DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Machine learning surrogate for charged particle beam dynamics with space charge based on a recurrent neural network with aleatoric uncertainty

Journal Article · · Physical Review Accelerators and Beams

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 the transport of a space-charge-dominated 750 keV proton beam through a lattice of 22 quadrupole magnets. Our ML model is developed based on data generated by a Kapchinsky–Vladimirsky (KV) envelope model of beam transport. We show that a recurrent neural network can be used as a dynamical surrogate model for fast prediction of the LEBT beam envelope. Furthermore, we endow the model with the prediction of aleatoric uncertainty and compare three different approaches. We demonstrate that the ML-based uncertainty quantification models are well calibrated and produce good estimates of the regions where the model is less certain about its predictions. This ML framework is a necessary step in the development of a real-time virtual diagnostic tool with uncertainty quantification that can be integrated into more complex downstream tasks (e.g., adaptive control or learning flexible control policies via reinforcement learning) for improved efficiency in beam operations. In future work, we plan to expand on this preliminary study by considering more realistic envelope models that include longitudinal momentum spread and dispersive effects in bending magnets, as well as particle tracking codes with 3D space charge (such as and ). Published by the American Physical Society 2024

Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA)
OSTI ID:
2315712
Alternate ID(s):
OSTI ID: 2406574
Journal Information:
Physical Review Accelerators and Beams, Journal Name: Physical Review Accelerators and Beams Journal Issue: 2 Vol. 27; ISSN 2469-9888; ISSN PRABCJ
Publisher:
American Physical SocietyCopyright Statement
Country of Publication:
United States
Language:
English

References (29)

Optimizing a superconducting radio-frequency gun using deep reinforcement learning journal October 2022
Physics-constrained 3D convolutional neural networks for electrodynamics journal April 2023
Uncertainty quantification for virtual diagnostic of particle accelerators journal July 2021
Simulation study of the space charge limit in heavy-ion synchrotrons journal May 2022
Particle-core model for transverse dynamics of beam halo journal December 1998
Genetic algorithm enhanced by machine learning in dynamic aperture optimization journal May 2018
Explainable machine learning for breakdown prediction in high gradient rf cavities journal October 2022
Identification of magnetic field errors in synchrotrons based on deep Lie map networks journal June 2023
Tolerance regions for a multivariate normal population journal December 1964
High-Fidelity Prediction of Megapixel Longitudinal Phase-Space Images of Electron Beams Using Encoder-Decoder Neural Networks journal August 2021
Uncertainty aware anomaly detection to predict errant beam pulses in the Oak Ridge Spallation Neutron Source accelerator journal December 2022
Virtual-diagnostic-based time stamping for ultrafast electron diffraction journal May 2023
Transverse phase space tomography in an accelerator test facility using image compression and machine learning journal December 2022
Predicting the transverse emittance of space charge dominated beams using the phase advance scan technique and a fully connected neural network journal September 2022
SciPy 1.0: fundamental algorithms for scientific computing in Python journal February 2020
Long Short-Term Memory journal November 1997
Demonstration of Model-Independent Control of the Longitudinal Phase Space of Electron Beams in the Linac-Coherent Light Source with Femtosecond Resolution journal July 2018
A review of uncertainty quantification in deep learning: Techniques, applications and challenges journal December 2021
Bounded extremum seeking with discontinuous dithers journal July 2016
On Nonintrusive Uncertainty Quantification and Surrogate Model Construction in Particle Accelerator Modeling journal January 2019
Uncertainty aware machine-learning-based surrogate models for particle accelerators: Study at the Fermilab Booster Accelerator Complex journal April 2023
GPU accelerated online multi-particle beam dynamics simulator for ion linear particle accelerators journal March 2014
An adaptive approach to machine learning for compact particle accelerators journal September 2021
Superconducting radio-frequency cavity fault classification using machine learning at Jefferson Laboratory journal November 2020
Adaptive autoencoder latent space tuning for more robust machine learning beyond the training set for six-dimensional phase space diagnostics of a time-varying ultrafast electron-diffraction compact accelerator journal April 2023
Self-consistent long-term dynamics of space charge driven resonances in 2D and 3D journal February 2021
rms Envelope Equations in the Presence of Space Charge and Dispersion journal July 1998
Physics-based deep neural networks for beam dynamics in charged particle accelerators journal July 2020
Machine-learning calibration of intense x-ray free-electron-laser pulses using Bayesian optimization journal May 2023