The electromagnetic heating of oil wells and reservoirs refers to thermal processes for the improved production of oil from underground reservoirs. The source of the heat, generated either in the wells or in the volume of the reservoir, is the electrical energy supplied from the surface. This energy is then transmitted to the reservoir either by cables or through metal structures that reach the reservoir. The main effect, because of the electrical heating systems used in practice in enhanced oil recovery, has been the reduction of the viscosity of heavy and extra heavy crudes and bitumens, with the corresponding increase in production. Focus is centered on systems (and the models that describe their effects) that have been used for the electromagnetic heating in the production of extra heavy petroleum and bitumen.
In this paper, we divide the steps to explore, identify, and monitor a SAGD site into six stages and show how electromagnetic methods can be used at each stage. Three-dimensional inversion of airborne EM data provide large-scale, regional geologic trends and delineate paleo-channels and the caprock thickness at a newly-developed property in the Athabasca oil sands. We use semi-synthetic models from resistivity logging and the airborne data in conjunction with ground-based and borehole EM surveys to characterize the oil-rich McMurray Formation and monitor steam chamber growth over time. Periodic EM data collection and three-dimensional time-lapse inversion allow for high-resolution interpretations throughout the SAGD process.
Presentation Date: Wednesday, October 19, 2016
Start Time: 4:00:00 PM
Presentation Type: ORAL
Park, Changhyup (Kangwon National University) | Choi, Jiyeon (Kangwon National University) | Lee, Changsoo (Kangwon National University) | Ahn, Taewoong (Korea Institute of Geoscience and Mineral Resources) | Jang, Ilsik (Chosun University)
This paper determined the optimum operations of steam assisted gravity drainage covering steam interference between vapor chambers in a heterogeneous oil-sands deposit. The objective value was the minimum cumulative steam to oil ratio that represented energy efficiency. Three wellpairs, i.e. six horizontal wells, were installed to take steam interference between the chambers into consideration. The optimum operations showed the small difference of bottom hole pressure between an injector and a producer that released a small amount of steam into the reservoir. The lower injection could decrease the steam interference and lateral fluid movement by generating a similar size of chamber. A sensitivity analysis showed the key factors to cumulative steam to oil ratio were bottom hole pressure of the injectors and surface steam rate at the producers. To satisfy the limit of surface steam rate at the producer, the bottom hole pressure at the producer increased sharply and maintain the small difference between the producer and the injector.
A management decision-making and planning tool has been developed to provide a quick, high-level resource quality assessment of oil sands assets to enable economics-based ranking of investment opportunities. It is systematic and transparent, largely avoiding the subjectivity and human bias often associated with ranking assets for capital allocation.
The model utilizes the available reservoir characterization information (API gravity and petrophysical analysis including oil saturation, effective porosity, V shale, pay thickness), expected operating conditions (steam injection pressure, horizontal well lengths), and a reservoir risk assessment to predict the key performance metrics for an in situ oil sands project using SAGD (steam assisted gravity drainage) including oil rates, steam-oil ratio and recovery factors. The reservoir risk factor is a quantification of the production impact (lower expectations and/or increased uncertainty) from reservoir impairments based on expert opinions and reservoir simulation. These performance metrics can then be used to estimate expected overall economic potential (IRR) for a given asset. The ranking can be done at various levels: land sections, defined prospects, wellpad drainage areas, or at the individual (delineation) well level.
Predictive analytics techniques, in this case multi-variable linear regression, were used to construct the model. It was initially based on thermal recovery theoretical models for the SAGD process for predicting oil rates and a simple energy balance for predicting steam-oil ratio (SOR). Subsequently it has been updated via industry production data "fitting", or applying the actual performance data of various mature, operating wellpads to improve the confidence level of the model. The result is a hybrid model; science-based but influenced by real operating and production experience.
