Journal Home > Volume 4 , Issue 2
Friction 2016, 4(2): 105-115 https://doi.org/10.1007/s40544-016-0104-z
Research Article | Open Access | Issue | Published: 28 March 2016

Establishing quantitative structure tribo-ability relationship model using Bayesian regularization neural network

Show Author's Information Xinlei GAO1,  Kang DAI2( ),  Zhan WANG,  Tingting WANG,  Junbo HE3
School of Chemical and Environmental Engineering, Wuhan Polytechnic University, Wuhan 430023, China
College of Pharmacy, South-Central University for Nationalities, Wuhan 430074, China
College of Food Science and Engineering, Wuhan Polytechnic University, Wuhan 430023, China
Keywords:
lubricant additive, quantitative structure tribo-ability relationship, antiwear, Bayesian regularization neural network
Cite this article:
GAO X, DAI K, WANG Z, et al. Establishing quantitative structure tribo-ability relationship model using Bayesian regularization neural network. Friction, 2016, 4(2): 105-115. https://doi.org/10.1007/s40544-016-0104-z
Download citation
EndNote(RIS)
BibTeX

547

Views

22

Downloads

Citations

21

Crossref

N/A

WoS

23

Scopus

5

CSCD

Abstract Full text Electronic supplementary material About this article

Quantitative structure-activity relationship methods are used to study the quantitative structure tribo- ability relationship (QSTR), which refers to the tribology capability of a compound from the calculation of structure descriptors. Here, we used the Bayesian regularization neural network (BRNN) to establish a QSTR prediction model. Two-dimensional (2D) BRNN–QSTR models can flexibly and easily estimate lubricant-additive antiwear properties. Our results show that electron transfer and heteroatoms (such as S, P, O, and N) in a lubricant-additive molecule improve the antiwear ability. We also found that molecular connectivity indices are good descriptors of 2D BRNN–QSTR models.


menu
Abstract
Full text
Outline
Electronic supplementary material
About this article

Establishing quantitative structure tribo-ability relationship model using Bayesian regularization neural network

Show Author's information Xinlei GAO1, Kang DAI2( ), Zhan WANG, Tingting WANG, Junbo HE3
School of Chemical and Environmental Engineering, Wuhan Polytechnic University, Wuhan 430023, China
College of Pharmacy, South-Central University for Nationalities, Wuhan 430074, China
College of Food Science and Engineering, Wuhan Polytechnic University, Wuhan 430023, China

Abstract

Quantitative structure-activity relationship methods are used to study the quantitative structure tribo- ability relationship (QSTR), which refers to the tribology capability of a compound from the calculation of structure descriptors. Here, we used the Bayesian regularization neural network (BRNN) to establish a QSTR prediction model. Two-dimensional (2D) BRNN–QSTR models can flexibly and easily estimate lubricant-additive antiwear properties. Our results show that electron transfer and heteroatoms (such as S, P, O, and N) in a lubricant-additive molecule improve the antiwear ability. We also found that molecular connectivity indices are good descriptors of 2D BRNN–QSTR models.

Keywords: lubricant additive, quantitative structure tribo-ability relationship, antiwear, Bayesian regularization neural network

References(24)

