Modelling a Skin-Pass Rolling Process by Means of Data Mining Techniques and Finite Element Method

An experience is presented using the finite element method (FEM) and data mining (DM) techniques to develop models that can be used to optimize the skin-pass rolling process based on its operating conditions. A FE model based on a real skin-pass process is built and validated. Based on this model, a group of FE models is simulated with different adjustment parameters and with different materials for the sheet; both variables are chosen from pre-set ranges. From all FE model simulations, a database is generated; this database is made up of the above mentioned adjustment parameters, sheet properties and the variables of the process arising from the simulation of the model, Various types of data mining algorithms are used to develop predictive models for each of the variables of the process. The best predictive models can be used to predict experimentally hard-to-measure variables (internal stresses, internal strains, etc.) which are useful in the optimal design of the process or to be applied in real time control systems of a skin-pass process in-plant.