Steam-assisted Gravity Drainage Production Research Articles

Numerical simulation study of uniform steam injection method for SAGD horizontal wells

Steam assisted gravity drainage (SAGD) has a better effect on heavy oil development than other thermal recovery methods. However, during the SAGD production practice, the problem of uneven temperature distribution of long horizontal well sections has appeared. In response to this problem, the reservoir numerical simulation software CMG is used to simulate different steam injection methods for SAGD horizontal well sections. The results show that the conventional steam injection steam cavity only develops at the heel end, the development length is 1/3 of the length of the horizontal section, and the steam suction profile is uneven. The double-pipe steam injection steam cavity develops at the heel and toe ends, and the steam suction profile is uniform. Improved, multi-point steam injection steam chambers are developed along horizontal wells, and the steam suction profile is the most uniform.

Comparison of different machine learning algorithms for predicting the SAGD production performance

Steam Assisted Gravity Drainage (SAGD) is a typical thermal recovery process consisting of an upper horizontal injector and a lower horizontal producer, which heats and removes bitumen from oil sands deposits by steam. Serious research has been conducted on data-driven models to analyze the cumulative production performance of the SAGD process, and one of the most common machine learning methods utilized is the Artificial Neural Network (ANN). It is important to test other machine learning methods like Extreme Gradient Boosting (XGBoost) and Light Gradient Boosting Machine (LightGBM) in different conditions of data samples. In this paper, a series of SAGD models based on typical oil sands reservoir properties and operational conditions were constructed. Three different data groups were simulated by the Computer Modelling Group (CMG) thermal software STARS. Three different machine learning methods, including ANN, XGBoost, and LightGBM were constructed to calibrate a relationship between the input parameters and the output parameters in the different simulated data groups. A series of final models were tested and compared. The conclusion shows that data-driven training improves as the number and randomness of the data samples increase, and the LightGBM has the best prediction performance with such data samples. The ANN model may be the best choice with data samples of the worst randomness.

Development of a deep learning-based model for the entire production process of steam-assisted gravity drainage (SAGD)

Steam-assisted gravity drainage (SAGD) has been widely proven to enhance heavy oil recovery effectively. However, the performance evaluation of SAGD in heavy oil reservoirs usually involves time-consuming and tedious static and dynamic numerical simulations, which is not suitable for practical decision-making and prediction. As a reliable alternative modeling tool, the deep learning-based model has great potential for comprehensive data analysis and system prediction of machine learning methods. Still, it has not been widely used in SAGD development planning. In this paper, a recurrent neural network model is applied to the prediction of SAGD production information. A large number of simulation results are collected using CMG software to build a training database, which includes static parameters describing various characteristics of the reservoir and dynamic parameters for production control. The deep learning-based model is divided into two parts. One part is responsible for predicting the thermal communication in the preheating stage, and the other part is responsible for predicting the production information in the production stage, using the principal component analysis method to reduce the dimension of the input vector, limiting the overfitting and improving the quality of the prediction. Besides, an intelligent history matching technology for oilfield SAGD production is proposed. The model is designed to be controlled by a user-friendly graphical interface for user convenience and can be directly integrated into reservoir management programs. This research highlights the enormous potential of replacing traditional modeling tools with neural network models for decision-making and predicting SAGD production projects in oilfields.

An Improved Set of Design Criteria for Slotted Liners in Steam Assisted Gravity Drainage Operation

Slotted liners are widely used in steam-assisted gravity drainage (SAGD) wells to control sand production and sustain wellbore productivity. The slotted liner can provide desirable performance when appropriately designed. A literature review indicates a limited number of studies that offer design criteria specifically for SAGD wells. Moreover, past criteria seem to neglect some key factors, which may lead to inadequate slot design. This paper proposes a set of graphical design criteria for slotted liners in SAGD production wells, using prepacked sand retention testing (SRT) data. The SRT is designed to incorporate several essential factors that are not present in the past design criteria, such as slot density, steam breakthrough, and particle size distribution (PSD). The proposed design criteria are presented graphically for normal and aggressive conditions, where the aggressive condition accounts for the potential occurrence of the steam breakthrough. It is found that the upper bound of the design window is substantially lower for the aggressive condition due to the higher sand production after the steam breakthrough. The design criteria also indicate that the slotted liner is suitable only for the formations with low fines content.

