Classification of crude oil gas chromatograms by pattern recognition techniques

Abstract

Pattern recognition methods have been employed to classify crude oils based on their gas chromatograms. Four oil types were represented by gas chromatograms taken before and after artificial weathering. The chromatograms were hand digitized and coded with 13 descriptors each-peak areas for the normal alkanes for C16 through C25 plus pristane and phytane and also one descriptor characterizing the unresolved background. A statistical algorithm was used to find the best subsets of descriptors to use with Bayesian discriminant analysis. A variety of different partitions of the data set showed the similarities of some classes of oils and some dissimilarities for others. The nonparametric linear learning machine method of discriminant training was also applied to various partitions of the data. Predictive abilities as high as 100% were obtained for some of the partitions of the data set.

Pattern Recognition Used to Investigate Multivariate Data in Analytical Chemistry

“…Materials or mixtures characterized by many measurements can be classified into categories, for example, origin, by pattern recognition methods. Examples of identification or classification problems drawn from diverse areas are found in the literature: manufacturers and grades of papers (22), quarry sites of archeological artifacts (23), sources of atmospheric particulate matter (24), classification ofwines (25), determination of the origin of olive oil samples (26), identification of crude oil samples (27), selection of adsorbates for chemical sensor arrays (28), determination of the clinical status of patients from urine samples (29), classification ofcancer cells (30), the study of acute lymphocytic leukemia (31), classification ofhuman brain tissues (32), detection ofcystic fibrosis heterozygotes (33), and the classification of bacteria (34).…”

Stochastic model on the mobility of petroleum in the soils of the Ecuadorian Amazon

“…Petroleum is a complex and heterogeneous mixture of aliphatic, alicyclic and aromatic hydrocarbons having between one and sixty carbon atoms, contains oxygen, nitrogen and sulfur in changing proportions, is considered as a pollutant because of its complex chemical composition, characterized by being persistent and its ability to bioaccumulate [6]. Oil can be classified based on its density, which is measured on a scale developed by the American Petroleum Institute (API) known as API grades [7]. At one end, it is a pale, mobile, straw-colored liquid with a density between 30 and 40 ° API (called light crude).…”

Characterization of Fuel Blends by GC-MS and Multi-Way Chemometric Tools

“…Accordingly, it has traditionally been very difficult to elucidate mechanisms of fuel degradation and to relate fuel properties such as thermal stability to fuel composition. Automated classification algorithms have been successfully used to characterize GC data (Clark and Jurs, 1979) from crude oil samples according to their origin. Chemometric analysis of near infrared spectroscopy (NIR) data is successfully used to obtain estimates of octane number and aromatic content of gasoline (Swarin and Drumm, 1991).…”