Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
University of Edinburgh
Safe Underground Hydrogen Storage in Porous Subsurface Reservoirs (SHINE): A New European Interdisciplinary Project Exploring the Hydrogen Interaction with Porous Reservoir
Iacopini, David (University of Naples - Federico II) | Parente, Mariano (University of Naples - Federico II) | Edelmann, Katriona (University of Edinburgh) | Hadibeijy, Hadi (TU Delft) | Alcalde, Juan (CSCI) | Vilarrasa, Victor (CSCI) | Donze, Frederic (University of Grenobles Alpes) | Truche, Laurent (University of Grenobles Alpes) | Giovannelli, Donato (University of Naples - Federico II Naples) | di Benedetto, Almerinda (University of Naples - Federico II)
Abstract We present a project named SHINE (Safe underground Hydrogen storage IN porous subsurface rEservoirs) recently funded under the Horizon 2020 scheme, which involve a consortium of five Universities and six energy companies. The main target is to explore the feasibility and address technical, geological, biological and hydrogeological challenges related to hydrogen storage across subsurface porous reservoirs by training 10 new PhD students. In this contribution we will introduce the main scope of the proposal and the research package structure. The novelty behind SHINE is that it focuses at integrating analytical, monitoring and computing techniques to explore how hydrogen may react with the subsurface minerals, fluids and microbial community potentially affecting the storage operations; it will also model the stress field changes across hydrogen reservoir/caprocks and monitor its geomechanics response during repeated injection/production cycles. This approach will radically improve our understanding of this technology, implement and de-risk its application to potential sites providing the necessary insights into underground hydrogen storage for decision makers in government and industry. SHINE will contribute at training a future generation of geo-researcher ready to have an impact into the EU transition energy challenges.
Robotic Fish Enabled Offshore Pipeline Inspection
Wu, Xuqing (University of Houston) | Sood, Tushar (University of Edinburgh) | Chen, Zheng (University of Houston) | Chen, Jiefu (University of Houston)
Abstract Timely inspection of subsea infrastructure, especially subsea pipelines, is the key to the prevention of oil spills. In this paper, a transformative offshore pipeline inspection technology is presented by using a bio-inspired autonomous robotic system equipped with a processing unit for underwater computer vision processing and edge computing. The goal is to build a time-efficient and cost-effective system for underwater pipeline inspection that can detect oil leakage at early stages and prevent disastrous results. In this paper, we introduced a bio-inspired autonomous underwater vehicle (BAUV) equipped with video cameras and mobile edge computing devices. We deploy a deep neural network (DNN) specially trained for a variety of underwater image/video processing tasks. The intelligent computer vision processing unit allows us to navigate and track objects even when the visibility is poor. This time-efficient and cost-effective solution will detect pipeline leakage and rupture at an early stage and allow operators to make timely and informed decisions to minimize environmental impacts.
- Energy > Oil & Gas > Midstream (1.00)
- Government > Regional Government > North America Government > United States Government (0.68)
- Facilities Design, Construction and Operation > Pipelines, Flowlines and Risers > Offshore pipelines (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Neural networks (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Artificial intelligence (1.00)
GPR data imaging and interpretation โ Introduction
Bano, Maksim (University of Strasbourg) | Economou, Nikos (Technical University of Crete, Sultan Qaboos University) | Bradford, John (Colorado School of Mines) | Giannopoulos, Antonios (University of Edinburgh) | Klotzsche, Anja (Agrosphere (IBG-3)) | Slob, Evert (Delft University of Technology) | Tsoflias, George (The University of Kansas)
- Europe (1.00)
- North America > United States > Colorado > Jefferson County > Golden (0.16)
- Geophysics > Electromagnetic Surveying (0.93)
- Geophysics > Seismic Surveying > Seismic Modeling > Velocity Modeling (0.31)
ABSTRACT: Zonal isolation is key to the safety of drilling, injection, and production. If zonal isolation fails, significant financial losses or environmental damage can occur. Cement integrity is critical to maintain zonal isolation. The objective of this paper is to study the influence of wellbore shape on cement sheath integrity and zonal isolation. To diagnose cement sheath integrity, cylindrical wellbore models which contain circular-shape casing, cement, and formation layers have been built. However, the wellbore geometry in real drilling process is elliptical instead of circular. In this study, we built an elliptical-geometry Casing-Cement -Formation System (CCFS) analytical model by applying complex variable method. Two case studies were performed with the elliptical-geometry CCFS model and circular-shape CCFS model that does not consider the wellbore shape factor. 1. INTRODUCTION Quality of cement is an integral part of well integrity. The change of downhole condition and post well site operation may lead to change of the stress-strain state of casingcement- formation system (CCFS) [1]. Cement sheath failure is manifested by inter zonal annular fluid movement and abnormally high annular pressure at some point behind casing up to and at the surface [2]. The detrimental effects of cement sheath failure are numerous and may cause lost revenue from lost production, potentially hazardous rig operations and producing operations. Losing zonal isolation may lead to contamination of freshwater table [3]. To understand the failure behavior of cement is significant for maintaining zonal isolation. Previous research has built analytical/numerical models for predicting the stress distribution around the circular wellbore [4] [5]. However, wellbore geometry in real drilling process is elliptical instead of circular. The geometrical irregularities can result from a combination of geology and/or drilling/completion operations [6]. Two reasons that may cause elliptical geometry were listed here: 1) Tilt angle is the angle between the centerline of the hole and the centerline of the bit. Figure 1 is a simple sketch of tilt angle. The red line shows the plane that is perpendicular to the hole axis. The geometry for this plane is elliptical because of tilt angle.
- North America > United States > South Dakota > Williston Basin > Bakken Shale Formation (0.99)
- North America > United States > North Dakota > Williston Basin > Bakken Shale Formation (0.99)
- North America > United States > Montana > Williston Basin > Bakken Shale Formation (0.99)
Approach for Real-Time Prediction of Pipe Stuck Risk Using a Long Short-Term Memory Autoencoder Architecture
Nakagawa, Yujin (Japan Agency for Marine-Earth Science and Technology) | Inoue, Tomoya (Japan Agency for Marine-Earth Science and Technology) | Bilen, Hakan (University of Edinburgh) | Mopuri, Konda R. (Indian Institute of Technology) | Miyoshi, Keisuke (Japan Oil, Gas and Metals National Corporation) | Abe, Shungo (Japan Oil, Gas and Metals National Corporation) | Wada, Ryota (University of Tokyo) | Kuroda, Kouhei (Japan Petroleum Exploration Co., Ltd) | Nishi, Masatoshi (INPEX corporation) | Ogasawara, Hiroyasu (INPEX corporation)
Abstract Pipe-sticking during drilling operations causes severe difficulties, including economic losses and safety issues. Therefore, real-time stuck-pipe predictions are an important tool to preempt this problem and avoid the aforementioned troubles. In this study, we have developed a prediction technique based on artificial intelligence, in collaboration with industry, the government, and academia. This technique was developed by combining an unsupervised learning model built using an encoder-decoder, long short-term memory architecture, with a relative error function. The model was trained with the time series data of normal drilling operations and based on an important hypothesis: reconstruction errors between observed and predicted values are higher around the time of pipe sticking than during normal drilling operations. An evaluation method of stuck-pipe possibilities using a relative error function reduced false predictors caused by large variations of drilling parameters. The prediction technique was then applied to 34 actual stuck-pipe events, where it was found that reconstruction errors calculated with the relative error function increased 0.5-10 hours prior to the pipe sticking for 17 out of 34 stuck-pipe events (thereby partly confirming our hypothesis).
