Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
The University of New South Wales
Inclusion of Microporosity in Numerical Simulation of Relative Permeability Curves
Lanetc, Zakhar (The University of New South Wales) | Zhuravljov, Aleksandr (Tyumen State University) | Shapoval, Artur (The University of New South Wales) | Armstrong, Ryan T. (The University of New South Wales) | Mostaghimi, Peyman (The University of New South Wales)
Abstract Advances in high-resolution micro computed tomography (micro-CT) allow obtaining high-quality rock images with a resolution of up to a few micrometres. Novel direct numerical simulation methods provide the opportunity to precisely predict the flow properties in the resolved pore space. However, a large fraction of porosity lies below the resolution of modern micro-CT scanners. These, so called, micro-pores may significantly affect the physics of flow in geologically complex dual-porosity heterogeneous formations (carbonates, shales, and coals) and are currently not accounted for in traditional micro-CT based simulations. In this work, we have employed a multiphase multi-scale Darcy-Brinkman approach to simulate immiscible two-phase flow in a hybrid system containing both macro-porous solid-free regions and a micro-porous permeable matrix. This approach solves the Navier-Stokes based volume of fluid equations system in macro-pores and accounts for multiphase Darcy equations in micro-porous regions. By combining available information on micro-porosity with relative permeability curves estimated from the synthetically generated image with both macro- and micro-porous regions fully resolved, we solve the inverse problem to account for micro-porous contribution in our Darcy-Brinkman simulation. This approach allows us to estimate relative-permeability curves in the micro-porous region and correct the multi-scale simulation so it coincides with the data from the fully-resolved image. As a result, we were able to account for the impact of micro-porosity on the residual saturation and correct the shape of relative permeability curves and their end-points in the micro-porous domain. The proposed approach provides a workflow which can be used to history-match the Darcy-Brinkman pore-scale simulation with core-scale petrophysical data with respect to the relative permeability. Thus, it is possible to account for heterogeneity in complex rock formations by incorporating the whole range of porosity. The inclusion of micro-porosity in pore-scale image-based simulations for predicting relative permeability curves may help in a more reliable modelling and estimation of filed-scale subsurface flows, production profiles, recoverable reserves and carbon capture and storage mechanisms.
The Wave System of the Sydney Harbor RiverCat Ferry
Doctors, Lawrence J. (The University of New South Wales)
We revisit here the hydrodynamics supporting the design and development of the RiverCat class of catamaran ferries operating in Sydney Harbor since 1991. The design of the ferry was strongly influenced by the requirement to limit the erosion of the banks on the sides of the Parramatta River along which these ferries operate. More advanced software is used here. This software accounts for the hydrodynamics of the transom stern which create a hollow in the water. This hollow causes an effective hydrodynamic lengthening of the vessel. This leads to a reduction in the generation of the wave system and the consequent wave resistance. It is demonstrated that lengthening the RiverCat would substantially reduce its wave generation.
- Europe (0.46)
- North America > United States (0.29)
- Oceania > Australia > New South Wales (0.28)
Mechanisms of Confining Pressure Dependence of Resistivity Index for Tight Sandstones
Dai, Hongyi (The University of New South Wales) | Shikhov, Igor (The University of New South Wales) | Arns, Ji-Youn (CJEL Digital Imaging Education Solution Pty Ltd) | Li, Rupeng (The University of New South Wales) | Arns, Christoph Hermann (The University of New South Wales)
Abstract Resistivity measurements are a major input into hydrocarbon reserve estimation and are usually described by Archie's laws. In this study digital rock physics (DRP) is utilized to analyse the mechanisms of non-Archie and Archie behaviour of formation factor (FF) and resistivity index (RI) of low porosity Fontainebleau sandstone for ambient conditions and under high confining pressure, respectively. Fontainebleau sandstone was imaged by micro-CT at a resolution of one micron and sub-resolution details of grain contact width distribution along with their length, extracted from a set of scanning electron microscope (SEM) images. The nano-scale aperture of grain contacts, which is below image resolution, is accounted for in micro-CT based numerical calculations by assigning effective porosity and conductivity to individual voxels of the extracted grain contact network. A porosity reduction of grain contacts and open pore space as function of applied confining pressure is introduced, capturing the pressure dependence. The concept was implemented by grain-contact labelling and introducing an additional phase derived from a Euclidean distance transform. Sub-voxel stress-strain effects were incorporated by attributing all compressibility effects to the pore space (open pore space and grain contacts), treating the solid phase as perfectly rigid. Voxel-scale input conductivities are assigned using Archie's law followed by solving the Laplace equation for sample-scale effective rock resistivity and resistivity index directly on the segmented image using the finite element method (FEM). For the numerically modelling of Formation Factor and Resistivity Index of low porosity Fontainebleau sandstone as a function of confining pressure, which depends on sub-resolution features, a set of hypotheses were tested: (1) two segmentation scenarios based on the measured contact aperture distribution from SEM analysis โ a homogeneous grain contact aperture based segmentation (single grain contact network) by assuming all grain contacts as a average constant value and a heterogeneous grain contact aperture based segmentation (dual grain contact network) by assigning two groups of grain contact aperture (wide and narrow); (2) homogeneous and heterogeneous morphological change of grain contacts due to confining pressure effect; (3) partial saturation of grain contacts. In all cases strong water-wetness was assumed and discretization effects were analysed carefully. The numerical results highlight the relative contribution of each of two conductive components of Fontainebleau sandstone (open pores vs. grain contacts) over the full range of partial saturations. Of importance is the connectivity of the system, with discretization effects having a significant effect on formation factor, but small effect on resistivity index. Grain contacts and confining pressure are found to have a significant impact on RI behaviour of low-porosity Fontainebleau sandstone. Both the grain contact network with homogeneous aperture and the heterogeneous grain contact network are able to describe experimental observations. However, it is not sufficient to assume a homogeneous change in contact area and an inhomogeneous deformation of grain contact zones is required to match experiment.
