Diagnosing electron temperature using machine learning and neutral tungsten spectral emission

C. A. Johnson; E. A. Unterberg; D. A. Ennis; G. J. Hartwell; D. A. Maurer

doi:10.1016/j.nme.2022.101304

Diagnosing electron temperature using machine learning and neutral tungsten spectral emission

C. A. Johnson, E. A. Unterberg, D. A. Ennis, G. J. Hartwell, D. A. Maurer

Power Exhaust and Particle Control

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

Current spectroscopic based erosion diagnostics require both T_e and n_e measurements in addition to detailed atomic physics and collisional radiative (CR) modeling. Machine Learning (ML) techniques are used to address the temperature measurement requirement for erosion diagnosis. ML techniques are combined with tungsten spectroscopic diagnosis trained with co-located Langmuir probe measurements in the Compact Toroidal Hybrid (CTH) to obtain a spectroscopic based local electron temperature diagnostic. Initial analysis using synthetic data and a Neutral Network (NN) suggests a temperature diagnostic obtained with experimental data is feasible. ML methods have the potential to bypass sources of error in traditional tungsten erosion diagnosis by taking the place of required atomic and CR modeling which introduce inherent uncertainties. Temperature diagnosed could be used as input to current erosion diagnosis techniques (the S/XB method).

Original language	English
Article number	101304
Journal	Nuclear Materials and Energy
Volume	34
DOIs	https://doi.org/10.1016/j.nme.2022.101304
State	Published - Mar 2023

Funding

This research was supported by the U.S. Department of Energy Fusion Energy Sciences Postdoctoral Research Program administered by the Oak Ridge Institute for Science and Education (ORISE) for the DOE . ORISE is managed by Oak Ridge Associated Universities (ORAU), USA under DOE contract number DE-SC0014664 . Work also supported under DOE, USA grants DE-AC05-00OR22725 , DE-FG02-00ER54610 and DE-SC0015877 . All opinions expressed in this paper are the author’s and do not necessarily reflect the policies and views of DOE, ORAU, or ORISE. Notice: This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan ( http://energy.gov/downloads/doe-public-access-plan ).

Funders	Funder number
U.S. Department of Energy Fusion Energy Sciences
U.S. Department of Energy
Oak Ridge Associated Universities	DE-AC05-00OR22725, DE-SC0015877, DE-FG02-00ER54610, DE-SC0014664
Oak Ridge Institute for Science and Education

Keywords

Collisional radiative
Line ratios
Machine learning
Spectroscopy
Tungsten

Access to Document

10.1016/j.nme.2022.101304

Cite this

@article{52a11976632f4c548f9e7c08e0559cd4,

title = "Diagnosing electron temperature using machine learning and neutral tungsten spectral emission",

abstract = "Current spectroscopic based erosion diagnostics require both Te and ne measurements in addition to detailed atomic physics and collisional radiative (CR) modeling. Machine Learning (ML) techniques are used to address the temperature measurement requirement for erosion diagnosis. ML techniques are combined with tungsten spectroscopic diagnosis trained with co-located Langmuir probe measurements in the Compact Toroidal Hybrid (CTH) to obtain a spectroscopic based local electron temperature diagnostic. Initial analysis using synthetic data and a Neutral Network (NN) suggests a temperature diagnostic obtained with experimental data is feasible. ML methods have the potential to bypass sources of error in traditional tungsten erosion diagnosis by taking the place of required atomic and CR modeling which introduce inherent uncertainties. Temperature diagnosed could be used as input to current erosion diagnosis techniques (the S/XB method).",

keywords = "Collisional radiative, Line ratios, Machine learning, Spectroscopy, Tungsten",

author = "Johnson, {C. A.} and Unterberg, {E. A.} and Ennis, {D. A.} and Hartwell, {G. J.} and Maurer, {D. A.}",

note = "Publisher Copyright: {\textcopyright} 2022 The Author(s)",

year = "2023",

month = mar,

doi = "10.1016/j.nme.2022.101304",

language = "English",

volume = "34",

journal = "Nuclear Materials and Energy",

issn = "2352-1791",

publisher = "Elsevier",

}

TY - JOUR

T1 - Diagnosing electron temperature using machine learning and neutral tungsten spectral emission

AU - Johnson, C. A.

AU - Unterberg, E. A.

AU - Ennis, D. A.

AU - Hartwell, G. J.

AU - Maurer, D. A.

PY - 2023/3

Y1 - 2023/3

N2 - Current spectroscopic based erosion diagnostics require both Te and ne measurements in addition to detailed atomic physics and collisional radiative (CR) modeling. Machine Learning (ML) techniques are used to address the temperature measurement requirement for erosion diagnosis. ML techniques are combined with tungsten spectroscopic diagnosis trained with co-located Langmuir probe measurements in the Compact Toroidal Hybrid (CTH) to obtain a spectroscopic based local electron temperature diagnostic. Initial analysis using synthetic data and a Neutral Network (NN) suggests a temperature diagnostic obtained with experimental data is feasible. ML methods have the potential to bypass sources of error in traditional tungsten erosion diagnosis by taking the place of required atomic and CR modeling which introduce inherent uncertainties. Temperature diagnosed could be used as input to current erosion diagnosis techniques (the S/XB method).

AB - Current spectroscopic based erosion diagnostics require both Te and ne measurements in addition to detailed atomic physics and collisional radiative (CR) modeling. Machine Learning (ML) techniques are used to address the temperature measurement requirement for erosion diagnosis. ML techniques are combined with tungsten spectroscopic diagnosis trained with co-located Langmuir probe measurements in the Compact Toroidal Hybrid (CTH) to obtain a spectroscopic based local electron temperature diagnostic. Initial analysis using synthetic data and a Neutral Network (NN) suggests a temperature diagnostic obtained with experimental data is feasible. ML methods have the potential to bypass sources of error in traditional tungsten erosion diagnosis by taking the place of required atomic and CR modeling which introduce inherent uncertainties. Temperature diagnosed could be used as input to current erosion diagnosis techniques (the S/XB method).

KW - Collisional radiative

KW - Line ratios

KW - Machine learning

KW - Spectroscopy

KW - Tungsten

UR - http://www.scopus.com/inward/record.url?scp=85143688651&partnerID=8YFLogxK

U2 - 10.1016/j.nme.2022.101304

DO - 10.1016/j.nme.2022.101304

M3 - Article

AN - SCOPUS:85143688651

SN - 2352-1791

VL - 34

JO - Nuclear Materials and Energy

JF - Nuclear Materials and Energy

M1 - 101304

ER -

Diagnosing electron temperature using machine learning and neutral tungsten spectral emission

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this