Development of the service for calculating the material and thermal balance of steel smelting

Authors:

Abstract
The literature review highlights innovative approaches to modeling: the use of kinetic equations, computational fluid dynamics (CFD) and machine learning algorithms (XGBoost, Random Forest) to predict carbon and oxygen content. Special attention is paid to adapting models to new environmental technologies, in particular the use of hydrogen and scrap metal processing, the share of which in the charge can reach 40%. The experience of world leaders (Danieli Corus, Kuttner) and domestic scientific schools (NMetAU) in creating decision support systems, such as the «Technologist-Operator Advisor», is separately considered. The own development of the Department of Metallurgical Technologies of DUET and «Metinvest Polytechnics» is presented – a universal service for calculating smelting balances. The software product is built on the basis of a modern technology stack (HTML, CSS, JavaScript) using the Node.js platform for the server part. This choice of tools provides cross-platform compatibility, easy integration with existing industrial systems, and the possibility of remote access, which is critically important for training engineers in conditions of pandemic restrictions. The service is implemented in two modes: «light» (for educational purposes) and "industrial" (with full consideration of all elements of the charge). The program allows you to quickly enter data on the composition of pig iron, scrap, ferroalloys and receive final balance tables with an error of no more than 0.1-0.5%. The authors provide examples of the program code and the system interface, demonstrating the mechanisms for checking the chemical composition of the charge and generating final reports

This publication has 13 references indexed in Scilit: