含铁炉料良好的冶金性能是保障高炉炉况顺行，节燃增产的前提条件。熔滴试验对于高炉含铁炉料冶金性能把握具有重要意义。但由于熔滴试验本身成本较高且检测耗时，多数企业仅在高炉炉况出现重大变化时通过试验追溯炉料冶金性能。因此，操作者难以根据炉料结构变化预判其冶金性能，进而调整操作方针。在对莱钢含铁炉料进行熔滴性能试验检测的基础上，通过遗传算法优化最小二乘支持向量机关键参数，以含铁料化学成分对其熔滴性能指标建立优化预测模型。模型预测精度高，且避免了建模过程中的主观性，可指导生产配料及调整高炉操作。
Abstract
Iron-bearing material with good metallurgical properties of is the guarantee of blast furnace operation. Based on melting and dropping (MD) tests for iron-bearing material in Laigang Group, a chemical composition on metallurgical performance indexes prediction model was built. Based on the model, key parameters of least square support vector machine (LS-SVM) were optimized used genetic algorithm (GA). The model has high prediction accuracy and avoids the subjectivity in the modeling process. It can be applied in blast furnace iron-making.
关键词
高炉 /
含铁炉料 /
熔滴性能 /
预测 /
遗传算法 /
最小二乘支持向量机
{{custom_keyword}} /
Key words
blast furnace /
iron-bearing material /
MD index /
prediction /
GA /
LS-SVM
{{custom_keyword}} /
中图分类号：
TF521
{{custom_clc.code}}
({{custom_clc.text}})
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 周传典.高炉炼铁生产技术手册[M].4版.北京：冶金工业出版社，2008.
[2] Paulo F. Nogueira ,Richard J. Fruehan. Blast furnace burden softening and melting phenomena: Part I. Pellet bulk interaction observation[J]. Metallurgical and materials transactions B,2004,35(5):829-838.
[3] 郑德玲，梁瑞鑫，付冬梅，等.人工免疫系统及人工免疫遗传算法在优化中的应用[J]. 北京科技大学学报，2003，25(3):284-288.
Zheng D L, Liang R X, Fu D M, et al. Application of artificial immune system and artificial immune genetic algorithm to optimization[J]. J Univ Sci Technol Beijing, 2003, 25(3): 284-288.
[4] Lu D G, Bai C G, Liang D. Scheduling Model for Section of Sintering-Blast Furnace[C].Symposium on Extraction and Processing Division held at the TMS 2009 Annual Meeting and Exhibition, San Francisco, America, 2009:475-478.
[5] Kao Y T, Erwie Z. A hybrid genetic algorithm and particle swarm optimization for multimodal functions[J]. Applied soft computing, 2008,8(2):849-857.
[6] 张军红，金永龙，沈峰满，等. 应用优化BP神经网络建立铁水硅含量的预测模型[J]. 钢铁研究学报，2007,19(11):60-64.
Zhang J H, Jin Y L, Shen F M, et al. Prediction model of silicon content in hot metal using optimized BP network[J]. Journal of iron and steel research, 2007,19(11):60.
[7] Kemal P, Salih G, Ahmet A A. A cascade learning system for classification of diabetes :generalized discriminant analysis and least square support machine[J]. Expert system and applications,2008,34(1):482-486.
[8] Stanislaw O, Konrad G. Forecasting of the daily meteorological pollution using wavelets and support vector machine[J]. Engineering Applications of artificial intelligence,2006(10):1-11.
[9] Liang D, Bai C G, Shi H Y, et al. Research on intellectual prediction for permeability index of blast furnace[C]. 1st WRI global congress on intelligent systems. Xiamen China, 2009:299-303.
[10] Nello C, John S T. An introduction to support vector machines and other kernel based learning methods[M]. 1st ED. Beijin: Metallurgical Industry Press,2005.
[11] Keiichi A, Takazo K, Yoshikazu N. Effect of test condition on decrepitation index and Test Repeatability for lump iron ore[J]. ISIJ International,2010, 50(10): 1511-1513.
[12] Chin E L, Leanne T M ,Damien P O. Lump ore and sinter behaviour during softing and melting[J]. ISIJ International, 2011,51(6): 930-938.
[13] Ritz V. J, Kortmann H.A , Koch K. Reduction, softening and melting properties of pellets, sinters, lumpy ore and mixed blast furnace burden[C].2nd International Congress on the Science and Technology of Ironmaking and 57th Ironmaking Conference, Toronto, Canada, 1998:1635-1640.
[14] Wu S L, Han H L, Xu H F，et al. Increasing lump ores proportion in blast furnace based on the high-temperature interactivity of iron bearing materials[J]. ISIJ International,2010, 50(5): 686-691.
[15] Rahamatulla J, Angamuthu A, Vijay K J，et al. Reduction and degradation behaviour of sinter under simulated vertical probe trial condition[J]. ISIJ International,2008, 48(7): 918-925.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
