A new three-dimensional drillstring model has been developed that determines the static and dynamic behavior of bottom hole assemblies (BHAs) in realistic wellbores. The analysis approach has been validated with field data, and shows a strong agreement between observed and calculated BHA behavior. Several case studies are presented that show the practical use and benefit of the advanced model for, among other applications, unconventional horizontal drilling. A framework is also provided to show how the model can be incorporated into automated engineering processes for operations support. Validation tests were conducted using high-frequency down-hole data measured within a motor- assisted rotary-steerable BHA. The gathered data was used to verify the calculated mechanical loads, predicted lateral natural frequencies of the BHA, estimated directional performance of the down-hole assembly, as well as torsional resonance resulting from High-Frequency Torsional Oscillations (HFTO). Using the validated model, various analyses have been conducted for operators around the globe, in a multitude of different drilling environments, to aid in identifying drilling dysfunctions and optimizing BHA performance. Several case studies are presented that highlight the benefit of the modeling techniques in US unconventional shale plays as well as in the Canadian heavy-oil sands, with noticeable improvements in drilling efficiencies, tool design, and reduced non-productive time (NPT). Results from the field tests show a strong correlation between measured and calculated bending moment values, as well as lateral natural frequencies of the BHA with an average of 3% error across all data sets. The primary source of error is thought to be borehole spiraling, which is quantified through analysis of the down-hole bending moment data. In addition, the model is shown to provide close estimates to actual directional performance of both steerable mud motor and Rotary-Steerable BHAs. However, the directional calculation-measurement comparison does reveal a need to incorporate an ROP-dependency within the directional prediction algorithms. Nevertheless, even with these sources of discrepancy, the modeling approach provides a sensible prediction of the BHA's mechanical and dynamic behavior and, as shown through case studies, can be used as a planning tool for BHA design, an investigative tool for root-cause analysis, or potentially as a real-time optimization tool for avoiding harmful operating conditions.

In the current market conditions, foreseeing drilling risks and mitigating all the surprises beforehand will greatly reduce development costs. Rig personnel use the risk and message object in WITSML schema and daily drilling reports to capture rig activity, findings, unfavorable events, 24 hours well summary, etc. All the information from these objects are not extracted and effectively utilized in well planning exercise. This paper focuses on a collaborative approach of using text mining and knowledge management practices to mine unfavorable events and best practices from daily drilling reports and integrating them with subsurface information management system for maintaining risk inventory and Geospatial analysis. The proposed approach applies Information retrieval, Extraction, Clustering, Pattern Identification and Knowledge Management on daily drilling reports from offset wells to identify any data that might be relevant to drilling anomalies and best practices. Text mining converts unstructured information within the reports to structured data that can be then stored within existing subsurface data management systems. The results of text mining can be integrated with Geospatial techniques to perform risk assessment by leveraging spatially distributed large datasets from offset wells. This collaborative approach improves extracting and sharing of the risks and best practices among experts across the globe and alerting users of upcoming risks. This approach also enables building an effective risk inventory that can be reused for future well planning purposes. Using this approach, risk assessment can be executed for every rig activity, user defined depth range, hole section / size and upcoming formation markers. The results from text mining approach can also be visualized on advanced User Interfaces providing a bird's eye view of well operations. This approach streamlines the evaluation of risks in the planning stage of the well and enables effective collaboration between the drilling, geological and geophysical teams during execution. It also improves risk management and knowledge management practices. With the proper training, a well-defined drilling process, sufficient data and tools for interpretation; drilling a well should become a routine process.

