An Adaptive Cascade Structure for the Estimation and Control of Perfusion Animal Cell Cultures

This paper presents an effective and robust structure for the estimation and control of perfusion cell cultures, in which the cells and glucose concentrations are simultaneously controlled by manipulating the dilution and bleed rates. Firstly, a partially linearizing feedback controller is designed to ensure an approximately linear decoupled dynamics between the controlled outputs and the manipulated inputs. Then the model of the inner loop is used to design an extended Kalman filter, which estimates all the system states used in the implementation of the linearizing feedback control law from the (possibly noisy) measurements of cells and glucose concentrations. Two PI controllers are used in the outer loop for a good tracking performance, which are tuned using a receding horizon optimization procedure. The proposed structure shows good performance and robustness with respect to parameter uncertainties, non-cancelled nonlinearities and measurement noise.