Adaptive prestack inversion method based on quadratic encoder-decoder network
SHAN Bo1,2(), XING Yu-Xin1,2(), ZHANG Fan-Chang3, LI Zhi-Wei1,2, CHEN Mo1,2
1. Oil & Gas Resources Survey, China Geological Survey, Beijing 100083, China 2. Key Laboratory of Unconventional Oil and Gas, China Geological Survey, Beijing 100083, China 3. School of Geosciences, China University of Petroleum (East China), Qingdao 266580, China
AVO inversion, based on the Zoeppritz equation, extracts various hidden petrophysical parameters from pre-stack seismic data. In seismic data, angle information is recorded in the form of offset values, and converting between offset values and angles is prone to generate errors. In addition, using the same approximate formula for different acreage types may lead to reduced applicability due to varying actual geological conditions. The exact Zoeppritz equation will lead to increased computational demands due to its high complexity. Therefore, this study developed an adaptive prestack inversion method based on the quadratic Encoder-Decoder network. This inversion method used the high feature and relationship extraction abilities of deep learning to replace traditional relationships, thereby reducing the angle errors and adapting to varying acreage types and geological conditions. The quadratic Encoder-Decoder network used a quadratic algorithm as the optimization method, maximizing the efficiency of the standard Encoder-Decoder structure. Additionally, the Xavier initialization method was incorporated to enhance the randomness of model initialization, thus improving the robustness of the network. The results indicate that the quadratic Encoder-Decoder network, selected through orthogonal experiments, outperforms the single-decoder network in prediction and exhibits greater consistency with actual log curves. The P-wave velocity, S-wave velocity, and density profiles obtained from inversion are consistent with the geological conditions of the study area, exhibit strong lateral continuity, and can effectively achieve high-precision prestack inversion.
Zoeppritz K. On the reflection and propagation of seismic waves at discontinuities[J]. Erdbebenwellen VII B,1919:66-84.
[2]
Bortfeld R. Approximations to the reflection and transmission coefficients of plane longitudinal and transverse WAVES[J]. Geophysical Prospecting, 1961, 9(4):485-502.
[3]
Aki K, Richards P G. Quantitative seismology:Theory and methods[M]. San Francisco: Freeman W H,1980:133-154.
[4]
Shuey R T. A simplification of the Zoeppritz equations[J]. Geophysics, 1985, 50(4):609.
[5]
Fatti J L, Smith G C, Vail P J, et al. Detection of gas in sandstone reservoirs using AVO analysis:A 3-D seismic case history using the geostack technique[J]. Geophysics, 1994, 59(9):1362-1376.
[6]
Goodway B, Chen T W, Downton J. Improved AVO fluid detection and lithology discrimination using Lamé petrophysical parameters; “λρ”,“μρ”,&“λ/μ fluid stack”,from P and S inversions[C]// SEG Technical Program Expanded Abstracts 1997.Society of Exploration Geophysicists,1997:148-151.
[7]
Smith G C, Gidlow P M. Weighted stacking for rock property estimation and detection of gas[J]. Geophysical Prospecting, 1987, 35(9):993-1014.
Zong Z Y, Yin X Y, Zhang F, et al. Reflection coefficient equation and pre-stack seismic inversion with Young’s modulus and Poisson ratio[J]. Chinese Journal of Geophysics, 2012, 55(11):3786-3794.
[9]
宋晓雨. 准噶尔盆地三叠系致密砂岩甜点地震预测[D]. 东营: 中国石油大学(华东), 2020.
[9]
Song X Y. Earthquake prediction of Triassic tight sandstone dessert in Junggar basin[D]. Dongying: China University of Petroleum (Huadong), 2020.
[10]
Zhang F C, Yang J Y, Li C H, et al. Direct inversion for reservoir parameters from prestack seismic data[J]. Journal of Geophysics and Engineering, 2020, 17(6):993-1004.
[11]
宋宝山. 多波AVO叠前反演及应用研究[D]. 北京: 中国石油大学(北京), 2021.
