According to current experience of the Bazhenov formation developing in the West Siberian petroleum province it is evident that there are great prospects for oil recovery efforts in the region. They go along with the challenges of identifying producing intervals and the selection oil recovery methods. The aim of this work is to develop a mineral component model (MCM) to be used for identifying potential producing intervals in the Bazhenov formation.
Current approach to developing MCM involves usage of either a large set of well logging methods for complex MCMs which can make it more complicated to apply the model to a large set of wells due to fewer common well logs or the standard set of well logging methods which can make it impossible to develop a model with complex mineral composition. Moreover when tuning a volumetric mineralogical model with the standard set of well logging methods some errors related to normalization of neutron and gamma ray logs may occur. When a full set of well logs is not available it is proposed to use a two-step method which is based on developing a continuous model with a conditionally "extended" set of well logs with the petrophysical constants tuned to certain core macrocomponents in the first step and tuning a model developed with the standard set of well logs to the continuous model with a conditionally "extended" set of well logs in the second step. The process of highlighting of core macrocomponents is done in such a way so as to allow their identification on the available well logs and to be able to match the initial MCM with a linear combination of the macrocomponents. After the volume fraction MCM is obtained, the relationships between the mineralogical composition of rocks, their geomechanical and geochemical properties and well performance are established. Producing intervals are then identified by machine learning algorithms using the established relationships and the volume fraction MCM. The results obtained can be further used as input data for petro elastic modeling, seismic interpretation problems, as well as for forecasting the current oil-generating potential, the amount of hydrocarbons present in the pore space (both occluded and free), brittleness and productivity indices with further ability to forecast productive thicknesses.
1. Dakhnov V.N., Geofizicheskie metody opredeleniya kollektorskikh svoystv i neftegazonasyshcheniya gornykh porod (Geophysical methods for the determination of reservoir properties and oil and gas saturation of rocks), Moscow: Nedra Publ., 1985, 310 p.
2. Merkulov V.P., Posysoev A.A., Otsenka plastovykh svoystv i operativnyy analiz karotazhnykh diagramm (Evaluation of reservoir properties and operational analysis of well logs), Tomsk: Publ. of TPU, 2004, 176 p.
3. Fizicheskie svoystva gornykh porod i poleznykh iskopaemykh (petrofizika). Spravochnik geofizika (Physical properties of rocks and minerals (petrophysics). Geophysics Reference): edited by Dortman N.B. Moscow: Nedra Publ., 1984, 455 p.
4. Biblioteka neobsazhennogo stvola (Open hole library), Ufa, 2017, URL: http://www.primegeo.ru/assets/files/bns.pdf
5. Kulyapin P.S., Razrabotka interpretatsionnoy ipetrouprugoy modeley porod-kollektorov mnogokomponentnogo sostava i slozhnoy struktury emkostnogo prostranstva (Development of interpretative and petroelastic models of reservoir rocks of multicomponent composition and complex structure of capacitive space): thesis of geological and mineralogical science, Moscow, 2016.
6. Dan'ko D.A., Razrabotka printsipov izucheniya netraditsionnykh glinistykh kollektorov na osnove petrouprugogo modelirovaniya i amplitudnoy inversii seysmicheskikh dannykh (Development of principles for the study of unconventional clay reservoirs based on petroelastic modeling and amplitude inversion of seismic data): thesis of candidate of technical science, Moscow, 2018.
8. Hartigan J.A., Wong M.A., A k-means clustering algorithm, Applied Statistics, 1979, V. 28, pp. 100–108.9. Tou J.T., Gonzalez R.C., Pattern recognition principles, Addison-Wesley, 1974.