NAPAC2022 - List of Keywords (network)

Paper	Title	Other Keywords	Page
MOPA41	Diagnostics for LINAC Optimization with Machine Learning	linac, DTL, controls, diagnostics	139
	R.V. Sharankova, M.W. Mwaniki, K. Seiya, M.E. Wesley Fermilab, Batavia, Illinois, USA
	The Fermilab Linac delivers 400 MeV H⁻ beam to the rest of the accelerator chain. Providing stable intensity, energy, and emittance is key since it directly affects downstream machines. To operate high current beam, accelerators must minimize uncontrolled particle loss; this is generally accomplished by minimizing beam emittance. Ambient temperature and humidity variations are known to affect resonance frequency of the accelerating cavities which induces emittance growth. In addition, the energy and phase space distribution of particles emerging from the ion source are subject to fluctuations. To counter these effects we are working on implementing dynamic longitudinal parameter optimization based on Machine Learning (ML). As an input for the ML model, signals from beam diagnostic have to be well understand and reliable. We have been revisiting diagnostics in the linac. In this presentation we discuss the status of the diagnostics and beam studies as well as the status and plans for ML-based optimization.
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-MOPA41
About •	Received ※ 05 August 2022 — Accepted ※ 06 August 2022 — Issue date ※ 07 September 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

MOPA89	RHIC Electron Beam Cooling Analysis Using Principle Component and Autoencoder Analysis	luminosity, electron, ECR, beam-cooling	260
	A.D. Tran, Y. Hao FRIB, East Lansing, Michigan, USA X. Gu BNL, Upton, New York, USA
	Funding: Work supported by the US Department of Energy under contract No. DE-AC02-98CH10886. Principal component analysis and autoencoder analysis were used to analyze the experimental data of RHIC operation with low energy RHIC electron cooling (LEReC). This is unsupervised learning which includes electron beam settings and observable during operation. Both analyses were used to gauge the dimensional reducibility of the data and to understand which features are important to beam cooling.
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-MOPA89
About •	Received ※ 02 August 2022 — Revised ※ 05 August 2022 — Accepted ※ 06 August 2022 — Issue date ※ 12 August 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