A Novel sand control testing facility to evaluate the impact of radial flow regime on screen performance and its verification

Optimum design of the sand control devices in oil sand reservoirs plays a vital role in minimizing the sand production and increasing the reservoir productivity in Steam-Assisted Gravity Drainage (SAGD) operations. Various sand control testing facilities have been developed to evaluate the performance of sand control screens, such as the pre-packed Sand Retention Test (SRT). Current testing apparatuses are based on the linear flow regime. However, fluid flow around SAGD production wells is radial flow, not linear. This study introduces a Full-scale Completion Test (FCT) facility to emulate the radial-flow condition in SAGD wells. Instead of using a disk-shaped screen coupon, this facility utilizes a cylindrical-shaped screen. A couple of tests were carried out to determine the flow uniformity inside the cell and identify the test repeatability. Test results show that flow is distributed uniformly inside the cell, and experiments are repeatable in terms of differential pressures, fines production, and sanding levels. Therefore, this innovative FCT experimental setup and procedure allows a more realistic evaluation of the liner performance by emulating the real SAGD flow regime around the liner. Testing results obtained from the FCT can be used to complement and validate the current testing procedures. These tools can be adopted for an objective custom-design and selection of standalone screens in SAGD.

Protocol for optimal size selection of punched screen in Steam Assisted Gravity Drainage operations

Punched screen as a stand-alone screen is gaining more and more popularity in Steam Assisted Gravity Drainage (SAGD) operations. However, the literature contains limited studies for proper aperture size selection and the performance of the punched screen. This paper aims at investigating the performance of punched screen and generating an aperture size selection protocol for SAGD production wells.

Predicting Geomechanical Dynamics of Steam-Assisted Gravity-Drainage Process. Part II: Modified Cam-Clay Model

SummaryIn Part I of this study (Irani 2018), the geomechanical effects in the reservoir associated with steam-assisted gravity drainage (SAGD) steam chamber growth was evaluated on the basis of two core assumptions: reservoir yield behavior follows that of the Mohr-Coulomb (MC) dilative behavior, and the reservoir stress response follows that of a drained sand. In Part I, it was shown that although the dilative model nicely described the shearing and the sheared zone thickness at the front of the SAGD steam chamber, it could not predict the displacements associated with cold dilation in SAGD reservoirs, in which cold dilation refers to vertical displacement created in the zone ahead of the heated zone caused by isotropic unloading generated by the pore pressure increase and the increase in far-field horizontal stress. In cold dilation, the stresses do not reach the critical state line (CSL), which defines the yield surface and should, therefore, be analyzed considering elastic behavior. A modified Cam-Clay (MCC) model, however, can be used to describe the behavior of the oil sand in the cold dilation zone before reaching the CSL. In this study and as an extension to the results presented in Part I, strains developed in the reservoir during SAGD operation are calculated using an MCC model, and the associated oil rate enhancement and displacements are evaluated. The vertical strains and displacements are compared with measured values from the extensive monitoring program conducted at the Underground Test Facility (UTF) in the late 1980s. Two aspects of geomechanical effects are compared between the cap models (Part II) and dilative models (Part I): first, prediction of the sheared zone thickness and its effect on SAGD production enhancement, and second, prediction of vertical and horizontal displacements. It is shown that consideration of the material model effects on production rates are negligible for both models and that the MCC model can predict displacements in both the heated and cold zones of the reservoir reasonably accurately. Although dilative constitutive models can be used to predict horizontal and vertical displacements in the heated zone quite accurately, they lack the ability to predict the response in the “cold dilation zone.” Another main advantage of using an MCC model is that the MCC model provides a better description of a stress path and how the reservoir mobility can affect reservoir dilation, especially in the cold dilation zone.