Dielectric logging is used in oil and gas exploration for estimating the hydrocarbon content of the reservoir and rock texture. This information is essential for making decisions on development and production of reservoirs. This paper looks at using dielectric dispersive behavior of formation and how it can be related to widely used parameters of Archie's law in petro-physics. In dielectric logging, the electromagnetic properties, permittivity and conductivity of the formation are obtained over a frequency range. These properties exhibit a variation with respect to frequency, which is called dispersion. The dispersive behavior of formation as a whole is different from that of its constituents; water, oil and dry rock. This dispersive data can be used to obtain various petrophysical properties like water saturation, water resistivity and rock texture. Alternatively, Archie's law is widely used to relate the resistivity of formation to the resistivity of water, porosity and water saturation. The so-called Archie parameters, cementation and saturation constants, are widely used in petrophysics. This paper discusses a novel approach that uses the dielectric dispersive behavior of a formation to provide a link to parameters of Archie's law. There exist many mixing laws that predict the dielectric dispersive behavior of a rock as a mixture of water, oil and dry rock. The proposed approach can be applied to any mixing law to provide parameters of Archie's law. Detailed derivation of the Archie's parameters from known dispersive laws is provided through examples. The method is applied to published data on dispersive behavior of sample rocks and its prediction of Archie's parameters is compared with the values obtained from experiment. The proposed method is capable of relating textural parameters of rock used in mixing laws to Archie's parameters.

Borehole acoustic tools that use broadband source excitation functions can optimally excite multiple frequencies at nearly the same time, yielding a wide range of different formation acoustic response properties in the received signal. However, the analysis of received broadband signals is not straightforward in the time-domain because standard time semblance algorithms suffer from interference when received acoustic modes with different slownesses and frequency content arrive at a similar time. Another method, frequency semblance, is computationally expensive and is sensitive to noise and to other effects. We propose using specially designed phase filters that can move (delay) individual frequency contributions differently in time without changing the spectral amplitudes of the received signal, thus enabling existing time semblance methods to perform optimally at many frequencies by separating the received spectra in time.

Abstract The Charlton 30/31 field, located in the northern reef trend of the Michigan Basin was developed within the stratigraphic unit historically referred to as the Niagaran Brown. The field covers 300 acres with a structural closure of approximately 300 feet and produced 2.6 million Bbl. of oil during its primary production. A total of 6 wells were drilled during the 1970's to produce the field. This reservoir is a low porosity, low permeability limestone matrix with irregular dolomitized intervals of higher porosity and permeability. Core Energy selected this field for enhanced oil recovery (EOR) using CO2. The source of the CO2 is a nearby natural gas processing plant which removes the CO2 from the produced gas stream of local Antrim Shale formation production in order to meet pipeline quality specifications. To monitor the flooding of this reef a full field study is being conducted which includes earth model construction, reservoir simulation with history matching and the acquisition of a 4D seismic survey. This study is being funded by the US Department of Energy and the first 3D survey has been acquired. Basic and advanced geophysical analyses were performed on this initial 3D survey. These included wavelet extraction, well to seismic tie generation, horizon interpretation, seismic variance analysis and investigations into a number of other seismic attributes. The results of this interpretation along with the well data were used in static model construction. Reservoir simulation was performed using this static model and a 25-year production history. During preparation of the field for injection, it was discovered that the reservoir had been an inadvertent dump-flood with water from the overlying Dundee formation, entering the reef through corroded well casings. This significantly complicated the forward reservoir simulation that was used to predict how the CO2 would flood through this reservoir. Introduction Recent increases in oil price have focused renewed attention on older reservoirs and their potential for enhanced oil recovery. The use of CO2 in enhanced oil recovery projects has taken place for a number of years. At the same time there has been a growing global concern about the need to sequester green houses gases, such as CO2. These processes may be combined to provide not only the additionally energy needed by our society but also some of the sequestration sites needed for CO2 storage. One producing area that could benefit considerably from this combination is the Silurian reef trend of the Northern Michigan Basin, see Figure 1.

