Parameter identification and online estimation of a reduced kinetic model

We present the estimation techniques used to update the model parameters in a reduced kinetic model describing the oxidation-reduction reactions in a hydrothermal oxidation reactor. This model is used in a nonlinear model-based controller that minimizes the total aqueous nitrogen in the reactor effluent. Model reduction is accomplished by combining similar reacting compounds into one of four component groups and considering the global reaction pathways for each of these groups. The reduced kinetic model developed for this reaction system provides a means to characterize the complex chemical reaction system without considering each chemical species present and the reaction kinetics of every possible reaction pathway. For the reaction system under study, model reduction is essential in order to reduce the computational requirement so that online implementation of the nonlinear model-based controller is possible and also to reduce the amount of a priori information required for the model