- North America > United States (0.93)
- Asia > Middle East > UAE > Abu Dhabi Emirate > Abu Dhabi (0.16)
- Well Drilling > Wellbore Design > Wellbore integrity (1.00)
- Well Drilling > Drilling Operations (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Neural networks (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Artificial intelligence (1.00)
A sliced-3D approach to finite-difference time-domain modeling by optimizing perfectly matched layers
Delf, Richard (University of Edinburgh) | Giannopoulos, Antonios (University of Edinburgh) | Bingham, Robert G. (University of Edinburgh) | Curtis, Andrew (University of Edinburgh)
ABSTRACT Finite-difference time-domain forward modeling is often used to gain a more quantitative understanding of the interactions between electromagnetic fields and targets. To undertake full 3D simulations, the computational demands are challenging, so simulations are often undertaken in two dimensions, in which assumptions in the propagation of electromagnetic fields and source type can result in errors. We have developed the concept of a sliced-3D simulation, wherein a thin slice of a 3D domain with strictly 2D geometry is used to minimize computational demands while obtaining synthetic waveforms that contain full 3D propagation effects. This approach requires optimization of perfectly matched layer (PML) boundary condition parameters so as to minimize the errors associated with the source being located close to the boundary, and as a result of grazing-incident angle wave conversion to evanescent energy. We explore the frequency dependence of PML parameters, and we establish a relationship between complex frequency stretching parameters and effective wavelength. The resultant parameter choice is shown to minimize propagation errors in the context of a simple radioglaciological model, in which 3D domains may be prohibitively large, and for a near-surface cross-borehole survey configuration, a case in which full-waveform inversion may typically be used.
- Information Technology > Software (0.50)
- Information Technology > Software Engineering (0.40)
Bayesian full-waveform inversion with realistic priors
Zhang, Xin (University of Edinburgh) | Curtis, Andrew (University of Edinburgh)
ABSTRACT Seismic full-waveform inversion (FWI) uses full seismic records to estimate the subsurface velocity structure. This requires a highly nonlinear and nonunique inverse problem to be solved; therefore, Bayesian methods have been used to quantify uncertainties in the solution. Variational Bayesian inference uses optimization to efficiently provide solutions. However, previously the method has only been applied to a transmission FWI problem and with strong prior information imposed on the velocity such as is never available in practice. We have found that the method works well in a seismic reflection setting and with realistically weak prior information, representing the type of problem that occurs in reality. We conclude that the method can produce high-resolution images and reliable uncertainties using data from standard reflection seismic acquisition geometry, realistic nonlinearity, and practically achievable prior information.
- Information Technology > Artificial Intelligence > Representation & Reasoning > Uncertainty > Bayesian Inference (0.87)
- Information Technology > Artificial Intelligence > Machine Learning > Statistical Learning (0.69)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Optimization (0.68)
- Information Technology > Artificial Intelligence > Machine Learning > Learning Graphical Models > Directed Networks > Bayesian Learning (0.48)
Early Stuck Pipe Sign Detection with Depth-Domain 3D Convolutional Neural Network Using Actual Drilling Data
Tsuchihashi, Naoki (University of Tokyo) | Wada, Ryota (University of Tokyo (Corresponding author) | Ozaki, Masahiko (email: r_wada@k.u-tokyo.ac.jp)) | Inoue, Tomoya (University of Tokyo) | Mopuri, Konda Reddy (Japan Agency for Marine-Earth Science and Technology) | Bilen, Hakan (University of Edinburgh) | Nishiyama, Tazuru (University of Edinburgh) | Fujita, Kazuhiro (Japan Petroleum Exploration Co., Ltd.) | Kusanagi, Kazuya (INPEX Corporation)
Summary A real-time stuck pipe prediction using the deep-learning approach is studied in this paper. Early signs of stuck pipe, hereinafter called stuck, are assumed to show common patterns in the monitored data set, and designing a data clip that well captures these features is critical for efficient prediction. With the valuable input from drilling engineers, we propose a 3D-convolutional neural network (CNN) approach with depth-domain data clip. The clip illustrates depth-domain data in 2D-histogram images with unique abstraction of the time domain. Thirty field well data prepared in multivariate time series are used in this study--20 for training and 10 for validation. The validation data include six stuck incidents, and the 3D-CNN model has successfully detected early signs of stuck in three cases before the occurrence. The portion of the data clip contributing to anomaly detection is indicated by gradient-weighted class activation map (grad-CAM), providing physical explanation of the black box model. We consider such explanation inevitable for the drilling engineers to interpret the signs for rational decision-making. Introduction Stuck is one of the major drilling problems that accounts for nonproductive time. In the past, one-third of the wells drilled in the Gulf of Mexico and the North Sea experience stuck pipe problems (Howard et al. 1994). Moreover, stuck may lead to wellbore abandonment in the worst scenario. Regarding its severity, there is a strong demand for early stuck sign-detection methods to avoid stuck incidents. The drillstring is considered to be stuck when it loses the freedom of movement; that is, it can neither rotate nor move up and down. Also, it cannot be retrieved from the wellbore due to external force. According to the mechanisms, stuck is roughly classified into three types, namely mechanical stuck, differential stuck, and geometrical stuck (Alshaikh et al. 2018). Each stuck type could be separated into smaller classes, for example, mechanical stuck includes both packoff and bridging.
