Adaptive predictive control for the fed batch fermentation process

The industrial fed batch fermentation process is economically important being used in the large scale production of foodstuffs and pharmaceuticals. The process is routinely subject to disturbances originating in variations in raw materials or in micro-organism activity levels. These disturbances mean that batches vary in productivity and the poorly understood nature of the process, its nonlinearity, the wide range of operating states passed through during a batch, and the immeasurability of key process variables means that it is not easy to design a controller that will enforce batch-to-batch and in-batch consistency and high productivity. In this paper, conventional biotechnological theory has been used to develop a set of equations to model the Cyathus striatus fermentation process. To control the state variables of the process around a defined trajectory, model based predictive control has been adopted. This paper compares the performance of the open loop feedback optimal controller with the dynamic matrix control to control the fed batch fermentation process.