Bellorin, William Enrique (Petrobras Energia de Venezuela) | Castillo, Jose Antonio (Petrobras Energia de Venezuela) | Lopez Kovacs, Sonia Isabella (Petrobras) | Riera, Ladimir Alberto (Petrobras Energia de Venezuela)
The study area has produced over 457 MMBbl of oil since its discovery in 1954. This study is focused on the R sand reservoir, one of the most important of the area. The original oil in place (OOIP) is 140 MMBbls and before the begining of the project, the accounted recovery of OOIP was 7.2%.
The field reactivation was performed after a thorough analysis that included a geological model review and creating a numeric simulation model. This resulted in an increment of the field production which reached up to 7000 Bbls/d of oil and an enhanced recovery factor of 13.6% by March 31st, 2005.
The main producing zone of the study area comprises Early to Mid Miocene (Oficina Formation); it is interpreted as interbedded sandstone, deposited in a fluvial environment with marine to shallow marine influence. Core analysis in one well shows that R sand was deposited in coastal plain influenced by tides and characterized by connected channels and outlets. The 3D seismic information available in the area made it possible to accurately define the structural model. After defining the static model and with the support of the engineering study, the reservoir development plan could be defined, which was oriented to improve and to increase the field productivity. It mainly consisted in drilling horizontal wells where succesful results were obtained after drilling and completing seven wells. This shows the impact of this type of wells in terms of economic benefits resulting in an increment in production and the recovery factor.
The study area, is located approximately 400 km to the south east of Caracas, in the Area Mayor de Oficina. Structurally, the field is located in the southern flank of the Eastern Venezuelan Basin, in the foreland platform zone (Parnaud et al., 1995) (Fig. 1). The reactivation project of the R Sand reservoir resulted from a previous study of characterization that allowed elaborating a plan of operation constituted mainly by horizontal wells producing with high volume electrical submersible pumps from 7,000 to 20,000 BFPD. To date 7 horizontal wells throughout the reservoir have been drilled; they produce between 300 and 1100 BPPD. At the end of 2003 an electrical submersible pump with capacity to produce 20,000 BFPD was installed representing the first in its class of Petrobras in Venezuela. The definition of the stratigraphic and the structural model as well as the simulation of the reservoir played an important role in deciding the location of the horizontal section of wells. The reservoir is delimited to the south by a main fault east-west oriented, that has a vertical throw of 400 feet, to the north a water oil contact at -6760 feet, to the east by a secondary fault of direction NE-SO of 280 feet of vertical throw and to the west by a structural closure against the main fault of the area.
The producing sands of the Oficina Formation of which the R sand is part of, are included in a foreland megasequence of the Maturin Sub basin which in turn is part of the Eastern Venezuelan Basin (Fig. 1). This Neogene foreland basin is superposed to a Mesozoic passive margin (Di Croce, 1995). The Eastern Venezuelan Basin is subdivided in the Guárico and Maturin Sub basins; they are separated by the Anaco fault system (Di Croce, 1995). The South and east limits of the Eastern Venezuelan Basin are, respectively, the Guayana Shield and the Deltana Platform (Di Croce, 1995). The Maturin Sub basin constitutes the main hydrocarbon unit of the basin. To the south of this Sub-basin, the important reservoirs are in the Merecure and Oficina Formation (Fig.2), with seals of shale within these units overlain by an important and extensive seal of shale of regional character corresponding to the Freites Formation (Upper Miocene). The API gravity of the crude is very diverse, varying from light crude to heavy and extra-heavy crude. As to the oil systems of the Maturin Sub basin, one of the most important ones is the denominated Guayuta-Oficina that is related to the fields of the South flank of the sub basin and includes the main source rocks of Late Cretaceous age, Querecual and San Antonio Formations.
In order to improve the seismic imaging and delineation of a reservoir, an OVSP was run in an exploratory well; Neuquén Basin; Argentina. A second purpose of this acquisition was to accurately locate an Intermediate Casing above the main target. Acoustic Impedance Inversion of the corresponding Zero Offset VSP plus the image interpretation generated from the OVSP data helped to define the position of a sill, which was one of the objectives in this project.
