On the optimization of the non-linear lattice of BESSY III

Helmholtz-Zentrum Berlin plans to construct a fourth- generation greenfield synchrotron light source in the early 2030s to replace BESSY II, a 1.7 GeV machine that has been running since 1998. The optimization of the linear lattice already considers non-linear aspects, such as minimizing the necessary sextupole strength and, for the minimal case of two families of sextupoles, phase cancellation to reduce the resonant driving terms. In preparation for the final optimization of 8 sextupole families and the single octupole, different approaches are compared: multi-objective genetic optimization, for a lattice with given error sets and orbit correction on the one hand, and the minimization of the resonant driving terms and the detuning terms on the other hand. Here, analytic formulas are used, so after a single evaluation of the Twiss parameters, the driving terms can be determined for different combinations of sextupole and octupole values. The results will determine the strategy for optimizing the lattice’s non-linear behavior, i.e., dynamic aperture and momentum acceptance, taking the efficiency of the optimization into account.