Review on the effective recovery of SAGD production for extra and super heavy oil reservoirs

Industrial development of extra and super heavy oil as an important energy resource substitute has been achieved worldwide. Steam-assisted gravity drainage (SAGD) is one of the most important recovery methods for extra and super heavy oil reservoirs. In this review paper, the following points of SAGD have been discussed: (1) mechanism, influencing factors and application criteria; (2) field application and experimental study; (3) analytical and reservoir simulation models; and (4) current research progress and future application of multi-medium-assisted SAGD. This study gives an overview of the current status and challenges of and solutions for the SAGD technology and provides reference for the effective recovery of extra and super heavy oil reservoirs.

Distributed Fiber-Optic Sensors Characterize Flow-Control-Device Performance

This article, written by JPT Technology Editor Judy Feder, contains highlights of paper SPE 195869, “Characterization of Flow-Control-Device Performance With Distributed Fiber-Optic Sensors,” by Ben Banack, SPE, Halliburton; Lyle H. Burke, SPE, Canadian Natural Resources; and Daniel Booy, SPE, C-FER Technologies, et al., prepared for the 2019 SPE Annual Technical Conference and Exhibition, Calgary, 30 September–2 October. The paper has not been peer reviewed. The complete paper describes piloting the collection and analysis of distributed temperature and acoustic sensing (DTS and DAS, respectively) data to characterize flow-control-device (FCD) performance and help improve understanding of steam-assisted gravity drainage (SAGD) inflow distribution. Fiber-optic-based instrumentation was deployed within FCD-equipped active wells using permanently installed coiled tubing. Logs were performed on multiple wells during stable and transient flowing conditions. Additionally, acoustic recording using flow-loop testing was completed with accelerometers, geophones, and fiber-optic cables during FCD characterization. The goal was to cross-reference the acquired acoustic signals for quantification of flow at devices and validation of performance. An overview of the flow-loop FCD acoustic characterization program is described. Introduction Installation of inflow control devices (ICDs) along SAGD production liners is common to enhance temperature conformance and accelerate depletion. Additionally, some operators advocate the installation of similar outflow control devices (OCDs) along the injection well of the SAGD well pair. Collectively, these inflow and outflow devices are often referred to as FCDs. Several FCD devices are commercially available for use in SAGD. Methodology In an effort to optimize FCD design and selection, a joint industry partnership (JIP) was formed and flow-loop testing conducted to establish FCD performance curves and erosion tolerance over wide pressure, temperature, and steam-quality ranges consistent with a typical SAGD well environment. In conjunction with flow-loop testing, several full-scale FCD deployments were completed at the JIP fields, including pilot wells at the production company’s SAGD facility. These wells were logged with fiber-optic technology. Fiber-optic-based instrumentation was deployed within FCD-equipped wells using permanently installed coiled tubing. Well-architecture-design changes to a typical completion were not required because fiber-optic sensors are used for most non-FCD wells to collect DTS data. Although DTS is a common tool for optimizing SAGD production, it has certain limitations. Specifically, temperature changes along production wells typically do not allow a detailed definition or quantification of the inflow distribution along the wellbore. In addition to DTS, DAS was performed periodically on the FCD wells. DAS logging of SAGD producers has several potential uses, including flow profiling, steam breakthrough or noncondensable gas (NCG) detection, multiphase flow characterization, electric submersible pump (ESP) performance, completion failure analysis, and 4D seismic analysis. Although FCD characterization with DAS appears promising, a knowledge gap exists regarding how to move beyond qualitative analysis to quantitative analysis of FCD performance and the lateral emulsion inflow distribution. Pending satisfactory results, DAS logging on active wells potentially can be completed to accelerate improvements of SAGD FCD performance and design as well as increase the efficiency of SAGD recovery through improved steam/oil ratio and an associated reduction in greenhouse gases.

