Steamflood Pilot Research Articles

Integrated Optimization Process Advances Steamflood Performance

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 196249, “Advancing Steamflood Performance Through a New Integrated Optimization Process: Transforming the Concept Into Practice,” by Henri Silalahi, Muhamad Aji, SPE, and Afrilia Elisa, SPE, Chevron, et al., prepared for the 2019 SPE/IATMI Asia Pacific Oil and Gas Conference and Exhibition, Bali, Indonesia, 29-31 October. The paper has not been peer reviewed. Duri Field in Indonesia is the largest active steamflood project in the world. The field produces 73,000 BOPD, and 10,000 optimization jobs are executed annually to support base production. The operator identified an opportunity to implement a new process to improve the quality of reservoir analyses, define short- and long-term reservoir depletion strategies, and increase the ability to develop a prioritized queue of executable opportunities to increase production. Introduction Duri Field was discovered in 1941 in the central Sumatra Basin in Riau Province. The first production-optimization effort, in 1967, used cyclic steam stimulation. That led to a successful steamflood pilot trial in 1975 and implementation of a full steamflood areal expansion in 1985. Thirteen production areas currently exist in Duri Field as a result of steamflooding. Geological Background The geological setting of the field involves an anticline structure 20 km long and 14 km wide that covers more than 21,000 acres. The historical tectonic process results in shallow reservoir depths. The field is composed of multiple sand deposits of a tidally influenced estuarine depositional system. The reservoirs are characterized by stacked sand lobes with a total thickness of approximately 240 ft, porosity ranges from 25 to 35%, and relatively high permeability of 600 md to 2 darcy. As a giant field, Duri had an original oil in place of approximately 5 billion bbl. The oil is characterized as heavy, with gravity from 17 to 22 °API and high oil viscosity (300 cp) at 100°F.

Phased Pilot Approach Reduces Uncertainty in Carbonate Steamflood Development

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 191695, “Results From a Phased Piloting Approach in the First Eocene Reservoir of Wafra: Reducing Uncertainties in Carbonate Steamflood Development,” by David Barge, SPE, Joel Brown, Raushan Kumar, Chris Lolley, SPE, Ruurd Bartlema, and Saleh Al-Ghamdi, Chevron, prepared for the 2018 SPE Annual Technical Conference and Exhibition, Dallas, 24–26 September. The paper has not been peer reviewed. The First Eocene is a multibillion-barrel heavy-oil carbonate reservoir in the Wafra field, located in the Partitioned Zone (PZ) between Saudi Arabia and Kuwait. After more than 60 years of primary production, expected recovery is low and provides a good target for enhanced-oil-recovery (EOR) processes. A phased piloting approach has been used to reduce the uncertainties (subsurface and surface) related to application of thermal EOR processes in this field. Introduction Wafra is one of four major fields located in the PZ (Fig. 1). Because of the low primary oil recovery and large original oil in place of the Wafra Eocene reservoirs, a significant EOR opportunity exists. A detailed geological description of the reservoir is provided in the complete paper. Because of key surface and sub-surface uncertainties, along with the high cost of implementing a full-field steamflood-development (FFSFD) project, a phased, sequential approach, from piloting to FFSFD, was implemented. This approach was required so that critical uncertainties could be evaluated efficiently and sequentially in a time-frame consistent with overall reservoir- development objectives while managing capital exposure. Pilot Planning and Design Steamflood Pilots. A specific set of objectives, primary and secondary, was developed for each of the continuous-steam-injection (CSI) pilots. Primary objectives were identified as being critical for progression to the next phase (piloting/FFSFD); secondary objectives were regarded as important in understanding identified uncertainties, but not necessarily project roadblocks. Success measures for the pilots were derived from the key metrics that were prerequisites for the next phase. These success measures were formulated to ensure that learnings gained will be used to support the design and operational planning of the next phase. A five-spot pattern on a 1.25-acre steamflood-pilot, small-scale test (SST) was a continuation of the phased approach in managing and mitigating technical uncertainties related to implementing steamflooding. The primary objectives of the SST were to test whether Eocene-produced water could be treated to produce suitable feedwater for steam generation and to test steam injectivity in the First Eocene reservoir. The SST was conducted in First Eocene C Zone/Steam Zone 3 and successfully proved sustained injectivity using steam generated from the produced water. The SST was a pre-lude to a multipattern steamflood pilot known as the large-scale pilot (LSP).

