4. Control description
Two types of control have been developed and put into service by ALSTOM on industrial sites:
neural network-based non-linear filtering to eliminate disturbances caused by roll eccentricity from the rolling force measurement signal;
active compensation of cylinder eccentricity using neural networks.
4.1 Non-linear neural network filtering
Principle
Eccentricity is modeled using a neural network (figure 12); this network is fed by the angular position of each rolling mill roll and produces a signal representative of roll eccentricity.
The filtered rolling force is obtained by subtracting...
Exclusive to subscribers. 97% yet to be discovered!
Already subscribed? Log in!
Control description
Theoretical study and development of the new control
References
Articles available on the Web
- (1) - YASMINE (Y.) - Stage sur les réseaux de neurones. - CICT, 1999. http://www.cict.fr.
- (2) - HARDY (J.-M.), STRASSERA (A.) - Les réseaux de neurones. - 1998.
- (3)...
Exclusive to subscribers. 97% yet to be discovered!
Already subscribed? Log in!
Article included in this offer
"Control and systems engineering"
(
145 articles
)
Updated and enriched with articles validated by our scientific committees
A set of exclusive tools to complement the resources
