High-resolution discretizations can be advantageous in compositional simulation to reduce excessive numerical diffusion that tends to mask shocks and fingering effects. In this work, we outline a fully implicit, dynamic, multilevel, high-resolution simulator for compositional problems on unstructured polyhedral grids. We rely on four ingredients: (i) sequential splitting of the full problem into a pressure and a transport problem, (ii) ordering of grid cells based on intercell fluxes to localize the nonlinear transport solves, (iii) higher-order discontinuous Galerkin (dG) spatial discretization with order adaptivity for the component transport, and (iv) a dynamic coarsening and refinement procedure. For purely cocurrent flow, and in the absence of capillary forces, the nonlinear transport system can be perturbed to a lower block-triangular form. With counter-current flow caused by gravity or capillary forces, the nonlinear system of discrete transport equations will contain larger blocks of mutually dependent cells on the diagonal. In either case, the transport subproblem can be solved efficiently cell-by-cell or block-by-block because of the natural localization in the dG scheme. In addition, we discuss how adaptive grid and order refinement can effectively improve accuracy. We demonstrate the applicability of the proposed solver through a number of examples, ranging from simple conceptual problems with PEBI grids in two dimensions, to realistic reservoir models in three dimensions. We compare our new solver to the standard upstream-mobility-weighting scheme and to a second-order WENO scheme.
Cement holds the most critical role for providing long-term zonal isolation for permanent abandonment phase. The loss of cement integrity is undesirable as it may threaten the surrounding environment and safety on the surface. The quality of cured cement is commonly associated with the properties of cement material and cement placement in the wellbore. However, there are still limited investigations that link these factors specifically to the sealing ability of cement plug, especially with the lack of proper equipment in the past.
In the present work, a small-scale laboratory setup has been constructed to test the sealing performance of a cement plug. The cement plug is contained inside a test cell, connected to a pressurizing system and placed inside a heating cabinet. Consequently, the test can be simulated at downhole conditions in a controlled manner. By using this setup, it is possible to monitor the minimum pressure required for the plug to fail and the gas leak rate.
Two different cement systems, neat- and silica-cement, were prepared as plugging materials. Both cement systems are placed inside pipes with three different levels of surface roughness and then tested. Results show that the inner surface roughness of the pipes affects cement plug sealing significantly, and the effect is independent of the type of cement systems. Plugs placed inside a very-rough pipe significantly reduce the gas leak rate. Our results also show that an immediate gas leak occurs in all samples from leak paths formed at the cement/steel interface.
Saelevik, Gunnstein (NORCE Norwegian Research Centre AS) | Skadsem, Hans Joakim (NORCE Norwegian Research Centre AS) | Kragset, Steinar (NORCE Norwegian Research Centre AS) | Gardner, Dave (NORCE Norwegian Research Centre AS) | Randeberg, Erlend (NORCE Norwegian Research Centre AS) | Hjelstuen, Magnus (SINTEF)
The micro-sonde well logging system is a concept for obtaining along-well measurements of temperature, changes in fluid velocity and pressure while drilling. The micro-sondes are encapsulated, self-contained measurement devices designed to follow the flow from the bottom hole assembly to surface. Once retrieved at surface, the measurements can provide information about well conditions such as the presence of cuttings beds, washed out zones and gain/loss zones. Post-processing of the data requires knowledge of the forces acting on the micro-sonde and the resulting trajectories from the release point at the bottom hole assembly and to surface. The main objectives of the studies presented in this work are to understand the micro-sonde motion in the mud flow and to design and construct a demonstration device. We have performed a literature study, CFD-modelling and experiments to better understand the behavior of a micro-sonde in a typical annular flow domain. The trajectory of a single micro-sonde has been investigated as it passes tool-joints under the influence of inner string rotation. Further, the experiments investigated the likelihood of transporting the micro-sonde intact to surface for a drilling application. For the experimental studies, two annular flow loops, one vertical without rotation of the drill-string and one horizontal with a rotating drill-string, including tool-joints and a model bottom hole assembly have been constructed and utilized.
Cement sheaths are among the most important barrier elements in petroleum wells. However, the cement may lose its integrity due to repeated pressure variations in the wellbore, such as during pressure tests and fluid injections. Typical cement sheaths failure mechanisms are formation of radial cracks and microannuli, and such potential leak paths may lead to loss of zonal isolation and pressure build-up in the annulus. To prevent such barrier failures, it is important to study and understand cement sheath failure mechanisms.
This paper describes a series of experiments where we have used a tailor-built laboratory set-up to study cement sheath integrity during pressure cycling, where the set-up consists of down-scaled samples of rock, cement and casing. Cement integrity before and during casing pressurization is characterized by X-ray computed tomography (CT), which provides 3D visualization of radial cracks formed inside the cement and rock. We have studied how contextual well conditions, such as rock stiffness, casing stand-off and presence of mudfilm, influence cement sheath integrity.
