Since decades, steam-assisted oil recovery processes have been successfully deployed in heavy oil reservoirs to extract bitumen/heavy oil. Current resource allocation practices mostly involve reservoir model-based open loop optimization at the planning stage and its periodic recurrence. However, such decades-old strategies need a complete overhaul as they ignore dynamic changes in reservoir conditions and surface facilities, ultimately rendering heavy oil production economically unsustainable in the low-oil-price environment. Since steam supply costs account for more than 50% of total operating costs, a data-driven strategy that transforms the data available from various sensors into meaningful steam allocation decisions requires further attention.
In this research, we propose a purely data-driven algorithm that maximizes the economic objective function by allocating an optimal amount of steam to different well pads. The method primarily constitutes two components: forecasting and nonlinear optimization. A dynamic model is used to relate different variables in historical field data that were measured at regular time intervals and can be used to compute economic performance indicators (EPI). The variables in the model are cumulative in nature since they can represent the temporal changes in reservoir conditions. Accurate prediction of EPI is ensured by retraining regression model using the latest available data. Then, predicted EPI is optimized using a nonlinear optimization algorithm subject to amplitude and rate saturation constraints on decision variables i.e., amount of steam allocated to each well pad.
Proposed steam allocation strategy is tested on 2 well pads (each containing 10 wells) of an oil sands reservoir located near Fort McMurray in Alberta, Canada. After exploratory analysis of production history, an output error (OE) model is built between logarithmically transformed cumulative steam injection and cumulative oil production for each well pad. Commonly used net-present-value (NPV) is considered as EPI to be maximized. Optimization of the objective function is subject to distinct operating conditions and realistic constraints. By comparing results with field production history, it can be observed that optimum steam injection profiles for both well pads are significantly different than that of a field. In fact, the proposed algorithm provides smooth and consistent steam injection rates, unlike field injection history. Also, the lower steam-oil ratio is achieved for both well pads, ultimately translating into ~19 % higher NPV when compared with field data.
Inspired from state-of-the-art control techniques, proposed steam allocation algorithm provides a generic data-driven framework that can consider any number of well pads, EPIs, and amount of past data. It is computationally inexpensive as no numerical simulations are required. Overall, it can potentially reduce the energy required to extract heavy oil and increase the revenue while inflicting no additional capital cost and reducing greenhouse gas emissions.
Yong, Wen Pin (PETRONAS Research Sdn. Bhd.) | Azahree, Ahmad Ismail (PETRONAS Research Sdn. Bhd.) | M Ali, Siti Syareena (PETRONAS Research Sdn. Bhd.) | Jaafar Azuddin, Farhana (PETRONAS Research Sdn. Bhd.) | M Amin, Sharidah (PETRONAS Research Sdn. Bhd.)
This paper presents a two-way coupled modelling approach to simulate CO2 movement and containment with geochemical reactions and geomechanical effects. CO2 storage simulation studies cover three main disciplines, reservoir engineering, geochemistry and geomechanics. This new approach of coupled modelling simulation, by simultaneously simulate both effects of geochemistry and geomechanics, is considered as a more representative and better predictive modelling practice.
The integration of geochemistry and geomechanics effects is important for CO2 sequestration modelling. There are a number of published studies on coupled modelling for CO2 storage. However, the majority of the studies has only covered dynamic-geomechanics or dynamic-geochemistry interaction, without considering any direct geomechanics-geochemistry interaction in a reservoir condition. It is crucial to understand the integrated effects when injected CO2 dissolves into formation water and interacts with formation rock. Depending on in-situ conditions, the formation water with dissolved CO2 could weak or strengthen the formation stress due to geochemical reactions of formation minerals. Therefore, coupled modelling is needed to ensure the long-term safety of CO2 containment at a CO2 storage site with the interactions among geomechanical, geochemical and dynamic fluid flow, and especially to understand the slow and not experimentally accessible mineral reactions.
