BOOSTR: A Dataset for Accelerator Control Systems

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 controllable devices the Booster, Fermilab’s Rapid-Cycling Synchrotron (RCS) operating at 15 Hz. BOOSTR provides 55 device that pertain most directly high-precision regulation Booster’s gradient magnet power supply (GMPS). To our knowledge, this is one first well-documented datasets accelerator parameters made publicly available. We are releasing it in hopes can be used demonstrate aspects artificial intelligence advanced control systems, such as reinforcement learning autonomous anomaly detection.