Successful Steamflood Pilot in 1st Eocene Carbonate Reservoir of Wafra, PZ Mitigates Key Technical Uncertainties

Summary The 1st Eocene is a multibillion-barrel heavy-oil carbonate reservoir in Wafra field in the Partitioned Zone (PZ) between the Kingdom of Saudi Arabia and Kuwait. A large-scale steamflood pilot has been successfully completed in the 1st Eocene reservoir. The large-scale pilot (LSP) was the first multipattern steamflood in a carbonate reservoir in the Middle East, and consisted of sixteen 2.5-acre inverted 5-spot patterns with associated steamflood and production facilities. The primary objective of the LSP was to identify and mitigate technical and economic risks and uncertainties in carbonate steamflooding to assist in the broader Wafra full-field steamflood development (FFSFD) decisions. The key technical uncertainties related to the steamflooding in the 1st Eocene reservoir were identified, categorized, and prioritized. These were then used as a basis to create surveillance and subsurface response plans. Success measures were developed to assess success in steamflooding this carbonate reservoir. These success measures were derived from the key metrics that were prerequisites for the FFSFD. The pilot met all the success measures, thereby mitigating the key technical uncertainties, and opened the path to FFSFD. This paper describes the elements of pilot planning and the results achieved during pilot execution. The emphasis, specifically, is on the achievements against the success measures set for the project; the insights into the pilot behavior from detailed analysis of production, pressure, and temperature data; and the progress made in identifying and mitigating key uncertainties in carbonate steamflooding.

Calcium Sulfate Scaling Risk and Inhibition for a Steamflood Project

Summary Steamflooding is a widely used technique for heavy-oil recovery. Scale control during steamflooding, however, can be challenging because the high temperature of the steamflood can decompose thermally unstable inhibitors and/or lead to the precipitation of metal-inhibitor pseudoscale. In this paper, we present the analysis of the scaling risk and scale inhibition for a pilot steamflood project in a Middle Eastern oil field. The formation of this field is a dolomite formation interbedded with anhydrite (CaSO4) streaks. Anhydrite has been observed to be the predominant scale form. Anhydrite scale was presumably formed by the increased production-system temperature resulting from steamflooding and/or the mixing of steam condensate with connate water at equilibrium with calcium sulfate minerals at lower temperature and higher solubility. Anhydrite is inherently difficult to control because of its high solubility and the high-temperature (HT) conditions under which it forms. Compared with barite and calcite, only limited knowledge has been acquired for anhydrite control. To predict the scaling tendency and inhibitor need in different wells of this field with different supersaturation levels and temperatures, a scaling-risk model has been developed. To build such a model, detailed and revised laboratory procedures have been developed to study nucleation and precipitation kinetics of anhydrite at 125–175°C, different supersaturation, different water composition, and long reaction time. Predictions of this scaling-risk model suggest a saturation index (SI) of 0.8 as a critical SI for anhydrite control at >125°C. For example, when the SI is above 0.8, anhydrite will be difficult to control in the presence of threshold inhibitor. Model predictions were benchmarked with the water-chemistry data from a total of more than 20 wells from this field, and were found to be consistent with field observations of scale occurrence in different wells. With the recommended inhibitor concentrations, anhydrite scale has been controlled in this field, which provides validation that the proposed scaling-risk model is a powerful tool to optimize the scale-treatment plan for anhydrite.

Failure of large-scale pilot evaporator duct nozzle

Failure investigation was carried out on a 12'' water evaporator duct nozzle of a large-scale steam flood pilot which is used to inject steam into a heavy crude oil reservoir. The nozzle contained cracks at the pipe, shell, pad and welds. All the materials are made of UNS31803 grade 2205 duplex stainless steel. The cracks were initially formed at the welds and then propagated through the pipe, pad and shell. The cracks were running almost parallel in a straight manner. The cracks were transgranular in nature in the form of fissures associated with second phase precipitates. The results indicated that the nozzle was subjected to repeated welding thermal cycles and the heat input was relatively high, as manifested by the presence of intense heat tint, high percentage of austenite in the weld region (about 70%), and the presence of embrittling intermetallic phases. The investigation concluded that the primary reason of failure was poor welding practice, which facilitated brittle cracking. Fatigue was the mode of failure, which resulted from vibration in the piping system.