In this paper, a high CO2 content gas field in Malaysia with high temperature (150°C) and high pressure (350 bar) has been studied using integrated coupled modelling approach. The simulation input parameters are first investigated and collected from literature and laboratory studies. A two-way coupled modelling simulation with the consideration of geochemistry and geomechanics effects is desirable because it allows the updates of reservoir properties back and forth in every time step. Different CO2 trapping mechanisms, long term fate analysis, subsidence and heaving analysis, and changes of porosity and permeability are investigated. The time frame of simulation studies consists of CO2 injection period (15 years) and post CO2 injection period (500 years).
During the first 15 years of CO2 injection, 95.13% of injected CO2 is structurally trapped, 3.67% of CO2 is soluted in formation water and 1.2% is trapped by mineralization. About 0.041m of heaving is observed at the injection area while about 0.05m of subsidence is observed at the production area. In the investigation of long-term CO2 fate, it is observed that CO2 gas will be trapped between the lighter hydrocarbon gas layer and aquifer due to density difference.
Shale heterogeneities often impede the development of 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. DL, such as convolutional neural network, together with other data analytics techniques are 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 decline 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 complementary and computationally-efficient tool for inference of 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 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.
Xiong, Hao (University of Oklahoma) | Huang, Shijun (China University of Petroleum, Beijing) | Devegowda, Deepak (University of Oklahoma) | Liu, Hao (China University of Petroleum, Beijing) | Li, Hao (University of Oklahoma) | Padgett, Zack (Univiersity of Oklahoma)
Hao Xiong, University of Oklahoma; Shijun Huang, China University of Petroleum, Beijing; Deepak Devegowda, University of Oklahoma; Hao Liu, China University of Petroleum, Beijing; and Hao Li and Zack Padgett, University of Oklahoma Summary Steam-assisted gravity drainage (SAGD) is the most-effective thermal recovery method to exploit oil sand. The driving force of gravity is generally acknowledged as the most-significant driving mechanism in the SAGD process. However, an increasing number of field cases have shown that pressure difference might play an important role in the process. The objective of this paper is to simulate the effects of injector/producer-pressure difference on steam-chamber evolution and SAGD production performance. A series of 2D numerical simulations was conducted using the MacKay River and Dover reservoirs in western Canada to investigate the influence of pressure difference on SAGD recovery. Meanwhile, the effects of pressure difference on oil-production rate, stable production time, and steam-chamber development were studied in detail. Moreover, by combining Darcy's law and heat conduction along with a mass balance in the reservoir, a modified mathematical model considering the effects of pressure difference is established to predict the SAGD production performance. Finally, the proposed model is validated by comparing calculated cumulative oil production and oil-production rate with the results from numerical and experimental simulations. The results indicate that the oil production first increases rapidly and then slows down when a certain pressure difference is reached. However, at the expansion stage, lower pressure difference can achieve the same effect as high pressure difference. In addition, it is shown that the steam-chamber-expansion angle is a function of pressure difference. Using this finding, a new mathematical model is established considering the modification of the expansion angle, which (Butler 1991) treated as a constant. With the proposed model, production performance such as cumulative oil production and oil-production rate can be predicted. The steam-chamber shape is redefined at the rising stage, changing from a fanlike shape to a hexagonal shape, but not the single fanlike shape defined by (Butler 1991). This shape redefinition can clearly explain why the greatest oil-production rate does not occur when the steam chamber reaches the caprock.
