RESEARCH OF CLUSTERIZATION METHODS OF OIL DEPOSITS IN THE DNIPRO-DONETSK DEPRESSION WITH THE PURPOSE OF CREATING THEIR CLASSIFICATION BY METAL CONTENT (ON THE VANADIUM EXAMPLE)

Abstract

Purpose. Substantiate the method of typification of oil deposits of the Dnipro-Donetsk basin to create their classification by metal content. The relevance of such studies is due to the strengthening of requirements for environmental protection and comprehensive processing of mineral resources, which is determined by a number of Ukraine laws, resolutions of the Cabinet of Ministers, as well as the requirements of the State Commission on Mineral reserves to the quality and content of geological materials. Methodology. The factual basis of the work was the results of X-ray fluorescence analyzes of metal content in oils from the deposits of the Dnipro-Donetsk basin. These results were statistically processed with calculations of descriptive statistics and assessment of the peculiarities of the distribution of metal contents in oil from each field to justify the choice and calculation of normalized contents. Given that one of the most effective procedures for simplifying and minimizing the data set to create natural classifications is cluster analysis, to achieve this goal, clustering of oil deposits was carried out by various methods, which are implemented in the most popular professional statistical software platforms “STATISTICA” and “SPSS”. Results. To perform a cluster analysis of metal contents of 36 deposits of the Dnipro-Donetsk basin: Bakhmachsky, Pryluky, Krasnozayarsky, Kachalivsky, Kremenovsky, Karaykozovsky, Korobochkinsky, Kulychikhinsky, Lipovodolinsky, Monastyrychovsky, Marastyrishchevsky, Malostyrishchevsky, Malostyrishchevsky, Malostyrishchevsky, , Raspashnovsky, Sofiyivsky, Sukhodolivsky, Solontsivsky, Solokhivsky, Talalayevsky, Trostyanetsky, Turutynsky, Kharkivtsy, Shchurynsky, Yuryevsky, Yaroshivsky, Khukhryansky, Sagaidatsky №1, Sagaidatsky №13, Kybybitsy, Kyshtsy №5 №56 and Kybytsivsky №1 by different methods, analysis of clustering results and substantiation of selection of the most optimal of them for further development of natural classification of deposits by metal content in oils. Scientific novelty. To establish a weighted centroid method of cluster analysis, as the most optimal for the subject-free researcher division of oil fields by metal content (including vanadium) into taxa. Practical significance. The practical significance of the results of the work is that the constructed dendrograms of clustering of deposits by vanadium content in oils can be used as a basis for developing classifications of these deposits for environmental and technological assessment in the future. The proposed methodological approach can be used in the future to divide other groups of geological objects into natural taxa. Key words: oil fields, X-ray fluorescence analysis, metals, cluster analysis, dendrogram, weighted centroid method.