Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Results
Improving Subsurface Characterization Utilizing Machine Learning Techniques
Koray, Abdul-Muaizz (New Mexico Institute of Mining and Technology) | Bui, Dung (New Mexico Institute of Mining and Technology) | Ampomah, William (New Mexico Institute of Mining and Technology) | Appiah-Kubi, Emmanuel (New Mexico Institute of Mining and Technology) | Klumpenhower, Joshua (New Mexico Institute of Mining and Technology)
Abstract The objective of this paper is to present a framework that applies machine learning to reservoir characterization. Machine learning applications in the oil and gas industry is rapidly becoming popular and in recent years has been utilized for the characterization of various reservoirs. Conventional reservoir characterization employs core data measurements and local correlations between porosity and permeability as input data for reservoir property modeling. However, a strong correlation between porosity and permeability as well as reliable core measurements are not always available. The proposed approach uses both well logs and core data to construct different models to predict permeability using three distinct methods including a parametric, non-parametric, and machine learning technique. The parametric method employed the known relationship between porosity and the natural log of permeability. The non-parametric regression method utilized the alternating conditional expectation (ACE) algorithm. The third approach involved machine learning workflow implemented within a commercial software. The reservoir was first classified into distinct hydraulic flow units using the flow zone indicator (FZI) approach and k-means clustering. Permeability was then predicted using a supervised machine-learning framework. A field case study was then utilized to ascertain the effectiveness of these approaches by validating the model with data from one of the wells. The results of these three approaches were compared using the mean absolute error (MAE) and mean squared error (MSE) values in the validation process. An examination of the error calculated found the support vector machine (SVM) and linear regression algorithms in characterizing the upper reservoir region and the SVM for the lower reservoir characterization yielding the best results when using the machine learning approach thus, yielding the least error as compared to the other two approaches. Additional validation was performed by comparing different models based on permeability fields through numerical model calibration to historical data. It was found that machine learning-based permeability had the least error compared to calibration data prior to the history matching process. The investigated reservoir consists of two distinct productive oil zones separated by an impermeable shale. There are 15 existing wells that have been producing from both the upper and lower zones since 1997. Using machine learning permeability-based model, the history matching process was conducted successfully to match both observed production data and pressure data of 15 wells with less than 10% global deviation. This study presents the feasibility of applying several different approaches in predicting permeability based on gamma ray, bulk density, and deep resistivity logs. The machine learning approach proves its high potential and readiness in supporting reservoir characterization and history matching compared to the other approaches.
- Geology > Rock Type > Sedimentary Rock > Clastic Rock (0.50)
- Geology > Geological Subdiscipline > Geomechanics (0.46)
- North America > United States > Texas > Anadarko Basin (0.99)
- North America > United States > Oklahoma > Anadarko Basin (0.99)
- North America > United States > Kansas > Anadarko Basin (0.99)
- Asia > Indonesia > Java > East Java > Pangkah PSC > Kujung Formation > Kujung-1 Formation (0.99)
Improved Model for Predicting the Productivity of Multi-Fractured Shale Wells. TMS and EFS Field Data as Case Studies
Fadairo, Adesina Samson (University of North Dakota, United States) | Egenhoff, Sven (University of North Dakota, United States) | Adeyemi, Gbadegesin Abiodun (University of North Dakota, United States) | Ling, Kegang (University of North Dakota, United States) | Tomomewo, Olusegun Stanley (University of North Dakota, United States) | Oladepo, Adebowale David (Circular One Resources, United States) | Oni, Opeyemi (University of North Dakota, United States) | Nwaokwu, Richmond Nduka (Litewell Completions Services Limited, Nigeria)
