Model-based monitoring and control of batch pulp digester

In a recent work by Lee and Datta (1994), the feasibility of applying a model-based dynamic estimation and control technique was tested. It was shown that reliable monitoring and control of the Kappa number can be achieved, even in the presence of significant errors in the initial state estimates, given that the model parameters can be adapted efficiently. However, since the available fundamental pulping models are of very high-order involving many parameters, simultaneous update of the states and parameters resulted in an ill-conditioned estimation problem, leading to unacceptably slow convergence or divergence. This indicated a need for a reduced-order model that captures the essential aspects of the pulping dynamics. Development of such a model and testing of its suitableness for online use are topics of this paper. We verify the adequacy of the proposed reduced-order model and its online adaptability by applying a standard model-based nonlinear estimation and predictive control technique.