1,720,977 research outputs found
Supporting Data for: Direct multi-modal inversion of geophysical logs using deep learning
No full textThe dataset contains realizations of geological stratigraphic curves which were generated using the included script.
This data is required to replicate results in Sergey Alyaev and Ahmed H. Elsheikh. "Direct multi-modal inversion of geophysical logs using deep learning." arXiv preprint arXiv:2201.01871 (2021).This stratigraphy-realization dataset consists of randomly generated stratigraphic vertical depth functions b∗(x) which follow a known trend (here zero).
They need to be combined with an offset well-log, e.g. gamma-ray log from the Geosteering World Cup: https://doi.org/10.18710/20VIVT.
The training data consists of triples: a reference offset-well log which is trimmed
randomly to a short section of 64 cells (32 feet TVD); a sample of b∗(x) with 32 points
(32 feet); and an observed well-log corresponding to the first 16 feet of b∗(x), obtained
using the code supplied in trajectories_data_set.py.
The full training dataset, if read with overlap, contains 28 million samples, stored in train.nc. Additionally, we use a test dataset generated with the same rules containing 560 thousand samples, stored in test.nc.Abstract of the publication
Geosteering of wells requires fast interpretation of geophysical logs which is a non-unique
inverse problem. Current work presents a proof-of-concept approach to multi-modal probabilistic inversion of logs using a single evaluation of an artificial deep neural network
(DNN). A mixture density DNN (MDN) is trained using the ”multiple-trajectory-prediction”
(MTP) loss functions, which avoids mode collapse typical for traditional MDNs, and allows multi-modal prediction ahead of data. The proposed approach is verified on the real-time stratigraphic inversion of gamma-ray logs. The multi-modal predictor outputs several likely inverse solutions/predictions, providing more accurate and realistic solutions
compared to a deterministic regression using a DNN. For these likely stratigraphic curves,
the model simultaneously predicts their probabilities, which are implicitly learned from
the training geological data. The stratigraphy predictions and their probabilities obtained
in milliseconds from the MDN can enable better real-time decisions under geological uncertainties.</p
The typelog from the Geosteering World Cup 2020 semi-finals
No full textThis dataset contains the offset-well gamma-ray log (typelog) from the Geosteering World Cup 2020 semi-finals organized by Rogii Inc., the unconventional well described in Tadjer et al. 2021. The data is in gr.csv. This synthetic log is built based on observation in the Middle Woodford formation, located in the South Central Oklahoma Oil Province (SCOOP) in the United States. The log is discretized every half a foot and we normalize the values of the log to 0-1 interval
Applied Transfer Learning in Drilling Engineering
Full text linkPhD thesis in Information technologyDrilling in the oil and gas industry generates multimodal data crucial for decision-making in both operational and administrative units. The sheer volume of data produced throughout the lifecycle of a well presents opportunities and challenges. Deep learning (DL) has made significant progress in computer vision and language modeling. However, its adoption in niche industries like oil and gas drilling lags due to practical constraints such as limited on-site computational resources, high costs of developing models, and large data requirements to capture meaningful relationships.
In the dissertation, we explore transfer learning to address the DL application bottlenecks. We cover two areas: sequential drilling data for rate of penetration (ROP) prediction and language modeling for efficient data retrieval. In the first part, we leverage simulated data from physics-based simulators as supplemental data. Then, we explore the idea and techniques of transferring knowledge from pre-trained models to adapt to specific wells. Second, we examine the capabilities of generic large language models for drilling text data. Subsequently, we adapt a generic language model in the drilling domain to improve a document retriever. We show that transfer learning enables DL applications in the drilling more accessible. Finally, we aim to foster the development of applications by sharing Our collated and generated datasets
A Probabilistic Reinforcement Learning Framework for Optimized Decision-Making in Geosteering
Full text linkGeosteering refers to the process of intentionally adjusting the drilling trajectory to navigate subsurface environment and maintain alignment with target formations. It is a key process used while drilling oil and gas wells but is also gaining traction in other areas, such as geothermal and civil tunnels drilling. Geosteering is fundamentally a sequential decision-making process, where a series of steering decisions are made under uncertain conditions to optimize the drilling trajectory in real-time. By framing it as a sequential process, operators or decision-makers gain the flexibility to make their decisions as new data becomes available and uncertainties resolved during the drilling operation.
In the oil and gas industry, the utilization of real-time logging-while-drilling (LWD) data has significantly enhanced the ability to make informed decisions during the geosteering process. Most recent research efforts focus on automating and improving the accuracy of LWD data interpretation to better estimate subsurface conditions. However, the challenge remains to effectively use these estimates in a way that optimizes steering decisions and enhances overall operational efficiency. This dissertation tackles this challenge by developing an automated decision-making framework for geosteering. The framework uses reinforcement learning (RL), a subset of machine learning, to optimize and automate the decision-making process.
