Production in most wells follows a predictable pattern dictated by the decline curve. Initially high production is quickly followed by a long, measured decline. This potentially long decline in production, together with an ever-increasing demand for energy, has resulted in many mature, low production fields. To stem the decline in production in these wells and extend the viable economic life of these assets, operators are increasingly turning to advances in technology. Technologies such as improved slimhole re-entry drilling bottomhole assemblies (BHAs), enhanced reservoir navigation systems, and improved conventional drilling techniques are successfully meeting the challenges of developing mature fields in established oil and gas basins around the world. This combination of techniques and technologies is being used in the Williston Basin to increase the recoverable reserves and improve the present economic value of each asset.
Production from the Williston Basin declined from the time of its initial discovery through the early 1990s. The introduction of horizontal re-entry drilling technologies then revitalized this region.
Targeting these thin beds typically requires extending 4½-in. laterals from existing or new vertical 7-in. cased wells. In the recent past, re-entry performance was confined by well placement restrictions, water and salt zones, and available drilling technologies. These requirements restricted the wellpath to relatively tight radius build sections. Together with the drilling difficulties associated with slimhole tubulars, these tight builds often resulted in high drag in the hole, limiting the lateral section to 3,000 to 4,000 ft due to weight transfer and drag constraints. However, in recent applications that did not suffer from such restrictions or the consequences of the water/salt formation hole instability issues, larger radius curves could be incorporated into the wellplan. This increased radius reduced drag in the well program, which together with improved motor technologies and experienced wellsite execution, allowed the 4½-in. re-entry section to achieve world record open slimhole multilaterals. The successful completion of such multilateral wells provide for over 10,000 ft of potential reservoir exposure in this thin bed.
The Williston Basin is a large, roughly circular sedimentary basin covering several hundred thousand square miles along the eastern edge of the Rocky Mountains straddling the northern U.S.A. states of western North Dakota, eastern Montana, as well as southern Saskatchewan, and Manitoba in Canada, as shown in Fig. 1. The scope of this paper will concentrate on the North Dakota portion of this basin.
Oil production began in earnest in the North Dakota Williston Basin in 1951 when Amerada Hess completed and produced its first commercial well. Drilling continued in a cyclical nature for the next 50+ years with production mainly based upon the Mississippian Madison Group formations, although some Mesozoic strata are productive (see Fig. 2). Currently, 3,300 oil wells are still producing in North Dakota. Additionally, there are over 14,000 wells existing in the North Dakota Williston Basin, having produced over 1.5 billion barrels of oil. Annual oil production peaked in 1984 at more than 52 million barrels of oil, followed by the expected and inevitable steep decline, as shown in Fig. 3.
The decline was temporarily arrested in the late 1980s with the introduction of horizontal drilling of the Bakken formation. Subsequent advances in horizontal drilling techniques and technologies allowed for further exploitation of other formations to alter the production decline again in the late 1990s and mid-2000s. Although production will probably never again reach the peaks achieved 30 years ago, the reversal is so drastic that production in North Dakota is currently on the rise (see Fig. 3).
The Birdbear (Nisku) formation oil-producing payzones are characterized by compartmentalized secondary porosity thin beds locked in deep muddy limestone and dolomite formations. The Nisku ‘A' dolomite is encased between two impermeable anhydrite beds, creating a large regional stratigraphic trap, ideally suited for horizontal drilling and subsequent viable economic development. In this particular area of the Williston Basin, formation thicknesses vary from 2 to 4 ft (0.7 to 1.2 m) when exposed with vertical wells. Traditionally, the Birdbear is relatively low on the list of most prolific oil-producing zones, ranking 10th overall in Table 1.
The factors that control the performance of sand control screens that use woven metal mesh as the filter media; i.e, commonly called "premium screens??, are not generally well understood by the end user. The end user is provided a premium screen with a woven metal mesh with some weave type, and told the sand control screen is better than the other premium sand control screens on the market. The end user is told that the premium sand control screen uses such material as Dutch, Dutch Twill or a Reverse Dutch Twill woven metal mesh, and some cases the manufacturer will promote the benefit of multiple layers of woven metal mesh filter media in the sand control screen to control the sand. All this information is interesting, but the end user really does not understand the significance of the information.
