Subspace identification based inferential control applied to a continuous pulp digester

The idea of constructing a data-driven stochastic system model through subs pace identification for the purpose of inferential control is investigated. Various available methods for designing an inferential controller are disc ussed and their limitations are brought out, particularly in applications i nvolving multi-variable processes. Practical issues that arise in identifyi ng a system model geared toward inferential control using a subspace method are discussed. They include: handling of nonstationary disturbances, handl ing of multi-rate measurements/missing data, and secondary measurement sele ction. With the identified stochastic system model, a multi-rate Kalman fil ter can be designed and coupled with a model predictive controller. The met hod is applied to a continuous pulp digester, which is a complex distribute d parameter system involving heterogeneous reactions. The application study indicates much potential for the data-based approach. (C) 1999 Published b y Elsevier Science Ltd. Ail rights reserved.