Abstract Multi-fractured horizontal wells have been an admirable completion technique for unconventional resources such as in Tuscaloosa Marine shale (TMS) and Eagle Ford Shale (EFS) plays located in the United States. Studies have shown that the productivity of multi-fractured wells of these two shale plays are majorly based on the fracture conductivity, which may be dependent on the type of the geometrical shape of the fractures connecting the fluid to the well. A reliable model is desirable to the operator to accurately capture the productivity of multi-fractured shale wells. Several mathematical models have been adopted with various assumptions that include simple slot geometry for fracture shape in the derivation of production rate models. These assumptions significantly simplify the existing model's applications but limit the efficiency of the models to accurately predict the fluid production rate. Failure to utilize an elliptical fracture shape and a correct drive mechanism-based model for analyzing flow rate have been considered as a vital reason for the disparity between the calculated results by the past investigators and the exact values obtained from TMS and EFS field measurements. In this study, an elliptical model based on the fracture geometry has been derived to analyze the productivity of multi-fractured shale wells considering the accurate drive mechanism for the shale play. The model validation has been achieved using field data from the Tuscaloosa Marine shale (TMS) and the Eagle Ford Shale (EFS) plays. The results generated from the newly improved model resulted in more accurate outcomes when compared with results presented by Yang and Guo (2019) and Guo and Schechter (1997); all these authors assumed the cross-sectional area of the induced fractures as being a slot showed nonconformity using real life values from the Tuscaloosa Marine shale (TMS) and the Eagle Ford Shale (EFS) plays as benchmarks. The newly improved model reduces the prediction percentage error to 0.55% and 0.43% compared to the percentage error reported by Yang and Guo (2019) as 9.1% and 3.5% and by Guo and Schechter (1997) respectively as 29.7% and 47.2 % using the actual oilfield results as their benchmark. The accurate prediction of the long-term productivity of multi-fractured oil shale depends on the ability to determine fracture geometry and the drive mechanisms that dominantly control flow in the shale play considered. Sample calculations of flow rate of the two fields considered and the controllable parameters influencing the flow rate have also been identified. The study would serve as a tool for accurate assessment of flow rate in multi-fractured wells of shale plays and analyzes its performance.
- North America > United States > Texas (0.67)
- North America > United States > Mississippi (0.66)
- North America > United States > Louisiana (0.66)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (1.00)
- Geology > Petroleum Play Type > Unconventional Play > Shale Play (1.00)
- North America > United States > Texas > West Gulf Coast Tertiary Basin > Eagle Ford Shale Formation (0.99)
- North America > United States > Texas > Sabinas - Rio Grande Basin > Eagle Ford Shale Formation (0.99)
- North America > United States > Texas > Maverick Basin > Eagle Ford Shale Formation (0.99)
- (2 more...)
Abstract Oil production enhancement from mature fields through polymer injection has gained elevated interest due to the improved mobility and conformance controls. The suitability of polymer with harsh reservoir condition and its insitu performance dictate the success of polymer-augmented waterflooding. This motivates thorough evaluation of polymer to build optimum injection strategy for the targeted reservoir. This study aims to evaluate the impact of polymer and chase water injections in high salinity carbonate rock using single-phase coreflood experiments coupled with real-time saturation monitoring. A sulfonated polymer, acrylamido tertiary-butyl sulfonate (ATBS), was utilized and polymer solutions at different concentrations were prepared in 20 wt% brine. Coreflooding integrated with computed tomography (CT) scanning was used to generate 3D images during polymer flooding in carbonate outcrop (95.2 mD) at 70ยฐC. Polymer injection was also evaluated in a tapered injection scheme in which polymer slug concentration was stepwise reduced. Polymer injectivity, retention, flow patterns, and inaccessible pore volume (IPV) were analyzed using pressure drop, average saturation, and CT images in real time. The results showed that the selected polymer has favorable injectivity behavior with acceptable injectivity losses (0.5-0.85) at all tested concentrations and injection rates. Polymer injection at higher concentration provided higher resistance factor, lower injectivity, and higher injectivity reduction. In-situ saturation monitoring showed polymer breakthrough before 0.8 PV with an IPV of 20%. The brine post-flush exhibited 74.1% decrease in polymer saturation after 1 PV and 99% of the polymer was recovered after 10 PVs of brine injection. For polymer augmented waterflooding with a concentration tapering mode, the IPV was reduced to 26.8%. Moreover, the chase water after tapered polymer injection showed 4.5 times high flow resistance compared to that of pre-flush brine. The chase water injection for about 1 PV reduced the retained polymer to 20% due to the increased flow resistance. This study assessed polymer injectivity and retention behavior for mobility control performance in carbonate under moderate temperature and salinity conditions. The findings of this work would guide future studies on the optimization of polymer-augmented waterflooding by using different injection schemes to improve the efficiency of mobility control process in carbonates, which would further aid in designing successful field projects.
