Cavity resonator tuning using perturbation-based extremum seeking control

Perturbation-based extremum seeking (ES) is a non model-based approach to maximize or minimize the output of a dynamic system without requiring any explicit information about the dynamics of the system, steady state map, or the value of the extremum. In this work, perturbation-based ES is used to minimize power reflection from a cavity resonator in order to provide maximum accelerating field for particles travelling through the cavity. Power reflection is minimized by equalizing the frequency of the cavity and RF source. The controller sends commands to a step motor setup connected to the cavity. The step motor moves the tuner plate which changes the frequency of the cavity. Simulation and experimental results are presented which indicate that the algorithm can successfully minimize power reflection despite variations in the value of the minimum.