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...

Deep Learning Applied to Fault Interpretation and Attribute Computation

Search and Discovery.com

Abstract: Unsupervised Segmentation of Rock MicroCT Scans Using Deep Learning;

Fernando Bordignon, Giovanni Formighieri, Eduardo Burgel, Bruno Rodrigues

Search and Discovery.com

... of DNNs when working with images are the Convolutional Neural Networks (CNN), usually employed in conjunction with supervised training, which needs...

Using Machine Learning Methods to Identify Coals from Drilling and Logging-While-Drilling LWD Data

Ruizhi Zhong, Raymond L. Johnson Jr., Zhongwei Chen

Unconventional Resources Technology Conference (URTEC)

... (Schmidhuber 2015). Supervised learning is learning a predictive model that maps certain inputs to a desired output. To build the supervised learning...

Well Performance and Completion Efficiency Assessment in the Delaware Basin using the Diffusive Time of Flight

Jaeyoung Park, Yuxing Ben, Vivek Muralidharan

Unconventional Resources Technology Conference (URTEC)

... et al. 2019, Park et al. 2020, Santos et al. 2020, Zalavadia and Gildin 2021), yet the numerical simulation still requires a static geologic model...

Perceptual quality-based model training under annotator label uncertainty

Chen Zhou, Mohit Prabhushankar, Ghassan AlRegib

International Meeting for Applied Geoscience and Energy (IMAGE)

...Perceptual quality-based model training under annotator label uncertainty Chen Zhou, Mohit Prabhushankar, Ghassan AlRegib Perceptual quality-based...

Introduction to Deep Learning: Part I

Hongbo Zhou, Lasse Amundsen, Martin Landrø

GEO ExPro Magazine

... that showed that computers could perform tasks once thought to be solely the domain of human capability. However, lack of computer power soon stopped...

Enhancing Lithology Classification through a Deep Learning Framework

P. Zhang, T. Gao, R. Li

Unconventional Resources Technology Conference (URTEC)

..., more data typically improves model accuracy but also increases costs, this research optimizes the utility of existing and common logs. To leverage...

Amplitude enhancement of far-offset refractions via machine learning

Lurun Su, Han Wang, Jie Zhang

International Meeting for Applied Geoscience and Energy (IMAGE)

... refractions using a physical model. Such methods are time-consuming, especially for 3D applications. In this study, we apply a machine learning method...

Generating high-quality labels for deep learning CO2 monitoring using local orthogonalization

Shuang Gao, Sergey Fomel, Yangkang Chen

International Meeting for Applied Geoscience and Energy (IMAGE)

... label generation. We implement a modified 3D U-Net deep learning model to interpret the seismic attributes associated with CO2 injections for complex...

AAPG RM Section Meetings

Search and Discovery.com

Comparative Algorithm Machine Learning Approaches for Predictive Analysis of Well Log Data: A Case Study in the Central Sumatra Basin

Nungga Saputra, Hafid Rizki Nur Rohman, Puspa Alifya, Patria Ufaira Aprina

Indonesian Petroleum Association

... of Long-Short Term Memory (LSTM) and Bi-LSTM. The evaluation of each algorithm model uses field data from the Central Sumatra Basin. Classification metrics...

Strong Foundations, Deep Integration, Infinite Possibilities

James Lowell, Peter Szafian, Nicola Tessen

GEO ExPro Magazine

... years and is effective for building a conceptual model of the geology and QC’ing the individual faults and overall interpretation whilst picking...

Seismic Characterization of Large-scale Sandstone Intrusions

Mads Huuse, Joe Cartwright, Andrew Hurst, Noralf Steinsland

AAPG Special Volumes

... resolution varies with bed thickness, velocity, and frequency. (a) Wedge model convolved with a Ricker wavelet. The vertical seismic resolution is defined...

