Nonlinear high-gain observers with integral action: Application to bioreactors

Abstract

This paper deals with the problem of estimating the states in nonlinear systems with parameters that are not precisely known through the implementation of proportional nonlinear high gain observers with additional integral action. It is known that non-linear observers (NL-P) show excellent results in the estimation of state variables but these depend heavily on precise knowledge of model parameters, this is not always possible because in many process the parameters may suffer sudden shifts as in the biological processes, which in turn will certainly broke the asymptotical convergence observer property. We propose the addition of a integral action in order to deal with possible change of parameters. In this work it is showed that the so called nonlinear integral high gain observers (NL-PI) are able to cancel the dynamical disturbances due to parameter changes. We studied the substrate estimation in a wastewater treatment bioreactor. The simulation study was carried on the basis of experimental data of a batch culture of Spirulina maxima for the pollutant removal present in the culture media.