[1]
Liu X Q, Zhou F, Liang Y M, Liu W M. Tribological performance of phosphonium based ionic liquids for an aluminum-on-steel system and opinions on lubrication mechanism. Wear 261(10): 1174–1179 (2006)
DOI Google Scholar
[2]
Yu G Q, Zhou F, Liu W M, Liang Y M, Yan S Q. Preparation of functional ionic liquids and tribological investigation of their ultra-thin films. Wear 260(9):1076–1080 (2006)
DOI Google Scholar
[3]
Jiménez A E, Bermúdez M D, Iglesias P, Carrión F J, Martínez-Nicolás G. 1-N-alkyl-3-methylimidazolium ionic liquids as neat lubricants and lubricant additives in steel– aluminium contacts. Wear 260(7): 766–782 (2006)
DOI Google Scholar
[4]
Singh H, Gulati I B. Tribological behaviour of some hydrocarbon compounds and their blends. Wear 139(2): 425–437 (1990)
DOI Google Scholar
[5]
Martin J M, Grossiord C, Varlo K, Vacher B, Igarashi J. Synergistic effects in binary systems of lubricant additives: A chemical hardness approach. Tribol Lett 8: 193–201 (2000)
DOI Google Scholar
[6]
Hansch, C., and Steward, A. R.The Use of Substituent Constants in the Analysis of the Structure-Activity Relationship in Penicillin Derivatives. J. Med. Chem., 7: 691–694 (1964)
DOI Google Scholar
[7]
Li F, Chen J W, Wang Z J, Li J, Qiao X L. Determination and prediction of xenoestrogens by recombinant yeast-based assay and QSAR. Chemosphere 74(9): 1152–1157 (2009)
DOI Google Scholar
[8]
Tintori C, Magnani M, Schenone S, Botta M. Docking, 3D-QSAR studies and in silico ADME prediction on c-Src tyrosine kinase inhibitors. European Journal of Medicinal Chemistry 44(3): 990–1000 (2009)
DOI Google Scholar
[9]
Sharma D, Narasimhan B, Kumar P, Jalbout A. Synthesis and QSAR evaluation of 2-(substituted phenyl)-1H-benzimidazoles and [2-(substituted phenyl)-benzimidazol-1-yl]-pyridin-3-yl- methanones. Eur J Med Chem 44(3): 1119–1127 (2009)
DOI Google Scholar
[10]
Gao X L Wang Z, Zhang H, Dai K. A three dimensional quantitative structure-tribological relationship model. J Tribol 137(2): 021802-1–021802-2 (2015)
DOI Google Scholar
[11]
Dai K, Gao X L. Estimating antiwear properties of lubricant additives using a quantitative structure tribo-ability relationship model with back propagation neural network. Wear 306(1–2): 242–247 (2013)
DOI Google Scholar
[12]
Aggarwal K K, Singh Y, Chandra P, Puri M. Bayesian regularization in a neural network model to estimate lines of code using function. J Comput Sci 1: 505–509 (2005)
DOI Google Scholar
[13]
Zhang J Y. The relationship between additives molecular structure and their tribological properties and the mechanism of boundary lubrication. Ph.D Thesis. Lanzhou: Lanzhou Institute of Chemical Physics, Chinese Academy of Sciences, 1999.
[14]
Gao X, Wang Z, Zhang H, Dai K, Wang T. A quantitative structure tribo-ability relationship model for ester lubricant base oils. J Tribol 137(2): 021801-1–021801-7 (2015)
DOI Google Scholar
[15]
Gao X, Wang R, Wang Z, Dai K. BPNN-QSTR friction model for organic compounds as potential lubricant base oils. J Tribol, in press, .
DOI Google Scholar
[16]
Gao X, Liu D, Wang Z, Dai K. Quantitative structure tribo- ability relationship for organic compounds as lubricant base oils using CoMFA and CoMSIA. J Tribol (Accepted)
Google Scholar
[17]
Gao X L, Dai K, Gao W Z, Wang Z, Wang T T. The application of quantitative structure tribo-ability relationship model. In World Tribology Congress, Turin, Italy, 2013.
[18]
MacKay D J C. A practical Bayesian framework for backprop networks. Neural Comput 4(3): 448–472 (1992)
DOI Google Scholar
[19]
Foresee F D, Hagan M T. Gauss-Newton approximation to Bayesian learning. In Proceedings of the 1997 International Joint Conference on Neural Networks. Houston, 1997: 1930–1935.
[20]
MacKay D J C. Bayesian interpolation. Neural Comput 4(3): 415–447 (1992)
DOI Google Scholar
[21]
Bonchev D. Information Theoretic Indices for Characterization of Chemical Structures, Chemometrics Series. New York: Research Studies Press Ltd, 1983.
[22]
Balaban A T. Highly discriminating distance-based topological index. Chem Phys Lett 89: 399–404 (1982)
DOI Google Scholar
[23]
Müller W R, Szymanski K, Knop J V, Trinajstić N. An algorithm for construction of the molecular distance matrix. J Comput Chem 8(2): 170–173 (1987)
DOI Google Scholar
[24]
Kier L B, Hall L H. Molecular Connectivity Indices in Chemistry and Drug Research, Medicinal Chemistry. New York: Academic Press, 1976.
File
40544_2016_104_MOESM1_ESM.pdf (326.9 KB)
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 11 December 2015
Revised: 19 January 2016
Accepted: 22 January 2016
Published: 28 March 2016
Issue date: June 2021

Copyright

© The author(s) 2016

Acknowledgements

This work was supported by the National Basic Research (973) Program of China (No. 2013CB632303) and the National Natural Science Foundation of China (NSFC, No. 51075309). The authors gratefully acknowledge Prof. Junyan Zhang for providing critical data on tribological properties.

Rights and permissions

This article is published with open access at Springerlink.com

Open Access: The articles published in this journal are distributed under the terms of the Creative Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Return