TX 75083-3836, U.S.A., fax 01-972-952-9435. AbstractWaterflood automation at a field scale is a complex coupled problem. In our analysis, data from injecting and producing wells, as well as satellite differential interferograms (InSAR) are used as the inputs. Some information processing is carried out on-line automatically, and other needs personnel expertise. Dynamic adaptive control is performed in the mixed openloop-feedback mode. Our surveillance-control system is being implemented in Sections 32 and 33 of the Lost Hills diatomite field, CA, USA.

Hall's method is a tool for evaluation of an injecting well performance.It is based on the assumption of radial steady-state flow.Besides time series of historical injection pressures and rates, rigorous implementation of Hall's method requires information about the ambient reservoir pressure.In addition, it is assumed that the influence domain radius is constant during the observation period.Neither of these parameters can be measured directly. This paper discusses a new method called slope analysis.It is based on analysis of the variations of the slope of the plot of the time integral of pressures versus cumulative injection volume.In particular, it produces an estimate of an apparent average reservoir pressure.This method requires only injection pressures and rates data, which are routinely collected in the course of a waterflood.Thus, slope analysis requires no interruption of regular field operations. The method has been verified with numerically generated pressure and rate data, and tested with field data.In both cases it proved to be accurate, efficient, and simple.The estimated ambient reservoir pressure can be used to correct the Hall plot analysis or to map the average reservoir pressure over several patterns or an entire waterflood project.Such maps can then be used to develop an efficient waterflood policy, which will help to arrest subsidence andimprove oil recovery. Introduction Monitoring and control of performance of each individual well is an important component of successful oil recovery operations.The dramatic progress in information technology over the past decade has made it possible to collect and store huge volumes of high-quality production and injection data.These data, if appropriately interpreted, provide new insights into reservoir dynamics across multiple temporal and spatial scales.Therefore, efficient processing and interpretation of the high-frequency field measurements is a task of crucial importance to modern management of oil & gas recovery projects. This paper deals with problems related to monitoring and control of waterflood operations.The necessity of collecting and processing numerous measurements was understood decades ago[1].Recently, ChevronTexaco with participation of the Lawrence Berkeley National Laboratory and the University of California, Berkeley, developed a concept of field-wide surveillance and control of waterflood.This concept is being implemented in the Lost Hills oil field[2]. The work reported here is a part of this effort. Waterflood performance in an entire oilfield sums up from operations at each individual well.The global project objectives are derived from a field-scale analysis, such as inspection of satellite images for surface subsidence and uplift, and calculation of the fluid injection-withdrawal balance.[3]But the subsurface reservoir can only be accessed and controlled through the wells.Therefore, it is critically important to have efficient tools for regular well performance monitoring, and methods for adequate interpretation of this performance for assessment of the reservoir conditions near the wellbore. Traditional transient well tests have been used to evaluate the average near-wellbore formation transmissivity[4].Such tests interrupt regular field operations.Their interpretation is based on analysis of transient effects taking place at time scales which are short relative to those of fluid injection and production.On one hand, unless there is a pipe-like circulation of injected fluid between an injector and the surrounding producers, the injection results in ever-changing reservoir conditions.On the other hand, these changes may be almost imperceptible over a typical observation time interval.Therefore, the field-scale reservoir processes can be called quasi steady-state.In real life operations, short-time fluctuations of the injection pressures and rates at the well are inevitable.Separation of these short-time transient effects and long-time quasi steady-state processes is one of the most important tasks of well performance monitoring and diagnostics,[5–7] crucial to the information-driven oilfields of the future.

This paper will discuss how Chevron U.S.A. Inc., a ChevronTexaco Company (ChevronTexaco), is implementing a field trial that will use research developed software integrated with Supervisory Control and Data Acquisition (SCADA) on injection wells, in conjunction with satellite images to measure ground elevation changes, to perform real-time reservoir management in the Lost Hills Field. Implementation of a new software control model to restrict hydrofracture growth in water injection wells is being field tested.1 Synthetic Aperture Radar Interferograms (InSAR) are being obtained on an approximately 60-day interval to determine subsidence rates and will be used as an input variable for pattern voidage calculations. Incorporating new and innovative technologies is helping ChevronTexaco produce a very unique and challenging diatomite reservoir. Introduction The Lost Hills Field was discovered in 1910 and is located approximately 45 miles northwest of Bakersfield, California (Figure 1). ChevronTexaco implemented a waterflood in the Lost Hills field in 1992 to increase recovery and mitigate subsidence.2 Waterflood project oil production from the Belridge diatomite reservoir is approximately 10,000 BOPD and has been relatively flat for the last seven years (Figure 2). The majority of the 520 acre waterflood area is developed on 2 1/2 acre staggered line drive patterns. Injection has been confined to the intervals from the F point to the K point (Figure 3).