This paper presents the results of a study to evaluate the effect of different woven metal mesh weaves on the performance; i.e., dirt holding capacity and plugging tendency, of media commonly used in sand control screens for deep water and heavy oil completions. The work also introduces a new approach to designing sand control screens that use woven metal mesh material. The performance criteria used in the evaluation of different designs for the filter media include control of solids, and the impact of particle size distribution of the formation on the plugging on the surface and in the filtration media. Laboratory testing demonstrates that use of a different type of construction of the retention/filtration media allows for greater performance than currently used media in sand control screens.
The significance of the approach discussed in the paper is the ability to design and engineer a sand control screen with a retention/filtration media for a given particle size distribution of the formation that allows for maximum oil production and minimum solids production. Critical to the success of this approach to designing sand control screens is the thorough evaluation/testing of multiple media types to create the "formation specific?? design.
The main pay of Guantao Formation in Gudao Field is a large-scale thick positive-rhythm channel sand oil layer. At the extra high water-cut stage, the thick oil layer is waterflooded seriously, but the top remaining oil is still enrichment due to influence of reservoir heterogeneity. The research shows that the development effectiveness of tapping the potential with vertical well is bad and the economic benefit is low . If horizontal well is used, the producing degree and the recovery of the remaining reserve can be improved effectively. Among various factors of affecting development effectiveness of horizontal well, the influence of intraformational bed development is much more important. The development of intraformational bed and its control on remaining oil distribution should be understood first, then the technology policy research of tapping the potential with horizontal well was conducted in order to ensure the good development effectiveness with horizontal well. In this paper, the remaining oil distribution features under the different developmental conditions of intraformational bed are expounded in detail, the intraformational bed was divided into two kinds, developed and undeveloped; according to its developmental conditions, the technology policies of horizontal well were studied respectively, and the commensurate design criteria of horizontal well was proposed. The application of the achievements in tapping the potential of remaining oil in thick positive-rhythm oil layer during the extra-high water cut period has achieved preferable effectiveness, which provides reference for that in the same type of reservoir.
Extra-high water cut reservoir, Thick positive rhythm oil layer, Intraformational bed, Remaining oil, Horizontal well
Gudao oilfield is one of four uncompartmentalized oilfields in Shengli petroleum province, characterized by fluvial reservoir which is the primary type of reservoirs found in China . At present, Gudao oilfield has been developed for more than 30 years and is at the extra high water cut stage, the distribution of remaining oil is very complicated due to the influence of reservoir heterogeneity and long-term waterflooding, and the challenge is how to tapping the potential of remaining oil at extra-high water cut stage. The main pay of Guantao Formation in Gudao Field is a large-scale thick positive-rhythm channel sand oil layer. At the extra-high water cut stage, the thick oil layer is waterflooded seriously, but the top remaining oil is still enrichment due to influence of reservoir heterogeneity. The research shows the problem existing in the thick positive rhythm sand oil layer could be resolved at extra high water cut stage by horizontal well can be solved, the potential of low permeability interval can be tapped and the producing reserves and the oil recovery can be improved. The development condition of intraformational bed is the key of many factors influencing the development effectiveness of horizontal well because the distribution of remaining oil is significantly influenced by intraformational bed during extra-high water cut period. Based on the development condition of intraformational bed in the thick positive rhythm oil layer and its influence on the distribution of remaining oil , the technology polices of horizontal well in tapping the potential had been worked out. Therefore the advantage of horizontal well can be brought into play.
By giving a definition of general enterprise's e-information structure model, each level in the structure is described. A real enterprise's e-information is analysed corresponding to those levels. Based on a standard Information Maturity Model (IMM) that is used in IT industry currently, the special characteristics of five levels in IMM are discussed in detail. According to those features, a real enterprise's information maturity status is introduced. Compared with the theory of IMM, a full-scale analysis is bringing up for this enterprise, such as information maturity, personnel training, traditional thinking, management, IT service, consultant service and evaluation. The highest level in IMM can be achieved in the future if more attention to be paid and existing problems to be resolved.
