Machine learning for design and control of particle accelerators: A look backward and forward
Particle accelerators are extremely complex machines that are challenging to simulate,
design, and control. Over the past decade, artificial intelligence (AI) and machine learning …
design, and control. Over the past decade, artificial intelligence (AI) and machine learning …
Adaptive autoencoder latent space tuning for more robust machine learning beyond the training set for six-dimensional phase space diagnostics of a time-varying …
We present a general adaptive latent space tuning approach for improving the robustness of
machine learning tools with respect to time variation and distribution shift. We demonstrate …
machine learning tools with respect to time variation and distribution shift. We demonstrate …
An adaptive approach to machine learning for compact particle accelerators
Abstract Machine learning (ML) tools are able to learn relationships between the inputs and
outputs of large complex systems directly from data. However, for time-varying systems, the …
outputs of large complex systems directly from data. However, for time-varying systems, the …
A conditional latent autoregressive recurrent model for generation and forecasting of beam dynamics in particle accelerators
Particle accelerators are complex systems that focus, guide, and accelerate intense charged
particle beams to high energy. Beam diagnostics present a challenging problem due to …
particle beams to high energy. Beam diagnostics present a challenging problem due to …
Adaptive machine learning for time-varying systems: low dimensional latent space tuning
A Scheinker - Journal of Instrumentation, 2021 - iopscience.iop.org
Abstract Machine learning (ML) tools such as encoder-decoder convolutional neural
networks (CNN) can represent incredibly complex nonlinear functions which map between …
networks (CNN) can represent incredibly complex nonlinear functions which map between …
Conditional guided generative diffusion for particle accelerator beam diagnostics
A Scheinker - Scientific Reports, 2024 - nature.com
Advanced accelerator-based light sources such as free electron lasers (FEL) accelerate
highly relativistic electron beams to generate incredibly short (10s of femtoseconds) …
highly relativistic electron beams to generate incredibly short (10s of femtoseconds) …
Specification and design for full energy beam exploitation of the compact linear accelerator for research and applications
EW Snedden, D Angal-Kalinin, AR Bainbridge… - … Review Accelerators and …, 2024 - APS
The compact linear accelerator for research and applications (CLARA) is a 250 MeV
ultrabright electron beam test facility at STFC Daresbury Laboratory. A user beamline has …
ultrabright electron beam test facility at STFC Daresbury Laboratory. A user beamline has …
[HTML][HTML] Physics-constrained 3D convolutional neural networks for electrodynamics
A Scheinker, R Pokharel - APL Machine Learning, 2023 - pubs.aip.org
We present a physics-constrained neural network (PCNN) approach to solving Maxwell's
equations for the electromagnetic fields of intense relativistic charged particle beams. We …
equations for the electromagnetic fields of intense relativistic charged particle beams. We …
Experimental safe extremum seeking for accelerators
We demonstrate the recent designs of safe extremum seeking (Safe ES) on the 1-km-long
charged particle accelerator at the Los Alamos Neutron Science Center (LANSCE). Safe ES …
charged particle accelerator at the Los Alamos Neutron Science Center (LANSCE). Safe ES …
Machine learning surrogate for charged particle beam dynamics with space charge based on a recurrent neural network with aleatoric uncertainty
C Garcia-Cardona, A Scheinker - Physical Review Accelerators and Beams, 2024 - APS
In this work, we develop a machine learning (ML) model with aleatoric uncertainty for the low
energy beam transport (LEBT) region of the LANSCE linear accelerator in which we model …
energy beam transport (LEBT) region of the LANSCE linear accelerator in which we model …