Accelerating Mature Field EOR Evaluation Using Machine Learning Uncertainty Workflows Integrating Subsurface And Economics

2021 ◽  
Author(s):  
Mathias Bayerl ◽  
Pascale Neff ◽  
Torsten Clemens ◽  
Martin Sieberer ◽  
Barbara Stummer ◽  
...  
Keyword(s):  
Machine Learning ◽  
Economic Evaluation ◽  
Dynamic Modeling ◽  
Oil Recovery ◽  
History Matching ◽  
Development Planning ◽  
Pattern Generation ◽  
Oil Field ◽  
Production History ◽  
Roll Out

Abstract Field re-development planning for tertiary recovery projects in mature fields traditionally involves a comprehensive subsurface evaluation circle, including static/dynamic modeling, scenario assessment and candidate selection based on economic models. The aforementioned sequential approach is time-consuming and includes the risk of delaying project maturation. This work introduces a novel approach which integrates subsurface geological and dynamic modeling as well as economics and uses machine learning augmented uncertainty workflows to achieve project acceleration. In the elaborated enhanced oil recovery (EOR) evaluation process, a machine learning assisted approach is used in order to narrow geological and dynamic parameter ranges both for model initialization and subsequent history matching. The resulting posterior parameter distributions are used to create the input models for scenario evaluation under uncertainty. This scenario screening comprises not only an investigation of qualified EOR roll-out areas, but also includes detailed engineering such as well spacing optimization and pattern generation. Eventually, a fully stochastic economic evaluation approach is performed in order to rank and select scenarios for EOR implementation. The presented workflow has been applied successfully for a mature oil field in Central/Eastern Europe with 60+ years of production history. It is shown that by using a fully stochastic approach, integrating subsurface engineering and economic evaluation, a considerable acceleration of up to 75% in project maturation time is achieved. Moreover, the applied workflow stands out due to its flexibility and adaptability based on changes in the project scope. In the course of this case study, a sector roll-out of chemical EOR is elaborated, including a proposal for 27 new well candidates and 17 well conversions, resulting in an incremental oil production of 4.7MM bbl. The key findings were: A workflow is introduced that delivers a fully stochastic economic evaluation while honoring the input and measured data.The delivered scenarios are conditioned to the geological information and the production history in a Bayesian Framework to ensure full consistency of the selected subsurface model advanced to forecasting.The applied process results in substantial time reduction for an EOR re-development project evaluation cycle.

scholarly journals Practical CO2—WAG Field Operational Designs Using Hybrid Numerical-Machine-Learning Approaches

Energies ◽  
2021 ◽  
Vol 14 (4) ◽  
pp. 1055
Author(s):  
Qian Sun ◽  
William Ampomah ◽  
Junyu You ◽  
Martha Cather ◽  
Robert Balch
Keyword(s):  
Machine Learning ◽  
Oil Recovery ◽  
History Matching ◽  
Optimization Problems ◽  
Learning Technologies ◽  
Petroleum Engineering ◽  
Support Vector ◽  
Learning Approaches ◽  
Field Development ◽  
Proxy Models

Machine-learning technologies have exhibited robust competences in solving many petroleum engineering problems. The accurate predictivity and fast computational speed enable a large volume of time-consuming engineering processes such as history-matching and field development optimization. The Southwest Regional Partnership on Carbon Sequestration (SWP) project desires rigorous history-matching and multi-objective optimization processes, which fits the superiorities of the machine-learning approaches. Although the machine-learning proxy models are trained and validated before imposing to solve practical problems, the error margin would essentially introduce uncertainties to the results. In this paper, a hybrid numerical machine-learning workflow solving various optimization problems is presented. By coupling the expert machine-learning proxies with a global optimizer, the workflow successfully solves the history-matching and CO2 water alternative gas (WAG) design problem with low computational overheads. The history-matching work considers the heterogeneities of multiphase relative characteristics, and the CO2-WAG injection design takes multiple techno-economic objective functions into accounts. This work trained an expert response surface, a support vector machine, and a multi-layer neural network as proxy models to effectively learn the high-dimensional nonlinear data structure. The proposed workflow suggests revisiting the high-fidelity numerical simulator for validation purposes. The experience gained from this work would provide valuable guiding insights to similar CO2 enhanced oil recovery (EOR) projects.

