Optimizing integrated steelworks process off-gas distribution through Economic Hybrid Model Predictive Control and Echo State Networks

Steel production in integrated steelworks involves the simultaneous of various byproducts, including process off-gases that are usually exploited for generating electricity internal power plant, heat and steam. Their discontinuous is managed through complex network, gasholders torches, which must be with stringent operational constraints. In this paper we present a supervision control system designed to optimize economic management distribution also allows minimizing environmental impact. The implements digital twin based mainly on machine learning techniques, Echo State Networks, hierarchical optimization system, first level an model predictive approach second hybrid control. This effectively maximize use while impact their up 97%.