Abstract Today, the major challenge in reservoir characterization is integrating data coming from different sources in varying scales, in order to obtain an accurate and high-resolution reservoir model. The role of seismic data in this integration is often limited to providing a structural model for the reservoir. Its relatively low resolution usually limits its further use. However, its areal coverage and availability suggest that it has the potential of providing valuable data for more detailed reservoir characterization studies through the process of seismic inversion. In this paper, a novel intelligent seismic inversion methodology is presented to achieve a desirable correlation between relatively low-frequency seismic signals, and the much higher frequency wireline-log data. Vertical seismic profile (VSP) is used as an intermediate step between the well logs and the surface seismic. A synthetic seismic model is developed by using real data and seismic interpretation. In the example presented here, the model represents the Atoka and Morrow formations, and the overlying Pennsylvanian sequence of the Buffalo Valley Field in New Mexico. Generalized regression neural network (GRNN) is used to build two independent correlation models between; 1) Surface seismic and VSP, 2) VSP and well logs. After generating virtual VSP's from the surface seismic, well logs are predicted by using the correlation between VSP and well logs. The values of the density log, which is a surrogate for reservoir porosity, are predicted for each seismic trace through the seismic line with a classification approach having a correlation coefficient of 0.81. The same methodology is then applied to real data taken from the Buffalo Valley Field, to predict interwell gamma ray and neutron porosity logs through the seismic line of interest. The same procedure can be applied to a complete 3D seismic block to obtain 3D distributions of reservoir properties with less uncertainty than the geostatistical estimation methods. The intelligent seismic inversion method should help to increase the success of drilling new wells during field development. Introduction Reservoir characterization requires building a spatial model of the reservoir by using appropriate data gathered from previous studies. This spatial model is then used in flow simulators, which can predict reservoir performance. An accurate and reliable reservoir characterization study is indispensable in reservoir management. The major challenge in today's reservoir characterization is to integrate all different kinds of data to obtain an accurate and high-resolution reservoir model. The concept of data analysis forms the basis of reservoir characterization. Uncertainty, unreliability, and large variety of scales due to the different origins of the data must be taken into consideration. Together with the immense size of the data sets that must be dealt with, these issues bring complex problems, which are hard to address with conventional tools. That's why unconventional computation tools have gained much interest in data analysis in recent years. Among those modern tools; intelligent systems, which mimic the mechanism of the human mind, are a way of dealing with imprecision and partial truth[1]. It should not surprise us that using intelligent systems in reservoir characterization studies has become a widely-used method in the petroleum engineering literature. Some previous intelligent reservoir characterization applications include, but are not limited to, synthetic log generation[2,3,4], permeability estimation from logs[5,6], and predicting bulk volume of oil[7]. Let us consider different types of data used in reservoir characterization: core samples provide very high resolution information about the reservoir (fraction of inches), while seismic data have a resolution in tens of feet, and well logs have in one of inches. Because of its low resolution, seismic data is routinely used only to attain a structural view of the reservoir. On the other hand, unlike core samples or well logs, which are only available at isolated localities of a reservoir, seismic data frequently provides 3D coverage over a large area. Because of this areal coverage, researchers have always aimed to use seismic data in reservoir description. Inverse modeling of reservoir properties from the seismic data is known as seismic inversion in the literature. The process presented in this paper includes modeling of the well logs from seismic data, which is also an inverse modeling process (Figure 1). This approach attracts a lot of interest and is very important because of the necessary shift from exploration to development of existing fields[8].

Abstract The underground gas storage (UGS) industry uses more than 400 reservoirs and 15,000 wells to store and withdraw natural gas and is thus a significant contributor to gas supply in the United States. 1 Previous work indicates that many gas storage wells show a loss of deliverability each year due to numerous damage mechanisms and it is estimated that tens of millions of dollars are spent each year to recover or replace this lost deliverability. 