The first contribution of the dissertation is validating the suitability of RL for geosteering decision-making by performing a comparison study against existing methods. The study shows that our RL-based geosteering framework, particularly with the deep Q-network (DQN) algorithm, consistently outperforms greedy optimization, which focuses on short-term gains. Furthermore, the framework provides results comparable to approximate dynamic programming (ADP), but with significantly reduced computational demands, especially after the training phase is complete. We also introduce the RL-Sensor method, which optimizes geosteering decision-making by utilizing data from behind the decision points, thereby eliminating the need for the Bayesian framework for estimating data ahead of the decision points. This significantly reduces computational demands, particularly during the training phase.
The second contribution of this dissertation focuses on extending the RL-based geosteering framework and applying it to realistic, field-scale scenarios.
This includes the development of the RL-Estimation method, which integrates the particle filter (PF), a state estimation method, into the framework. By combining real-time state estimation with probabilistic estimates, the RL-Estimation method enhances decision-making under uncertainty, significantly improving the robustness and reliability of the decision-making process. Additionally, the dissertation introduces the “Pluralistic” geosteering robot, which applies the extended RL-based geosteering framework to realistic geosteering contexts. This robot adapts the framework to industry-standard geosteering software, incorporating targetline action spaces and realistic DLS constraints. Trained on stochastic geological models informed by human experts interpretations, the robot has demonstrated performance that exceeds that of top-quartile human experts in synthetic test environments.
In summary, this dissertation bridges the gap between the theoretical potential of RL and its practical application in real-time geosteering decision-making. It provides a solid foundation for future advancements in the RL-based geosteering framework, contributing to more efficient and reliable automated geosteering decision-making in the oil and gas industry, as well as other drilling sectors
High-Precision Geosteering via Reinforcement Learning and Particle Filters in Python
No full text<div>
<div># High-Precision Geosteering via Reinforcement Learning and Particle Filters in Python</div>
<br>
<div> This GitHub repository hosts our Python code for integrating Reinforcement Learning (RL) with Particle Filter (PF) to improve decision-making in geosteering, as detailed in "High-Precision Geosteering via Reinforcement Learning and Particle Filters" [ressi2024rlpf](https://arxiv.org/abs/2402.06377).</div>
<br>
<div>## How to cite:</div>
<br>
<div>If you want to adopt the code in your research, please cite the original paper:</div>
<br>
<div>Muhammad, R. B., Srivastava, A., Alyaev, S., Bratvold, R. B., & Tartakovsky, D. M. (2024). High-Precision Geosteering via Reinforcement Learning and Particle Filters. arXiv preprint [arXiv:2402.06377](https://arxiv.org/abs/2402.06377).</div>
<br>
<div>```</div>
<div>@article{alyaev2021direct,</div>
<div> doi = {https://doi.org/10.48550/arXiv.2402.06377},</div>
<div> url = {https://arxiv.org/abs/2402.06377},</div>
<div> author = {Muhammad, Ressi Bonti and Srivastava, Apoorv and Alyaev, Sergey and Bratvold, Reidar B. and Tartakovsky, Daniel M.},</div>
<div> title = {High-Precision Geosteering via Reinforcement Learning and Particle Filters},</div>
<div> journal = {arXiv 2402.06377},</div>
<div> year = {2024},</div>
<div>}</div>
<div>```</div>
<div>## Acknowledgements</div>
<br>
<div>This work is part of the Center for Research-based Innovation DigiWells: Digital Well Center for Value Creation, Competitiveness and Minimum Environmental Footprint (NFR SFI project no. 309589, https://DigiWells.no). The center is a cooperation of NORCE Norwegian Research Centre, the University of Stavanger, the Norwegian University of Science and Technology (NTNU), and the University of Bergen. It is funded by Aker BP, ConocoPhillips, Equinor, TotalEnergies, Vår Energi, Wintershall Dea, and the Research Council of Norway.</div>
</div>
High-Precision Geosteering via Reinforcement Learning and Particle Filters in Python
No full text<div>
<div># High-Precision Geosteering via Reinforcement Learning and Particle Filters in Python</div>
<br>
<div> This GitHub repository hosts our Python code for integrating Reinforcement Learning (RL) with Particle Filter (PF) to improve decision-making in geosteering, as detailed in "High-Precision Geosteering via Reinforcement Learning and Particle Filters" [ressi2024rlpf](https://arxiv.org/abs/2402.06377).</div>
