Speaker
Description
We present a beam dynamics study aimed at optimizing the transverse emittance of electron bunches in a photoinjector, motivated by the performance requirements of Free-Electron Lasers. The study is conducted using the accelerator configuration of the Photo Injector Test Facility at DESY in Zeuthen (PITZ), with the overarching goal of developing emittance optimization strategies for the European XFEL photoinjector. To this end, we perform large-scale beam dynamics simulations using the simulation code ASTRA, systematically sampling from a low-dimensional latent representation of temporal laser pulse profiles. This latent space is learned from a broad set of physically plausible pulses using a Wasserstein Autoencoder (WAE), enabling compact and structured exploration of pulse shape variations. The ability to efficiently sample from this representation supports targeted emittance studies that would be computationally prohibitive in the original high-dimensional shaping parameter space. For each simulation, beam quality metrics such as normalized projected emittance and slice mismatches are recorded. The study reveals meaningful correlations between latent coordinates and beam quality, demonstrating the utility of WAE-based representations in guiding laser pulse design. We briefly outline future directions involving neural surrogate models to accelerate beam emittance optimization.
Funding Agency
German Federal Ministry of Research, Technology and Space within the project OPAL-FEL, grant number 05D23GT1
| In which format do you inted to submit your paper? | LaTeX |
|---|---|
| Preprint marking on your proceeding paper | I wish my paper to be marked as preprint. |