06.基于高炉炉喉煤气分布的煤气利用率预测模型

石琳刘文磊李江鹏赵娜

(1.内蒙古科技大学数理与生物工程学院内蒙古包头 0140102.内蒙古科技大学材料与冶金学院内蒙古包头 014010

关键词：煤气利用率；煤气流中心分布；Elman神经网络

中图分类号：TF57     文献标识码：A

摘要：本文以煤气利用率为能耗评价指标，利用在线煤气流分布的红外监控图像预测高炉煤气利用率。首先，运用红外图像处理的高炉煤气流中心分布特征识别方法对红外监控图像进行识别以及量化统计。其次，根据煤气流中心分布特征及对应的煤气利用率，建立了Elman神经网络预测模型，与BP神经网络模型和径向基神经网络进行比较,预测结果显示Elman神经网络模型命中率为90.2%，BP神经网络命中率80.5%，径向基神经网络命中率只有46.3%。因此，运用实时煤气流分布特征数据，采用Elman神经网络模型预测煤气利用率，为高炉的在线控制和节能减排提供了有效的方法。

Prediction Model of Gas Utilization Rate Based on Blast Furnace Throat Gas Distribution

SHI Lin LIU Wen-LeiLI Jiang-PengZHAO Na

(1.School of Mathematics, Physics and Biological Engineering, University of Science and Technology of Inner Mongolia, Baotou, 014010, Inner Mongolia, China; 2.School of Materials and Metallurgy, University of Science and Technology of Inner Mongolia, Baotou, 014010, Inner Mongolia, China;)

Key words: gas utilization rate; gas flow center distribution; Elman neural network

Abstract: The online infrared monitoring image of gas flow distribution was used to predict the gas utilization rate of blast furnace, which is studied as the energy consumption index . Firstly, based on the center distribution of blast furnace gas flow of the infrared image processing ,the feature identification method was adopted to identify the infrared monitoring image and its statistical analysis was carried out .Next, according to the center distribution of the gas flow and corresponding gas utilization, Elman neural network prediction model was established, which was compared with the BP neural network model and RBF neural network model. The prediction results show that the hit rate of Elman neural network model ,BP neural network model，and RBF neural network model is 90.2%, 80.5% and 46.3%,respectively. Therefore, using the real-time characteristic data of the gas flow distribution, the Elman neural network model is applied to predict the gas utilization, providing an effective method for the online blast furnace control and emission reduction.