Gas utilization ratio prediction of blast furnace based on intelligent model
ZHAO Jun1,2, LI Hong-yang3, LI Xin3, LIU Xiao-jie3, LI Hong-wei3, LÜ Qing1,3
1. College of Metallurgy, Northeastern University, Shenyang 110819, Liaoning, China; 2. Tangshan Branch of HBIS Group Co., Ltd., Tangshan 063020, Hebei, China; 3. College of Metallurgy and Energy, North China University of Technology, Tangshan 063210, Hebei, China
Abstract:Gas utilization ratio is a key parameter that affects the energy utilization of blast furnace, and it has an important relationship with the stable running of blast furnace. After time interval analysis of production parameters of blast furnace, several simple prediction models (adaptive boosting, random forest, neural network) are used to predict gas utilization rate. Firstly, the historical data of blast furnace production is collected and sorted. Secondly, the pretreatment of characteristic parameters is carried out considering the abnormal status of blast furnace production (such as overhaul and wind rest). Finally, using time-delay relationship between parameters of blast furnace, gas utilization ratio prediction model is set up. By parameter adjustion and optimization and evaluation of prediction model, gas utilization ratio prediction model is determined, to realize gas utilization ratio prediction in advance. At the same time, the same model parameters are used to predict the original data set and the pretreatment data set, and the results are compared. The results show that the data pretreatment has a great influence on prediction results, and the random forest model has a better prediction effect on gas utilization rate, which can be used to guide the actual production of blast furnace.
赵军, 李宏扬, 李欣, 刘小杰, 李红玮, 吕庆. 基于智能模型的高炉煤气利用率预测[J]. 中国冶金, 2021, 31(3): 93-100.
ZHAO Jun, LI Hong-yang, LI Xin, LIU Xiao-jie, LI Hong-wei, LÜ Qing. Gas utilization ratio prediction of blast furnace based on intelligent model[J]. China Metallurgy, 2021, 31(3): 93-100.