Research on Gas Channeling Identification Using the Fuzzy Comprehensive Evaluation Method
Abstract
:1. Introduction
2. Comprehensive Fuzzy Evaluation Method of Gas Channeling Pathways
2.1. Determination of Static and Dynamic Indicators
2.2. Determination of Indicator Weights
2.2.1. Construction of the Weight Discrimination Matrix
2.2.2. Weight Calculation Method
2.3. Calculation of Indicator Membership Degree
2.3.1. Membership Function
2.3.2. Calculation Functions for Indicator Membership Degrees
2.4. Gas Channel Grading Standards
3. Application Example
4. Conclusions
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Kang, W.-L.; Zhou, B.-B.; Issakhov, M.; Gabdullin, M. Advances in enhanced oil recovery technologies for low permeability reservoirs. Pet. Sci. 2022, 19, 1622–1640. [Google Scholar] [CrossRef]
- Shiyi, Y.; Qiang, W. New progress and prospect of oilfields development technologies in China. Pet. Explor. Prod. 2018, 45, 698–711. [Google Scholar]
- Hu, Y.; Hao, M.; Chen, G.; Sun, R.; Li, S. Technologies and practice of CO2 flooding and sequestration in China. Pet. Explor. Dev. 2019, 46, 753–766. [Google Scholar] [CrossRef]
- Yan, S.; Ma, D.; Li, J.; Zhao, T.; Ji, Z.; Han, H. Progress and prospects of carbon dioxide capture, EOR-utilization and storage industrialization. Pet. Explor. Prod. 2022, 49, 955–962. [Google Scholar] [CrossRef]
- Afzali, S.; Rezaei, N.; Zendehboudi, S. A comprehensive review on Enhanced Oil Recovery by Water Alternating Gas (WAG) injection. Fuel 2018, 227, 218–246. [Google Scholar] [CrossRef]
- Gbadamosi, A.O.; Kiwalabye, J.; Junin, R.; Augustine, A. A review of gas enhanced oil recovery schemes used in the North Sea. J. Pet. Explor. Prod. Technol. 2018, 8, 1373–1387. [Google Scholar] [CrossRef]
- Gao, Y.; Zhao, M.; Wang, J.; Zong, C. Performance and gas breakthrough during CO2 immiscible flooding in ultra-low permeability reservoirs. Pet. Explor. Dev. 2014, 41, 88–95. [Google Scholar] [CrossRef]
- Xu, T.; Wei, Y.; Ma, T. Experiment and Simulation of Gas Channeling Control Technique during CO2 Flooding in Normal Pressure Tight Oil Reservoir of South Ordos Basin. In Proceedings of the Abu Dhabi International Petroleum Exhibition & Conference, Abu Dhabi, United Arab Emirates, 9–12 November 2020. [Google Scholar]
- Rein, E.; Schulz, L.K. Applications of natural gas tracers in the detection of reservoir compartmentalisation and production monitoring. J. Pet. Sci. Eng. 2007, 58, 428–442. [Google Scholar] [CrossRef]
- Brigham, W.E.; Abbaszadeh-Dehghani, M. Tracer testing for reservoir description. Pet. Technol 1986, 1, 519–527. [Google Scholar] [CrossRef]
- Abbaszadeh-Dehghani, M.; Brigham, W.E. Analysis of Well-to-Well Tracer Flow to Determine Reservoir Layering. J. Pet. Technol. 1984, 36, 1753–1762. [Google Scholar] [CrossRef]
- Ren, B.; Ren, S.; Zhang, L.; Chen, G.; Zhang, H. Monitoring on CO2 migration in a tight oil reservoir during CCS-EOR in Jilin Oilfield China. Energy 2016, 98, 108–121. [Google Scholar] [CrossRef]
- Chen, M.; Tang, W.; Wang, L.; Sun, X.; Li, N.; Tan, J.; Zhang, W. Tracer Research Methods for Hydrocarbon Substances in Shale Oil Reservoirs. Chem. Technol. Fuels Oils 2024, 60, 197–206. [Google Scholar] [CrossRef]
