作者：邱东, 张楠, 赵晨旭, 戴文娟

作者单位：长春工业大学电气与电子工程学院, 吉林 长春 130012

关键词：转炉冶炼; 灰色预报; BP神经网络; 硫含量; 预报模型

摘要：

针对复吹转炉冶炼中低碳铬铁终点硫含量在线监测困难、取样检测无法实时指导生产的现状,考虑影响终点硫含量的供氧强度、铁水温度等因素,采用以灰色预报模型结合BP神经网络的方法实现中低碳铬铁终点硫含量的预报。仿真实验表明:中低碳铬铁终点硫含量预报绝对误差值在0.004%以内的命中率为95%,相对误差值在15%以内的命中率达到85%,验证了该预报模型的有效性。

Research on prediction model of medium-low carbon ferrochrome sulfur end-point content based on gray neural network

QIU Dong, ZHANG Nan, ZHAO Chen-xu, DAI Wen-juan

Department of Electrical and Electronic Engineering, Changchun University of Technology, Changchun 130012, China

Abstract: It was difficult to detect sulfur end-point content through on-line measurement in the medium-low carbon ferrochrome combined-blowing converter and hardly to guide production in real time through sampling and inspection. Considering the impaction of the sulfur endpoint content of oxygen intensity, temperature of hot metal and other factors and combining the grey model with BP neural network prediction model, the prediction of medium-low carbon ferrochrome sulfur endpoint content was realized. The simulation results indicated that the prediction hitting rate of the sulfur endpoint content in the low-carbon ferrochrome was 95% when the absolute error was less than 0.004%, and the heating rate was 85% when the relative error was less than 15%. Thus, the validity of the prediction model was proved.

Keywords: converter smelting; gray prediction; BP neural network; sulfur content; prediction model

2014, 40(4): 67-70  收稿日期: 2013-8-5;收到修改稿日期: 2013-10-11

基金项目: 吉林省科技发展计划项目(20120420)

作者简介: 邱东(1969-),男,吉林长春市人,副教授,博士,研究方向为智能测试技术。

参考文献

[1] 贾振海. 氩氧炉生产中低微碳铬铁的新工艺[J]. 铁合金,2005(2):11-16.
[2] Amlna D, Mavoori H, Perm K. Adaptive neural net models for desulfurization of hot metal and steel[J]. Steel Research,1994,65(11):466-471.
[3] Zhan D P, Zhang H S, Jiang Z H. Prediction model of magnesium powder consumption during hot metal pre-desulfurization[C]//Magnesium Technology Symposium,2009(2):51-55.
[4] 但斌斌,陈奎生,张华,等. 基于神经网络的铁水KR脱硫预报模型[J]. 计算机应用与软件,2011,28(1):118-120.
[5] 吕俊杰,游洪滨,田继成,等. 高碳铬铁冶炼脱硫的实践[J].铁合金,2005(1):22-25.
[6] Tathavadkarv S. Predicting the performance of submerged arc furnace with varied raw material combinations using artificial neural network[J]. Journal of Materials Processing Technology,2007(183):111-116.
[7] 胡敏艺,马荣骏. 神经网络在冶金工业中的应用[J]. 湖南有色金属,2000,16(9):16-19.
[8] 王炜,陈畏林,叶勇,等. 神经网络在高炉铁水硫含量预报中的应用[J]. 钢铁,2006,41(10):19-22.
[9] 李光辉,刘青. 转炉炼钢过程工艺控制的发展与展望[J].钢铁研究学报,2013,25(1):1-4.