The results confirm that the rock stiffness and casing stand-off determine how much casing pressure the cement can withstand before radial cracks are formed in the cement sheath, where the rock stiffness is significantly more important than casing stand-off. Furthermore, it is seen that the radial cracks in the cement sheath continue into the rock as well. However, when a thin mudfilm is present at the rock surface, the cracks stop at the cement-rock interface, and the cement sheath withstands less pressure before failure. The bonding towards the rock is thus of importance.
Shale plays an important role as cap rock above oil and gas reservoirs and above e.g. CO2 storage sites, as well as being source and reservoir rock in development of so-called unconventional reserves. Shale anisotropy needs to be accounted for in geophysical as well as geomechanical applications. This paper presents a brief description of anisotropic poroelasticity theory, and compares it to its more familiar isotropic counterpart. Experiments performed with field shales are presented, and the static mechanical behavior in terms of drained versus undrained moduli, Skempton parameters and Biot coefficients are shown to be consistent with the poroelastic approach. The necessary steps to provide static properties from seismic data and further link these measurements to laboratory ultrasonic data are briefly discussed.
Presentation Date: Tuesday, October 16, 2018
Start Time: 8:30:00 AM
Location: 202A (Anaheim Convention Center)
Presentation Type: Oral
ABSTRACT: Fluid production from or injection into a subsurface reservoir leads to stress pressure changes inside and outside the reservoir. The overburden often consists of low permeability shale, and the undrained pore pressure response may be estimated from Skempton’s empirical equations. The key parameters are Skempton’s A and B, controlling the impact of shear stress and mean stress, respectively. In this paper we show how these parameters can be deduced from anisotropic poroelasticity theory and how they are influenced by plasticity. Laboratory data verify Skempton’s relationship for shales for various stress paths, and show the predicted dependence of A on the orientation of the stress field with respect to the symmetry axis of the sample. Data also show decrease of A as failure is approached, and links this to measured volumetric strain. Assisted by results of geomechanical modeling, the Skempton parameters are used to estimate in situ pore pressure changes. Pore pressure increase is likely during injection, but may also occur for certain stress paths during depletion. The results may have impact on infill drilling and on induced seismicity.
Stress-induced pore pressure change during undrained loading of porous and permeable rocks and soils is described by Skempton’s empirically based equation (1954):
Here Δσι and Δσ3 denote changes in the maximum and minimum principal stresses, respectively, while the parameters A and B are named after Skempton. Clearly; both A and B play important roles for pore pressure evolution in the subsurface. Eq. (1) may help to quantify in situ pore pressure in sedimentary rocks, established over geological time. It may also quantify induced pore- pressure changes in formations surrounding depleting or inflating subsurface reservoirs, and near well pore pressure changes caused by stress changes during drilling or production operations.
Skempton’s original experiments were performed with various clays. While the value of B was consistently close to 1 (> 0.96), the measured A-values varied between −0.5 and + 1.5. The values of A varied between different types of clays, but also varied with stress and strain and proximity to failure. Skempton pointed to the non-elastic nature of clays for the observed variability. B-values are commonly reported in the literature, since they are relatively easy to measure, only requiring hydrostatic tests. However, there are very limited published data on A-values, particularly for rocks such as shales. Moreover, it appears that fundamental understanding of this parameter is lacking.
Sezgin, Jean-Gabriel (AIST-Kyushu University Hydrogen Materials Laboratory (HydroMate)) | Fjær, Hallvard G. (Institute for Energy Technology) | Matsunaga, Hisao (Kyushu University) | Yamabe, Junichiro (AIST-Kyushu University Hydrogen Materials Laboratory (HydroMate), Kyushu University) | Olden, Vigdis (SINTEF)
Diffusion measurements and thermal desorption measurements have been performed on a X70 pipeline steel, in as received and normalized and quenched condition, after being hydrogen charged for 200 h at 100 MPa hydrogen gas pressure and 85°C. Numerical simulations based on the assumption of thermodynamic equilibrium were performed, aiming to compare the trapping energies when fitting to the TDS spectra. TDS experiments revealed a reversible trap site with activation energy of 26.2 kJ.mol−1. Irreversible trap sites with an activation energy of >100 kJ.mol−1 were observed from both as-received and heat-treated condition. In contrast, a reversible trap site with an activation energy of 49.2 kJ.mol−1 was observed only from the heat-treated condition. The numerical modelling based on the assumption of equilibrium between hydrogen in traps and hydrogen in lattice is seen to provide a good fit to the experimental data.
