Speaker
Description
We present progress on developing a physics-informed machine learning framework that reconstructs two-dimensional traverse particle distributions. In this framework, the distribution is inferred from sparse electric-field measurements obtained with a simulated electro-optic beam profile monitor consisting of eight crystals arranged around the beam. A key component of the approach is the use of physics-informed loss terms that supplement standard image-based losses and mean-squared-error training, as well as moment-matching penalties based on centroid and covariance matrix, enforcing physically plausible bunch shapes. Additionally, physics-based constraints are included to ensure that the reconstructed distributions produce boundary-field patterns consistent with electric field measurements.
| 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. |
| I no longer wish to present this contribution, please withdraw it. | Please withdraw my contribution |