Machine learning-based prediction of the shale barrier size and spatial location using key features of SAGD production curves

The shale barrier (SB) tends to impede the steam chamber development because of the low permeability of shale during the steam-assisted gravity drainage (SAGD) process. The interpretation of 4-D seismic data in geostatistical reservoir modelling is traditionally used to understand better the uncertainty of distribution of SB. On the other hand, long-term 4-D seismic survey to monitor the steam chamber development incurs additional cost.In this study, machine learning (ML) approaches, such as multiple linear regression (MLR), Random Forest, and artificial neural network (ANN), were performed to predict the horizontal location, width, and length of the SB using the inflection points of the SAGD production data; oil rate, cumulative oil, steam-oil ratio, and cumulative steam-oil ratio and the reservoir and operating parameters; reservoir thickness, horizontal permeability, vertical-horizontal permeability ratio, oil saturation, and steam injection pressure without the application of 4-D seismic data.Data pre-processing was carried out to examine and select 38 parameters (5 reservoir and operating parameters, the vertical location and thickness of the SB, and 31 inflection points), and the dataset (2131 cases) was divided into 75% of the train set and 25% of the test set. The k-fold cross validation and grid search process for hyperparameter tuning were performed to optimize the ML models. A comparison of the ML models revealed the ANN models to have larger coefficient of determination (R2) values than those of the MLR and Random Forest models. In the optimal ANN model, the R2 values of the test set for predicting the width (0.869) and length (0.918), but the R2 value for predicting the horizontal location was as low as 0.569.For verification, the optimal ANN model was tested using the production data from Husky's Sunrise and Suncor's Firebag SAGD projects in Canada. The field production data and simulation results using the reconstructed SB showed a similar trend, and the similarity between the field data and simulation results was improved compared to the results reported by Kim and Shin (2018).

Integration of deep learning and data analytics for SAGD temperature and production analysis

Shale heterogeneities often impede the development of a steam chamber in many steam-assisted gravity drainage (SAGD) projects. Unfortunately, static data alone is generally insufficient for inferring the corresponding distribution of shale barriers. This study presents a novel data-driven modeling workflow, which integrates deep learning (DL) and data analytics techniques to analyze production profiles from horizontal well pairs and temperature profiles from vertical observation wells, for the inference of shale barrier characteristics. Field data gathered from several Athabasca oil sands projects are extracted to build a set of synthetic SAGD models, where the geometries, proportions, and spatial distribution of shale barriers are modeled stochastically. Numerical flow simulation is performed on each realization; the corresponding production/injection time series data, as well as temperature profiles from one vertical observation well, are recorded. A large dataset is assembled for the development of data-driven models: wavelet analysis and other data analysis techniques are performed to extract relevant input features from the temperature and production profiles; a novel parameterization scheme is also proposed to formulate the output variables that would effectively describe the detailed distribution of shale barriers. Convolutional neural network, one type of DL, is applied to capture the complex and nonlinear relationships between these input and output variables. The feasibility of the developed workflow is validated using synthetic test cases. Salient features capturing the impacts of shale barriers are extracted. It is observed from the production time series data that, as the steam chamber approaches a shale barrier, a declining pattern is noticeable until the steam chamber advances around the shale barrier. An obstruction in the steam chamber development can also be noted in the temperature profiles, as steam is trapped by shale barriers that are located reasonably close to the horizontal well pair. This observation is confirmed by comparing the petrophysical logs and the temperature profiles at the observation wells. Analyzing both temperature and production data could help to infer the size of shale barriers in the inter-well regions. Finally, the model outputs are used to generate an ensemble of heterogeneous SAGD realizations that correspond to the input production and temperature time series data. This study offers a computationally efficient tool for inferring stochastically distributed shale barriers in SAGD models, which can be subjected to detailed history matching workflows. It is the first time that data-driven models are used to analyze both production data from horizontal production well pairs and temperature profiles from a vertical observation well for inferring SAGD reservoir heterogeneities. The results illustrate the potential for the application of data analytics in reservoir modeling and flow simulation analysis. The developed workflow also can be extended to characterize reservoir heterogeneities in other recovery processes.

Experimental Correlations for the Performance and Aperture Selection of Wire-Wrapped Screens in Steam-Assisted Gravity Drainage Production Wells

Summary Wire-wrapped screens (WWSs) are one of the most-commonly used devices by steam-assisted gravity drainage (SAGD) operators because of the capacity to control plugging and improve flow performance. WWSs offer high open-to-flow area (OFA) (6 to 18%) that allow a high release of fines, hence, less pore plugging and accumulation at the near-screen zone. Over the years, several criteria have been proposed for the selection of aperture sizes on the basis of different industrial contexts and laboratory experiments. Generally, existing aperture-sizing recommendations include only a single point of the particle-size distribution (PSD). Operators and academics rely on sand-control testing to evaluate the performance of sand-control devices (SCDs). Scaled laboratory testing provides a straightforward tool to understand the role of flow rate, flowing phases, fluid properties, stresses, and screen specifications on sand retention and flow impairment. This study employs large-scale prepacked sand-retention tests (SRTs) to experimentally assess the performance of WWSs under variable single-phase and multiphase conditions. The experimental results and parametric trends are used to formulate a set of empirical equations that describe the response of the WWS. Several PSD classes with various fines content and particle size are tested to evaluate a broad range of PSDs. Operational procedures include the coinjection of gas, brine, and oil to emulate aggressive conditions during steam-breakthrough events. The experimental investigation leads to the formulation of predictive correlations. Additional PSDs were prepared to verify the adequacy of the proposed equations. The results show that sanding modes are both flow-rate and flowing-phase dependent. Moreover, the severity or intensity of producing sand is greatly influenced by the ratio of grain size to aperture size and the ability to form stable bridges. During gas and multiphase flow, a dramatic amount of sanding was observed for wider apertures caused by high multiphase flow velocities. However, liquid stages displayed less-intense transient behaviors. Remarkably, WWSs rendered an excellent flow performance even for low-quality sands and narrow apertures. Although further and more complete testing is required, empirical correlations showed good agreement with experimental results.

A steam rising model of steam-assisted gravity drainage production for heavy oil reservoirs

The steam chamber rising process is an essential feature of steam-assisted gravity drainage. The development of a steam chamber and its production capabilities have been the focus of various studies. In this paper, a new analytical model is proposed that mimics the steam chamber development and predicts the oil production rate during the steam chamber rising stage. The steam chamber was assumed to have a circular geometry relative to a plane. The model includes determining the relation between the steam chamber development and the production capability. The daily oil production, steam oil ratio, and rising height of the steam chamber curves influenced by different model parameters were drawn. In addition, the curve sensitivities to different model parameters were thoroughly considered. The findings are as follows: The daily oil production increases with the steam injection rate, the steam quality, and the degree of utilization of a horizontal well. In addition, the steam oil ratio decreases with the steam quality and the degree of utilization of a horizontal well. Finally, the rising height of the steam chamber increases with the steam injection rate and steam quality, but decreases with the horizontal well length. The steam chamber rising rate, the location of the steam chamber interface, the rising time, and the daily oil production at a certain steam injection rate were also predicted. An example application showed that the proposed model is able to predict the oil production rate and describe the steam chamber development during the steam chamber rising stage.

A set of graphical design criteria for slotted liners in steam assisted gravity drainage production wells

Slotted liners have been widely used in steam-assisted gravity drainage (SAGD) wells owing to their low cost and superior mechanical integrity. Multiple factors affect the performance of slotted liners, such as particle size distribution (PSD) of formation sands, aperture size, slot density, fluid flow rate, and wellbore operational conditions. Currently, most of the existing design criteria formulate the lower and upper bounds of the aperture based on one or several points on the particle size distribution curve of oil sands. Most of these design criteria neglect the slot density, wellbore operational conditions, and shape of PSD curve.This study carries out a series of large-scale pre-pack sand retention tests (SRT) in step rates. The aim is to investigate the impacts of aperture size, slot density, and fluid flow rate on the slotted liner performance. Comprehensive design criteria for determining the safe aperture window are presented to maintain the sanding and the wellbore plugging of the zone near the slotted liners within an acceptable level. Sand production governs the upper bound of the aperture size, and flow performance guides the lower bound of the aperture size. The new criteria are presented graphically to illustrate the optimal slot window as a function of the sand PSD, slot density, and fluid flow rate. The results of separate tests are used to demonstrate the performance of the new design criteria. The optimal slot window obtained via the new design criteria guides the slot liner selection in the SAGD process.

Modeling of reservoir temperature upon preheating in SAGD wells considering phase change of bitumen

Steam circulation prior to production is usually employed to establish inter-well thermal communication and create an expected initial steam chamber in the process of bitumen recovery using steam assisted gravity drainage (SAGD). To realistically evaluate the efficiency of steam circulation, and to find the indicator for terminating the steam circulation and switching to SAGD production, modeling of the temperature propagation considering the bitumen-rich oil sand reservoir properties and oilfield operations is required. This study proposed two new modified methodologies, including an analytical model as well as a numerical model, in which the latent heat effect caused by the phase change of bitumen was involved for thermo-phase change coupled analyses. The modeling was conducted for a real SAGD steam circulation project in an oilfield located in Karamay, northwest of China, using oilfield operation parameters and thermo-dynamical properties obtained from laboratory tests. The models can predict the temperature propagation and bitumen phase change behavior in the domain encompassing dual wells. In general, the predicted results obtained from both analytical and numerical approaches agree well with those in published articles in terms of the cooling time derived from temperature falloff tests and inter-well midpoint temperature. The modeling draws major conclusions that the temperature drops rapidly with the distance to wellbore in the fusion zone, but decreases slowly in the solid zone. The heat flux on wellbore exhibits an exponential reduction in the beginning, but ultimately tends to be a constant. The moving speed of phase change interface decreases with injection time, and finally increases up to only a short distance. The inter-well midpoint temperature is heated to exceed the threshold temperature, and the whole inter-well area is ultimately occupied by the fusion zone. These findings not only can provide a criterion for evaluating the efficiency of steam circulation and an indicator for terminating the steam circulation and switching to production, but also the approaches proposed may be used to analyze these heat injection and phase change problems occurring in other types of oil and gas-rich reservoirs like the natural gas hydrate formation.

Bayesian Biclustering by dynamics: A clustering algorithm for SAGD time series data

Steam-Assisted Gravity Drainage (SAGD) is an extra heavy oil recovery process consisting of an upper horizontal steam injector and a lower horizontal producer that removes fluids from the reservoir. Given its costs and environmental impact, SAGD injection strategies must be examined to find ways to improve performance. In this paper, a new Bayesian Biclustering by Dynamics (BBCD) method is proposed, which finds and differentiates groups of SAGD wells based on their oil production response to steam injected over time. This is a greedy algorithm that automatically clusters both rows and columns of SAGD injection and production data and then generates a descriptive summary for each cluster. Clusters are described with probability distributions that capture the likelihood of transitioning between discrete steam-to-oil ratio states. In addition, BBCD incorporates background knowledge on SAGD directly into the clustering process via prior transition probability distributions. Real SAGD operation data from sites in Alberta, Canada are used for this analysis. The results reveal nine production responses to two different steam injection strategies, and new insights into SAGD process performance.

An experimental and numerical study of a steam chamber and production characteristics of SAGD considering multiple barrier layers

Steam-assisted gravity drainage (SAGD) is an efficient technology that has been used to develop oil sand resources, and it has been successfully and maturely applied in Canada. However, oilfield production has demonstrated that reservoir heterogeneity has a serious impact on SAGD development, among which the most drastic impact is caused by multiple barrier layers in the reservoir. These barrier layers can severely impede the development of a steam chamber and the drainage of oil. Hence, it is important to investigate their influence mechanism. In this study, oil samples from the Long Lake oil field were used in laboratory experiments to study the development of the steam chamber and the residual oil distribution under the effect of multiple barrier layers. Then, a theoretical numerical simulation model was established to describe the fluid flow more precisely. In addition, the production characteristics of SAGD under the influence of different numbers of barrier layers were analyzed by comparing the development of the steam chamber. Finally, the impacts of multiple barrier layers with different modes of combinations were studied for their effects on SAGD production. The results indicated that for impermeable barrier layers of the same length, the steam primarily developed upward after flowing around the first barrier layer. After the steam chamber reached the top of the reservoir, the steam began to enter into the interbedded zone. Based on an unchanged first barrier layer, the number of barrier layers had little influence on the overall shape of the steam chamber and SAGD production, which confirmed that the first barrier layer played a dominant role in the influence of multiple barrier layers. In addition, characteristic points (P1, P2, P3), which corresponded to changes in the stage of steam chamber development, were established and used to evaluate the effect of the barrier layers. Different combinations of barrier layers were realized by changing the relative properties of the first barrier layer. The results showed that the longer the length of the first barrier layer, the closer it was to the steam injection well and the lower the permeability; hence, the more obvious its hysteresis effect on the SAGD process.

