Virtual MWIR spectra for improved quantification of fuel mixtures

Barreiro Elorza, Pilar and Maldonado Gil, María del Sagrario and Gutierrez.Osuna, Ricardo and Roger, Jean Michel and Vergara, Germán (2021). Virtual MWIR spectra for improved quantification of fuel mixtures. In: "European Conference on Agricultural Engineering AgEng2021", 4-8 Julio, Evora. ISBN 978-972-778-214-7.

Description

Title: Virtual MWIR spectra for improved quantification of fuel mixtures
Author/s:
  • Barreiro Elorza, Pilar
  • Maldonado Gil, María del Sagrario
  • Gutierrez.Osuna, Ricardo
  • Roger, Jean Michel
  • Vergara, Germán
Item Type: Presentation at Congress or Conference (Speech)
Event Title: European Conference on Agricultural Engineering AgEng2021
Event Dates: 4-8 Julio
Event Location: Evora
Title of Book: Proceedings of the European Conference on Agricultural Engineering AgEng2021
Date: 2021
ISBN: 978-972-778-214-7
Subjects:
Freetext Keywords: VOC, Spectroscopy, quality assurance
Faculty: E.T.S. de Ingeniería Agronómica, Alimentaria y de Biosistemas (UPM)
Department: Ingeniería Agroforestal
UPM's Research Group: Técnicas Avanzadas en Agroalimentación LPF-TAGRALIA
Creative Commons Licenses: Recognition - No derivative works - Non commercial

Full text

[img] PDF - Requires a PDF viewer, such as GSview, Xpdf or Adobe Acrobat Reader
Download (2MB)

Abstract

Previous research of the authors has demonstrated that uncooled low-cost MWIR Spectroscopy is a fast and non-invasive method for the qualitative and quantitative evaluation of fuels and their blending with alcohols: methanol, ethanol and propanol among others. However impressive, the preliminary tests have failed to reach the minimum required accuracy + 1%. A main difficulty, in order to reach the expected estimation performance, lays on the generation of very narrow volume of mixture, of ever decreasing additive concentration. This paper proposes a new chemometric procedure, as an in-silico experimental procedure that would help the generation of stable virtual spectra of ultra-low additive concentration. To do so, a previous MANOVA model is used to generate noiseless points in the canonical plane of binary mixtures of fuel and alcohols. Afterwards the Cartesian coordinates of each mixture are de-projected into the estimated original spectra, which shall be onwards called virtual. The so called virtual spectra are then fed into a PLS model for the identification of the additive and corresponding estimation of the concentration, which is validated with regard to the experimental samples.

More information

Item ID: 71740
DC Identifier: https://oa.upm.es/71740/
OAI Identifier: oai:oa.upm.es:71740
Deposited by: Catedrática Pilar Barreiro
Deposited on: 20 Sep 2022 06:00
Last Modified: 20 Sep 2022 06:00
  • Logo InvestigaM (UPM)
  • Logo GEOUP4
  • Logo Open Access
  • Open Access
  • Logo Sherpa/Romeo
    Check whether the anglo-saxon journal in which you have published an article allows you to also publish it under open access.
  • Logo Dulcinea
    Check whether the spanish journal in which you have published an article allows you to also publish it under open access.
  • Logo de Recolecta
  • Logo del Observatorio I+D+i UPM
  • Logo de OpenCourseWare UPM