- Du, D.; Hao, F.; Li, Y.; Li, D.; Tang, Y. Study on Interpretation Method of Multistage Fracture Tracer Flowback Curve in Tight Oil Reservoirs. ACS Omega 2024, 9, 11628–11636. [Google Scholar] [CrossRef]
- Albertoni, A.; Lake, L.W. Inferring Interwell Connectivity from Well-Rate Fluctuations in Waterfloods. In Proceedings of the SPE/DOE Thirteenth Symposium on Improved Oil Recovery, Tulsa, OK, USA, 13–17 April 2002. [Google Scholar]
- Zhang, X. Microscopic Characterization and Identification Criteria for Water Flow Dominance Channels. Inn. Mong. Petrochem. Ind. 2013, 16, 146–149. [Google Scholar]
- Jeong, J.; Park, E.; Emelyanova, I.; Pervukhina, M.; Esteban, L.; Yun, S.T. Interpreting the Subsurface Lithofacies at High Lithological Resolution by Integrating Information from Well-Log Data and Rock-Core Digital Images. J. Geophys. Res.-Solid Earth 2020, 125, e2019JB018204. [Google Scholar] [CrossRef]
- Mahdy, A.; Zakaria, W.; Helmi, A.; Helaly, A.S.; Mahmoud, A.M.E. Machine learning approach for core permeability prediction from well logs in Sandstone Reservoir, Mediterranean Sea, Egypt. J. Appl. Geophys. 2024, 220, 105249. [Google Scholar] [CrossRef]
- Lunny, C. Well Log Channel Matching. U.S. Patent 202017247580, 17 December 2020. [Google Scholar]
- Anees, A.; Shi, W.; Ashraf, U.; Xu, Q. Corrigendum to “Channel identification using 3D seismic attributes and well logging in lower Shihezi Formation of Hangjinqi area, northern Ordos Basin, China. J. Appl. Geophys. 2022, 197, 104554. [Google Scholar] [CrossRef]
- Sun, L.; Li, B.; Li, Y.; Jiao, Y. Method and device for identification of layer section with advantageous channel development based on logging curve. China. Patent 201811508998.0, 11 December 2018. [Google Scholar]
- Wood, D.A. Predicting Total Organic Carbon from Few Well Logs Aided by Well-Log Attributes; DWA Energy Limited: Lincoln, UK, 2023. [Google Scholar]
- Verma, S.; Yılmaz, E.Y.; Sanguino, L.O.; Yandell, J.; Henderson, M.; Trentham, R. Seismic attribute and well-log analysis for channel characterization in the upper San Andres and Grayburg formations of the Midland Basin, Texas. Energy Geosci. 2023, 5, 100188. [Google Scholar] [CrossRef]
- Feng, Q.; Wang, S.; Gao, G.; Li, C. A new approach to thief zone identification based on interference test. J. Pet. Sci. Eng. 2010, 75, 13–18. [Google Scholar] [CrossRef]
- Li, S. Investigation and Application of Integration Approaches for Thief Zone Identification. Master’s Thesis, China University of Petroleum (East China), Dongying, China, 2013. [Google Scholar]
- Izgec, B.; Chevron, S.; Kabir, C.S.; Hess, S. Identification and Characterization of High-Conductive Layers in Waterfloods. SPE Reserv. Eval. Eng. 2011, 14, 113–119. [Google Scholar] [CrossRef]
- Lei, X. Studying the Advantage Channels of Waterflooding Reservoir by Using Reservoir Engineering Methods. Master’s Thesis, China University of Petroleum, Beijing, China, 2018. [Google Scholar]
