2019 Pittsburgh AISTech

Sessions Speakers Exhibitors Products Press Releases Show Specials Floor Plan Buzz

Data-Driven Dynamic Feedforward Compensation Method for Coiling Temperature Control in Hot Strip Mills (Room 306)

A data-driven feedforward compensation method based on system identification was applied on coiling temperature control in a hot strip mill. System identification is a statistical method to build mathematical models of dynamic systems from measured input-output data, and it is expected to be one of the most useful methods of modeling for big data. By using system identification, the dynamics of coiling temperature fluctuation in bar was predicted on-line, and it was compensated for by feedforward control. As a result, the accuracy of coiling temperature control was successfully improved. This presentation will explain the basic idea and actual results in the paper.