17–22 May 2026
C.I.D
Europe/Zurich timezone

Optimization of Fermilab Booster using hybrid Bayesian/RL framework

MOP6312
18 May 2026, 16:00
2h
C.I.D

C.I.D

Deauville, France
Poster Presentation MC6.D13: Instrumentation: Artificial Intelligence Poster session

Speaker

Nikita Kuklev (Fermi National Accelerator Laboratory)

Description

PIPII project will raise Fermilab Booster intensity and ramp rate. Beam losses will limit average power and are hard to simulate. Presently, Booster uses operator-guided empirical tuning - a challenging task due to high dimensionality, multiple objectives, critical safety constraints, and drifts. We developed a synergistic suite of Bayesian optimization (BO) and reinforcement learning (RL) tools to optimize and stabilize beam losses. First, active learning was used to build a rough model. Data was collected parasitically using two novel safety constraint types – nonlinear input space restrictions (based on optics model), and uncertainty constraints (to stop bad steps/beam aborts). We then applied online multi-objective BO with scalarized objectives and fast fitting to improve/rebalance losses, increasing safety margins by 25%. Using BO model as a safety veto, we tried several on/off-policy RL agents for long term stabilization; SAC had best medium-term performance. Adding contextual information from linac further improved performance; eventually we integrated knobs like linac phase into the parameter space. Long term testing is ongoing to enable operational use.

In which format do you inted to submit your paper? LaTeX

Author

Nikita Kuklev (Fermi National Accelerator Laboratory)

Co-authors

Jeffrey Eldred (Fermi National Accelerator Laboratory) Michael Balcewicz (Fermi National Accelerator Laboratory)

Presentation materials

There are no materials yet.