Abstract: Recovering Low Frequencies for Impedance Inversion by Frequency Domain Deconvolution; #90224 (2015)

Sina Esmaeili and Gary Frank

Search and Discovery.com

... reflectivity. We start by reintroducing the convolutional model for normal incident seismograms and then show how reflectivity can be estimated...

Multi-conditional seismic velocity synthesis using an autoregressive transformer network

Randy Harsuko, Shijun Cheng, Tariq Alkhalifah

International Meeting for Applied Geoscience and Energy (IMAGE)

... in a generative model. These stored velocity model distributions can be utilized to regularize inverse problems, like full waveform inversion. However, most...

Uncertainty quantification of single and multi-parameter full-waveform inversion through a variational autoencoder

Abdelrahman Elmeliegy, Mrinal Sen, Jennifer Harding, Hongkyu Yoon

International Meeting for Applied Geoscience and Energy (IMAGE)

.... The input to the network is seismic shot gathers and the output are samples (distribution) of model parameters. We then use these samples to estimate...

Finding the reservoir in an intensely scattered wavefield: A case study from the Central Sichuan Basin, China

John Coffin, Mengmeng Zhang, Panos Doulgeris, Peter Haffinger, Angie Kelsay, Jinsong Li

International Meeting for Applied Geoscience and Energy (IMAGE)

... an elastic modelling scheme including interbed multiples and multiple mode conversions for the well-tie. A linear convolutional model results in perfectly...

Survey merging using CycleGAN and patchy seismic images

Chaoshun Hu, Fan Jiang, Konstantin Osypov, Julianna Toms

International Meeting for Applied Geoscience and Energy (IMAGE)

... Resnet convolutional neural network. Discriminator is using the VGG16 pretrained model. Loss function is based on the SRGAN perceptional content loss...

Abstract: Fast and Accurate Impedance Inversion by Well-Log Calibration; #90171 (2013)

Igor B. Morozov and Jinfeng Ma

Search and Discovery.com

...) the convolutional equation; 2) time-depth constraints from the seismic data, 3) background low-frequency model from the logs or seismic/geological interpretation...

Anti-aliasing seismic data interpolation by dip-informed self-supervised learning

Shirui Wang, Xuqing Wu, Jiefu Chen

International Meeting for Applied Geoscience and Energy (IMAGE)

... an event of angle α is marked. (b) A traditional (3 × 3) convolutional kernel. (c) A deformed (3 × 3) convolutional kernel offset along the event...

Abstract: Real-time Bit Wear Prediction and Deployment Validation in Challenging Hard and Heterogeneous Sandstones using 3D Detailed and Simplified Physics-Based Progressive Wear Models; #91204 (2023)

Guodong (David) Zhan, William B. Contreras Otalvora, Xu Huang, Reed Spencer, John Bomidi

Search and Discovery.com

... formulas based on decades of drilling mechanics research are implemented. The detailed physics-based model was used in the design workflow to simulate...

Incorporating Artificial Intelligence into Traditional Exploration Workflows in the Cooper-Eromanga Basin, South Australia

H. M. Garcia, W. G. "Woody" Leel Jr., M. Riehle, P. Szafian

International Meeting for Applied Geoscience and Energy (IMAGE)

... a detail geological model at the seismic resolution. This was complemented with the structural information from the AI. Combining all the information...

Understanding the Seismic Wavelet; Steven G. Henry; Search and Discovery Article #40028 (2001)

Search and Discovery.com

Predicting Coiled-Tubing Drilling Dynamics Using Transformers

Carlos Urdaneta, Cheolkyun Jeong, Xuqing Wu, Jiefu Chen

Unconventional Resources Technology Conference (URTEC)

... tailored for CTD drilling time series data. This model leverages its ability to learn long-term dependencies and subtle patterns within the data...

Machine-Learning Driven Prediction of Geological Marker Signatures

Shashin Sharan, Kaustubh Shrivastava, Per Irgens, Tatjana Scherschel

Unconventional Resources Technology Conference (URTEC)

... to predict markers. The procedure involves three steps: data conditioning, feature extraction, and model training. Initially, log data is preprocessed...

Seismic random noise attenuation via enhanced similarity self-supervised learning

Jiale Wang, Naihao Liu, Yihuai Lou, Jinghuai Gao

International Meeting for Applied Geoscience and Energy (IMAGE)

... the requirement of noise-free labels, we propose an enhanced similarity self-supervised learning (ESSL) model by effectively utilizing the self-similarity...