2 Prior Gas Technology Institute (GTI) studies 3,4 have been aimed at qualitatively reviewing completion and deliverability enhancement techniques used in the storage industry, identifying specific damage mechanisms present in storage reservoirs, and establishing procedures for damage mechanism identification. This paper presents key results of a major new study sponsored by the Gas Technology Institute (GTI) that focused on a comprehensive, quantitative assessment of the effectiveness and longevity of the various stimulation treatments employed in the UGS industry. Specifically, this paper summarizes the major results of the most comprehensive quantitative study conducted to date of stimulation treatments employed in the UGS industry. The study involved seven operators and 23 reservoirs (14 carbonate and 5 sandstone). Data were collected, entered into the database, and reviewed for 381 stimulations in 365 wells. Deliverability data were entered for 159 stimulations in 155 wells (67 in carbonate reservoirs and 88 in sandstone reservoirs) with sufficient data for inclusion in the study. The most popular stimulation treatments currently employed in the UGS industry are identified, the effectiveness and longevity of various treatment types are quantified and compared, and a new quantitative indicator designed to capture the results of effectiveness and longevity calculations in a single benchmark indicator is presented. In addition, the frequency and duration of cleanup effects for various treatment types and the post-stimulation decline rates typically observed for various treatments are also quantified and compared. The implications of conclusions reached in the study, as they relate to future UGS R&D efforts, are also discussed, and specific areas of additional study are recommended. Introduction In prior GRI studies, similar issues to those addressed in our study have been examined to a very limited extent. Specifically, in a prior GRI investigation by Mauer et al., 5 Mauer conducted a broad survey of operators to determine deliverability decline rates, the types of deliverability enhancement treatments employed in the UGS industry, and the success of these techniques. All of this work was based solely on the qualitative opinions/estimates of the operators polled. No quantitative, well-specific data were collected to verify these estimates. One primary objective of the new GTI study is to perform a comprehensive, quantitative assessment of the effectiveness and longevity of the various stimulation treatments employed in the UGS industry. In the process of achieving this objective, we also gained considerable insight into two related areas important to UGS operators: post-stimulation cleanup effects and post-stimulation deliverability decline rates. This study represents the most comprehensive quantitative assessment to date of stimulation treatments employed, cleanup times experienced, and post-stimulation decline rates observed in the UGS industry. Methodology To accomplish our objective, we developed a methodology to quantify a storage well's deliverability over time using standard backpressure test data. This methodology requires determination of the well's deliverability indicator, DI (defined as the deliverability potential at a specified delta-pressure squared value, P r 2 - P wf 2 , that is representative of actual operating conditions), at least once before stimulation treatments and several times after a stimulation treatment. This was accomplished by using backpressure curves and/or the associated equations supplied by the operators. Typically, operators collect DI information only once or twice a year, and often less frequently.

Abstract Pressure relief valves are essential components of natural gas pipelines. To ensure safe operation of the pipeline it is imperative that these valves function properly and reliably. Improper installation or setup can lead to potentially catastrophic problems in the field. In this study the performance of a commonly used pilot operated safety relief valve was examined under both normal and abnormal circumstances. The fundamental operating principles of the valve are reviewed, followed by presentation of experimental data illustrating the blowdown performance of the valve. The experimental techniques employed in the laboratory to simulate pipeline conditions are also described. While the performance of the relief valve was consistent with the manufacturer's specifications under normal setup and operation, significant problems were found to exist when the pilot was incorrectly adjusted, or when pressure pulsations were present. Examples of both cyclic venting and premature relief are shown in the paper. To assist in determining when a pulsation suppressor is required a semi-emperical relationship between pulsation strength, and premature opening pressure, is proposed. Introduction Safety relief valves are installed on pipelines, typically at the discharge of compressor stations, to make certain that the operating pressure within the line never exceeds the conditions for which the system was designed. These valves are essential to the safe operation of the pipeline and must be properly installed in order to function effectively. Several factors that can affect the performance of relief valves are valve design, valve settings, sense line installation and the presence of pressure pulsations in the system to which they are attached. Numerous researchers have investigated the performance of both pressure and safety relief valves under a variety of conditions. Several have investigated the effect of flow induced pulsation on direct acting safety relief valve performance from the perspective of eliminating the pulsations. Others have analyzed the stability of valves connected to piping, or vessels. Two papers and deal with pressure pulsation affecting the opening pressure of a relief valve. While the information presented in these papers is helpful in understanding relief valve performance, they do not explore the dynamics of pilot operated relief valves. To our knowledge this is the first time that the dynamics of pilot internals have been discussed in the literature as a cause of premature release. This paper specifically focuses on the behavior of a typical safety relief valve pilot and its effect on overall valve performance. It will be shown that under normal setup and operation, the pilot behaved consistently with the manufacturer's specifications; however, it will also be shown that significant problems result when the pilot is incorrectly set or when pressure pulsation is present at the inlet of the pilot. We will begin by briefly reviewing the normal operation of the pilot. P. 183^

