Process analytics and machine learning to predict arc losss in an electric arc furnace

Published in Conference of Metallurgists - Under review, 2020

Recommended citation: Rippon, L. D., Yousef, I., Hosseini, B., Beaulieu, J. F., Prevost, C., Shah, S. L., & Gopaluni, R. B. (2020). "Process analytics and machine learning to predict arc losss in an electric arc furnace." Conference of Metallurgists. Under review. https://com.metsoc.org/

Abstract

Stable smelter operation is critical for successful production of base metals from particulate ore. This work studies the operation of an industrial direct current electric arc furnace that operates as a smelter in a large-scale metallurgical process. Specifically, unexpected loss of the plasma arc is an important unresolved problem with a significant impact on the production efficiency of the process. Moreover, given that electric arc furnaces are highly energy intensive units, even minimal improvements to the overall production efficiency represent meaningful reductions in the environmental footprint of the process over the lifetime of operation. To reduce the overall environmental footprint a predictive inferential sensor is proposed to identify high risk operating regimes. Once a high-risk situation is identified the alarm instructs operators to take corrective actions to avoid the loss of arc. Large amounts of historical industrial process data have been collected, pre-processed and leveraged in a cross-validated supervised learning framework that trains the inferential sensor model. New data, previously unseen by the model, are drawn from the historical database and used to test the ability of the model to generalize. This work showcases our progress to date including the training, validation and testing of competing inferential sensor models and their ability to predict arc loss on industrial data.

Download paper here

Recommended citation: Rippon, L. D., Yousef, I., Hosseini, B., Beaulieu, J. F., Prevost, C., Shah, S. L., & Gopaluni, R. B. (2020). “Process analytics and machine learning to predict arc losss in an electric arc furnace.” Conference of Metallurgists. Under review.