Probabilistic Forecasting and Model Validation for the First-Eocene Large-Scale Pilot Steamflood, Partitioned Zone, Saudi Arabia and Kuwait

Summary The First-Eocene heavy-oil reservoir (1E) in the Wafra field is a candidate for steamflooding because of its world-class resource base and low-estimated primary recovery. However, industry has little experience in steamflooding carbonate reservoirs, which has prompted the staging of several 1E steamflooding tests, the latest of which is the large-scale pilot (LSP) started in 2009. To assist in facilities design, to help understand expected performance in a very heterogeneous reservoir, and to provide input to early-decision analyses, numerical thermal simulation was used to generate probabilistic forecasts. When adequate pilot history was available, the model was validated with probabilistic methods. The LSP model contained 1.5 million cells, which allowed the maintenance of adequate resolution and proper boundary conditions in the pilot area. Parallel computation enabled a probabilistic workflow to be implemented with this large thermal model. In this paper, we highlight the methodologies and inputs used to generate the probabilistic forecasts and validate the model. Major results of this work include the following: In contrast to many greenfield forecasts, the LSP forecasts were conservative, likely because of the unique aspects of the forecasting methodology, proper selection of uncertainty ranges, and the relatively high density of input data for model construction; wide variations in production metrics were forecast, indicative of a highly heterogeneous reservoir; results indicated that the validated model adequately captured the global or statistical pilot heterogeneity, enabling proper capture of steamflood flow/drainage mechanisms; and despite this heterogeneity, forecast oil-recovery levels were comparable with those observed in steamfloods in sandstone reservoirs.

Characterization of a Complex Carbonate Heavy-Oil Reservoir

This article, written by Assistant Technology Editor Karen Bybee, contains highlights of paper SPE 120423, ’Characterization of Complex Carbonate Heavy Oil Reservoir - A Case Study,’ by Afzal Iqbal,John Smith, Ali Reza Zahedi, Deemer Arthur, and Falah M. Al-Yami, Saudi Arabian Chevron; W. Scott Meddaugh, Chevron Energy Technology Company; Mansoor A. Rampurawala, SPE, Bingjian Li, SPE, and Ihsan Gok, SPE, Schlumberger Oilfield Services; and Talal Al Enazi, Kuwait Gulf Oil Company, originally prepared for the 2009 SPE Middle East Oil and Gas Show and Conference, Bahrain, 15-18 March. The First Eocene reservoir is the shallowest producing interval of the Wafra field in the partitioned Neutral Zone between Saudi Arabia and Kuwait. Characterization of this heavy-oil reservoir is challenging because of observed variations in oil viscosity, heterogeneity related to complex mineralogy, a possible dual-porosity system, and the presence of fractures of varing scales. The full-length paper presents several aspects of an integrated approach to characterize the reservoir. Introduction Currently, a large-scale steamflood pilot (LSP) consisting of sixteen 2.5-acre inverted-5-spot patterns is being installed. This will be the first multipattern steamflood of a carbonate reservoir in the Middle East. Oil was discovered in the First Eocene in 1954, and full-scale development began in 1956. The First Eocene has produced only 4% of original oil in place (OOIP), and primary recovery is not expected to exceed 10%. The oil gravity in this heavy-oil reservoir varies from 14 to 20ºAPI. The First Eocene is a depletion-drive reservoir with partial solution-gas drive and limited aquifer support. The estimated gross OOIP within the currently delineated field limit is more than 10 billion reservoir bbl. The reservoir mainly consists of dolomitized packstones and grainstones deposited under arid-to-semiarid conditions on a shallow, very gently dipping, low- to moderate-energy inner shelf or ramp setting in a gently dipping, restricted ramp environment. The presence of minor interbedded evaporites suggests restriction was occasionally sufficient for the development of hypersaline lagoons and sabkhas. The shallowing-upward cycles are capped by mud-dominated rocks, hardgrounds, and exposure surfaces. The average porosity is 35% and the average permeability is 250 md over the gross interval. Well-log and core-plug porosity values greater than 50% are common, and measured permeability values range up to 5,000 md. The First Eocene reservoir has an average depth of approximately 1,000 ft and a gross thickness of approximately 750 ft. Three wells were used in the study to demonstrate the approach of integrating log and core data to evaluate reservoir heterogeneity.

