When planning production drilling within fields with a complex geological structure, the task was to determine the intervals of the productive reservoir and assess their distribution by area. A complicating factor was the core recovery mainly from the impermeable matrix, which made it difficult to build a petrophysical model of the fractured reservoir. For the analysis of well logging materials presented by the standard Russian Well-logging complex, in the conditions of a limited number of high-tech well-logging methods, it became necessary to use neural networks to determine the lithology of the section and the type of void space. The well with the HI-TECH complex acted as a training data set. The k-means algorithm was used as the most simple and widespread in software products. The parameters for the interval time of the longitudinal wave, the spectrum of uranium and thorium, and the photoelectric factor turned out to be significant for clustering. In total, 6 clusters were identified, three of which are lithologically represented by limestone, the rest are dolomite. Clusters of intermediate differences associated with secondary processes such as fracturing, shale, and dolomitization were identified. The results of the section typification are well comparable with the lithological description of the core. Cluster 4, which is represented by fractured dolomite has the best reservoir properties. Conductive fractures, according to the electric scanner, are most common in clusters 1 and 3. In smaller quantities, they are observed in cluster 4. Productivity was confirmed by production logs and well tests from clusters 1, 3 and 4. It was also revealed that increased well flow rates corresponded to a higher dolomite content. According to the results of the analysis a pilot hole accompanied by such well logs as spectrometry, cross-dipole broadband acoustics and lithology density logging is recommended to minimize the drilling risks and improve the efficiency of development. Horizontal completion is recommended in the intervals of the section composed of rocks similar to clusters 1, 3 and 4.
References
1. Dobrynin V.M., Vendelshteyn B.Yu., Kozhevnikov D.A., Petrofizika (Fizika gornykh porod) (Petrophysics (Physics of rocks)), Moscow: Nedra Publ., 1991, 368 p.
2. Itenberg S.S., Shnurman G.A., Interpretatsiya rezul'tatov karotazha skvazhin (Interpretation of complex reservoirs logging data), Moscow: Nedra Publ., 1978, 389 p.
3. Khusainova A.M., Burikova T.V., Privalova O.R. et al., Raspredelenie neftenasyshchennosti v karbonatnykh kollektorakh v zavisimosti ot strukturnykh osobennostey porody (Oil saturation distribution in carbonate reservoirs depending on rock structural features), Collected papers “Aktual'nye nauchno-tekhnicheskie resheniya neftedobyvayushchego potentsiala PAO ANK “Bashneft'” (Current scientific and technical solutions to the oil production potential of Bashneft PJSC), 2016, V. 124, pp. 62–67.
4. Khusainova A.M., Burikova T.V., Privalova O.R., Nugaeva A.N., The influence of structural and lithological features on the saturation model of Middle Carboniferous reservoirs of the Republic of Bashkortostan (In Russ.), Neftyanoe khozyaystvo = Oil Industry, 2016, no. 8, pp. 74-77.
5. Akhmetzyanov R.V., Privalova O.R., Burikova T.V., Khusainova A.M., Sozdanie unifitsirovannoy petrofizicheskoy modeli srednego karbona po mestorozhdeniyam Respubliki Bashkortostan (Creation of a unified petrophysical model of Middle Carbon in deposits of the Republic of Bashkortostan), Proceedings of XI “Aktual'nye problemy razvitiya neftegazovogo kompleksa Rossii” (Actual problems of the development of the oil and gas complex of Russia), Moscow: Publ. of Gubkin University, 2016, p. 12.
