Oil Saturation Distribution Research Articles

Spatial distribution of remaining movable and non-movable oil fractions in a depleted Maastrichtian chalk reservoir, Danish North Sea: Implications for CO2 storage

Depleted oil and gas fields constitute potentially important storage sites for CO2 in the subsurface, but large-scale injection of supercritical (sc) CO2 in chalk has not yet been attempted. One of the risks is the adverse effect of the substantial amount of remaining oil in the chalk reservoirs on scCO2 injection. In order to counter an undesired effect on injectivity, a fundamental understanding of the spatial distribution and quantity of the movable, semi-movable, and non-movable oil, and solid bitumen/asphaltenes fractions of the remaining oil is critical. In this study a combination of organic geochemistry (gas chromatography of the saturated fraction and programmed pyrolysis), and reflected light microscopy was applied to evaluate and measure the spatial distribution, volume, and saturation of different oil fractions in a well-defined reservoir interval of a waterflooded Maastrichtian chalk reservoir in the Danish Central Graben, North Sea. A total of 127 samples from a slightly deviated vertical well and two ∼5 km-long horizontal wells from the Halfdan and Dan fields were analyzed. An original uneven distribution of oil saturation and composition or different production efficiency of different levels in the reservoir may account for variations in the total oil and oil fraction saturations. Gas chromatography shows that the solvent extractable oil is quite similar in composition, characterized by a dominance of polar compounds and a high content of asphaltenes. Extended slow heating (ESH) pyrolysis reveals that most of the remaining oil saturation consists of semi-movable oil and total non-movable oil (non-movable oil plus solid bitumen/asphaltenes). Reduced oil gravity values (API) are related to evaporation loss of the lightest hydrocarbon fraction during core storage and increase of the relative proportion of the heavier oil fractions by waterflooding during production. Microscopy disclosed three forms of oil: i) Patchy distributed lighter, movable oil showing a bluish fluorescence, ii) Brownish staining with a dark orange to brownish fluorescence, and iii) Dark brown non-fluorescing oil and black solid bitumen/asphaltenes occurring in microfossils and along deformation bands and stylolites, constituting the heavy non-movable oil fractions. There is a general correlation between bulk rock porosity and the total non-movable oil saturation. It thus appears that the heavy non-movable oil fractions preferentially occur in association with low-permeability heterogeneities within high-permeability stratigraphic intervals. These intervals appear to favor accumulation of non-movable oil and solid bitumen/asphaltenes and may carry a higher risk for impeding scCO2 flow.

METHOD FOR ASSESSING THE IMPACT OF GEOLOGICAL HETEROGENEITY ON OIL SATURATION DISTRIBUTION

This article examines a method for assessing the impact of reservoir heterogeneity on the distribution of oil saturation in pay zones using the example of the Bajossian and Bathonian stages of the Jurassic system of the Kalamkas field. Within this field, the Jurassic oil and gas stratum under consideration lies on the eroded surface of various horizons of the Middle and Lower Triassic and is represented by uneven alternation of sand-siltstone and clay rocks, as a result of which these deposits have a complex geological structure, due in the lithological plan to uneven alternation of sand-siltstone and clay rocks. As a result, a rather complex distribution of oil (water) saturation may occur in the formation, which differs from the standard ideas about the geological structure of the object under study, i.e. the presence of such «unconventional» reservoir boundaries (oil-bearing contour) as capillary barriers of the first kind is possible. As a result of this approach, the following is obtained. For the conditions of occurrence of productive formations of the Bajossian stage of the field in question, in order to ensure that the dimensions of the capillary barrier in the reservoir do not exceed the height of 1 m at an average value of the porosity coefficient Kп = 27,6 % it is necessary to fulfill the inequality of Kпр ≥ 6,1∙10−3 μm2. For the conditions of occurrence of productive formations of the Bathonian stage, in order for the dimensions of the capillary barrier in the reservoir not to exceed the height of 1 m with an average value of the porosity coefficient Kп = 28,0 % it is necessary to Kпр ≥ 7,5∙10−3 μm2. Also, on the basis of this method, various ratios of Kпр permeability coefficients and porosity were obtained Kп at which the dimensions of the capillary barrier in the deposit will exceed the heights of 1.0, 1.5 and 2.0 m above the level of the accepted oil-water contact (OWC) for the conditions of occurrence of the Kalamkas field in question. It was noted that this approach makes it possible to assess uncertainties in the position of fluid contacts when justifying the boundaries of the oil potential contour and the distribution of oil reserves in formations with a wide range of reservoir properties.

