Prediction of Energy Consumption in an Electric Arc Furnace Using Weka

Abstract

Industry 4.0 and digital transformation have managed to integrate technology into production processes with the aim of improving their automation levels. The purpose of this research was to identify some potentially useful and understandable patterns from the energy consumption data of an Electric Arc Furnace (EAF), starting from the hypothesis that the final energy consumption of an EAF depends on the different types of waste that are used to power the oven. The process was applied as part of the methodology Knowledge Discovery in Databases in order to collect, select and transform data. Then Weka software was used to discover predictive models and rules that were evaluated and interpreted to obtain knowledge. The methodological process to build the appropriate model that fits the collected data is described in this work, able to generate an effective prediction of energy used in steel production processes with an EAF. Through simulations, the prediction models were tested, and some conclusions were reached regarding the accuracy of the models. The results about the models are presented by means of a comparative table, in which the model M5P would have greater accuracy at the time of predicting the energy consumption for identifying which would be the optimal composition of the material to feed the furnace and therefore improve the efficiency of the metal melting process.