- Europe (1.00)
- Asia > Middle East (1.00)
- North America > United States > California (0.46)
- North America > United States > Texas > Dawson County (0.24)
Realistic microearthquake magnitudes and locations from surface monitoring of hydrofacturing at Preston New Road, UK
Roy, Corinna (University of Leeds) | Nowacki, Andy (University of Leeds) | Zhang, Xin (University of Edinburgh) | Curtis, Andrew (University of Edinburgh, and ETH Zurich;) | Bapie, Brian (British Geological Survey)
Traffic light systems are often used to reduce the probability of damaging seismicity during anthropogenic activities such as industrial mining, geothermal energy and hydraulic fracturing operations. Under such system operations are continued (โgreenโ), amended (โamberโ) or stopped (โredโ) based on the local event magnitude. Accessing accurate microseismic magnitudes is challenging due to unquantified uncertainties, which cannot be neglected in TLS because they can exceed a whole magnitude unit - large enough to make a difference between a continuation as planned (โgreenโ) and an immediate stop (โredโ) of operations. A way to account for these uncertainties in the choice of TLS thresholds was demonstrated such that an operator or regulator can choose between a system which minimizes either the risk of future larger magnitude events or the risk of incorrectly halting operations (C. Roy, pers. comm., 2020). The purpose of this study is to assess the impact of these two different TLS strategies on decisionmaking for induced seismicity at Preston New Road, UK. Presentation Date: Monday, October 12, 2020 Session Start Time: 1:50 PM Presentation Time: 3:30 PM Location: 360A Presentation Type: Oral
Variational Bayesian inversion of seismic attributes jointly for geologic facies and petrophysical rock properties
Nawaz, Muhammad Atif (University of Edinburgh) | Curtis, Andrew (University of Edinburgh, Exploration and Environmental Geophysics Group) | Shahraeeni, Mohammad Sadegh (Total) | Gerea, Constantin (Geoscience Research Center)
ABSTRACT Seismic attributes (derived quantities) such as P-wave and S-wave impedances and P-wave to S-wave velocity ratios may be used to classify subsurface volume of rock into geologic facies (distinct lithology-fluid classes) using pattern recognition methods. Seismic attributes may also be used to estimate subsurface petrophysical rock properties such as porosity, mineral composition, and pore-fluid saturations. Both of these estimation processes are conventionally carried out independent of each other and involve considerable uncertainties, which may be reduced significantly by a joint estimation process. We have developed an efficient probabilistic inversion method for joint estimation of geologic facies and petrophysical rock properties. Seismic attributes and petrophysical properties are jointly modeled using a Gaussian mixture distribution whose parameters are initialized by unsupervised learning using well-log data. Rock-physics models may be used in our method to augment the training data if the existing well data are limited; however, this is not required if sufficient well data are available. The inverse problem is solved using the Bayesian paradigm that models uncertainties in the form of probability distributions. Probabilistic inference is performed using variational optimization, which is a computationally efficient deterministic alternative to the commonly used sampling-based stochastic inference methods. With the help of a real data application from the North Sea, we find that our method is computationally efficient, honors expected spatial correlations of geologic facies, allows reliable detection of convergence, and provides full probabilistic results without stochastic sampling of the posterior distribution.
- Europe > United Kingdom > North Sea (0.34)
- Europe > United Kingdom > Scotland (0.28)
- Europe > Norway > North Sea (0.25)
- (2 more...)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Sandstone (0.47)
- Information Technology > Artificial Intelligence > Representation & Reasoning > Uncertainty > Bayesian Inference (1.00)
- Information Technology > Artificial Intelligence > Machine Learning > Learning Graphical Models > Directed Networks > Bayesian Learning (0.93)
- Information Technology > Artificial Intelligence > Machine Learning > Learning Graphical Models > Undirected Networks > Markov Models (0.68)