Kalman Filter Based Glucose Control at Small Set Points during Fed-Batch Cultivation of Saccharomyces cerevisiae

Abstract

A glucose control system is presented, which is able to control cultivations of Saccharomyces cerevisiae even at low glucose concentrations. Glucose concentrations are determined using a special flow injection analysis (FIA) system, which does not require a sampling module. An extended Kalman filter is employed for smoothing the glucose measurements as well as for the prediction of glucose and biomass concentration, the maximum specific growth rate, and the volume of the culture broth. The predicted values are utilized for feedforward/feedback control of the glucose concentration at set points of 0.08 and 0.05 g/L. The controller established well-defined conditions over several hours up to biomass concentrations of 13.5 and 20.7 g/L, respectively. The specific glucose uptake rates at both set points were 1.04 and 0.68 g/g/h, respectively. It is demonstrated that during fed-batch cultivation an overall pure oxidative metabolism of glucose is maintained at the lower set point and a specific ethanol production rate of 0.18 g/g/h at the higher set point.

Comparison of polysialic acid production in Escherichia coli K1 during batch cultivation and fed-batch cultivation applying two different control strategies

“…In order to minimize the glucose measurement noise and to estimate the biomass concentration, the maximal specific growth rate and the volume of the reaction broth, an extended Kalman filter (EKF) was applied (Arndt and Hitzmann, 2004). This software was running on a separate PC and received the glucose measurement values through a serial connection.…”

Control of cultivation processes for recombinant protein production: a review

“…[87,88]). In recent years they were also applied to recombinant protein production processes [6,8,57].…”

Monitoring of fed-batch E. coli fermentations with software sensors

“…The first class includes the classical observers, which are based on the perfect knowledge of both model structure and parameters, such as the Luenberger and the Kalman observers, as well as the non-linear observers. Several works have been published concerning the application of such observers, mainly the Extended Kalman Observer, to biological processes [12][13][14][15][16][17]. Nevertheless, in spite of the satisfactory results reported, an uncertainty in the model parameters can generate a large bias in the estimation of unmeasured state(s) with these methodologies [2,10].…”