It has served as a primary tool used for strategic planning, in the setting of high-level performance targets (and probabilistic distributions thereof) for each of the assets. This tool has enabled a resource driven development strategy, allowing the company to focus technical resources on the assets that possess the greatest economic potential. Resulting business decisions include capital allocation (for additional delineation data) and more rigorous technical efforts (reservoir modeling and simulation) on the highest ranking prospects.
Data driven well models were constructed for a steam injected field producing 800 m3/d of API 8 bitumen from 100+ beam pumped wells. The modelling technique is known as Production Universe(PU) which gives real time estimates of oil and water flows for each of the wells during steady state and transient operations. PU tracks rates when wells are starting-up, closing-in and for transient flow oscillations, resulting in automatic daily reports of total production and deferment (low and off) for each well.
The deferments reports drive an exception-based surveillance (EBS) process by flagging the engineers of the largest production gain opportunities. EBS analysis and associated well remedial actions have resulted in an 8% production gain.
Continuous estimation of well bitumen and water flows has allowed real time optimization of the entire field. This is achieved by continuously maximizing production within the plant water handling and economic constraints. The system knows how much oil and water the wells are producing and recommends rates for each of the wells to maximize production within the water handling capability.
The EBS process has been successfully running for more than a year. The RTO process has just been introduced, hence it is too early to estimate the incremental gain.
This paper will detail the surveillance and optimization techniques, learnings and associated work processes.
Background and Field Description
The field is located in Northern Alberta, Canada and produces 1,200 m3/day of API 8 bitumen with an in-situ viscosity of 100,000 cp at reservoir pressure. The production facility consists of 8 production pads, each with 8 -16 wells.
Steam injection is used to reduce the bitumen viscosity such that it can flow to the well bores. Bitumen and water have similar densities at operational temperatures. The wells operate in a cyclic steam stimulation mode and steam drive mode. In CSS steam is injected into each well at a given rate for a given time, after which natural production commences, without artificial lift with temperatures in excess of 200 degrees C. Once the pressure drops below a certain level, rod pump artificial lift commences and continues until the temperature falls to a level that does not support adequate flow rates and steam injection again commences and the cycle repeats. In steam drive steam is continuously injected via dedicated injectors to mobilize bitumen and drive it to the adjacent producers.
During the CSS production phase and in steam drive, bitumen is rod-pumped (or flows naturally) as an emulsion to the treatment plant. Treatment consists of degassing, separating into bitumen/ water phases and spiking with diluent to allow effective pipeline transportation to the refinery. The produced gas is compressed and injected into the formation for future usage. Produced water is disposed off via water disposal wells.
During the production phase well tests are performed frequently.
Graphical depictions of the field location and the overall process flow are shown in the following diagrams.
Connacher's first oil sands project, the Pod One facility at Great Divide, has been operational since 2007. The successful SAGD project has produced approximately 7 million barrels of bitumen. During the past three and a half years, the impacts of certain predicted reservoir challenges and opportunities have become apparent.
While the quality of the oil sands in this first phase of Pod One is generally good, Pad 101 South in particular has geological zones that affect SAGD operation. This includes a bitumen lean zone, and a gas cap overlying the main bitumen channel/s. Early field results matched with detailed simulations have shown positive results in maximizing well pair production. For the purposes of this paper a lean bitumen zone differs from an aquifer in two ways. The lean zone is not charged, and is limited in size. The operation is also complicated by the fact the gas bearing zone has been depleted through earlier production.
Connacher's operating practice at Great Divide attempts to achieve a pressure balance between the 3 zones (rich oil sands, lean zone, gas cap) to reduce steam loss and maximize production rates. Reducing the pressure encourages steam chamber development growth horizontally and ensures that steam contacts the highly saturated bitumen areas. How this is achieved with the highest positive impact on well productivity is illustrated with operational data and analysis including the results of simulations that recommended the optimum operating strategies.
