Application of an extended Kalman filter for state estimation of a yeast fermentation

A macrokinetic and regulator model for myeloma cell culture based on metabolic balance of pathways

“…At the beginning of cultivation, if r glc,R < r glc,M , then r glc = r glc,R and the regulator becomes a positive feedback loop, to imitate an accelerating establishment of the enzyme pool for glycolysis. A negative feedback of m is introduced in the regulator model because the enzyme pool may be diluted by the cell growth [36]. If r glc,R > r glc,M , then r glc = r glc,M and the regulator will work in a servo mode to follow up r glc,M .…”

“…The specific glucose uptake rate is influenced by the concentrations of glucose and glutamine and cell density [4,5,[34][35][36].…”

Adaptive multirate state and parameter estimation strategies with application to a bioreactor

“…This motivates the use of an iterated version of the extended Kalman filter equations (Denham and Pines, 1966;Jazwinski, 1970). The iterated extended Kalman filter (IEKF) has been recommended for strong nonlinearities in the output equation and has been successfully used for fermentation processes by Bellgardt et al (1986) and Sargantanis and Karim (1994). The iterated version of the extended Kalman filter equations can be expressed as shown below: Assume that the state prediction x(i) is available at instant i, then the local recursive scheme can be invoked by initializing y(1) to be equal to x ( i ) and then recursing over the equation,…”

Adaptive nonlinear cell mass state estimator for a continuous yeast fermentation

“…Stephanopoulos and San (1984) and Bellgardt et al (1986) discuss experimental results of the application of an extended Kalman filter (EKF) to fed-batch and batch fermentations, respectively. The EKF structure is well suited for fermentation systems because of its capability to deal with nonlinear process relations and to provide combined state and parameter estimation, which can be very attractive.…”