State estimators for better bioprocesses operation

Abstract

Abstract Nowadays many advanced control tools are available to improve industrial processes (e.g. virtual analyzers, state estimators, predictive and multivariable controllers, etc.). Particularly, state estimation techniques have a long development history focused mainly to supply the lack of system measurements. Between the applications involved by state estimators we can highlight: more detailed process monitoring, mathematical model fitting and update, transient data reconciliation and feedback control. The goal of our work is showing the advantages of using state estimators as a tool for better operation of a bioprocess and to illustrate how a state estimator can be easily implemented. We evaluate these applications using three state estimation techniques in a bioreactor: Extended Kalman Filter (EKF), Constrained Extended Kalman Filter (CEKF), and Moving Horizon Estimator (MHE). Further we introduce the CEKF & Smoother (CEKF&S) as a simple and efficient alternative to MHE. As benchmark case study we have chosen the continuous glucose fermentation with Zymomonas mobilis bacteria to produce ethanol. Our results clearly show the relevance of state estimators as a tool to improve the bioprocesses operation.