Journal of Computers, Vol 9, No 2 (2014), 396-403, Feb 2014
doi:10.4304/jcp.9.2.396-403

Variable Selection Method for Aluminum Electrolytic Process Based on FNN and RM in KPLS Feature Space

Lizhong Yao, Taifu Li, Jun Peng, Deyong Wu, Yingying Su, Jun Yi

Abstract


Selecting the process variables is an important prerequisite for establishing an accurate model of aluminum electrolytic process. A variable selection method is researched and proposed based on the False Nearest Neighbors (FNN) and Randomization Method (RM) (FR) in KPLS(Kernel Partial Least Squares) feature space. Firstly, the KPLS is employed to transform the original space to the PLS feature space; secondly, in the new feature space, the FNN is used to calculate the similarity measure of each variable which first is retained and then reset to zero for evaluateing the importance to the dependent variable; then, the RM is utilized to test the significance level of the importance for each variable in turn, so the redundant variables would be excluded; lastly, technical energy consumption model of aluminum electrolytic process is built to verify the presented method. The experimental results show that the method selects out the best process variables of aluminum electrolytic process. Therefore, the research provides a new method of the variable selection for metallurgical industrial processes.


Keywords


variable selection; aluminum electrolysis; false nearest neighbors(FNN);randomization method(RM); KPLS

References


[1] G. Huchet, M. Boussuge, V. Maurel, H. Roustan, "Mechanical properties of a spinel-ferrite-based cermet: effects of temperature and oxidation," Journal of Materials Science, Springer, vol.48,no.8,pp. 3264-3271, 2013
http://dx.doi.org/10.1007/s10853-012-7111-x
 
[2] C.-Y. Cheung, C. Menictas, J. Bao, M. Skyllas-Kazacos, and B. J. Welch, "Spatial temperature profiles in an aluminum reduction cell under different anode current distributions," American Institute of Chemical Engineers, Wiley-Blackwell, vol.59,no.5,pp. 1544–1556, 2013
http://dx.doi.org/10.1002/aic.13942
 
[3] M. Abbasi, M. A. Abduli, B. Omidvar, A. Baghvand, "Forecasting municipal solid waste generation by hybrid support vector machine and partial least square model," International Journal of Environmental Research, Univ Tehran,vol.7, no.1,pp. 27-38,2013.
 
[4] R. D. Pereira, J. Sousa, S. Vieira, S. Reti, S. Finkelstein, "Modified Sequential Forward Selection Applied to Predicting Septic Shock Outcome in the Intensive Care Unit," in Soft Methods in Probability and Statistics 2013, 2013 6th International Conference on,vol.190,2013,pp. 469-477
 
[5] K. Y. Chan, C. K. Kwong, T. S. Dillon, Y. C. Tsim, "Reducing overfitting in manufacturing process modeling using a backward elimination based genetic programming," Applied soft computing, Elsevier, vol.11, no.2, pp. 1648-1656, 2011.
 
[6] N. Zhou, J. W. Pierre, D. Trudnowski, "A Stepwise Regression Method for Estimating Dominant Electromechanical Modes," Power Systems, IEEE Transactons on, vol.27, no.2, pp. 1051-1059, 2012.
http://dx.doi.org/10.1109/TPWRS.2011.2172004
 
[7] F. Zhdanov, Y. Kalnishkan, "An identity for kernel ridge regression," Theoretical Computer Science, Elsevier, vol.473, no. SI, pp. 157-178, 2013.
 
[8] C. X. Wu, J. K. Zhu, D. Cai, C. Chen, J. J. Bu, "Semi-Supervised Nonlinear Hashing Using Bootstrap Sequential Projection Learning," Knowledge andData Enineering, IEEE Transactons on, vol.25, no.6, pp. 1380-1393, 2013.
 
