Intercriteria analysis of simple genetic algorithms performance

Abstract

Recently developed approach of InterCriteria Analysis is here applied aiming at an assessment of the performance of such a promising stochastic optimization technique as simple genetic algorithms. Considered algorithms, as representatives of the biologically-inspired ones, are chosen as an object of investigation since they are proven as quite successful in solving of many challenging problems in the field of complex dynamic systems optimization. In this investigation simple genetic algorithms are applied for the purposes of parameter identification of a fermentation process. Altogether six simple genetic algorithms are here considered, differ from each other in the execution order of main genetic operators, namely selection, crossover and mutation. The apparatuses of index matrices and intuitionistic fuzzy sets, underlying the InterCriteria Analysis, are implemented to assess the performance of simple genetic algorithms for the parameter identification of Sac-charomyces cerevisiae fed-batch fermentation process. The obtained results after the InterCriteria Analysis application are thoroughly analysed towards the algorithms outcomes, such as convergence time and model accuracy.