Speakers
Christopher Hall
(RadiaSoft LLC)
Jonathan Edelen
(RadiaSoft LLC)
Kathryn Wolfinger
(Colorado University at Boulder)
Description
Online models are becoming increasingly more important for the tuning of particle accelerators. Photo-injectors are especially notorious due to there sometimes finicky nature and the difficulties involved in developing models that are suitable for online use. Surrogate models using artificial neural networks are increasingly popular for this application due to their ability to model nonlinear behavior and their execution speed. In support of photo-injector commissioning we have developed a neural network surrogate model and an associated GUI to deploy this model during operations. Here we present our results on the simulations, surrogate model development, and GUI deployment.
| I have read and accept the Privacy Policy Statement | Yes |
|---|
Primary authors
Alex Murokh
(RadiaBeam Technologies)
Christopher Hall
(RadiaSoft LLC)
Jonathan Edelen
(RadiaSoft LLC)
Kathryn Wolfinger
(Colorado University at Boulder)
Maksim Kravchenko
(RadiaBeam)
Marcos Ruelas
(RadiaBeam)
