Title :
Adaptive neural network filter for steel rolling
Author :
Fechner, Thomas ; Neumerkel, Dietmar ; Keller, Ivo
Author_Institution :
Daimler-Benz AG, Berlin, Germany
fDate :
27 Jun- 2 Jul 1994
Abstract :
Rolling mill control systems which use measurements of the rolling force (gauge control) must compensate for eccentricity of the rolls. The proposed neural eccentricity filter provides this compensation without any information about the position of the rolls. This application requires fast online adaptation of the filter due to time-variant behavior of the process which is provided by a recursive least squares learning algorithm
Keywords :
adaptive filters; neural nets; process control; rolling mills; steel industry; adaptive neural network filter; eccentricity compensation; gauge control; neural eccentricity filter; recursive least squares learning algorithm; rolling force measurement; rolling mill control systems; steel rolling; time-variant behavior; Adaptive filters; Adaptive systems; Control systems; Force control; Force measurement; Information filtering; Information filters; Milling machines; Neural networks; Steel;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374837
