Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
44th U.S. Rock Mechanics Symposium and 5th U.S.-Canada Rock Mechanics Symposium
Prediction of Standard Penetration Tests Via Microtremor Array Using Artificial Neural Networks
Angorani, S. (School of Mining Engineering, University of Tehran) | Memarian, H. (School of Mining Engineering, University of Tehran) | Panahi, M. Shariat (School of Mechanical Engineering, University of Tehran)
ABSTRACT: In recent years, artificial neural networks (ANNs) have been applied to many geotechnical engineering problems with some degree of success. In this paper, an ANN model is developed for predicting standard penetration tests (SPT) by microtremor array results. SPT gives an indication of the soil stiffness, which can be empirically related to many engineering properties. Microtremor is a low amplitude ambient vibration of the ground, caused by man-made or atmospheric disturbances and gives useful information on dynamic properties of the studied site, such as ground predominant period and amplitude. Using microtremor array shear wave velocity profile was obtained. An ANN was trained with Vs and SPT measurements in 7 boreholes with 100m depth as input and output, respectively. The present study shows that using an optimized ANN model, the SPT can successfully be estimated from a microtremor array, which eliminates the necessity of having excessive boreholes. 1- INTRODUCTION Microtremor Array (MA) observations provide substantial information about such dynamic properties of a site as its predominant period and amplitude. The shear wave velocity (Vs) could be determined using the dispersion characteristics of the surface wave part. The MA technique requires no boreholes. This particular characteristic makes MA significantly more convenient and inexpensive compared to the traditional borehole method [3]. Many empirical relations have been suggested to estimate Vs from the Standard Penetration Test Number (SPT-N) with limited success, including those suggested by Ohta and Goto (1978), Iyisan (1996) and Kiku et al. (2001). In this article we propose a methodology for the prediction of SPT-N from microtremor-provided Vs using Artificial Neural Networks (ANNs). ANNs are biologically inspired networks of interconnected artificial Neurons (or processing units) that can be trained to approximate highly complex relations simply by being exposed to a finite number of verified examples. The proposed strategy builds on the development and training of an ANN using a combination of Vs values obtained from MA and corresponding SPT-N from boreholes. The resulting network would be able to predict, accurately enough, the SPT-N for other, new boreholes and thus to reduce the time and cost of geotechnical exploration. 2. MICROTREMOR Microtremor is a low amplitude (in the order of micrometres) ambient vibration of the ground caused by man-made or atmospheric disturbances. Observation of microtremors can give useful information on dynamic properties of the site such as predominant period and amplitude. The major advantages of these ambient vibration techniques are easy to perform, inexpensive the low cost exploration and monitoring capabilities, the possibility to perform non-destructive measurements at every place of a densely populated city, and the relatively large penetration depth [3]. More detailed information on the shear wave velocity profile of the site can be obtained from microtremor array observation. In recent years, the passive recording of microtremors at single stations [12, 13] or at small-scale arrays [14, 15, and 16] has been used to determine shallow shear wave velocity profiles. Thus, it is of key interest to determine the shallow shear wave velocity structure [6, 11].
- Geology > Geological Subdiscipline > Geomechanics (0.35)
- Geology > Structural Geology > Tectonics > Plate Tectonics > Earthquake (0.31)
- Reservoir Description and Dynamics > Reservoir Characterization > Seismic processing and interpretation (1.00)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Open hole/cased hole log analysis (1.00)
- Data Science & Engineering Analytics > Information Management and Systems > Neural networks (1.00)
ABSTRACT: Deformation bands in high porosity sandstone are an important geological feature for geologists and petroleum engineers; however, formation of these bands is not fully understood. The theoretical framework for deformation band formation in high porosity geomaterials is well established. It suggests that the intermediate principal stress influences the predicted deformation band type; however, these predictions have yet to be fully validated through experiments. Therefore, this study investigates the influence of the intermediate principal stress on failure and the formation of deformation bands in Castlegate sandstone. Mean stresses for these tests range from 30 to 150 MPa, covering brittle to ductile behavior. Deformation band orientations are measured with external observation as well as through acoustic emission locations. Results of experiments conducted at Lode angles of 30 and 14.5 degrees show trends that qualitatively agree with localization theory. The band angle (between the band normal and maximum compression) decreases with increasing mean stress. For tests at the same mean stress, band angle decreases with increasing Lode angle. 1. INTRODUCTION Deformation bands in high porosity sandstone are features of interest to both the petroleum industry as well as geologists. Deformation bands can serve as barriers or conduits for fluid flow depending on their compactant or dilatant nature. Additionally, deformation bands act as an indicator of the loading history of a particular outcrop. This study utilizes Castlegate sandstone, a commonly used reservoir analog, to investigate deformation band formation under true triaxial stress conditions. Comprised mostly of quartz grains bonded with calcite, Castlegate was selected due to its high porosity (27%) and the sizeable existing body of work on the material. Additionally, since Castlegate is known to form compaction bands (not all high porosity sandstones do), the full range of deformation band types can be examined. The Rudnicki and Rice localization criterion [1] suggests that the inception of a band of deformation can be modeled as a bifurcation from uniform deformation due to a constitutive instability. The constitutive model used by Rudnicki and Rice is dependent on the first invariant of stress, I1, and the second invariant of deviatoric stress, J2. An additional goal of this work is to determine if the Castlegate behavior depends on the third invariant of deviatoric stress, J3. A dependence on J3 would cause a change in the shape of the failure surface from a circle to a rounded triangle when viewed in the deviatoric (pi) plane. It is well known that sands are J3 dependent. However, it unknown if porous sandstone is also J3 dependent, since testing has historically been done under axisymmetric stress states (intermediate principal stress equal to the maximum or minimum principal stress), from which J3 dependence cannot be determined. Thus, because the J3 evaluation cannot be completed until all Lode angle tests are run for a selected mean stress, no J3 results are presented in this paper. To evaluate the theoretical predictions outlined above, and to assess J3 dependence, the following experiments were designed.
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Sandstone (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- Government > Regional Government > North America Government > United States Government (0.69)
- Energy > Oil & Gas > Upstream (0.67)
ABSTRACT: In preparation for well stimulation activities and the development of an enhanced geothermal system (EGS) in the Desert Peak geothermal field, a series of petrologic and rock mechanical tests were conducted on selected core samples to represent the planned stimulation interval within Well 27-15. The interval consists of Tertiary rhyolite tuffs that overlie metamorphic basement rocks consisting of fractured metasedimentary rocks. Hydraulic stimulation of the well is intended to enhance formation permeability through self-propping shear failure along the most optimally oriented and critically stressed of pre-existing fractures. Rock mechanical testing was conducted on core samples to determine mechanical properties of the various lithologies including: radial versus axial volumetric strain, stress-strain relationships, dynamic versus static Young's moduli, and frictional strengths and failure responses under a variety of confining conditions. The results of the laboratory tests were used to construct Mohr-Coulomb failure envelopes for the proposed reservoir rocks. Comparison of the test results indicate overall moderate rock strengths; with unconfined compression stress estimates of 17,000-27,000 psi (117-186 MPa) for the more siliceous lithologies, and about 12,000- 16,000 psi (82-110 MPa) for argillaceous and devitrified rhyolites. Quasi-static values for Young's modulus and Poisson's ratio ranged from 6538 MPa (in argillaceous rhyolite) to 41,700 MPa (in siliceous metamudstone), and 0.08 to 0.24, respectively. Residual compressive strength measurements were used to evaluate the propensity for frictional failure along natural fractures seen in the stimulation interval within Well 27-15. In the test samples, residual friction angles from constructed Mohr circles determined coefficients of sliding friction (ยต) in the range of 0.66 to 0.96. Pre- and post- test measurements on the core plugs indicate up to a 20-fold permeability enhancement in originally tight rhyolite units as a result of shear failure. Assuming that failure occurs on the same structural features in the well as in the core, these laboratory studies directly test the shear dilation concept in these clay-rich rocks, and are being used in combination with borehole stress measurements and fracture logging to predict fluid pressures required for initiation of shear dilation and permeability development within the geothermal reservoir. INTRODUCTION Petrologic evaluation was conducted on selected core and well cuttings samples collected from two drillholes in the Desert Peak geothermal field in northwestern Churchill County, Nevada (Figure 1). Petrologic characteristics of rocks within the proposed stimulation interval of Well 27-15 (Figure 2) are related to the geomechanical character and fracturing potential of these stratigraphic units where ambient temperatures are ~180 to 195ยฐ C. Mechanical testing of analogous core samples from nearby Well 35-13 (Figure 1) was conducted in preparation for mechanical and chemical stimulation of Well 27-15, as part of DOE-Ormat's Enhanced Geothermal System (EGS) Project (Robertson-Tait et al., 2004; Zemach et al., 2009, 2010). METHODS The proposed stimulation interval within Well 27-15 consists of variably clayey to silicified Tertiary rhyolite ash flow tuffs that overlie weakly metamorphosed mudstones in the pre-Tertiary basement. Petrographic and mineralogic analyses were conducted on well cuttings samples from 27-15 that represent the stimulation interval (3000-3500 ft; 930-1085 m).
