Speaker
Description
The Proton Synchrotron (PS) at CERN is equipped with numerous RF systems allowing for evolved longitudinal beam manipulations to adapt the number of bunches and their spacing. The beam produced for the LHC undergoes several bunch splittings, merging and batch compression. Each manipulation must be carefully adjusted to minimize the spread in bunch parameters at PS extraction. The design of these complex RF processes relies on the non-linearities of the longitudinal equations of motion. This step is usually performed applying numerical optimization with dedicated simulation codes and requires manual processing for deployment in operation. To improve the design of existing manipulations and explore new schemes, surrogate models are proposed in view of improved integration in the accelerator control system. In this work, surrogate models in form of artificial neural networks are built to control the parameters of the RF bucket based on the desired bucket areas and shapes, and generating the expected RF parameters (voltage and phase) to the control system.