- Asia > Middle East (1.00)
- North America > United States > California (0.47)
- North America > United States > Oklahoma (0.46)
- North America > United States > California > West Coyote Field (0.99)
- North America > United States > California > Los Angeles Basin > Wilmington Field (0.99)
- North America > United States > California > Dos Cuadras Field (0.99)
Abstract Artificial lift is the backbone of unconventional field production. Lifting oil and gas in an optimal manner and economically is one of the most challenging phases of field development with depleting reservoir energy. Traditional approaches of lift selection are not sufficient to manage unconventional wells effectively, with high decline rates initially. It is of prime interest to understand production behavior under different lift conditions since the decision on timing and design of lift method are crucial for optimizing the well performance. This paper presents an artificial-lift timing and selection (ALTS) methodology based on a hybrid data-driven and physics-based workflow to maximize the value of unconventional oil and gas assets. Our formulation employs a reduced physics model that is based on identification of Dynamic Drainage Volume (DDV) using commonly measured data (daily production rates and wellhead pressure) to calculate reservoir pressure depletion, transient productivity index (PI) and dynamic inflow performance relationship (IPR). Transient PI as the forecasting variable allows normalizing both surface pressure effects and considers phase behavior, thus reducing noise. The PI-based forecasting method is used to predict future IPRs and subsequently oil, water, and gas rates for any bottom hole pressure condition. The workflow allows estimating well deliverability under different artificial lift types and design parameters. The ALTS workflow was applied to real field cases for wells flowing under different operating conditions to optimize the best strategy to produce the well amongst several candidate scenarios. Transient PI and dynamic IPR results provided valuable insights on how and when to select different AL systems. The workflow is run periodically with everchanging subsurface and wellbore conditions against each candidate scenario with various pump models and other operating parameters (pressure, speed etc.). The method was applied to several wells in a hindcasting mode to evaluate lost production opportunity and validate the results. In certain cases, the optimal recommendation pointed to selecting a different artificial lift system than the chosen method in the field to significantly improve long term well performance. In addition, optimal artificial lift operating point recommendations are made for wells including optimal gas lift rates for gas lifted wells, optimal pump unit selection and speed for wells on ESP and SRP. The proposed method allows predicting future unconventional reservoir IPR consistently and has allowed continuous evaluation of artificial lift timing and selection scenarios for multiple lift types and designs in unconventional reservoirs. This can transform incumbent practices based on broad field heuristics, which are often ad hoc, inefficient, and manually intensive, towards well-specific ALTS analysis to improve field economics. Continuous application of this process is shown to improve production, reduce deferred production and increase life of lift equipment.
- North America > United States > Texas (0.28)
- North America > United States > Mississippi > Marion County (0.24)
- Reservoir Description and Dynamics > Unconventional and Complex Reservoirs (1.00)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Well performance, inflow performance (1.00)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Drillstem/well testing (1.00)
- (2 more...)
