开发了较为完善的VD终点温度在线预报系统。采用MINITAB软件确定影响VD过程温降的主要因素为抽真空时间、保压时间、吹氩时间、非真空时间、VD搬入钢水的过热度、LF处理时间以及转炉出钢至VD初始测温之间的钢包运输时间。应用神经网络方法对VD处理终点的钢水温度进行在线预报,系统在线连续预报了95罐,预报温度与实际测量温度之差在±5℃范围内的比例达到93.7%。
The perfect VD end-point temperature on-line forecast system was established.The main factors that influenced the VD process temperature drop were found out by MINITAB software,such as pumping time,vacuum keeping time,argon-blowing time,normal time,VD original superheat degrees,LF treatment time and ladle transferring time from converter to the first temperature measurement of VD.The forecasting system was applied to forecast the VD end-point temperature on-line based on the neural network method.For continuous 95 heats,the forecast accuracy for the difference less than 5℃ between the forecasting temperature and measured temperature is up to 93.7%.
参考文献
| [1] | 氧化物沉积的特性和镇静钢凝固过程中的生长[J].ISIJ International,1998(1):46-52. |
| [2] | 左秀荣,姜茂发,薛向欣,张庆才,顾文俊.基于人工神经网络预报性能的齿轮钢LF-VD过程钢水成分微调[J].炼钢,2001(04):18-21. |
| [3] | 李亮,姜周华,王文忠,刘晓,顾文兵,徐荣军.VD炉终点钢液温度预报[J].钢铁,2003(01):16-18. |
| [4] | 赵国威,樊海峰,郭峻岭.神经元网络在VD温度预报模型中的应用[J].控制工程,2006(03):227-229. |
| [5] | 刘晓,徐荣军,顾文兵,杨宝权,王军.基于模式识别和神经网络的VD温度预报模型[C].2003中国钢铁年会论文集,2003:1787-1792. |
| [6] | Margrave F U;Algas K;Braldey D A .The Use of Neural Net-works in Ultrasonic Flaw Detection[J].Measurement,1999,25(02):143. |
| [7] | Andrew Alleyne;Michael Pomykalski .Control of a class of nonlinear systems subject to periodic exogenous signals[J].IEEE transactions on control systems technology: A publication of the IEEE Control Systems Society,2000(2):279-287. |
| [8] | Dobrzannski L A;Sitek W .The Modelling of Hardenability Using Neural Networks[J].Journal of Materials Processing Technology,2000,104(01):74. |
| [9] | 冯明霞,李强,邹宗树.转炉终点预测模型中异常数据检验的研究[J].中国冶金,2006(09):27-31. |
- 下载量()
- 访问量()
- 您的评分：
-
10%
-
20%
-
30%
-
40%
-
50%