Automation and control of laser wakefield accelerators using Bayesian optimization

R. J. Shalloo; S. J.D. Dann; J. N. Gruse; C. I.D. Underwood; A. F. Antoine; C. Arran; M. Backhouse; C. D. Baird; M. D. Balcazar; N. Bourgeois; J. A. Cardarelli; P. Hatfield; J. Kang; K. Krushelnick; S. P.D. Mangles; C. D. Murphy; N. Lu; J. Osterhoff; K. Põder; P. P. Rajeev; C. P. Ridgers; S. Rozario; M. P. Selwood; A. J. Shahani; D. R. Symes; A. G.R. Thomas; C. Thornton; Z. Najmudin; M. J.V. Streeter

doi:10.1038/s41467-020-20245-6

Automation and control of laser wakefield accelerators using Bayesian optimization

R. J. Shalloo^*, S. J.D. Dann, J. N. Gruse, C. I.D. Underwood, A. F. Antoine, C. Arran, M. Backhouse, C. D. Baird, M. D. Balcazar, N. Bourgeois, J. A. Cardarelli, P. Hatfield, J. Kang, K. Krushelnick, S. P.D. Mangles, C. D. Murphy, N. Lu, J. Osterhoff, K. Põder, P. P. RajeevC. P. Ridgers, S. Rozario, M. P. Selwood, A. J. Shahani, D. R. Symes, A. G.R. Thomas, C. Thornton, Z. Najmudin, M. J.V. Streeter

^*Corresponding author for this work

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

87 Citations (Scopus)

62 Downloads (Pure)

Abstract

Laser wakefield accelerators promise to revolutionize many areas of accelerator science. However, one of the greatest challenges to their widespread adoption is the difficulty in control and optimization of the accelerator outputs due to coupling between input parameters and the dynamic evolution of the accelerating structure. Here, we use machine learning techniques to automate a 100 MeV-scale accelerator, which optimized its outputs by simultaneously varying up to six parameters including the spectral and spatial phase of the laser and the plasma density and length. Most notably, the model built by the algorithm enabled optimization of the laser evolution that might otherwise have been missed in single-variable scans. Subtle tuning of the laser pulse shape caused an 80% increase in electron beam charge, despite the pulse length changing by just 1%.

Original language	English
Article number	6355
Journal	Nature Communications
Volume	11
DOIs	https://doi.org/10.1038/s41467-020-20245-6
Publication status	Published - 11 Dec 2020
Externally published	Yes

Bibliographical note

Funding Information:
We gratefully acknowledge the hard work of the staff at the Central Laser Facility in the planning and execution of the experiment. R.J.S., J-N.G., M.B., S.P.D.M., Z.N. and M.J.V. S. acknowledge funding from Science and Technology Facilities Council Grant number ST/P002021/1 and the EU Horizon 2020 research and innovation programme grant number 653782. A.G.R.T. acknowledges funding from US NSF grant number 1804463 and US DOE/FES grant number DE-SC0020237. A.G.R.T. and K.K. acknowledge funding from US DOE/High Energy Physics grant number DE-SC0016804. C.T. acknowledges funding from the Engineering and Physical Sciences Research Council grant number EP/S001379/1.

Publisher Copyright:
© 2020, The Author(s).

ASJC Scopus subject areas

General Chemistry
General Biochemistry,Genetics and Molecular Biology
General Physics and Astronomy

Access to Document

10.1038/s41467-020-20245-6Licence: CC BY

Automation and control of laser wakefield accelerators using Bayesian optimization
Copyright 2020 the authors. This is an open access article published under a Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium, provided the author and source are cited
Final published version, 2.22 MBLicence: CC BY

Code and data associated with "Automation and control of laser wakefield accelerators using Bayesian optimisation"
Streeter, M. (Creator), Zenodo, 03 Nov 2020
DOI: 10.5281/zenodo.4229537
Dataset

Automation and modelling of laser-driven plasma accelerators
Matthew Streeter (Advisor)
07 Mar 2023
Activity: Talk or presentation types › Invited or keynote talk at national or international conference