Enhanced Oil Recovery by Steamflooding, Cruse 'E' Field, Trinidad

This article, written by Technology Editor Dennis Denney, contains highlights of paper SPE 89411, "Enhanced Oil Recovery by Steamflooding in a Recent Steamflood Project, Cruse 'E' Field, Trinidad," by V. Ramlal, SPE, Petrotrin, prepared for the 2004 SPE/DOE Symposium on Improved Oil Recovery, Tulsa, 17-21 April. On the basis of the success of the Cruse ‘E’ Pilot steamflood in Trinidad, Petrotrin decided to expand the thermal project. The project was commissioned in January 1996 when steam injection began. In November 1998, environmental concerns in a residential area near the steamflood required project shutdown. This paper discusses the performance of the steamflood during steam injection and during the shutdown period, the operational and economic aspects of the project, and the environmental upgrade conducted to restart the project. Introduction The Cruse ‘E’ steamflood is in the Parrylands area of southwestern Trinidad, 8 miles from the town of Point Fortin. It is one of several active thermal projects operated by Petrotrin and is the most recent steamflood developed by the company. The Cruse ‘E’ Expansion steamflood was undertaken following the success of the Cruse ‘E’ Pilot Project. With loan financing from the Inter-American Development Bank (IADB), 28 new patterns were formed by drilling 60 wells and incorporating 35 existing wells in the thermal area. Steam-generation and -distribution equipment and production facilities were installed during 1995, and steam injection began in January 1996. The shallow Cruse ‘E’ sands were ideal for steamflooding, and response was obtained in only 9 months. After 22 months, production reached 1,200 BOPD. For environmental reasons, the flood was shut down in November 1998, and approval was not obtained to restart until March 2002. Factors that contribute to the success of this project include large reserves of shallow heavy oil (average gravity of 17°API); reliable, adequate supplies of fresh water for injection and natural gas for steam generation; good reservoir continuity within the project limits; 30 years’ accumulated company expertise in enhanced oil recovery; and the success of the pilot project.

Predicting Natural or Induced Fracture Azimuths From Shear-Wave Anisotropy

Summary Prediction of fracture orientation to determine depletion geometry is a key step in reservoir characterization and development optimization. Wireline crossed-dipole anisotropy logs measure shear-wave velocity anisotropy and provide a relatively inexpensive alternative to tiltmeter surveys and nine-component vertical seismic profiles (9C VSP's) for such predictions. A multiwell study in Cymric oil field, California, was conducted to map the variations in the predicted azimuth of induced fractures in a shallow Miocene siliceous shale reservoir (>500 millions of barrels of oil in place). Results from anisotropy data showed that the azimuths predicted by anisotropy logs and VSP's agreed. The agreement is significant because these two technologies measure very different kinds of seismic waves and have scales of investigation differing by two orders of magnitude. Within the productive zone, the predicted fracture azimuths from anisotropy data differed by about 60° from the regional stress in the NW portion of the Cymric development, while they were approximately aligned with the regional stress in the SE portion. Marked changes in azimuth also occurred as a function of depth across a regional Plio-Miocene (P-M) unconformity. Independent evidence, including observed steam breakthrough, natural fracture azimuths, and curvature analysis, supports the azimuths predicted from anisotropy data. Time-lapse temperature data from observation wells in a pilot steamflood showed anisotropy in horizontal permeability. Microresistivity borehole imaging logs allowed measurements of borehole breakouts and large, open natural fractures. The curvature analysis indicated a NW trending region of high curvature in the NW part of the Cymric structure. The large changes in azimuth with both depth and location in the development area show that regional stress maps are inadequate for planning oilfield development when stress patterns are complex, especially when deviated wells or induced fracturing are planned. Local stress configurations, therefore, must also be determined.

Integrated, Multidisciplinary Reservoir Characterization, Modeling and Engineering Leading to Enhanced Oil Recovery from the Midway-Sunset Field, California: ABSTRACT

The Pru Fee property is developed in a heavy oil, Class III (slope and basin clastic sand), reservoir of the Midway-Sunset field, San Joaquin Basin, California. Wells on the property were shut-in with an estimated 85% of the original oil remaining in place because the reservoir failed to respond to conventional cyclic steaming. Producibility problems are attributed to the close proximity of the property to the margin of the field. Specific problems include complex reservoir geometry, thinning pay, bottom water, and dipping beds. These problems are likely common at the margins of the Midway-Sunset and other Class III reservoirs. This project forms the first step in returning the property to production and explores strategies that might be applied elsewhere. Reservoir characterization, modeling, and engineering methods are integrated to design, simulate, and implement a pilot steam flood. A new drillhole provides good quality, core through the pay zone and a full suite of geophysical logs. Correlations between geological and petrophysical data are used to extrapolate reservoir conditions from older logs and yield a 3-dimensional petrophysical model. Numerical results illustrate how each producibility problem might influence production and provide a framework for designing the pilot steam flood. This first phase illustrates howmore » a multidisciplinary team can use established technologies in developing the detailed petrophysical, geological, and numerical models needed to enhance oil recovery from marginal areas of Class III reservoirs.« less