- Phanerozoic > Mesozoic (0.46)
- Phanerozoic > Cenozoic > Tertiary (0.34)
- Geology > Mineral > Silicate > Phyllosilicate (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock (0.91)
- Energy > Renewable > Geothermal > Geothermal Resource (1.00)
- Energy > Oil & Gas > Upstream (1.00)
- Energy > Renewable > Geothermal > Geothermal Resource for Power Generation > Enhanced Geothermal System (0.80)
ABSTRACT: This study explores the interaction between crack initiation and nanomechanical properties in the crack tip process zone (zone of microcracking at the tip of a propagating crack) of a brittle material. Samples of Carrara marble with pre-existing cracks (โflawsโ) were loaded in a uniaxial testing machine until the process zone appeared at the tips of the pre-existing cracks in the form of โwhite patchingโ. Two techniques were then used to obtain nanomechanical properties of the process zone and relate them to macroscale crack initiation: digital photography, to visually assess the macrostructure and crack formation, and nanoindentation, to yield nanomechanical properties and assess nano/microheterogeneities. Nanoindentation testing was comprised of lines and grids of single nanoindentations located both near and far from the process zone. The purpose of nanoindentation testing is to investigate the underlying trend in nanomechanical property change between intact and process zone marble. Analysis of nanoindentation testing results showed a decrease of both modulus and hardness (a) near grain boundaries in intact material, and (b) with closeness to the process zone. Ultimately, the study confirms that the crack tip process zone manifests itself as an area of reduced nanoindentation hardness and nanoindentation modulus in marble. 1. INTRODUCTION The study of geomaterial cracking at its most fundamental scale, nano and microscales, is critical to predicting crack propagation. With this in mind, this study explores the interaction between crack initiation and nanomechanical properties in the crack tip "process zone" of Carrara marble. Nanomechanical properties of material in the fracture process zone (FPZ), i.e. the material within the zone of microcracking around the crack tip, are compared with nanomechanical properties of the "intact material". The differences in nanomechanical properties between intact and process zone material reveal the existence of nanoscale damage in the FPZ that may well contribute to the fracture propagation. The literature contains extensive information on the three defining aspects of this investigation - process zone, nanomechanical properties, and geomaterials - but this study represents the first occasion to bring these aspects together in such a way. Many theories exist regarding the process zone, but the experimental investigation of this theoretical region in rock is a relatively recent development. In Linear Elastic Fracture Mechanics (LEFM) the process zone is defined as the core region of plastic-behaving material. Rice investigated the process zone region from a theoretical perspective to ultimately configure the J-Integral, which expresses the energy release rate. However, the integral does not consider the interplay between nano/micromechanical properties and the energy release rate or fracture energy [2]. The Dugdale model for the size of the process zone, or โinelastic zone,โ has been applied even at the fault scale (several kilometers), but does not consider the variability of properties in the region [3]. Application of the closely related Cowie and Scholz model to existing faults reveals a logarithmically decreasing microfracture density within and moving away from the โcohesion zoneโ (a low-strength area surrounding a displaced fault), and a finite scale-independent limiting microfracture density [4].
