Frank Vavak

Comparison of steady state and generational genetic algorithms for use in nonstationary environments

The objective of this study is a comparison of two models of the genetic algorithm, the generational and incremental/steady state genetic algorithms, for use in nonstationary/dynamic environments. It is experimentally shown that the choice of a suitable version of the genetic algorithm can improve its performance in such environments. This can extend the ability of the genetic algorithm to track environmental changes which are relatively small and occur with low frequency without the need to implement an additional technique for tracking changing optima.

A comparative study of steady state and generational genetic algorithms for use in nonstationary environments

The objective of this study is a comparison of two models of a genetic algorithm — the generational and incremental/steady state genetic algorithms — for use in the nonstationary/dynamic environments. It is experimentally shown that selection of a suitable version of the genetic algorithm can improve performance of the genetic algorithm in such environments.This can extend ability of the genetic algorithm to track the environmental changes which are relatively small and occur with a low frequency without need to implement an additional technique for tracking changing optima.

A Genetic Algorithm with Variable Range of Local Search for Tracking Changing Environments

In this paper we examine a modification to the genetic algorithm — a new adaptive operator was developed for two industrial applications using genetic algorithm based on-line control systems. The aim is to enable the control systems to track optima of a time-varying dynamic system whilst not being detrimental to its ability to provide sound results for the stationary environments. When compared with the hypermutation operator, the new operator matched the level of diversity introduced into the population with the “degree” of the environmental changes better because it increases population diversity only gradually. Although the new technique was developed for the control application domain where real variables are mostly used, a possible generalization of the method is also suggested. It is believed that the technique has the potential to be a further contribution in making genetic algorithm based techniques more readily usable in industrial control applications.

Load Balancing Application of the Genetic Algorithm in a Nonstationary Environment

There is a wide range of industrial activities involving load balancing problems and there is currently no general approach to such problems. The genetic algorithm, (Holland 1975), has proved to be successful on optimisation problems and is often used in conjunction with other methods. This paper describes and compares two methods which use the genetic algorithm to balance the load of the presses in a sugar beet pressing station. Because the station is a time varying system, possibilities of tracking changing environment has to be considered and an adaptive strategy is needed.The first approach uses the genetic algorithm to optimise an on-line mathematical model of the station and the use of this model to maximize the percentage of dry substances in the pressed pulp produced. The second approach implements the genetic algorithm in direct control of the station with the identical goal of reducing the moisture content of the pressed pulp. This reduction improves energy efficiency of the driers which dry the pulp to be used as animal feed.