Cite this

Shalloo, R. J., Dann, S. J. D., Gruse, J. N., Underwood, C. I. D., Antoine, A. F., Arran, C., Backhouse, M., Baird, C. D., Balcazar, M. D., Bourgeois, N., Cardarelli, J. A., Hatfield, P., Kang, J., Krushelnick, K., Mangles, S. P. D., Murphy, C. D., Lu, N., Osterhoff, J., Põder, K., ... Streeter, M. J. V. (2020). Automation and control of laser wakefield accelerators using Bayesian optimization. Nature Communications, 11, Article 6355. https://doi.org/10.1038/s41467-020-20245-6

@article{528091f0544241afb6974e96fe888574,

title = "Automation and control of laser wakefield accelerators using Bayesian optimization",

abstract = "Laser wakefield accelerators promise to revolutionize many areas of accelerator science. However, one of the greatest challenges to their widespread adoption is the difficulty in control and optimization of the accelerator outputs due to coupling between input parameters and the dynamic evolution of the accelerating structure. Here, we use machine learning techniques to automate a 100 MeV-scale accelerator, which optimized its outputs by simultaneously varying up to six parameters including the spectral and spatial phase of the laser and the plasma density and length. Most notably, the model built by the algorithm enabled optimization of the laser evolution that might otherwise have been missed in single-variable scans. Subtle tuning of the laser pulse shape caused an 80% increase in electron beam charge, despite the pulse length changing by just 1%.",

author = "Shalloo, {R. J.} and Dann, {S. J.D.} and Gruse, {J. N.} and Underwood, {C. I.D.} and Antoine, {A. F.} and C. Arran and M. Backhouse and Baird, {C. D.} and Balcazar, {M. D.} and N. Bourgeois and Cardarelli, {J. A.} and P. Hatfield and J. Kang and K. Krushelnick and Mangles, {S. P.D.} and Murphy, {C. D.} and N. Lu and J. Osterhoff and K. P{\~o}der and Rajeev, {P. P.} and Ridgers, {C. P.} and S. Rozario and Selwood, {M. P.} and Shahani, {A. J.} and Symes, {D. R.} and Thomas, {A. G.R.} and C. Thornton and Z. Najmudin and Streeter, {M. J.V.}",

note = "Funding Information: We gratefully acknowledge the hard work of the staff at the Central Laser Facility in the planning and execution of the experiment. R.J.S., J-N.G., M.B., S.P.D.M., Z.N. and M.J.V. S. acknowledge funding from Science and Technology Facilities Council Grant number ST/P002021/1 and the EU Horizon 2020 research and innovation programme grant number 653782. A.G.R.T. acknowledges funding from US NSF grant number 1804463 and US DOE/FES grant number DE-SC0020237. A.G.R.T. and K.K. acknowledge funding from US DOE/High Energy Physics grant number DE-SC0016804. C.T. acknowledges funding from the Engineering and Physical Sciences Research Council grant number EP/S001379/1. Publisher Copyright: {\textcopyright} 2020, The Author(s).",

year = "2020",

month = dec,

day = "11",

doi = "10.1038/s41467-020-20245-6",

language = "English",

volume = "11",

journal = "Nature Communications",

issn = "2041-1723",

publisher = "Nature Publishing Group",

}

Shalloo, RJ, Dann, SJD, Gruse, JN, Underwood, CID, Antoine, AF, Arran, C, Backhouse, M, Baird, CD, Balcazar, MD, Bourgeois, N, Cardarelli, JA, Hatfield, P, Kang, J, Krushelnick, K, Mangles, SPD, Murphy, CD, Lu, N, Osterhoff, J, Põder, K, Rajeev, PP, Ridgers, CP, Rozario, S, Selwood, MP, Shahani, AJ, Symes, DR, Thomas, AGR, Thornton, C, Najmudin, Z & Streeter, MJV 2020, 'Automation and control of laser wakefield accelerators using Bayesian optimization', Nature Communications, vol. 11, 6355. https://doi.org/10.1038/s41467-020-20245-6

TY - JOUR

T1 - Automation and control of laser wakefield accelerators using Bayesian optimization

AU - Shalloo, R. J.

