Adaptive regularization parameter estimation in image domain least-squares migration

Shen Sheng, Huazhong Wang

International Meeting for Applied Geoscience and Energy (IMAGE)

... image domain LSM. = Fx * rlocal , i where * denotes convolution operator， rlocal is the active local patch of the full model r . By introducing...

Multiple prediction and removal via zero offset SRME: a case study from the Tarim Basin

Yalin Li, Haijun Yang, Gengxin Peng, Duoming Zheng, Ruirui Zhao, Chuanwen Sun

International Meeting for Applied Geoscience and Energy (IMAGE)

... convolution of multiple reflection is consistent with the actual second reflection in terms of reflection time, phase and amplitude spectrum. (a) (a) (b...

Research on fault-karst reservoir identification method based on deep convolutional network

Zhipeng Gui, Junhua Zhang, Hong Zhang, Dong Chen, Pengbo Yin

International Meeting for Applied Geoscience and Energy (IMAGE)

...) in this paper, as shown in Figure 1c. It consists of a 3x3x3 grouped convolution and a 1x1x1 convolution layer. The number of groups in the grouped...

S-wave velocity prediction using a deep learning scheme and attention mechanism

Gang Feng, Wen-Qin Liu, Zhe Yang, Wei Yang, Jian-Hua Wang

International Meeting for Applied Geoscience and Energy (IMAGE)

... Where, Re LU ( x ) = max ( x ) , Wc and bc represent the weight matrix and bias, respectively, and * represents the convolution operation, X represents...

Time-lapse seismic joint inversion based on Zoeppritz equation in depth domain

Wenjin Li, Jingye Li, Lei Han, Zhen Liao

International Meeting for Applied Geoscience and Energy (IMAGE)

.... Then, the precise Zoeppritz equation is used to calculate the reflection coefficient in the depth domain and perform non-stationary convolution...

Characterization of Early-Time Performance of a Well with a Vertical Fracture Producing at Constant Pressure

Nicholas Wiewiorowski, Alex Valdes-Perez, Thomas A. Blasingame

Unconventional Resources Technology Conference (URTEC)

..., then apply the convolution integral to produce the constant pressure condition — all in the Laplace domain. In order to generate various scenarios...

Deconvolution of Time-Varying Bottomhole Pressure Improves Rate-Time Models History Matches and Forecasts of Tight-Oil Wells Production

Leopoldo M. Ruiz Maraggi, Larry W. Lake, Mark P. Walsh

Unconventional Resources Technology Conference (URTEC)

...-phase slightly compressible fluid solution. Second, we present the concepts of time superposition, convolution, and deconvolution. Third, we present...

3D CNN for channel identification in seismic volume

Haishan Li, Wuyang Yang, Xiangyang Zhang, Xinjian Wei, Xin Xu

International Meeting for Applied Geoscience and Energy (IMAGE)

... according to convolution principle. Finally, the labeled dataset is use to train the 3D ResNet, so as to give the 3D ResNet efficient and accurate...

A deep-learning workflow for weak reflection extraction in pitch-catch measurements in cased-hole

Qiang Wang, Hua Wang, Shaopeng Shi

International Meeting for Applied Geoscience and Energy (IMAGE)

... colors are corresponding to A0+casing reflection and TIE, respectively. We use a 1-D convolution kernel M (kernel size is 256) to convolute with each...

Deep-learning application of salt geometry detection in deep water Brazil

Ruichao Ye, Anatoly Baumstein, Kirk A. Wagenvelt, Erik R. Neumann

International Meeting for Applied Geoscience and Energy (IMAGE)

...-sided arrow indicates sharing parameters for deepest 3 layers. Green block after Decoder stands for additional layers with 2D convolution and sigmoid...

Optimum Deconvolution for Seismic Inversion Purposes

M. Suprajitno, L. Nutt, P. Busono

Southeast Asia Petroleum Exploration Society (SEAPEX)

...; Tribolet, 1975) is particularly useful for separating signals that have been combined through convolution as the case with the seismic trace...

