DataSheet1_Deep Learning Based Superconducting Radio-Frequency Cavity Fault Classification at Jefferson Laboratory.docx

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.

本研究探讨了深度学习（deep learning, DL）在杰斐逊实验室（Jefferson Lab）的连续电子束加速器装置（Continuous Electron Beam Accelerator Facility, CEBAF）中对C100超导射频（superconducting radio-frequency, SRF）腔故障进行分类的有效性。CEBAF是一台大型高功率连续波循环直线加速器，搭载418个SRF腔，可将电子加速至12 GeV。CEBAF近期的升级项目包括安装11台全新低温模块（cryomodules，含88个腔体），其配套的低电平射频系统可在射频故障触发时，记录每个腔体的射频时间序列数据。通常情况下，领域专家（subject matter experts, SME）会对该类数据进行分析，以确定故障类型并定位故障起源腔体。后续可利用该信息识别故障趋势，并对出现故障的腔体采取纠正措施。然而，针对频繁系统故障产生的大规模时间序列数据开展人工检查，不仅繁琐且耗时，这推动了借助机器学习（machine learning, ML）自动化该任务的研究。本研究拓展了此前基于传统机器学习方法开发的系统（Tennant与Carpenter、Powers、Shabalina Solopova、Vidyaratne及Iftekharuddin, Phys. Rev. Accel. Beams, 2020, 23, 114601），并探究了深度学习方法的应用效果。转向深度学习模型的核心动机，是开发一款推理速度足够快的系统，使其能够在故障发生前（约数百毫秒级别）预测故障事件并提供可操作的应对信息。相较于需手动提取特征的传统机器学习，深度学习通过自主学习特征，具备潜在的性能优势。本研究探索了两类经典深度学习架构：深度循环神经网络（recurrent neural networks, RNN）与深度卷积神经网络（convolutional neural networks, CNN）。我们基于CEBAF过往运行数据构建的射频波形数据集，对各模型的性能展开了详细分析。具体而言，本文分析了结合长短期记忆网络（long short-term memory, LSTM）的循环神经网络模型的性能，同时也分析了卷积神经网络的性能。此外，将这些深度学习模型与当前最先进的机器学习故障检测模型进行对比后发现：深度学习架构在腔体定位任务上性能相当，但在故障分类任务上表现稍逊，不过在推理速度上具备显著优势。