Control of ethanol concentration in a fed-batch cultivation of Acinetobacter calcoaceticus RAG-1 using a feedback-assisted iterative learning algorithm

A control scheme using a feedback-assisted iterative learning control algorithm to maintain the optimal ethanol concentration for the enhancement of cell growth and emulsan production in fed-batch cultivation of Acinetobacter calcoaceticus RAG-1 is presented. The optimal concentration of ethanol for cell growth and emulsan production was determined by considering the inhibitory effect of ethanol. To maintain the optimal concentration of ethanol, continuous feeding of ethanol was adopted using the feedback-assisted iterative learning control algorithm. A mathematical kinetic model is used to simulate the objective process. The transfer function of the objective process was approximated to the FOPDT (First Order Plus Dead Time) model and model parameters were determined by the LSE (Least Square Estimation) method. In the transfer function modeling, to evaluate the process model parameters which can describe the objective process, the disturbance included in the process was eliminated by linearization of two data sets of preliminary fed-batch runs. Fed-batch cultivation experiments using a feedback-assisted learning control algorithm were performed. The results showed that the convergence performance was improved as the run was iterated and the emulsan yield was increased compared with that of the runs controlled by only the feedback loop.