Machine learning assisted Fiber Bragg Grating based temperature sensing

Martin Søren Engmann Djurhuus*, Stefan Werzinger, Anders Clausen, Darko Zibar

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

179 Downloads (Pure)

Abstract

This paper proposes an alternative approach to the signal processing of temperature measurements based on Fiber Bragg Gratings (FBG) using the machine learning tool Gaussian Process Regression (GPR). Experimental results show that for a majority of the cases under consideration the reported technique provides a more accurate calculation of the temperature than the conventional methods. Furthermore, the GPR can give the uncertainty of an estimate together with the estimate itself, which for example is useful when it is important to know the worst case scenario of a measurand. GPR also has the potential to improve the measurement speed of FBG based temperature sensing compared to current standards.
Original languageEnglish
JournalIEEE Photonics Technology Letters
Volume31
Issue number12
Pages (from-to)939-942
Number of pages4
ISSN1041-1135
DOIs
Publication statusPublished - 2019

Keywords

  • Machine learning
  • Gaussian processes
  • Optical fiber sensors
  • Bragg gratings

Cite this

@article{131e529d40d4411893a1487d9bf48418,
title = "Machine learning assisted Fiber Bragg Grating based temperature sensing",
abstract = "This paper proposes an alternative approach to the signal processing of temperature measurements based on Fiber Bragg Gratings (FBG) using the machine learning tool Gaussian Process Regression (GPR). Experimental results show that for a majority of the cases under consideration the reported technique provides a more accurate calculation of the temperature than the conventional methods. Furthermore, the GPR can give the uncertainty of an estimate together with the estimate itself, which for example is useful when it is important to know the worst case scenario of a measurand. GPR also has the potential to improve the measurement speed of FBG based temperature sensing compared to current standards.",
keywords = "Machine learning, Gaussian processes, Optical fiber sensors, Bragg gratings",
author = "Djurhuus, {Martin S{\o}ren Engmann} and Stefan Werzinger and Anders Clausen and Darko Zibar",
year = "2019",
doi = "10.1109/LPT.2019.2913992",
language = "English",
volume = "31",
pages = "939--942",
journal = "I E E E Photonics Technology Letters",
issn = "1041-1135",
publisher = "Institute of Electrical and Electronics Engineers",
number = "12",

}

Machine learning assisted Fiber Bragg Grating based temperature sensing. / Djurhuus, Martin Søren Engmann; Werzinger, Stefan; Clausen, Anders; Zibar, Darko.

In: IEEE Photonics Technology Letters, Vol. 31, No. 12, 2019, p. 939-942.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Machine learning assisted Fiber Bragg Grating based temperature sensing

AU - Djurhuus, Martin Søren Engmann

AU - Werzinger, Stefan

AU - Clausen, Anders

AU - Zibar, Darko

PY - 2019

Y1 - 2019

N2 - This paper proposes an alternative approach to the signal processing of temperature measurements based on Fiber Bragg Gratings (FBG) using the machine learning tool Gaussian Process Regression (GPR). Experimental results show that for a majority of the cases under consideration the reported technique provides a more accurate calculation of the temperature than the conventional methods. Furthermore, the GPR can give the uncertainty of an estimate together with the estimate itself, which for example is useful when it is important to know the worst case scenario of a measurand. GPR also has the potential to improve the measurement speed of FBG based temperature sensing compared to current standards.

AB - This paper proposes an alternative approach to the signal processing of temperature measurements based on Fiber Bragg Gratings (FBG) using the machine learning tool Gaussian Process Regression (GPR). Experimental results show that for a majority of the cases under consideration the reported technique provides a more accurate calculation of the temperature than the conventional methods. Furthermore, the GPR can give the uncertainty of an estimate together with the estimate itself, which for example is useful when it is important to know the worst case scenario of a measurand. GPR also has the potential to improve the measurement speed of FBG based temperature sensing compared to current standards.

KW - Machine learning

KW - Gaussian processes

KW - Optical fiber sensors

KW - Bragg gratings

U2 - 10.1109/LPT.2019.2913992

DO - 10.1109/LPT.2019.2913992

M3 - Journal article

VL - 31

SP - 939

EP - 942

JO - I E E E Photonics Technology Letters

JF - I E E E Photonics Technology Letters

SN - 1041-1135

IS - 12

ER -