Prediction of Relative Density, Distillation Temperatures, Flash Point, and Cetane Number of S500 Diesel Oil Using Multivariate Calibration of Gas Chromatographic Profiles

In the present study, hierarchical cluster analysis was used to select 150 S500 diesel fuel samples from an initial set of 1320 samples assayed through official standards according to ANP Brazilian Regulation No. 50/2013. Four physicochemical properties were analyzed, namely, relative density, distillation temperatures (T10%, T50%, and T85%), flash point, and cetane number. Selected samples were also analyzed by gas chromatography with flame ionization detection (GC-FID), a very common technique used for fuel quality control due to its convenience, accuracy, simplicity, and possible association of the chromatographic profiles with multivariate analyses. PLS regression models were obtained aiming at predicting the four physicochemical properties of the diesel fuel samples. From a maximum chromatographic analysis time of 108 min, regression models with unbiased predictions and good prediction capability for all properties were obtained, with average relative errors lower than 6%.