Journals of Accelerator Conferences Website (JACoW)

JACoW is a publisher in Geneva, Switzerland that publishes the proceedings of accelerator conferences held around the world by an international collaboration of editors.

URLhttps://doi.org/10.18429/JACoW-ICALEPCS2025-TUPD113
TitleUnsupervised Anomaly Detection in ALS EPICS Event Logs
Authors
  • A. Sulc, S. Hunt, T. Hellert
    Lawrence Berkeley National Laboratory
AbstractThis paper introduces an automated fault analysis framework for the Advanced Light Source (ALS) that processes real-time event logs from its EPICS control system. By treating log entries as natural language, we transform them into contextual vector representations using semantic embedding techniques. A sequence-aware neural network, trained on normal operational data, assigns a real-time anomaly score to each event. This method flags deviations from baseline behavior, enabling operators to rapidly identify the critical event sequences that precede complex system failures.
Paperdownload: TUPD113.pdf
CiteBibTeX, LaTeX, Text/Word, RIS, EndNote
Conference20th Int. Conf. Accel. Large Exp. Phys. Control Syst. (ICALEPCS'25)
Series
LocationChicago, IL, USA
Date20-26 Sep 2025
PublisherJACoW Publishing, Geneva, Switzerland
Editorial BoardKent Wootton (ANL), Jeffery Biss (FNAL), Gustavo Bruno (ANL), Jeff Kirchman (ANL), Kelly Jaje (ANL), Denise Skiadopoulos (ANL), SuYin Grass Wang (ANL)
Online ISBN978-3-95450-255-4
Online ISSN2226-0358
Received05 September 2025
Revised24 September 2025
Accepted29 October 2025
Issued25 November 2025
DOI10.18429/JACoW-ICALEPCS2025-TUPD113
Pages763-768