BOOSTR: A Dataset for Accelerator Control Systems

TL;DR

BOOSTR is a comprehensive, publicly available dataset capturing cycle-by-cycle device readings from Fermilab's Booster accelerator, aimed at advancing AI-driven control and anomaly detection methods.

Contribution

This paper introduces BOOSTR, one of the first detailed datasets of accelerator device parameters for AI research in control systems.

Findings

Provides a detailed time series dataset of 55 device parameters.

Enables development of AI methods for accelerator control and anomaly detection.

Facilitates research in reinforcement learning for accelerator optimization.

Abstract

The Booster Operation Optimization Sequential Time-series for Regression (BOOSTR) dataset was created to provide a cycle-by-cycle time series of readings and settings from instruments and controllable devices of the Booster, Fermilab's Rapid-Cycling Synchrotron (RCS) operating at 15~Hz. BOOSTR provides a time series from 55 device readings and settings that pertain most directly to the high-precision regulation of the Booster's gradient magnet power supply (GMPS). To our knowledge, this is one of the first well-documented datasets of accelerator device parameters made publicly available. We are releasing it in the hopes that it can be used to demonstrate aspects of artificial intelligence for advanced control systems, such as reinforcement learning and autonomous anomaly detection.