South Casper Creek Field: A Study in Reservoir Heterogeneity

The Tensleep Sandstone, an eolian and marine deposit, has produced low-gravity oil at South Casper Creek field (Natrona County, Wyoming) since the 1920s. Until recently, the reservoir was considered a relatively homogeneous sandstone body and was modeled as such for secondary recovery operations initiated during the late 1970s and early 1980s. Poor secondary recovery performance led to extensive reevaluation of the reservoir. Characterization studies by the field operator, Union Oil of California (UNOCAL), and through the Reservoir Characterization Project at the Colorado School of Mines produced an entirely revised picture of the Tensleep interval. Significant stratigraphic, diagenetic, and structural heterogeneities were identified, mapped, and correlated against productivity patterns under pilot steamflood programs. The results of this integrated geologic and geophysical effort are significant and have implications for secondary recovery operations elsewhere in the Tensleep and its regional correlatives.

スチーム攻法シミュレーション技術の実油田への適用例 （その２）

In this paper, application of simulation techniques for the evaluation of a steam flood pilot test is described. The pilot test has been conducted with a single inverted five spot pattern in a heavy oil reservoir.Thermal reservoir simulators generally take more calculation time than isothermal ones because both energy balance equations and material balance equations have to be solved; small time step sizers are required to establish numerical stability due to steep reduction in heavy oil viscosity with temperature increment. Thus, most of the application of the simulation to real field has been aimed at such a limited area as 1/8 symmetric element of a whole flood pattern.In the previous issue, the process and the results of simulation study using 1/8 symmetric model was presented. The results were examined and proved to be basically reasonable in comparison with actual reservoir performance observed by several monitoring methods.In this paper, a full pattern model including all pattern wells is constructed. History match for about two years of steamflood performance is done where some knowledge obtained in the previous issue is utilized. Furthermore, influence of fluid conductivity across a fault suggested in a geological study of the field is investigated with the model. At last, the importance of reservoir monitoring, for instance, bottomhole temperature survey, for the restriction of degree of freedom of the full pattern model, is discussed.

Steam-Soak Performance in South Oman

Summary With about 2 × 109 stock-tank m3 [12.6 × 109 STB] of medium/heavy oil originally in place (OOIP) in south Oman, considerable scope exists for increasing oil recovery by thermal methods. The viability of thermal recovery in south Oman was tested with a steamflood pilot in the Al Khlata sands of the Marmul field and a 2-year steam-soak project to test the applicability of steam soak in five south Oman oil fields producing heavy oil. This paper describes the performance of the latter project. The wells selected for the test program included a wide range of south Oman reservoir and oil characteristics—i.e., the main reservoir drive mechanisms of depletion, solution-gas, and edge- and bottomwater drive, the reservoir sandbody types, and oil viscosities from 80 to 4000 mPa·s [80 to 4, 000 cp]. Steam-soak operations were successful, and oil production accelerated significantly, with an average stimulated production rate twice that before stimulation. Acceleration was less marked in wells where reservoir energy is limited or where the primary (cold) water cut is more than 30%. At primary (cold) water cuts >50%, no increase in oil production rate was observed. The process was simulated numerically for several wells, with the results in close agreement with performance. Improved understanding of the process resulting from the simulation allowed the most important factors influencing performance to be identified and aided process optimization in the field test. Two small-scale steam-soak projects currently are being assessed for implementation in the early 1990's.

