An accurate physics simulation model is key to accelerator operation because all beam control and optimization algorithms require good understanding of the accelerator and its elements. For the AGS Booster, discrepancies between the real physical system and online simulation model have been a long-standing issue. Due to the lack of a reliable model, the current practice of beam control relies mainly on empirical tuning by experienced operators, which may be inefficient or sub-optimal. In this work, we investigate two main factors that can cause discrepancies between simulation and reality in the AGS Booster: magnet misalignments and magnet transfer functions. We developed a orbit response measurement script that collects real machine data in the Booster for model calibration. By matching simulated data with real data, we can develop a more accurate simulation model for future polarization optimizations, and build the foundation for a fully functional digital-twin.

The RHIC heavy ion program relies on a series of RF bunch merge gymnastics to combine individual source pulses into bunches of suitable intensity. Intensity and emittance preservation during these gymnastics require careful setup of the voltages and phases of RF cavities operating at several different harmonic numbers. The optimum setting tends to drift over time, degrading performance and requiring operator attention to correct. We describe a reinforcement learning approach to learning and maintaining an optimum configuration, accounting for the relevant RF parameters and external perturbations (e.g., a changing main dipole field) using a physics-based simulator at Brookhaven Alternating Gradient Synchrotron (AGS).

An accurate physics simulation model is key to accelerator operation because all beam control and optimization algorithms require good understanding of the accelerator and its elements. For the AGS Booster, discrepancies between the real physical system and online simulation model have been a long-standing issue. Due to the lack of a reliable model, the current practice of beam control relies mainly on empirical tuning by experienced operators, which may be inefficient or sub-optimal. In this work, we investigate two main factors that can cause discrepancies between simulation and reality in the AGS Booster: magnet misalignments and magnet transfer functions. We developed a orbit response measurement script that collects real machine data in the Booster for model calibration. By matching simulated data with real data, we can develop a more accurate simulation model for future polarization optimizations, and build the foundation for a fully functional digital-twin.