The thick (hundreds of meters) and extensive Auca Mahuida volcanic complex which covers most of the region, considerably reduces the signal to noise ratio of the seismic data; strongly attenuating the high frequencies of the data due to wavefields dispersion, hampering their separation at the processing stage. This effect is stronger in surface seismic than in borehole seismic, due to greater offset and two-way traveltime, so that VSP is used as a tool in this environment. Therefore, Q factor estimation (and consequently Q inverse filtering application) computed directly from Zero offset VSP, is a very valuable technique which allowed us to recover high
frequencies in the field area improving vertical seismic resolution and helped to define the reservoir with greater accuracy.
An acoustic impedance inversion of the VSP trace was later obtained, and its interpretation was very useful to predict impedance variations, related to lithology and formation changes in the deeper portion of the stratigraphic column. This was the first deep exploratory well (4750 meters) in this part of the basin, so the implemented technique led to define deep formation tops.
The acquisition was performed by Schlumberger using the multilevel 3C high-fidelity downhole tool, Versatile Seismic Imager (VSI). It was used two vibrators simultaneously in flipflop configuration for both source positions in order to reduce the operational time.
In this paper we describe the objectives, methodology and results of a Borehole Seismic Job performed in an exploratory well, YPF.Nq.LoAm.x-1 (Loma Amarilla), La Banda Block, Neuquén Basin. This block is entirely covered by volcanics from Auca Mahuida Igneous Complex. At the well location it was estimated 100 to 150 meters of surface volcanics.
So that, in this area, the 3D seismic volume acquired in 2003, presents a poor to fair signal to noise ratio; and low frequency content (Fmax ˜40 Hz). Fig. 1.
The main targets of this project were two igneous bodies, intruded as sills; the deeper one in Cuyo Gr.; and the shallower in the Vaca Muerta shales.
Fig 1: Visualization of Borehole, 3D Seismic, Satellite Image and Topography.
The presence of basalts in the area is affecting the recovery of high frequencies reflected in the subsurface; for this reason, surface seismic is very poor in terms of resolution and it is expected that the image obtained through the OVSP helps in the interpretation of the area. The job was planned in 2 phases, intending to acquire a ZVSP from surface to 2480m in the first phase, for the second phase a ZVSP from 2480m to 3580m and also an OVSP. The first VSP data was used to calibrate the model and determine the best position for the OVSP (based on survey design). In the second phase, Q factor was estimated using ZVSP, a migrated image was obtained
from the OVSP, and an Acoustic Impedance Inversion was performed using the ZVSP. Finally, all the information was merged to obtain a complete time-depth relationship, corridor stack and Q factor estimation.
Artificial neural networks are becoming increasingly popular in the oil and gas industry. In the past, studies have been done on the use of artificial neural networks in reservoir characterization, field development and formation damage prediction, to name a few. The aim of this study is to provide guidelines to successfully develop and train an artificial neural network (ANN) that will predict reservoir properties that can give an improved history match when input into a reservoir simulation model. An ANN was developed to improve the history match with a ‘small' number of simulation runs for a reservoir that produced oil, gas and water for a period of ten years. Due to a lack of specific protocols for this type of study, the trial and error process was utilized to establish guidelines and suggestions.
The neural network was developed by using an inverse solution method to formulate the training and testing data. Normalization of the data simplified the neural network, improved its effectiveness and enhanced its performance. The feed-forward network with back-propagation and the hyperbolic tangent sigmoid function (tansig) in the hidden layers of the network proved to be most effective in the training/learning process.
Results indicated that functional links and eigenvalues of various system related matrices were effective in the training/learning process. These provided the network with the necessary connections that linked the inputs to the required outputs. It was necessary to input production differences between the historical and simulated performances at specific times to successfully train the network and predict realistic property values for the reservoir. Data structure and production time intervals influenced the training time as well as the accuracy of the predictions. If time intervals were too short, training times were longer, memorization occurred, and the network could not accurately predict the reservoir's properties. Most of the effective functional links that were successful in the training/learning process included relationships between permeability and other factors such as porosity, areas of the regions in the reservoir and the distances from the producer to the boundaries of the reservoir.
The M4.1 reservoir in the Tahoe Field located in the Gulf of Mexico was used as a case study to illustrate the use of ANNs in decreasing the amount of numerical reservoir simulations required to obtain an improved history match. The effective parameters, obtained from network development, were applied to data from the M4.1 reservoir simulations to determine which functional links and architecture would be most effective in training the network. It was observed that some of the functional links and network structures that were effective in network development were also effective in the ANN developed for the M4.1 reservoir while some were not.