MOPA90	Relating Initial Distribution to Beam Loss on the Front End of a Heavy-Ion Linac Using Machine Learning	simulation, LEBT, emittance, controls	263
	A.D. Tran, Y. Hao FRIB, East Lansing, Michigan, USA J.L. Martinez Marin, B. Mustapha ANL, Lemont, Illinois, USA
	Funding: This work was supported by a sub-reward from Argonne National Laboratory and supported by the U.S. Department of Energy, under Contract No. DE-AC02-06CH11357. This work demonstrates using a Neural Network and a Gaussian Process to model the ATLAS front-end. Various neural network architectures were created and trained on the machine settings and outputs to model the phase space projections. The model was then trained on a dataset, with non-linear distortion, to gauge the transferability of the model from simulation to machine.
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-MOPA90
About •	Received ※ 02 August 2022 — Revised ※ 05 August 2022 — Accepted ※ 06 August 2022 — Issue date ※ 11 September 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUYE3	An Open-Source Based Data Management and Processing Framework on a Central Server for Scientific Experimental Data	framework, experiment, software, data-management	307
	A. Liu, J.R. Callahan, S. Poddar, W. Si Euclid TechLabs, Solon, Ohio, USA J. Gao AJS Smartech LLC, Naperville, TX, USA
	Funding: This work is supported by the US DOE SBIR program under contract number DE-SC0021512. The ever-expanding size of accelerator operation and experimental data including those generated by electron microscopes and beamline facilities renders most proprietary software inefficient at managing data. The Findability, Accessibility, Interoperability, and Reuse (FAIR) principles of digital assets require a convenient platform for users to share and manage data on. An open-source data framework for storing raw data and metadata, hosting databases, and providing a platform for data processing and visualization is highly desirable. In this paper, we present an open-source, infrastructure-independent data management software framework, named by Euclid-NexusLIMS, to archive, register, record, visualize and process experimental data. The software was targeted initially for electron microscopes, but can be widely applied to all scientific experimental data.
	Slides TUYE3 [5.891 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUYE3
About •	Received ※ 04 August 2022 — Revised ※ 07 August 2022 — Accepted ※ 08 August 2022 — Issue date ※ 24 August 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUYE4	Machine Learning for Anomaly Detection and Classification in Particle Accelerators	injection, linac, operation, controls	311
	I. Lobach, M. Borland, K.C. Harkay, N. Kuklev, A. Sannibale, Y. Sun ANL, Lemont, Illinois, USA
	Funding: The work is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. We explore the possibility of using a Machine Learning (ML) algorithm to identify the source of occasional poor performance of the Particle Accumulator Ring (PAR) and the Linac-To-PAR (LTP) transport line, which are parts of the injector complex of the Advanced Photon Source (APS) at Argonne National Lab. The cause of reduced injection or extraction efficiencies may be as simple as one parameter being out of range. Still, it may take an expert considerable time to notice it, whereas a well-trained ML model can point at it instantly. In addition, a machine expert might not be immediately available when a problem occurs. Therefore, we began by focusing on such single-parameter anomalies. The training data were generated by creating controlled perturbations of several parameters of PAR and LTP one-by-one, while continuously logging all available process variables. Then, several ML classifiers were trained to recognize certain signatures in the logged data and link them to the sources of poor machine performance. Possible applications of autoencoders and variational autoencoders for unsupervised anomaly detection and for anomaly clustering were considered as well.
	Slides TUYE4 [9.534 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUYE4
About •	Received ※ 03 August 2022 — Revised ※ 07 August 2022 — Accepted ※ 08 August 2022 — Issue date ※ 28 August 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA29	Machine Learning for Predicting Power Supply Trips in Storage Rings	storage-ring, power-supply, quadrupole, sextupole	413
	I. Lobach, M. Borland, G.I. Fystro, A. Sannibale, Y. Sun ANL, Lemont, Illinois, USA A. Diaw, J.P. Edelen RadiaSoft LLC, Boulder, Colorado, USA
	Funding: The work is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357. In the Advanced Photon Source (APS) storage ring at Argonne National Lab, trips in the magnet power supplies (PSs) lead to a complete electron beam loss a few times a year. This results in unexpected interruptions of the users’ experiments. In this contribution, we investigate the historical data for the last two decades to find precursors for the PS trips that could provide an advance notice for future trips and allow some preventive action by the ring operator or by the PS maintenance team. Various unsupervised anomaly detection models can be trained on the vast amounts of available reference data from the beamtime periods that ended with an intentional beam dump. We find that such models can sometimes detect trip precursors in PS currents, voltages, and in the temperatures of magnets, capacitors and transistors (components of PSs).
	Poster TUPA29 [2.116 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA29
About •	Received ※ 03 August 2022 — Revised ※ 07 August 2022 — Accepted ※ 10 August 2022 — Issue date ※ 18 August 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA34	Model-Based Calibration of Control Parameters at the Argonne Wakefield Accelerator	controls, gun, simulation, wakefield	427
	I.P. Sugrue, B. Mustapha, P. Piot, J.G. Power ANL, Lemont, Illinois, USA N. Krislock Northern Illinois University, DeKalb, Illinois, USA
	Particle accelerators utilize a large number of control parameters to generate and manipulate beams. Digital models and simulations are often used to find the best operating parameters to achieve a set of given beam parameters. Unfortunately, the optimized physics parameters cannot precisely be set in the control system due to, e.g., calibration uncertainties. We developed a data-driven physics-informed surrogate model using neural networks to replace digital models relying on beam-dynamics simulations. This surrogate model can then be used to perform quick diagnostics of the Argonne Wakefield accelerator in real time using nonlinear least-squares methods to find the most likely operating parameters given a measured beam distribution.
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA34
About •	Received ※ 05 August 2022 — Revised ※ 09 August 2022 — Accepted ※ 10 August 2022 — Issue date ※ 24 September 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA41	Applications of Machine Learning in Photo-Cathode Injectors	laser, electron, controls, cathode	441
	A. Aslam UNM-ECE, Albuquerque, USA M. Babzien BNL, Upton, New York, USA S. Biedron Element Aero, Chicago, USA
	To configure a photoinjector to reproduce a given electron bunch with the desired characteristics, it is necessary to adjust the operating parameters with high precision. More or less, the fine tunability of the laser parameters are of extreme importance as we try to model further applications of the photoinjector. The laser pulse incident on the photocathode critically affects the electron bunch 3D phase space. Parameters such as the laser pulse transverse shape, total energy, and temporal profile must be controlled independently, any laser pulse variation over both short and long-time scales also requires correction. The ability to produce arbitrary laser intensity distributions enables better control of electron bunch transverse and longitudinal emittance by affecting the space-charge forces throughout the bunch. In an accelerator employing a photoinjector, electron optics in the beamline downstream are used to transport, manipulate, and characterize the electron bunch. The adjustment of the electron optics to achieve a desired electron bunch at the interaction point is a much better understood problem than laser adjustment, so this research emphasizes laser shaping.
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA41
About •	Received ※ 30 July 2022 — Revised ※ 12 August 2022 — Accepted ※ 13 August 2022 — Issue date ※ 07 September 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA53	Modeling of Nonlinear Beam Dynamics via a Novel Particle-Mesh Method and Surrogate Models with Symplectic Neural Networks	simulation, electron, radiation, synchrotron-radiation	462
	C.-K. Huang, O. Beznosov, J.W. Burby, B.E. Carlsten, G.A. Dilts, J. Domine, R. Garimella, A. Kim, T.J. Kwan, H.N. Rakotoarivelo, R.W. Robey, B. Shen, Q. Tang LANL, Los Alamos, New Mexico, USA F.Y. Li New Mexico Consortium, Los Alamos, USA
	Funding: Work supported by the LDRD program at Los Alamos National Laboratory and the ASCR SciML program of DOE. The self-consistent nonlinear dynamics of a relativistic charged particle beam, particularly through the interaction with its complete self-fields, is a fundamental problem underpinning many accelerator design issues in high brightness beam applications, as well as the development of advanced accelerators. A novel self-consistent particle-mesh code, CoSyR [1], is developed based on a Lagrangian method for the calculation of the beam particles’ radiation near-fields and associated beam dynamics. Our recent simulations reveal the slice emittance growth in a bend and complex interplay between the longitudinal and transverse dynamics that are not captured in the 1D longitudinal static-state Coherent Synchrotron Radiation (CSR) model. We further show that surrogate models with symplectic neural networks can be trained from simulation data with significant time-savings for the modeling of nonlinear beam dynamics effects. Possibility to extend such surrogate models for the study of spin-orbital coupling is also briefly discussed. [1] C.-K. Huang et al., Nucl. Instruments Methods Phys. Res. Sect. A, vol. 1034, p. 166808, 2022.
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA53
About •	Received ※ 25 July 2022 — Revised ※ 03 August 2022 — Accepted ※ 09 August 2022 — Issue date ※ 11 August 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA62	LANSCE Control System’s 50th Anniversary	controls, EPICS, timing, data-acquisition	482
	M. Pieck, C.D. Hatch, J.O. Hill, H.A. Watkins, E.E. Westbrook LANL, Los Alamos, New Mexico, USA
	After almost exactly 50 years in service, the LANSCE (Los Alamos Neutron Science Center) control system has achieved a major milestone, replacing its original and reliable RICE (Remote Instrumentation and Control Equipment) with a modern customized control system. The task of replacing RICE was challenging because of its technology (late 1960’s), number of channels (>10,000), unique characteristics (all-modules data takes, timed/flavored data takes) and that it was designed as an integral part of the whole accelerator. We discuss the history, RICE integral architecture, upgrade efforts, and the new system providing cutting-edge capabilities. The boundary condition was that upgrades only could be implemented during the annual four-month accelerator maintenance outage. This led to a multi-phased project which turned out to be about an 11-year effort.
	Poster TUPA62 [1.985 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA62
About •	Received ※ 02 August 2022 — Revised ※ 09 August 2022 — Accepted ※ 10 August 2022 — Issue date ※ 25 September 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

