Is this project an undergraduate, graduate, or faculty project?
Undergraduate
Project Type
group
Campus
Daytona Beach
Authors' Class Standing
Antonio Cascio, Senior John Scheer, Senior Ioannis Paraschos, Junior Lynette Ramirez, Junior Laina Tallman, Junior Lexi Azoulay, Junior Virginia Sacchi, Junior Ahnika Ghee, Sophomore
Lead Presenter's Name
Antonio Cascio
Lead Presenter's College
DB College of Arts and Sciences
Faculty Mentor Name
Dr. Mihhail Berezovski
Abstract
The Nevada National Security Site (NNSS) is a United States Department of Energy complex located about 65 miles north of Las Vegas, Nevada and is home to the Cygnus X-Ray diagnostic machine. They perform hydrodynamic and shockwave physics experiments and use Cygnus to capture radiographic images. Considering that conducting these experiments costs the company money, being able to detect failures of machinery is vital when performing large-scale experiments on the timescale of a few months to even a year. In this report we will be using machine diagnostic data from the Nevada National Security Site’s Cygnus dual-beam X-Ray machine to preemptively determine if part of the machine is in danger of potential failure. To complete this, our team has sorted the 28 signals produced by the Cygnus machines (1 and 2) with 2861 records in total and have extracted quantities of interest such as the maximum and minimum values from the signals using Python scripts. From this, we plotted all these quantities onto a graph based off the pipeline of Cygnus to determine outliers and note where some of the shots may have failed. Our models have been able to correctly classify failures in the provided data.
Did this research project receive funding support (Spark, SURF, Research Abroad, Student Internal Grants, Collaborative, Climbing, or Ignite Grants) from the Office of Undergraduate Research?
No
Predicting Machine Failure in Large Machine Diagnostics
The Nevada National Security Site (NNSS) is a United States Department of Energy complex located about 65 miles north of Las Vegas, Nevada and is home to the Cygnus X-Ray diagnostic machine. They perform hydrodynamic and shockwave physics experiments and use Cygnus to capture radiographic images. Considering that conducting these experiments costs the company money, being able to detect failures of machinery is vital when performing large-scale experiments on the timescale of a few months to even a year. In this report we will be using machine diagnostic data from the Nevada National Security Site’s Cygnus dual-beam X-Ray machine to preemptively determine if part of the machine is in danger of potential failure. To complete this, our team has sorted the 28 signals produced by the Cygnus machines (1 and 2) with 2861 records in total and have extracted quantities of interest such as the maximum and minimum values from the signals using Python scripts. From this, we plotted all these quantities onto a graph based off the pipeline of Cygnus to determine outliers and note where some of the shots may have failed. Our models have been able to correctly classify failures in the provided data.