Segmentation of X-Ray Images of Rocks Using Deep Learning
Ar Rushood, Ibrahim (Saudi Aramco D&WO) | Alqahtani, Naif (The University of New South Wales) | Wang, Ying Da (The University of New South Wales) | Shabaninejad, Mehdi (The Australian National University) | Armstrong, Ryan (The University of New South Wales) | Mostaghimi, Peyman (The University of New South Wales)
Abstract Segmentation of X-ray images of rocks is an important step in digital rock technology. Current segmentation methods often suffer from operator bias and can be time-consuming. To overcome these limitations, an automated image segmentation model is created for pore scale images through the use of convolutional neural networks. A dataset of micro-Computed Tomography (CT) images of sandstones is considered. To create the ground truth data for training, Scanning Electron Microscopy 2D and micro-CT 3D images are used to obtain accurate segmentation masks. Three models are trained with the available data: One with limited images, one with the full dataset, and one with augmented data. The data augmentation is achieved by increasing sample size through image partitioning. The data set for each model is divided into training, validation, and testing with a 60/30/10 split, respectively. The U-Net architecture, designed to work with limited training data, is used to develop the models. Further validation of the models is performed on a different dataset unseen by the models. The Minkowski functionals and permeability of the volume generated by the segmented images are computed and compared with the ground truth segmentation. Against the unseen dataset, the models scored a dice coefficient of 0.9479, 0.9518, and 0.9599, respectively. We discuss the potential for improvement by data augmentation and fine-tuning. One limitation of the models is with the deficiency of variety in training data as both SEM and micro-CT is required to obtain the segmentation masks. We also discuss the performance of the model on unseen samples and show the potential improving efficiency in the digital rock technology. The models provide a quick and accurate segmentation for images of sandstones without the influence of operator bias, and the method shows promise for further development and improvement.
- Health & Medicine (1.00)
- Energy > Oil & Gas > Upstream (1.00)
Deterministic Pipe Network Modelling for Fractured Rocks
Jing, Yu (The University of New South Wales) | Armstrong, Ryan (The University of New South Wales) | Mostaghimi, Peyman (The University of New South Wales)
Abstract The fracture network plays a critical role in controlling flow pathways in fractured rock. Thus, transmissibility study of fracture networks via different flow modelling methods is of importance. Compared with direct flow simulation, the pipe network model is an effective means of modelling fluid flow in fracture network due to its computational efficiency. However, the characterisation of the fracture network topology as well as the equivalent conductivity of the pipes are still challenging to credibly predict the permeability. Also, pipe network models are commonly constructed based on stochastic Discrete Fracture Network (DFN) models with more uncertainties, which also requires a number of stochastic DFN realisations to be created. In this paper, we develop a novel Pipe Network Modelling (PNM) framework for fractured media, where the PNM is constructed based on deterministic DFN models that are directly derived from micro-CT images. By comparing permeability values obtained from PNMs and results from micro-CT images and voxelised DFNs, we conclude that PNM modelling can effectively estimate the permeability of original fracture networks, while requiring significantly less computational cost. In addition to the advantage of computational efficiency, PNM is more preferable for the challenging multi-phase flow simulation.