WEPA29	Real-Time Cavity Fault Prediction in CEBAF Using Deep Learning	cavity, cryomodule, SRF, experiment	687
	M. Rahman, K.M. Iftekharuddin ODU, Norfolk, Virginia, USA A. Carpenter, T.S. McGuckin, C. Tennant, L.S. Vidyaratne JLab, Newport News, Virginia, USA
	Funding: Authored by Jefferson Science Associates, LLC under U.S. DOE Contract No. DE-AC05-06OR23177. Data-driven prediction of future faults is a major research area for many industrial applications. In this work, we present a new procedure of real-time fault prediction for superconducting radio-frequency (SRF) cavities at the Continuous Electron Beam Accelerator Facility (CEBAF) using deep learning. CEBAF has been afflicted by frequent downtime caused by SRF cavity faults. We perform fault prediction using pre-fault RF signals from C100-type cryomodules. Using the pre-fault signal information, the new algorithm predicts the type of cavity fault before the actual onset. The early prediction may enable potential mitigation strategies to prevent the fault. In our work, we apply a two-stage fault prediction pipeline. In the first stage, a model distinguishes between faulty and normal signals using a U-Net deep learning architecture. In the second stage of the network, signals flagged as faulty by the first model are classified into one of seven fault types based on learned signatures in the data. Initial results show that our model can successfully predict most fault types 200 ms before onset. We will discuss reasons for poor model performance on specific fault types.
	Poster WEPA29 [1.339 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-WEPA29
About •	Received ※ 02 August 2022 — Revised ※ 07 August 2022 — Accepted ※ 11 August 2022 — Issue date ※ 17 August 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

WEPA38	Progress on Machine Learning for the SNS High Voltage Converter Modulators	klystron, linac, electron, simulation	715
	M.I. Radaideh, S.M. Cousineau, D. Lu ORNL, Oak Ridge, Tennessee, USA T.J. Britton, K. Rajput, M. Schram, L.S. Vidyaratne JLab, Newport News, Virginia, USA G.C. Pappas, J.D. Walden ORNL RAD, Oak Ridge, Tennessee, USA
	The High-Voltage Converter Modulators (HVCM) used to power the klystrons in the Spallation Neutron Source (SNS) linac were selected as one area to explore machine learning due to reliability issues in the past and the availability of large sets of archived waveforms. Progress in the past two years has resulted in generating a significant amount of simulated and measured data for training neural network models such as recurrent neural networks, convolutional neural networks, and variational autoencoders. Applications in anomaly detection, fault classification, and prognostics of capacitor degradation were pursued in collaboration with the Jefferson Laboratory, and early promising results were achieved. This paper will discuss the progress to date and present results from these efforts.
	Poster WEPA38 [1.320 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-WEPA38
About •	Received ※ 25 July 2022 — Revised ※ 08 August 2022 — Accepted ※ 11 August 2022 — Issue date ※ 03 October 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

