Speaker
Description
Traditional phase scans at LANSCE are useful for tuning longitudinal capture but provide only indirect information about the bunch distribution. This work extends a deep neural network-based reconstruction method to the first two modules of the side-coupled cavity linac. Simulated two-dimensional phase scans were generated with HPSim by varying the RF phases of Modules 5 and 6 and recording the transmitted current after the absorber/collector diagnostic. A retrained network reconstructed correlated Gaussian longitudinal phase space distributions from these scans, recovering their approximate size, orientation, and centroid. These results support further development using realistic distributions and measured CCL phase scans.
Footnotes
The research presented in this article was supported by the Laboratory Directed Research and Development program of Los Alamos National Laboratory under project number 20250783MFR.
Funding Agency
U.S. DOE and NNSA
| Paper status | Resubmitted proceeding files received and assigned to an editor. Accepted by Submitter. |
|---|