Tuning Particle Accelerators with Safety Constraints using Bayesian Optimization

TL;DR

This paper introduces a safe Bayesian optimization method with step-size constraints for tuning particle accelerators, effectively handling safety-critical constraints and demonstrating promising results on real facilities.

Contribution

It presents a novel step-size limited safe Bayesian optimization approach and evaluates it on two large-scale particle accelerators, addressing safety constraints during tuning.

Findings

Successful tuning of 16 parameters with 224 constraints

Effective handling of safety-critical constraints during optimization

Promising experimental results on SwissFEL and HIPA facilities

Abstract

Tuning machine parameters of particle accelerators is a repetitive and time-consuming task that is challenging to automate. While many off-the-shelf optimization algorithms are available, in practice their use is limited because most methods do not account for safety-critical constraints in each iteration, such as loss signals or step-size limitations. One notable exception is safe Bayesian optimization, which is a data-driven tuning approach for global optimization with noisy feedback. We propose and evaluate a step-size limited variant of safe Bayesian optimization on two research facilities of the Paul Scherrer Institut (PSI): a) the Swiss Free Electron Laser (SwissFEL) and b) the High-Intensity Proton Accelerator (HIPA). We report promising experimental results on both machines, tuning up to 16 parameters subject to 224 constraints.