Prediction Model of Steel Mechanical Properties Based on Integrated KPLS
In this paper, an integrated KPLS (Kernel Partial Least Square) prediction model for steel mechanical property is proposed. To eliminate the heterogeneity among variables in the hot rolling process, the KFA (Kernel Factor Analysis) is used to obtain the latent factor load vectors. Then the variables with large factor load were clustered into subsets, and the KPLS components are extracted respectively for each subset variable and target variable. Finally, the KPLS results of all subsets were integrated as input, and an integral KPLS prediction model is constructed with the target variables. An application study was carried out on the real production data of a steel-making plant. The experimental result shows that the precision of the presented method is greatly improved.
KeywordsKernel factor analysis Kernel partial least square Steel mechanical property
This research work was supported by the National Natural Science Foundation of China (Grant No. 61572073), National Key R&D Program of China (NO. 2017YFB0306403) and the Fundamental Research Funds for the China Central Universities of USTB (FRF-BD-17-002A).
- 2.Y. Zongshen, Y. Zexi, L. Shiqi, et al. Quantitative relationship of composition, residual elements and Properties of Steel (Metallurgical Industry Press, 2001)Google Scholar
- 4.T. Jiali, C. Qiuru, L. Yijun, Prediction of Material Mechanical Properties with Support Vector Machine, in International Conference on Machine Vision and Human-Machine Interface (IEEE, 2010), pp. 592–595Google Scholar
- 5.R. Rosipal, Kernel partial least squares for nonlinear regression and discrimination. Neural Netw. World 13(3), 291–300 (2002)Google Scholar
- 6.R. Rosipal, L.J. Trejo, Kernel partial least squares regression in reproducing kernel hilbert space. JMLR.org (2002)Google Scholar
- 10.X. Guoen, S. Peiji, Factor analysis algorithm with Mercer Kernel, in International Symposium on Intelligent Information Technology and Security Informatics (IEEE, 2009), pp. 202–205Google Scholar
- 11.W. Ling, F. Dongmei, L. Qing, Samples selection based on SVR for prediction of steel mechanical property, in International Conference on Intelligent System Design and Engineering Application (IEEE Computer Society, 2012), pp. 909–912Google Scholar