Regular Article

Dynamic tracking prediction control of exit strip thickness based on improved fractal

¹ Institute of Advanced Control Technology, Dalian University of Technology, Dalian 116024, PR China
² School of Information, Liaoning University, Shenyang 110036, PR China

^* Corresponding author: zhang_li@lnu.edu.cn

Received: 13 October 2016
Received in final form: 7 April 2017
Accepted: 9 May 2017

Abstract

The dynamic tracking prediction of the exit strip thickness (EST) has important contribution to solve the problem of low strip thickness accuracy caused by the delay of the monitor AGC in the hot rolling mill. Fractal extrapolation-interpolation method is presented to apply for EST prediction. The key problem to be solved is determining the vertical scaling factors in fractal interpolation prediction, the vertical scaling factor optimization function is established, which is designed by three items containing of the interpolation error, model tracking performance and the constraint condition of the vertical scaling factors in the fractal theory. In addition, chaos optimization algorithm is used to solve the optimal vertical scaling factors. The optimal fractal method is embedded in the traditional monitor AGC, which forms a new prediction control method of EST (denoted as IFEIP-AGC). Compared with traditional monitor AGC and Smith-AGC, the simulation results verify that IFEIP-AGC method has better dynamic tracking performance and robustness.