ABSTRACT: Fracture propagation, especially for fractures embedded in the body, simultaneously involves Mode I, Mode II and Mode III fracture pattern. Therefore, a two-dimensional simulation has limited applicability in modeling realistic fracture behavior, and a 3D propagation model is necessary. In this paper we present a VMIB model for simulating 3D fracture propagation. Fracture behavior is naturally three-dimensional, involving macro fracture initiation and propagation rooted in bond rupture at the micro scale. The virtual multidimensional internal bond (VMIB) model bridges the processes of macro fracture and micro bond rupture. The macro three-dimensional constitutive relation in VMIB is derived from the one-dimensional bond in the micro scale and is implemented in a 3D finite element method. To represent the contact and friction between fracture faces, the three-dimensional element partition method is employed. The model is applied to simulate fracture propagation and coalescence in typical laboratory experiments, and to analyze the propagation on an embedded fracture. Numerical simulations show good agreement with experimental observations. 1. INTRODUCTION Simulation of fracture propagation is often conducted in 2D. Although useful, a 2D approach is subjected to simplifying assumptions such as plane-stress or plane strain, and is limited in revealing real fracture mechanisms since in reality, only few situations strictly meet the plane-stress or plane strain postulates. Propagation of an embedded fracture illustrates the differences between 3D fracture propagation mechanism and the 2D case. When an embedded inclined 3D fracture is subjected to compressive stress, the fracture tips are restrained by the surrounding materials so that it does not propagates as predicted by 2D fracture models. Experiments by Germanovich et al[1] on 3-D crack growth under compressive loads have shown that a single wing crack cannot grow extensively to cause ultimate failure of the specimen. Simulation of 3D fracture propagation is complex as it often simultaneously involves all three fracture modes over a contour. Whereas in 2D case the zone of interest is only a point (fracture tips), in 3-D case the fracture tip is a closed boundary making it difficult to develop a fracture criterion for predicting propagation at different points along its edge as each point has a different stress intensity factor. Another major challenge in modeling fracture propagation using the linear elastic finite element method is the need for remeshing as the fracture propagates. To avoid the this problem, Frangi [2] developed the symmetric Galerkin boundary element method to simulate 3D fracture propagation. Lazarus [3] has theoretically studied the 3D fracture propagation subjected to uniform remote tensile loading. It was concluded that a closed crack tends to propagate to become circular and that for the material of heterogeneous toughness, the crack tends to propagate to ensure that the stress intensity factor reaches the toughness along the whole front. A macro constitutive relation is then derived from the cohesive law between material particles, which makes the separate fracture criterion unnecessary. Zhang and Ge [6-8] have accounted the shear effect between material particles and developed the virtual multidimensional internal bond (VMIB) model.
ABSTRACT: Field scale rock mass constitutive behaviour remains an area of intense debate. The Generalized Hoek-Brown [GHB] criterion remains the most popular and commonly used rock mass behaviour criteria for practical applications. The GHB as originally developed did not did not incorporate post-peak behaviour and provides no guidance as to how such parameters should be derived. It is, in fact, uncertain whether derivation of so called post-peak GHB parameters is a valid or appropriate approach, although at present there appears to be few viable alternate options. Numerous attempts have been made to derive post-peak GHB parameters and to use these to model rock mass behaviour in the post failure regime. This paper critically discusses several of these approaches and tests two of these against two field case studies. 1. INTRODUCTION The field scale behaviour of fractured rock masses dramatically impacts constructability and serviceability of infrastructure and production facilities in civil, petroleum and mining enterprises. Unanticipated rock mass failure can result in significant construction delays and cost over runs in infrastructure projects and in the potential loss of production facilities. In underground mining a significant operational cost component stems from ground support in the form of rock bolts, mesh and screen, cable bolts, and shotcrete. As mining progresses much of the critical mine infrastructure, [e.g. footwall haulage drives; stope access development, etc], are subject to changing mine induced stress conditions. Should any of this support approach or reach failure as a result of these changing stress conditions, (e.g. stripped or broken cable bolts, severe bagging of screen, large numbers of broken rock bolts, etc.), then the support will require rehabilitation. This rehabilitation is often as or more expensive than the initial support installation, not due simply to the ground support cost, but to loss of access, down-time of that area of the mine, potential for injuries, the slow nature of the work, and possible lost mining revenues and/or reserves. The question then becomes: can support design be made more robust? That is, can the initial design of mining support be done to a level such that the type, length and capacity of support initially installed can survive the changing mine induced stresses, hence greatly reducing overall direct mining costs? Most ground support elements are installed as dowels, [i.e. un-tensioned tendons such as passive cables, rebar, etc.]. The issue here is that support is only provided when sufficient dilation or movement of the rock mass has occurred to generate resisting tensile forces in the support element. Significant stress-driven movement only occurs in a rock mass after the peak strength of the rock has been exceeded and it moves into a state of yield or failure. In other words, based on experience, stress driven ground movements are not sufficient to load the support to any significant amount prior to yield of the rock mass. These same stress induced yield processes can also dramatically impact the stability of mine production stopes and pillars with both cost and potential safety implications.