A method of relative impedance holography-inversion based on reflection coefficient inversion

Jiangfeng Zheng, Jialin Sun, Zongyu Zhen, Shaoxuan Li

International Meeting for Applied Geoscience and Energy (IMAGE)

...(a) is the reservoir model, (b) is the reflectivity of model, (c) is the trace integration of reflection coefficient, (d) is the trace integration of actual...

Unlocking hidden potential in shallow water Gulf of Mexico legacy data for carbon capture and storage exploration

Rachel Collings, Igor Marino, Adriana Arroyo Acosta, Jack Kinkead, Hugo Medel, Trong Tang, Gabriela Suarez, Brett Sellers

International Meeting for Applied Geoscience and Energy (IMAGE)

... deployed a comprehensive wavelet processing workflow. To obtain a high-resolution velocity model, a seismic inversion workflow was implemented...

Prestack Seismic Data Inversion for Shale Gas Reservoir Characterization in China; #41829 (2016)

Gang Yu, Yusheng Zhang, Uwe Strecker, Maggie Smith

Search and Discovery.com

... of Horizontal Transverse Isotropy (HTI) the model is parallel vertical cracks (Thomsen, 1988). P waves travel with different velocities depending...

Subsurface hydrogen storage: A feasibility study

Dwaipayan Chakraborty, Subhashis Mallick, Mortezaa Dejam

International Meeting for Applied Geoscience and Energy (IMAGE)

... models. The original model before H2 replacement is in cyan, the model after replacement is in black and the inverted model is in red. (b) observed...

Improving subsalt imaging using mode-converted waves and volumetric coherent noise attenuation in southern Gulf of Mexico

Riaz Alai, Nik Nur Halim C. Soh, M Shah B. Sulaiman, M. Iqbal Supardy, Kien Kok Lee, Sandeep Kumar, Syazwani Bt Suhairi, Christian Brinzer, Frederico Xavier de Melo, Hongyan Li, Iestyn Williams, Hugo Enrique Munoz Cuenca, Nolan Brand, Kate Glaccum, George Zhao, Saeeda Hydal, Khaled Abdelaziz, Emmi Sanchez Vargas

International Meeting for Applied Geoscience and Energy (IMAGE)

... originating from the base of salt bodies. All models are derived from convolutional-based (surface multiples), dip guided selective stacking...

Interbed Multiple Suppression in Carbonate Sequences

Gabriel Gil

Unconventional Resources Technology Conference (URTEC)

... transform. The presented workflow shows a model-based approach to predict and suppress the presence of interbed multiples from migrated pre-stack data...

Abstract: Cost Efficient Acquisition to Reduce Coarse Land 3D Line Spacings Through Beyond Nyquist Interpolation and Wavefield Reconstruction for Signal and Noise; #90187 (2014)

Bill Goodway

Search and Discovery.com

... not exceed Nyquist. Both authors concluded that the assumption of a smoothly varying linear model for the wavefield (or a plane wave decomposition...

Hierarchical machine learning workflow for conditional and multiscale deep-water reservoir modeling

Wen Pan, Honggeun Jo, Javier E. Santos, Carlos Torres-Verdín, and Michael J. Pyrcz

AAPG Bulletin

... a diverse ensemble of conditional, multiscale, architectural and petrophysical, property model realizations reproducing the patterns. We demonstrate...

Earth Model Building in Real-Time with an Automated Machine Learning Framework - A Midland Basin Example

Altay Sansal, Muhlis Unaldi, Edward Tian, Gareth Taylor

Unconventional Resources Technology Conference (URTEC)

...Earth Model Building in Real-Time with an Automated Machine Learning Framework - A Midland Basin Example Altay Sansal, Muhlis Unaldi, Edward Tian...

Research on first break picking based on deep learning for DAS-VSP data

Naijian Wang, Yinpo Xu, Yuxin Hou, Yingjie Pan, Mingxing Wang, Chun Zhang, Tianfu Yang

International Meeting for Applied Geoscience and Energy (IMAGE)

... of whole zone are covered; thirdly, the U-Net network is improved by adjusting the model hierarchy, optimizing the incentive function and adding batch...

Label-defect tolerance ability of deep learning inversion networks and its applications

Yang Ping, Xu Hunqun, Liu Di, Tao Chunfeng, Wang Chengxiang, Yue Changqing

International Meeting for Applied Geoscience and Energy (IMAGE)

... inversion accuracy. Through model testing, we confirm that Deep Learning (DL) networks have the ability to tolerate labels’ defects, which we name...

Enabling grassroots digital transformation with a Python-Excel ML toolkit

Andrew C. Silver

International Meeting for Applied Geoscience and Energy (IMAGE)

... regression, support-vector machine (SVM), decision tree, random forest, boosted tree (including AdaBoost and XGBoost), and convolutional neural network (CNN...