PETRA III Storage Ring Performance Improvement Based on Multi-Objective Genetic Algorithms (MOGA)

The performance of the 3th generation light sources relies on the beam lifetime and the injection efficiency, both related with the beam dynamic properties of the storage ring as momentum acceptance and dynamic aperture, respectively. High values of beam lifetime and injection efficiency are desirable to reduce the storage ring instabilities during injection, the radiation losses and the energy consumption of the facility. This paper reports the first application of Multi-Objective Genetic Algorithms (MOGA) to optimize the linear and non-linear beam dynamics of PETRA III storage ring. Genetic algorithms are a heuristic search that mimics the process of natural evolution to optimize problems with a high level of complexity, as in the case of PETRA III storage ring composed by hundreds of magnets. This computational method uses hundreds of CPUs. MOGA is used to maximize the dynamic aperture and the momentum acceptance finding new combinations of quadrupole and sextupole settings in a multi-dimensional parameter space maintaining the solutions in the level of ten picometers