Using Machine Learning Tools to Predict Accelerator Failure
M. Reščič, R. Seviour
University of Huddersfield, Huddersfield, United Kingdom
W. Blokland
ORNL, Oak Ridge, Tennessee, USA
Modern particle accelerator facilities are continuously introducing improved beam diagnostics, data acquisition, storage and analysis capabilities. Although the increased volume of data this generates makes manual analyse and understand the acquired data very difficult. In this paper we propose the use of machine learning to better understanding of beam failures. The proposed methods allow for both precognitive failure prediction and failure classification, determined using existing beam diagnostics infrastructure. Where we present the concept of tuning classifier parameters and pulse properties to refine datasets. As a demonstrator we apply our machine learning algorithm to analysis the vast data generated by the Oakridge Spallation Neutron Source (SNS) Differential Beam Current Monitoring (DBCM) diagnostics system. We show that analysis of the SNS DBCM data using machine learning, particle accelerator failure can be identified prior to the actual machine failure with 92% accuracy. Importantly, our research shows that emergent behavior regarding machine failure is encoded in the beam pulses prior to failure actually occurring.
Right click on video for Picture-in-Picture mode or Full screen display.
Feasibility Study of a Non-Rad Camera for the SNS* Ring Injection Dump Imaging System
W. Blokland, A. Rakhman
ORNL, Oak Ridge, Tennessee, USA
Funding:* This manuscript has been authored by UT-Battelle, LLC, under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The Proton Power Upgrade (PPU) increases the the Spallation Neutron Source (SNS) accelerator power from 1.4 MW to 2.8 MW and calls for a modification of the Ring Injection Dump beam line. The charge exchange injection technique to accumulate proton beam in the SNS ring results in multiple beam spots on the ring injection dump window. To properly setup the new injection beam line, the size and locations of the beam spots must be measured. We plan to use a camera to look at a fluorescent coating made of Chromium Oxide doped Aluminum Oxide. To simplify the optical path, we want to place the camera in the tunnel. While radhard cameras are available, they typically are more expensive and have worse performance. To study the feasibility of non-radhard cameras, we measured the radiation in the tunnel in unshielded and shielded locations. We compare the radiation measurements with results from a CERN HiRadMat study and tested the cameras during full power beam to show that the non-radhard camera is an option for the Ring Injection Dump Imaging System.
