Abstract
The endpoint parameters are very important to the process of EAF steel-making, but their on-line measurement is difficult. The soft sensor technology is widely used for the prediction of endpoint parameters. Based on the analysis of the smelting process of EAF and the advantages of support vector machines, a soft sensor model for predicting the endpoint parameters was built using multiple support vector machines (MSVM). In this model, the input space was divided by subtractive clustering and a sub-model based on LSSVM was built in each sub-space. To decrease the correlation among the sub-models and to improve the accuracy and robustness of the model, the submodels were combined by Principal Components Regression. The accuracy of the soft sensor model is perfectly improved. The simulation result demonstrates the practicability and efficiency of the MSVM model for the endpoint prediction of EAF.
Similar content being viewed by others
References
Morales R D. A Mathematical Model for the Reduction Kinetics of Iron Oxide in Electric Furnace Slags by Graphite Injection [J]. ISIJ Int, 1997, 37(11): 1072–1080.
Bekker J G. Modeling and Simulation of Electric Arc Furnace [J]. ISIJ Int, 1999, 39(1): 23–32.
ZHANG Jun-jie. Terminal Adaptive Predication and Expert Directing Operation for the Steelmaking Process of Electric Arc Furnace [J]. Acta Automatic Sinica, 1993, 19(4): 463–467.
LIU Kun, LIU Liu, HE Ping, et al. Application of Increment Artificial Neural Network Model to Prediction of Endpoint Carbon, Phosphorus and Temperature for an 100 t EAF Steel Making [J]. Special Steel, 2004, 25(3): 40–43 (in Chinese).
ZHANG Xue-gong. Introduction to Statistical Learning Theory and Support Vector Machines [J]. Acta Automatica Sinica, 2000, 26(1): 34–42 (in Chinese).
WANG Ding-cheng, FANG Ting-jian, GAO Li-fu, et al. Support Vector Machines Regression on-Line Modeling and Its Application [J]. Control and Decision, 2003, 18(1): 89–95 (in Chinese).
Dasaratha VS, Richard CS. EricBB Process Modeling Using Stacked Neural Networks [J]. AIChE Journal, 1996, 42(9), 89–95.
Cho S B, Kim J H. Combining Multiple Neural Networks by Fuzzy Integral for Recognition [J]. IEEE Transon System, Man and Cybern, 1995, 25(2): 380–384.
WANG Xu-dong, SHAO Hui-he, LUO Rong-fu. The Distributed RBF Neural Network and Its Application in Soft Sensor [J]. Control Theory and Applications, 1998, 15(4): 558–563 (in Chinese).
XIONG Zhi-hua, WANG Xiong, XU Yong-mao. Nonlinear Software Sensor Modeling Using Mulitple Neural Network [J]. Control and Decision, 2000, 15(2): 173–187 (in Chinese).
CHANG Yu-qing, WANG Fu-li. Distributed RBF Network Soft Sensor Model Based on Fuzzy Rule Classification [J]. ACTA Metrologica Sinica, 2002, 23 (2): 131–133 (in Chinese).
Chiu S L. Fuzzy Model Identification Based on Cluster Estimation [J]. Journal of Intelligent Fuzzy Systems, 1994, 2(3): 267–278.
Suykens J A K, Vandewalle J. Least Squares Support Vector Machines Classifiers [J]. Neural Processing Letters, 1999, 19(3): 293–300.
YAN Wei-wu, ZHU Hong-da, SHAO He-hui. Soft Sensor Modeling Based on Support Vector Machines [J]. Journal of Systemsimulation, 2003, 15(10): 1494–1496 (in Chinese).
Demirli K, Cheng S X, Muthukumaran P. Subtractive Clustering Based Modeling of Job Sequencing With Parametric Search [J]. Fuzzy Sets and Systems, 2003, 137(2): 235–270.
YAN Hui, ZHANG Xue-gong, MA Yun-qian, et al. The Parameter Estimation of RBF Kernel Function Based on Vario-gram [J]. Acta Automation Sinica, 2002, 28(3): 450–455 (in Chinese).
Author information
Authors and Affiliations
Corresponding author
Additional information
Foundation Item: Item Sponsored by National Natural Science Foundation of China (60374003)
Rights and permissions
About this article
Cite this article
Yuan, P., Mao, Zz. & Wang, Fl. Endpoint Prediction of EAF Based on Multiple Support Vector Machines. J. Iron Steel Res. Int. 14, 20–24 (2007). https://doi.org/10.1016/S1006-706X(07)60021-1
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1016/S1006-706X(07)60021-1