Hadavand, Mostafa (University of Alberta) | Carmichael, Paul (ConocoPhillips Canada) | Dalir, Ali (ConocoPhillips Canada) | Rodriguez, Maximo (ConocoPhillips Canada) | Silva, Diogo F. S. (University of Alberta) | Deutsch, Clayton Vernon (University of Alberta)
Mostafa Hadavand, University of Alberta; Paul Carmichael, Ali Dalir, and Maximo Rodriguez, ConocoPhillips Canada; and Diogo F. S. Silva and Clayton V. Deutsch, University of Alberta Summary 4D seismic is one of the main sources of dynamic data for heavy-oil-reservoir monitoring and management. Thus, the large-scale nature of fluid flow within the reservoir can be evaluated through information provided by 4D-seismic data. Such information may be described as anomalies in fluid flow that can be inferred from the unusual patterns in variations of a seismic attribute. During steam-assisted gravity drainage (SAGD), the steam-chamber propagation is fairly clear from 4D-seismic data mainly because of changes in reservoir conditions caused by steam injection and bitumen production. Anomalies in the propagation of the steam chamber reflect the quality of fluid flow within the reservoir. A practical methodology is implemented for integration of 4D seismic into SAGD reservoir characterization for the Surmont project. Introduction One of the main objectives in petroleum-reservoir modeling is to predict the future performance of the reservoir under a recovery process. It is not possible to establish the true spatial distribution of reservoir properties using limited data. Thus, the modeling process is ill-posed and subject to uncertainty (Pyrcz and Deutsch 2014). Geostatistical simulation provides a framework to quantify geological uncertainty that is represented by multiple equally probable realizations of the reservoir model. The uncertainty can be reduced by integration of all available sources of data, including static and dynamic (time-variant) data. However, each source of data provides information at different scales and levels of precision. Although there are well-established geostatistical techniques to generate stochastic realizations of the reservoir conditioned to static data, such as local measurements from wells and 2D/3D-seismic data, effective integration of dynamic data remains a major challenge. Time-lapse seismic, or 4D seismic, is one of the main dynamic sources of data for heavy-oil-reservoir monitoring and management. It contains valuable information regarding fluid movement, temperature, pressure buildup, and quality of fluid flow within the reservoir during a recovery process (Lumley and Behrens 1998; Gosselin et al. 2001). For SAGD, the evolution of the steam chamber over time is fairly clear in 4D-seismic images.
Mahmoudi, Mahdi (RGL Reservoir Management) | Fattahpour, Vahidoddin (RGL Reservoir Management) | Velayati, Arian (University of Alberta) | Roostaei, Morteza (RGL Reservoir Management) | Kyanpour, Mohammad (RGL Reservoir Management) | Alkouh, Ahmad (College of Technological Studies) | Sutton, Colby (RGL Reservoir Management) | Fermaniuk, Brent (RGL Reservoir Management) | Nouri, Alireza (University of Alberta)
Sand control and sand management require a rigorous assessment of several contributing factors including the sand facies variation, fluid composition, near-wellbore velocities, interaction of the sand control with other completion tools and operational practices. A multivariate approach or risk analysis is required to consider the relative role of each parameter in the overall design for reliable and robust sand control. This paper introduces a qualitative risk factor model for this purpose.
In this research, a series of Sand Retention Tests (SRT) was conducted, and results were used to formulate a set of design criteria for slotted liners. The proposed criteria specify both the slot width and density for different operational conditions and different classes of Particle Size Distribution (PSD) for the McMurray oil sands. The goal is to provide a qualitative rationale for choosing the best liner design that keeps the produced sand and skin within an acceptable level. The test is performed at several flow rates to account for different operational conditions for Steam Assisted Gravity Drainage (SAGD) and Cyclic Steam Stimulation (CSS) wells. A Traffic Light System (TLS) is adopted for presenting the design criteria in which the red and green colors are used to indicate, respectively, unacceptable and acceptable design concerning sanding and plugging. Yellow color in the TLS is also used to indicate marginal design.
Testing results indicate the liner performance is affected by the near-wellbore flow velocities, geochemical composition of the produced water, PSD of the formation sand and fines content, and composition of formation clays. For low near-wellbore velocities and typical produced water composition, conservatively designed narrow slots show a similar performance compared to somewhat wider slots. However, high fluid flow velocities or unfavorable water composition results in excessive plugging of the pore space near the screen leading to significant pressure drops for narrow slots. The new design criteria suggest at low flow rates, slot widths up to three and half times of the mean grain size will result in minimal sand production. At elevated flow rates, however, this range shrinks to somewhere between one and a half to three times the mean grain size.
This paper presents novel design criteria for slotted liners using the results of multi-slot coupons in SRT testing, which is deemed to be more realistic compared to the single-slot coupon experiments in the previous tests. The new design criteria consider not only certain points on the PSD curve (e.g., D50 or D70) but also the shape of the PSD curve, water cut, and gas oil ratio and other parameters.
