Information-modeling system for prediction of the composition and properties of final slag in a blast furnace in real time

Abstract

The article considers general characteristics of the algorithm for prediction of the composition of the final slag in a blast furnace in real time. This algorithm is based on fundamental knowledge on the processes occurring in the furnace and general laws of transient processes. It allows predicting at the current moment of time and for every hour ten hours ahead. A linearized model of the blast furnace process and a natural-mathematical approach are used. The model takes into account the dynamic characteristics of blast furnaces in various impact channels, which change and depend on the type of impact, operating parameters of the furnaces and properties of the melted raw material. This makes it possible to adjust the model to operating conditions of the object, to take into account changes in the composition and properties of iron ore and coke, blast and regime parameters of blast furnace smelting when modeling. The software of the information-modeling system for prediction of the composition and properties of the final slag in a blast furnace in real time was developed in the C# programming language based on the ASP.NET MVC framework using the .NET 5 cross-platform. The web application includes the following main functions: visualization of change APCS parameters and design parameters over time; slag mode diagnostics; modeling of transient processes of composition and properties of slag; prediction of slag composition and properties in real time and prediction history. The software architecture is described and its operation is illustrated. An assessment of the accuracy and reliability of the simulation results based on statistical indicators was carried out. The root-mean-square deviation of the predicted basicity of the CaO/SiO2 slag from that measured at taps is 0.023, the prediction reliability is 92 %, which indicates a satisfactory agreement between the predicted and actual values of the content of individual components in the slag. The information modeling system developed on the basis of the presented algorithm is integrated into the information system of the blast furnace shop of PJSC Magnitogorsk Iron and Steel Works.