A geophysical prior knowledge guided semisupervised deep learning framework for AVA inversion

Lei Zhu

International Meeting for Applied Geoscience and Energy (IMAGE)

... forward model. This reduces the dependence of the framework on training data. This GPKGS framework preserves the physical process of AVA inversion, making...

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)

... a novel workflow based on a deep convolutional neural network for automatically detecting salt geometry from a seismic image. By developing...

An integrated machine learning-based fault classification workflow

Jie Qi, Carolan Laudon, Kurt Marfurt

International Meeting for Applied Geoscience and Energy (IMAGE)

... on the human interpreter. We first compute a 3D fault probability volume from pre-conditioned seismic amplitude data using a 3D convolutional neural network...

Automatic microseismic event detection in downhole DAS data through convolutional neural networks: A comparison of events during and post-stimulation of the well

Paige Given, Fantine Huot, Ariel Lellouch, Bin Luo, Robert G. Clapp, Biondo L. Biondi, Tamas Nemeth, Kurt Nihei

International Meeting for Applied Geoscience and Energy (IMAGE)

... present a convolutional neural network (CNN) which takes inputted images from DAS arrays and accurately detects microseismic events. Our model is able...

Seismic resolution enhancement with self-supervised learning

Shijun Cheng, Tariq Alkhalifah, Haoran Zhang

International Meeting for Applied Geoscience and Energy (IMAGE)

... et al. (2023) incorporated the convolutional model into the loss function in a self-supervised manner to constrain the network’s predicted outcomes...

Abstract: Machine Learning and Deep Learning in Oil and Gas Industry: A Review Ofapplications, Opportunities and Challenges; #91204 (2023)

Tejas Balasaheb Sabale, Syed Aaquib Hussain, Mohd Zuhair, Mohammad Saud Afzal, Arnab Ghosh

Search and Discovery.com

...] S. Madasu and K. P. Rangarajan, “Deep recurrent neural network DRNN model for real-time multistage pumping data,” OTC Arct. Technol. Conf. 2018, 2018...

Application of Machine Learning and Deep Learning for Complex Fault Network Characterizationon the North Slope, Alaska

Search and Discovery.com

Unsupervised deep learning for seismic data reconstruction

Gui Chen, Yang Liu, Mi Zhang

International Meeting for Applied Geoscience and Energy (IMAGE)

... for reconstructing missing traces in observed seismic data. While many DL-based reconstruction methods employ convolutional neural networks (CNNs...

Seismic data augmentation for automatic faults picking using deep learning

Nam Pham, Sergey Fomel

International Meeting for Applied Geoscience and Energy (IMAGE)

... these newly generated data for training a convolutional neural network for faults picking, as the training data will resemble the field test data...

Unconventional Reservoir Microstructural Analysis Using SEM and Machine Learning

Amanda S. Knaup, Jeremy D. Jernigen, Mark E. Curtis, John W. Sholeen, John J. Borer IV, Carl H. Sondergeld, Chandra S. Rai

Unconventional Resources Technology Conference (URTEC)

... an adequate statistical database, a large number of images need to be processed which is labor intensive, time consuming, and requires trained technicians...

Time-lapse seismic data shaping with transformer encoder neural networks

Jorge E. Monsegny, Daniel O. Trad, Don C. Lawton

International Meeting for Applied Geoscience and Energy (IMAGE)

... repeatability. The datasets were processed with the same flow and reverse time migration (RTM) images were generated using the same velocity model. We...

Deep convolutional neural networks for generating grain-size logs from core photographs

Thomas T. Tran, Tobias H. D. Payenberg, Feng X. Jian, Scott Cole, and Ishtar Barranco

AAPG Bulletin

... and time. In this paper, a method is presented to characterize grain size rapidly and consistently from core photographs using deep convolutional neural...

Abstract: Machine Learning Receiver Deghosting - Shallow Water OBN Data Example; #91204 (2023)

Rolf Baardman, Rob Hegge, Jewoo Yoo

Search and Discovery.com

...: The proposed supervised ML-method uses a convolutional neural network (CNN) with a two-channel input layer (P and Vz) and an output layer containing the up...