WEPA40	The L-CAPE Project at FNAL	controls, operation, linac, alignment	719
	M. Jain, V.C. Amatya, G.U. Panapitiya, J.F. Strube PNNL, Richland, Washington, USA B.F. Harrison, K.J. Hazelwood, W. Pellico, B.A. Schupbach, K. Seiya, J.M. St. John Fermilab, Batavia, Illinois, USA
	The controls system at FNAL records data asynchronously from several thousand Linac devices at their respective cadences, ranging from 15Hz down to once per minute. In case of downtimes, current operations are mostly reactive, investigating the cause of an outage and labeling it after the fact. However, as one of the most upstream systems at the FNAL accelerator complex, the Linac’s foreknowledge of an impending downtime as well as its duration could prompt downstream systems to go into standby, potentially leading to energy savings. The goals of the Linac Condition Anomaly Prediction of Emergence (L-CAPE) project that started in late 2020 are (1) to apply data-analytic methods to improve the information that is available to operators in the control room, and (2) to use machine learning to automate the labeling of outage types as they occur and discover patterns in the data that could lead to the prediction of outages. We present an overview of the challenges in dealing with time-series data from 2000+ devices, our approach to developing an ML-based automated outage labeling system, and the status of augmenting operations by identifying the most likely devices predicting an outage.
	Poster WEPA40 [1.870 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-WEPA40
About •	Received ※ 03 August 2022 — Revised ※ 12 August 2022 — Accepted ※ 17 August 2022 — Issue date ※ 31 August 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

THXD2	6D Phase Space Diagnostics Based on Adaptive Tuning of the Latent Space of Encoder-Decoder Convolutional Neural Networks	controls, solenoid, feedback, electron	837
	A. Scheinker LANL, Los Alamos, New Mexico, USA
	We present a general approach to 6D phase space diagnostics for charged particle beams based on adaptively tuning the low-dimensional latent space of generative encoder-decoder convolutional neural networks (CNN). Our approach first trains the CNN based on supervised learning to learn the correlations and physics constrains within a given accelerator system. The input of the CNN is a high dimensional collection of 2D phase space projections of the beam at the accelerator entrance together with a vector of accelerator parameters such as magnet and RF settings. The inputs are squeezed down to a low-dimensional latent space from which we generate the output in the form of projections of the beam’s 6D phase space at various accelerator locations. After training the CNN is applied in an unsupervised adaptive manner by comparing a subset of the output predictions to available measurements with the error guiding feedback directly in the low-dimensional latent space. We show that our approach is robust to unseen time-variation of the input beam and accelerator parameters and a study of the robustness of the method to go beyond the span of the training data.
	Slides THXD2 [19.086 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-THXD2
About •	Received ※ 18 July 2022 — Revised ※ 05 August 2022 — Accepted ※ 08 August 2022 — Issue date ※ 09 August 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

FRXE1	Bayesian Algorithms for Practical Accelerator Control and Adaptive Machine Learning for Time-Varying Systems	controls, feedback, experiment, electron	921
	A. Scheinker LANL, Los Alamos, New Mexico, USA R.J. Roussel SLAC, Menlo Park, California, USA
	Particle accelerators are complicated machines with thousands of coupled time varying components. The electromagnetic fields of accelerator devices such as magnets and RF cavities drift and are uncertain due to external disturbances, vibrations, temperature changes, and hysteresis. Accelerated charged particle beams are complex objects with 6D phase space dynamics governed by collective effects such as space charge forces, coherent synchrotron radiation, and whose initial phase space distributions change in unexpected and difficult to measure ways. This two-part tutorial presents recent developments in Bayesian methods and adaptive machine learning (ML) techniques for accelerators. Part 1: We introduce Bayesian control algorithms, and we describe how these algorithms can be customized to solve practical accelerator specific problems, including online characterization and optimization. Part 2: We give an overview of adaptive ML (AML) combining adaptive model-independent feedback within physics-informed ML architectures to make ML tools robust to time-variation (distribution shift) and to enable their use further beyond the span of the training data without relying on re-training.
	Slides FRXE1 [34.283 MB]
DOI •	reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-FRXE1
About •	Received ※ 08 August 2022 — Revised ※ 10 August 2022 — Accepted ※ 12 August 2022 — Issue date ※ 27 September 2022
Cite •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

