On Optimization of Multiparametric Systems with Distributed Global Extremum

Optimization methods for multiparametric systems with a distributed global extremum are considered. A typical example of such a system is a multi-cavity klystron. It is shown that optimization of multiple-cavity klystron can be successfully performed by the macrostep method. The results of optimization of klystrons with extreme efficiency values and gain bands are presented. The possibilities of development of the macrostep method are considered. The problem of forming a uniform discrete neighborhood of a point in a multidimensional space for the formation of a hyperspheric working area is considered. Options for solving this problem are given.