An Empirical Oil, Steam, and Produced-Water Forecasting Model for Steam-Assisted Gravity Drainage With Linear Steam-Chamber Geometry

Summary Presented here is a steam–assisted–gravity–drainage (SAGD) forecasting technique for oil and water production and steam injection based on linear steam–chamber geometry and an empirical form for the progression of the steam/oil interface with time. The oil forecast model generates full–cycle SAGD production profiles using three empirical inputs (the initial plateau oil rate qo,si, an exponent n, and the rising–phase–chamber angle θ) and volumetric parameters (well length, pay thickness, porosity, half–well spacing, and initial and residual oil saturation). Steam and produced–water forecasts are derived analytically using the resulting steam–chamber geometry and the heat required for the fluids, rock, and overburden. Peak rate is a direct input in this methodology, and as a result only volumetric and some thermodynamic parameters are required, but not fluid or transmissibility inputs, such as viscosity and permeability. This allows for direct use of commercial SAGD production data in the forecasting process. The model predictions are validated at a high level with field data from Devon Canada Corporation's Jackfish SAGD project and numerical simulation.

A Proxy Model for Predicting SAGD Production From Reservoirs Containing Shale Barriers

Artificial intelligence (AI) tools are used to explore the influence of shale barriers on steam-assisted gravity drainage (SAGD) production. The data are derived from synthetic SAGD reservoir simulations based on petrophysical properties and operational constraints gathered from the Suncor's Firebag project, which is representative of Athabasca oil sands reservoirs. The underlying reservoir simulation model is homogeneous and two-dimensional. Reservoir heterogeneities are modeled by superimposing sets of idealized shale barrier configurations on this homogeneous reservoir model. The individual shale barriers are categorized by their location relative to the SAGD well pair and by their geometry. SAGD production for a training set of shale barrier configurations was simulated. A network model based on AI tools was constructed to match the output of the reservoir simulation for this training set of shale barrier configurations, with a focus on the production rate and the steam-oil ratio (SOR). Then the trained AI proxy model was used to predict SAGD production profiles for arbitrary configurations of shale barriers. The predicted results were consistent with the results of the SAGD simulation model with the same shale barrier configurations. The results of this work demonstrate the capability and flexibility of the AI-based network model, and of the parametrization technique for representing the characteristics of the shale barriers, in capturing the effects of complex heterogeneities on SAGD production. It offers the significant potential of providing an indirect method for inferring the presence and distribution of heterogeneous reservoir features from SAGD field production data.

Development and application of proxy models for predicting the shale barrier size using reservoir parameters and SAGD production data

The modelling of an interbedded shale barrier in heterogeneous oil sands reservoirs has high uncertainty, which has a critical impact on the steam-assisted gravity drainage (SAGD) performance. Generally, an interpretation of 4D seismic data is the most widely used method for understanding the uncertainty of the shale barrier size.In a shale interbedded oil sands reservoir, the SAGD production data, such as the oil rate and steam-oil ratio (SOR), increase to a peak point and decrease to a trough point; the point intermediate between these two points is called the inflection point (IP).In this study, proxy models were developed to predict the shale barrier size using reservoir parameters and the IP of the SAGD production data without utilizing a 4D seismic data interpretation. In addition, a field application was carried out to test the developed proxy models using the data from Suncor's Firebag SAGD project, Alberta, Canada.The proxy models for predicting the shale width (WD), shale length (LN), and vertical location of the shale (VL) were developed using the statistical analysis of the reservoir parameters, shale size and location, and IP from production data. As a verification procedure, homogeneous 3-D simulations were performed to reconstruct the predicted WD, LN, and VL, and the oil production profiles of the SAGD field were compared with the simulation results. The oil production profiles showed a similar trend between the simulation results of the reconstructed model and the field SAGD production data. Therefore, the developed proxy models may use to predict the shale barrier size properly, and might be useful for field applications.