Induced gamma ray spectroscopy (IGRS) logs from two cyclic-steam-stimulation observation wells in Cold Lake were analyzed to determine the vertical resolution and repeatability of data derived from gamma rays of inelastic and capture neutron reactions. Time-series analysis, a technique that uses the Fourier representation of the log data, was used to quantify the vertical resolution and the signal/noise characteristics of various IGRS log curves and .to compute the coherence vs. spatial frequency of data collected on multiple IGRS passes. The coherence function ranges from 1.0 for perfect repeatability to 0.0 for incoherent noise. Because no real variations in the measured data were expected for the 12- to 24-hour data-collection period, any deviation of the coherence function from 1.0 is attributed to incoherent noise. Generally, coherence is high at low frequencies (large vertical scales) and low at high frequencies (small vertical scales). By noting the frequency at which the coherence level decreases to the expected value of random noise, we can quantify the vertical resolution of a log curve.
Data analysis from these wells indicates that both the vertical resolution and repeatability of individual capture and inelastic curves differ. We found that the H-yield and capture curves have the highest vertical resolution (˜0.3 m) and the best signal/noise ratios (FSN˜ 30:1). In contrast, the capture Ca and Si yields are of significantly lower quality (FSN˜ 2 : 1). Only a small difference exists between the vertical resolution of the inelastic C (1.0 m) and 0 (1.3 m) yields, but the FSN of the 0 yield is only one-half that of the C yield. Fortunately, the vertical resolution and the repeatability of the C/O ratio, FCO are determined primarily by the quality of the inelastic C data. Specific to our application of monitoring heavy-oil saturations during cyclic-steam stimulation in a relatively uniform, high-porosity unconsolidated sand, we developed a simplified FCO interpretation that does not rely on the elemental yields from capture data. This linear model permits us to assess the uncertainties in computed saturations from the statistical noise in the inelastic FCO data.
IGRS logs are used routinely in Cold Lake to monitor bitumen saturations on observation wells and occasionally to evaluate new wells. Esso Resources Canada Ltd. uses cyclic-steam stimulation to recover bitumen from the Cretaceous unconsolidated Clearwater formation. Throughout the commercial operations, bitumen is produced from directional wells drilled from a central pad location.1 These wells generally are logged before being cased; however, sometimes a key well cannot be logged before casing, so IGRS logs may be run to quantify bitumen saturations and to provide data for reservoir description. Reservoir descriptions are derived primarily from capture gamma ray data (i.e., Si, Ca, Fe, H, S, and Cl yields and the capture cross section, S), and bitumen saturations are obtained from the inelastic FCO data.
Besides large commercial operations, Esso also operates several pilots1-4 to test various production scenarios and to study the physics of cyclic-steam stimulation.5 Observation wells playa key role in the pilots, and much has been learned from temperature observations.4 To monitor changes in bitumen saturations at the observation wells, we rely primarily on the inelastic FCO data from IGRS logs. Two observation wells located 15 m apart on one pad have been logged 10 times with IGRS logs during various phases of cyclic-steam stimulation. An 8D-m interval in the Clearwater formation and a 30-m interval in the shallower Grand Rapids formation are logged routinely to monitor tool performance and stability. A specially made 3-m casing joint that carries 22.5 kg of graphite was installed on one of the wells below the Clearwater sands to provide an in-situ, stable, high-carbon FCO calibration facility.
Calibration and the statistical precision of the IGRS data must be understood. Porosities are very high (f˜33%) in the unconsolidated Clearwater sands, making it feasible to collect continuous IGRS data by logging slowly. Generally, we make four inelastic passes at 0.5 m/min and four capture passes at 2 m/min. These data are averaged to obtain a single continuous FCO curve. To decrease the statistical noise inherent in the FCO data further, we average the data from multiple vertical levels. We use time-series analysis7 to quantify the statistical noise in the inelastic and capture data and to determine the optimal vertical averaging.