- Europe > United Kingdom > England (0.46)
- North America > Canada > Ontario (0.28)
- Geology > Rock Type (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- Materials > Metals & Mining (1.00)
- Energy > Oil & Gas > Upstream (0.93)
- Oceania > Australia > Western Australia > North West Shelf > Carnarvon Basin > Exmouth Basin > Pyrenees Development > Block WA-42-L > Harrison Field (0.89)
- Oceania > Australia > Western Australia > North West Shelf > Carnarvon Basin > Exmouth Basin > Pyrenees Development > Block WA-12-R > Harrison Field (0.89)
- Asia > Thailand > Chiang Mai > Fang Basin (0.89)
ABSTRACT: In today's increasingly complex well designs, often the range in drilling fluid densities required to prevent hole collapse without fracturing the wellbore, i.e., the safe drilling window, is narrow. This is often the case for wells having high angles of deviation, and is of particular concern in extended-reach drilling (ERD) wells. With increasing step-outs, the drilling equivalent circulating density (ECD) continues to climb with increasing measured depth (MD) while the true vertical depth (TVD) remains fairly constant. The net result can be that as horizontal departure lengths continue to increase, the drilling ECD violates the safe drilling window. A recent study was presented to the industry, in which the effect of changes in the water phase salinity (WPS) of invert emulsion drilling fluids (IEF) were investigated as a function of rock shear failure for two very different shales: a deepwater West Africa shale and a more-competent Oklahoma shale. After a 3-hr exposure time, the changes in rock strength with different WPS levels were measured directly and results were qualitatively consistent for the two shales. In this paper, the effect of changes in the WPS of IEF on the safe drilling window is demonstrated in an ERD drilling scenario. INTRODUCTION The drilling of challenging wells is often characterized by a narrow safe drilling window, with the fracture initiation pressure as the upper bound and the wellbore hole collapse pressure as the lower bound. For those cases where the collapse pressures are below the pore pressure, the pore pressure then serves as the safe drilling window lower bound. The safe drilling window is commonly gauged in terms of equivalent mud weight (EMW) and represents the acceptable range of density for maintaining wellbore stability. When wellbore is circulated, for example during drilling operations, the EMW is equivalent in value to the ECD. When the wellbore is static, the EMW is equivalent in value to the equivalent static density (ESD), which may or may not be equivalent in value to the surface density. Depending on the downhole pressure and temperature conditions, the ESD is often not equal to the density measured under ambient conditions, especially for IEFs. In the modeling of stresses around an intact wellbore, the safe drilling window becomes more narrow with increasing hole angle. Specifically, a deviated wellbore profile requires increased EMW to prevent collapse and reduced EMW to initiate fractures. Hence in ERD wells, the safe drilling window can become very narrow at the higher angles of deviation, as is shown in Fig. 1. In the drilling of ERD wells having long departures, the ECD will continue to increase with increasing measured depth, as the frictional pressure produced while circulating increases with hole length while the TVD does not increase by much, if at all. The effect of ECD in these kinds of wells serves to increase the operating EMW to the point where the upper bound of the safe drilling window can be violated and initiation of wellbore fracturing is predicted, as shown in Fig. 2.
- North America > United States (1.00)
- Asia > Middle East > Saudi Arabia (0.68)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- North America > United States > Oklahoma > Anadarko Basin > Cana Woodford Shale Formation (0.99)
- Africa > West Africa (0.93)