SVR-Based Predictive Model for Purity of the Mg-Al-Hydrotalcite

Abstract:

In this study, a prediction model based on support vector regression (SVR) to describing the complex nonlinear relationship between the purity of the resultant (Mg-Al-hydrotalcite) and the raw material amount of reactant (NaOH, Mg²⁺ and Al³⁺) was developed. And the partarticle swarm optimization (PSO) algorithm was employ for optimizing the parameters of SVR model.The predicton results from SVR compared with that of BP (back- propagation) neuarl network prediction model by applying identical training and test samples stongly support that, SVR-based predictive model for Mg-Al-hydrotalcite purity is feasible and effective.This study suggest that SVR may be an efficient and novel methodology for conduct the productive process to produce synthetic hydrotacite in high purity and in high yield.

• Periodical: Advanced Materials Research (Volumes 189-193)
• Pages: 1482-1485
• DOI: https://doi.org/10.4028/www.scientific.net/AMR.189-193.1482
• Online since: February 2011
• Authors: J.L. Tang, C.Z. Cai, X.J. Zhu, G.L. Wang
• Keywords: Mg-Al-Hydrotalcite, Modeling, Purity, Regression Analysis, Support Vector Machine (SVM)

References

[1] V. N. Vapnik: The Nature of Statistical Learning Theory (Springer, New York 1995).

[2] C.Z. Cai, L.Y. Han, Z.L. Ji, X. Chen and Y.Z. Chen: Nucleic Acids Research 31 (2003), pp.3692-3697.

[3] D. Mao, Z.Y. Zeng, C. Wang and W.H. Lin: Proceedings of Third International Conference on Natural Computation 1 (2007), pp.656-660.

[4] H.Z. Yang, X.G. Shao, G. Chen and F. Ding: Dyn. Cont. Discr. Imp. Syst. Ser. A 13(2006), pp.1417-1425.

[5] H. Yoshida, K. Kawata, Y. Fukuyama, S. Takayama and Y. Nakanishi: IEEE Trans. Power Syst. 15 (2000), pp.1232-1239.

DOI: 10.1109/59.898095

[6] Q. Luo, Q.L. Ren, L. Luo and H.J. He: International Symposium on Information Science and Engineering 2 (2008), pp.544-547.