A dynamic model for basis weight variations in paper machines is established and estimated recursively from scanned measurements using a combination of an exponential forgetting and resetting least squares estimator and an extended Kalman filter. The algorithm provides online estimation of basis weight variations in cross-machine direction and machine direction, as well as estimates of the dynamic parameters and noise characteristics of the process
Published in:
Control Applications, 1992., First IEEE Conference on
Date of Conference: 13-16 Sep 1992