Multi-Objective FEL Design Optimisation Using Genetic Algorithms

Abstract

Simulation studies were carried out to optimise the performance of various FEL designs, with examples including longitudinal current profile shaping for a seeded FEL, and selection of the chicane delays for the High-Brightness SASE technique. In these examples multi-objective genetic algorithms were applied to a single section of the overall facility simulation, i.e. the undulator, as is the common approach. Further studies are also reported in which a full start-to-end simulation chain was optimised, with the aim of delivering a more holistic facility design optimisation.