On-line identification of the state and parameters for fed-batch penicillin fermentation process.

A mathematical model with noises which is suitable for penicillin fed-batch fermentation in factory has been presented. By taking the carbon dioxide production rate as an online measured variable, an extended Kalman filter (EKF) was used for the real-time identification of the state and parameters. The study offers a basis for the adaptive optimal control of fermentation processes. It is shown that the estimated values of state by EKF coincide with the experimental data. The results prove that the filter is a better online observer for the state of fermentation process and dynamical parameters.