Articles & Issues

Language: English

Conflict of Interest: In relation to this article, we declare that there is no conflict of interest.

Publication history: Received November 3, 2011
Accepted September 22, 2012

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/bync/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

All issues

Simulation and analysis of dehydration distillation column based on distillation mechanism integrated with neural network

Qingtao Sun Chunjian Pan Xuefeng Yan^†

Key Laboratory of Advanced Control and Optimization for Chemical Processes of Ministry of Education, East China University of Science and Technology, Shanghai 200237, P. R. China

Korean Journal of Chemical Engineering, March 2013, 30(3), 518-527(10), 10.1007/s11814-012-0163-4

Download PDF

Abstract

In an industrial solvent dehydration distillation column (SDDC) model, the Murphree efficiency represents the separation ability of a distillation tray and the SDDC model's performance depends on the value accuracy of the Murphree efficiency. Because there are many operation conditions having nonlinear effect on Murphree efficiency, it is difficult to determine its value. To develop a precise and robust SDDC model, a novel hybrid model combining distillation mechanism with neural network is proposed. In the SDDC hybrid model, the neural network is employed to model the nonlinear relationship between the operation conditions and Murphree efficiency, which is embedded into the SDDC mechanistic model. The results showed a good predicting and robust performance of the hybrid model under different operation conditions. Based on the hybrid model, the effect of the operation conditions on SDDC was analyzed to obtain some useful guiding rules for the SDDC operation.

Keywords

Hybrid Model Murphree Efficiency Neural Networks Dehydratic Distillation

References

Jang JY, Kim HJ, Wu K, US Patent, 20,030,150,706 (2003)
Lee MM, Kwon DJ, Lee DW, Lee AJ, US Patent, 20,070,027,340 (2007)
Othmer DF, Chem. Metall. Eng., 40, 91 (1941)
Chien IL, Zeng KL, Chao HY, Liu JH, Chem. Eng. Sci., 59(21), 4547 (2004)
Widagdo S, Seider WD, AIChE J., 42(1), 96 (1996)
Hornik K, Stinchombe M, White H, Neural Networks., 2, 359 (1989)
Russell E, Shi YH, Computational intelligence, Morgan Kaufman Publication, Massachusetts (2007)
Hopfield JJ, P Natl. Acad. Sci. USA., 79, 2554 (1982)
Cybenco G, Math. Control. Signal., 2, 303 (1989)
Girosi F, Poggio T, Biol. Cybern., 63, 169 (1990)
Rumelhart DE, Hinton GE, Williams RJ, Nature (London)., 323, 533 (1986)
Vogl TP, Mangis JK, Rigler AK, Zink WT, Alkon DL, Biol. Cybern., 59, 256 (1988)
Scales LE, Introduction to non-linear optimization, Springer-Verlag, New York (1985)
Abrams DS, Prausnitz JM, AIChE J., 21, 116 (1975)
Hayden JG, O’Connell JP, Ind. Eng. Chem., Process Des.Dev., 14, 209 (1975)
Aspen Technology, Inc., Aspen Plus®, http://www.aspentech.com/core/aspen-plus.aspx.
Yan XF, Chemometr. Intell. Lab. Syst., 103, 152 (2010)
MathWorks, Inc., Matlab®, http://www.mathworks.cn/products/matlab/.
Levenberg K, Q. Appl. Math., 2, 164 (1944)