Sucker Rod Wear In Directionally Drilled Steamflood Wells

Abstract Husky Oil's Thermal Operations District in the Lloydminster, Saskatchewan area, has had extensive experience with sucker rod wear in directionally drilled steamflood wells. Husky's two major steamflood projects, the Aberfeldy Steamflood Pilot (steam drive) and the Pikes Peak Pilot (cyclic with some post-cyclic stream drive), have drilled and produced directional wells from pads since 1983. As expected, the frequency of wear-related sucker rod failures in directionally drilled wells is much higher than in conventional vertical wells. Through careful data gathering and analyses, Husky has attempted to quantify the degree of incremental wear as well as identify the controlling mechanism(s) (i.e. maximum dog-leg, maximum hole angle, etc.). Rod wear-related problems are described in depth. The frequency and location in the rod string of the wear-related failures are compared with a number of parameters to determine the major factors contributing to rod wear. The various methods and items tested by Husky to reduce and control rod wear are described and discussed. A summary of conclusions based on the data as well as a suggested rod/tubing maintenance program are included in this paper. Introduction Husky Oil Operations Ltd. has been active in the Lloydminster, Saskatchewan / Alberta, heavy oil fields since 1946 and currently operates over 2000 producing wells and five active enhanced oil recovery (EOR) pilots. The Lloydminster operators produce from a series of lower Cretaceous sands at depths between 400 m and 600 m. The sands are reasonably clean, moderately sorted, and poorly consolidated. Porosity can vary between 25% and 35%. Permeability ranges from 0.5 darcies to 8 darcies. The oil produced in the area is dense and viscous. Density is generally between 911 kg/m3 and 980 kg/m3. Viscosity can be as high as 131 Pa •s. The original-oil-in-place in the Lloydminster area has been estimated at 500 million m3(1). Probably less than 10% is recoverable by conventional means, EOR recoveries have been estimated at 25%. Pad Well Completions Husky began drilling EOR wells from pads in 1983. The advantages of pad drilling are well documented and accepted by the operating companies in the area. The close spacing used in the EOR pilots (as low as 1 ha), the reduced cost of fluid injection and production lines, the centralization of facilities, and the large reduction in surface area requirements make pad drilling of EOR wells especially attractive. Husky currently has over 100 pad wells in its EOR pilots. Pad sizes vary between 3 and 14 wells. The wells are typically equipped with 178 mm casing, 89 mm EUE tubing, 22 mm grade C sucker rods, and 64 mm rod (insert) pumps. When the first pad wells were brought on stream in 1983 (at the Pikes Peak Cyclic Steam Pilot), the field personnel were expecting to have significant rod wear problems. The frequency of rod wear-related servicing was higher for the new wells but not alarmingly so. The installation of rod rotators and the implementation of a rod maintenance program kept the problem under control.

A Kinetic Model of CO2 Generation and Mineral and Isotopic Alteration During Steamflooding

Summary CO2 is commonly produced during steamflooding. In higher-temperature steamfloods the amount of CO2 produced is such that the only viable source is carbonate minerals. We show by reference to CO2 concentrations in natural geothermal systems that in most common rocks and sediments, CO2 is generated naturally under hydrothermal (~572°F) conditions by the dissolution of carbonate minerals. CO2 solubility is a strong positive function of temperature and, at 572°F, is more than sufficient to account for high CO2 concentrations like those encountered at Chevron's Buena Vista (BV) Hills steamflood pilot test. A linear kinetic model is developed to describe chemical equilibration and CO2 generation in the hot parts of the reservoir. The model also describes CO2 removal through carbonate precipitation in the cooler parts of the rock formation where the fluid moves ahead of the thermal front. Changes in carbon isotope concentration are included in the model. The small difference between the 13C in the produced gas and the source carbonates at BV Hills suggests that little CO2 is precipitated as carbonate within that reservoir, a prediction that can be tested by poststeamflood coring.

スチーム攻法シミュレーション技術の実油田への適用例 その１

In this paper, application of simulation techniques for the evaluation of a steam flood pilot test is described. The pilot test has been conducted with a single inverted five spot pattern in a heavy oil reservoir.Thermal reservoir simulators generally take more calculation time than isothermal ones because both energy balance equations and material balance equations have to be solved; small time step sizers are required to establish numerical stability due to steep reduction in heavy oil viscosity with temperature increment. Thus, most of the application of the simulation to real field has been aimed at such a limited area as 1/8 symmetric element of a whole flood pattern.In the previous issue, the process and the results of simulation study using 1/8 symmetric model was presented. The results were examined and proved to be basically reasonable in comparison with actual reservoir performance observed by several monitoring methods.In this paper, a full pattern model including all pattern wells is constructed. History match for about two years of steamflood performance is done where some knowledge obtained in the previous issue is utilized. Furthermore, influence of fluid conductivity across a fault suggested in a geological study of the field is investigated with the model. At last, the importance of reservoir monitoring, for instance, bottomhole temperature survey, for the restriction of degree of freedom of the full pattern model, is discussed.

