Research and Application of Improved Support Vector Machine in Temperature Control of Heat Treatment Furnace

Abstract:

According to the process characteristics of the medium thickness steel plate temperature control in the heat treatment furnace, a new steel temperature predictive model of heat treatment furnace based on particle swarm optimization algorithm and genetic algorithm to optimize the parameters of regression support vector machine (PSO-SVR and GA-SVR) is established. Based on the new model, the design steps are given. The very good forecast effect is obtained when on-site production process data are taken as the training samples to train the model, and then the data samples of model test are selected to simulate it. The comparison of the PSO-SVR and GA-SVR forecast results indicates that GA-SVR is able to obtain better regression result and stronger predictive ability.

• Periodical: Applied Mechanics and Materials (Volumes 380-384)
• Pages: 430-434
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.380-384.430
• Online since: August 2013
• Authors: Jing Li, Xin Jian Wang, Sen Liu
• Keywords: Ga-SVR, Parameter Optimization, Predictive Model, PSO-SVR, Regression Support Vector Machine, Steel Plate Temperature

References

[1] G. Yuan, G.D. Wang, R.Q. Wang, et al, Heat treatment technology of the medium thickness steel plate and equipment development., Journal of Iron and Steel Research, 2009, 21(5): 1-7.

[2] J.G. Wei. Application and analysis in the heat treatment furnace., Wide and Heavy Plate, 2008, 14(3): 16-19.

[3] Z.W. Shao, Development status and prospect of the medium thickness heat treatment technology and equipment, Shandong Metallurgy, 2006, 28(3): 11-15.

[4] D.S. Zhu, Analysis and measures of the factor improving production capacity of heat treatment furnace, NanSteel Science Technology and Management, 2009(2): 32-34.

[5] J. Niu, Z. Wen, J.S. Wang, et al, Mathematical model optimization control simulation system of the continuous roller-hearth heat treatment furnace, Energy for Metallurgical Industry, 2007, 26(2): 15-19.

[6] L.P. Shan, J. Wang, J. Li, Research and Application of the roller hearth heat treatment furnace steel temperature prediction model, Journal of Iron and Steel Research, 2008, 20(4): 30-33.

[7] S.F. Tian, H.K. Huang, A simplified algorithm of support vector machine regression, Journal of software, 2002, 13（6）：1169-1172.

[8] T. Xie, J. Li, Predictive model of steel plate temperature in heat treatment furnace based on LS-SVM, Science Technology Review2012，30（08）: 41-44.

[9] L. Shi, Research on function fitting based on least squares support vector machine, Computer knowledge and Technology (academic exchange), 2007, 4(20): 440-441.

[10] Cortes C，V. Vapnik, Support Vector Networks, Machine Learning，1995，20：273—293.

[11] O. Chapelle，V Vapnik et al, Choosing multiple parameters for support vector machines, Machine Learning，46(2002)，131-160.

[12] X.B. Wang, X.P. Zhang, H.H. Wang, Optimization and application of support vector machine parameter based on particle swarm optimization pattern search, Computer application, 2011, 12 ( 31): 3302-3304.

[13] J. Li, Research on parameter optimization of support vector machine, Master degree thesis. Huazhong Normal University, 2011. swarm optimization pattern search. Computer application, 2011, 12 ( 31): 3302-3304.