Microaqueous Interesterification Reaction Control by a Feedforward/Feedback Controller with an On-line Enzyme Activity Estimator

Abstract

To maintain constant composition of triglyceride in the effluent from a microaqueous bioreactor using a filamentous fungus immobilized within biomass support particles as lipase catalysts, a feedforward/feedback control system with an online enzyme activity estimator was developed. Feed flow rate as a manipulate value is calculated from the improved glyceride-enzyme complex model. The parameter in the model, enzyme activity, is adjusted repeatedly by the on-line estimator, in which the discrepancy between observed triglyceride data and calculated values is minimized by a trial-and-error method. To evaluate the performance of the control system, both simulation and experimental studies of the uncertainty of model parameters such as the deactivation constant and initial enzyme activity, and of stepwise change of substrate concentration of the feed medium as a disturbance were carried out. These results have proved that the above system successfully controls the glyceride composition of continuous interesterification reaction.