Analysis of Case History on Enterprise's E-Information Management Levels
In general, an internal enterprise's e-information model can be structured as Fig.1. The lowest level in this model is linked to E-information of Manufacturing or Operating Level which is made up of information system units on an enterprise's basic
operation. A variety of operating information systems being applied internally contribute to the whole enterprise's e-information. In Tuha Oilfield Company of PetroChina, the e-information of operating level is referred to a series of information system owned by those production factories and science research units, such as Exploration Dynamic System, Production Real Time Database, Logging Database, Remote Mud Logging Integrative Information System, Experiment Analysis Database, and Management System for Storage, Transportation and Sale of Oil & Gas. The upper level above the Operating Level in this e-information structure model is Management E-Information. By having various of management tools, managers can work more efficiency. In Tuha Oilfield Company, they are OA, Device Management System, Human Resources Management System, Financial Management System, Handling of Goods and Materials System, Security Information Management, Project Management System, Contract Management System etc. The level above the management in this model is called Knowledgebase and Knowledge Share System that is served for whole enterprise. For instance, Tuha Oilfield Company portal site is a kind of representation of a part of Knowledge Share System. The top level in enterprise's e-information model is Decision Support System. This system is built for policy-makers in order to make quick and right decisions. Data mining application based on data warehouse is one function of DSS.
The Duri field in Sumatra, Indonesia shown in Figure 1 operated by Chevron Pacific Indonesia (CPI) is one of the largest steam flood operations in the world. Producing heavy oil (API gravity˜25) from an essentially unconsolidated reservoir with a depth that ranges from 300 - 700 feet using steam injection @ 300 - 400°F poses a unique challenge in designing an effective yet economic completion. One of the biggest problems associated with the production of the crude oil in this environment is the production of solids, i.e., sand. It is reported that greater than 1,000,000 lbs of sand per day are produced from the field. In addition to the cost of the re-completions, problems associated with disposing of this amount of sand; and the effect the produced solids have on the facilities such as stabilization of emulsions is a large cost to operations.
A program was initiated in 2002 to evaluate the effectiveness of the completions in the Duri field. This effort involved evaluated field data such as the frequency and type of workovers, the amount and size of produced solids, the nature and number of failed liners, and the frequency of stuck pumps to better understand the efficiency of the sand control completions. Local sand control gravel used in the completions was evaluated for steam dissolution and adherence to API gravel pack sand standards. An audit of the primary sand control screen manufacturer was also conducted to evaluate the quality of the sand control screens used in the completions. In addition, on-site inspections of the operational aspects of the completions were done. The results of the program showed the completion designs, and sand control screens were up to industry standards. However, several operational aspects were identified as opportunities for improvement. The results of the program are presented, and items for improvement of the completion for thermal wells in the Duri field are discussed in this paper.
Understanding the Produced Solids
The origin and cause of solids production was not well understood. A field wide effort was initiated to sample produced fluids and solids from wells, test stations and gathering stations. Bailed sands from well services and workovers were also collected for analysis. The analysis objective was to understand size, nature and amount of gathered solids at surface.
Several particle size analyses were done to understand the particle size distribution on both core and produced/gathered solids in order to get a better picture of the produced solids origin. Figure 2 shows a laser particle size distribution (LPSA) of a typical formation in the Duri field. Figures 3 & 4 show the histograms of typical produced and bailed samples from wells in the field.
Inspection of Figures 3 & 4 reveals that most of the solids produced to the surface are consistently the very small particles; whereas, the bailed samples were mostly the larger particles. This observation was not unexpected. Some of the larger material analyzed was gravel pack sand which indicated that the sand control placed in the well was compromised to some level.
Origin of Produced Solids
A major concern for the project was to understand the source of the "fine?? solids produced from the wellbore. It was not clear if the production of the fines was just characteristic of the formation or as a result of some type of interaction of the steam and the formation. Typical mineralogy percentages of the Duri formation are given in Table 1.
Several tests were conducted which consisted of initially flowing simulated formation brine through a core sample, and then stepwise decreasing the salinity. The decrease in salinity simulates the decreasing salinity of the water in the formation as a result of the injecting steam to stimulate the oil.
