Estimation strategies for kinetic parameters of an alcoholic fermentation bioprocess

This work deals with the design of on-line estimation strategies for imprecisely known kinetics of an alcoholic fermentation bioprocess, which takes place inside a Fed-batch Bioreactor. The kinetic rates of this process are highly nonlinear, uncertain and time-varying. In order to obtain on-line estimates useful in control design, three kinetic parameter estimation algorithms are implemented: an observer-based estimator, a regressive estimator and a high-gain observer. If necessary, the unavailable states are provided by an asymptotic observer. Several comparisons regarding the performance and behavior of estimation strategies are achieved by numerical simulations.