Conditioning Model Ensembles to Various Observed Data (Field and Regional Level) by Applying Machine-Learning-Augmented Workflows to a Mature Field with 70 Years of Production History

2021 ◽  
pp. 1-18
Author(s):  
Gisela Vanegas ◽  
John Nejedlik ◽  
Pascale Neff ◽  
Torsten Clemens
Keyword(s):  
Machine Learning ◽  
Oil Recovery ◽  
Numerical Models ◽  
Operating Conditions ◽  
Model Parameters ◽  
Large Set ◽  
Model Parameter ◽  
Production History ◽  
Hydrocarbon Fields ◽  
Parameter Distributions

Summary Forecasting production from hydrocarbon fields is challenging because of the large number of uncertain model parameters and the multitude of observed data that are measured. The large number of model parameters leads to uncertainty in the production forecast from hydrocarbon fields. Changing operating conditions [e.g., implementation of improved oil recovery or enhanced oil recovery (EOR)] results in model parameters becoming sensitive in the forecast that were not sensitive during the production history. Hence, simulation approaches need to be able to address uncertainty in model parameters as well as conditioning numerical models to a multitude of different observed data. Sampling from distributions of various geological and dynamic parameters allows for the generation of an ensemble of numerical models that could be falsified using principal-component analysis (PCA) for different observed data. If the numerical models are not falsified, machine-learning (ML) approaches can be used to generate a large set of parameter combinations that can be conditioned to the different observed data. The data conditioning is followed by a final step ensuring that parameter interactions are covered. The methodology was applied to a sandstone oil reservoir with more than 70 years of production history containing dozens of wells. The resulting ensemble of numerical models is conditioned to all observed data. Furthermore, the resulting posterior-model parameter distributions are only modified from the prior-model parameter distributions if the observed data are informative for the model parameters. Hence, changes in operating conditions can be forecast under uncertainty, which is essential if nonsensitive parameters in the history are sensitive in the forecast.

Experimental Design and Evaluation of Surfactant Polymer for a Heavy Oil Field in South of Sultanate of Oman

2021 ◽  
Author(s):  
Ali Reham Al-Jabri ◽  
Rouhollah Farajzadeh ◽  
Abdullah Alkindi ◽  
Rifaat Al-Mjeni ◽  
David Rousseau ◽  
...  
Keyword(s):  
Oil Recovery ◽  
Heavy Oil ◽  
History Matching ◽  
Oil Field ◽  
Reservoir Rock ◽  
Sultanate Of Oman ◽  
Industrial Chemicals ◽  
Oil Viscosity ◽  
Injection Strategy ◽  
Operational Conditions

Abstract Heavy oil reservoirs remain challenging for surfactant-based EOR. In particular, selecting fine-tuned and cost effective chemical formulations requires extensive laboratory work and a solid methodology. This paper reports a laboratory feasibility study, aiming at designing a surfactant-polymer pilot for a heavy oil field with an oil viscosity of ~500cP in the South of Sultanate of Oman, where polymer flooding has already been successfully trialed. A major driver was to design a simple chemical EOR method, to minimize the risk of operational issues (e.g. scaling) and ensure smooth logistics on the field. To that end, a dedicated alkaline-free and solvent-free surfactant polymer (SP) formulation has been designed, with its sole three components, polymer, surfactant and co-surfactant, being readily available industrial chemicals. This part of the work has been reported in a previous paper. A comprehensive set of oil recovery coreflood tests has then been carried out with two objectives: validate the intrinsic performances of the SP formulation in terms of residual oil mobilization and establish an optimal injection strategy to maximize oil recovery with minimal surfactant dosage. The 10 coreflood tests performed involved: Bentheimer sandstone, for baseline assessments on large plugs with minimized experimental uncertainties; homogeneous artificial sand and clays granular packs built to have representative mineralogical composition, for tuning of the injection parameters; native reservoir rock plugs, unstacked in order to avoid any bias, to validate the injection strategy in fully representative conditions. All surfactant injections were performed after long polymer injections, to mimic the operational conditions in the field. Under injection of "infinite" slugs of the SP formulation, all tests have led to tertiary recoveries of more than 88% of the remaining oil after waterflood with final oil saturations of less than 5%. When short slugs of SP formulation were injected, tertiary recoveries were larger than 70% ROIP with final oil saturations less than 10%. The final optimized test on a reservoir rock plug, which was selected after an extensive review of the petrophysical and mineralogical properties of the available reservoir cores, led to a tertiary recovery of 90% ROIP with a final oil saturation of 2%, after injection of 0.35 PV of SP formulation at 6 g/L total surfactant concentration, with surfactant losses of 0.14 mg-surfactant/g(rock). Further optimization will allow accelerating oil bank arrival and reducing the large PV of chase polymer needed to mobilize the liberated oil. An additional part of the work consisted in generating the parameters needed for reservoir scale simulation. This required dedicated laboratory assays and history matching simulations of which the results are presented and discussed. These outcomes validate, at lab scale, the feasibility of a surfactant polymer process for the heavy oil field investigated. As there has been no published field test of SP injection in heavy oil, this work may also open the way to a new range of field applications.

scholarly journals Reservoir Model and Production Strategy of Mishrif Reservoir-Nasryia Oil Field Southern Iraq

2020 ◽  
Vol 10 (3) ◽  
pp. 54-85
Author(s):  
Hamzah Amer Abdulameer ◽  
Dr. Sameera Hamd-Allah
Keyword(s):  
Oil Recovery ◽  
Carbonate Rock ◽  
History Matching ◽  
Water Saturation ◽  
Oil Field ◽  
Neutron Density ◽  
Reservoir Model ◽  
Water Contact ◽  
Medium Permeability ◽  
Production Strategy