This paper describes the experience and lessons learned to acquire logging while drilling (LWD) formation pressure and near-wellbore mobility data in Bohai Bay. This area is known to be difficult in terms of measuring key parameters for reservoir description in conventional wireline logging (WL) programs.
While comparisons between WL and LWD, including costs savings associated with the LWD approach, are common today in operators' minds, the intangible benefits gained by real-time acquisition of these critical data are often neglected. Incorporating formation pressure testing into the drilling process, on the other hand, creates challenges to perform measurements in a timely manner as well as the need for continuous circulation while testing to ensure wellbore safety.
Formation testing at Bohai Bay is difficult because of the unconsolidated formations and all aspects associated with this type of environment, such as borehole stability, hole washouts, sanding while testing, or lost seals.
This paper describes successful test procedures, like the orientation of the probe into any direction, and discusses test examples from various hole sizes in detail. The key to achieve a high sealing success rate seems to be the ability to control and adjust the pad contact forces against the formation. Analyzing each drawdown sequence in the tool downhole and optimizing the drawdown rate and speed in the consecutive test reduces effects like sanding and improves the overall success rate.
Providing this type of formation evaluation data with an LWD tool allows a continuous approach to data evaluation and decision-making. The ability to measure accurate LWD formation pressure data in a variety of hole sizes represents a significant opportunity for safe and cost-efficient wellbore construction, especially in environments like Bohai Bay.
With the introduction of LWD formation pressure testers, it has become possible to acquire formation pressure and mobility data during short breaks in the drilling process. Formation pore pressure and near-wellbore mobility are key parameters for reservoir description. Traditionally, these data are acquired with wireline formation testers upon reaching section or well total depth (TD). In high-angle wells, this is a time-consuming operation, as the tools must be conveyed by drillpipe. Providing this type of formation evaluation with an LWD tool allows for a continuous approach to data acquisition and decision-making and represents a significant opportunity for safe and cost-efficient wellbore construction.
The success of the discussed LWD formation tester is in particular based on smart, self-learning operating processes, which improve the accuracy of the pressure and mobility data as well as the sealing success rate. In addition to mobility-dependent test times, this smart test function reduces shock effects while drawing down on tight formations and also avoids sanding in highly unconsolidated formations.
A closed loop control of the pad contact force enables optimum sealing efficiency, saving significant time for "lost seal?? retesting and avoids formation damage. Initial LWD formation pressure test results have been good. However, the drive has been to decrease test times and improve seal success and accuracy. Fig. 1 shows the improvement (on a global basis) in seal success since the introduction of the discussed smart technologies1.
A method was tried to analyze interwell heterogeneity economically and conveniently by correlation of injection and production data. The basic idea of this method is that the fluctuations of injection rate of injector and production rate of producer can be regarded as lines of wave changing with time. A correlative coefficient is used to estimate the correlation between the two lines of waves.
Taking a group of injection production wells as a study unit, centered on injection well unit, we calculate the correlative coefficient of performance parameters between production well and injection well in "correlative time window??, using the correlative coefficient to reflect the relative extent between the two lines of waves.
We study the reservoir in the following aspects according to correlativity and the delay time when the correlative coefficient reaches best: (1) correlative coefficient was used to distribute the injection rate, and to analyze the effect between injection rate and production rate. (2) The delay time is the effective response time for water flooding when the correlative coefficient reaches best, then we can determine which well has little effective response time, and which well is the main flow direction of injection well flow to. Only if the delay time is zero, can there be obvious flee channels between production well and injection well. (3) The analysis of the reason for minor production, output reducing and lack of injection (4) Evaluation of formation interconnection and identify big pore wormhole. This method has been made into software, it has visual operation interface, which is easy in operation and studying. The diagnostic example of big pore wormhole in china oilfield made an excellent agreement with the tracer result.
During the waterflooding period, heterogeneity is an intrinsic factor that determines swept volume and sweep efficiency. To some extent, injection performance of injector and production performance of its offsetting producers are the appearance of heterogeneity. A reservoir is in a system of dynamic after it has been brought into production. The chief parameters that characterize output changes of the system are oil and water production rate. The chief parameter that characterizes input changes of the system is water injection rate. The oil and water production rate and the water injection rate changes with production time. For the past decades, lots of study, which is based on the analysis of performance data, has been conducted to describe reservoir characteristics. However, there almost have been no methods which fully meet the demands of the practical waterflooding management.
