Generalized Predictive Control of Flatness Using BP Neural Network

摘要

Flatness control in rolling process is intractable due to the multivariate, nonlinearity, time-variation and coupling of the process. In this paper, a generalized predictive control scheme using a feedforward neural network as the predictive model is introduced to control the strip flatness on a reversal UC (Universal Crown-Control) mill. Results of some comparison experiments are given to verify the validity of the algorithm. Experiment results show that the resulting control algorithm performs asatisfactory control behavior.