Method for controlling optimization through self-organized extremum optimization process

Abstract

A method for controlling optimization through self-organized extremum optimization process is used for solving continuous function optimization through the self-organized extremum optimization process, and the solved optimum solution is used as control parameters. The self-organized extremum optimization process comprises the following steps: step 1, determining the local fitness function of a continuous function f(x) according to a control objective; step 2, generating the initial solution s of the continuous function f(x) at random and setting the current optimum solution sbest; step 3, working out the local fitness of each individual si (i equals to 1, ..., m) aiming at the current initial solution; step 4, finding out the local fitness with the maximum fitness value; step 5, structurally changing the neighbourhood space N(s) of the individual sj corresponding to the local fitness with the maximum fitness value, selecting a neighbourhood solution s' belonging to N(s), and taking the selected neighbourhood solution as a new solution unconditionally; step 6, substituting the new solution and the current optimum solution sbest into the continuous function respectively so as to judge that whether the formula that f(s') is less than or equal to f(sbest) is workable or not, if the formula that f(s') is less than or equal to f(sbest) is workable, setting the new solution to be the final optimum solution, and if the formula that f(s') is less than or equal to f(sbest) is not workable, returning to step 2.

Description

technical field [0001] The invention relates to the field of intelligent control, and more specifically, the invention relates to a method for controlling optimization by using self-organized extremum optimization processing. Background technique [0002] In the field of intelligent control, the extreme value optimization algorithm has been widely used in complex optimization problems such as classical combinatorial optimization, complex networks, and production scheduling. The common ones are EO, τ-EO and their improved algorithms, which have fast convergence speed and local optimization capabilities. strong features. However, continuous optimization problems are difficult to analyze intuitively from the perspective of the system. [0003] The current continuous optimization algorithms, such as continuous extreme value optimization algorithm, group-based EO algorithm, and algorithms mixed with other intelligent algorithms, ignore the theoretical basis of EO algorithm to so...

AI Technical Summary

Problems solved by technology

The GEO algorithm simply extends the EO algorithm to continuous function optimization problems, and does not fundamentally analyze its self-organizing optimization process, resulting in a large number of iterations and easy to fall into local optimal problems.

Images