Influence of pore structure heterogeneity on channeling channels during hot water flooding in heavy oil reservoir based on CT scanning

Hot water flooding is an effective way to develop heavy oil reservoirs. However, local channeling channels may form, possibly leading to a low thermal utilization efficiency and high water cut in the reservoir. The pore structure heterogeneity is an important factor in forming these channels. This study proposes a method that mixes quartz sand with different particle sizes to prepare weakly heterogeneous and strongly heterogeneous models through which hot water flooding experiments are conducted. During the experiments, computer tomography (CT) scanning identifies the pore structure and micro remaining oil saturation distribution to analyze the influence of the pore structure heterogeneity on the channeling channels. The oil saturation reduction and average pore size are divided into three levels to quantitatively describe the relationship between the channeling channel distribution and pore structure heterogeneity. The zone where oil saturation reduction exceeds 20% is defined as a channeling channel. The scanning area is divided into 180 equally sized zones based on the CT scanning images, and three-dimensional (3D) distributions of the channeling channels are developed. Four micro remaining oil distribution patterns are proposed, and the morphology characteristics of micro remaining oil inside and outside the channeling channels are analyzed. The results show that hot water flooding is more balanced in the weakly heterogeneous model, and the oil saturation decreases by more than 20% in most zones without narrow channeling channels forming. In the strongly heterogeneous model, hot water flooding is unbalanced, and three narrow channeling channels of different lengths form. In the weakly heterogeneous model, the oil saturation reduction is greater in zones with larger pores. The distribution range of the average pore size is larger in the strongly heterogeneous model. The network remaining oil inside the channeling channels is less than outside the channeling channels, and the hot water converts the network remaining oil into cluster, film, and droplet remaining oil.

The impact of heterogeneity and pore network characteristics on single and multi-phase fluid propagation in complex porous media: An X-ray computed tomography study

This study investigates the impact of pore network characteristics on fluid flow through complex and heterogeneous porous media, providing insights into the factors affecting fluid propagation in such systems. Specifically, high-resolution or micro X-ray computed tomography (CT) imaging techniques were utilized to examine outcrop stromatolite samples of the Lagoa Salgada, considered flow analogous to the Brazilian Pre-salt carbonate reservoirs. The petrophysical results comprised two distinct stromatolite depositional facies, the columnar and the fine-grained facies. By generating pore network model (PNM), the study quantified the relationship between key features of the porous system, including pore and throat radius, throat length, coordination number, shape factor, and pore volume. The study found that the less dense pore network of the columnar sample is typically characterized by larger pores and wider and longer throats but with a weaker connection of throats to pores. Both facies exhibited less variability in the radius of the pores and throats in comparison to throat length. Additionally, a series of core flooding experiments coupled with medical CT scanning was designed and conducted in the plug samples to assess flow propagation and saturation fields. The study revealed that the heterogeneity and presence of disconnected or dead-end pores significantly impacted the flow patterns and saturation. Two-phase flow patterns and oil saturation distribution reveal a preferential and heterogeneous displacement that mainly swept displaced fluid in some regions of plugs and bypassed it in others. The relation between saturation profiles, porosity profiles, and the number of fluid flow patterns for the samples was evident. Only for the columnar plug sample was the enhancement in recovery factor after shifting to lower salinity water injection (SB) observed.

Remaining oil potential and adjustment countermeasures in fault X well area of Lamadian Oilfield

The distribution characteristics of remaining oil in fault well area X of Lamadian oilfield are studied by numerical simulation, and adjustment countermeasures are formulated. Through the establishment of numerical model, the geological reserves and residual reserves of each small layer are analyzed, and the oil saturation distribution diagram output by the model is calculated, and the distribution characteristics of different types of remaining oil are summarized. The results are as follows :(1) the residual reserves are mainly distributed in the channel, accounting for 57.0 % of the total residual reserves, followed by the main sheet sand; (2) The remaining oil is mainly affected by imperfect injection-production relationship, fault occlusion and plane interference, accounting for 95 % of the total reserves of the type. Imperfect injection-production relationship accounted for 73.58 % of remaining oil types. The second is the enrichment of remaining oil from plane disturbance, accounting for 12.3 %. Residual oil of fault occlusion accounted for 9.16 %. The optimal well spacing was 200m, and the final recovery was 2.84 % higher than the original well pattern.

A deep learning based surrogate model for reservoir dynamic performance prediction

Predicting the reservoir dynamic performance is an important basis for exploiting the remaining oil and improving the recovery. Traditional prediction techniques primarily include numerical simulation and reservoir engineering methods, which show technical drawbacks such as low computational efficiency and too simplified assumptions. To tackle this challenge, we propose a surrogate model for reservoir dynamic performance prediction based on a Variety of View deep convolutional encoder-decoder (VoV-DCED) network. The influence of production time is accounted for using dimension expansion. Our model directly predicts the remaining oil saturation distribution with known reservoir permeability maps and varying well schedules. We examine the performance of our model in 2D and 3D waterflooding reservoirs, which shows that the prediction results are in good consistence with the reservoir numerical simulation. The model can be extended to account for the impacts of other factors, such as initial oil saturation distribution, well patterns, and non-frequent heterogeneities. Although deep learning takes a longer time for dataset preparation in comparison to the traditional simulation, the prediction efficiency of our trained model is almost two orders of magnitude higher than that of numerical simulation. Therefore, this surrogate model can act as an efficient tool for automatic history matching and production optimization—which require multiple simulation runs.

3D-MODELING TECHNOLOGY OF INITIAL OIL SATURATION IN TRANSITIONAL OIL-WATER ZONE BY COMPLEX OF CAPILLARIMETRY AND ELECTRIC LOGGING METHODS

Link for citation: Potekhin D.V., Shiryaev E.O., Galkin S.V. 3D-modeling technology of initial oil saturation in transitional oil-water zone by complex of capillarimetry and electric logging methods. Bulletin of the Tomsk Polytechnic University. Geo Аssets Engineering, 2023, vol. 334, no. 10, рр. 98-107. In Rus. The relevance. The 3D distribution of the initial oil saturation in the reservoir volume is one of the key elements of geological and technological modeling, largely determining the quality of subsequent development design solutions. The publication presents an analysis of the current state of the problem of studying the oil-water transition zone and the distribution of oil saturation. For terrigenous hydrophilic reservoirs, electric lateral logging quite reliably controls the fluid saturation of the void space of the oil part of the deposit, which suggests the possibility of its integration with widely used capillarimetry methods. Aim: to evaluate the possibility of reliably determining the oil saturation of terrigenous hydrophilic reservoirs by combining the data of electrical lateral logging, filtration-capacitance properties of reservoirs and capillarimetric core studies, to build a 3D model of the distribution of the initial oil saturation of reservoirs in oil deposits, taking into account the allocation of the oil-water transition zone. Object: oil-saturated reservoirs of the Visean oil deposits of the Perm region. Method: methods of multidimensional mathematical modeling in the development of a methodology for determining the initial oil saturation of reservoirs based on a complex of well logging and capillarimetric studies of the core, construction of 3D distribution of the initial oil saturation of the reservoir, taking into account the parameters of the oil-water transition zone. Results. The authors, based on the developed methodology for the deposit of the reservoir Bb of the Aspinskoe field, built a 3D model of the distribution of the initial oil saturation. For the studied reservoir, they carried out the analysis of distribution of zones of free saturation, undersaturation, transition zone and limiting oil saturation. The resulting 3D model can be used in geological and technological modeling of the development of an oil deposit.

NMR and CT characterizing the influence of O2 in air injection performance in shale oil cores

BackgroundThe primary objective of this paper is to evaluate the performance of extracting shale oil resources by air injection experiments. In this paper, the shale cores were acquired from Xinjiang Permian Lucaogou formation, with an average porosity of 14% and an average permeability of 0.3mD. 2) Method: The remaining oil distribution and mobilization of oil saturation was characterized by online NMR spectra and X-ray CT scanner. Nitrogen, 8% oxygen of air and 16% oxygen of air injection were evaluated for their enhanced shale oil recovery performance. 3) Significant findings: With the increase of oxygen concentration in the air, a higher oil recovery factor was yielded. Under different O2 concentration, oil recovery factor by nitrogen drive is 45.31%, and the highest (16% oxygen) is 69.81%. Most of hydrocarbons were produced from the 1～5 μ m pore size. Approximately 50% of oil was recovered by the first 0.6 PV of air injection. Oil recovery efficiency is remarkably declined for successive 0.6 PV of air injection. From the CT images of air injection into the fractured shale cores, air injection has the potential to recover incremental oil in the presence of fractures.

Establishment of sample pool and prediction of saturation field in complex fault block reservoir based on deep learning algorithm

The remaining oil prediction of complex fault block reservoir is the key to guide the well deployment in the late high water cut stage. We formed the construction methods of 20,000 numerical simulation forward models, including the dynamic and static attributes of the reservoir, such as structural depth, reservoir thickness, permeability, porosity, injection and production fluid volume, etc. The saturation field distribution of each model is obtained through numerical simulation, so as to form the saturation field characteristic data sample database. Through the deep convolutional generative adversative neural network model, the data of sample pool was trained, in which 70% of the data was randomly selected as the training set and 30% of the data as the test set. Finally, the saturation field prediction method that could be applied to an actual block was established. The result showed that: the new method did not need to carry out the numerical simulation research on the actual region, the remaining oil saturation distribution at different time step that obtained by the deep convolutional generative adversative neural network model only need to input the reservoir physical parameters, well location coordinates, injection production and other information of the actual block, and the prediction accuracy of the method was more than 90%. The test showed that the deep convolutional generative adversative neural network model had good generalization ability, and the model could be widely used in the prediction of residual oil saturation field of complex fault block reservoirs, which greatly improved the efficiency of reservoir research.

Further discussion of CO2 huff-n-puff mechanisms in tight oil reservoirs based on NMR monitored fluids spatial distributions

Due to the poor physical properties of tight reservoirs, CO2 huff-n-puff (HNP) is considered a potential enhanced oil recovery (EOR) method after primary depletion. Optimization plays a critical role in the effective implementation of CO2 huff-n-puff. But the optimization requires a good understanding of the EOR mechanisms. In this work, the spatial distribution of oil saturation under different experimental conditions was analyzed by the NMR method to further discuss the HNP mechanisms. According to the variation of 1D frequency signal amplitude, we divided the core into the hardly movable area and movable area, the region with the obvious signal decline was defined as the movable area, and the hardly movable area was the region with limited signal decline. Based on that the recovery characteristics of different scenarios were evaluated. Firstly, the necessity of the soaking stage was studied, where three scenarios with different soaking times were carried out. Secondly, the injection pressure was adjusted to investigate the effect of the pressure gradient. The T2 spectra show that soaking has significantly improved the production of crude oil in small pores, and higher oil recovery in a single cycle is observed, but it is lower when the elapsed time (total operation time) is the same. 31.03% of oil can be recovered after 3 cycles HNP, which increases to 33.8% and 37.06% for the 4 cycles and 6 cycles cases. As the pressure gradient increases, more oil is removed out of the matrix, and the oil in the deep part of the reservoir can be effectively recovered. During the CO2 huff-n-puff process, the oil distributions are similar to the solution gas drive, the residual oil is distributed at the close end of the core and the range that the oil can be efficiently recovered is limited.

Effect of CO2 flooding in an oil reservoir with strong bottom-water drive in the Tahe Oilfield, Tarim Basin, Northwest China

The dissolution and diffusion of CO2 in oil and water and its displacement mechanism were investigated by laboratory experiment and numerical simulation for Block 9 in the Tahe oilfield, a sandstone oil reservoir with strong bottom-water drive in Tarim Basin, Northwest China. Such parameters were analyzed as solubility ratio of CO2 in oil, gas and water, interfacial tension, in-situ oil viscosity distribution, remaining oil saturation distribution, and oil compositions. The results show that CO2 flooding could control water coning and increase oil production. In the early stage of the injection process, CO2 expanded vertically due to gravity differentiation, and extended laterally under the action of strong bottom water in the intermediate and late stages. The CO2 got enriched and extended at the oil-water interface, forming a high interfacial tension zone, which inhibited the coning of bottom water to some extent. A miscible region with low interfacial tension formed at the gas injection front, which reduced the in-situ oil viscosity by about 50%. The numerical simulation results show that enhanced oil recovery (EOR) is estimated at 5.72% and the oil exchange ratio of CO2 is 0.17 t/t.

Optimizing Colloidal Stability and Transport of Polysaccharide-Coated Magnetic Nanoparticles for Reservoir Management: Effects of Ion Specificity

In this work we explore the mechanisms of ion-specific stabilization of a polysaccharide-based coating for colloidal nanomaterials used within the oil & gas industry. While nanotechnology has wide prevalence across multiple industries, its utility within this sector is largely undeveloped but has potential applications in areas including (but not limited to) exploration, drilling and production processes. For example, reservoir contrast agents in the form of superparamagnetic nanoparticles could be used to accurately determine the residual oil saturation distribution in a reservoir and thus advise enhanced oil recovery (EOR) efforts. However, deployment of such materials in oil reservoirs proves challenging in cases where high salinity subsurface environments induce nanoparticle aggregation, leading to loss of mobility. Here, we report the synthesis and characterization of dextran-coated superparamagnetic iron oxide nanoparticles (Dex-SPIONs), the colloidal stability of which was evaluated in various brine formulations at elevated temperatures. Initial dynamic light scattering (DLS) measurements reveal a lack of contingency between particle stability and total electrolyte concentration for samples comprised of synthetic seawater and low-salinity brine, the latter fluid of which possesses higher ionic strength yet preserves colloidal integrity to a much greater extent than its seawater counterpart. Further experiments point to a calcium (Ca2+) ion-specific stabilization effect wherein surface complexation of Ca2+ ions to the dextran periphery improves carbohydrate hydration and thus enhances colloidal stability. Ion selective electrode (ISE) measurements provide additional evidence of the Ca2+ - dextran binding interaction, the role of which also factors significantly into mitigation of polysaccharide degradation [as demonstrated through gel permeation chromatography (GPC)]. Finally, we assess the transport of Dex-SPIONs through porous media, including examination of retention properties with respect to variances in ionic composition.

Quantitative study of CO2 huff-n-puff enhanced oil recovery in tight formation using online NMR technology

Recent evidence suggests that CO2 huff-n-puff can significantly enhance the oil recovery in tight/shale oil reservoirs after primary depletion. However, the key driving forces of CO2 huff-n-puff for oil displacement, especially their effect on the enhanced oil recovery are still unclear. In addition, the CO2 effective distance in matrix, which is a critical parameter impacting the performance of huff-n-puff, has not been accurately studied. In this work, to identify and quantify the contribution of the related enhanced-oil-recovery (EOR) mechanisms, an online nuclear magnetic resonance (NMR) instrument was used to monitor the real-time migration of the matrix oil at all stages of the CO2 huff-n-puff process. The experiment results show that the dissolved CO2 drive controlled by molecular diffusion is the dominant mechanism during the huff-n-puff process. This EOR mechanism mainly works on the oil stored in the macropores, while the oil stored in the micropores is hardly mobilized. Base on the dynamic oil saturation distributions converted from the 1-D NMR profiles during huff-n-puff, we designed a novel 2-D evaluation system, which focuses on microscopic displacement efficiency and CO2 effective distance, to investigate the effect of the operation parameters on oil production. We found that simply increasing the injection pressure mainly improves the displacement efficiency but has less effect on the expansion of the sweep area, which limits the CO2 huff-n-puff EOR on the reservoir scale. Lengthening the soaking period extends the effective distance significantly but delays oil production. A cyclic huff-n-puff scheme with variable soaking time was proposed to maintain a high oil production rate and a large sweep area. Periodic optimization of the soaking time in the cyclic CO2 huff-n-puff process is needed for an optimal development plan.

Laboratory comparison of tertiary N2, CH4, and CO2 injection into an Inland oil field sample

Few laboratory studies compared tertiary gas injection using N2, CH4, and CO2. However, these studies only focussed on oil recovery without examining CO2 storage. This study experimentally compares the oil recovery performance of tertiary N2, CH4, and CO2 injection into a composite Hutton Sandstone core taken from the Inland oil field, Australia. CO2 storage during tertiary CO2 injection is recorded. The core receives an initial injection, which is oil; then a secondary injection, which is water; and finally a tertiary injection, which is N2, CH4, or CO2. Pressure difference across the core and fluid production are recorded. Compositional analysis is performed on the produced oil samples. To estimate oil saturation distribution in the core, numerical simulations are run to simulate the injection experiment.Tertiary oil recovery for CO2 injection is found to be 23.1% of the original oil in-place, which far exceeds the 7.3% for CH4 injection and 5.3% for N2 injection. Ultimate oil recovery was 76.5% for tertiary CO2 injection, 59.6% for tertiary CH4 injection, and 58.3% for tertiary N2 injection. We observed that tertiary oil recovery for each injected gas begins with the oil bank it has generated being produced at a reasonably constant rate after which, the production rate declines reflecting the declining oil concentration in the oil bank. For tertiary CO2 injection, mass transfer between oil and CO2 is observed, which is found to lead to more production of the lighter oil after the oil bank production stage. In addition, tertiary CO2 injection is found to store 0.57 pore volume of the injected CO2 in the core.

Experimental investigation on the recovery performance and steam chamber expansion of multi-lateral well SAGD process

Multi-lateral well SAGD process is a newly proposed recovery process in recent years, in which a multi-lateral well is applied to replace the horizontal well in classic SAGD process. In this paper, a three-dimensional physical model is applied to simulate the recovery performance of multi-lateral well SAGD process in heavy oil reservoirs. First, on the basis of Pujol-Boberg similarity criterion, the Forchheimer's law is introduced to deal with the similarity of model permeability in 3D experiment. Thus, from the actual properties of Long Lake heavy oil reservoir, the lab scale experimental parameters are obtained. Then, by using a multi-lateral wellbore model, four scaled 3D SAGD experiments are performed, including one conventional dual horizontal well SAGD experiment (base case) and three multi-lateral well based SAGD experiments. From the experimental observation, the advantages of multi-lateral well are discussed from liquid production and steam chamber expansion. Finally, by the collected images, the distribution of residual oil saturation after SAGD process is also discussed. Results indicate that the introduction of Forchheimer's law can effectively handle with the problem of kinetic similarity damage caused by high permeability during a 3D experiment. Compared with the conventional dual horizontal well SAGD process, the application of multi-lateral well can increase the oil recovery factor by above 15%. To some extent, the application of multi-lateral well can increase the duration time of plateau stage or wind down stage. Simultaneously, compared with the single application of multi-lateral well in SAGD well pair, a dual multi-lateral well SAGD process has a higher expansion rate of steam chamber and a shorter recovery time. For steam chamber, under the dragging mechanism of branch wellbore, the shape of steam chamber is a “inverted trapezoid” shape instead of the “inverted triangle” shape for conventional dual horizontal well configuration. Under the effect of branch wellbore, steam chamber can effectively reach to the reservoir boundary and increase the volume of steam chamber. The residual oil saturation mainly distributed around the reservoir boundary, which is far from the main wellbore and branch wellbore. This paper further clarifies the EOR mechanisms and steam chamber expansion rules of multi-lateral well SAGD process.

Experimental research on the mechanisms of improving water flooding in fractured-vuggy reservoirs

Fractured-vuggy reservoirs have extremely complex reservoir characteristics, leading to a large amount of remaining oil not being recovered after water flooding. In order to solve the problem of the low recovery in fractured-vuggy reservoirs, we utilized 3D printing technology to fabricate two-dimensional and three-dimensional experimental models based on the typical reservoir structure. A series of experiments with different models on improving water flooding is conducted to study the production improvement effect. The difference in oil-water distribution or oil saturation distribution before and after improved water flooding is also compared, and the mechanism is analyzed. The results show that the oil-water gravity differentiation is the most critical factor affecting the remaining oil distribution, and changing the injected pressure or its propagation direction is the root of improving water injection. Reversing injection-production wells, increasing injection strength, water plugging, and cyclic water injection can effectively increase the sweeping scope or displacement efficiency. The attic oil, residual oil in filling medium, bypass remaining oil, and water cone remaining oil can be produced by one or more improving water flooding methods. This paper provides effective measures and mechanisms for the efficient and rapid development of fractured-vuggy reservoirs.

GeoDIN - Geoscience-Based Deep Interaction Networks for Predicting Flow Dynamics in Reservoir Simulation Models

Summary Network graphs represent a general language for describing complex systems and a framework for knowledge discovery. Graph learning is a new concept with applications emerging in biomedicine, pharmacology, smart mobility, and physical reasoning. When applied to petroleum systems, such as reservoir models, graphs provide unique differentiators for the abstraction of reservoir connectivity to facilitate “reservoir-centric” machine learning (ML) applications. In this paper, we demonstrate, for the first time, the application of geoscience-based deep interaction networks (GeoDIN) to learn complex physics relationships from 3D reservoir models for fast and accurate prediction of subsurface spatio-temporal flow dynamics. We build the network graph with embedded subsurface and physics representations and train the ML model to “act like the reservoir simulator.” We use a simulation benchmark model for two-phase incompressible flow, with approximately 1.1 million grid size, one central injector, and four corner producers. Static 3D grid properties include porosity and permeability. We use full-physics simulation output to construct the interaction network (IN) graph, where graph nodes objects (nodes) represent reservoir grid cells. We embed the feature vector combining pore, oil and water volumes, and pressure and relative permeability. The graph objects representing wells are connected with well completion factors. The producing wells have embedded oil and water production rates, while the objects representing injecting wells have embedded water injection rates. We represent graph relations (edges) with bidirectional transmissibility of the source cell. To preprocess the data for ML, we scale the graph object attributes using “min-max” normalization and we normalize the graph relation attributes using Box-Cox transformation. We train the GeoDIN framework to predict oil and water saturation dynamics in space and time. When benchmarked with full-physics simulation, the INs ran on two V100 graphics processing units and substantially accelerated the prediction phase compared to the physics-based simulator running on 70 Intel Xeon E5 CPU cores. On average, the error in GeoDIN predicted spatio-temporal distribution of oil saturation remains within 5% of full-physics simulation for 90% of model grid cells, while the error in water saturation remains within 2.5% of full-physics simulation. The spatio-temporal propagation of pressure is more sensitive to local embeddings of INs, which communicate on node-to-node information transfer. This results in a larger prediction error of the GeoDIN model when benchmarked to full-physics simulation. On average, the error distribution suggests that the great majority (90 to 95%) of grid cells fall within 10 to 30% error bound relative to full-physics simulation. The presented GeoDIN approach to network learning carries a game-changing potential for the prediction of subsurface flow dynamics. As the way forward, we will investigate the implementation of graph neural networks with automated feature learning, generalization, and scaleup.

Multimodal Carbonates: Distribution of Oil Saturation in the Microporous Regions of Arab Formations

Perhaps as much as 50% of the oil-in-place in carbonate formations around the world is locked away in the easy to bypass microporosity. If some of this oil is unlocked by the improved recovery processes focused on tight carbonate formations, the world may gain a major source of lower-rate power over several decades. Here, we overview the Arab D formation in the largest oil field on earth, the Ghawar. We investigate the occurrence of microporosity of different origins and sizes using scanning electron microscopy (SEM) and pore casting techniques. Then, we present a robust calculation of the probability of invasion and oil saturation distribution in the nested micropores using mercury injection capillary pressure data available in the literature. We show that large portions of the micropores in Arab D formation would have been bypassed during primary drainage unless the invading crude oil ganglia were sufficiently long. We also show that, under prevailing conditions of primary drainage of the strongly water-wet Arab formations in the Ghawar, the microporosity there was invaded and the porosity-weighted initial oil saturations of 60–85% are expected. Considering the asphaltenic nature of crude oil in the Ghawar, we expect the invaded portions of the pores to turn mixed-wet, thus becoming inaccessible to waterflooding until further measures are taken to modify the system’s surface chemistry and/or create substantial local pore pressure gradients.

Study on the influence of fracture lengths and fracture angles on residual oil distribution based on the slab model

After fracturing in the pilot area, channeling occurs at a low fracture angle (15°). Based on the resistance-water saturation relationship, three-dimensional physical simulation methods are used in the laboratory to study the effect of different fracture angles and lengths on the residual oil distribution during the displacement process. Meanwhile, recovery percent, displacement efficiency and expanding sweep coefficient to the improvement of recovery percent are also discussed. The results show that the fracture angle and length are closely related to the oil saturation distribution in these models. As the fracture angle increases, the sweep coefficient decreases (0.2412 → 0.1463), and the recovery percent also increases (46.16 → 56.88%), but the extent of increase has been reduced (7.96 → 2.96%). The extension of the fracture length is more prone to have a cross-flow phenomenon; the sweep coefficient is reduced (0.2412 → 0.1463). Compared to the model with 1/2 oil-water well spacing, the recovery percent is decreased by 14.29%. In different fracture models, the increasable sweep coefficient has a greater impact on oil recovery than the increasable displacement efficiency (71.30 → 28.70%).

Integral effects of initial fluids configuration and wettability alteration on remaining saturation: characterization with X-ray micro-computed tomography

Wettability alteration is the key process in low salinity waterflooding to enhance oil recovery. However, wettability alteration cannot necessarily lead to better oil recovery as other factors such as initial fluids morphology plays critical roles. We used Micro-CT X-ray imaging to investigate the effects of initial water morphology and wettability alteration on final remaining oil saturation and oil ganglia size distribution. The results show that presence of percolating water pathways impedes the oil recovery and oil can be produced mostly from large pores at tertiary mode, whereas the secondary low salinity waterflooding unlocks oil from a much broader range of pores. Moreover, the in-situ contact angles decrease towards less oil-wet regardless of initial water saturation, suggesting that wettability alteration is not dependent on the initial water saturation. We observed that at the end of secondary low salinity waterflooding, larger trapped ganglia were identified which is in agreement with the ganglia mobilisation theory. This work provided novel insights into the source of additional oil recovery during low salinity waterflooding and addressed the unrecognized role of initial fluids morphology in hydrocarbon recovery and multiphase flow.