Production deferment due to wellbore sanding issues is a major risk for heavy oil field development. The heavy oil reservoir in Kuwait is a multi-stacked unconsolidated formation, which is prone to sanding. Currently there are two steam flood pilots in inverted 5-spot pattern configuration with pattern areas of 5 and 10 acres. The wells were operated for two cycles of Cyclic Steam Stimulation (CSS), before their conversion to steam flood.
Different Sand Control equipment is field tested in some pilot wells to optimize production in this viscous oil-saturated unconsolidated sandstone reservoir. This paper will discuss the operational challenges and the difference in the performance of cold production and after thermal CSS cycles of the installed Stand-Alone Sand Screens (SAS), which were retrofitted in the pilot wells. The mesh size of the SAS was designed based on the particle size distribution and well operating condition.
A comprehensive reservoir and well surveillance program was conducted to monitor and gather necessary data to characterize the reservoir and well performance in the pilot wells equipped with sand control equipment. The primary objective was to determine the optimal sand control strategy, moving forward in to the commercial phase of field development.
This paper will discuss the learnings from the pilot wells. The various SAS used were partially successful in mitigating sand production. More piloting with advanced sand controlling technologies at both laboratory and field levels may be required to reach to the optimum design for the field-specific cases.
The well trials using SAS was mainly to assess the screens regain permeability (skin) versus crude oil production while minimizing sand movement within the wellbore. This helped to improve the artificial lift pump run life and minimize sand debris from entering the pump.
Abdelfatah, Elsayed (Canada Excellence Research Chair in Material Engineering for Unconventional Oil Reservoirs, Chemical and Petroleum Engineering Department, University of Calgary) | Berton, Paula (Canada Excellence Research Chair in Material Engineering for Unconventional Oil Reservoirs, Chemical and Petroleum Engineering Department, University of Calgary) | Rogers, Robin (525 Solutions, Inc.) | Bryant, Steven (Canada Excellence Research Chair in Material Engineering for Unconventional Oil Reservoirs, Chemical and Petroleum Engineering Department, University of Calgary)
Steam injection is widely used for bitumen recovery. However, steam is not efficient for shallow or thin reservoirs due to heat losses in the wellbore or to surrounding formations. Numerous alternatives have been proposed, including the addition of solvents and replacing steam with volatile solvents. Here we propose a new technology combining nonvolatile ionic liquids and waterflooding for bitumen recovery that can deliver very high recovery at ambient temperature.
Different ionic liquids were designed for complete dissolution of bitumen at ambient temperature. The designed ionic liquids were tested in coreflood experiments into high-grade oil sand from Alberta. Two different scenarios were tested; continuous injection of ionic liquids at different injection rates or injection of a slug of ionic liquids followed by water injection. Different slug volumes were tested at a constant injection rate. After ionic liquids injection, the oil sand was removed from the column and the remaining bitumen was quantified using a modified Dean-Stark method. Viscosity and solid content measurements of the recovered samples at breakthrough were conducted.
Bitumen recovery by injection of ionic liquids through the column is kind of miscible displacement process. Tuning the physical and chemical properties of the ionic liquids is the most important aspect for achieving the desired interaction with the oil sand system. These properties of the designed ionic liquid depend on the selected cation and anion and the strength of their intermolecular interaction. The cation and anion chosen for ionic liquid 1 (IL1) form a viscous that can recover bitumen with slight bitumen remaining behind, but with large pressure gradient. Changing the cation produces significantly less viscous IL2 and IL3 which completely recover the bitumen in the oil sand pack. Moreover, the cation can be tailored to significantly minimize the fines migration and viscosity of the recovered bitumen. In all cases, these recoveries are significant, compared to the currently used technologies.
This work proves that bitumen recovery from oilsands is possible at low temperature via a miscible displacement process, in contrast to viscosity reduction processes achieved by thermal methods, with the design of the proper ionic liquids. The properties of these ionic liquids can be tuned for different recovery mechanisms. Thus, this work establishes the basis for developing a new class of in situ recovery processes with high recovery efficiencies and low environmental impact.
In 2014, number of exploratory wells were drilled and tested in Umm Niqa (UN) Field located in north Kuwait NE which approved a new discovery in Lower Fars (LF) reservoir.
