Comparative Study of Diesel Analysis by Ftir, Ftnir and Ft-Raman Spectroscopy Using Pls and Artificial Neural Network Analysis

Topics: Cetane number, Raman spectroscopy, Spectroscopy Pages: 14 (4991 words) Published: October 5, 2010
Analytica Chimica Acta 547 (2005) 188–196

Acomparative study of diesel analysis by FTIR, FTNIR and FT-Raman spectroscopy using PLS and artificial neural network analysis Vianney O. Santos Jr., Flavia C.C. Oliveira, Daniella G. Lima, Andrea C. Petry, Edgardo Garcia, Paulo A.Z. Suarez, Joel C. Rubim ∗ Laborat´ rio de Materiais e Combust´veis (LMC), Instituto de Qu´mica da Universidade de Brasilia, C.P. 04478, 70904-970 Bras´lia, DF, Brazil o ı ı ı Received 10 January 2005; received in revised form 28 April 2005; accepted 17 May 2005 Available online 24 June 2005

Abstract Diesel properties determined by ASTM reference methods as cetane index, density, viscosity, distillation temperatures at 50% (T50) and 85% (T85) recovery, and the total sulfur content (%, w/w) were modeled by FTIR-ATR, FTNIR, and FT-Raman spectroscopy using partial last square regression (PLS) and artificial neural network (ANN) spectral analysis. In the PLS models, 45 diesel samples were used in the training group and the other 45 samples were used in the validation. In the ANN analysis a modular feedforward network was used. Sixty diesel samples were used in the neural network training and other 30 samples were used in the validation. Two different ATR configurations were compared in the FTIR, a conventional (ATR1) and an immersion (ATR2) cell. The ATR1 cell presented the best results, with smaller prediction errors (root mean square error of prediction, RMSEP). The comparison of the three PLS models (FTIR-ATR1, FTNIR, and FT-Raman) shows that reasonable values of R2 and RMSEP were obtained by the FTIR-ATR1 and FTNIR models in the evaluation of density, viscosity, and T50. The PLS/FT-Raman models presented reasonable results only for the T50 property. None of the techniques was able to generate suitable PLS calibration models for the determination of sulfur content. The ANN/FT-Raman models presented the best performances, with all models presenting R2 -values above 85% some of them with RMSEP values significantly smaller than those obtained with FTIR-ATR and FTNIR. The ANN/FT-Raman and ANN/FTIR-ATR1 models were able to estimate the total sulfur content of diesel with 0.01% (w/w) accuracy. © 2005 Elsevier B.V. All rights reserved. Keywords: FTIR; FTNIR; FT-Raman; PLS; Artificial neural network; Diesel

1. Introduction Vibrational spectroscopies as NIR, IR, and Raman, using dispersive or interferometric instruments, have been extensively used in the last two decades in different kinds of analytical applications, including the analysis of fuels [1–28] as gasoline [1–3,5–13,15,16,18,24], kerosene [4,14,16,25], diesel [2,13,15–17,21,22,26], alcohol fuel [19,20,27], and biodiesel [23,28]. The high light throughput presented by optical fibers in the NIR region and its chemical stability in different solvents is the main factor for its large application in fuel analysis

Corresponding author. Tel.: +55 61 3072162; fax: +55 61 2734149. E-mail address: (J.C. Rubim).

[1–3,7,10,14–24,26,27] including on line monitoring of fuels [2,3,16]. In the medium infrared (mid-IR) spectral region the state of the art technologies of fiber optics demands infrared detectors with high sensitivity in order to get better signalto-noise (S/N) ratios. The mid-IR optical fibers, those used in attenuated total reflectance (ATR) configuration, are not transparent below 950 cm−1 , thus restricting the number of infrared bands to be considered in the analysis. Albeit these limitations mid-IR has found application in fuel quality control, and examples of its application in different fuels can be found in refs. [4–7,10,12–14]. Raman spectroscopy has also been used in the analysis of different kind of fuels [7–11,25,27], including gasoline [7–11], kerosene [25], and alcohol fuel [27]. It is worth to mention that we did not find any reference

0003-2670/$ – see front matter © 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.aca.2005.05.042

V.O. Santos Jr. et al....
Continue Reading

Please join StudyMode to read the full document

You May Also Find These Documents Helpful

  • Artificial Neural Network: Usage Essay
  • Stock Prices Prediction Using Artificial Neural Networks Essay
  • Essay about Artificial Neural Network
  • Essay about artificial Neural Networks
  • Artificial Neural Network
  • Artificial Neural Network Essay
  • Artificial Neural Network Essay

Become a StudyMode Member

Sign Up - It's Free