Neural Computing and Applications

, Volume 18, Issue 6, pp 557–565 | Cite as

Universal technique for optimization of neural network training parameters: gasoline near infrared data example

  • Roman M. Balabin
  • Ravilya Z. Safieva
  • Ekaterina I. Lomakina
Original Article

Abstract

The universal technique of finding optimum training parameters for multi-layer perceptron—such as percentage of samples in a cross-validation set and quantities of training iterations with various initial values—is offered. This technique is aimed at the searching of optimum values of two complex factors depending on accuracy and convergence of a network, and also on the time of its training. Their conventional names are “cross-validation coefficient” and “training iteration coefficient”. Near infrared spectroscopy data for gasoline samples are used to evaluate the efficiency of the method.

Keywords

Artificial neural network (ANN) Multi-layer perceptron (MLP) Parameters optimization Cross-validation coefficient (CVC) Training iteration coefficient (TIC) Gasoline Near infrared (NIR) spectroscopy Density 

References

  1. 1.
    Côcco LC, Yamamotob CI, von Meien OF (2005) Study of correlations for physicochemical properties of Brazilian gasoline. Chemom Intell Lab Syst 76:55–63. doi:10.1016/j.chemolab.2004.09.004 CrossRefGoogle Scholar
  2. 2.
    van Leeuwen JA, Jonker RJ, Gill R (1994) Octane number prediction based on gas chromatographic analysis with non-linear regression techniques. Chemom Intell Lab Syst 25:325–390. doi:10.1016/0169-7439(94)85051-8 CrossRefGoogle Scholar
  3. 3.
    Andrade JM, Sánchez MS, Sarabia LA (1999) Applicability of high-absorbance MIR spectroscopy in industrial quality control of reformed gasolines. Chemom Intell Lab Syst 46:41–55. doi:10.1016/S0169-7439(98)00156-7 CrossRefGoogle Scholar
  4. 4.
    Yang H, Ring Z, Briker Y, McLean N, Friesen W, Fairbridge C (2002) Neural network prediction of cetane number and density of diesel fuel from its chemical composition determined by LC and GC–MS. Fuel 81:65–74. doi:10.1016/S0016-2361(01)00121-1 CrossRefGoogle Scholar
  5. 5.
    Meusingera R, Morosb R (2001) Determination of octane numbers of gasoline compounds from their chemical structure by 13C NMR spectroscopy and neural networks. Fuel 80:613–621. doi:10.1016/S0016-2361(00)00125-3 CrossRefGoogle Scholar
  6. 6.
    Rumelhart DE, Hinton GE, Williams RJ (1986) Learning internal representations by error propagation. In: Rumelhart D, McClelland J (eds) Parallel data processing, vol I, Chap. 8. The MIT Press, Cambridge, pp 318–362Google Scholar
  7. 7.
    Wang W, Paliwal J (2006) Generalisation performance of artificial neural networks for near infrared spectral analysis. Biosyst Eng 94:7–18. doi:10.1016/j.biosystemseng.2006.02.001 CrossRefGoogle Scholar
  8. 8.
    Burns JA, Whitesides GM (1993) Feed-forward neural networks in chemistry: mathematical systems for classification and pattern recognition. Chem Rev 93:2583–2601. doi:10.1021/cr00024a001 CrossRefGoogle Scholar
  9. 9.
    Balabin RM, Safieva RZ, Lomakina EI (2008) Wavelet neural network (WNN) approach for calibration model building based on gasoline near infrared (NIR) spectra. Chemometr Intell Lab 93:58. doi:10.1016/j.chemolab.2008.04.003 CrossRefGoogle Scholar
  10. 10.
    Balabin RM, Safieva RZ (2008) Motor oil classification by base stock and viscosity based on near infrared (NIR) spectroscopy data. Fuel 87:2745. doi:10.1016/j.fuel.2008.02.014
  11. 11.
    Balabin RM, Safieva RZ (2007) Capabilities of near infrared spectroscopy for the determination of petroleum macromolecule content in aromatic solutions. J Near Infrared 15(Spec):343. doi:10.1255/jnirs.749 CrossRefGoogle Scholar
  12. 12.
    Russian Standard: GOST R 51069-97Google Scholar
  13. 13.
    Moody J, Utans J (1992) Principled architecture selection for neural networks: application to corporate bond rating prediction. In: Moody J, Hanson SJ, Lippmann RP (eds) Advances in neural information processing systems, vol IV. Morgan Kaufmann, San Mateo, pp 683–690Google Scholar
  14. 14.
    Larsen J, Goutte C (1999) On optimal data split for generalisation estimation and model selection neural networks for signal processing. In: IX Proceedings of the 1999 IEEE signal processing society workshop, pp 225–234Google Scholar
  15. 15.
    Jurs PC, Bakken GA, McClelland HE (2000) Computational methods for the analysis of chemical sensor array data from volatile analytes. Chem Rev 100:2649–2678. doi:10.1021/cr9800964 CrossRefGoogle Scholar
  16. 16.
    Balabin RM, Syunyaev RZ (2008) Petroleum resins adsorption onto quartz sand: near infrared (NIR) spectroscopy study. J Colloid Interface Sci 318:167. doi:10.1016/j.jcis.2007.10.045 CrossRefGoogle Scholar
  17. 17.
    Balabin RM, Safieva RZ (2008) Gasoline classification by source and type based on near infrared (NIR) spectroscopy data. Fuel 87:1096. doi:10.1016/j.fuel.2007.07.018 CrossRefGoogle Scholar
  18. 18.
    Balabin RM, Safieva RZ, Lomakina EI (2007) Comparison of linear and nonlinear calibration models based on near infrared (NIR) spectroscopy data for gasoline properties prediction. Chemometr Intell Lab 88:183. doi:10.1016/j.chemolab.2007.04.006 CrossRefGoogle Scholar
  19. 19.
    Balabin RM, Syunyaev RZ, Karpov SA (2007) Molar enthalpy of vaporization of ethanol–gasoline mixtures and their colloid state. Fuel 86:323. doi:10.1016/j.fuel.2006.08.008 CrossRefGoogle Scholar
  20. 20.
    Balabin RM (2008) Dispersed structure of ethanol–gasoline fuel according to dynamic light scattering method. J Dispersion Sci Tech 29:457. doi:10.1080/01932690701718925 CrossRefGoogle Scholar
  21. 21.
    Balabin RM, Syunyaev RZ, Karpov SA (2007) Quantitative measurement of ethanol distribution over fractions of ethanol-gasoline fuel. Energy Fuels 21:2460. doi:10.1021/ef070081l CrossRefGoogle Scholar
  22. 22.
    Allen MP, Tildesley DJ (1989) Computer simulation of liquids. Oxford University Press, New YorkGoogle Scholar
  23. 23.
    Balabin RM (2008) Intermolecular dispersion interactions of normal alkanes with rare gas atoms: van der Waals complexes of n-pentane with helium, neon, and argon. Chem Phys 352:267. doi:10.1016/j.chemphys.2008.06.015 CrossRefGoogle Scholar

Copyright information

© Springer-Verlag London Limited 2009

Authors and Affiliations

  • Roman M. Balabin
    • 1
  • Ravilya Z. Safieva
    • 1
  • Ekaterina I. Lomakina
    • 2
  1. 1.Gubkin Russian State University of Oil and GasMoscowRussia
  2. 2.Faculty of Computational Mathematics and CyberneticsLomonosov Moscow State UniversityMoscowRussia

Personalised recommendations