Speaker
Description
Due to their large size, and number of devices, particle accelerators are rich sources of data, providing real opportunities for developing new equipment monitoring and automation tools using data analysis and machine learning.
The ready availability of low-cost computers and microcontrollers, such as Raspberry Pi and ESP32 devices, could enable a flexible data acquisition system for short-term applications that do not require, or cannot justify, the development and installation of permanent acquisition infrastructure.
This paper presents the initial work from a pilot project to develop such a system, and assesses its use for several applications, including speculative investigations of environmental conditions, such as temperature and high-energy hadron flux, as well as assessing the feasibility of detecting arc faults in power converters, and gathering datasets for training machine learning models.
Key considerations for the implementation of this system are also discussed, including concerns around network security, data quality, data availability, hardware configurations, deployment conditions, and avoiding control system dependency, with initial recommendations given for each.
| In which format do you inted to submit your paper? | LaTeX |
|---|