Summary The reservoir engineering aspects of the design of a major west Texas CO2 flood are presented. The design included a detailed fieldwide geologic study, a CO2 injectivity test, laboratory work, and reservoir simulation. CO2 floodingis predicted to recover an additional 8% of the original oil in place(OOIP). Introduction The North Ward Estes (NWE) field, located in Ward and Winkler Counties, TX(Fig. 1), was discovered in 1929. Cumulative oil produced is more than 320million bbl (25% OOIP). The field has produced is more than 320 million bbl(25% OOIP). The field has been waterflooded since 1955. Geologically, the NWEfield resides on the western flank of the Central Basin Platform. The Yates, the dominant producing formation, includes up to seven major reservoirs and iscomposed of very-fine-grained sandstones to siltstones separated by densedolomite beds. Within the 3,840-acre Sect area, average depth is 2,600 ft. Porosity and permeability average 16% PV and 37 md, respectively. Reservoirtemperature is 83 degrees F. The flood patterns are 20-acre five-spots andlinedrives. patterns are 20-acre five-spots and linedrives. CO2 flooding wasimplemented in early 1989 in a 6-section project area located in the betterpart of the field in terms of project area located in the better part of thefield in terms of cumulative oil production and reservoir-rock quality. Thispaper reviews the field history and reservoir geology and focuses on thereservoir engineering aspects of the design of this CO2 flood-laboratory work, CO2 injectivity test, and CO2 flood performance predictions. performancepredictions. Field History and Development The NWE field was discovered in 1929. Except for the most productive parts, which were drilled on 10-acre spacing, the field was initially parts, whichwere drilled on 10-acre spacing, the field was initially developed on 20-acrespacing. Until the early 1950's, a typical completion consisted of drilling tothe top of the Yates, drilling ahead and checking for gas caps, setting casingthrough the gas sands, drilling to total depth, shooting the producing sectionwith nitroglycerine, cleaning out the hole, and hanging a perforated liner fromthe casing. Practices changed in the early 1950's to eased-hole completions, hydraulic fracturing, and acidizing. About half the current injectors are shot, open-hole completions. Vertical sweep has been adversely affected because ofthe inability to measure and control the injection profiles. Fig. 2 shows theproduction and injection history of the project area. Primary production peakedin 1944 and was approaching the economic limit in the mid-1950's. A 960-acrepilot waterflood began in 1954. Oil production responded quickly, and the floodwas expanded to the rest of the project area during the next 2 years. Theprevailing flood patterns were 4O-acre five-spots. Oil production increasedsteadily after 1954, reached a peak in 1960, and then declined at 11 %/yr until1979, when it began to stabilize as a result of drilling infill and replacementwells, injection profile modifications by means of polymer treatments, andpattern profile modifications by means of polymer treatments, and patterntightening and realignment (Sections 3 and 6 through 8 were converted to20-acre five-spot patterns and Sections 9 and 10 to 20acre linedrive patterns). By the end of 1988, the 6 sections had produced 29% of the OOIP. Waterfloodingthe Yates has been very successful, as evidenced by the 2.3 ratio of ultimatesecondary to ultimate primary production from wells existing at the beginningof waterflooding. The production from wells existing at the beginning ofwaterflooding. The favorable mobility ratio in these reservoirs indicates goodarea sweep efficiency. Because of the high Dykstra-Parsons coefficient (0.85)and permeability contrast among the major sands, the vertical conformance hasbeen poor. Even after injection of 2.6 waterflood-movable PV, only 50% of theoil recoverable by waterflooding has been produced. Reservoir Geology and Properties A comprehensive geologic study and reservoir characterization was conductedto characterize the individual reservoirs of the Yates, which consist ofvery-fine-grained sandstones to siltstones separated by dense dolomite beds. Indescending order, these sands are Sands BC, D, E, Strays, J1 and J2 (Fig. 3). The general depositional environment was a tidal-flat-to-lagoona settingsituated to the east of and behind the shelf margin. The reservoirs weredeposited as sand and silt in the