Enhancing Resolution of Anisotropic Resistivity Using a Pixel Inversion with Data-Driven Regularization
- Gong Li Wang (Schlumberger) | Koji Ito (Schlumberger) | Xiaobo Hong (Schlumberger) | Mohammad Taghi Salehi (Schlumberger) | ZhanGuo Shi (Schlumberger) | David Allen (Schlumberger) | Michael Rabinovich (BP) | Jeffrey Meyer (Repsol)
- Document ID
- Society of Petroleum Engineers
- SPE Annual Technical Conference and Exhibition, 26-29 October, Virtual
- Publication Date
- Document Type
- Conference Paper
- 2020. Society of Petroleum Engineers
- Resistivity anisotropy, Pixel inversion, Resolution, Triaxial induction
- 36 in the last 30 days
- 108 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 28.00|
Triaxial induction is a powerful tool to identify thin-bed reservoirs that would be missed with conventional techniques and to improve the accuracy of net pay estimation. On the other hand, it has been noticed that with the standard processing, the resolution of horizontal and vertical resistivities cannot match that of the 2-foot array induction resistivity. Sometimes large apparent anisotropy can be seen in high-resistivity clean sands, which has caused confusion in data interpretation and reserve evaluation.
Researchers have long known that triaxial induction data contains high-frequency information from the formation through the abrupt change and spikes of transverse coupling logs near bed boundaries. In this paper, we present a novel pixel inversion equipped with a revamped cost function and a data-driven regularization scheme to better resolve thin beds by using the high-frequency information that had not been used to its full potential.
The pixel inversion is a variant of the maximum entropy inversion that has proved to be superior for conventional induction data. However, the direct use of the method for triaxial induction data tends to give a slowly varying vertical resistivity log that fails to resolve thin beds as desired. This issue is resolved by means of two adaptive relaxation terms for the smoothness regularization determined by utilizing the data sensitivity to horizontal and vertical resistivities that evolve continuously with the iteration.
In favorable conditions, results on a variety of synthetic models show that a thin bed of less than 1 ft can be detected with deep arrays. In contrast, a bed less than 2 ft can hardly be seen with previous inversions. Results also show that apparent anisotropy ratio is reduced significantly in high-resistivity isotropic cases emulating clean sands. Moreover, the horizontal and vertical resistivities compare favorably with array induction logs in terms of accuracy in the clean sand cases.
Field cases confirm that the pixel inversion is clearly superior to the standard triaxial 1D inversion as far as the vertical (or along-hole) resolution is concerned. The thin beds that can now been seen on horizontal and vertical resistivity logs of the pixel inversion are in good agreement with nuclear logs. In the cases of high-resistivity clean sands, the large apparent anisotropy is largely eliminated with the pixel inversion.
|File Size||2 MB||Number of Pages||22|
Barber, T. D., and Rosthal, R. A. 1991. Using a multiarray induction tool to achieve high-resolution logs with minimum environmental effects, Presented at the SPE Annual Technical Conference and Exhibition, Dallas, Texas, USA, 6–9 October. SPE-22725-MS. https://doi.org/10.2118/22725-MS.
Freedman, R., and Minerbo, G. N., 1991, Maximum entropy inversion of induction-log data, SPE Formation Evaluation, 6, 259–268. https://doi.org/10.2118/19608-PA.
Wang, G. L., Dean M. Homan, Natalie Uschner-Arroyo, Ping Zhang, Wael Abdallah, and Nasar Khan, 2019, Determining resistivity and low-frequency dielectric constant using induction data in the presence of strong induced polarization: Paper PP in Transactions of the 60th SPWLA Annual Logging Symposium.
Wang, H., T.Barber, R.Rosthal, J.Tabanou, B.Anderson, and T.Habashy, 2003, Fast and rigorous inversion of triaxial induction logging data to determine formation resistivity anisotropy, bed boundary position, relative dip and azimuth angle: Society of Exploration Geophysicists Expanded Abstracts, 22, 514.
Wang, H., T.Barber, K.Chen, S.Davydycheva, M.Frey, D.Homan, G.Minerbo, C.Morriss, R.Rosthal, J.Smits, and G.Tumbiolo, 2006a, Triaxial induction logging: Theory, modeling, inversion, and interpretation: Paper SPE 103897 presented at the SPE International Oil & Gas Conference and Exhibition, Beijing, China, December 5–7.
Wang, H., T.Barber, C.Morriss, R.Rosthal, R.Hayden, and M.Markley, 2006b, Determining anisotropic formation resistivity at any relative dip using a multiarray triaxial induction tool: Paper SPE 103113 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA., September 24–27.