Estimation of anisotropic parameters from semblance picking using dynamic programming

Hong Liang, Houzhu (James) Zhang, Dongliang Zhang, Hongwei Liu, Xu Ji

International Meeting for Applied Geoscience and Energy (IMAGE)

.... Conventioanlly, seismic model building starts from time-domain velocity analysis using semblance scanning (Taner and Koehler, 1969). The stacking...

A robust approach for shear log predictions using deep learning on big data sets from a carbonate reservoir for integrated reservoir characterization projects

Aun Al Ghaithi

International Meeting for Applied Geoscience and Energy (IMAGE)

... reservoir. The trained model was then used to predict shear logs in wells drilled previously without acquired shear log data. Deep learning provided added...

Seismic fault proximity to production

Jesse Lomask, Toby Burrough, Allison Gilmore, Michael Pyrcz

International Meeting for Applied Geoscience and Energy (IMAGE)

... learning can be utilized to learn and model the relationships between well performance, proximity to faults and the various associated fault attributes...

Deep-Learning-Based Prediction of Post-Fracturing Permeability Field for Development Strategy Optimization in Unconventional Reservoirs

Jiehao Wang, Yunhui Tan, Baosheng Liang, Xinhui Min, Chaoshun Hu, Chao Zhao, Yuyu Wang, Mike Li, Yuguang Chen, Gerardo Jimenez, Shahzad Khan

Unconventional Resources Technology Conference (URTEC)

... number of wells and geological scenarios. A deep-learning-based model (Artificial Learning Fracture, or ALF) was developed to accelerate the hydraulic...

Abstract: Automatic Fault Tracking from 3D Seismic Data Using the 2D Continuous Shearlet Transform with an Example from the Algerian Sahara; #91210 (2025)

Sid-Ali Ouadfeul

Search and Discovery.com

... transforms have introduced more powerful tools for seismic image analysis. The wavelet transform, while effective in time-frequency localization...

Machine learning and explainable AI for predicting missing well log data with uncertainty analysis: A case study in the Norwegian North Sea

Sushil Acharya, Karl Fabian

International Meeting for Applied Geoscience and Energy (IMAGE)

..., followed by PCR and PLSR. Linear regression provides a baseline model for predicting DTC logs. PCR and PLSR are utilized to explore the potential...

What Broke? Microseismic analysis using seismic derived rock properties and structural attributes in the Eagle Ford play

Robert Meek, Bailo Suliman, Robert Hull, Hector Bello, Doug Portis

Unconventional Resources Technology Conference (URTEC)

... regression analysis technique. Rock properties and structural attributes are combined with an ellipsoid stimulation model around the well bore...

Background noise suppression for DAS-VSP data using attention-based deep image prior

Yang Cui, Umair bin Waheed, Yangkang Chen

International Meeting for Applied Geoscience and Energy (IMAGE)

... classical (such as wavelet and curvelet) and dictionary-learning techniques, rely on distinguishing signal components in the transform domain, albeit...

Abstract: A Deep Learning Saturation Imaging Framework to Optimize Reservoir Contact While Drilling; #91204 (2023)

Abdallah AlShehri, Klemens Katterbauer, Ali AlYouesf

Search and Discovery.com

... framework for the optimization of hydrocarbon contact while drilling. The framework utilizes a deep learning convolutional imaging framework in order...

Seismic-based paleoenvironmental analysis of the Paleocene carbonate shelf in Ajdabiya Trough, north-central of Libya

Abdeladim M. Asheibi

Bulletin of Canadian Energy Geoscience (CEGA)

..., constrained sparse spike inversion (Debeye and Riel, 1990). All these methods resolve the components of the convolutional model in each seismic trace...

Multi-Level of Fracture Network Imaging: A HFTS Use Case and Knowledge Transferring

Guoxiang Liu, Abhash Kumar, Song Zhao, Chung Yan Shih, Veronika Vasylkivska, Paul Holcomb, Richard Hammack, Jeffery Ilconich, Grant Bromhal

Unconventional Resources Technology Conference (URTEC)

... making in reservoir management (Left, C). This machine learning model optimizes relatively rapidly (Right, A), matches well with observed data (Right...