Subsea oil and gas structural steel pipelines are exposed to hydrogen at the steel surface due to cathodic protection towards corrosion. Hydrogen reduces the fracture toughness in the base metal as well as in welded joints, which may be critical for the structural integrity of the pipeline. As atomic hydrogen enters the steel it occupies lattice sites and traps, as dislocations, grain boundaries and precipitates, often categorized as reversible and irreversible traps according to their trapping energy.
To be able to build predictive models for the fracture susceptibility of pipelines under operation conditions, knowledge of the amount of diffusible hydrogen and trapped hydrogen that may contribute to fracture, are vital information. Thus, knowledge of the trapping energies is essential.
In the present work, results from Thermal Desorption Spectrometry (TDS) measurements of X70 structural steel will be presented. As-received steel and heat-treated (normalized and quenched) steel, representative of the coarse grain heat affected zone of a welded joint, are investigated. Finally, the measured trapping energies, diffusivity and hydrogen concentration are discussed and compared to a numerical model, where the trapping energies are assessed.
The objective is to compare two different transient models for evaluating kick management in backpressure managed pressure drilling systems and to analyse numerical uncertainties and impact on simulation results.
Two different numerical methods will be compared with respect to how accurate they describe the maximum surface rates occurring during a kick scenario with water based mud. The importance of being aware of the uncertainty in results due to numerical diffusion is demonstrated. In addition, different techniques for reducing the numerical diffusion will be discussed and the impact on the predicted rates will be demonstrated. This will include a study of grid refinement and application of front tracking or slope limiter techniques. In addition, a comparison of a kick in oil based mud vs. water based mud in a HPHT MPD scenario will also be shown to highlight the main difference between these systems.
A backpressure managed pressure drilling system makes it possible to manage small pressure margins since bottomhole pressure can be controlled by choke pressure adjustments. However, when operating close to pore pressure, small kicks can be taken. These influxes can be circulated to surface if the surface equipment can handle the pressure and the mud gas separator have sufficient capacity. How large kicks one can handle is a decision that can be supported by transient simulations. Here it will be demonstrated that it is important to be aware of and reduce the numerical diffusion to improve prediction of the maximum rates that will occur. It will be shown that by increasing number of boxes in the discretization and introducing methods for reducing numerical diffusion, a more accurate prediction of the maximum rates occurring can be obtained.
The main contribution of this work will be to make engineers aware of that there can be uncertainties involved in the simulation tools they are using and to share knowledge about how one can reduce those uncertainties.
Plug and abandonment (P&A) of subsea wells is very costly and usually requires semi-submersible drilling rigs (SSR). To reduce total costs of the subsea P&A campaigns, it is beneficial to perform P&A operations with riserless light well intervention (RLWI) vessels instead of rigs. Currently, a drilling rig is required for performing P&A operations in the reservoir section and overburden, whereas intervention vessels can be used for preparatory work and wellhead removal.
This paper discusses how it can be technologically feasible to perform full P&A of subsea wells with RLWI vessels. It is shown that, for wells of simple and medium complexity, innovate approaches with use of existing technologies can enable full P&A of the entire well with RLWI vessels. This is demonstrated by thorough analyses of operational procedures using available technologies, where RLWI operations are compared to rig operations for different well scenarios. Furthermore, to quantify the cost benefits of the innovative approaches, a cost-optimization tool has been used to estimate the resulting cost and time durations of the different approaches and scenarios.
Libre, Jean-Marie (Total) | Collin-Hansen, Christian (Statoil) | Kjeilen-Eilertsen, Grethe (Total) | Rogstad, Tonje Waterloo (Statoil) | Stephansen, Cathrine (Akvaplan-niva) | Brude, Odd Willy (DNV GL) | Bjorgesaeter, Anders (ACONA) | Brönner, Ute (SINTEF)
Energy companies, like Statoil and TOTAL conduct Environmental Risk assessments (ERAs) as part of their risk management processes to ensure acceptable environmental risk for all operations. In some parts of the world, ERAs are required by regulators to assess risk and as a basis for evaluating risk reducing measures. A standardized ERA Acute method has been developed, providing quantitative assessment of environmental impact and risk of acute oil spills covering four environmental compartments: sea surface, shoreline, water column and seafloor. The method usesoil drift simulations and Valued Ecosystem Components (VECs) data as input. Based on a selection of relevant oil spill incidents, impact and recovery timesare calculated for VECs in all compartmentsusingcontinuous functions. Several endpoints are provided including the "Resource Damage Factor" (RDF) combining the extent of an impact with recovery time, the risk matrix, and a risk comparison tool, e.g. for quantifying effects of risk reducing measures. The methodology has been benchmarked and compared to the current industry standard ERA used on the Norwegian Continental Shelf (NCS), the MIRA method (