Improving the Reliability of Newsprint Quality Data Using Integrated Factor Networks for Paper Machine Control and Analysis, 1992 Engineering Conference Proceedings
Previous studies have shown how factor analysis can be used to improve the reliability of pulp handsheet tests by quantifying the interrelationships among strength and optical properties. In this study, this technique has been extended to newsprint quality. In addition to the use of factor models for data reconciliation, techniques have been developed for interpreting the underlying factor networks to determine the physical reasons for the interrelationships among the newsprint quality variables. Using these techniques, changes in paper quality can be traced back to the underlying causes. Once the cause for the disturbance is identified using the factor network, actions can be taken to minimize the disturbance. In this study, the application of these techniques to a newsprint paper machine will be presented including the use of integrated factor networks for data reconciliation, disturbance detection and identification.
Improving the Reliability of Newsprint Quality Data Using Integrated Factor Networks for Paper Machi
