Técnicas quimiométricas de reconocimiento de patrones para el control de calidad de turbocombustibles Jet A-1 a partir de sus propiedades físico químicas.

Abstract

Jet fuels are assessed by rigorous quality standards in order to guarantee the optimum performance of aircrafts. The process includes the checking of three main areas: raw material, refining process and final products. The present paper shows the feasibility of using the results achieved in physicalchemical trials, together with chemometric pattern recognition techniques, for quality control in the refining process of jet fuels. A set of 70 samples of JET A-1, collected in a period of 2 years, was used to calculate the model of principal component analysis (PCA). The calculation of PCA model was based on three principal components that described 73 % of data variance. It allowed differentiating two classes of Jet A-1 fuel along the first principal component. The variables with the highest modeling power were: distillation temperatures (final boiling point, temperatures at 50 and 90 % of recovery), density, kinematic viscosity and naphthalene hydrocarbons. A training set of 30 samples was used to calculate the SIMCA classification model. Besides, an independent validation set of 30 samples was used to evaluate their performance, five of them coming from another refinery. The comparative study carried out through a validation with external samples proved that the SIMCA model had an adequate sensitivity and selectivity: it didn’t show false negatives and it was unable to classify the turbo fuel samples alien to the modeled system. The achieved results demonstrated that the PCA and SIMCA models are valuable tools that allow the fast obtaining of useful information for quality control