Abstract The development of shale plays requires accurate forecasting of production rates and expected ultimate recoveries, which is challenging due to the complexities associated with production from hydraulically fractured horizontal wells in unconventional reservoirs. Traditional empirical models like Arps decline are inadequate in capturing these complexities, and long-term forecasting is hindered by the challenges posed by 3 phase flow. In response, a new physics-augmented, data-driven forecasting method has been proposed that efficiently captures these complexities. The proposed PI-based forecasting (PIBF) method combines data-driven techniques with the physics of propagation of dynamic drainage volume under transient flow conditions observed by unconventional wells for a prolonged period. The model requires only routinely measured inputs such as production rates and wellhead pressure, and efficiently captures the trend shift in gas-to-oil ratio caused by free gas liberation in the near-wellbore region. By using material balance and productivity index models, the proposed approach can forecast well performance and handle changing operational conditions during the well's lifecycle. Compared to existing empirical or analytical methods like Arps decline and RTA, the proposed method yields more accurate forecasting results, while still using easily available inputs. Empirical methods like Arps decline have low input requirements but lack physical insights, leading to inaccuracies and inability to handle changing operational conditions. Pure physics-based methods like RTA and reservoir simulation require more input properties that are often difficult to obtain, resulting in a low range of applicability. Overall, the proposed method offers a promising alternative to existing methods, effectively combining data-driven techniques with physics-based insights to accurately forecast well performance and handle changing operational conditions in unconventional reservoirs.
- Europe (0.68)
- Africa (0.68)
- North America > United States > Texas (0.48)
- Geology > Petroleum Play Type > Unconventional Play > Shale Play (0.48)
- Geology > Rock Type (0.30)
Abstract In this work, a robust and pragmatic method has been developed, validated, and applied to describe two-phase flow behaviour of a multifractured horizontal well (MFHW) in a shale gas formation. As for a fracture subsystem, its permeability modulus, non-Darcy flow coefficient, and slippage factor have been defined and embedded into the governing equation, while an iterative method is applied to update the gas/water saturation in each fracture segment within discrete fracture networks. For a matrix subsystem, a skin factor on a fracture face is defined and introduced to represent the change in relative permeability in the matrix domain at each timestep, while the adsorption/desorption term is incorporated into the diffusivity equation to accurately calculate the shale gas production by taking the adsorbed gas in nanoscale porous media into account. Then, the theoretical model can be applied to accurately capture the two-phase flow behaviour in different subdomains. The accuracy of this newly developed model has been confirmed by the numerical simulation and then it is extended to field applications with excellent performance. The stress-sensitivity, non-Darcy flow, and slippage effect in a hydraulic fracture (HF) are found to be obvious during the production, while the initial gas saturation in a matrix and HFs imposes an evident influence on the production profile. As for an HF with a high gas saturation, the dewatering stage is missing and water from the matrix can be neglected during a short production time. For the matrix subsystem, a high-water saturation in the matrix near an HF can affect gas production during the entire stage as long as gas relative permeability in the HF remains low. In addition, the adsorption/desorption in the matrix subsystem can increase gas production but decrease water production. Compared to the observed gas/water production rates for field applications, the solutions obtained from the method in this work are found to be well matched, confirming its reliability and robustness.
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (1.00)
- Geology > Geological Subdiscipline > Geomechanics (0.93)
- Well Completion > Hydraulic Fracturing (1.00)
- Reservoir Description and Dynamics > Unconventional and Complex Reservoirs > Shale gas (1.00)
- Reservoir Description and Dynamics > Unconventional and Complex Reservoirs > Naturally-fractured reservoirs (1.00)
- Reservoir Description and Dynamics > Reservoir Fluid Dynamics > Flow in porous media (1.00)
