Development and pilot testing of modular dynamic thermomechanical pulp mill model to develop energy reduction strategies. Final report

With the development of on-line and real-time process simulations, one obtains the ability to predict and control the process; thus, the opportunity exists to improve energy efficiency, decrease materials wastes, and maintain product quality. Developing this capability was the objective of the this research program. A thermomechanical pulp mill was simulated using both a first principles model and a neural network. The models made use of actual process data and a model that calculated the mass and energy balance of the mill was successfully implemented and run at the mill on an hourly basis. The attempt to develop a model that accurately predicted the quality of the pulp was not successful. It was concluded that the key fro a successful implementation of a real-time control model, such as a neural net model, is availability of on-line sensors that sufficiently characterize the pulp