Investigation on realtime deterministic chaos control by means of evolutionary algorithms

Abstract

This contribution introduces an investigation on deterministic spatiotemporal chaos real-time control, based on evolutionary algorithms use. Problem is defined so that spatiotemporal chaos model would behave like real-time process. Two evolutionary algorithms are used for chaos control here: differential evolution and self-organizing migrating algorithm in total number of 10 versions (SOMA in 4, DE in 6). Like a model of spatiotemporal chaos so called coupled map lattices based on logistic equation are used. Main aim of this investigation was to show that evolutionary algorithms are capable of real-time control of deterministic chaos when cost function is properly defined. Investigation consists of three different case studies with increasing calculation complexity. For both algorithms each simulation was 100 times repeated to show and check robustness of used methods. Copyright © 2006 IFAC.