Using a Neural Network Control Policy for Rapid Switching Between Beam Parameters in an FEL

A.L. Edelen; S.G. Biedron; J.P. Edelen; S.V. Milton; P.J.M. van der Slot

doi:10.18429/JACoW-FEL2017-WEP031

Using a Neural Network Control Policy for Rapid Switching Between Beam Parameters in an FEL

A.L. Edelen
, S.G. Biedron
, J.P. Edelen
, S.V. Milton
, P.J.M. van der Slot

Laser Physics & Nonlinear Optics

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

13 Citations (Scopus)

113 Downloads (Pure)

Abstract

FEL user facilities often must accommodate requests for a variety of beam parameters. This usually requires skilled operators to tune the machine, reducing the amount of available time for users. In principle, a neural network control policy that is trained on a broad range of operating states could be used to quickly switch between these requests without substantial need for human intervention.
We present preliminary results from an ongoing study in which a neural network control policy is investigated for rapid switching between beam parameters in a
compact THz FEL.

Original language	English
Title of host publication	Proceedings of the 38th International Free Electron Laser Conference (FEL 2017)
Place of Publication	Santa Fe, New Mexico
Pages	406-409
DOIs	https://doi.org/10.18429/JACoW-FEL2017-WEP031
Publication status	Published - 20 Aug 2017
Event	38th International Free Electron Laser Conference, FEL 2017 - Santa Fe Community Convention Center (SFCCC), Santa Fe, United States Duration: 20 Aug 2017 → 25 Aug 2017 Conference number: 38 http://www.lanl.gov/conferences/free-electron-laser/

Conference

Conference	38th International Free Electron Laser Conference, FEL 2017
Abbreviated title	FEL
Country/Territory	United States
City	Santa Fe
Period	20/08/17 → 25/08/17
Internet address	http://www.lanl.gov/conferences/free-electron-laser/

Access to Document

10.18429/JACoW-FEL2017-WEP031

Edelen2017usingFinal published version, 450 KBLicence: CC BY

Cite this

@inproceedings{9abe378fdf2b4f248444ac8154707470,

title = "Using a Neural Network Control Policy for Rapid Switching Between Beam Parameters in an FEL",

abstract = "FEL user facilities often must accommodate requests for a variety of beam parameters. This usually requires skilled operators to tune the machine, reducing the amount of available time for users. In principle, a neural network control policy that is trained on a broad range of operating states could be used to quickly switch between these requests without substantial need for human intervention. We present preliminary results from an ongoing study in which a neural network control policy is investigated for rapid switching between beam parameters in a compact THz FEL.",

author = "A.L. Edelen and S.G. Biedron and J.P. Edelen and S.V. Milton and \{van der Slot\}, P.J.M.",

year = "2017",

month = aug,

day = "20",

doi = "10.18429/JACoW-FEL2017-WEP031",

language = "English",

isbn = "978-3-95450-179-3",

pages = "406--409",

booktitle = "Proceedings of the 38th International Free Electron Laser Conference (FEL 2017)",

note = "38th International Free Electron Laser Conference, FEL 2017, FEL ; Conference date: 20-08-2017 Through 25-08-2017",

url = "http://www.lanl.gov/conferences/free-electron-laser/",

}

Edelen, AL, Biedron, SG, Edelen, JP, Milton, SV & van der Slot, PJM 2017, Using a Neural Network Control Policy for Rapid Switching Between Beam Parameters in an FEL. in Proceedings of the 38th International Free Electron Laser Conference (FEL 2017)., WEP031, Santa Fe, New Mexico, pp. 406-409, 38th International Free Electron Laser Conference, FEL 2017, Santa Fe, New Mexico, United States, 20/08/17. https://doi.org/10.18429/JACoW-FEL2017-WEP031

Using a Neural Network Control Policy for Rapid Switching Between Beam Parameters in an FEL. / Edelen, A.L.; Biedron, S.G.; Edelen, J.P. et al.
Proceedings of the 38th International Free Electron Laser Conference (FEL 2017). Santa Fe, New Mexico, 2017. p. 406-409 WEP031.

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

TY - GEN

T1 - Using a Neural Network Control Policy for Rapid Switching Between Beam Parameters in an FEL

AU - Edelen, A.L.

AU - Biedron, S.G.

AU - Edelen, J.P.

AU - Milton, S.V.

AU - van der Slot, P.J.M.

N1 - Conference code: 38

PY - 2017/8/20

Y1 - 2017/8/20

N2 - FEL user facilities often must accommodate requests for a variety of beam parameters. This usually requires skilled operators to tune the machine, reducing the amount of available time for users. In principle, a neural network control policy that is trained on a broad range of operating states could be used to quickly switch between these requests without substantial need for human intervention. We present preliminary results from an ongoing study in which a neural network control policy is investigated for rapid switching between beam parameters in a compact THz FEL.

AB - FEL user facilities often must accommodate requests for a variety of beam parameters. This usually requires skilled operators to tune the machine, reducing the amount of available time for users. In principle, a neural network control policy that is trained on a broad range of operating states could be used to quickly switch between these requests without substantial need for human intervention. We present preliminary results from an ongoing study in which a neural network control policy is investigated for rapid switching between beam parameters in a compact THz FEL.

U2 - 10.18429/JACoW-FEL2017-WEP031

DO - 10.18429/JACoW-FEL2017-WEP031

M3 - Conference contribution

SN - 978-3-95450-179-3

SP - 406

EP - 409

BT - Proceedings of the 38th International Free Electron Laser Conference (FEL 2017)

CY - Santa Fe, New Mexico

T2 - 38th International Free Electron Laser Conference, FEL 2017

Y2 - 20 August 2017 through 25 August 2017

ER -

Using a Neural Network Control Policy for Rapid Switching Between Beam Parameters in an FEL

Abstract

Conference

Access to Document

Fingerprint

Cite this