[9] J. C. Boulet, D. Bertrand, G. Mazerolles, R. Sabatier, "A family of regression methods derived from standard PLSR," Chemometrics and Intelligent Laboratory Systems, Elsevier, vol.120, pp. 116-125, 2013.
http://dx.doi.org/10.1016/j.chemolab.2012.11.002
 
[10] J. Yi, T. F. Li, Y. Y. Su, W. J. Hu, T. Gao, "A variable selection method based on KPCA and FNN for nonlinear system modeling,"in BMEI, 2011 Proceedings IEEE, May 2011, pp. 832-835.
 
[11] J. B. Li, H. J. Gao, "Sparse data-dependent kernel principal component analysis based on least squares support vector machine for feature extraction and recognition," Neural Computing& Applications, Springer, vol.21, no.8, pp. 1971-1980, 2013.
http://dx.doi.org/10.1007/s00521-011-0600-z
 
[12] R. C. Zhang, L. H. Wang, S. H. Zhang, "Windows memory analysis based on KPCR," in Information Assurance and Security 2009, 2009 5th International Conference on, vol.2, 2009, pp. 677-680.
 
[13] J. S. Wang, Q. P. Guo, "D-FNN based soft-sensor modeling and migration reconfiguration of polymerizing process," Applied Soft Computing, Elsevier, vol.13, no.4, pp. 1892-1901, 2013.
http://dx.doi.org/10.1016/j.asoc.2012.12.018
 
[14] W. Suphamitmongko, G. L. Nie, R. Liu, "An alternative approach for the classification of orange varieties based on near infrared spectroscopy," Computers and Electronics in Agriculture, Elsevier, vol.91, no.2, pp. 87-93, 2013.
http://dx.doi.org/10.1016/j.compag.2012.11.014
 
[15] J. H. Li, P. L. Cui, "Improved kernel fisher discriminant analysis for fault diagnosis," Expert Systems with Applications, Elsevier, vol.36, no.2, pp. 1423-1432, 2009.
http://dx.doi.org/10.1016/j.eswa.2007.11.043
 
[16] Z. X. Li, X. P. Yan, C. Q. Yuan, "Intelligent fault diagnosis method for marine diesel engines using instantaneous angular speed," Journal of Mechanical Science and Technology, Korean Soc Mechanical Engineers, vol.26, no.8, pp. 2413-2423, 2012.
 
[17] H. D. Abarbane, R. Brown, J. J. Sidorowich, "The Analysis of Observed Chaotic Data in Physical Systems," Rev. Mod. Phy., American Physical Soc, vol.65, no.4, pp. 1331–1392, 1993.
 
[18] J. D. Olden, D. A. Jackson, "Illuminating the ''black box'': a randomization approach for understanding variable contributions in artificial neural networks," Ecological Modelling, Elsevier, vol.154, no.1-2, pp. 135-1502, 2002.
http://dx.doi.org/10.1016/S0304-3800(02)00064-9
 
[19] P. Susanne, K. Daniel, H. Thomas, "Significant amino acids in aroma compound profiling during yeast fermentation analyzed by PLS regression," Lwt-Food Science and Technology, Elsevier, vol.51, no.2, pp. 423-4322, 2013.
http://dx.doi.org/10.1016/j.lwt.2012.11.022
 
[20] N. Hadi, F. Abbas, N. Mehrab, "Application of GA-KPLS and L-M ANN calculations for the prediction of the capacity factor of hazardous psychoactive designer drugs," Medicinal Chemistry Research, Springer, vol.21, no.9, pp. 2680-2688, 2012.
http://dx.doi.org/10.1007/s00044-011-9794-y
 
[21] J. J. Wang, B. L. Chen, Y. C. Yang. Mehrab, "Approximation of algebraic and trigonometric polynomials by feedforward neural networks," Neural Computing& Applications, Springer, vol.21, no.1, pp. 73-80, 2012.
http://dx.doi.org/10.1007/s00521-011-0617-3
 


Full Text: PDF


Journal of Computers (JCP, ISSN 1796-203X)

Copyright @ 2006-2014 by ACADEMY PUBLISHER – All rights reserved.