Speaker
Corinna Steffen
(Heidelberg Ion Beam Therapy Centre)
Description
Previous work has shown that the efficiency of multi-turn injection (MTI) is heavily influenced by the injection bumper curve. By implementing a neural network with a custom loss function based on the given machine limitations from the HIT Medical Synchrotron, the curve was optimized via a sequence of third-order polynomial functions by online training. Experiments with variable epoch lengths demonstrated an improved current output with up to a 20% increase over the previously established maximum. These results underscore the importance of bumper curve optimization in maximizing synchroton efficiency, demonstrating a more effective alternative to a curve tuned by operators regarding time spent and outcome.
| In which format do you inted to submit your paper? | LaTeX |
|---|
Author
Corinna Steffen
(Heidelberg Ion Beam Therapy Centre)
Co-author
Marcel Hun
(Heidelberg Ion Beam Therapy Centre)