Abstract
Accurate prediction of chemical composition of vacuum gas oil (VGO) is essential for the routine operation of refineries. In this work, a new approach for auto-design of artificial neural networks (ANN) based on a genetic algorithm (GA) is developed for predicting VGO saturates. The number of neurons in the hidden layer, the momentum and the learning rates are determined by using the genetic algorithm. The inputs for the artificial neural networks model are five physical properties, namely, average boiling point, density, molecular weight, viscosity and refractive index. It is verified that the genetic algorithm could find the optimal structural parameters and training parameters of ANN. In addition, an artificial neural networks model based on a genetic algorithm was tested and the results indicated that the VGO saturates can be efficiently predicted. Compared with conventional artificial neural networks models, this approach can improve the prediction accuracy.
Article PDF
Similar content being viewed by others
References
El-Hadi D and Bezzina M. Improved empirical correlation for petroleum fraction composition quantitative prediction. Fuel. 2005. 84(5): 611–617
Funahashi K. On the approximate realization of continuous mappings by neural networks. Neural Networks. 1989. 2(3): 183–192
Goldberg D E. Genetic Algorithms in Search, Optimization and Machine Learning. Boston, MA: Addison-Wesley Longman Publishing Co. Lnc. 1989
Holland J. Adaptation in Natural and Artificial Systems. Ann Arbor, MI: University of Michigan Press. 1975
Hornik K, Stinchcombe M and White H. Universal approximation of an unknown mapping and its derivatives using multilayer feedforward networks. Neural Networks. 1990. 3(5): 551–560
Liu S B, Tian S B and Liu Y R et al. Study on the prediction of hydrocarbon type compositions of VGO based on its conventional characteristics. Petroleum Processing and Petrochemicals. 2007. 38(9): 18–22 (in Chinese)
Michalewicz Z. Genetic Algorithms + Data Structures = Evolution Programs (third ed.). Berlin: Springer-Verlag. 1996
Nes K V and Westen H V. Aspects of the Constitution of Mineral Oils. New York: Elsevier. 1951
Nwadinigwe C A and Okoroji K A. Novel equations for quantitative hydrocarbon-type analysis of petroleum fractions. Fuel. 1990. (69): 340–343
Patuwo E, Hu M Y and Hung M S. Two group classification problem using neural networks. Decision Sciences. 1993. 24(4): 825–846
Riazi M R and Daubert T E. Prediction of the composition of petroleum fractions. Ind. Eng. Chem. Process Des. Dev. 1980.19(2): 289–294
Riazi M R and Daubert T E. Prediction of molecular-type analysis of petroleum fraction and coal liquids. Ind. Eng. Chem. Process Des. Dev. 1986. 25: 1009–1015
Rojas R. Neural Networks—A Systematic Introduction. Berlin, New-York: Springer-Verlag. 1996
Sun R J, Wang S C and Zhao S Q. Study on prediction of sour vacuum gas oil (VGO) saturates by using artificial neural networks. Computers and Applied Chemistry. 2008. 25(11): 1393–1396 (in Chinese)
Waterman H I, Boehower C and Cornelissen J. Correlation between physical constants and chemical structure. New York: Elsevier. 1958
Xing B and Cheng Z. Studies on the prediction of hydrocarbon series composition of VGO. Qilu Petrochemical Technology. 2006. 34(2): 173–177 (in Chinese)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Dong, X., Wang, S., Sun, R. et al. Design of artificial neural networks using a genetic algorithm to predict saturates of vacuum gas oil. Pet. Sci. 7, 118–122 (2010). https://doi.org/10.1007/s12182-010-0015-y
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12182-010-0015-y