Abstract The natural fractures existing in the Devonian Shale of the Appalachian Basin provide the enhanced permeability necessary to make the shale economically permeability necessary to make the shale economically productive. Lineament maps are the predominant productive. Lineament maps are the predominant means of tracing fractures at the surface, but interpretation errors that are associated with their construction limit their accuracy. This study uses an electromagnetic surveying technique to eliminate these inaccuracies. The first five wells sited with the aid of this method have all intersected fractures. Such surveys appear to be useful for optimizing the location of wellsites. Introduction In recent years, the Devonian Shale has become one of the most widely explored formations in the Appalachian Basin. This formation's capability of yielding high initial flowrates and its longevity of production have caused many companies to concentrate on the shale. Because of the impermeable nature of the Devonian Shale, the production of a shale well can be substantially increased by natural fracturing in the reservoir. The key to successfully drilling the Devonian Shale is the precise location of drillsites in areas where natural fractures are highly concentrated. A typical well which is sited away from natural fracturing in the shale is unlikely to produce more than 1–3 bbl/D (0.2–0.5 m /d) and 10–50 Mcf/D to .3–1.4 m /d). Conversely, a well which penetrates a naturally in excess of 500 bbl/D (80 m /d) and 1000 Mcf/D (28 m /d). The tetonic forces involved in creating the naturally fractured areas in the Appalachian Basin have expressed many fractures to the surface of the earth. Some of the tetonic movements formed linear, slumped regions at the surface which served as drainage paths in the form of creeks. Over time, erosion has altered many of these paths significantly, moving the creekbeds away from the fracture traces. In addition, the drainage paths of some creeks never did follow fracture traces, but rather low regions produced by jointing, failing and folding near the surface. Therefore, the practice of "creekology", or locating wellsites along the creekbeds, is no more accurate than randomly selecting drillsites. Studies have shown that the width of a typical fractured trace in the Devonian Shale is 25 ft (7.6 m) and that wells located within 150 ft (46 m) of a fracture can have initial production rates double that of an average shale well. Combine these facts with the relatively high costs involved in exploring the Devonian Shale and one can see that considerable effort should be spent in trying to locate a Devonian well in a naturally fractured area. Several techniques can be used to locate naturally fractured areas using surface measurements. Some of the more common methods are lineament mapping, geochemical surveying, contact resistivity profiling and electromagnetic surveying. profiling and electromagnetic surveying. This paper will focus on the combination of lineament mapping and electromagnetic surveying as a means of detecting natural fractures. The successful application of the natural fracturing requires an understanding of the natural fracturing process that occurred in the Appalachian Basin, the construction and limitations of lineament maps, and the theory and practice of electromagnetic surveying. ORIGIN OF NATURAL FRACTURING Regardless of the surface technique used to locate natural fractures, it is assumed that a fracture detected at the surface will also exist at depth. On examining the creation of natural fractures in the Appalachian Basin, this assumption is supported. The fractured oil and gas producing structures of the Appalachian Basin were created by the thrusting folding and faulting of rock strata during the formation of the Appalachian Mountains. p. 79

This paper was prepared for the Eastern Regional Meeting to be held in Columbus, Ohio, November 8–9, 1972. Permission to copy is restricted to an abstract of not more than 300 words. Illustrations may not be copied. The abstract should contain conspicuous acknowledgment of where and by whom the paper is presented. Publication elsewhere after publication in the JOURNAL OF PETROLEUM TECHNOLOGY or the SOCIETY OF PETROLEUM ENGINEERS JOURNAL is usually granted upon requested to the Editor of the appropriate journal, provided agreement to give proper credit is made. Discussion of this paper is invited. Three copies of any discussion should be sent to the Society of Petroleum Engineers office. Such discussion may be presented at the above meeting and, with the paper, may be considered for publication in one of the two SPE magazines. Abstract The "VIBROSEIS" system of exploration had its beginnings in 1950, its first application on a production basis in 1957, its first