Artificial neural networks are information processing systems that are a rough approximation and simplified simulation of the biological neuron network system. The first practical application of ANNs came in the late 1950s when Frank Rosenblatt and his colleagues demonstrated their ability to perform pattern recognition1. However, interest in neural networks dwindled due to its limitations as well as the lack of new ideas and powerful computers1. With some of these hurdles overcome in the 1980s, and with the development of the back-propagation algorithm for training multilayer perceptron networks, there was a renewed interest in the field. Since then, ANNs have been improved and applied in aerospace, automotive, defense, transportation, tele-communications, electronics, entertainment, manufacturing, financial, medical and the oil and gas industry, to name a few.
In recent years, there has been a growing interest in applying ANNs to petroleum engineering. ANNs in the oil and gas industry are based on supervised training algorithms that have the potential for solving many of the challenging and complex problems in the oil and gas industry2. Previously, some of the studies done on the applications of neural networks have been in reservoir characterization, field development, two-phase flow in pipes, and identification of well test interpretation models, completion analysis, formation damage prediction, permeability prediction and fractured reservoirs.
Oil identification and quantification in low resistivity laminated sand-shale sequences is a major challenge for petrophysics and reservoir engineers; essentially because the thickness of the sand laminas is usually bellow the vertical resolution of the resistivity logging tool. The presence of this lithology generates electrical anisotropy where horizontal resistivity is highly affected by the conductivity of the laminar shale volume, while vertical resistivity is higher and more sensitive to the laminar sand electrical properties. Once identified the productive low resistivity problem the prediction of movable water, creates enormous uncertainty when it comes to decide if this laminated sand should be open to production in the well. All this issues have caused the underestimation of Oil-In-Situ volumes and the lost of thousands of oil production per day in the upper Misoa Formation Reservoirs in western Venezuela. The incorporation of resistive image logs in the geological analysis of upper Misoa Reservoirs, have shown the existence of thinly laminated sand-shale sequences with laminations of an inch thick and less.
Traditionally the sands with resistivity values of 20 ohm-m and above are considered as potential oil bearing reservoirs. This assumption has been made with conventional induction resistivity tools with transmitter and receiver orientation parallel to the borehole axis, therefore providing horizontal resistivity. Recently a multi-component transmitter-receiver induction tool has been used to derive both horizontal and vertical resistivity in a section of upper Misoa Formation with resistivity values in the very limit of the pay cuttof. The 3D array induction tool showed the vertical resistivity of this reservoir, to be as high as 80 ohm-m, adding up to 30 % more of oil saturation in the laminated sands. In the decision process of shooting this sands, the mobility of the water remained unknown; until, magnetic resonance measurements where included in the study. 2D analysis of T2 and diffusivity indicated that 90% of the water contained in the reservoir was irreducible, so it would not be produced.
After completing the low resistive sands, production logging tests and well production showed 1300 BBD with 4% of water. This case opened a great opportunity in western Venezuela fields where this type of lithology can be found in most of the wells drilled trough Eocene reservoirs.
Petrophysical evaluations of thinly bedded, laminated reservoirs, such as the upper Misoa Formation in Maracaibo Lake Basin, are incorrect if traditional, empirically derived ‘bulk volume' effective porosity effective saturation models are employed. Resistivity measurements in these reservoirs, with moderate laminar shale content, are besieged by the high electrical ‘parallel conductivity' effect of the laminated shales that results in the classic low contrast, low resistivity shaly sand problem as noted by Worthington (1997). These laminated shales however may have little effect on the intrinsic reservoir properties of the sand laminae. The use of improper models, in many cases, may result in underestimation of reservoir potential and hydrocarbon reserves by as much as 40%, as noted by van den Berg and Sandor (1996). Mollison (1999) developed an analytical model and sensitivity analysis, using an orthogonal tensor resistivity model, based on electrical anisotropy (RV/RH), that is a far more accurate and robust solution than single scalar parallel conductivity models. The tensor model is easily implemented for isotropic and anisotropic shales with isotropic sands. For the solution of anisotropic sand and shale, the laminar shale volume must be determined from some external model such as Thomas-Stieber and can be geologically validated with image log data. True laminar sand porosity must also be derived from the Thomas-Stieber (1975) model and is essential to true laminar reservoir characterization. Once determined the total porosity and water saturation of the laminar sands, irreducible water saturations must be consider and compared to total water saturation in order select the intervals with reduced chance of producing water. I order to achieve this task NMR and 2D analysis can provide an accurate volume of non-movable water bounded to clays and capillary forces.
