Reconstruction of the spatial distribution of filtration properties of heterogeneous geologic media based on variations of microseismicity resulting from fluid injection

Мұқаба

Дәйексөз келтіру

Толық мәтін

Ашық рұқсат Ашық рұқсат
Рұқсат жабық Рұқсат берілді
Рұқсат жабық Рұқсат ақылы немесе тек жазылушылар үшін

Аннотация

Determining the properties of heterogeneous reservoirs based on microseismic evolution data is an important task in field development. Analyzing the propagation of microseismic events occurring during fluid injection/withdrawal provides valuable information about permeability and stress state of the reservoir. In this paper, we consider the inverse problem of determining reservoir filtration properties from microseismic event propagation data. For this purpose, the influence of various geological factors on the distribution of microseismic event sources is investigated. Machine learning methods were used to identify correlations between geologic model parameters and microseismicity evolution. Due to the insufficient variability of in-situ data, an artificial database of catalogs of microseismic events containing the coordinates of sources and their occurrence times was created to train the model. For this purpose, numerical modeling of fluid injection and generation of microseismic events in synthetic models of permeable media with different geological structure was carried out. Thus, a comprehensive approach to the restoration of filtration properties of heterogeneous reservoirs from microseismicity evolution data using machine learning methods is proposed. The proposed methodology can be applied to optimize field development, improve the efficiency of fluid extraction and reduce the risks associated with the occurrence of undesirable anthropogenic seismic activity.

Толық мәтін

Рұқсат жабық

Авторлар туралы

Е. Novikova

Sadovsky Institute of Geospheres Dynamics of Russian Academy of Sciences

Хат алмасуға жауапты Автор.
Email: e.novikova@idg.ras.ru
Ресей, Moscow

N. Barishnikov

Sadovsky Institute of Geospheres Dynamics of Russian Academy of Sciences

Email: e.novikova@idg.ras.ru
Ресей, Moscow

S. Turuntaev

Sadovsky Institute of Geospheres Dynamics of Russian Academy of Sciences

Email: e.novikova@idg.ras.ru
Ресей, Moscow

M. Trimonova

Sadovsky Institute of Geospheres Dynamics of Russian Academy of Sciences

Email: e.novikova@idg.ras.ru
Ресей, Moscow

Әдебиет тізімі

  1. Адушкин В. В., Турунтаев С. Б. Техногенная сейсмичность — индуцированная и триггерная. М.: ИДГ РАН. 2015a. 364 с.
  2. Баренблатт Г. И., Ентов В. М., Рыжик В. М. Движение жидкостей и газов в природных пластах. М.: Недра. 1984. 211 с.
  3. Dichiarante A.M., Langet N., Bauer R.A., Goertz-Allmann B.P., Williams-Stroud S.C., Kühn D., Oye V., Greenberg S.E., Dando B.D.E.Identifying geological structures through microseismic cluster and burst analyses complementing active seismic interpretation // Tectonophysics. 2021. Т. 820. С. 229107.
  4. Jessell M., Guo J., Li Y., Lindsay M., Scalzo R., Giraud J., Pirot G., Cripps E., Ogarko V. Into the Noddyverse: a massive data store of 3D geological models for machine learning and inversion applications // Earth Syst Sci Data. 2022. Т. 14. № 1. С. 381–392.
  5. Ritz V., Rinaldi A. P., Wiemer S. Transient evolution of the relative size distribution of earthquakes as a risk indicator for induced seismicity // Commun Earth Environ. 2022. Т. 3.
  6. Scibek J. Multidisciplinary database of permeability of fault zones and surrounding protolith rocks at world-wide sites // Sci Data. 2020. Т. 7. № 1. С. 95.
  7. Shapiro S. Fluid-Induced Seismicity. Cambridge: Cambridge University Press. 2015.
  8. Turuntaev S., Eremeeva E., Zenchenko E. Laboratory study of microseismicity spreading due to pore pressure change // J. Seismol. 2013. Т. 17.
  9. Vaswani A., Shazeer N.M., Parmar N., Uszkoreit J., Jones L., Gomez A.N., Kaiser L., Polosukhin I. Attention is all you need. Adv Neural Inf Process Syst. 2017. Т. 2017-Decem. № Nips. С. 5999–6009.
  10. Zoback M. D. Managing the seismic risk posed by wastewater disposal // Earth Magazine. 2012. С. 38–43.
  11. Zwicker D. A Python package for solving partial differential equations // J. Open Source Softw. 2020. Т. 5. № 48. С. 2158.
  12. idglgfd/permeability_reconstruction_network [Электронный ресурс].
  13. URL: https://github.com/idglgfd/permeability_reconstruction_network (дата обращения: 15.08.2024).

