SeyedBijan Mahbaz, Maurice B. Dusseault, Giovanni Cascante, Philippe Vanheeghe, June 2017, Journal of Environmental & Engineering Geophysics
Abstract Defects in steel reinforcement are critical factors in the evaluation of the service life of reinforced concrete (RC). Steel reinforcement (bar) defects or deterioration may lead to crack propagation and strength decrease in RC structural members. Deterioration also changes the steel magnetic properties; the evaluation of these changes can be investigated by an indirect passive and non-invasive method to locate and quantify defect in RC structures. Herein, a passive magnetic inspection (PMI) method is modified and used to examine its potential as a non-destructive testing (NDT) method for condition assessment of steel reinforcement. The passive magnetic flux density of steel bar with three small holes in three different positions and locations along the bar is measured in the laboratory. A signal processing methodology based on frequency spectrum analysis and filtering is applied to the experimental data, and the results are compared with the numerical simulation. The processed data from the experimental test shows the potential of PMI method to detect, locate and evaluate bar condition. Both experimental results (after signal processing) and simulation results show a good similarity for top and bottom holes. Cross-correlation of numerical simulation with experimental data was necessary to reveal detection of the side hole.
Molecular Dynamic Model Applications in Reservoir Geomechanics and Fracture Propagation in Pure Calcium Carbonate
Milad Mosharafi, SeyedBijan Mahbaz, Maurice B. Dusseault, ARMA 2017, San Francisco, California, USA, Volume: 51
Fracture initiation and propagation are important research topics in reservoir geomechanics. A high proportion of the oil and gas reservoirs around the world are carbonate reservoirs, dominated by brittle and fractured carbonate rocks such as calcium carbonate (CaCO3) and dolomite (CaMg(CO3)2). Investigation of fracture behavior has been extensively carried out by different approaches but usually involving macro-scale continuum-based methods. Such methods have limitations for analyzing discontinuous porous media like carbonate rocks. Dis-continuum based methods are needed for understanding fracture complexity challenges in naturally fractured strata. Micro-scale dis-continuum methods such as Molecular Dynamic (MD) modeling have been widely used in physics and mechanics, but much less so in rock mechanics. This article contains a literature review of MD modeling and dis-continuum methods for fracture simulation, addressing their advantages and disadvantages. Subsequently, the context of MD as a micro-scale dis-continuum method is described. Then, an MD model of pure calcium carbonate (Calcite) is developed and a numerical simulation of a uniaxial test is carried out. Results are compared to simple uniaxial experiments, and results showed fair agreement. This preliminary application of MD at a small scale may aid in understanding and implementing fracture initiation and small-scale fracture propagation during early stages of the fracture stimulation process.
Scanning Electron Microscopy (SEM) and Profilometer Scanning Microscopy to Estimate In Situ Stresses in a Dolomite Core Specimen
SeyedBijan Mahbaz, Ashley Goddard, Maurice B. Dusseault, ARMA 2013
Knowledge of the in situ stress state is an essential element of a geomechanics study. Current methods of evaluating in situ stress are based on direct and indirect measurements on the centimeter to kilometer scale. A new core-based technique to estimate in situ stress magnitudes and orientations is tested in this study. The method is based on the nanometer to micrometer scale study of samples to identify nano-features and micro-features that have a direct or indirect relation to in situ stress magnitudes and orientations. A dense dolomite core was divided into three parts; each sample was loaded with a different uniaxial load ranging from 0-45MPa, and then scanning electron microscope (SEM) images were taken and analyzed for sections from each of the samples. More specifically, microcracks and micro-laminations were detected, classified, and quantified at the nano- to micro-scale. Lastly, the surface roughness of the sample sections was studied by taking advantage of Profilometer scanning microscope images in relation to the applied stress load. This paper details the experimental procedures used and the results obtained to date.
Low strain measurement of shear modulus with resonant column and bender element tests – Frequency effects
Hassan Ali, SeyedBijan Mahbaz, Giovanni Cascante, Murray Garbinsky, GEO Montreal 2013
Laboratory determination of shear modulus at small strains is usually done through resonant column (RC) or bender element (BE) tests. However, the exact determination of the arrival time in BE tests is not always well defined. In this paper, BE and RC tests are performed on three different soils using a modified Stokoe-type RC test setup. Time domain analysis of BE signals is typically done; however, the change in the frequency content of the signals during testing is not well understood. The frequency domain analysis is required for the consistent and accurate determination of arrival times because of near-field and boundary condition effects on the results. In this study, both frequency domain and time domain analyses are used for the interpretation of BE tests. The increase in resonant frequency is compared with the increase in the main frequency component present in the BE response. Both frequencies increase according to the exponential law between the frequency (f) and the confining pressure (sigma_o) (f = a(sigma_o)^b) however, the exponents are different and the BE test shows higher sensitivity to the confining pressure. In RC testing, the change in frequency does not affect the interpretation of the results, as the wavelengths are much larger than the size of the specimen. Conversely in BE test, the change in frequency affects the participation of compressional waves in the response and the arrival of shear waves is masked. The is an increasing difference in the shear wave velocities computed from BE and RC tests with the increase in confining pressure. According the results from the thee soils studied, BE results should be corrected for the increase in frequency content specially at high confining pressures (sigma_o > 400 kPa).