Perez, Laura Elena (Ecopetrol SA) | Gonzalez Mosquera, Julio Gabriel (ECOPETROL) | Gomez Ramirez, Vicente (ECOPETROL) | Lozano Guarnizo, Eduardo (ECOPETROL) | Tirado, Luis Sarmiento (ECOPETROL) | Vargas Medina, Jose Arnobio (ECOPETROL)
In the current and future scenario of an increasing demand for hydrocarbons, many companies have oriented their efforts to maximize the recovery in mature fields. This paper presents the implementation and results of an integrated reservoir management strategy that allowed revitalizing a field, which was previously considered as a marginal and currently is one of the main assets of the company.
Yarigui-Cantagallo Field, Colombia, is a compartmentalized, varying-dip monocline, with three main tertiary reservoirs. The field was discovered in the 1940`s and reached its production peak of 20,400 STB/day in 1962 after two aggressive drilling programs. A third drilling campaign in the 1980's had poor results and no additional wells were drilled. At 1999 production declined to 5,000 STB/day. In order to mitigate production decline, and maximize final recovery, integrated reservoir characterization including structural, stratigraphic, and petrophysical reinterpretation, geostatistic modeling, advanced production analysis, PVT and pressure reinterpretation, and reservoir simulation have been conducted.
An effective reservoir management has been implemented, including infill drilling, optimized well completion, hydraulic fracturing and production optimization. As a result, production levels had increased up to 13,000 STB/day and 40 MMSTB of reserves have been incorporated. Future implementation of a waterflooding project and additional infill program will incorporate 35 MMSTB of reserves.
Discontinuities of rock and fluid properties are very common in the reservoir due to variations of formation lithology and oil-water gravity differentiation. Fluid discontinuity also exists at the water front in the water-flooding reservoir. Interfaces with discontinuous fluid properties are called jump interfaces. It is known that fluid flow at these jump interfaces is neither continuous flow, which is dealt with conventional multiphase flow model in porous media, nor purely piston-like drive, which is the problem of dynamic interface tracking. In practice, however, by neglecting jump interfaces, fluid flow is widely treated as continuous flow by adding relative permeability into each phase equation in black oil simulation. Also, single-point upstream weighting and high order schemes have no improvement on flux rate and moving speed of phase interfaces.
This paper starts from analyzing two-phase flow with consideration of jump interfaces. In porous media, flow channels of oil and water are relatively stable at low rates: one fluid always takes first priority of displacing the same type of fluid and then displaces the other fluid when the channels of the same type of fluid are not sufficient. Therefore, the flow of a specified fluid through jump interfaces includes two parts: displacing the same type of fluid and displacing the other type of fluid. Based on this analysis, we develop a new two-phase flow model, considering discontinuities of flow properties. Unlike currently available flow models, our model takes account the effect of one fluid displacing the other fluid across jump interfaces.
We apply to a 1-D case and compare with Buckley-Leverett equation. The example demonstrates that our new model has more capabilities to describe the physical flow at the jump interface. We also discuss applications of our model on connection condition construction for jump interfaces in the numerical model and relative permeability measurements.
Fluid flow in the reservoir involves porous media and fluid. Discontinuities of rock and fluid properties are very common. Many examples exist in the reservoir porous media: faults interrupt the connection between upthrown side and downthrown side; rock properties abruptly change among multiple reservoir layers; the changes or differences of sedimentary facies during diagenetic event cause the break of lithology and property. For fluid, it is also usual to have the discontinuity of fluid properties: the shock in the injected water or gas front; property change around the oil-water and oil-gas contacts under gravity differentiation. These discontinuities bring about two main consequences. First, it is natural for the existence of numerous jumps in the reservoir and fluid flow in porous media is under the discontinuity condition. In this paper, the interfaces with discontinuous fluid properties are called jump interfaces. Second, mass transfer usually occurs through the jump interfaces. Therefore, flow in the reservoir is not a pure piston-like behavior.