In the paper, an economical and convenient method was provided to characterize reservoir heterogeneity, which is based on the calculation of correlative coefficient between injection performance of injector and production performance of its offsetting producer. In this way, reservoir characteristics of the well groups can be learnt. Then, waterflooding performance can be predicted and such treatments as well pattern modification, stimulation and water control can be recommended scientically and rationally.
2. Theory for the Dynamic Correlation Method
An injector and its offsetting production wells are taken as one study unit and injection and production of the wells are carried out in the same perforation zones. Injection performance such as monthly or daily injection rates are regarded as one line of wave changing with time. At the same time, production performance such as monthly or daily water production rate, oil production rate, water cut etc are regarded as other line of wave changing with time. Two intervals of wave with the same length are selected from the two lines of wave respectively. Correlative coefficient of two intervals of wave, which is between -1 and +1, can be calculated. The more it approaches one, the closer the correlativity of the two lines of wave is .
For the practical application of the theory mentioned above, a term named "time window?? must be put forward first. It is the length of a time interval during which injection rate of the injector has a sharp fluctuation. Time window can be determined according to injection performance curve. Once time window is specified, one study interval can be picked from injection performance and its length is the same with the length of time window. Based on the study interval of injection performance, corresponding study interval of production performance can also be known. Compared with the study interval of injection performance, the study interval of production performance can have a delay time. In other words, the time for the study interval of production performance may be later than the time for the study interval of injection performance.
Geologists often generate highly heterogeneous descriptions of reservoirs, containing complex structures which are likely to give rise to very tortuous flow paths. However, these models contain too many grid cells for multiphase flow simulation and the number of cells must be reduced by upscaling for reservoir simulation. Conventional upscaling methods often have difficulty in the representation of tortuous
flow paths, mainly due to the inappropriate assumptions concerning the boundary conditions. An accurate and practical upscaling method, which is able to preserve the flow behaviour caused by highly heterogeneous fine scale geological features, is therefore required.
In this paper, the problems encountered in routinely used upscaling approaches are outlined, and a more accurate and practical way of performing upscaling is proposed. The new upscaling method, Well Drive Upscaling (WDU), employs the actual reservoir boundary conditions. The main advantage of this method is that the dominant flow paths can be preserved, and thus the geological knowledge can be assimilated appropriately. The new method has firstly been applied to a synthetic model with a tortuous channel, and is shown to have
significant improvement over the traditional approach. The sensitivity study on the scale up factor using a bench mark model shows the advantage of the method with various scale up factors. The method was then applied to a model of a field in the central North Sea, which involves three phase flow. In the cases studied, the WDU method produced a comparable result to the dynamic Pore Volume Weighted approach which involves running the fine grid simulation and computing appropriate relative permeabilities and inter-block ransmissibilities. The new method makes the upscaling process both practical and accurate and ensures that field predictions can be made more confidently with greater geological realism.
The heterogeneity observed in a field is generally high and the geological structures therein can be complex. From a geological point of view, it would be ideal to represent each facies boundary, both vertically and horizontally, by a gridblock boundary1, 2. Also, if distinct layering exists within a genetic unit, a further split into subunits is also desirable. In practice, reservoir models are usually created at the scale of metres or less vertically and 100 metres or less areally (and each block itself may have involved small-scale upscaling3).
In many cases, detailed reservoir modelling for a highly heterogeneous reservoir may result in a large number of grid cells, e.g. 106 grid cells or more. This large number of grid cells prohibits direct simulation of the reservoir, especially for a very heterogeneous reservoir model. This is because, apart from the limitation of computational power, the high level of heterogeneity often causes instability in the simulation, i.e. it is difficult to obtain a convergenced solution. The problem becomes more severe when simulations involve three-phase
flow. In order to perform reservoir simulation on a highly heterogeneous geological model within a reasonable time frame, we have to apply appropriate upscaling techniques to reduce the number of grid cells so as to speed up the reservoir simulation and thus field development planning process.
