TUBITAK Destekli Projelerden Çıkan Yayınlar Topluluğu
Published January 1, 2023 | Version v1
Journal article Open

Spatio-Temporal Anomaly Detection with Graph Networks for Data Quality Monitoring of the Hadron Calorimeter

Creators

  • 1. Univ Agder, Ctr Artificial Intelligence Res, Dept Informat & Commun Technol, N-4879 Grimstad, Norway
  • 2. Univ Maryland, Dept Phys, College Pk, MD 20742 USA
  • 3. Brown Univ, Dept Phys, Providence, RI 02912 USA
  • 4. Univ Rochester, Dept Phys & Astron, Rochester, NY 14627 USA
  • 5. Baylor Univ, Dept Phys, Waco, TX 76706 USA
  • 6. Univ Calif Riverside, Dept Phys & Astron, Riverside, CA 92521 USA
  • 7. Riga Tech Univ, Inst Particle Phys & Accelerator Technol, LV-1048 Riga, Latvia
  • 8. Bari Univ, Dept Phys, I-70121 Bari, Italy
  • 9. Univ Ghent, Dept Phys & Astron, B-9000 Ghent, Belgium
  • 10. Univ Alabama, Dept Phys & Astron, Tuscaloosa, AL 35487 USA
  • 11. Texas A&M Univ, Dept Phys & Astron, College Stn, TX 77843 USA
  • 12. Univ Oviedo, Inst Univ Ciencias & Tecnol Espaciales Asturias, Oviedo 33004, Spain
  • 13. Fermilab Natl Accelerator Lab, Batavia, IL 60510 USA

Description

The Compact Muon Solenoid (CMS) experiment is a general-purpose detector for high-energy collision at the Large Hadron Collider (LHC) at CERN. It employs an online data quality monitoring (DQM) system to promptly spot and diagnose particle data acquisition problems to avoid data quality loss. In this study, we present a semi-supervised spatio-temporal anomaly detection (AD) monitoring system for the physics particle reading channels of the Hadron Calorimeter (HCAL) of the CMS using three-dimensional digi-occupancy map data of the DQM. We propose the GraphSTAD system, which employs convolutional and graph neural networks to learn local spatial characteristics induced by particles traversing the detector and the global behavior owing to shared backend circuit connections and housing boxes of the channels, respectively. Recurrent neural networks capture the temporal evolution of the extracted spatial features. We validate the accuracy of the proposed AD system in capturing diverse channel fault types using the LHC collision data sets. The GraphSTAD system achieves production-level accuracy and is being integrated into the CMS core production system for real-time monitoring of the HCAL. We provide a quantitative performance comparison with alternative benchmark models to demonstrate the promising leverage of the presented system.

Files

bib-a88c27e8-1db7-4630-9d92-36ed7a29a069.txt

Files (319 Bytes)

Name Size Download all
md5:b3646c4bd1f9cce8ef28f84e18634bda
319 Bytes Preview Download