Study of the conditions of formation of spectral and absorbption multidimensional images of motor fuels and liquid mineral technical oils

Abstract

Effect of experimental conditions on the possibility of formation of multidimensional images of motor fuels and liquid mineral technical oils used for the classification and identification of these samples by the method of pattern recognition is studied. Transmission spectra are considered as primary images of the studied samples. Multidimensional statistical analysis of the results of spectroscopic measurements is performed by principal component analysis. A noticeable increase in the separation of spectroscopic images on the statistical plane in the optimization of the spectral range of the primary image and the thickness of the absorbing layer is shown.