Paper

Title

Other Keywords

Page

MOPA41

Diagnostics for LINAC Optimization with Machine Learning

linac, DTL, controls, diagnostics

139

R.V. Sharankova, M.W. Mwaniki, K. Seiya, M.E. Wesley
Fermilab, Batavia, Illinois, USA

The Fermilab Linac delivers 400 MeV H⁻ beam to the rest of the accelerator chain. Providing stable intensity, energy, and emittance is key since it directly affects downstream machines. To operate high current beam, accelerators must minimize uncontrolled particle loss; this is generally accomplished by minimizing beam emittance. Ambient temperature and humidity variations are known to affect resonance frequency of the accelerating cavities which induces emittance growth. In addition, the energy and phase space distribution of particles emerging from the ion source are subject to fluctuations. To counter these effects we are working on implementing dynamic longitudinal parameter optimization based on Machine Learning (ML). As an input for the ML model, signals from beam diagnostic have to be well understand and reliable. We have been revisiting diagnostics in the linac. In this presentation we discuss the status of the diagnostics and beam studies as well as the status and plans for ML-based optimization.

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-MOPA41

About •

Received ※ 05 August 2022 — Accepted ※ 06 August 2022 — Issue date ※ 07 September 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

MOPA89

RHIC Electron Beam Cooling Analysis Using Principle Component and Autoencoder Analysis

luminosity, electron, ECR, beam-cooling

260

A.D. Tran, Y. Hao
FRIB, East Lansing, Michigan, USA
X. Gu
BNL, Upton, New York, USA

Funding: Work supported by the US Department of Energy under contract No. DE-AC02-98CH10886.
Principal component analysis and autoencoder analysis were used to analyze the experimental data of RHIC operation with low energy RHIC electron cooling (LEReC). This is unsupervised learning which includes electron beam settings and observable during operation. Both analyses were used to gauge the dimensional reducibility of the data and to understand which features are important to beam cooling.

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-MOPA89

About •

Received ※ 02 August 2022 — Revised ※ 05 August 2022 — Accepted ※ 06 August 2022 — Issue date ※ 12 August 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

MOPA90

Relating Initial Distribution to Beam Loss on the Front End of a Heavy-Ion Linac Using Machine Learning

simulation, LEBT, emittance, controls

263

A.D. Tran, Y. Hao
FRIB, East Lansing, Michigan, USA
J.L. Martinez Marin, B. Mustapha
ANL, Lemont, Illinois, USA

Funding: This work was supported by a sub-reward from Argonne National Laboratory and supported by the U.S. Department of Energy, under Contract No. DE-AC02-06CH11357.
This work demonstrates using a Neural Network and a Gaussian Process to model the ATLAS front-end. Various neural network architectures were created and trained on the machine settings and outputs to model the phase space projections. The model was then trained on a dataset, with non-linear distortion, to gauge the transferability of the model from simulation to machine.

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-MOPA90

About •

Received ※ 02 August 2022 — Revised ※ 05 August 2022 — Accepted ※ 06 August 2022 — Issue date ※ 11 September 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUYE3

An Open-Source Based Data Management and Processing Framework on a Central Server for Scientific Experimental Data

framework, experiment, software, data-management

307

A. Liu, J.R. Callahan, S. Poddar, W. Si
Euclid TechLabs, Solon, Ohio, USA
J. Gao
AJS Smartech LLC, Naperville, TX, USA

Funding: This work is supported by the US DOE SBIR program under contract number DE-SC0021512.
The ever-expanding size of accelerator operation and experimental data including those generated by electron microscopes and beamline facilities renders most proprietary software inefficient at managing data. The Findability, Accessibility, Interoperability, and Reuse (FAIR) principles of digital assets require a convenient platform for users to share and manage data on. An open-source data framework for storing raw data and metadata, hosting databases, and providing a platform for data processing and visualization is highly desirable. In this paper, we present an open-source, infrastructure-independent data management software framework, named by Euclid-NexusLIMS, to archive, register, record, visualize and process experimental data. The software was targeted initially for electron microscopes, but can be widely applied to all scientific experimental data.

Slides TUYE3 [5.891 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUYE3

About •

Received ※ 04 August 2022 — Revised ※ 07 August 2022 — Accepted ※ 08 August 2022 — Issue date ※ 24 August 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUYE4

Machine Learning for Anomaly Detection and Classification in Particle Accelerators

injection, linac, operation, controls

311

I. Lobach, M. Borland, K.C. Harkay, N. Kuklev, A. Sannibale, Y. Sun
ANL, Lemont, Illinois, USA

Funding: The work is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.
We explore the possibility of using a Machine Learning (ML) algorithm to identify the source of occasional poor performance of the Particle Accumulator Ring (PAR) and the Linac-To-PAR (LTP) transport line, which are parts of the injector complex of the Advanced Photon Source (APS) at Argonne National Lab. The cause of reduced injection or extraction efficiencies may be as simple as one parameter being out of range. Still, it may take an expert considerable time to notice it, whereas a well-trained ML model can point at it instantly. In addition, a machine expert might not be immediately available when a problem occurs. Therefore, we began by focusing on such single-parameter anomalies. The training data were generated by creating controlled perturbations of several parameters of PAR and LTP one-by-one, while continuously logging all available process variables. Then, several ML classifiers were trained to recognize certain signatures in the logged data and link them to the sources of poor machine performance. Possible applications of autoencoders and variational autoencoders for unsupervised anomaly detection and for anomaly clustering were considered as well.

