초록
<P><B>Abstract</B></P><P>In this study, we introduce a two-phase design approach to optimize a fed-batch fermentation process using a mixture of sugars to produce ethanol. In the first phase, the batch time-series observations were collected and used to construct a kinetic model of <I>Saccharomyces diastaticus</I> LORRE 316. The primary kinetic model was then applied in the second phase to determine the optimal control policy for a fed-batch fermentation process and to retune the kinetic model parameters using the fed-batch information. Two runs were performed in the second phase to achieve the maximum ethanol production rate for the fed-batch fermentation process. Parameter estimation and process optimization were carried out sequentially with a run-to-run approach, which yielded useful estimated parameters even when the model is extrapolated, as observed from cross-validation.</P>