Cuckoo Search Algorithm for Parameter Identification of Fermentation Process Model
Parameter identification of non-linear dynamic processes, among them fermentation ones, is rather difficult and non-trivial task to be solved. Failure of conventional optimization methods to provide a satisfactory solution provokes the idea some stochastic algorithms to be tested. As such, the promising metaheuristic algorithm Cuckoo search (CS) has been adapted and applied for a first time to a parameter identification of S. cerevisiae fed-batch fermentation process model. Aiming to improve the model accuracy and the algorithm convergence time, several pre-tests adjustments of CS have been done according to the specific optimization problem. Obtained results confirm the effectiveness and efficacy of the applied CS algorithm. In addition, a comparison between CS and simple genetic algorithm, proved as successful in parameter identification of fermentation process model, has been done. Algorithms advantages and disadvantages have been outlined and the more reliable one have been distinguished.
KeywordsCuckoo search Genetic algorithm Parameter identification Fermentation process
The work is partially supported by the National Science Fund of Bulgaria under grants DM 07/1 “Development of New Modified and Hybrid Metaheuristic Algorithms” and DN02/10 “New Instruments for Knowledge Discovery from Data, and Their Modelling”.
- 5.Goldberg, D.E.: Genetic Algorithms in Search, Optimization and Machine Learning. Addison Wesley Longman, London (2006)Google Scholar
- 6.Karaboga, D.: An idea based on honey bee swarm for numerical optimization. Technical report-TR06, Erciyes University, Engineering Faculty, Computer Engineering Department (2005)Google Scholar
- 7.Kennedy, J., Eberhart, R.: Particle swarm optimization. In: Proceedings of IEEE International Conference on Neural Networks, vol. IV, pp. 1942–1948 (1995). https://doi.org/10.1109/ICNN.1995.488968
- 10.Pencheva, T., Roeva, O., Hristozov, I.: Functional State Approach to Fermentation Processes Modelling. Prof. M. Drinov Academic Publishing House, Sofia (2006)Google Scholar
- 11.Petrov, M., Ilkova, T., Vanags, J.: Modelling of a batch whey cultivation of Kluyveromyces marxianus var. lactis MC 5 with investigation of mass transfer processes in the bioreactor. Int. J. Bioautomation 19(1), S81–S92 (2015)Google Scholar
- 12.Roeva, O., Fidanova, S., Paprzycki, M.: Population size influence on the genetic and ant algorithms performance in case of cultivation process modeling. In: Fidanova, S. (ed.) Recent Advances in Computational Optimization. SCI, vol. 580, pp. 107–120. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-12631-9_7CrossRefGoogle Scholar
- 13.Roeva, O., Pencheva, T., Tzonkov, St., Hitzmann, B.: Functional state modelling of cultivation processes: dissolved oxygen limitation state. Int. J. Bioautomation 19(1), S93–S112 (2015)Google Scholar
- 14.Roeva, O., Atanassova, V.: Cuckoo search algorithm for model parameter identification. Int. J. Bioautomation 20(4), 483–492 (2016)Google Scholar
- 16.Yang, X.-S., Deb, S.: Cuckoo search via Lévy flights. In: Proceedings of World Congress on Nature & Biologically Inspired Computing (NaBIC), USA, pp. 210–214. IEEE Publications (2009)Google Scholar