Slides TUYE4 [9.534 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUYE4

About •

Received ※ 03 August 2022 — Revised ※ 07 August 2022 — Accepted ※ 08 August 2022 — Issue date ※ 28 August 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA29

Machine Learning for Predicting Power Supply Trips in Storage Rings

storage-ring, power-supply, quadrupole, sextupole

413

I. Lobach, M. Borland, G.I. Fystro, A. Sannibale, Y. Sun
ANL, Lemont, Illinois, USA
A. Diaw, J.P. Edelen
RadiaSoft LLC, Boulder, Colorado, USA

Funding: The work is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DE-AC02-06CH11357.
In the Advanced Photon Source (APS) storage ring at Argonne National Lab, trips in the magnet power supplies (PSs) lead to a complete electron beam loss a few times a year. This results in unexpected interruptions of the users’ experiments. In this contribution, we investigate the historical data for the last two decades to find precursors for the PS trips that could provide an advance notice for future trips and allow some preventive action by the ring operator or by the PS maintenance team. Various unsupervised anomaly detection models can be trained on the vast amounts of available reference data from the beamtime periods that ended with an intentional beam dump. We find that such models can sometimes detect trip precursors in PS currents, voltages, and in the temperatures of magnets, capacitors and transistors (components of PSs).

Poster TUPA29 [2.116 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA29

About •

Received ※ 03 August 2022 — Revised ※ 07 August 2022 — Accepted ※ 10 August 2022 — Issue date ※ 18 August 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA34

Model-Based Calibration of Control Parameters at the Argonne Wakefield Accelerator

controls, gun, simulation, wakefield

427

I.P. Sugrue, B. Mustapha, P. Piot, J.G. Power
ANL, Lemont, Illinois, USA
N. Krislock
Northern Illinois University, DeKalb, Illinois, USA

Particle accelerators utilize a large number of control parameters to generate and manipulate beams. Digital models and simulations are often used to find the best operating parameters to achieve a set of given beam parameters. Unfortunately, the optimized physics parameters cannot precisely be set in the control system due to, e.g., calibration uncertainties. We developed a data-driven physics-informed surrogate model using neural networks to replace digital models relying on beam-dynamics simulations. This surrogate model can then be used to perform quick diagnostics of the Argonne Wakefield accelerator in real time using nonlinear least-squares methods to find the most likely operating parameters given a measured beam distribution.

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA34

About •

Received ※ 05 August 2022 — Revised ※ 09 August 2022 — Accepted ※ 10 August 2022 — Issue date ※ 24 September 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA41

Applications of Machine Learning in Photo-Cathode Injectors

laser, electron, controls, cathode

441

A. Aslam
UNM-ECE, Albuquerque, USA
M. Babzien
BNL, Upton, New York, USA
S. Biedron
Element Aero, Chicago, USA

To configure a photoinjector to reproduce a given electron bunch with the desired characteristics, it is necessary to adjust the operating parameters with high precision. More or less, the fine tunability of the laser parameters are of extreme importance as we try to model further applications of the photoinjector. The laser pulse incident on the photocathode critically affects the electron bunch 3D phase space. Parameters such as the laser pulse transverse shape, total energy, and temporal profile must be controlled independently, any laser pulse variation over both short and long-time scales also requires correction. The ability to produce arbitrary laser intensity distributions enables better control of electron bunch transverse and longitudinal emittance by affecting the space-charge forces throughout the bunch. In an accelerator employing a photoinjector, electron optics in the beamline downstream are used to transport, manipulate, and characterize the electron bunch. The adjustment of the electron optics to achieve a desired electron bunch at the interaction point is a much better understood problem than laser adjustment, so this research emphasizes laser shaping.