- Zheng, W.; Jiang, H.; Chen, M.; Zhang, F. Fuzzy comprehensive evaluation for injection-production effectiveness in medium-high water cut stage. In Proceedings of the 2011 Asia-Pacific Power and Energy Engineering Conference, Washington, DC, USA, 25–28 March 2011. [Google Scholar]
- Liu, H. Study on the Identification of Dominant Channel and Profile Control of CO2 Foam in 530 Well Area. Master’s Thesis, Southwest Petroleum University, Chengdu, China, 2019. [Google Scholar]
- Li, Y. Identification and Forewarning Method of Gas Channeling in Gas Injection Development of Clastic Reservoir. Master’s Thesis, China University of Petroleum (East China), Dongying, China, 2021. [Google Scholar]
- Klir, G.J.; Yuan, B. Fuzzy Sets, Fuzzy Logic, and Fuzzy Systems: Selected Papers by Lotfi A Zadeh; Advances in Fuzzy Systems-Applications and Theory; World Scientific: Singapore, 1996; p. 842. [Google Scholar]
- Lu, C.; Liu, H.; Pang, Z.; Dong, X.; Xiu, W. A new profile control design based on quantitative identification of steam breakthrough channel in heavy oil reservoirs (Article). J. Pet. Explor. Prod. Technol. 2014, 4, 17–35. [Google Scholar] [CrossRef]
- Lei, Z.; Qingdong, C. Study and application of identification method of channeling-path in the fractured ultra-low permeability reservoir. Drill. Prod. Technol. 2015, 38, 29–32+8. [Google Scholar]
- Kuenen, P.H.; Migliorini, C.I. Turbidity Currents as a Cause of Graded Bedding. J. Geol. 1950, 58, 91–127. [Google Scholar] [CrossRef]
- Wang, J.B.; Gao, Y.C.; Wang, K.Z. Law of gas breakthrough on CO2 flooding well in ultra-lower permeability reservoir of Yao Yingtai Oilfield. Fault-Block Oil Gas Field 2013, 20, 118–122. [Google Scholar]
- Saaty, T.L. Highlights and critical points in the theory and application of the Analytic Hierarchy Process. Eur. J. Oper. Res. 1994, 74, 426–447. [Google Scholar] [CrossRef]
Scale Value | Meaning |
---|---|
1 | Equal importance |
3 | Moderate importance |
5 | Strong importance |
7 | Very strong importance |
9 | Extreme importance |
2, 4, 6, 8 | Intermediate values between adjacent judgments |
Reciprocal | If the judgment value for factor i relative to factor j is bij, then the judgment value for factor j relative to factor i is the reciprocal 1/bij |
Decision Layer | Static Indicators | Dynamic Indicators |
---|---|---|
Static Indicators | 1 | 1/2 |
Dynamic Indicators | 2 | 1 |
Effective Thickness | Permeability | Planar Heterogeneity | Vertical Heterogeneity | Angle | Sedimentary Rhythm | |
---|---|---|---|---|---|---|
Effective Thickness | 1 | 3 | 1/3 | 1/4 | 1/2 | 2 |
Permeability | 1/3 | 1 | 1/4 | 1/6 | 1/4 | 1/2 |
Planar Heterogeneity | 3 | 5 | 1 | 1/2 | 2 | 4 |
Vertical Heterogeneity | 4 | 6 | 2 | 1 | 3 | 5 |
Angle | 2 | 4 | 1/2 | 1/3 | 1 | 3 |
Sedimentary Rhythm | 1/2 | 2 | 1/4 | 1/2 | 1/3 | 1 |
Injection–Production Gas Ratio | Production GOR | Injection WGR | Production GLR | Injection Intensity | Distance | |
---|---|---|---|---|---|---|
Injection–Production Gas Ratio | 1 | 3 | 5 | 3 | 4 | 2 |
Production GOR | 1/3 | 1 | 3 | 1 | 2 | 1/2 |
Injection WGR | 1/5 | 1/3 | 1 | 1/3 | 1/2 | 1/4 |
Production GLR | 1/3 | 1 | 3 | 1 | 2 | 1/2 |
Injection Intensity | 1/4 | 1/2 | 2 | 1/2 | 1 | 1/3 |
Distance | 1/2 | 2 | 4 | 2 | 3 | 1 |
