Application of the Artificial Neural Networks in Strip Shape Defect Identification

An intelligent pattern recognition used artificial neural networks is presented in this paper to meet the requirement of controlling stripe shape in cold rolling. The cold-rolled products are characterize into several types based on its irregularity, 'left wave', 'right wave', 'center buckle', 'edge wave', 'W-type', and 'M-type'. The developed identification algorithm calculates for each type of irregular strip shape using neural network and experiment data. The work is studied by taking a double-stand reversing cold rolling mill as example. The method improves the speed and the accuracy of strip shape identification.