Akihito Yamane

Synchronous Learning Algorithm for Time-Variant and Rolling Conditional Factor

Abstract A modelling error consists of two factors; one is a time-variant factor like the tools wear, and the other is a factor of rolling conditions like the material characteristic. If both factors are not differently identified when the modelling errors are learned and reflected in the adaptation of a setup control, controlling accuracy gets worse. This paper proposes the synchronous learning algorithm for both factors with learning gain scheduling to improve learning efficiency. The application simulation using hot-rolled width data shows that standard deviation for modelling error can be reduced 20% compared with the conventional method.