Modelos para predição de blaine e finura no processo de moagem de cimento

Abstract

Artificial intelligence has been advancing in several areas, such as process automation, digital security, and everyday human tasks. In contrast to the traditional practices of cement production, technology is not only replacing manual processes but also assisting the entire production chain. Currently, in the cement industry, the measurement of cement fineness and Blaine is performed manually, with intervals of two hours between each measurement. In this context, the study of machine learning models can open the door to a future integration of artificial intelligence with the factory’s logic panels, since part of the operator’s decision-making is based on these two parameters. Therefore, this work proposes the study of machine learning models for the prediction of fineness and porosity. It was observed that some models achieved good interpretation of industrial data for predicting these two variables. It is expected that this research may support future integration of AI in the cement industry, making the process more modern and efficient.