State observation of a nonlinear system: Application to (bio)chemical processes
Résumé
The design of simple nonlinear observer-based estimators for the on-line estimation of the reaction rates in chemical and biochemical processes is presented. The proposed estimators do not assume or require any model for the reaction kinetics and are very successful for accurately estimating these variables. They are first presented in a continuous-time version. Then, the stability and convergence of a discrete-time version, obtained by a direct forward Euler discretization of these estimators, are established. A main characteristic of these estimators lies in the easiness of their implementation and in particular their calibration. Indeed, their gain does not necessitate the resolution of any dynamical system and is explicitly given. Moreover, its tuning is reduced to the calibration of a single parameter, and this is true regardless of the number of reaction rates to be estimated. The performances of the proposed estimators are highlighted through experimental results dealing with biological and chemical reactions performed in stirred-tank reactors.