Control and estimation algorithms for physico-chemical models of pH-processes in stirred tank reactors

Abstract

Abstract The paper presents a new approach to pH-process modelling and control based on the basic physico-chemical laws. The model includes hypothesized chemical species, the dissociation constants of which can be chosen by static fitting with the aid of so-colled titration curves of real liquid samples. With respect to the chosen concentration variables of the species, the mixing dynamics of-the process is linear. The pH-variables have a highly non-linear momentary relation to the concentrations, but with the aid of a certain transformation a linearity can be achieved. In control applications the unknown concentrations are estimated from pH measurements by a linear algorithm constructed via a deterministic formulation. The estimated concentrations are used to determine the input to the Pi-controller. Several simulations of the algorithms as well as comparisons with the conventional feedback pH-control have been performed and are presented. In addition, simulations using samples from a real industrial process have been made.