IPAC2021 - List of Authors (Peters, C.C.)

Paper	Title	Page
TUPAB328	Machine Learning for Time Series Prediction of an Accelerator Beam to Recognize Equipment Malfunction	2272
	C.C. Peters ORNL RAD, Oak Ridge, Tennessee, USA W. Blokland, D.L. Brown, F. Liu, C.D. Long, D. Lu, P. Ramuhalli, D.E. Womble, J. Zhang, A.P. Zhukov ORNL, Oak Ridge, Tennessee, USA
	Funding: ORNL is managed by UT-Battelle, LLC, under contract DE-AC05- 00OR22725 for the U.S. Department of Energy. The Spallation Neutron Source (SNS) is an accelerator based pulsed neutron source based on a 1 GeV pulsed proton Superconducting Radio Frequency (SRF) linear accelerator (linac). Since beginning high power beam operation in 2006 correlations have been found linking abrupt beam loss events to SRF cavity instabilities. With the planned upgrades to double the beam power we expect increased rates of degradation and the importance of minimizing these beam loss events will become ever more important. To further limit degradation, we are developing machine learning approaches to monitor the beam and to detect, predict and prevent beam loss events. Initial research has shown that precursors to beam loss events are detectable. The initial steps are to use ML-based classification to recognize anomalies and to use Long Short-Term Memory (LSTM) autoencoders to predict beam loss. In this paper, we describe recent progress in applying machine learning for recognizing anomalies and predicting beam loss and present initial results of our research using acquired data from different diagnostics and the Machine Protection System (MPS).
DOI •	reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB328
About •	paper received ※ 23 May 2021 paper accepted ※ 28 May 2021 issue date ※ 15 August 2021
Export •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

TUPAB329	Pattern Based Parameter Setup of the SNS Linac	2276
	C.C. Peters ORNL RAD, Oak Ridge, Tennessee, USA A.P. Shishlo ORNL, Oak Ridge, Tennessee, USA
	Funding: ORNL is managed by UT-Battelle, LLC, under contract DE-AC05- 00OR22725 for the U.S. Department of Energy. Theoretical and practical aspects of beam tuning procedures used for the SNS linac are discussed. The SNS linac includes two sections of beam acceleration. Acceleration in the first section up to 185.5 MeV is done with a room temperature copper linac which consists of both Drift Tube Linac (DTL) and Coupled Cavity Linac (CCL) Radio Frequency (RF) cavities. The second section consists of 81 Superconducting RF (SRF) cavities which accelerate the beam to the final beam energy of 1 GeV. The linac is currently capable of delivering an average beam power output of 1.44 MW with typical yearly operating hours of around 4500 hours. Due to the high power output and high availability of the linac, activation of accelerator equipment is a significant concern. The linac tuning process consists of three stages: model based setup of amplitudes and phases of the RF cavities, empirical beam loss reduction, and then documentation of the final amplitudes and phases of RF cavities after the empirical tuning. The final step is needed to ensure fast recovery from an SRF cavity failure. This paper discusses models, algorithms, diagnostic tools, software, and practices that are used for these stages.
DOI •	reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB329
About •	paper received ※ 22 May 2021 paper accepted ※ 28 May 2021 issue date ※ 25 August 2021
Export •	reference for this paper using ※ BibTeX, ※ LaTeX, ※ Text/Word, ※ RIS, ※ EndNote (xml)

Paper

Title

Page

Machine Learning for Time Series Prediction of an Accelerator Beam to Recognize Equipment Malfunction

2272

C.C. Peters
ORNL RAD, Oak Ridge, Tennessee, USA
W. Blokland, D.L. Brown, F. Liu, C.D. Long, D. Lu, P. Ramuhalli, D.E. Womble, J. Zhang, A.P. Zhukov
ORNL, Oak Ridge, Tennessee, USA

Funding: ORNL is managed by UT-Battelle, LLC, under contract DE-AC05- 00OR22725 for the U.S. Department of Energy.
The Spallation Neutron Source (SNS) is an accelerator based pulsed neutron source based on a 1 GeV pulsed proton Superconducting Radio Frequency (SRF) linear accelerator (linac). Since beginning high power beam operation in 2006 correlations have been found linking abrupt beam loss events to SRF cavity instabilities. With the planned upgrades to double the beam power we expect increased rates of degradation and the importance of minimizing these beam loss events will become ever more important. To further limit degradation, we are developing machine learning approaches to monitor the beam and to detect, predict and prevent beam loss events. Initial research has shown that precursors to beam loss events are detectable. The initial steps are to use ML-based classification to recognize anomalies and to use Long Short-Term Memory (LSTM) autoencoders to predict beam loss. In this paper, we describe recent progress in applying machine learning for recognizing anomalies and predicting beam loss and present initial results of our research using acquired data from different diagnostics and the Machine Protection System (MPS).

DOI •

reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB328

About •

paper received ※ 23 May 2021 paper accepted ※ 28 May 2021 issue date ※ 15 August 2021

Export •

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

TUPAB329

Pattern Based Parameter Setup of the SNS Linac

2276

C.C. Peters
ORNL RAD, Oak Ridge, Tennessee, USA
A.P. Shishlo
ORNL, Oak Ridge, Tennessee, USA

Funding: ORNL is managed by UT-Battelle, LLC, under contract DE-AC05- 00OR22725 for the U.S. Department of Energy.
Theoretical and practical aspects of beam tuning procedures used for the SNS linac are discussed. The SNS linac includes two sections of beam acceleration. Acceleration in the first section up to 185.5 MeV is done with a room temperature copper linac which consists of both Drift Tube Linac (DTL) and Coupled Cavity Linac (CCL) Radio Frequency (RF) cavities. The second section consists of 81 Superconducting RF (SRF) cavities which accelerate the beam to the final beam energy of 1 GeV. The linac is currently capable of delivering an average beam power output of 1.44 MW with typical yearly operating hours of around 4500 hours. Due to the high power output and high availability of the linac, activation of accelerator equipment is a significant concern. The linac tuning process consists of three stages: model based setup of amplitudes and phases of the RF cavities, empirical beam loss reduction, and then documentation of the final amplitudes and phases of RF cavities after the empirical tuning. The final step is needed to ensure fast recovery from an SRF cavity failure. This paper discusses models, algorithms, diagnostic tools, software, and practices that are used for these stages.

DOI •

reference for this paper ※ https://doi.org/10.18429/JACoW-IPAC2021-TUPAB329

About •

paper received ※ 22 May 2021 paper accepted ※ 28 May 2021 issue date ※ 25 August 2021

Export •

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