Akademik Veri Yönetim Sistemi
BIOMASS CONVERSION AND BIOREFINERY, cilt.13, sa.12, ss.10985-11003, 2023 (SCI-Expanded)
The current study aimed to model using various primary models the Aspergillus niger inulinase fermentation in the sugar-beet molasses-based medium, validate the selected functions, and analyze in terms of sensitivity the selected functions. In the selection of models, model comparisons including root-mean-square-error, mean-absolute-error, mean-standard-deviation, determination coefficient, slope, Akaike Information Criterion, bias factor, accuracy factor, and f-testing were used. The results indicated that the best models were Morgan-Mercer-Flodin and Cone for sugar consumption, Stannard, Morgan-Mercer-Flodin, Fitzhugh, and Cone for inulinase production (I), modified Gompertz and Richards for invertase-type production (S), Weibull for S/I ratio, modified Richards and Asymmetric for specific inulinase production, and Asymmetric for specific invertase-type production. By using an independent set of the experimental data, the validation of the selected models gave gratifying results with a high determination coefficient (> 0.968). According to the sensitivity analysis made by increasing or decreasing each parameter in the selected models by 5%, satisfactory results with a high determination coefficient were also yielded in both situations. But with a 5% rise in each parameter of the selected models, higher product formation and higher production and consumption rates are predicted to be achieved. Overall, the results showed that the selected models can serve as universal equations for describing A. niger inulinase fermentation.