Joint inversion of magnetotelluric and seismic travel time data with intelligent interpretation of geophysical models

Hongyu Zhou, Rui Guo, Maokun Li, Fan Yang, Shengheng Xu, Zuzhi Hu, Deqiang Tao

International Meeting for Applied Geoscience and Energy (IMAGE)

... post-stack data can be calculated from the convolution between reflectivity and ricker wavelet, that is Si = Ri ∗Wi + ni (9) (13a) G (sk ) = G (sk−1...

Know Before You Go!

Lars Zühlsdorff, Trond Haugland

GEO ExPro Magazine

..., as generated from ray-based modeling, may act as 3D convolution wavelets that can be directly applied to reservoir models (or, more specifically, to incident...

An MT data inversion method constrained by seismic texture

Hongyu Zhou, Rui Guo, Maokun Li, Fan Yang, Shengheng Xu, Aria Abubakar

International Meeting for Applied Geoscience and Energy (IMAGE)

... of the seismic section. FORMULATION Texture extraction operator The feature maps generated by convolution layers of a CNN G is a high level and abstract...

Multichannel seismic deconvolution via 2D K-SVD and convolutional sparse coding

Guiqian Zhang, Xiayu Gao, Bangli Zou, Yaojun Wang, Yingzhu Chen

International Meeting for Applied Geoscience and Energy (IMAGE)

..., each trace of the poststack seismic profile is considered to be the convolution of the source wavelet with a reflectivity series. The single channel...

Strike-slip fault skeletonization based on deep learning cascade ant tracking method

Zhipeng Gui, Junhua Zhang, Rujun Wang, Yintao Zhang, Chong Sun, Mei Yang

International Meeting for Applied Geoscience and Energy (IMAGE)

... convolution layers in the path of receiving the encoder feature map, and the number of these convolution layers depends on pyramid-level blocks. Dense...

Bi-directional LSTM-based non-causal deconvolution

G. Roncoroni, I. Deiana, E. Forte, M. Pipan

International Meeting for Applied Geoscience and Energy (IMAGE)

... traces, respectively; r(t) is a randomly distributed reflection coefficient series and n1(t) represents a random noise component added before convolution...

Application of fractional Gaussian fractal-based stochastic seismic inversion to F3-Netherland dataset

Srishti Singh, Sreeja Kalyankar, Ajay Malkoti, Nimisha Vedanti

International Meeting for Applied Geoscience and Energy (IMAGE)

...). The inverted impedance model are used to generate synthetic seismograms through forward convolution modelling. These inverted synthetic Data...

New Joint Categorical/Continuous Simultaneous Inversion Technology (Paper C9)

M. Kemper

Geological Society of Malaysia (GSM)

... angles θ, typically in some least squares sense (solving for AI, SI and ρ directly is possible also; not discussed here). Because of the convolution...

Venn Diagram To Represent Rock Attributes: METHOD PAPER

Carol J. Pudsey

Journal of Sedimentary Research (SEPM)

..., convolution, bioturbation, etc.; 3) surface features of pebbles--ice striae, percussion marks, borings, overgrowths, etc.; 4) framework constituents...

Abstract: Particle Swarm and Differential Evolution Optimization - Global Optimization for Geophysical Inversion, by Puneet Saraswat; #90105 (2010)

Search and Discovery.com

What Is Deconvolution?, by Robert E. Sheriff, #40132 (2004).

Search and Discovery.com

Unraveling Complex Oil Mixtures in the Williston Basin; #40941 (2012)

John M. Moldowan, David Zinniker, and Jeremy E. Dahl

Search and Discovery.com

... correlation of both black oil and cracked oil sources and de-convolution of their mixtures. Unraveling Complex Oil Mixtures in the Williston Basin J...

Reservoirs Characterizing Based on Spectral Difference Anomaly at Lower-Frequency on Multi-Angle Stacking Gathers: Case Studies from China; #41256 (2013)

Xuehua Chen

Search and Discovery.com

..., and we do not decimate the results of the convolution Cvetkovic and Marfurt (2007), inspired by Yu et al. (2002), found that applying filters based...

Spectral Decomposition Helps Define Channel Morphology, by Rongfeng Zhang; #41272 (2013).

Search and Discovery.com