Target Layer | Weight | Criterion Layer | Weight | Evaluation Layer |
---|---|---|---|---|
Gas Channeling Identification | 0.33 | Static Indicators | 0.101 | Effective Thickness |
0.042 | Permeability | |||
0.252 | Planar Heterogeneity | |||
0.381 | Vertical Heterogeneity | |||
0.160 | Angle | |||
0.064 | Sedimentary Rhythm | |||
0.67 | Dynamic Indicators | 0.364 | Injection–Production Gas Ratio | |
0.136 | Production GOR | |||
0.051 | Injection WGR | |||
0.136 | Production GLR | |||
0.080 | Injection Intensity | |||
0.232 | Distance |
Indicator | Membership Function Type |
---|---|
Effective Thickness | Semi-ascending |
Permeability | Semi-ascending |
Planar Heterogeneity | Semi-ascending |
Vertical Heterogeneity | Semi-ascending |
Angle | Semi-descending |
Injection–Production Gas Ratio | Semi-descending |
Production GOR | Semi-ascending |
Injection WGR | Semi-descending |
Production GLR | Semi-ascending |
Injection Intensity | Semi-ascending |
Distance | Semi-descending |
Evaluation Result | Degree of Gas Channel Development |
---|---|
F < 0.25 | No obvious gas channeling |
0.25 ≤ F < 0.4 | Slight gas channeling |
0.4 ≤ F < 0.6 | Weak gas channeling |
0.6 ≤ F < 0.8 | Strong gas channeling |
0.8 ≤ F ≤ 1.0 | Complete gas channeling |
Inj. Well | Prod. Well | Eff. Thickness (m) | Perm. (mD) | Heterogeneity Coeff. of Perm. | Perm. Ratio | Angle (°) | Sedimentary Rhythm |
---|---|---|---|---|---|---|---|
H + 1 | H1 | 5.00 | 2.159 | 1.601 | 34.868 | 18 | Composite Rhythm |
H2 | 10.35 | 3.036 | 1.695 | 34.868 | 75 | ||
H3 | 5.75 | 1.545 | 1.476 | 34.868 | 20 | ||
H4 | 13.85 | 2.929 | 1.221 | 34.868 | 73 | ||
H + 2 | H2 | 7.70 | 3.033 | 1.533 | 45.271 | 17 | Composite Rhythm |
H5 | 4.85 | 1.857 | 1.550 | 45.271 | 70 | ||
H6 | 4.55 | 3.969 | 1.776 | 45.271 | 20 | ||
H3 | 6.00 | 1.545 | 1.391 | 45.271 | 69 | ||
H + 3 | H3 | 7.50 | 2.586 | 1.368 | 63.826 | 20 | Composite Rhythm |
H6 | 10.05 | 4.728 | 1.515 | 63.826 | 45 | ||
H7 | 8.30 | 4.940 | 1.414 | 63.826 | 79 | ||
H + 4 | H4 | 6.55 | 3.019 | 1.245 | 26.464 | 20 | Composite Rhythm |
H3 | 6.55 | 2.438 | 1.567 | 26.464 | 60 | ||
H7 | 4.95 | 4.463 | 1.240 | 26.464 | 22 | ||
H8 | 7.10 | 4.651 | 1.165 | 26.464 | 72 |
Inj. Well | Prod. Well | Inj.-Prod. Gas Ratio (m3/m3) | Production GOR (m3/t) | Injection WGR (t/m3 × 104) | Production GLR (m3/t) | Injection Intensity (m3/d/m) | Distance (m) |
---|---|---|---|---|---|---|---|
H + 1 | H1 | 47.717 | 713.499 | 1.433 | 57.634 | 11.212 | 266 |
H2 | 17.773 | 2583.187 | 1.433 | 170.145 | 9.868 | 146 | |
H3 | 13.231 | 2261.285 | 1.433 | 148.786 | 12.839 | 202 | |
H4 | 25.539 | 1590.093 | 1.433 | 109.349 | 6.772 | 159 | |
H + 2 | H2 | 12.895 | 2583.187 | 1.513 | 170.145 | 8.893 | 158 |
H5 | 23.804 | 528.980 | 1.513 | 81.236 | 8.646 | 258 | |
H6 | 10.331 | 2040.638 | 1.513 | 174.904 | 11.111 | 214 | |
H3 | 9.600 | 2261.285 | 1.513 | 148.786 | 12.610 | 143 | |
H + 3 | H3 | 14.395 | 2261.285 | 1.324 | 148.786 | 11.385 | 190 |
H6 | 15.491 | 2040.638 | 1.324 | 174.904 | 9.964 | 162 | |
H7 | 20.713 | 1759.276 | 1.324 | 192.175 | 9.822 | 199 | |
H + 4 | H4 | 27.545 | 1590.093 | 1.441 | 109.349 | 11.920 | 206 |
