Intelligent judgment of copper end-point in on-line converter for copper smelting
HU Jin-bao1, YU Xiang-yang2
1. Research and Development Department, LIWODE Technology Co., Ltd., Nanchang 330000, Jiangxi, China; 2. School of Physics, State Key Laboratory of Optoelectronic Materials and Technologies, Sun Yat-sen University, Guangzhou 510275, Guangdong, China
Abstract:The copper smelting converter blowing is a complex process with multi-variable, nonlinear, strong coupling, great inertia and uncertainty, complex mechanism, a wide range of material changes and many influencing factors, which brings great difficulty to the prediction of blowing endpoint. At present, the end-point judgment of copper-matte converting process at home and abroad is still dominated by manual experience judgment, which not only increases the working intensity, but also makes the end-point judgment of copper-matte converting heavily dependent on experience and working attitude, which may easily lead to under-blowing or over-blowing, affect the normal production, and cause copper loss and even furnace injection accidents in a serious accident. Based on the accurate correspondence between the end point of converter blowing and the content of SO2 and O2 in flue gas, combined with artificial experience and converter process principle, the end of copper smelting could realize the accurate judgment online. The relationship between the SO2 concentration and flue gas temperature, air supply volume, air supply pressure, oxygen-enriched volume, internal factors (sulfur ratio of raw material, the weight of raw material, quality of raw material, etc.) was used to realize dynamic compensation, Elman recursive neural network model was used to realize self-adjustment and self-learning, so that the accuracy of judgment was more than 98%, which was of great significance to guide actual production.
ZHANG Ren-hua,LI Zhao-liang,SUN Xiao-min,et al. On the applicability of Kirchoff′s law and the principle of heat balance in thermal infrared remote sensing:A non-isothermal system[J]. Science in China Series D:Earth Sciences,2005,48(1):53.
Smola A J,Kohlmorgen J,Vapnik V. Using support vector machines for time series prediction[J]. Chemometrics and Intelligent Laboratory Systems,2003,69(1/2):35.
HE Yuan,MENG Zhi-yi,XU Hong,et al. A dynamic model of evaluating differential automatic method for solving plane problems based on BP neural network algorithm[J]. Physica A:Statistical Mechanics and its Applications,2020,556:124845.