Iterative learning control to cancel beam loading effect in particle accelerators

Iterative learning control (ILC) is an approach to improve the performance of a system that carries out the same operation repeatedly. It stores error and control output throughout every iteration, and incorporates the stored information into generating new control output. In a cavity resonator, a standing wave electromagnetic field is formed to accelerate charged particles passing through the cavity. In the process of the acceleration, energy is transferred from the accelerating field to the beam. As a result, the cavity voltage drops. This effect is referred to as beam loading. A feedback controller is responsible to maintain constant amplitude and phase for the cavity field oscillation. However, for certain applications, a feedback controller is not fast enough. In this project, we aim to use ILC to reduce beam loading effect in cavity resonators. Since beam loading is a repeating disturbance, the error is repetitive as well, and the controller can use a non-causal learning function to pre-emptively counteract beam loading effect.

Faculty Supervisor:

Guy Dumont

Student:

Partner:

Lund University

Discipline:

Engineering

Sector:

Education

University:

The University of British Columbia

Program: