Valov, A. V.Golovin, S. V.Shcherbakov, V. V.Kuznetsov, D. S....
17页
查看更多>>摘要：In the paper, we analyze factors that can lead to a failure of the cement sheath of a cased wellbore. By using a fully coupled linear thermo-poro-elastic model, we describe mechanical compression of the casing by fluid pressure and inhomogeneous geological stresses together with heating or cooling of the casing relative to the reservoir temperature. Due to the characteristic small ratio of the well diameter to its length, we observe the 2D problem in the section perpendicular to the well axis by adopting the plane strain approximation. We model the construction of single-or double-casings of the well, at that, both casings can be non-centered relative to the well axis. The cement sheath can be loaded by a prestress appeared during the cement hydration. The model is implemented numerically using the finite element method (FEM). Although many articles consider individual factors of the cement sheath failure, in our paper all these factors are modeled in a complex. Among the triggering effects for the failure we distinguish the strong contrast of geological stresses, the prestress in the cement sheath coupled with the action of the temperature variation. We also found that the thermo-poroelastic coupling may lead to the significant growth of the pore pressure due to temperature changes when the pore pressure cannot dissolve either in the case of the very low permeability of cement or for the double-cased wellbore. In many situations, it is a combination of various factors that causes the failure of the cement sheath of the well, and not each factor separately.
查看更多>>摘要：Particle settling in inclined channels is an important phenomenon that occurs during hydraulic fracturing of shale gas production. In order to accurately simulate the large-scale (field-scale) proppant transport process, constructing a fast and accurate sub-scale proppant settling model, or surrogate model, becomes a critical issue. However, mapping between physical parameters and proppant settling velocity is complex, which makes the model construction difficult. Previously, particle settling has usually been investigated via high-fidelity experiments and meso-scale numerical simulations, both of which are time-consuming. In this work, we propose a new method, i.e., the multi-fidelity neural network (MFNN), to construct a settling surrogate model, which could greatly reduce computational cost while preserving accuracy. The results demonstrate that constructing the settling surrogate with the MFNN can reduce the need for high-fidelity data and thus computational cost by 80%, while the accuracy lost is less than 5% compared to a high-fidelity surrogate. Moreover, the investigated particle settling surrogate is applied in macro-scale proppant transport simulation, which shows that the settling model is significant to proppant transport and yields accurate results. The framework opens novel pathways for rapidly predicting proppant settling velocity in reservoir applications. Furthermore, the method can be extended to almost all numerical simulation tasks, especially high-dimensional tasks.
查看更多>>摘要：Crude oil-water interfacial tension in petroleum reservoir is reduced or increased due to surfactant injection or surfactant retention, respectively. Changes in the interfacial tension crucially attribute to a governing capillary pressure and hence an oil displacement in spontaneous imbibition process. While a reduction in the interfacial tension has been highlighted as one of the underlying mechanisms for enhanced oil recovery, fluctuated surfactant concentration within reservoir promptly disturbs such interfacial phenomenon. The current study therefore attempts to elucidate an influence of such changes on spontaneous imbibition by replacing surfactant concentration consecutively with two approaches: sequential decrease and sequential increase in the interfacial tension. Two fluid flow directions were examined simultaneously: co-current and counter-current flows. Dimensionless numbers were analyzed to emphasize the fluid displacement. With strongly oil-wet wettability (contact angle >= 123 degrees), capillarity was a resisting element to oil displacement and therefore controlled by the oilwater interfacial tension. The sequential-reduced interfacial tension was found to weaken such resisting capillary force gradually and resulted in consecutive incremental oil production. On the contrary, the sequential-increased interfacial tension initiated the lowest interfacial tension fluid that produced an immediate large amount of oil, but did not much displace further oil. The current study also observed a greater oil recovery obtained from a sequential reduction in the interfacial tension scheme (26.9%) compared to a conventional single reduction scheme (22.4%), with both schemes attaining same interfacial tension at last. Variation in pore-filling events was believed to attribute to such discrepancy since an inertia hindrance to oil displacement developed differently. In counter-current imbibition, same characteristics of oil displacement were observed as in co-current imbibition, with less oil produced (<= 17.6% ultimate recovery) and less sensitive to fluid changes due to negligible gravitational contribution. The results emphasized how the sequential-reduced interfacial tension exhibits a greater oil recovery by imbibition as analogy to secondary oil production by surfactant injection after water flooding, while increasing interfacial tension is likely attributed from surfactant retention could produce less oil.
Gaitan Espinosa, MauricioZavala-Arriaga, MercedesRamirez-Gonzalez, Patsy, V
9页
查看更多>>摘要：D-Limonene, a compound consisting of terpenes in its dextrogyrous form, is commonly used as a degreaser, hydrocarbon waste cleaner and as a sand oil separator. In this work, the use of this chemical is proposed as an alternative for enhanced oil recovery (EOR). Two oils from the southeast basin of Mexico (one medium and one extra-heavy oil) were tested for viscosity reduction with D-Limonene. The viscosity of the heavier oil @25 degrees C is 117,000 cP. The addition of 5% D-Limonene decreased the viscosity to 55,000 cP at same conditions, which corresponds to 53% reduction. The addition of D-Limonene also reduced the percentage of asphaltenes and paraffins in oil and increased the percentage of aromatics and resins (according to SARA analysis). D-Limonene was diluted (3% v/v) in brine and EOR tests were performed on a bioclastic sandstone rock with the medium oil. The rock has a helium permeability of 2.6 mD, slightly higher than the carbonated rocks, both characteristic of the Mexican Republic. The rock was characterized by scanning electron microscopy (SEM), porosimetry and Xray diffraction. Contact angles were measured to know the humectability of rock with respect to oil, brine and DLimonene. Coreflooding tests were carried out in a CoreFlood equipment at 50 degrees C and an average of 62.6% oil recovery with 3% D-Limonene diluted in brine was achieved. It should be noted that D-Limonene is a biodegradable and harmless product.
Werneck, Rafael de OliveiraPrates, RaphaelMoura, RenatoGoncalves, Maiara Moreira...
16页
查看更多>>摘要：Production forecasting plays an important role in oil and gas production, aiding engineers to perform field management. However, this can be challenging for complex reservoirs such as the highly heterogeneous carbonate reservoirs from Brazilian Pre-salt fields. We propose a new setup for forecasting multiple outputs using machine-learning algorithms and evaluate a set of deep-learning architectures suitable for time-series forecasting. The setup proposed is called N-th Day and it provides a coherent solution for the problem of forecasting multiple data points in which a sliding window mechanism guarantees there is no data leakage during training. We also devise four deep-learning architectures for forecasting, stacking the layers to focus on different timescales, and compare them with different existing off-the-shelf methods. The obtained results confirm that specific architectures, as those we propose, are crucial for oil and gas production forecasting. Although LSTM and GRU layers are designed to capture temporal sequences, the experiments also indicate that the investigated scenario of production forecasting requires additional and specific structures.
Hussaini, Syed RizwaahDvorkin, Jack PetrovichAl-Mukainah, Hani Salman
10页
查看更多>>摘要：The question posed in this work is whether and how elemental data obtained on small and often sub representative fragments of natural rock can be exploited to generate a physically valid transform between two physical properties useable at a coarser spatial scale, such as the reservoir scale. The pair of such properties targeted here is the porosity and electrical formation factor. We use the process-based upscaling to prove, by example, that a coarse-scale transform between these two variables can be obtained from a dataset computed on microscopic digital volumes of natural rock, including unconsolidated sand, medium-porosity sandstones, and low-and high-porosity carbonates. This upscaling method is based on constructing a coarse-scale (effective) rock volume from the elemental volumes with a subsequent simulation of a relevant physical process in this effective object, specifically the electrical current. A numerical simulation of Laplace equation governing electrical current is used to compute the effective resistivity. The effective porosity is simply the arithmetically averaged porosity of the elements. By using hundreds of random realizations of the effective object, we generate tight formation factor versus porosity transforms for each of the natural rock samples under examination. Because the proof offered here is by-example, it is not general. However, the method offered is general and, as such, can be used to address the same question as posed for various situations.
查看更多>>摘要：Directional drilling decision making is complicated by the presence of multiple objectives and constraints. Existing directional drilling advisory methods find a balanced solution through optimization over a weighted sum of the objectives, but the weights are often chosen arbitrarily or based on rough estimates. In addition, existing methods only return a single solution to a multi-objective problem that likely does not have a single, best solution. The work presented here overcomes these limitations by identifying a set of feasible, diversely performing (i.e., Pareto optimal) solutions for the directional driller to choose from. In this novel approach to directional drilling steering optimization, the directional drilling steering problem is framed as a constrained optimal control problem over a finite horizon. A wellbore propagation model that is calibrated at set interval using field data was used to estimate system responses for a bent-sub mud motor drilling assembly. Pareto front approximation was used to solve the problem and a multi-objective evolutionary search method was leveraged to solve this optimization problem and present the directional driller with a set of Pareto optimal solutions that satisfy the problem constraints. The proposed Pareto front approximation approach was validated against a set of field test cases. These demonstrate that the proposed method is capable of finding feasible, high-performing solutions to these real-world problems sufficiently fast to be of use for real-time directional drilling advisory in the field.