LF is unconsolidated, sub –hydrostatic- sand stone reservoir with highly sour and moderate corrosive environment (H2S 8% and CO2 4%). Subsequently additional wells were drilled to evaluate the production potential of UN field. With rig on location UN wells are completed with test (Progressive Cavity Pump) PCP and tested. During the initial testing period the PCP is run at different speeds to evaluate the well productivity, water cut, and determine sand-free draw down to enable selection of suitable completion PCP for production.
Well UN-X is one of the developed wells which is perforated in LF sand in overbalanced condition using 4-1/2" (High Shot Density) HSD guns with 0.83" entrance hole diameter at 12 shoots per foot.
During initial testing with test PCP, the pump tripped due to high torque because of sand production (up to 60%). Five runs were performed to clean out the wellbore and repeated test PCP runs failed due to high sand production.
Coordination between FDHO (Field Development Heavy Oil) and Discovery Promotion Team was conducted to perform quick sand analysis to LF sands from offset sand distribution since subject well has no available sieve analysis. Based on the outcome of sieve analysis, decision was made to utilize one of the available SAS (Stand Alone Screen) designed for LF sand in another field to control sand production. It was agreed by both teams to install SAS in the subject well to mitigate the sand problem and minimize cost due to NPT (Non-Productive Time) of the rig. SAS was installed and the potential zone in UN-X could be tested successfully with tubing PCP. No sand problem was observed during testing and after testing while clean out operation there was no sand.
Well test showed an average liquid rate of 124 BFPD with 37% WC (predominantly completion brine). The well was put on production on November 2016 and producing till date without any sand problem.
This paper will include discussion on the approach used to select a sand control method for cold and heavy oil production. The results of sieve analysis was in the middle between sand screen and gravel pack but based on the team experience in sand control and the nature of heavy oil and its relatively low oil production rate, the decision was made to install SAS and that was proved to be prudent decision.
Sidahmed, Anas (University of Alberta) | Nouri, Alireza (University of Alberta) | Kyanpour, Mohammad (RGL Reservoir Management Inc.) | Nejadi, Siavash (University of Alberta) | Fermaniuk, Brent (RGL Reservoir Management Inc.)
Canada has enormous oil reserves which ranks third worldwide with proven oil reserves of 171 billion barrels. Alberta alone contributes with 165.4 billion barrels found in oil sands. However, the oil in oil sands is extremely viscous, and only 10% is recoverable through open-pit mining. In-situ thermal recovery methods such as Steam-Assisted Gravity Drainage (SAGD) have been developed and adopted as an efficient means to unlock the oil sands reserves.
Different reservoir geological settings and long horizontal wells impose limitations and operational challenges on the implementation of SAGD technology. Wellbore trajectory excursions or undulations- unintentionally generated trajectory deviations due to suboptimal drilling operations- are some of the complications that lead to non-uniform steam chamber conformance, high cumulative Steam-Oil Ratio (cSOR) and low bitumen recovery.
Conventional dual-string completion scheme (a short tubing landed at the heel, and a long tubing landed at the toe) has been widely adopted in most of the SAGD operations. Such configurations allow steam injection at two points: the toe and the heel sections of the horizontal well. However, these completions have demonstrated poor efficiency when reservoir/well complications exist. Tubing-deployed Flow Control Devices (FCD's) have been introduced to offer high flexibility in delivering specific amounts of steam to designated areas (such as low permeability zones) and ensure uniform development of steam chamber in the reservoir. The work in this thesis presents the results of a numerical effort for optimizing the design of Outflow Control Devices (OCD's) in SAGD wells for different scenarios of well pair trajectory excursions.
A coupled wellbore-reservoir SAGD simulation model was constructed to optimize the placement and number of ports in every single OCD. Three different cases were generated from the constructed basic SAGD model with each case having a specific well pair trajectory which causes variable lateral distances between the well pair.
Results of the optimized OCD's cases demonstrate a higher SAGD efficiency compared to their corresponding conventional dual-string cases. Those enhancements resulted in a higher steam chamber conformance, a higher cumulative oil production, and an improved Net Present Value (NPV).