| Internal Note | |
| Report number | CERN-ACC-NOTE-2020-0001 |
| Title | MD 4510 : Working point exploration for use in lifetime optimization by machine learning |
| Author(s) | Coyle, Loic Thomas Davies (EPFL - Ecole Polytechnique Federale Lausanne (CH)) ; Pieloni, Tatiana (EPFL - Ecole Polytechnique Federale Lausanne (CH)) ; Rivkin, Lenny (Paul Scherrer Institut (CH)) ; Salvachua Ferrando, Belen Maria (CERN) |
| Corporate author(s) | CERN. Geneva. ATS Department |
| Publication | 2019 |
| Imprint | 12 Dec 2019 |
| Number of pages | 16 |
| Subject category | Accelerators and Storage Rings |
| Accelerator/Facility, Experiment | CERN LHC |
| Keywords | Optimization ; Parameter Exploration ; Machine Learning |
| Abstract | Supervised learning based Machine Learning models are fundamentally reliant on the data on which they are trained. Previous to this MD, the data available although plentiful, was lacking variety as the working point is rarely changed. We have a large amount of data, however, many of the beam and machine parameters are left unchanged during operation, and from fill to fill. Therefore, previous work done with this data at injection energy show promising results but restricted pre- diction power due to this lack of exploration. This MD will serve to generate a wider data training sample in a more exotic configurations, at injection energy. The goal is to explore the possibility to optimize the beam lifetime of the LHC by the use of machine learning algorithm. Previous Machine Learning studies have predicted some tentative trends which were confirmed with this MD. |
| Submitted by | [email protected] |
Rekord stworzony 2020-01-07, ostatnia modyfikacja 2020-01-21