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA41

About •

Received ※ 30 July 2022 — Revised ※ 12 August 2022 — Accepted ※ 13 August 2022 — Issue date ※ 07 September 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA53

Modeling of Nonlinear Beam Dynamics via a Novel Particle-Mesh Method and Surrogate Models with Symplectic Neural Networks

simulation, electron, radiation, synchrotron-radiation

462

C.-K. Huang, O. Beznosov, J.W. Burby, B.E. Carlsten, G.A. Dilts, J. Domine, R. Garimella, A. Kim, T.J. Kwan, H.N. Rakotoarivelo, R.W. Robey, B. Shen, Q. Tang
LANL, Los Alamos, New Mexico, USA
F.Y. Li
New Mexico Consortium, Los Alamos, USA

Funding: Work supported by the LDRD program at Los Alamos National Laboratory and the ASCR SciML program of DOE.
The self-consistent nonlinear dynamics of a relativistic charged particle beam, particularly through the interaction with its complete self-fields, is a fundamental problem underpinning many accelerator design issues in high brightness beam applications, as well as the development of advanced accelerators. A novel self-consistent particle-mesh code, CoSyR [1], is developed based on a Lagrangian method for the calculation of the beam particles’ radiation near-fields and associated beam dynamics. Our recent simulations reveal the slice emittance growth in a bend and complex interplay between the longitudinal and transverse dynamics that are not captured in the 1D longitudinal static-state Coherent Synchrotron Radiation (CSR) model. We further show that surrogate models with symplectic neural networks can be trained from simulation data with significant time-savings for the modeling of nonlinear beam dynamics effects. Possibility to extend such surrogate models for the study of spin-orbital coupling is also briefly discussed.
[1] C.-K. Huang et al., Nucl. Instruments Methods Phys. Res. Sect. A, vol. 1034, p. 166808, 2022.

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA53

About •

Received ※ 25 July 2022 — Revised ※ 03 August 2022 — Accepted ※ 09 August 2022 — Issue date ※ 11 August 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPA62

LANSCE Control System’s 50th Anniversary

controls, EPICS, timing, data-acquisition

482

M. Pieck, C.D. Hatch, J.O. Hill, H.A. Watkins, E.E. Westbrook
LANL, Los Alamos, New Mexico, USA

After almost exactly 50 years in service, the LANSCE (Los Alamos Neutron Science Center) control system has achieved a major milestone, replacing its original and reliable RICE (Remote Instrumentation and Control Equipment) with a modern customized control system. The task of replacing RICE was challenging because of its technology (late 1960’s), number of channels (>10,000), unique characteristics (all-modules data takes, timed/flavored data takes) and that it was designed as an integral part of the whole accelerator. We discuss the history, RICE integral architecture, upgrade efforts, and the new system providing cutting-edge capabilities. The boundary condition was that upgrades only could be implemented during the annual four-month accelerator maintenance outage. This led to a multi-phased project which turned out to be about an 11-year effort.

Poster TUPA62 [1.985 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-TUPA62

About •

Received ※ 02 August 2022 — Revised ※ 09 August 2022 — Accepted ※ 10 August 2022 — Issue date ※ 25 September 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

WEPA29

Real-Time Cavity Fault Prediction in CEBAF Using Deep Learning

cavity, cryomodule, SRF, experiment

687

M. Rahman, K.M. Iftekharuddin
ODU, Norfolk, Virginia, USA
A. Carpenter, T.S. McGuckin, C. Tennant, L.S. Vidyaratne
JLab, Newport News, Virginia, USA

Funding: Authored by Jefferson Science Associates, LLC under U.S. DOE Contract No. DE-AC05-06OR23177.
Data-driven prediction of future faults is a major research area for many industrial applications. In this work, we present a new procedure of real-time fault prediction for superconducting radio-frequency (SRF) cavities at the Continuous Electron Beam Accelerator Facility (CEBAF) using deep learning. CEBAF has been afflicted by frequent downtime caused by SRF cavity faults. We perform fault prediction using pre-fault RF signals from C100-type cryomodules. Using the pre-fault signal information, the new algorithm predicts the type of cavity fault before the actual onset. The early prediction may enable potential mitigation strategies to prevent the fault. In our work, we apply a two-stage fault prediction pipeline. In the first stage, a model distinguishes between faulty and normal signals using a U-Net deep learning architecture. In the second stage of the network, signals flagged as faulty by the first model are classified into one of seven fault types based on learned signatures in the data. Initial results show that our model can successfully predict most fault types 200 ms before onset. We will discuss reasons for poor model performance on specific fault types.

Poster WEPA29 [1.339 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-WEPA29

About •

Received ※ 02 August 2022 — Revised ※ 07 August 2022 — Accepted ※ 11 August 2022 — Issue date ※ 17 August 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

WEPA38

Progress on Machine Learning for the SNS High Voltage Converter Modulators

klystron, linac, electron, simulation

715

M.I. Radaideh, S.M. Cousineau, D. Lu
ORNL, Oak Ridge, Tennessee, USA
T.J. Britton, K. Rajput, M. Schram, L.S. Vidyaratne
JLab, Newport News, Virginia, USA
G.C. Pappas, J.D. Walden
ORNL RAD, Oak Ridge, Tennessee, USA

The High-Voltage Converter Modulators (HVCM) used to power the klystrons in the Spallation Neutron Source (SNS) linac were selected as one area to explore machine learning due to reliability issues in the past and the availability of large sets of archived waveforms. Progress in the past two years has resulted in generating a significant amount of simulated and measured data for training neural network models such as recurrent neural networks, convolutional neural networks, and variational autoencoders. Applications in anomaly detection, fault classification, and prognostics of capacitor degradation were pursued in collaboration with the Jefferson Laboratory, and early promising results were achieved. This paper will discuss the progress to date and present results from these efforts.