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- Geology > Rock Type (0.68)
- Well Completion > Hydraulic Fracturing (1.00)
- Reservoir Description and Dynamics > Unconventional and Complex Reservoirs > Naturally-fractured reservoirs (1.00)
- Reservoir Description and Dynamics > Reservoir Fluid Dynamics > Flow in porous media (1.00)
- Reservoir Description and Dynamics > Fluid Characterization > Fluid modeling, equations of state (1.00)
Analysis of Diffusion Coefficient and Fracture Aperture in Coal using Micro-Computed Tomography Imaging
Lu, Xiao (The University of New South Wales) | Armstrong, Ryan (The University of New South Wales) | Yuan, Meng (The University of New South Wales) | Zhang, Yulai (The University of New South Wales) | Mostaghimi, Peyman (The University of New South Wales)
Abstract Coalbed methane (CBM), also known as coal seam gas, is becoming an increasingly important energy resource in the global natural gas market. Gas transport in CBM reservoirs remains a crucial research topic that has not been fully understood. Two scales of gas flow are identified in coal cores: flow in fractures and diffusion within matrix. The diffusion process is quantified by the gas diffusion coefficient while flow in fractures is governed by fracture apertures. This paper aims to explore the diffusion process in coal using X-ray microcomputed tomography (micro-CT) imaging. The experiments are conducted at 100 psi effective stress to eliminate the impact of pressure. The images obtained are registered for visualisation and analysis of the diffusion process and comparisons of fracture. In the paper, the impact of increasing effective stress on fracture aperture is demonstrated. Also, the diffusion coefficient of Krypton in coal matrix is estimated and discussed.
- Oceania > Australia (0.68)
- North America > Canada (0.46)
- Asia > Middle East > Turkey (0.28)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- Geology > Rock Type > Sedimentary Rock > Organic-Rich Rock > Coal (0.89)
- Asia > Middle East > Turkey > Zonguldak Basin (0.99)
- North America > United States (0.89)
- North America > Canada (0.89)
Deep Learning Convolutional Neural Networks to Predict Porous Media Properties
Alqahtani, Naif (The University of New South Wales) | Armstrong, Ryan T. (The University of New South Wales) | Mostaghimi, Peyman (The University of New South Wales)
Abstract Digital rocks obtained from high-resolution micro-computed tomography (micro-CT) imaging has quickly emerged as a powerful tool for studying pore-scale transport phenomena in petroleum engineering. In such frameworks, digital rock analysis usually carries the problematic aspect of segmenting greyscale images into different phases for quantifying many physical properties. Fine pore structures, such as small rock fissures, are usually lost during segmentation. In addition, user bias in this process can lead to significantly different results. An alternative approach based on deep learning is proposed. Convolutional Neural Networks (CNN) are utilized to rapidly predict several porous media properties from 2D greyscale micro-computed tomography images in a supervised learning frame. A dataset of greyscale micro-CT images of three different sandstones species is prepared for this study. The image dataset is segmented, and pore networks are extracted to compute porosity, coordination number, and average pore size for training and validating our model predictions. The greyscale images (input) and the computed properties (output) are uploaded to a deep neural network for training and validation in an end-to-end regression scheme. Overall, our model estimates porosity, coordination number, and average pore size with an average error of 0.05, 0.17, and 1.8ฮผm, respectively. Training wall-time and prediction error analysis are also discussed. This is a first step to use artificial intelligence and machine learning methods for the robust prediction of porous media properties from unprocessed image-driven data.
DigiCoal: A Numerical Toolbox for Fractured Coal Characterisation
Jing, Yu (The University of New South Wales) | Armstrong, Ryan (The University of New South Wales) | Mostaghimi, Peyman (The University of New South Wales)
Abstract We present the DigiCoal toolbox that is an integrated set of numerical functions written in Matlab, dedicated to analyse 3D computed tomography (CT) images of coal and reconstructing representative digital models. The design is based on a comprehensive framework: CT image pre-processing, statistics extraction, digital coal modelling and structural analysis. This paper offers an overview of the structure and techniques used in the creation of the toolbox, together with code snippets and examples.
- North America > United States (0.68)
- Oceania > Australia (0.46)
- Geology > Geological Subdiscipline > Geomechanics (0.94)
- Geology > Rock Type > Sedimentary Rock > Organic-Rich Rock > Coal (0.47)
- Oceania > Australia > Queensland > Central Highlands > Bowen Basin (0.99)
- Europe > Germany > Ruhr Basin (0.99)
- Asia > China > Shanxi > Ordos Basin (0.99)
- (2 more...)