Steamflood Material-Balance Applications

Summary A simple steamflood material balance derived from a two-zone reservoir model is presented. Alternative forms of the material balance include the effects of an initial hydrocarbon gas saturation, PV compressibility, and a hot-water zone. In example applications, the material balance is used to calculate the amount of fluid migration from a 16-pattern steamflood pilot and the steam injection rate within a regular five-spot pattern.

Rock Creek Oil Field CO2, Pilot Tests, Roane County, West Virginia

This paper reports on the design and operation of two CO/sub 2/ EOR tests conducted in the Rock Creek field in Roane County, WV. The history, fluid properties, and geology of the Rock Creek field are presented first. The test area is then addressed more specifically with an evaluation of the cores and the geophysical logs of the injection, production, and observation wells. Finally, the injection history and the production response are documented. The first test was conducted in two 10-acre (4-ha), normal five-spot patterns, with 13,000 scf CO/sub 2//STB oil (2315 st m/sup 3//stock-tank m/sup 3/) injected. This test effort recovered 13,078 STB (2079 stock-tank m/sup 3/) of oil (3% of the original oil in place (OOIP)) but was terminated after 3 years before all oil capable of being mobilized was recovered. About 15% of an HCPV was injected. The first test was followed by a second, smaller test that, given the same amount of CO/sub 2/ to be purchased, would result in an increase in HCPV's of CO/sub 2/ injected and a greater potential for oil recovery. The second test was conducted in a 1.55-acre (0.63-ha), normal four-spot pattern contained within the original test pattern. This test lastedmore » 2 years, with 9,000 scf CO/sub 2//STB oil (1603 std m/sup 3//stock-tank m/sup 3/) injected. Recovery from this test was 3,821 STB (607 stock-tank m/sup 3/) of oil (11% of the OOIP). About 48% of an HCPV was injected. It appears that CO/sub 2/ miscible flooding is technically successful in Appalachian reservoirs.« less

Hot-Water and Steamflood Laboratory Experiments Under Reservoir Conditions

Summary Hot waterfloods at four temperatures (between 30 and 240°C [86 and 464°F]) and a steamflood were performed on the same porous medium (compacted silt) under reservoir conditions. Dynamic isothermal displacements were interpreted with a thermal simulator, taking into account capillary end effects. The corresponding oil/water relative permeability curves were obtained by matching observed pressure drop and oil production. Results show that temperature influences the endpoint saturations but not the shape of the curves. The steamflood experiment was carried out with an adiabatic core holder. The oil recovery was 70% higher than the recovery by the hot waterflood at the same temperature (240°C [464°F]). Oil stripping and production of a large amount of CO2 caused by dissolution of carbonates were pointed out. In the numerical interpretation of this experiment, we used oil/water relative permeabilities to provide the three-phase oil relative permeability. These experiments allow a better understanding of steam displacement mechanisms and supply essential data for numerical interpretation of a steamflood pilot.

Mineral reactions in quartzose rocks during thermal recovery of heavy oil, Lloydminster, Saskatchewan, Canada

X-ray diffraction analysis, thin section analysis and scanning electron microscopy were used to compare the mineralogy of pre- and post-recovery cores from the Lower Cretaceous Sparky interval in Aberfeldy steamflood and fireflood pilot projects. Lloydminster area, Saskatchewan, Canada. The objectives were to document the potential changes in mineralogy that may occur during recovery of heavy oil and to evaluate the risk of formation damage from mineral reactions. These unconsolidated sediments are very quartzose (90%) and they also contain feldspars, clays (mainly kaolinite) and carbonate cements (siderite and ankerite). The formation of new minerals in the burned zone and the plugging of pores by coke in the zones below the burned zones is likely to cause small reductions in porosity and permeability. Illite formation is observed in the steamflood pilot but is not extensive enough to cause formation damage. Carbonate scale is observed at the producing wells. During steamflooding illite and chlorite are formed; the smectite which is observed is due to mud invasion. During fireflooding, illite forms in the coked zones, whereas K-feldspar and hematite form in the burned zones. Thermodynamic calculations of the stability of mineral-forming reactions are in agreement with physical conditions during steaming. The decomposition of kaolinite in the burned zones indicates combustion temperatures greater than 510–540°C. It may be concluded that in a period of a few years only minor detrimental effects related to mineralogy occurred during thermal recovery in the Aberfeldy pilot projects.