A mathematical model of a cold rolling mill by symbolic regression alpha-beta

Improvement of processes in metallurgical industry is a constant of competitive enterprises, however, changes made in a process are risky and involves high cost and time, considering this, a model can be made even using inputs usually not presented in real process and its analysis could be useful for the improvement of the process. In this work, a mathematical model is built using only experimental data of a four high tandem cold rolling mill, a set of input variables involving characteristics of the process. The performance of the model is determined by residual analysis considering new data. Results are a non black box model with a good performance; by this way, the model is a good representation of the process under study.