H3 | 14.270 | 2261.285 | 1.441 | 148.786 | 13.794 | 178 | |
H7 | 20.534 | 1759.276 | 1.441 | 192.175 | 11.604 | 280 | |
H8 | 15.123 | 891.872 | 1.441 | 150.445 | 14.709 | 154 |
Indicators | Boundary Limits of Membership Degree | ||
---|---|---|---|
0 | (x − a)/(b − a) or (b − x)/(b − a) | 1 | |
Effective Thickness (m) | <1 | [1, 15] | >15 |
Permeability (mD) | <0.01 | [0.01, 5] | >5 |
Heterogeneity Coeff. of Perm. | <1 | [1, 5] | >5 |
Permeability Ratio | <10 | [10, 100] | >100 |
Angle (°) | >80 | [10, 80] | <10 |
Sedimentary Rhythm | Normal | Composite = 0.5 | Reverse |
Inj.-Prod. Gas Ratio (m3/m3) | >50 | [1, 50] | <1 |
Production GOR (m3/t) | <500 | [500, 2500] | >2500 |
Injection WGR (t/m3 × 104) | >2 | [0.01, 2] | <0.01 |
Production GLR (m3/t) | <25 | [25, 250] | >250 |
Injection Intensity (m3/d/m) | <2 | [2, 15] | >15 |
Distance (m) | >300 | [100, 300] | <100 |
Inj. Well | Prod. Well | Static Indicators Evaluation Results | Dynamic Indicators Evaluation Results | Comprehensive Evaluation Results | Gas Channeling Level | Evaluation Results Verification |
---|---|---|---|---|---|---|
H + 1 | H1 | 0.480 | 0.162 | 0.27 | Slight | True |
H2 | 0.400 | 0.687 | 0.59 | Weak | True | |
H3 | 0.317 | 0.663 | 0.55 | Weak | True | |
H4 | 0.353 | 0.515 | 0.46 | Weak | True | |
H + 2 | H2 | 0.419 | 0.725 | 0.62 | Strong | True |
H5 | 0.392 | 0.333 | 0.35 | Slight | True | |
H6 | 0.445 | 0.659 | 0.59 | Weak | True | |
H3 | 0.371 | 0.755 | 0.63 | Strong | True | |
H + 3 | H3 | 0.504 | 0.662 | 0.61 | Strong | True |
H6 | 0.565 | 0.679 | 0.64 | Strong | True | |
H7 | 0.538 | 0.587 | 0.57 | Weak | True | |
H + 4 | H4 | 0.243 | 0.477 | 0.40 | Weak | True |
H3 | 0.292 | 0.689 | 0.56 | Weak | True | |
H7 | 0.243 | 0.503 | 0.42 | Weak | True | |
H8 | 0.248 | 0.624 | 0.50 | Weak | True |
Inj. Well | Prod. Well | Layer No. | Eff. Thickness (m) | Perm. (mD) | Heterogeneity Coeff. of Perm. | Perm. Ratio | Comprehensive Evaluation Results |
---|---|---|---|---|---|---|---|
H + 3 | H6 | 1 | 2.90 | 5.741 | 1.723 | 2.029 | 0.538 |
2 | 0.40 | 0.348 | 1.364 | 1.552 | 0.178 | ||
3 | 3.55 | 5.033 | 1.577 | 4.702 | 0.685 | ||
4 | 2.15 | 5.494 | 1.294 | 8.879 | 0.632 | ||
5 | 1.05 | 1.005 | 1.239 | 1.453 | 0.162 |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Liu, Y.; Hao, M.; Bi, R.; Bian, C.; Wang, X. Research on Gas Channeling Identification Using the Fuzzy Comprehensive Evaluation Method. Energies 2024, 17, 3908. https://doi.org/10.3390/en17163908
Liu Y, Hao M, Bi R, Bian C, Wang X. Research on Gas Channeling Identification Using the Fuzzy Comprehensive Evaluation Method. Energies. 2024; 17(16):3908. https://doi.org/10.3390/en17163908
Chicago/Turabian StyleLiu, Yang, Mingqiang Hao, Ran Bi, Chaoliang Bian, and Xiaoqing Wang. 2024. "Research on Gas Channeling Identification Using the Fuzzy Comprehensive Evaluation Method" Energies 17, no. 16: 3908. https://doi.org/10.3390/en17163908
APA StyleLiu, Y., Hao, M., Bi, R., Bian, C., & Wang, X. (2024). Research on Gas Channeling Identification Using the Fuzzy Comprehensive Evaluation Method. Energies, 17(16), 3908. https://doi.org/10.3390/en17163908