NEURAL-NETWORK ANALYSIS OF STEEL PLATE PROCESSING

The process of rolling is very complicated and the number of parameter s which determines the final properties can be quite large. It is extr emely difficult therefore to develop a physical model for predicting v arious properties like yield and tensile strengths. In the present wor k, a neural network technique which can recognise complex relationship s was employed to develop a quantitative method for estimating the yie ld and tensile strengths as a function of steel composition and rollin g parameters. The model was tested extensively to confirm that the pre dictions are reasonable in the context of metallurgical principles and other data published in the literature. IS/1355. (C) 1998 The Institu te of Materials.