Experiences building and validating a large-scale dynamic paper machine model

Concerns the development of a dynamic model of a 5-ply board machine producing packaging board. Three models were developed in decreasing order of complexity. First a detailed dynamic model of the short circulations was built using Simons IDEAS, which is a special purpose dynamic simulation package intended for the pulp and paper industry. This full model contained a comprehensive differential and algebraic equation set so consequently quickly became too unwieldy for the intended uses of troubleshooting, aiding retrofits and model-based control studies. A slightly simplified version of this model was re-engineered in Simulink, dropping the algebraic constraint equations and implementing the transport delays in a more elegant manner. Finally transfer function models, suitable for online control and optimisation tasks were identified from production data with a structure given by semi-rigorous reasoning. All three models were adjusted and subsequently validated against almost one year of plant operating data. This paper highlights the difficulties in validating a large industrial system with widely varying time constants, dubious transducers and the ubiquitous noise. We also discuss the necessary, but delicate, task of performing model reduction; contrast different simulation tools used at the various levels of model hierarchy and finally the paper demonstrates the use of simplified models to improve the performance of the adaptive IMC controller for the basis weight in the finished board