Genetic Algorithm and Cuckoo Search Hybrid Technique for Parameter Identification of Fermentation Process Model

Abstract

This paper presents a hybrid scheme based on two population-based metaheuristic techniques, namely genetic algorithm (GA) and cuckoo search (CS). In particular, the hybrid is formed by the application of standard simple genetic algorithms (SGA) and CS, specifically adapted and for first time applied by the authors for the purposes of parameter identification of yeast fed-batch fermentation process model. The parameters of the hybrid technique SGA-CS have been thoroughly explored and tuned to meet the specific peculiarities of the considered here optimization problem. A comparison of SGA, CS and developed hybrid SGA-CS has been performed, outlining the advantages and disadvantages of each algorithm. Additionally, a new modification of SGA-CS hybrid technique, inspired by proven as very effective modification of SGA, working with implementation of main genetic operators in order crossover, mutation and selection, has been here elaborated. Presented modified hybrid technique has been tested, aiming at verification of the obtained promising results of developed SGA-CS technique.

Keywords

• genetic algorithm
• cuckoo search
• hybrid technique
• parameter identification
• fed-batch fermentation process