Premium
Optimisation of edging amount for high accuracy plan view control in plate mill
Author(s) -
Chun MyungSik,
Moon YoungHoon
Publication year - 2001
Publication title -
steel research
Language(s) - English
Resource type - Journals
eISSN - 1869-344X
pISSN - 0177-4832
DOI - 10.1002/srin.200100075
Subject(s) - mill , plan (archaeology) , engineering , control (management) , artificial neural network , standard deviation , computer science , mathematics , mechanical engineering , artificial intelligence , statistics , geology , paleontology
Algorithm to optimise the edging amount for high accuracy plan view control has been developed to improve plate yield. To produce rectangular shape of rolled plate, the edging amount that can minimise the crop length and width deviation must be determined from the operational analysis of plate mill. For the plan view control, new mathematical models to predict crop length at the top and bottom ends and width deviation at side ends are developed on the basis of statistics and back propagation learning algorithm of neural network. Using new mathematical models, under given rolling conditions of broadside and finishing pass and the permissible tolerance in width deviation, optimised algorithm which can achieve high accuracy plan view pattern is proposed. The application of newly developed algorithm in actual plate mill for plan view pattern control reduced width deviation by 20%, and crop length by 30% on average with a suitable fishtail crop shape.