30 August 2026 to 3 September 2026
Whistler Conference Centre
Canada/Pacific timezone
Poster ONLY abstracts are still being accepted -- contact IBIC2026@triumf.ca | Registration is Open

Complete 6-dimensional phase space reconstruction based on neural networks and differentiable simulation with standard elements

WERT09
2 Sept 2026, 15:10
30m
Rainbow Theatre (Whistler Conference Centre)

Rainbow Theatre

Whistler Conference Centre

4010 Whistler Way, Whistler, BC
Invited Oral Presentation MC08: Machine Parameter Measurements MC08: Machine Parameter Measurements

Speaker

Seongyeol Kim (Pohang Accelerator Laboratory)

Description

Comprehensive knowledge of the six-dimensional phase space distribution of particle beams is critical for optimizing accelerator performance. While conventional diagnostics such as transverse deflecting cavities enable detailed characterization, they require dedicated hardware and significant beamline space. Generative phase space reconstruction (GPSR) techniques have emerged as a promising alternative for beam diagnostics; however, previous implementations have continued to rely on specialized components.

An experimental implementation and validation of a GPSR methodology based solely on standard accelerator elements, including accelerating cavities and dipole magnets, is presented. The approach enables full six-dimensional phase space reconstruction without the need for dedicated longitudinal diagnostics. Simulation studies and experimental measurements conducted at the Pohang Accelerator Laboratory X-ray Free Electron Laser facility demonstrate successful reconstruction of complex and nonlinear beam structures.

The methodology is validated by predicting independent downstream beam measurements that are excluded from the reconstruction process, with the inferred phase space distributions showing close agreement with the corresponding reference data. These results establish a pathway toward predictive, hardware-efficient beam diagnostics applicable across multiple beamline segments and accelerator facilities.

I have read and accept the Privacy Policy Statement Yes

Author

Seongyeol Kim (Pohang Accelerator Laboratory)

Presentation materials

There are no materials yet.