Prediction of the cohesive zone in a blast furnace by integrating CFD and SVM modelling

Abstract

High efficiency and clean emission operation of a blast furnace are mainly determined by its operating stability, which is closely related to the position of the cohesive zone (CZ) in the blast furnace. In order to monitor the CZ position in real time, a prediction model of the CZ was proposed by combining an off-line computational fluid dynamics (CFD) calculation and support vector machine (SVM) modelling. An axisymmetric two-dimensional steady-state CFD model was established to simulate the fluid flow, heat and mass transfer in the blast furnace shaft. The sample dataset of the CZ information was established based on the CFD off-line calculation, and the prediction of the CZ position was realized by SVM modelling. The results show that the established model can predict the CZ position in real time with high accuracy, providing effective guidance to the blast furnace operation and control.