Modelling of fermentation kinetic processes by non-linear optimisation techniques.

Abstract

A non-linear optimisation procedure is proposed to model fermentation kinetic processes. The fundamentals of the method are based on the minimisation of an objective function which does not require the solution of the model equations. The enzyme kinetics expressions used to model pH effects on growth rate kinetic constants are analysed, A model for the pH effect (PHEM model) on growth rate associated and cell concentration associated reaction parameters is proposed. Model identification and parameter evaluation for the batch and continuous culture of Candida lipolytica in a glucose-ammonium medium at 30 °C and several pH levels (4 to 6 in continuous culture and 3 to 7 in batch culture) is carried out in order to examine the optimisation procedure proposed here. The optimisation approach was found to be highly efficient for the analysis of data from the continuous culture, but failed in the analysis of data from the batch culture. The continuous culture was found to be described adequately by a double kinetic Contois-Contois interactive model with glucose and ammonium as limiting nutrients. The batch culture was found to be described adequately by a single kinetics Monod model with glucose as the rate limiting nutrient and endogeneous respiration repressed by the carbon source. The pH affected both continuous and batch cultures, significantly. The pH effect is examined on the basis of the response of the model parameters to hydrogen ion concentration, pH effects on growth rate constants, stoichiometric coefficient of carbon and nitrogen sources, endogeneous respiration coefficients and inhibition constant of glucose (for endogeneous respiration in the batch culture) are discussed. Some examples of the PHEM model are given, from data found in the literature and from results found in this work.

Date of Award	1980
Original language	English
Awarding Institution	University of Bath