Fig. 1 demonstrates the statistical nature of the IGRS data with eight FCO passes across several heavy-oil sands of varying thicknesses in the Grand Rapids. Pass-to-pass variability is large, but the median-filtered data compare well with the openhole shallow resistivity data (Fig. 2). Because the Grand Rapids sands are not being produced in Cold Lake, data from two different runs (eight passes and five vertical levels, separated by 15.24 cm) were pooled and sorted for median filtering. The median value chosen from these data was used to represent the FCO value. The correspondence of the median-filtered FCO and the shallow resistivity data suggests that the vertical resolution of the averaged FCO data is 1 or 2 m.
The comparison of the averaged FCO with openhole shallow resistivity data also illustrates that the shallow depth of investigation of the IGRS measurements8 is not necessarily a problem at Cold Lake. Cold Lake bitumen is extremely viscous (100 Pa·s) and the oil sands do not invade. Furthermore, both wells were drilled with a 311-mm [12.25-in.] bit and cased with 178-mm [7-in.] casing, which results in a 50-mm-thick cement sheath and reduces the formation volume that the tool can investigate.
Fig. 3 shows four passes of inelastic and capture IGRS data from one run. The top (429 m) and base (480 m) of the Clearwater formation are evident on the C, O, FCO, S, and Fe- and Si-yield data. The S data exhibit good pass-to-pass repeatability; the S and Cl yields appear to consist mainly of statistical noise. An Fe-rich carbonate-cemented tight streak: (siderite) at 467 m is evident in the Ca, H, Si, and Fe yields. A radioactive tag is at 447 m, and the graphite calibrator is between 480 and 483 m. Fig. 4 shows the calibrator and data from eight repeat logs run across it.
Section I-Paper 36 INVESTIGATIONS INTO DIRECT OIL DETECTION METHODS BY V. A. SOKOLOV," F. A. ALEXEYEV," E. A. BARS," A. A. GEODEKYAN," G. A. MOGILEVSKY," Y. M. YUROVSKY" AND B. P. YASENEV" ABSTRACT. The report gives the results of comprehensive researches carried out during recent years in the U.S.S.R. on the development and application of geochemical, biochemical and radiometric methods of prospecting and exploring oil and gas fields. The direct geochemical indications of oil and gas considered are: natural oil gases, bitumens, bacteria-assimilating migrating hydrocarbons and other indications due to the effect of migrating gases on the surroundings, as well as organic substances of crude oil origin dissolved in underground waters. Extensive research on gaseous, bacterial and other geochemical anomalies under various geological conditions (geosynclinal areas, platforms and transitional zones) revealed the relationships of distri- bution of geochemical anomalies at various stratigraphie levels of sections and their relation to the deep sources of migration. The data on the forms of gaseous and bacterial anomalies and on the conditions of their formation are generalized. Results are given of practical work involving the use of gaseous and microbiological surveying, gas logging, and other methods of exploration. Analysis of the effectiveness of direct oil and gas detecting methods shows that under favourable geological and geochemical Conditions the proportion of correct predictions is as high as 70 per cent. For more extensive practical application of direct methods in oil and gas prospecting it is recommended: 1) to select objects of investigation more carefully; 2) to increase the depth of sampling in platform districts ; 3) to make extensive use of structural exploratory and seismic wells for gasometric surveying, and 4) in new regions to carry out in the reconnoitering stage primariIy regional gas microbiological and other investigations on soils and underground waters of the upper sedimentary layer. Gas logging techniques involving the use of chromatographic analysis are described. Direct methods deserve wider applications in prospecting for oil and gas deposits, especially in new regions; also further research is needed to improve these methods.
. Cette communication présente succinctement les résultats de recherches complexes pour- suivies ces dernières années en U.R.S.S. pour la mise au point et l'emploi de méthodes géochimiques, biochimiques et radiométriques de l'existence de gisements de pétrole et de gaz. Comme indice géochimiques directs de présence de pétrole et de gaz, on examine les gaz naturels de pétrole, les bitumes, les bactéries tests d'hydrocarbures en migration et autres indices conditionnés par l'action de gaz en migration sur le milieu environnant, aussi des matières organiques dissolues dans les eaux souterraines ayant le pétrole pour origine. Sur la base de nombreux tr