查看更多>>摘要：The most decisive constituent of operating expenditures in sucker-rod pumping operations is related to the system's electric power use as most installations are driven by electric motors. Consequently, the reduction of operating costs can be translated to the reduction of energy losses both downhole and on the surface. Therefore, the energy efficiency of the surface and downhole components of the pumping system as well as the overall system efficiency play a big role in maximizing profits. The paper presents a critical analysis of the energy efficiency of the individual components of the sucker-rod pumping system and introduces novel models to find the system's total efficiency. The typical energy losses in the sucker-rod pumping system's main components (the downhole pump, the sucker-rod string, the surface pumping unit, the gearbox, the V-belt drive, and the prime mover) are discussed in detail. It is demonstrated that the level of the pumping unit's counterbalancing as well as any inertial effects do not impact on the net gearbox torques and on the required average mechanical output power of the motor. The energy consumption of the electric motor, however, is affected by the pumping unit's counterbalancing and the inertial effects because of the changes in motor efficiency due its variable loading within the pumping cycle. The system's useful output power is performed by the downhole pump when it lifts the produced liquid to the surface. The paper demonstrates that the so-called hydraulic power, calculated from the increase of the produced liquid's potential energy, is a reliable indicator of the system's useful power. Using the hydraulic power together with the electric power taken from the power supply permits an easy way to assess the over-all energy efficiency of the pumping system. The paper proposes several other variants of system efficiency calculations; they are based on the fact that all components of the pumping system are connected in series to each other. Sizing of electric motors for sucker-rod pumping installations is normally done with the use of a cyclic load factor (CLF) that accounts for the fluctuations in motor load during the pumping cycle. Originally, CLFs were found from the variation of motor current, but mechanical CLFs based on net gearbox torques are much more practical to use. This practice is fully justified in the paper by proving that the correlation between the motor's real current and its net torque loading is nearly linear. It is further shown that the safety of sucker-rod pumping installation design is improved if the electric motor is sized with the use of a mechanical CLF because that case gives the highest required nameplate motor power.
查看更多>>摘要：Within polycyclic sedimentary basins, meteoric freshwater leaching plays an important role in the formation of secondary porosity. Whereas, when superimposed by other late diagenesis processes, it is often difficult to identify its diagenetic indicators. The Jurassic clastic reservoirs in the "Wmy buried hill" of the Huanghua Depression testify from such meteoric freshwater diagenesis. The "buried hill" is named after a village called "Wmy" and nowadays buried deeply into a Paleogene lacustrine sag, which displays like a highland or hill. These reservoir sandstones were studied based on a variety of approaches including 3-D seismic interpretation, core and thin section observations, X-ray fluorescence analysis, Electron Probe Microanalysis and stable isotope analysis, combined with reconstruction of the burial and thermal histories. At last, the evolutionary model of reservoir quality was addressed. Difference of sedimentary fabrics in sand bodies leads to the heterogeneity of pores distribution, making the permeability in coarse-grained and fine-grained sandstones largely different. The meteoric water leaching caused the loss of Fe and Ti, leading to a zonation of core colors with white and red zones. In addition, the dissolution by meteoric water of biotite and volcanic fragments as well as feldspar grains produced large amounts of secondary porosity without formation of diagenetic by-products, which thus points to an open diagenetic system. The calcite cement of meteoric freshwater origin displays relatively low Fe and high Mn content exhibiting a low Fe/Mn ratio. The origin of Mn is probably dissolution of volcanic fragments. The evolutionary process of reservoir quality can be divided into four stages. The second stage is the crucial period for formation of secondary porosity whilst faults derived from the compressional stress during tectonic uplif associate coal-bearing source rocks with Jurassic reservoirs. Moreover, the last stage is the critical period for petroleum and/or gas migration and accumulation because the underlying coal-measures entered the hydrocarbon generation window and overlying Paleogene sealing rocks were intensely compacted. Scale of potential reservoirs in the whole well profile are evaluated by statistical methods, the total thickness of which can reach about 25 m. Reservoirs close to the unconformity surface between the Mesozoic and Cenozoic display more effective porosity than the deeper ones, which should be the target for the subsequent hydrocarbon exploration.
查看更多>>摘要：The formation of water in oil (W/O) emulsion can result in flow issues in the reservoir and a high cost of demulsification throughout the water treatment process. Without treatment, the W/O emulsion takes a long time to settle, which results in blockage in porous media. CO2 has strong interaction with crude oil and so has the potential to alter the settling process of W/O emulsion. The purpose of this paper is to undertake a numerical and experimental investigation into the influence of CO2 on the emulsion stability. To anticipate the settling process of a W/O emulsion in a high-pressure environment, a mathematical model was created that included a CO2 diffusion model, a viscosity prediction model, and an emulsion settling model. A high-pressure reactor was used to model the emulsion demulsification process at three different pressure conditions and to evaluate the model's prediction capability. The results indicate that under a high-pressure CO2 atmosphere, the demulsification process of a W/O emulsion can be accelerated. The CO2 dissolution and diffusion into the oil phase might enhance the likelihood of droplets colliding, resulting in the demulsification of the W/O emulsion system. Pressure has a significant influence on the stability of the emulsion. High pressure accelerates the CO2 dissolution process and significantly reduces the time required for the emulsion system to reach equilibrium. Additionally, the oil/water ratio can affect the stability of a W/O emulsion. At the same pressure, an emulsion with a higher oil/water ratio required a shorter time to reach equilibrium and had a higher equilibrium viscosity.
NSTL
Elsevier