Poster WEPA38 [1.320 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-WEPA38

About •

Received ※ 25 July 2022 — Revised ※ 08 August 2022 — Accepted ※ 11 August 2022 — Issue date ※ 03 October 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

WEPA40

The L-CAPE Project at FNAL

controls, operation, linac, alignment

719

M. Jain, V.C. Amatya, G.U. Panapitiya, J.F. Strube
PNNL, Richland, Washington, USA
B.F. Harrison, K.J. Hazelwood, W. Pellico, B.A. Schupbach, K. Seiya, J.M. St. John
Fermilab, Batavia, Illinois, USA

The controls system at FNAL records data asynchronously from several thousand Linac devices at their respective cadences, ranging from 15Hz down to once per minute. In case of downtimes, current operations are mostly reactive, investigating the cause of an outage and labeling it after the fact. However, as one of the most upstream systems at the FNAL accelerator complex, the Linac’s foreknowledge of an impending downtime as well as its duration could prompt downstream systems to go into standby, potentially leading to energy savings. The goals of the Linac Condition Anomaly Prediction of Emergence (L-CAPE) project that started in late 2020 are (1) to apply data-analytic methods to improve the information that is available to operators in the control room, and (2) to use machine learning to automate the labeling of outage types as they occur and discover patterns in the data that could lead to the prediction of outages. We present an overview of the challenges in dealing with time-series data from 2000+ devices, our approach to developing an ML-based automated outage labeling system, and the status of augmenting operations by identifying the most likely devices predicting an outage.

Poster WEPA40 [1.870 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-WEPA40

About •

Received ※ 03 August 2022 — Revised ※ 12 August 2022 — Accepted ※ 17 August 2022 — Issue date ※ 31 August 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

THXD2

6D Phase Space Diagnostics Based on Adaptive Tuning of the Latent Space of Encoder-Decoder Convolutional Neural Networks

controls, solenoid, feedback, electron

837

A. Scheinker
LANL, Los Alamos, New Mexico, USA

We present a general approach to 6D phase space diagnostics for charged particle beams based on adaptively tuning the low-dimensional latent space of generative encoder-decoder convolutional neural networks (CNN). Our approach first trains the CNN based on supervised learning to learn the correlations and physics constrains within a given accelerator system. The input of the CNN is a high dimensional collection of 2D phase space projections of the beam at the accelerator entrance together with a vector of accelerator parameters such as magnet and RF settings. The inputs are squeezed down to a low-dimensional latent space from which we generate the output in the form of projections of the beam’s 6D phase space at various accelerator locations. After training the CNN is applied in an unsupervised adaptive manner by comparing a subset of the output predictions to available measurements with the error guiding feedback directly in the low-dimensional latent space. We show that our approach is robust to unseen time-variation of the input beam and accelerator parameters and a study of the robustness of the method to go beyond the span of the training data.

Slides THXD2 [19.086 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-THXD2

About •

Received ※ 18 July 2022 — Revised ※ 05 August 2022 — Accepted ※ 08 August 2022 — Issue date ※ 09 August 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

FRXE1

Bayesian Algorithms for Practical Accelerator Control and Adaptive Machine Learning for Time-Varying Systems

controls, feedback, experiment, electron

921

A. Scheinker
LANL, Los Alamos, New Mexico, USA
R.J. Roussel
SLAC, Menlo Park, California, USA

Particle accelerators are complicated machines with thousands of coupled time varying components. The electromagnetic fields of accelerator devices such as magnets and RF cavities drift and are uncertain due to external disturbances, vibrations, temperature changes, and hysteresis. Accelerated charged particle beams are complex objects with 6D phase space dynamics governed by collective effects such as space charge forces, coherent synchrotron radiation, and whose initial phase space distributions change in unexpected and difficult to measure ways. This two-part tutorial presents recent developments in Bayesian methods and adaptive machine learning (ML) techniques for accelerators. Part 1: We introduce Bayesian control algorithms, and we describe how these algorithms can be customized to solve practical accelerator specific problems, including online characterization and optimization. Part 2: We give an overview of adaptive ML (AML) combining adaptive model-independent feedback within physics-informed ML architectures to make ML tools robust to time-variation (distribution shift) and to enable their use further beyond the span of the training data without relying on re-training.

Slides FRXE1 [34.283 MB]

DOI •

reference for this paper ※ doi:10.18429/JACoW-NAPAC2022-FRXE1

About •

Received ※ 08 August 2022 — Revised ※ 10 August 2022 — Accepted ※ 12 August 2022 — Issue date ※ 27 September 2022

Cite •

reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)