Health Assessment and Performance Monitoring of Large Machine Diagnostics
Explore the source record for details and available documents.
Engineering topics
Publications and source records attributed to Howard, Marylesa M..
Explore the source record for details and available documents.
Since 1951, the Nevada National Security Site has ensured the security of the US and its allies by supporting the stewardship of the nuclear deterrent, providing emergency response capability and training, and contributing key nonproliferation and arms control initiatives.
This presentation will be given at the LANL P1 Physics & Theory Colloquium on July 7, 2022.
Presentation to be given at the CCS-6 Statistical Seminar Series. Regularly maintained and operated diagnostic machines are a backbone of data collection and are a vital component of the Stockpile Stewardship Program’s efforts to better understand nuclear physics. The Cygnus X-ray machine, stationed at the Nevada National Security Site’s U1a underground facility, is one such diagnostic that provides a radiographic capability for the subcritical experiment program. Component failures within Cygnus can result in catastrophic downtime for the diagnostic, affecting performance, schedules, and cost. However, over the years various measurements have been collected on the two Cygnus axes, including voltage and current measurements at different locations, which we believe have predictive power to indicate machine health. We will share preliminary insight into this data and the machine learning approaches we are taking to assess Cygnus’ health, observe declining performance, and predict failures.