Modeling and Predictive Control of Yeast Fermentation Process

This work mainly focus on development of advanced process control on the continuous fermentation process using Saccharomyces cerevisiae. Today, a lot of research has been done on renewable energy and it has been found that ethanol is one of the best alternatives fuels to substitute petroleum fuels. However, all of this can only be achieved if the production of ethanol is efficient and economical enough. A lot of industry nowadays uses fermentation in batch mode due to the problems occurred in 1970s such as low productivity, low yield, and high level of contamination. Recently, continuous fermentation processes are optimized based on kinetic models to achieve high productivities, high process flexibility and stability and less expensive production cost compared to batch processes. In addition, process control development for continuous fermentation is much better since a lot of research on advanced process is in the continuous mode and almost all kinetic models currently available for continuous fermentation with Saccharomyces cerevisiae are in steady state. One of the disadvantages of standard feedback controller is that the action can only be taken after the system has been affected by the disturbance. Thus, an advanced process control (APC) strategy will be developed based on this process. The objective of this work is to optimize the performance of the fermentation process in terms of yield and productivity by using model predictive control (MPC). In this process, the manipulated variable that has been considered is inlet temperature and inlet substrate concentration and the control variable is temperature in the reactor and ethanol concentration. The successful implementation of the controller is greatly affected by the accuracy of the process model.