针对传统高炉温度模型的固有缺陷,提出了一种基于灰色关联分析的ELM(极限学习机—extreme learning machine)温度预报模型。由于炼铁工艺的多变量、非线性、强耦合等特点,所以传统建模方法已经不能满足要求的高精度预报高炉温度。首先通过灰色关联分析对输入变量进行相关性分析,提高模型的性能,然后结合分析后的变量采用ELM学习算法训练神经网络,最后运用现场数据对该网络进行训练和测试,并与传统的模型相比较。结果表明该模型能快速、准确地预报高炉温度,并且能满足指导现场工人操纵高炉的要求。
Abstract
Aiming at the inherent defects of the traditional blast furnace temperature model, a kind of grey relational analysis based ELM (extreme learning machine) temperature prediction model was put forward. Due to the characteristics of ironmaking process with multivariable nonlinear, strong coupling, the traditional modeling methods were unable to meet the requirements of high precision prediction of blast furnace temperature. Firstly, the correlation of input variables was analyzed with the gray correlation analysis, and then the performance of the model was improved. Secondly, combined with analytical variables, the neural network was trained by ELM learning algorithm. Finally, the field data was used for training and testing of the network, and then compared with the traditional model. The results show that the model can predict the blast furnace temperature quickly and accurately, and also can meet the guide workers to manipulate the needs of blast furnace.
关键词
灰色关联 /
极限学习机 /
高炉 /
铁水温度预报 /
神经网络
{{custom_keyword}} /
Key words
Grey correlation /
extreme learning machine /
blast furnace /
temperature prediction of molten iron /
neural network
{{custom_keyword}} /
中图分类号:
TP183
{{custom_clc.code}}
({{custom_clc.text}})
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 桂卫华,阳春华,陈晓芳.复杂有色冶金生产过程智能建模、控制与优化[M].北京:科学出版社,2010:80-81.
[2] 崔桂梅,李静,张勇.基于T-S模糊神经网络模型的高炉铁水温度建模[J].钢铁研究学报,2013(11):12-16
[3] 崔桂梅,孙彤,李仲德.支持向量机在高炉铁水温度预测中的应用[J].控制工程,2013(5):809-813
[4]张俊明, 刘军, 康永林等.应用 神经网络预测冷连轧机轧制力[J].钢铁, 2007, 42(8):46-48
[5]陈治明, 罗飞, 曹建忠.基于小波分析的多神经网络轧制力设定模型[J].华南理工大学学报自然科学版, 2010, 38(2):142-147
[6]Huang G B, Zhu Q Y, Mao K Z, et al.Can threshold networks be trained directly?[J].IEEE Transactions on Circuits and Systems, 2006, 53(3):187-191
[7] Huang G B, Zhu Q Y, Siew C K.Extreme learning machine: a new learning scheme of feed forward neural networks[C]//Proceedings of International Joint Conference on Neural Networks (IJCNN2004). Piscataway: Institute of Electrical and Electronics Engineers Inc, 2004: 985-990.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
基金
基于数据驱动的白云鄂博矿高炉炼铁过程的建模及优化研究
{{custom_fund}}