Poisson Impedance as a Rock Physics Attribute for Developing Geomechanics Earth Models: Case Study from Southwest Iran
SeyedBijan Mahbaz, Maurice B. Dusseault, RockEng 2012
ABSTRACT: Development of Geomechanical Earth Models (GEM) is a prerequisite to reservoir management, involving issues such as stress path analysis, prediction of induced seismicity and sanding initiation, and coupled flow-stress modeling that leads to more realistic predictions of oil rates and rock response. Part of GEM development is the stipulation of the 3-D lithostratigraphic disposition (geometry) which is used to choose the geomechanical units. The digital geological model is then populated with quantitative and semi-quantitative data on rock properties, fluid properties, saturations, initial conditions (p, T, [σ]) and other information. Both fluid and solid properties are needed this requires integration of geomechanics skills with geophysical and rock physics skills, and this article deals with a subset of this integration process, with an example from two wells in southwest Iran. Rock physics approaches are used to predict reservoir rock and fluid properties with wellbore and 3-D seismic data, which are also used to interpolate rock mechanics properties to the inter-well regions. The first step in a generic rock physics approach is diagnostic and involves the introduction of a suitable Rock Physics model.Acoustic impedance (AI), shear impedance (SI), and density (ρ) are the fundamental rock properties usually derived using AVO equations, and these parameters, along with attenuation coefficients, can then be correlated to static rock mechanics properties that define deformability and perhaps even strength. In this study, Poisson Impedance is introduced as a new seismic attribute used for lithology and fluid differentiation in the FahliyanFormation (reservoir rock) in two oil wells in the southwest of Iran. To do this, a novel rock physics model was developed for this naturally fractured carbonate reservoir.Currently, these data are also being used to correlate to a limited rock mechanics data base to develop a more extensive GEM to serve as input to more general geomechanical and reservoir modeling tasks.
Determination of a rock physics model for the carbonate Fahliyan Formation in two oil wells in southwestern Iran
SeyedBijan Mahbaz, Hossein Memarian, Exploration Geophysics 43(10.1071/EGv43n1toc):47-57 · March 2012
Geophysical methods, especially seismic inversion, have improved considerably in recent years. The prediction of elastic behaviour is important to decrease risk in mining operations. The investigation of rock physics is a way to predict rock behaviours, especially reservoir geomechanical parameters. The first step in rock physics studies is to diagnose and introduce a suitable rock physics model. In this paper, we review rock physics models, such as the Rymer–Greenberg–Castagna model, and we compare them with real data trends in two oil wells of a carbonate reservoir (the Fahliyan Formation) in the Zagros Basin of southwestern Iran using sonic, density and porosity logs. After omitting the effect of water saturation and clay content, the best model for clean carbonate of the Fahliyan Formation was developed in two oil wells (A1 and A2).
Optimization of reservoir cut-off parameters: A case study in SW Iran
SeyedBijan Mahbaz, Yaser Mirzaahmadian, Petroleum Geoscience 17(4):355-363 · November 2011
Although reservoir quality cut-off criteria have been used formore than 50 years as a guide for economic decisions, there is still no rationalprocedure for identifying and applying them in Iranian oil and gas fields. Inother words, there are different 'rules-of-thumb' in different sections of theNational Iranian Oil Companies for determination of cut-off values. Forinstance, in one section, values of 10%, 50% and 50% are used for porosity,water saturation and shale content cut-offs, respectively; in another section,cut-off criteria are not used at all, simply an estimate of the time when 20%of oil-in-place could be produced. This paper addresses the optimization ofcut-off value estimation from raw and processed petrophysical data based onextracting the most appropriate relationship for permeability as a functionof porosity, water saturation and shale content - k = $(φ, Sw, Vsh). Theprocedure starts by looking at permeability as the key parameter in choosinga cut-off value because sometimes the minimum value (the permeabilitycut-off) is directly related to economic circumstances and is defined by theclient. Regression analysis coefficients of 0.936 and 0.870 were achieved forrelationships of the form k = $ (φ, Sw, Vsh) in the two petrofacies intervalsstudied. This leads to specification of minimum k values of permeability anddetermination of optimum cut-off values for φ, Sw and Vsh. This method isthen used to determine optimum cut-off values for the Burgan Member(sandstone) in the Kazhdumi Formation in an offshore oil field in the PersianGulf. The calculated cut-off values for this case for k = 1.0 mD are φ = 12.5%,Sw = 60% and Vsh = 27%, as opposed to the 'standard' corporate values ofφ = 10%, Sw = 50% and Vsh = 50%.
