Abstract
Approach to the solution of a problem of structural and parametrical identification of models of the soft sensors (SS) of technological plants on the basis of robust regression and information criteria is proposed. The robust regression is used for model parameter estimation, and choosing the best model structure in the sense of information criteria. SS is developed by means of the proposed approach which was tested in control systems for optimization of the process operation of gas separation section of fluid catalytic cracking unit of “OJSC Gazpromneft-Omsk Refinery.”
Similar content being viewed by others
References
Vlasov, S.S. and Shumihin, A.G., Simulation of the Oil Topping Process for Gasoline Quality Prediction, Vestn. SGTU, 2012, vol. 63, no. 1, pp. 90–94.
Olanrewaju, M.J., Huang, B., and Afacan, A., Online Composition Estimation and Experiment Validation of Distillation Processes with Switching Dynamics, Chemical Eng. Sci., 2010, vol. 65, no. 5, pp. 1597–1608.
Chatterjee, T. and Saraf, D.N., On-line Estimation of Product Properties for Crude Distillation Units, J. Process Control, 2004, vol. 14, pp. 61–77.
Holland, P.W. and Welsch, R.E., Robust Regression Using Iteratively Reweighted Least-Squares, Commun. Statis., Theor. Meth., 1977, vol. 6, no. 9, pp. 813–827.
Rawlings, J.O., Pantula, S.G., and Dickey, D.A., Applied Regression Analysis: A Research Tool, New York: Springer-Verlag, 1998.
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © G.B. Digo, N.B. Digo, A.V. Kozlov, S.A. Samotylova, A.Yu. Torgashov, 2015, published in Avtomatizatsiya v Promyshlennosti, 2015, No. 10, pp. 58–62.
Rights and permissions
About this article
Cite this article
Digo, G.B., Digo, N.B., Kozlov, A.V. et al. Structural and parametric identification of soft sensors models for process plants based on robust regression and information criteria. Autom Remote Control 78, 724–731 (2017). https://doi.org/10.1134/S0005117917040130
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0005117917040130