A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation

Yigit A. Yucesan; Willem Blokland; Pradeep Ramuhalli; Alexander Zhukov; Charles Peters; David Brown; Cary Long

doi:10.1016/j.nima.2024.169232

A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation

Yigit A. Yucesan, Willem Blokland, Pradeep Ramuhalli, Alexander Zhukov, Charles Peters, David Brown, Cary Long

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

Particle accelerators are extremely complex systems that are expected to operate on high availability. Predicting impending failures only by utilizing data collected from diagnostic equipment already on board can help operators to avoid installing expensive sensors, unscheduled downtime and associated costs. For this purpose we explore the predictive power of Machine Learning algorithms to detect faulty beams prior to the failure. In this study, we propose a Machine Learning approach to model mapping from a pair of sensors located across the accelerator. While the model is trained to represent normal operation, we evaluate the predictive performance on known faulty beam pulses. We also investigate the model performance on unseen data through k-fold cross-validation. Then we recap the analysis with a neural architecture search and hyperparameter optimization study to fine tune our initial model. This paper will also introduce a sustainable framework that can standardize Machine Learning workflow applied to particle accelerators.

Original language	English
Article number	169232
Journal	Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment
Volume	1063
DOIs	https://doi.org/10.1016/j.nima.2024.169232
State	Published - Jun 2024

Funding

The authors acknowledge the help of the collaborators from Thomas Jefferson National Accelerator Facility, Malachi Schram and Kishansingh Rajput, and Sarah Cousineau at ORNL, for making this grant work possible. This work was supported by the DOE Office of Science, United States under Grant No. DE-SC0009915 (Office of Basic Energy Sciences, Scientific User Facilities program). This research used resources at the Spallation Neutron Source, a DOE Office of Science User Facility operated by the Oak Ridge National Laboratory. This manuscript has been authored by UT-Battelle, LLC, under Contract No. DE-AC05-00OR22725. The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

Keywords

Data-driven fault detection
Neural architecture search
Particle accelerators
Sustainable machine learning framework

Access to Document

10.1016/j.nima.2024.169232

Cite this

Yucesan, Y. A., Blokland, W., Ramuhalli, P., Zhukov, A., Peters, C., Brown, D., & Long, C. (2024). A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation. Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 1063, Article 169232. https://doi.org/10.1016/j.nima.2024.169232

@article{1e95c52b24204f9aadb7add668cf5b01,

title = "A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation",

abstract = "Particle accelerators are extremely complex systems that are expected to operate on high availability. Predicting impending failures only by utilizing data collected from diagnostic equipment already on board can help operators to avoid installing expensive sensors, unscheduled downtime and associated costs. For this purpose we explore the predictive power of Machine Learning algorithms to detect faulty beams prior to the failure. In this study, we propose a Machine Learning approach to model mapping from a pair of sensors located across the accelerator. While the model is trained to represent normal operation, we evaluate the predictive performance on known faulty beam pulses. We also investigate the model performance on unseen data through k-fold cross-validation. Then we recap the analysis with a neural architecture search and hyperparameter optimization study to fine tune our initial model. This paper will also introduce a sustainable framework that can standardize Machine Learning workflow applied to particle accelerators.",

keywords = "Data-driven fault detection, Neural architecture search, Particle accelerators, Sustainable machine learning framework",

author = "Yucesan, \{Yigit A.\} and Willem Blokland and Pradeep Ramuhalli and Alexander Zhukov and Charles Peters and David Brown and Cary Long",

note = "Publisher Copyright: {\textcopyright} 2024",

year = "2024",

month = jun,

doi = "10.1016/j.nima.2024.169232",

language = "English",

volume = "1063",

journal = "Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment",

issn = "0168-9002",

publisher = "Elsevier",

}

Yucesan, YA, Blokland, W , Ramuhalli, P , Zhukov, A, Peters, C, Brown, D & Long, C 2024, 'A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation', Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, vol. 1063, 169232. https://doi.org/10.1016/j.nima.2024.169232

A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation. / Yucesan, Yigit A.; Blokland, Willem ; Ramuhalli, Pradeep et al.
In: Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 1063, 169232, 06.2024.

Research output: Contribution to journal › Article › peer-review

TY - JOUR

T1 - A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation

AU - Yucesan, Yigit A.

AU - Blokland, Willem

AU - Ramuhalli, Pradeep

AU - Zhukov, Alexander

AU - Peters, Charles

AU - Brown, David

AU - Long, Cary

PY - 2024/6

Y1 - 2024/6

N2 - Particle accelerators are extremely complex systems that are expected to operate on high availability. Predicting impending failures only by utilizing data collected from diagnostic equipment already on board can help operators to avoid installing expensive sensors, unscheduled downtime and associated costs. For this purpose we explore the predictive power of Machine Learning algorithms to detect faulty beams prior to the failure. In this study, we propose a Machine Learning approach to model mapping from a pair of sensors located across the accelerator. While the model is trained to represent normal operation, we evaluate the predictive performance on known faulty beam pulses. We also investigate the model performance on unseen data through k-fold cross-validation. Then we recap the analysis with a neural architecture search and hyperparameter optimization study to fine tune our initial model. This paper will also introduce a sustainable framework that can standardize Machine Learning workflow applied to particle accelerators.

AB - Particle accelerators are extremely complex systems that are expected to operate on high availability. Predicting impending failures only by utilizing data collected from diagnostic equipment already on board can help operators to avoid installing expensive sensors, unscheduled downtime and associated costs. For this purpose we explore the predictive power of Machine Learning algorithms to detect faulty beams prior to the failure. In this study, we propose a Machine Learning approach to model mapping from a pair of sensors located across the accelerator. While the model is trained to represent normal operation, we evaluate the predictive performance on known faulty beam pulses. We also investigate the model performance on unseen data through k-fold cross-validation. Then we recap the analysis with a neural architecture search and hyperparameter optimization study to fine tune our initial model. This paper will also introduce a sustainable framework that can standardize Machine Learning workflow applied to particle accelerators.

KW - Data-driven fault detection

KW - Neural architecture search

KW - Particle accelerators

KW - Sustainable machine learning framework

UR - https://www.scopus.com/pages/publications/85189085299

U2 - 10.1016/j.nima.2024.169232

DO - 10.1016/j.nima.2024.169232

M3 - Article

AN - SCOPUS:85189085299

SN - 0168-9002

VL - 1063

JO - Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment

JF - Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment

M1 - 169232

ER -

A machine learning approach for particle accelerator errant beam prediction using spatial phase deviation

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this