Automatic well-log baseline correction via deep learning for rapid screening of potential CO2 storage sites

Misael M. Morales, Carlos Torres-Verdín, Michael Pyrcz, Murray Christie, Vladimir Rabinovich

International Meeting for Applied Geoscience and Energy (IMAGE)

... convolutional U-Net model to estimate the baselinecorrected SP log from the raw SP log and a set of collocated predictor features based on feature engineering...

Pyseis: A high-performance, user-friendly Python package for GPU-accelerated seismic modeling and subsurface imaging

Stuart Farris, Guillaume Barnier, Ettore Biondi, Robert Clapp

International Meeting for Applied Geoscience and Energy (IMAGE)

... NVIDIA V100 GPU. In these tests, we kept the number of time steps constant while varying the size of the model domain, a realistic scenario where high...

Automated hyper-parameter optimization for deep learning framework to simulate boundary conditions for wave propagation

Harpreet Kaur, Sergey Fomel, Nam Pham

International Meeting for Applied Geoscience and Energy (IMAGE)

... of spurious reflections and the padded model used to simulate the unbounded domain (Figures1). We first define hyper-parameters for the deep learning...

Transformer-based network for an efficient ground roll suppression

Randy Harsuko, Omar Saad, Tariq Alkhalifah

International Meeting for Applied Geoscience and Energy (IMAGE)

...). This is attributed to the new components introduced to the model, namely the learnable positional encoding, the 1D convolutional encoder-decoder...

Introduction to Deep Learning: Part I

Hongbo Zhou, Lasse Amundsen, Martin Landrø

GEO ExPro Magazine

... 2017 Copyright © 2018 Geopublishing Limited AI Progression ARTIFICIAL INTELLIGENCE Machine Learning Date, time 0 – No Fraud Salary, m. spend 1...

Rock-physics based time-lapse inversion in Delivery4D: synthetic feasibility study for CO2CRC Otway Project

Stanislav Glubokokvskikh, James Gunning, Tess Dance, Roman Pevzner, Dmitry Popik, Christian Proud

Petroleum Exploration Society of Australia (PESA)

... into time-domain images. As a first step, we perform a well-tie with the log values extracted from the seismic model used for FDTD modelling. Figure 2...

Seismic Facies Segmentation Using Deep Learning

Search and Discovery.com

An unsupervised intelligent stacking velocity analysis based on clustering

Lide Wang, Xingrong Xu, Jie Wu, Huahui Zeng, Yundong Yong, Yanxiang Wang

International Meeting for Applied Geoscience and Energy (IMAGE)

... the 3D velocity of (a) manual and (b) the intelligent method we proposed. An intelligent stacking velocity analysis method in time domain based on Mean...

Abstract: 3-D Volumetric Interpretation with Computational Stratigraphy Models

Lisa Goggin, Tao Sun, Maisha Amaru, Ashley Harris, Anne Dutranois, Andrew Madof

Houston Geological Society Bulletin

... of a fluvially-dominated delta was created. The depositional model is converted into seismic volumes of various frequencies (1D convolutional approach...

A self-attention enhanced encoder-decoder network for seismic data denoising

Stefan Knispel, Jan Walda, Ruediger Zehn, Alexander Bauer, Dirk Gajewski

International Meeting for Applied Geoscience and Energy (IMAGE)

... noise. A fundamental step of seismic processing is the denoising of the measured seismic data, which is often highly time-consuming. In recent years...

Explainable machine learning for hydrocarbon prospect risking

Ahmad Mustafa, Ghassan AlRegib

International Meeting for Applied Geoscience and Energy (IMAGE)

... but perform poorly on test data; a distributional shift in the test data at run time could lead to significant degradation in model performance...

Abstract: Impedance Inversion of Blackfoot 3D Seismic Dataset; #90171 (2013)

A. Swisi and Igor B. Morozov

Search and Discovery.com

... by using the methods below. 2) Model-based inversion is also called blocky inversion. This method is based on the convolutional seismic model: S =W * R + n...