<br>
<div>## How to cite:</div>
<br>
<div>If you want to adopt the code in your research, please cite the original paper:</div>
<br>
<div>Muhammad, R. B., Srivastava, A., Alyaev, S., Bratvold, R. B., & Tartakovsky, D. M. (2024). High-Precision Geosteering via Reinforcement Learning and Particle Filters. arXiv preprint [arXiv:2402.06377](https://arxiv.org/abs/2402.06377).</div>
<br>
<div>```</div>
<div>@article{muhammad2024RLPF,</div>
<div> doi = {https://doi.org/10.48550/arXiv.2402.06377},</div>
<div> url = {https://arxiv.org/abs/2402.06377},</div>
<div> author = {Muhammad, Ressi Bonti and Srivastava, Apoorv and Alyaev, Sergey and Bratvold, Reidar B. and Tartakovsky, Daniel M.},</div>
<div> title = {High-Precision Geosteering via Reinforcement Learning and Particle Filters},</div>
<div> journal = {arXiv 2402.06377},</div>
<div> year = {2024},</div>
<div>}</div>
<div>```</div>
<br>
<div>### To cite this code</div>
<br>
<div>Please use the link/bibtex to its deposited version:</div>
<br>
<div><a href="https://doi.org/10.5281/zenodo.10727358"><img src="https://zenodo.org/badge/DOI/10.5281/zenodo.10727358.svg" alt="DOI"></a></div>
<br>
<div>```</div>
<div>@software{muhammad2024RLPFPy,</div>
<div> author = {Muhammad, Ressi Bonti and</div>
<div> Srivastava, Apoorv and</div>
<div> Alyaev, Sergey and</div>
<div> Bratvold, Reidar Brumer and</div>
<div> Tartakovsky, Daniel M.},</div>
<div> title = {{High-Precision Geosteering via Reinforcement</div>
<div> Learning and Particle Filters in Python}},</div>
<div> month = feb,</div>
<div> year = 2024,</div>
<div> publisher = {Zenodo},</div>
<div> version = {1.1},</div>
<div> doi = {10.5281/zenodo.10727358},</div>
<div> url = {https://doi.org/10.5281/zenodo.10727358}</div>
<div>}</div>
<div>```</div>
<br>
<div>## Acknowledgements</div>
<br>
<div>This work is part of the Center for Research-based Innovation DigiWells: Digital Well Center for Value Creation, Competitiveness and Minimum Environmental Footprint (NFR SFI project no. 309589, https://DigiWells.no). The center is a cooperation of NORCE Norwegian Research Centre, the University of Stavanger, the Norwegian University of Science and Technology (NTNU), and the University of Bergen. It is funded by Aker BP, ConocoPhillips, Equinor, TotalEnergies, Vår Energi, Wintershall Dea, and the Research Council of Norway.</div>
</div>
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Multiscale analysis of selected problems in fluid dynamics
Full text linkThe world around us is inherently multiscale. The variety of different models depending on the reference scale is particularly diverse in problems of fluid-solid systems. For those systems, the geometry, effective parameters and even equations vary between scales. Normally, on coarser scales the fundamental laws of physics are not sufficient to provide a closed-form model. This called for experiment-based closures, which often had limited domain of applicability. An alternative approach enabled by recent developments is so-called multiscale methods. By considering the two scales simultaneously, multiscale methods do not rely on experimental closures and therefore lay a physics-based foundation for engineering problems. In this work, we attempt to separate multiscale methods into larger classes based on a flowchart that contains questions about the structure of problems, for which they are designed. Analysis and methods, presented in the thesis, aim to fill the gaps in inter-disciplinary knowledge across several classes of multiscale methods. We conduct fit-for-purpose analysis of several selected multiscale problems in fluid dynamics: nonlinear single phase flows in porous media, fractal structures formation in freezing brine, particle-particle interaction affected by capillary bridging. The first problem is addressed by the control volume heterogeneous multiscale method (CVHMM). By coupling mass conservation with Navier-Stokes equations on the pore scale, the method captures nontrivial coarse-scale effects, which are not handled by standard single-scale models. To demonstrate the robustness of the presented methods, a multiscale convergence analysis and a derivation of a priori error estimates are developed for the method. For the second problem, simplifications and consequent self-similar analysis identifies compact and fractal regimes of ice formation. Moreover, combining the insights obtained from numerical and analytical solutions we introduce an empirical model that can qualitatively predict properties of the formed ice. For the third problem, we derive a new semi-analytical solution for particle interactions during viscous bridging. The solution provides basis for more efficient simulation of suspended flows of particles
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
- …
