/j.infrared.2019.103047

DO - 10.1016/j.infrared.2019.103047

M3 - Article

AN - SCOPUS:85072609756

VL - 102

JO - Infrared Physics and Technology

JF - Infrared Physics and Technology

SN - 1350-4495

M1 - 103047

ER -