Optimizing EAF/IF production - a concern for the entire company: What can we learn from development of digital twins

The last decades the development of computer science together with applied mathematics has given industry completely new tools. Artificial Intelligence, AI, has matured into among other things machine learning, Internet of things lets equipment talk to each other and cooperate, cloud computing makes it possible to build software in modules where resources are requested over the net only when needed. The systematic use of a combination of these new opportunities has been named the 4:th industrial revolution , Industry 4.0 or I4 . In traditional process or business control, experts have an idea of the status of processes or artefacts using mental models fed with data in both systematic and random ways. Automation systems may monitor and control a process but are often single function systems. In I4 some of these models are extended or replaced by multi-purpose digital models called Digital Twins, DT. A DT for metal consists of data and executable models put together in order to mimic relevant aspects of the metal for certain purposes. This paper is based on a pre-study, Omnimetal, performed at RISE SWECAST in Jönköping in spring 2019. The project aims at suggesting a data model for a metal digital twin, DT, to be used at a metal producer or foundry. The aim is to investigate where and when a DT can be used in the metals industry and what questions it may answer. In a digitalization project one of the first steps would be to identify how functions are performed The main method has been to describe the production system from different perspectives with regard to modelling and digitalization with the focus on the metal itself. There are models of furnaces or entire processes but in this case, these form an environment that can affect the metal. The study is further restricted to scrap-based iron, steel and metal production to feed a foundry or other metal working facilities.

کلیدواژه ها:

EAF ، IF ، Optimization ، Digital twin ، Model coherence ، Business Process