1. Study of the multicylinder drying section of the Klabin-PR MP-7 paper machine.
- Author
-
Da Motta Lima, Oswaldo Curty, Pereira, Nehemias Curvelo, and Deganutti, Flavio
- Subjects
PAPER dyeing ,PAPERMAKING ,PAPER industry ,MOISTURE ,PAPERMAKING machinery ,FACTORIES ,PAPER ,CONDENSATION ,RAINFALL - Abstract
Multicylinder sections of paper dryers consume the most energy in an industrial plant. They are also responsible for the final characteristics of the paper produced. In this context, the modeling of the paper drying process is very important, and is normally done by phenomenological approaches. An interesting and promising alternative, due to the possibilities of its application, is through the identification of an operational standard behavior, obtained by the use of neural networks in the simulation of the final paper moisture content. This alternative was implemented in the drying section of the KIabin Paraná Papeis S.A. (Klabin-PR) MP-7 paper machine. For this study, we collected operational data related directly and indirectly to the drying process, composing a database with 110 preselected variables. These data were analyzed with the help of Statistica Neural Network software, which, besides its use for training and choosing the more capable neural networks to model the chosen variable, allows the selection of the group of variables of larger influence in the studied process. The results obtained by the use of the neural network modeling were quite satisfactory, with an average error of less than 2% in the prediction of the final paper moisture content Application: Neural networks may be useful to help control moisture content and increase dryer efficiency. [ABSTRACT FROM AUTHOR]
- Published
- 2005