AU - Dann, S. J.D.

AU - Gruse, J. N.

AU - Underwood, C. I.D.

AU - Antoine, A. F.

AU - Arran, C.

AU - Backhouse, M.

AU - Baird, C. D.

AU - Balcazar, M. D.

AU - Bourgeois, N.

AU - Cardarelli, J. A.

AU - Hatfield, P.

AU - Kang, J.

AU - Krushelnick, K.

AU - Mangles, S. P.D.

AU - Murphy, C. D.

AU - Lu, N.

AU - Osterhoff, J.

AU - Põder, K.

AU - Rajeev, P. P.

AU - Ridgers, C. P.

AU - Rozario, S.

AU - Selwood, M. P.

AU - Shahani, A. J.

AU - Symes, D. R.

AU - Thomas, A. G.R.

AU - Thornton, C.

AU - Najmudin, Z.

AU - Streeter, M. J.V.

N1 - Funding Information: We gratefully acknowledge the hard work of the staff at the Central Laser Facility in the planning and execution of the experiment. R.J.S., J-N.G., M.B., S.P.D.M., Z.N. and M.J.V. S. acknowledge funding from Science and Technology Facilities Council Grant number ST/P002021/1 and the EU Horizon 2020 research and innovation programme grant number 653782. A.G.R.T. acknowledges funding from US NSF grant number 1804463 and US DOE/FES grant number DE-SC0020237. A.G.R.T. and K.K. acknowledge funding from US DOE/High Energy Physics grant number DE-SC0016804. C.T. acknowledges funding from the Engineering and Physical Sciences Research Council grant number EP/S001379/1. Publisher Copyright: © 2020, The Author(s).

PY - 2020/12/11

Y1 - 2020/12/11

N2 - Laser wakefield accelerators promise to revolutionize many areas of accelerator science. However, one of the greatest challenges to their widespread adoption is the difficulty in control and optimization of the accelerator outputs due to coupling between input parameters and the dynamic evolution of the accelerating structure. Here, we use machine learning techniques to automate a 100 MeV-scale accelerator, which optimized its outputs by simultaneously varying up to six parameters including the spectral and spatial phase of the laser and the plasma density and length. Most notably, the model built by the algorithm enabled optimization of the laser evolution that might otherwise have been missed in single-variable scans. Subtle tuning of the laser pulse shape caused an 80% increase in electron beam charge, despite the pulse length changing by just 1%.

AB - Laser wakefield accelerators promise to revolutionize many areas of accelerator science. However, one of the greatest challenges to their widespread adoption is the difficulty in control and optimization of the accelerator outputs due to coupling between input parameters and the dynamic evolution of the accelerating structure. Here, we use machine learning techniques to automate a 100 MeV-scale accelerator, which optimized its outputs by simultaneously varying up to six parameters including the spectral and spatial phase of the laser and the plasma density and length. Most notably, the model built by the algorithm enabled optimization of the laser evolution that might otherwise have been missed in single-variable scans. Subtle tuning of the laser pulse shape caused an 80% increase in electron beam charge, despite the pulse length changing by just 1%.

U2 - 10.1038/s41467-020-20245-6

DO - 10.1038/s41467-020-20245-6

M3 - Article

C2 - 33311487

AN - SCOPUS:85097486626

SN - 2041-1723

VL - 11

JO - Nature Communications

JF - Nature Communications

M1 - 6355

ER -

Automation and control of laser wakefield accelerators using Bayesian optimization

Abstract

Bibliographical note

ASJC Scopus subject areas

Access to Document

Fingerprint

Datasets

Code and data associated with "Automation and control of laser wakefield accelerators using Bayesian optimisation"

Activities

Automation and modelling of laser-driven plasma accelerators

Cite this