Fuzzy modelling of an alcoholic fermentation with Zymomonas mobilis

Abstract

This article presents the followed method for the obtaining of a dynamic fuzzy model that represents a bacterial fermentation process of bioethanol production. Specifically, this process works with Zymomonas mobilis bacterium. The alcoholic fermentation with this bacterium presents high yields of ethanol, but with highly complex behaviors. From the identification and control point of view, this represents a great difficulty, and to obtain a precise and suitable model is very important. The proposed identification methodology combines clustering and rules reduction methods to approximate a complex system by means of a set of local linear models, since the fuzzy modeling used is based on the Takagi-Sugeno (TS) model where the consequent ones are in fact the lineal functions. With the presented process model in this work is possible to capture the dynamic of the process of bioethanol obtaining by means of Zymomonas mobilis, and in addition, the model is compact and with a desirable structure for its future application in several model based control structures, for the sake of optimizing the process productivity and to achieve a diminution in the production cost.