Determination of similarity criteria in optimization tasks by means of neuro-fuzzy modelling

Abstract

A method is proposed for solving optimization problems with high complexity when searching for the function minimum by using methods and means of similarity theory and neuro-fuzzy modelling. The problem with nonlinear objective function and constraints is transformed into a task with a nonlinear objective function and linear constraints. In this task, the basic similarity criteria are presented in the form of membership functions. Dependent similarity criteria are defined through the base with the use of standard computational procedures.