Prediction Modeling of Hot Metal Silicon Content in Blast Furnace Based on PSO-LSSVM

Abstract:

The hot metal silicon content is important for the quality of the iron, but also as an indicator of the thermal state of the furnace. In order to stable operation of the blast furnace, a model is needed to predict hot metal silicon content more accurately. Towards this goal, a model based on least square support vector machine (LSSVM) is developed to model and predict hot metal silicon content, particle swarm optimization (PSO) is adopted to search the optimal set of the LSSVM model parameters. Prediction experiment is conducted based on the data obtained from a blast furnace past operation records in a steel tube plant and being preprocessed in several different ways. The experimental results show that the proposed methods yielded more accurate predictions than the neural network modeling methods.

• Periodical: Advanced Materials Research (Volume 721)
• Pages: 461-465
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.721.461
• Online since: July 2013
• Authors: Long Hui Wang, Song Gao, Xing Qu, Yao Geng Tang
• Keywords: Blast Furnace, Hot Metal Silicon Content, Least Squares Support Vector Machines (Ls-SVM), Particle Swarm Optimization (PSO), Prediction Model

References

[1] Henrik Saxén, Frank Pettersson, Nonlinear prediction of the hot metal silicon content in the blast furnace, SIJ international, 2007,47(12):1732-1737.

DOI: 10.2355/isijinternational.47.1732

[2] H Saxen, Short-term prediction of silicon content in pig iron, Canadian Metallurgical Quarterly, 1994 ,33(4): 319-326.

DOI: 10.1179/cmq.1994.33.4.319

[3] Matias Waller and Henrik Saxén, Time-varying event-internal trends in predictive modelingmethods with applications to ladlewise analyses of hot metal silicon content, Ind. Eng. Chem. Res., 2003, 42 (1): 85-90.

DOI: 10.1021/ie0201718

[4] Abhay B, Bulsari and Henrik Saxén, Artificial neural networks for predicting silicon content in raw iron from blast furnaces, Advances in Computing and InformationA- ICCI '91 Lecture Notes in Computer Science, 1991, 497: 641-644.

DOI: 10.1007/3-540-54029-6_212

[5] Singh, Himanshu and Sridhar, Nallamali and Deo, Artificial neural nets for prediction of silicon content of blast furnace hot metal, Steel Research, 1996. 67(12) :521-527.

DOI: 10.1002/srin.199605531

[6] Zhao Min, Liu Xiang-guan and Luo Shi-hua, An evolutionary artificial neural networks approach for BF hot metal silicon content prediction, Advances in Natural Discovery, Berlin Heidelberg, Springer-Verlag, 2005 :374-377.

DOI: 10.1007/11539087_46

[7] Vapnik V N, Statistical Learning Theory, John Wiley, New York, 1998.

[8] SUYKENS J A K and VANDEWALLE J, Recurrent least squares support vector machines, Transactions on Circuits and Systems, 2000, I.47:1109-1114.

DOI: 10.1109/81.855471

[9] Kennedy J and Eberhart R C. Particle Swarm Optimization Proc, IEEE Int. Conf. on Neural Networks, Perth: WA, 1995: 1942-1948.