Simultaneous adaptive predictive control of the partial pressures of dissolved oxygen (pO2) and dissolved carbon dioxide (pCO2) in a laboratory-scale bioreactor

Abstract In submerged cultures, the variation of the partial pressures of dissolved oxygen (pO 2 ) and of dissolved carbon dioxide (pCO 2 ) strongly affects micro-organism metabolism. Unfortunately, in a laboratory scale bioreactor these variables are very difficult to control using conventional controllers because of the poorly understood and constantly changing dynamics of the biological system. A control strategy based on the decoupling of these two variables has been developed. The agitation rate controls the pO 2 , and the air flow rate controls the pCO 2 . The cross influences are considered as external disturbances by both controllers. Each controller uses an adaptive-predictive algorithm consisting of an incremental linear model of the dynamics, a robust recursive estimator, an algebraic predictor, a Generalized Predictive Control optimizer and a monitor of the regulation smoothness. The strategy has been implemented on a PC computer under the visual programming environment Lab view and successfully used under various conditions. The major improvements, when compared with conventional controllers are as follows: no manual tuning is required during the culture or between experiments, efficient pO 2 and pCO 2 controls are achieved despite external disturbances, and finally since these variables remain stable, the agitation rate and the air flow rate curves provide a qualitative image of the oxygen consumption and of the carbon dioxide production rates by the micro-organisms, making it easier for the operator to spot otherwise hidden metabolic changes.