Poroelastic effects in reactivation of a finger-like hydraulic fracture
This paper studies the transient pressurization of a pre-existing, fingerlike crack in a poroelastic, permeable rock due to a fluid injection, the problem previously considered in the nonporoelastic reservoir context in the companion paper [Sarvaramini and Garagash, J. Appl. Mech., 2015]. Large-scale fluid leak-off from the crack and the associated pore-fluid diffusion within the permeable rock formation lead to dilation of the pore volume, which acts to additionally confine/close the crack. The influence of this so-called "poroelastic backstress" on the evolution of the fluid pressure in the crack and the onset of the fracture propagation are investigated. We first revisit the existing solution to an auxiliary problem of a poroelastic crack subjected to a step pressure increase, and generalize it to account for the full-space fluid leak-off diffusion. This solution is then used to formulate the solution to the transient pressurization of the crack due to a constant rate of fluid injection via the Green's function approach. Comparison to the reference solution for a fingerlike crack in a nonporoelastic reservoir shows that the poroelasticity has a minor effect on the fluid pressure evolution in the crack. However, the evolution of the fracture volume and the onset of the fracture propagation are shown to be significantly hindered by the poroelastic backstress at large enough injection time when fluid diffusion becomes fully two or three dimensional.
Breakdown of a Pressurized Fingerlike Crack in a Permeable Solid
This paper is concerned with the analysis of a low-viscosity fluid injection into a pre-existing, finger-like crack within a linear elastic, permeable rock, and of the conditions leading to the onset of the fracture propagation (i.e., the breakdown). The problem is of interest in reservoir waterflooding, supercritical CO2 injection for geological storage, and other subsurface fluid injection applications. Fluid injection into a stationary crack leads to its elastic dilation and pressurization, buffered by the fluid leak-off into the surrounding rock. The solution of the problem, therefore, requires coupling of the crack deformation and the full-space pore-fluid pressure diffusion in the permeable rock. Contrary to the case of propagating hydraulic fractures, when significant part of the energy input is dissipated in the viscous fluid flow in the fracture , we find that the viscous fluid pressure drop inside a stationary fracture can be often neglected (we establish the conditions when one can do so). This, in turn, allows for a semi-analytical solution of the problem using the Green's function method, and, furthermore, for the full analytical treatment of the small/large injection time asymptotics. We apply the transient pressurization solution to predict the the onset of the propagation based on the criteria derived from the energy considerations for a finger-like crack.
Initiation and propagation of a PKN hydraulic fracture in permeable rock: Toughness dominated regime
The present work investigates the injection of a low-viscosity fluid into a pre-existing fracture with constrained height (PKN), as in waterflooding or supercritical CO2 injection. Contrary to conventional hydraulic fracturing, where 'cake build up' limits diffusion to a small zone, the low viscosity fluid allows for diffusion over a wider range of scales. Over large injection times the pattern becomes 2 or 3-D, necessitating a full-space diffusion modeling. In addition, the dissipation of energy associated with fracturing of rock dominates the energy needed for the low-viscosity fluid flow into the propagating crack. As a result, the fracture toughness is important in evaluating both the initiation and the ensuing propagation of these fractures. Classical PKN hydraulic fracturing model, amended to account for full-space leak-off and the toughness [Garagash, unpublished 2009], is used to evaluate the pressure history and fluid leak-off volume during the injection of low viscosity fluid into a pre-existing and initially stationary. In order to find the pressure history, the stationary crack is first subject to a step pressure increase. The response of the porous medium to the step pressure increase in terms of fluid leak-off volume provides the fundamental solution, which then can be used to find the transient pressurization using Duhamel theorem [Detournay & Cheng, IJSS 1991]. For the step pressure increase an integral equation technique is used to find the leak-off rate history. For small time the solution must converge to short time asymptote, which corresponds to 1-D diffusion pattern. However, as the diffusion length in the zone around the fracture increases the assumption of a 1-D pattern is no longer valid and the diffusion follows a 2-D pattern. The solution to the corresponding integral equation gives the leak-off rate history, which is used to find the cumulative leak-off volume. The transient pressurization solution is obtained using global conservation of fluid injected into the fracture. With increasing pressure in the fracture due to the fluid injection, the energy release rate eventually becomes equal to the toughness and fracture propagates. The evolution of the fracture length is established using the method similar to the one employed for the stationary crack.