Қосымша файлдар

Қосымша файлдар
Әрекет
1. JATS XML
Жүктеу
2. Fig. 1. Scheme of the problem under consideration (left). Example of spatial distribution of permeability of a model reservoir (right).

Жүктеу (150KB)
Метадеректер
3. Fig. 2. Horizontal section of the permeability model in the plane of the wells (left). Distribution of pore pressure at a certain calculation step (right).

Жүктеу (92KB)
Метадеректер
4. Fig. 3. Scheme of recalculation of effective stress tensor into values ​​of normal and shear τ stress on a separate crack. The normal to the plane of the crack is specified by angles α, γ, — the vector of total stress.

Жүктеу (118KB)
Метадеректер
5. Fig. 4. Spatial distribution of events that have occurred since the start of injection. Events are shown in mutually perpendicular planes passing through the fluid injection point. The color scale reflects the density of seismic events — the ratio of the number of events in the elementary volume of the model to the number of event initiators N.

Жүктеу (122KB)
Метадеректер
6. Fig. 5. An example of synthesized models of a heterogeneous reservoir: a section of the permeability model in the XY plane containing the source location point.

Жүктеу (385KB)
Метадеректер
7. Fig. 6. Example of distribution of microseismic events of different magnitudes at different points in time.

Жүктеу (159KB)
Метадеректер
8. Fig. 7. Scheme of the implemented model. The dimensions of the data tensors obtained at each stage of the model operation are indicated in brackets, N=1000 is the number of events in one catalog.

Жүктеу (207KB)
Метадеректер
9. Fig. 8. Penalty mask - spatial distribution of the loss function multiplier. Shown are a horizontal and two vertical slices passing through the fluid injection point in the center of the reservoir.

Жүктеу (107KB)
Метадеректер
10. Fig. 9. Example of the original (top) and reconstructed (bottom) three-dimensional spatial distribution of permeability of the model formation. Shown are a horizontal and two vertical slices passing through the fluid injection point in the center of the formation. The contour marks the region with the highest reconstruction accuracy (σ≤log(2)≈0.3).

Жүктеу (288KB)
Метадеректер
11. Fig. 10. Result of reconstruction of filtration properties for a set of homogeneous reservoir models. The original permeability values ​​are shown on the left, and the reconstructed ones are shown on the right (section in the XZ plane).

Жүктеу (187KB)
Метадеректер
12. Fig. 11. Result of reconstruction of filtration properties for a set of homogeneous reservoir models with a more permeable rock layer. The initial permeability values ​​are shown on the left, and the reconstructed permeabilities are shown on the right (section in the XZ plane).

Жүктеу (182KB)
Метадеректер
13. Fig. 12. Comparison of true and reconstructed values ​​of the logarithm of permeability for homogeneous reservoirs (left) and for homogeneous reservoirs with a more permeable layer (right).

Жүктеу (158KB)
Метадеректер
14. Fig. 13. Spatial distribution of the parameter σ (standard deviation of the permeability order of magnitude error), averaged over the entire test sample. Shown are a horizontal and two vertical slices passing through the fluid injection point in the center of the reservoir.

Жүктеу (124KB)
Метадеректер
15. Fig. 14. Dependence of the parameter σ on the average number of events in each cell of the model layer for the test sample. The dashed line shows the linear regression.

Жүктеу (107KB)
Метадеректер

© Russian academy of sciences, 2025