Review of Current Empirical Approaches for Determination of the Weak Rock Mass Properties
Zhai, Hao (The University of New South Wales) | Canbulat, Ismet (The University of New South Wales) | Hebblewhite, Bruce (The University of New South Wales) | Zhang, Chengguo (The University of New South Wales)
Abstract Weak rock mass strength estimation is a long-lasting challenge associated with geotechnical engineering due to its complex nature and limited definition. Weak rock masses normally refer to low strength, highly fractured decomposed and tectonically disturbed rocks which have properties intermediate from brittle rocks to ductile soils. Since the behavior of weak rock mass has not been fully understood, it is a common practice to apply existing empirical approaches, which are developed for competent rock masses influenced by joints, to determine their mechanical properties. This paper reviewed the current empirical approaches, and detailed weak rock mass strength calculations based on rock matrix, joint layout, joint condition and external factors. The limitations associated with these methods are discussed, and suggestions are provided for the selection of suitable methods. 1. Introduction Determination of weak rock mass properties is a significant challenge in geotechnical engineering. In general, weak rocks are considered to be the transitional material between competent rocks and soil, therefore, their behavior converges to competent rock at its upper bound and soil at the lower bound. Despite significant amount of research, methods to estimate in-situ behavior and strength of weak rock masses remain to be relatively fragmented and incomplete. The difficulty of determining their behaviour is mostly caused by the complex nature and inadequate definitions. Different origin and alteration process of weak rocks result in variant properties that inevitably influence their overall behaviour. Hence, it is important to understand the differences in their property that inherited from both previous phase and alteration process and to adopt suitable approaches to estimate their strength according to these features. In practice, the term weak rock commonly refers to both young sedimentary rocks with low compressive strength and heavily altered hard rock with intense structures [1โ3]. Based on the origin and geological alterations, weak rock can be classified as young sedimentary rock, weathered competent rock and tectonically disturbed competent rock as shown in Fig. 1. Young sedimentary rocks such as mudstone and claystone contain poor lithification and weak particle cementation. The strength of them can be described by the ISRM definition of weak rocks with uniaxial compressive strength (UCS) being 0.5 MPa to 25 MPa [2, 3]. Weathered competent rocks such as sandstone can also be considered as weak rock. During prolonged exposure, some rock mass components start to break down and crack along pre-existing micro fractures. As a result of weathering, the well-developed, interconnected defect fabric deteriorates the integrity of the rock mass, thus lead to a reduction of the overall mechanical strength. This type of rock is well represented in Rock Mass Rating (RMR) and Geological Strength Index (GSI) classification systems as poor quality rocks with ratings lower than 25 and 20 respectively or less than 0.1 in Q system. In practice, there is a tendency to consider tectonically disturbed competent rocks, which preserves limited original structures formed in lithification, as weak rock mass [3]. Due to destruction of original structure during folding and shearing, it's common to observe widely existing intensive fractures. Thus, this type of rock has very low mechanical properties similar to other types of weak rock masses. Marinos and Heok's study of flysch in 1998 provides a good example of such weak rock [4โ6].
- Europe (1.00)
- North America > United States (0.93)
- Oceania > Australia (0.68)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock (0.54)
- Geology > Rock Type > Sedimentary Rock > Organic-Rich Rock > Coal (0.46)
- Energy > Oil & Gas > Upstream (1.00)
- Materials > Metals & Mining (0.94)
- North America > United States > Kentucky > Illinois Basin (0.99)
- North America > United States > Indiana > Illinois Basin (0.99)
- North America > United States > Illinois > Illinois Basin (0.99)
An Analytical Study of Failure of Transversely Isotropic Rock Discs Subjected to Various Diametrical Loading Configurations
Aliabadian, Z. (The University of New South Wales) | Zhao, G. F. (Tianjin University) | Russell, A. R. (The University of New South Wales)
Abstract The indirect (Brazilian) tensile strength test is commonly used in rock engineering and rock mechanics. Most of the research on how to interpret Brazilian test is only valid for isotropic rock. The load may be applied to disc shaped samples through flat platens or curved jaws and the different load configurations give rise to subtle but important differences to stress distributions throughout samples during testing and at the onset of failure. To reliably obtain the tensile strength accurate stress distributions, especially at the location of the initial crack, must be determined. However, most real rocks exhibit a certain type of anisotropy known as transverse isotropy due to the presence of preferred directions of grains, beddings, microcracks or pores. This study focuses on the failure modes when transverse isotropy is present. Previous studies for transverse isotropy indicate that the exact location of the initial crack is at the disc centre. These studies, however, ignore the effect of the load contact configuration on the stress distribution. This paper overcomes this limitation and applies Amadei's analytical solution to assess the effects of the load contact area on the stress distribution. The stress distribution is coupled with a transverse isotropy failure criterion in a mechanically consistent way to find the exact location of initial crack. It is shown that the location of the initial crack depends on the transverse isotropy orientation and load configuration. 1. Introduction The Brazilian test is a diametrical compression test that is used as an indirect tensile strength measurement. At first, the test was proposed by Carneiro [1] and Akazawa [2] to obtain tensile strength of concrete. In 1978, the International Society for Rock Mechanics (ISRM) officially presented Brazilian test as a standard method to indirectly determine the tensile strength of rock like materials (ISRM 2007). Nowadays the Brazilian test is widely used to determine tensile strength of rock like materials because of its simple specimen preparation, experimental performance and data reduction [3โ4]. In the test, a thin disc is diametrically compressed. The corresponding stress field throughout the specimen is highly non-uniform.