Self-cycling fermentation (SCF) is a process in which sequential batch fermentations are performed using a computerized feedback control scheme. During the fermentation, half of the reactor volume is periodically removed and replaced by fresh medium. The periodicity of the system is determined by continuously monitoring a growth associated parameter such as dissolved oxygen concentration or carbon dioxide evolution. The system cycles when an extremum in the monitored parameter is detected, corresponding to a decrease in metabolic activity. This results in very stable and repeatable growth cycles and synchronized cell populations.
In this work, a segregated, structured microbial population balance model is formulated and used to numerically simulate the SCF process. The model uses cell mass as the index of physiological state. In order to solve the cell mass population balance model for SCF, a numerical scheme using the Galerkin finite element method (GFEM) and the predictor-corrector Euler method was developed. The model outputs were compared and validated with previously obtained experimental data. The cell mass model as described by was able to capture the major growth profiles of the system. However, the model was not able to describe the cell number profile or cell synchrony. By introducing a feedback mechanism between the critical division mass m A and the limiting substrate into the classical cell population model, cell synchrony and qualitative agreement with experimental data are obtained.