Accelerator beam phase space tomography using machine learning to account for variations in beamline components

TL;DR

This paper presents a machine learning-based method for reconstructing the 4D transverse phase space of an accelerator beam, effectively accounting for unknown magnet errors and enabling rapid analysis with experimental validation.

Contribution

It introduces a novel machine learning approach for phase space tomography that simultaneously estimates beam distribution and magnet errors, improving accuracy and speed.

Findings

Successful reconstruction of beam phase space from experimental data

Effective estimation of magnet errors during reconstruction

Rapid and accurate phase space analysis demonstrated at CLARA facility

Abstract

We describe a technique for reconstruction of the four-dimensional transverse phase space of a beam in an accelerator beamline, taking into account the presence of unknown errors on the strengths of magnets used in the data collection. Use of machine learning allows rapid reconstruction of the phase-space distribution while at the same time providing estimates of the magnet errors. The technique is demonstrated using experimental data from CLARA, an accelerator test facility at Daresbury Laboratory.