A numerical model for the analysis of multiphase flow on vertical or slightly inclined wells has been developed. The model calculates flow properties (velocity of each phase, volumetric fraction of each phase, pressure and fluids properties) on gas-oil-water wells as function of depth. Fluids properties are obtained under the assumption of black oil
model by means of correlations taken from literature, requiring only petroleum °API and the gas specific gravity as
The model may be applied to simulate both liquid flow and gas-liquid flow. In this case, different flow patterns are taken
into account: -bubble, slug, dispersed bubble and annulardepending on flow conditions, which are determined from fluid properties and production rates of oil, gas and water. Flow in tubings consisting of several sections with different diameters and inclinations may also be simulated.
The model was validated by comparisons of measured and calculated the pressure variation along the well Good agreement was found between the numerically predicted pressure drop and measurements taken from different
databases from open literature. As a consequence the proposed model proves to be a reliable tool to describe the
flow on oil-gas-water wells.
The developed numerical model takes into account the most relevant effects that take place in a production well including multiphase flow, presence of different flow pattern, mass transfer from gaseous to liquid phase and influence of
gas-liquid flow pattern on wall friction. Special attention is paid to the velocity profile of each phase along the well. Ishii's
model for two-fluid flow is used to prescribe the slip velocity between liquid and gaseous phases and to determine the
acceleration term contribution to the pressure gradient. This model is actually being employed for corrosion rate calculations inside production wells.
The study of the multiphase flows (water - oil - gas) is of major importance in oil industry since it is found quite frequently during the production process. The physics involved in these flows is very complex due to interactions between the different phases. In order to deal with this complexity, sophisticated numerical models with several parameters (most of them determined from experiments) are required.
The complexity of the problem leads to a number of simplifying assumptions and to the use of correlations to model some terms of the equations. Many numerical methods have been proposed in order to prescribe flow variables (velocities of each phase, volume fraction of each phase, flow pattern, pressure gradient) along the tubing for vertical upward flow. There is a wide variety of numerical methods, including simple models where liquid and gas are supposed to have same velocity [1-3], models that account for slippage between gas and liquid but do not consider the existence of different flow patterns [4-6], models that take into account different flow patterns [7-12] to complex mechanistic models [13-17].
In this work an alternative numerical model to estimate the flow characteristics along a vertical or near vertical pipe is presented. The proposed method belongs to the class of models described in references [7-12]. However, instead of using a correlation for liquid hold up we use a correlation for the slip velocity between liquid and gaseous phases and
calculate the hold up from conservation equations. It was codified in a FORTRAN code named GOWflow.
In the next section the general equations of the model are introduced. Then the modeling of different terms taking part in the equations is presented, followed by the description of the algorithm. There is a section devoted to the validation and another one to the application. In the last section conclusions and future work are discussed.
Governing equations were obtained from mass conservation for each component and global momentum conservation
principles in steady state . The equations were averaged across the -assumed circular- section S of the pipe in order to obtain a one-dimensional model.
Recent studies have shown field data exhibiting a negative half slope trend in the pressure derivative that cannot be explained as spherical flow. In one case the well was located in an elongated fluvial reservoir bounded on one end by an aquifer acting as a constant pressure boundary. In another case the well was also in an elongated reservoir, this time crossed by a highly conductive fault. None has shown a rigorous derivation for the analytical equations for this flow regime.
This study derives flow regime equations for two new flow regimes encountered by a well near a constant pressure boundary. When there is no evidence of another nearby boundary, the pressure derivative trends are radial flow until the constant pressure boundary is encountered, and after that a straight trend with negative unit slope is observed. This flow regime is named here as dipolar flow. When the well is in an elongated reservoir near a constant pressure boundary perpendicular to the elongated direction, possible flow regimes include radial, dipolar flow, linear flow, and a flow regime with negative half slope, which is named here as dipole linear flow.