Abstract Routinely analyzing producing well performance in unconventional field is critical to maintain their profitability. In addition to continuous analysis, there is an increasing need to develop models that are scalable across entire field. Pure data-driven approaches, such as DCA, are prevalent but fail to capture essential physical elements, compounded by lack of key operational parameters such as pressures and fluid property changes across large number of wells. Traditional models such as numerical simulations face a scalability challenge to extend to large well counts with rapid pace of operations. Other widely used method is rate transient analysis (RTA), which requires identification of flow regimes and mechanistic model assumptions, making it interpretive and non-conducive to field-scale applications. The objective in this study is to build data-driven and physics-constrained reservoir models from routine data (rates and pressures) for pressure-aware production forecasting. We propose a hybrid data-driven and physics informed model based on sparse nonlinear regression (SNR) for identifying rate-pressure relationships in unconventionals. Hybrid SNR is a novel framework to discover governing equations underlying fluid flow in unconventionals, simply from production and pressure data, leveraging advances in sparsity techniques and machine learning. The method utilizes a library of data-driven functions along with information from standard flow-regime equations that form the basis for traditional RTA. However, the model is not limited to fixed known relationships of pressure and rates that are applicable only under certain assumptions (e.g. planar fractures, single-phase flowing conditions etc.). Complex, non-uniform fractures, and multi-phase flow of fluids do not follow the same diagnostics behavior but exhibits more complex behavior not explained by analytical equations. The hybrid SNR approach identifies these complexities from combination of the most relevant pressure and time features that explain the phase rates behavior for a given well, thus enables forecasting the well for different flowing pressure/operating conditions. In addition, the method allows identification of dominant flow regimes through highest contributing terms without performing typical line fitting procedure. The method has been validated against synthetic model with constant and varying bottom hole pressures. The results indicate good model accuracies to identify relevant set of features that dictate rate-pressure behavior and perform production forecasts for new bottom-hole pressure profiles. The method is robust since it can be applied to any well with different fluid types, flowing conditions and does not require any mechanistic fracture or simulation model assumptions and hence applicable to any reservoir complexity. The novelty of the method is that the hybrid SNR can resolve several modes that govern the flow process simultaneously that can provide physical insights on the prevailing multiple complex flow regimes.
- North America > United States > Texas > Fort Worth Basin > Barnett Shale Formation (0.99)
- Africa > Tanzania > Indian Ocean > K Formation (0.99)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Production forecasting (1.00)
- Reservoir Description and Dynamics > Formation Evaluation & Management > Drillstem/well testing (1.00)
- Production and Well Operations > Well & Reservoir Surveillance and Monitoring > Production logging (1.00)
- Data Science & Engineering Analytics > Information Management and Systems (1.00)
Abstract Nowadays, the only economic and effective way to exploit shale reservoirs is multi-stage fracturing of horizontal wells. The backflow after fracturing affects the damage degree of a fracturing fluid to a formation and fracture conductivity, and directly influences a fracturing outcome. At present, the backflow control of the fracturing fluid mostly adopts empirical methods, lacking a reliable theoretical basis. Therefore, it is of positively practical significance to reasonably optimize a flowback process and control the flowback velocity and flowback process of a fracturing fluid. On the other hand, the previous research on the productivity of multi-stage fracturing horizontal wells after fracturing is limited, and an equation derivation process has been simplified and approximated to a certain extent, so its accuracy is significantly affected. Based on previous studies, this paper established a new mathematical model. This model optimizes the flowback velocity after fracturing by dynamically adjusting a choke size and analyzes and predicts the production performance after fracturing. To maximize fracture clean-up efficiency, this work builds the model for a dynamic adjustment of choke sizes as wellhead pressure changes over time. It uses a two-phase (gas and liquid) flow model along the horizontal, slanted and vertical sections. The forces acting on proppant particles, filtration loss of water, the compressibility of a fracturing fluid, wellbore friction, a gas slippage effect, water absorption and adsorption are simultaneously considered. With the theories of mass conservation, we build a mathematical model for predicting production performance from multi-fractured horizontal wells with a dynamic two-phase model considering dual-porosity, stress-sensitivity, wellbore friction, gas adsorption and desorption. In this model, the gas production mechanisms from stimulated reservoir volume and gas and water relative permeabilities are employed. Based on shale reservoir parameters, wellhead pressure, a choke size, a gas/liquid rate, cumulative gas/liquid production, cumulative filtration loss and a flowback rate are simulated. In the simulations, the influential factors, such as shut-in soak time of the fracturing fluid, forced flowback velocity, fracturing stages and fracture half-length after fracturing, are studied. It is found by comparison that in the block studied, when a well is shut in four days after fracturing, the dynamic choke size is adjusted with wellhead pressure changing over time, the fracturing stage is 11, and the fracture half-length is 350 meters, the fracture conductivity after flowback is the largest, and the productivity of the horizontal well is the highest.
- North America > United States > Texas (1.00)
- Asia (0.68)
- Geology > Rock Type > Sedimentary Rock > Clastic Rock > Mudrock > Shale (1.00)
- Geology > Geological Subdiscipline > Geomechanics (1.00)
- North America > United States > Texas > Fort Worth Basin > Barnett Shale Formation (0.99)
- Asia > China > Sichuan > Sichuan Basin (0.99)
Production Forecasting in Conventional Oil Reservoirs Using Deep Learning
Temizel, Cenk Temizel (Saudi Aramco) | Odi, Uchenna (Aramco Americas) | Al-Sulaiman, Nouf (Saudi Aramco) | Reddy, Karri (Saudi Aramco) | Putra, Dike (Rafflesia Energy) | Yurukcu, Mesut (The University of Texas Permian Basin) | Aydin, Hakki (Middle East Technical University) | Yegin, Cengiz (Incendium Technologies, LLC)
Abstract Accurate estimation of the estimated ultimate recovery (EUR) is critical in decision making processes related to the development of conventional oil reservoirs. Existing methods have limitations when it comes to predicting such long-term production behaviors. This study analyzes the performance of deep learning methods such as long short-term memory (LSTM) neural networks on time-series data, and their effective application to accurately estimate the EUR in conventional reservoirs. Synthetic data that are realistic and representative of many major conventional oil reservoirs were generated for this study. The generated dataset was used by the LSTM model for the purpose of forecasting the EUR. The results of the LSTM model were compared with that of a reservoir simulation model from a full-physics reservoir simulator. EUR forecasts from the physics-based reservoir simulator is used as a benchmark and the LSTM model shows a good predictive accuracy while forecasting the long-term production behavior from a well in a conventional oil reservoir. The LSTM model-based deep learning method can be effectively used with real-field data obtained from wells in conventional reservoirs to accurately predict the EUR, and the study provides a comparative analysis of the results and factors affecting the EUR forecasts from the LSTM model and reservoir simulation model. Deep learning methods such as LSTMs have an inherent advantage in identifying trends in time-series data and making forecasts using the data. The existing literature has a limited number of studies that outline the use of deep learning methods for EUR forecasts and this study covers this gap by providing details analyses, best practices and workflows on the use of such methods for conventional oil reservoirs.
- North America > United States > Texas > Permian Basin > Yeso Formation (0.99)
- North America > United States > Texas > Permian Basin > Yates Formation (0.99)
- North America > United States > Texas > Permian Basin > Wolfcamp Formation (0.99)
- (21 more...)
Geomechanic Interferometry: Theory and Application to Time-Lapse InSAR Data for Separating Displacement Signal Between Overburden and Reservoir Sources
Shabelansky, Andrey Hanan (Chevron Technical Center, a division of Chevron U.S.A Inc.) | Nihei, Kurt (Chevron Technical Center, a division of Chevron U.S.A Inc.) | Zhang, Zhishuai (Chevron Technical Center, a division of Chevron U.S.A Inc.) | Bevc, Dimitri (Chevron Technical Center, a division of Chevron U.S.A Inc.) | Milliken, William (Chevron San-Joaquin Valley Business Unit) | Mali, Gwyn (Chevron San-Joaquin Valley Business Unit)
Abstract Interferometric synthetic aperture radar (InSAR) data provides a measurement of the Earth's surface displacements to monitor reservoir stresses, fluid pressure and volume changes. However, the InSAR measurements may suffer from poor sensitivity and resolution. To improve the sensitivity of the InSAR data and localize the effects of the near-surface overburden, we employ a Green's function retrieval (GFR) approach that uses time-lapse InSAR data. In this work, we derive the equations and compute the sensitivity between InSAR displacements caused by the reservoir changes with respect to observation points (i.e., virtual sources) at the surface. We present this method with time-lapse InSAR data from an oil field in the San-Joaquin Valley to demonstrate improved resolution of the GFR-InSAR measurements for subsurface imaging and continuous reservoir monitoring with applications to development, production, and subsurface integrity.