Finite Time Estimation for Switched Nonlinear Systems: Application to Stirred Tank Bioreactor

This work presents the design of a class of finite time observer applied to a nonlinear switched system. The proposed observer is applied to a stirred batch anaerobic bioreactor, described by classic mass balances, where a sulfate-reducing process took place and kinetic regimen alteration is induced by a change in the carbon source. The proposed observer has a simple structure and under an adequate choosing of the observeŕs gain the proposed methodology cancels the upper bounds of the system under the different kinetic regimens inducing the required finite time convergence. The performance of the proposed observer is showed via numerical simulations and for comparison purposes a sliding-mode observer is also implemented, both of them are conducted and experimentally corroborated. The convergence of the observer is done with a simple analysis of the estimation error dynamic equations for all the corresponding subspaces and is showed that the convergence of the estimator is reached under the required conditions.