Normally falling derivative behavior due to a constant pressure boundary is assumed to signal the end of any useful parameter estimation, but the new dipole flow regimes are sensitive to permeability and to the distances to the constant pressure boundary and to boundaries defining the elongated reservoir. This study shows how to use the flow regimes to determine distances to closed and constant pressure boundaries, and to identify bedding plane permeability anisotropy (kx from well to constant pressure boundary, ky parallel to constant pressure boundary plane). The new flow regimes are present in standard single fault and rectangle models for pressure transient behavior, but they have never been rigorously derived or described.
The plot of the log of the pressure change and its derivative with respect to superposition time as a function of the log of elapsed time was first introduced by Bourdet et al.1 as an aid to type-curve matching. Referring to the Bourdet plot as the log-log diagnostic plot, Ehlig-Economides2,3 summarized relationships between pressure derivative responses and flow geometries described as "flow regimes??. In the past 20 years, flow regime analysis has been accepted by industry and used widely in commercial interpretation software.
A commonly encountered flow regime, a negative half slope trend in the derivative is known as an indication of spherical or hemispherical flow. Ehlig-Economides et al.4 studied in detail this flow regime, which is associated with the limited entry completion.
However, some field data exhibit a negative half slope in the pressure derivative but cannot be interpreted as spherical or hemispherical flow. Two cases can be found in literature. One case is given by Escobar,5 who studied an elongated reservoir with a constant pressure boundary normal to the elongation. However, the author mistakenly named the negative half sloping flow regime parabolic flow because he observed a parabolic shape for the isobars created by a numerical simulator. In this work, we will show the correct model does not give parabolically shaped isobars. The other case was presented by Al-Ghamdi et al.6 This time a highly conductive fault serves as the constant pressure boundary in an elongated reservoir. This paper showed field examples illustrating the same negative 1/2 slope trend and provided a model to match the data, but it did not provide a flow regime equation for the behavior.
In both cases the well was located in an elongated reservoir which is bounded on one end by a constant pressure boundary. The schematic plan view is given in Fig. 1. Here the term "elongated reservoir?? means a long, narrow reservoir which could have a stratigraphic origin such as fluvial deposition, or a structural origin such as parallel sealing faults.5, 7-9
In addition, unlike the rapid drop in the pressure derivative seen when a well is surrounded by a circular or square constant pressure boundary, when the well is near a single lateral linear constant pressure boundary, the pressure derivative drops with a slope of -1. Physical examples of such a constant pressure boundary include a downdip aquifer10 or an updip gas cap, or a well near a finite conductivity fault.11 A diagram of a well near a single constant pressure boundary is shown in Fig. 2. This figure also illustrates that a well near a constant pressure boundary is equivalent to the well plus an image well with a rate of opposite sign at a distance twice that between the well and the constant pressure boundary.
CO2 injection is one of the most efficient methods used to improve oil recovery, and, as world statistics shows, its use has increased recently. Under a high crude oil price scenario, field applications of enhanced oil recovery (EOR) processes are becoming economic in today's environment. The natural CO2 sources come to be an excellent opportunity because of its low cost. Since 60 years ago, 2500 km2 of carbonate formations containing CO2 were discovered in North of Mexico.
The Quebrache region contains several occurrences of natural CO2 that have been discovered during exploration of oil fields. The CO2 that has been naturally trapped in carbonate formations in this region is present in concentrations that can exceed 90% purity. Due to the high concentrations of CO2, some wells were shut-in 60 years ago, others, have been developed for CO2 production intended for industrial uses and some others as a source of gas lift operations in nearby heavy oil fields.
Recently, a plan of acquisition of information and studies to evaluate the CO2 proven reserves has been designed. In addition, analysis of wells deliverability of these natural CO2 reservoirs, located in the southwestern portion of Tampico, has been carried out. In order to understand better this field, a geological model was built and its dynamic behavior and potential was examined through several well tests. Results of the interpretation of these tests showed excellent results associated with a reservoir of good permeability, high conductivity, large drainage radius, etc. According to the geology of this region and the geochemical signatures observed, the CO2 of Quebrache field has an inorganic origin.
This paper discusses the evaluation of potential supply of CO2 of Quebrache reservoirs for EOR projects in the North of Mexico. The main region studied contains estimated proven reserves of 1.9 Tscf of CO2; however, this volume could be extended to larger amounts associated to areas under study. The CO2 from Quebrache field could be the beginning of a new era of EOR projects in Mexico. A field example of potential EOR application in a mature oil field is shown.