LANL, Los Alamos, New Mexico, USA
This paper addresses an iterative particle beam phase and amplitude feedforward controller tuning method based on the gradient search approach. The method does not need an a priori plant model as it only needs data collected in previous experimental runs. The controller is implemented on a field programmable gate array (FPGA) equipped with a real-time operating system and a network connection. Data from each RF pulse is collected and sent via the network to the FPGA for processing. The controller tuning is performed between the RF pulses. Once the tuning is performed, the controller parameters are downloaded to the controller in the FPGA and new controller parameters are applied at the upcoming RF pulse