subtidal-to-beach environment and silt toclay in the supratidal environment. Depositional strike was parallel to theshelf margin, which is parallel to the present northwest/southeast sectionlines. Sand BC is a siltstone to fine-grained sandstone with detrital clay. Thedepositional environment was that of a shallow-water tidal flat with anabundant amount of windblown sediments. A zone of low porosity and permeabilitytrends northwest/southeast through the porosity and permeability trendsnorthwest/southeast through the middle of the project area. Most of Sand BC wasin the original gas cap. Sands D and E are similar to Sand BC, but theirporosities and permeabilities are more variable. The Strays sand is composed ofthin-bedded, lenticular, intertidal to subtidal siltstones and fine-grainedsandstones with the highest clay content of any Yates interval. Because ofthis, permeability and reservoir continuity suffer while porosity remains high. Sands J1 and J2 are composed of coarser sands with much less clay content andtherefore higher effective porosities and permeabilities. The depositionalenvironment was a beach to near-shore marine where turbulence winnowed finersilts and clays out of the strike-oriented sand deposits. Table 1 lists averagereservoir properties for the Yates. Laboratory Work Extensive laboratory work was conducted to support the evaluation of CO2 flooding in the NWE field.Black-oil PVT and oil/CO2 phase-behavior studiesof recombined separator oil and gas samples (Table 2) determined oil swelling, viscosity reduction, and phase transition pressures vs. mole percent CO2 (Fig.4). The PVT data show the typical complex phase behavior exhibited by CO2/light-crude-oil systems at low reservoir temperature.Slim-tubeexperiments determined minimum miscibility pressure (MMP). Fig. 5 shows theresults of the displacement of pressure (MMP). Fig. 5 shows the results of thedisplacement of reconstituted reservoir fluid by pure CO2 in a packed column atdifferent pressures. Additional displacement tests were conducted with fivedifferent CO2/hydrocarbon-gas mixtures. The MMP ranged from 1,010 to 1,350 psiavs. 937 psia for pure CO2. No significant changes in ultimate slim-tube oilrecovery were observed. These tests verified that a published correlationadequately estimates the MMP for NWE oil and impure CO2.CO2 flooding ofrestored-state composite cores determined the mobilization and recovery of thewaterflood residual oil saturation, S orw. The core assembly (Fig. 6) wasconstructed from 1-in.-diameter plugs drilled from NWE cores epoxied intoconfining stainless-steel sleeves. SPERE P. 11

This paper summarizes results to date of implementing i-field projects in selected assets in Chevron's San Joaquin Valley Business Unit (SJVBU) in California. The i-field projects include collaborative environments to transform operational processes at a basin-wide or asset level remote collaboration and visualization has been implemented to help execute reservoir management and major capital project targets reliably and efficiently in the field. Successful asset prototypes are standardized and replicated across the business unit. The results to date demonstrate business value and take-up of the technology and processes by oilfield operations. Introduction The major producing assets of Chevron's SJVBU are shown in red in Figure 1. SJVBU production in 2007 averaged over 220,000 barrels per day from approximately 15,000 producing wells. Approximately 83% of the production was heavy, 10% light, and 7% gas. The heavy oil is generally recovered through thermal operations, while the light oil is produced by waterflood. Unneland and Hauser (Reference 1) described the beginning of Chevron's digital oilfield program called "i-field". Ouimette and Oran (Reference 2) summarized the use of decision support software, integrated with improved instrumentation, workflow automation, and data architecture to enable more reliable and efficient field operation and execution of reservoir management targets at San Ardo within SJVBU. i-field has become critical to SJVBU's quest for operational transformation, in pursuit of a vision to:Operate from a centralized asset decision environment where the application of smart oilfield technologies result in industry leading margin performanceIntegrate reservoir management with operationsAutomate routine decisions by artificial intelligenceCreate a highly virtual organization where innovation and collaboration efficiently move ideas to applied technology