[11]
Song B S. Multi-wave AVO prestack inversion and its application[D]. Beijing: China University of Petroleum (Beijing), 2021.
[12]
Zhou L, Liu X Y, Li J Y, et al. Robust AVO inversion for the fluid factor and shear modulus[J]. Geophysics, 2021, 86(4):R471-R483.
Fan X H, Yang J Y, Shan B. Approximate equations of shear modulus and density reflection coefficient and prestack seismic inversion[J]. Geophysical Prospecting for Petroleum, 2022, 61(5):830-841.
[14]
Zhu X F, McMechan G. AVO inversion using the Zoeppritz equation for PP reflections[C]// SEG Technical Program Expanded Abstracts 2012. Society of Exploration Geophysicists,2012:1-5.
Huang H D, Wang Y C, Guo F, et al. High precision prestack inversion algorithm based on Zoeppritz equations[J]. Oil Geophysical Prospecting, 2013, 48(5):740-749,672,854.
[16]
Zhi L X, Chen S Q, Li X Y, et al. An improved strategy for exact Zoeppritz equations AVA inversion[C]// SEG Technical Program Expanded Abstracts 2015.New Orleans,Louisiana. Society of Exploration Geophysicists,2015:654-658.
Wang Z T, Shen G Q, Wang Y M, et al. Model driven high-precision prestack seismic inversion[J]. Geophysical Prospecting for Petroleum, 2020, 59(6):927-935.
doi: 10.3969/j.issn.1000-1441.2020.06.011
Wang H, Yan J Y, Fu G M, et al. Current status and application prospect of deep learning in geophysics[J]. Progress in Geophysics, 2020, 35(2):642-655.
[19]
Yu S W, Ma J W. Deep learning for geophysics:Current and future trends[J]. Reviews of Geophysics, 2021, 59(3):e2021RG000742.
Zhang L F, Wang J Y, Zhang M S, et al. Evaluation of regional landslide susceptibility assessment based on BP neural network[J]. Northwestern Geology, 2022, 55(2):260-270.
Qiu Q J, Ma K, Zhu H H, et al. BERT-based method and significance of constraint information extraction for 3D geological modelling[J]. Northwestern Geology, 2022, 55(4):124-132.
Yng L Q, Wang S D, Li J M. Few-shot pre-stack AVO inversion using a multi-task Transformer[J/OL]. Progress in Geophysics, 2024[2024-06-11]. http://kns.cnki.net/kcms/detail/11.2982.P.20240611.1523.062.html.
[23]
Biswas R, Sen M K, Das V, et al. Prestack and poststack inversion using a physics-guided convolutional neural network[J]. Interpretation, 2019, 7(3):SE161-SE174.
Sun Y H, Liu Y, Chen T S. Multi-wave amplitude-versus-offset inversion and reservoir fluid identification based on unsupervised deep learning[J]. Geophysical Prospecting for Petroleum, 2021, 60(3):385-394,413.
doi: 10.3969/j.issn.1000-1441.2021.03.004
[25]
Sun Y H, Liu Y, Zhang M, et al. Inversion of low-to medium-frequency velocities and densities from AVO data using invertible neural networks[J]. Geophysics, 2022, 87(3):A37-A42.
Shan B, Zhang F C, Ding J C. S-wave velocity prediction method for sand-shale formation based on quadratic optimization network[J]. Oil Geophysical Prospecting, 2022, 57(1):26-33,4-5.
[27]
Glorot X, Bengio Y. Understanding the difficulty of training deep feedforward neural networks[C]. Proceedings of the thirteenth international conference on artificial intelligence and statistics. JMLR Workshop and Conference Proceedings,2010:249-256.
[28]
常兆光, 王清河, 杜彩凤. 应用统计方法[M]. 北京: 石油工业出版社, 2009.
[28]
Chang Z G, Wang Q H, Du C F. Applied statistical method[M]. Beijing: Petroleum Industry Press, 2009.
