Orbital - Vol. 7 No. 2 - April-June 2015
FULL PAPERS

Application Self-organizing Map Type in a Study of the Profile of Gasoline C Commercialized in the Eastern and Northern Parana Regions

PDF
  • Lívia Ramazzoti Chanan Silva,
  • Karina Gomes Angilelli,
  • Hágata Cremasco,
  • Érica Signori Romagnoli,
  • Aline Regina Walkoff,
  • Dionisio Borsato
Lívia Ramazzoti Chanan Silva
State University Of Londrina, Chemistry Department, Fuels Analyses and Research Laboratory
Karina Gomes Angilelli
State University Of Londrina, Chemistry Department, Fuels Analyses and Research Laboratory
Hágata Cremasco
State University Of Londrina, Chemistry Department, Fuels Analyses and Research Laboratory
Érica Signori Romagnoli
State University Of Londrina, Chemistry Department, Fuels Analyses and Research Laboratory
Aline Regina Walkoff
State University Of Londrina, Chemistry Department, Fuels Analyses and Research Laboratory
Dionisio Borsato
State University Of Londrina, Chemistry Department, Fuels Analyses and Research Laboratory
Published June 29, 2015
Keywords
  • gasoline,
  • weight map,
  • topological map,
  • neural network
How to Cite
(1)
Silva, L. R. C.; Angilelli, K. G.; Cremasco, H.; Romagnoli, Érica S.; Walkoff, A. R.; Borsato, D. Application Self-Organizing Map Type in a Study of the Profile of Gasoline C Commercialized in the Eastern and Northern Parana Regions. Orbital: Electron. J. Chem. 2015, 7, 185-190.

Copyright (c) 2015 Orbital: The Electronic Journal of Chemistry

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Abstract

Artificial neural networks self-organizing map type (SOM) was used to classify samples of automotive gasoline C marketed in the eastern and northern regions of the state of Paraná, Brazil. The input order of parameters in the network were the values of temperature of the first drop, the 10, 50 and 90% distilled bulk, the final boiling point, density, residue content and alcohol content. A network with a topology of 25x25 and 5000 training epochs was used. The weight maps of input parameters for the trained network identified that the most important parameters for classifying samples were the temperature of the first drop and the temperature of the 10% and 50% of the distilled fuel.

DOI: http://dx.doi.org/10.17807/orbital.v7i2.732

PDF