Estimation of multiple specific growth rates: Experimental validation

Publisher Website
Google Scholar
Abstract
The experimental validation of on‐line estimation of multiple specific growth rates for the bakers' yeast fed‐batch process is presented. Pole placement based parameter estimation combined with an asymptotic biomass observer constitute the basic algorithm. The full process model being ill‐conditioned for estimation using the available measured state variables, the use of two partial models related to two different states of the process is suggested. An alternating procedure between two sets of estimation algorithms designed from the partial models is proposed. The performance of the alternating procedure is validated both with simulated and experimental data. The accuracy of the estimates of the three specific growth rates involved in this process is verified according to two criteria based on the respiratory quotient and on the evaluation of the ethanol production/consumption rate.