Superconducting radio-frequency cavity fault classification using machine learning at Jefferson Laboratory

Tennant, Chris; Carpenter, Adam F.; Powers, T.; Shabalina, Anna; Vidyaratne, Lasitha; Iftekharuddin, Khan M.

doi:10.1103/physrevaccelbeams.23.114601

Cited by 50 publications

(25 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The operation of large-scale scientific user facilities such as the European XFEL [1] is very challenging, as it is necessary to meet the specifications of various user experiments [2] and to be capable of switching the machine status rapidly. Machine learning, especially deep learning, has provided powerful tools for accelerator physicists to build fast-prediction surrogate models [3][4][5] and to extract essential information [6][7][8] from large amounts of data in recent years. These machine-learning models can be extremely useful for building virtual accelerators, which are capable of making fast predictions of the behavior of beams [9], assisting accelerator tuning by virtually bringing destructive diagnostics online [4], providing an initial guess of input parameters for model-independent adaptive feedback control algorithms [10,11], and driving modelbased feedback control algorithms [12].…”

Section: Introductionmentioning

confidence: 99%

High-Fidelity Prediction of Megapixel Longitudinal Phase-Space Images of Electron Beams Using Encoder-Decoder Neural Networks

Zhu

Chen²,

Brinker³

et al. 2021

Phys. Rev. Applied

View full text Add to dashboard Cite

Modeling of large-scale research facilities is extremely challenging due to complex physical processes and engineering problems. Here, we adopt a data-driven approach to model the longitudinal phase-spacediagnostic beamline at the photoinector of the European XFEL with an encoder-decoder neural-network model. A deep convolutional neural network (decoder) is used to build images, measured on the screen, from a small feature map generated by another neural network (encoder). We demonstrate that the model, trained only with experimental data, can make high-fidelity predictions of megapixel images for the longitudinal phase-space measurement without any prior knowledge of photoinjectors or electron beams. The prediction significantly outperforms existing methods. We also show the scalability and interpretability of the model by sharing the same decoder with more than one encoder, used for different setups of the photoinjector, and propose a pragmatic way to model a facility with various diagnostics and working points. This opens the door to a way of accurately modeling a photoinjector using neural networks and experimental data. The approach can possibly be extended to the whole accelerator and even other types of scientific facility.

show abstract

Section: Introductionmentioning

confidence: 99%

High-Fidelity Prediction of Megapixel Longitudinal Phase-Space Images of Electron Beams Using Encoder-Decoder Neural Networks

Zhu

Chen²,

Brinker³

et al. 2021

Phys. Rev. Applied

View full text Add to dashboard Cite

show abstract

“…We report quantitative performance figures for each DL system, and a performance comparison with the currently deployed ML pipeline ( Tennant et al, 2020 ). Following the typical DL workflow, the training and testing of the DL models are carried out using a data split of 60% (3,616 events) for training, 20% (1,205 events) for validation, and 20% (1,206 events) for testing (stratified).…”

Section: Resultsmentioning

confidence: 99%

“…The ML pipeline implements a feature extraction scheme based on autoregressive analysis of each signal to obtain 192 features representing each event ( Tennant et al, 2020 ). For the purpose of comparison, we implement this feature extraction scheme for each RF signal parallelized across a six-core CPU using six workers.…”

Section: Resultsmentioning

confidence: 99%

“…Parameters such as sampling frequency and the ratio of pre-fault to post-fault data in the captured waveforms are established by subject matter experts (SMEs) such that the data contain sufficient information to determine the nature of an RF failure event ( Tennant et al, 2020 ). Supplementary Appendix Table SA1 in Appendix A describes the 17 recorded RF signals.…”

Section: Introductionmentioning

confidence: 99%

“…A state-of-the-art ML based fault classifier model is currently deployed for use in CEBAF ( Tennant et al, 2020 ). Many classical machine learning (ML) methods have been developed and utilized for time-series analysis in a variety of domains such as economics ( Beveridge and Nelson, 1981 ; Caiado et al, 2006 ), the social sciences ( Meidinger, 1980 ; Shumway et al, 2000 ), and healthcare ( Chua et al, 2010 ; Shoeb and Guttag, 2010 ).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep Learning Based Superconducting Radio-Frequency Cavity Fault Classification at Jefferson Laboratory

Vidyaratne

Carpenter

Powers

et al. 2022

Front. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

This work investigates the efficacy of deep learning (DL) for classifying C100 superconducting radio-frequency (SRF) cavity faults in the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab. CEBAF is a large, high-power continuous wave recirculating linac that utilizes 418 SRF cavities to accelerate electrons up to 12 GeV. Recent upgrades to CEBAF include installation of 11 new cryomodules (88 cavities) equipped with a low-level RF system that records RF time-series data from each cavity at the onset of an RF failure. Typically, subject matter experts (SME) analyze this data to determine the fault type and identify the cavity of origin. This information is subsequently utilized to identify failure trends and to implement corrective measures on the offending cavity. Manual inspection of large-scale, time-series data, generated by frequent system failures is tedious and time consuming, and thereby motivates the use of machine learning (ML) to automate the task. This study extends work on a previously developed system based on traditional ML methods (Tennant and Carpenter and Powers and Shabalina Solopova and Vidyaratne and Iftekharuddin, Phys. Rev. Accel. Beams, 2020, 23, 114601), and investigates the effectiveness of deep learning approaches. The transition to a DL model is driven by the goal of developing a system with sufficiently fast inference that it could be used to predict a fault event and take actionable information before the onset (on the order of a few hundred milliseconds). Because features are learned, rather than explicitly computed, DL offers a potential advantage over traditional ML. Specifically, two seminal DL architecture types are explored: deep recurrent neural networks (RNN) and deep convolutional neural networks (CNN). We provide a detailed analysis on the performance of individual models using an RF waveform dataset built from past operational runs of CEBAF. In particular, the performance of RNN models incorporating long short-term memory (LSTM) are analyzed along with the CNN performance. Furthermore, comparing these DL models with a state-of-the-art fault ML model shows that DL architectures obtain similar performance for cavity identification, do not perform quite as well for fault classification, but provide an advantage in inference speed.

show abstract

Machine Learning for Beam Controls

Zhang

Simrock

2023

Particle Acceleration and Detection

View full text Add to dashboard Cite

Superconducting radio-frequency cavity fault classification using machine learning at Jefferson Laboratory

Cited by 50 publications

References 12 publications

High-Fidelity Prediction of Megapixel Longitudinal Phase-Space Images of Electron Beams Using Encoder-Decoder Neural Networks

High-Fidelity Prediction of Megapixel Longitudinal Phase-Space Images of Electron Beams Using Encoder-Decoder Neural Networks

Deep Learning Based Superconducting Radio-Frequency Cavity Fault Classification at Jefferson Laboratory

Machine Learning for Beam Controls

Contact Info

Product

Resources

About