One of the most important parameters calculated by reservoir engineers in a waterflood project is the variation of the average water saturation with time to estimate the recovery factor. For years, that average water saturation after breakthrough has been obtained by finding the intersection of the tangent to the fractional flow-water saturation curve, fw vs. Sw, at fw = 1.0. However, this technique is subjected to errors because it is difficult to determine the exact point where the tangent to the fractional flow curve intersects the curve.
This paper discusses the development of a new function that matches with all fractional flow calculation from relative permeability displacement data and oil-water viscosity ratios. That new function has four parameters that vary according to mobility ratio. The advantage of this new function is that its derivative can be used to calculate the average saturation after breakthrough and give more accurate results in estimating the waterflood predictions and the volume of oil displaced.
We present a procedure to determine the four parameters of the new general function and the advantages of its application with field examples from some reservoirs in the literature and from Maracaibo Lake, Venezuela.
Because the importance that for reservoir engineers has the estimate of waterflood project parameters such as: the volume of oil displaced at any time, the rate of oil production, and volume of water that must be handled per volume of oil once water production begins; models and aids to predict the displacement performance have been presented in the literature from time to time.1-6
Some of the prediction tools used by the engineers consist of graphical aids. Graphical methods are still useful to petroleum engineers because in some cases it is possible to get the same accuracy in a shorter time compared with computer methods. On the other hand, due to the increased knowledge of fluid mechanics through porous media, the advent of the high-speed computers, and better simulation software in the industry, it is possible to use fewer and better assumptions to adjust the models to real world scenarios.
For many years consultants, professors, students and engineers have used the intersection of the tangent to the fractional flow-water saturation curve, fw vs. Sw to calculate the average water saturation after breakthrough. In this case, we have seen in the last years how they use the new technology to expand the scale of the fw, vs. Sw plot after breakthrough for a better visualization and achieve more accurate results. Because we have been working on the waterflood predictions over a period of time, we know that this technique is subjected to errors because it is difficult to determine the exact point where the tangent to the fractional flow curve intersects the curve.
One of the problems to be solved in this case is to find a general function and its derivative that explains the displacement performance at and after the breakthrough. We have found a general function that matches with the fractional flow-water saturation curve. We have tested it with more than six sets of water-oil relative permeability data and oil-water viscosity ratios. The advantage of this new function is that its derivative can be used to calculate the average water saturation after breakthrough and give more accurate results in estimating waterflood predictions and volume of oil displaced. The new general function has four parameters that must be determined and they have some relationship between the mobility ratio and which fluid (oil/water) preferentially wets the rock. In this paper, we show the new function, its derivative and a procedure to determine the four parameters that fit any data set of water-oil permeability and viscosity ratio.
Formation evaluation in horizontal wells is always a challenge because factors affecting resistivity measurements are difficult to identify and evaluate. These factors include shoulder beds, anisotropy, borehole size, and mud-filtrate invasion. In thick carbonate reservoirs, uncertainties in resistivity have only a minor effect on the formation water saturation of producers due to high formation resistivity which dominate the saturation calculation. Formation evaluation of water injectors drilled in water swept areas of thick carbonate reservoir is different due to low formation resistivity; errors in formation resistivity directly translate to errors in calculation of formation water saturation. Therefore, accurate formation evaluation of water injectors requires an accurate estimate of formation resistivity.
In this study, 1D and 2D resistivity modeling was performed on two horizontal water injectors each drilled between two vertical water injectors. The 1Dmodeling was used to identify the dominant factors affecting resistivity measurement and correct for environmental effects. These results, along with logs from the two vertical wells were used to construct a local geological model as input to 2D modeling. This geological model was refined by using logging while drilling density images of the two horizontal wells. Both 1D and 2D modeling results were constrained with the two vertical wells.
Without modeling, formation water saturation calculated in the horizontal wells was too low. With the modeled formation resistivity, a more accurate water saturation was obtained in the horizontal injectors when compared to the vertical wells.
This study demonstrated the power and usefulness of resistivity modeling information evaluation of water swept areas.