public demonstration in Ponca City on April 10, 1958, its first published article in the February 1960 issue of Geophysics and its introduction as a tool for use by the general exploration industry in 1961. While the principles have remained the same, the intervening years have seen tremendous changes in vibrator, recorder and correlator equipment as well as in processing techniques. The main result of these changes has been continued improvement in signal quality and in m reliable and effective field operations. The "VIBROSEIS" system is an engineer system that uses hydraulically operated vibrators to send sound signals (instead of shock waves) down through the earth. The "VIBROSEIS" system input signal is a swept-frequency sinusoid (called the "sweep") that can be made to last for any length of time but in general practice has been from 3 to 24 seconds. The entire length of the sweep is partially reflected at each rock formation or seismic interface. Each reflected event then is essentially the length of the sweep duration. The total recorded field record time must be equivalent to the desired geologic record length plus the sweep duration time (18 seconds for a plus the sweep duration time (18 seconds for a 5 seconds of data and a 13 second sweep). The raw field data bears little resemblance to a conventional explosive (or impulse) seismic record, since it consists of a composite of overlapping 13-second reflections from each reflecting horizon. The "VIBROSEIS" system field recording is converted into a conventional type of seismogram by the "VIBROSEIS" or cross-correlation process. The cross-correlation of each trace of the field record with the transmitted sweep signal effectively results in pulse compression of the reflected wave trains to produce a normal appearing reflection record. The produce a normal appearing reflection record. The cross-correlation wavelet is exactly the same shape as the well known Ricker wavelet, therefore "VIBROSEIS" system and conventionally recorded data look exactly the same after processing. processing. Without resorting to complex mathematics, it is easy to demonstrate how a seismic record is correlated from the sweep frequency signal of the "VIBROSEIS" system.

Publication Rights Reserved This paper was to be presented at the Second Annual Eastern Regional Meeting of the Society of Petroleum Engineers of AIME, to be held in Charleston, West Va., Nov. 4–5, 1965, and is considered to an abstract of not more than 300 words, with no illustrations, unless the paper is specifically released to the press by the Editor of the Journal of Petroleum Technology or the Executive Secretary. Such abstract elsewhere after publication in the Journal of Petroleum Technology or Society of Petroleum Engineers Journal is granted on request, providing proper credit is given that publication and the original presentation of the paper. Discussion of this paper is invited. Three copies of any discussion should be sent to the Society of Petroleum Engineers office. Such discussion may be presented at the above meeting and, with the paper, may be considered for publication in one of the two SPE magazines. Abstract This paper summarizes the results of a preliminary investigation into the possible utilization of ultrasonics to promote the flow of fluids in porous media. The work reported is in essence a feasibility study of the ability of such waves to impart useful energy to a fluid-filled porous media, particularly for the case where one such fluid is being displaced by another. Also considered has been the possibility of selectively rejecting fluids in selected open-hole intervals by means of "focused" ultrasonic waves. A series of controlled tests was made on a sixteen-inch diameter by 1.82-in. thick piece of torpedo sandstone using diesel oil, SAE 10 oil and core test fluid as the hydrocarbons. These fluids were displaced by saline water in a series of "flood tests" designed to characterize the behavior of the system with and without the addition of sonic energy. The results of these tests [shown herein] serve to show that the addition of sonic energy increases displacement efficiency. Several reasons are hypothesized for this result. It is likewise indicated that the desired selective injectivity was feasible. Although the results of these experiments cannot be directly translated to field practice they serve as a basis for future studies. Introduction The presence of a medium is essential to the transmission of ultrasonic waves. Any material that has elasticity can propagate these waves. The propagation takes the form of a displacement of successive elements of the medium. Since all such media possess inertia, the particle continues to move after it returns to the position from which it started and finally reaches another, different, position, past the original one. From this second point it returns to its original position, about which it continues to oscillate with constantly diminishing amplitude. The elements of material will therefore execute different movements or orbits as the wave passes through them. It is the differences in these movements which characterize basic types of ultrasonic waves, but no matter what the wave type, the general properties of ultrasonics remain the same. As the wave travels through the material, successive elements in the material experience these displacements, each such element in the wave path moving a little later than its neighbor. In other words, the phase of the wave, or vibration, changes along the path of wave transmission.