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%.

Laser wakefield accelerators are compact sources of ultra-relativistic electrons which are highly sensitive to many control parameters. Here the authors present an automated machine learning based method for the efficient multi-dimensional optimization of these plasma-based particle accelerators.

Details

Title
Automation and control of laser wakefield accelerators using Bayesian optimization
Author
Shalloo, R J 1   VIAFID ORCID Logo  ; Dann S J D 2   VIAFID ORCID Logo  ; J-N, Gruse 1   VIAFID ORCID Logo  ; Underwood C I D 3   VIAFID ORCID Logo  ; Antoine, A F 4   VIAFID ORCID Logo  ; Arran, C 3 ; Backhouse, M 1 ; Baird, C D 5 ; Balcazar, M D 4 ; Bourgeois, N 2 ; Cardarelli, J A 4 ; Hatfield, P 6   VIAFID ORCID Logo  ; Kang, J 7 ; Krushelnick, K 4 ; Mangles S P D 1   VIAFID ORCID Logo  ; Murphy, C D 3   VIAFID ORCID Logo  ; Lu N 8 ; Osterhoff, J 9   VIAFID ORCID Logo  ; Põder, K 9   VIAFID ORCID Logo  ; Rajeev, P P 2   VIAFID ORCID Logo  ; Ridgers, C P 3 ; Rozario, S 1 ; Selwood, M P 3   VIAFID ORCID Logo  ; Shahani, A J 8 ; Symes, D R 2 ; Thomas, A G, R 4   VIAFID ORCID Logo  ; Thornton, C 2   VIAFID ORCID Logo  ; Najmudin, Z 1   VIAFID ORCID Logo  ; Streeter, M J, V 1   VIAFID ORCID Logo 

 Imperial College London, The John Adams Institute for Accelerator Science, London, UK (GRID:grid.7445.2) (ISNI:0000 0001 2113 8111) 
 STFC Rutherford Appleton Laboratory, Central Laser Facility, Didcot, UK (GRID:grid.76978.37) (ISNI:0000 0001 2296 6998) 
 York Plasma Institute, University of York, Department of Physics, York, UK (GRID:grid.5685.e) (ISNI:0000 0004 1936 9668) 
 University of Michigan, Center for Ultrafast Optical Science, Ann Arbor, USA (GRID:grid.214458.e) (ISNI:0000000086837370) 
 STFC Rutherford Appleton Laboratory, Central Laser Facility, Didcot, UK (GRID:grid.76978.37) (ISNI:0000 0001 2296 6998); York Plasma Institute, University of York, Department of Physics, York, UK (GRID:grid.5685.e) (ISNI:0000 0004 1936 9668) 
 University of Oxford, Clarendon Laboratory, Oxford, UK (GRID:grid.4991.5) (ISNI:0000 0004 1936 8948) 
 University of Michigan, Department of Chemical Engineering, Ann Arbor, USA (GRID:grid.214458.e) (ISNI:0000000086837370) 
 University of Michigan, Department of Materials Science and Engineering, Ann Arbor, USA (GRID:grid.214458.e) (ISNI:0000000086837370) 
 Deutsches Elektronen-Synchrotron DESY, Hamburg, Germany (GRID:grid.7683.a) (ISNI:0000 0004 0492 0453) 
Publication year
2020
Publication date
2020
Publisher
Nature Publishing Group
e-ISSN
20411723
Source type
Scholarly Journal
Language of publication
English
ProQuest document ID
2473304816
Copyright
© The Author(s) 2020. This work is published under http://creativecommons.org/licenses/by/4.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.