Open AccessResearch on Real-Time Prediction of Quality of the Slag Overflowing From Furnace Door Based On Image Processing
Author(s) -
Xin Xie,
Lingzhi Yang,
Yufeng Guo,
Zeshi Yang,
Shuai Wang,
Jiamiao Ni
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1288/1/012058
Subject(s) - slag (welding) , steelmaking , electric arc furnace , metallurgy , scrap , environmental science , waste management , materials science , lime , raw material , engineering , chemistry , organic chemistry
As an important substance (slag) with the absorption capacity to the harmful elemental phosphorus, saturated slag demanded discharging for both the requirement of the generation of foaming slag in the furnace and slag replacement to strengthen the effect of dephosphorization. This paper reviewed during the electric arc furnace (EAF) steelmaking process for real-time quality of the slag overflowing from furnace door based on image processing. Existing methods, due to high temperature and the on-site production environment, were not considered to weigh real-time spilled slag and the total mass. The real-time composition and quality of the slag in the furnace, which affected the steelmaking process and the quality of products, were significantly influenced by addition of various metallurgical raw materials such as ladle, scrap and lime. The real-time oxidation products and additives could be calculated through the PLC data in the furnace to learn the composition and quality of the slag but it suffered the amount of slag continuously decreasing, due to real-time slag overflowing from the door whose composition was consistent with ingredient in the furnace at that time. The image processing slag from the door serves as a calculating means for the area of the big and small slag areas by the three primary colors of the light and the adjusted pixel channel value. Based on the comprehensive analysis of the situation of the slag from the door, the area of the big and small slag areas were analyzed and extracted to explore the relationship between the area and mass. And the accuracy of calculating the real-time slag was characterized by calculating the volatility of the total mass of the slag from the door counted by each heat. The relationship was verified to calculate the mass of that accuracy was up to 90.48%, which provided a scientific method for predicting the quality of real-time spilled door slag and a scientific basis for predicting in real time the composition and quality of slag in the furnace.