Performance Evaluation of the Non-structured and Structured Kinetic Modelling for the ABE Process. From Batch to Continuous Fermentation

This work proposes the performance evaluation of three novel kinetic models of a biological process, one considering the traditional non-Structured Monod and Ludeking-Piret representations, also the named structured modelling approach which tries to include metabolic kinetic information, as possible and a final one utilising the so-called Non-Structured Phenomenological methodology that aim to become a middle-point between the both above mentioned, for the Acetone–Butanol–Ethanol (ABE) system as a study case. The obtained results under batch and continuous regimes with free cells and batch and continuous regimes with immobilized cells are considered. The above, for the determination of the reproducibility and predictability capabilities of the corresponding kinetic models via a statistical analysis based on the determination coefficient methodology (R2), the modelling performance index (Π) and the ANOVA analysis for mean comparison and significance of the simulation results generated with the three model structures versus experimental data. These analyses indicate that the traditional non-structured modeling approach is just statistically significant for the ABE fermentation process operating with free cells, still the performance of the phenomenological kinetic model is more reliable for the predicting of the behavior of fermentations utilizing immobilized cells.