Speaker
Guimei Wang
(Brookhaven National Laboratory)
Description
In NSLS-II storage ring, the 10 kHz orbit data are always available and can be collected at any time interval. At this moment, they are being collected every 10 minutes to review the machine stability status or investigate orbit related issues impacting user satisfaction. To improve the machine performance, we are studying the machine learning techniques with which we can detect any orbit stability issue as early as possible. As one of the strong candidates, we are testing models constructed by the long short-term memory (LSTM) autoencoder from the orbit data. In this paper, we present the optimized LSTM autoencoder parameters and the test results.
Funding Agency
DOE under contract No. DE-SC0012704
| In which format do you inted to submit your paper? | LaTeX |
|---|---|
| Preprint marking on your proceeding paper | I wish my paper to be marked as preprint. |
Author
Jinhyuk Choi
(Brookhaven National Laboratory)