Nasryia oil field is located about 38 Km to the north-west of Nasryia city. The field was discovered in 1975 after doing seismic by Iraqi national oil company. Mishrif formation is a carbonate rock (Limestone and Dolomite) and its thickness reach to 170m. The main reservoir is the lower Mishrif (MB) layer which has medium permeability (3.5-100) md and good porosity (10-25) %. Form well logging interpretation, it has been confirmed the rock type of Mishrif formation as carbonate rock. A ten meter shale layer is separating the MA from MB layer. Environmental corrections had been applied on well logs to use the corrected one in the analysis. The combination of Neutron-Density porosity has been chosen for interpretation as it is close to core porosity. Archie equation had been used to calculate water saturation using corrected porosity from shale effect and Archie parameters which are determined using Picket plot. Using core analysis with log data lead to establish equations to estimate permeability and porosity for non-cored wells. Water saturation form Archie was used to determine the oil-water contact which is very important in oil in place calculation. PVT software was used to choose the best fit PVT correlation that describes reservoir PVT properties which will be used in reservoir and well modeling. Numerical software was used to generate reservoir model using all geological and petrophysical properties. Using production data to do history matching and determine the aquifer affect as weak water drive. Reservoir model calculate 6.9 MMMSTB of oil as initial oil in place, this value is very close to that measured by Chevron study on same reservoir which was 7.1 MMMSTB. [1] Field production strategy had been applied to predict the reservoir behavior and production rate for 34 years. The development strategy used water injection to support reservoir pressure and to improve oil recovery. The result shows that the reservoir has the ability to produce oil at apparently stable rate equal to 85 Kbbl/d, also the recovery factor is about 14%.

An Innovative Workflow for the Dynamic Modeling and History Matching of a Mature Complex Carbonate Oil Field in the Middle East

2014 ◽  
Author(s):  
Caroline Tomio ◽  
Jeremie Fernagu ◽  
Luther Thomas Sullivan ◽  
Abdulaziz Rashid Al Naimi
Keyword(s):  
Middle East ◽  
Dynamic Modeling ◽  
History Matching ◽  
Oil Field

Waterflooding Surveillance and Monitoring: Putting Principles Into Practice

2007 ◽  
Vol 10 (05) ◽  
pp. 552-562 ◽  
Author(s):  
Rodolfo Martin Terrado ◽  
Suryo Yudono ◽  
Ganesh C. Thakur
Keyword(s):  
Oil Recovery ◽  
Reservoir Characterization ◽  
Real Life ◽  
Oil Field ◽  
Surveillance Program ◽  
Well Performance ◽  
Primary Input ◽  
Diagnostic Plot ◽  
Decline Rate ◽  
Production History

Summary This paper illustrates how practical application of surveillance and monitoring principles is a key to understanding reservoir performance and identifying opportunities that will improve ultimate oil recovery. Implementation of various principles recommended by industry experts is presented using examples from fields currently in production. Practices in processing valuable information and analyzing data from different perspectives are presented in a methodical way on the following bases: field, block, pattern, and wells. A novel diagnostic plot is presented to assess well performance and identify problem wells for the field. Results from the application of these practices in a pilot area are shared, indicating that the nominal decline rate improved from 33 to 18% per year without any infill drilling. The change in the decline rate is attributed primarily to effective waterflood management with a methodical approach, employing an integrated multifunctional team. Although the suggested techniques can be applied to any oil field undergoing a waterflood, they are of great value to mature waterfloods that involve significant production history. In these cases, prioritization is a key aspect to maintain focus on the opportunities that will add the most value during the final period of the depletion cycle. Case studies illustrating the best surveillance practices are discussed. Introduction Surveillance and monitoring techniques were first discussed in SPE literature in the early 1960s (Kunkel and Bagley 1965). Since then, several highly recognized authors have published related materials (Thakur 1991; Thakur and Satter 1998; Talash 1988; Gulick and McCain 1998; Baker 1997, 1998; SPE Reprint 2003). Industry experts recommend the following valuable principles:The key ingredients of any surveillance program are planning and accurate data collection.To understand reservoir flows and reduce nonuniqueness in interpretations, it is crucial to implement a multilevel surveillance effort.A single technique in isolation is not generally indicative because different parameters can cause similar plot signatures.Controlled waterflooding through the use of pattern balancing requires time and technical efforts —engineering and geological—during the life of the project.Valuable insights into the performance of the waterflood can be gained from individual-well plots such as Hall plots.Surveillance techniques should always be a precursor to in-depth studies, including numerical simulation. A process to consistently evaluate the performance of a reservoir—from field to block to pattern to well level—is discussed with the help of real-life examples. Type plots and maps are used to identify opportunities and promote team discussions to effectively manage a reservoir undergoing waterflood. Production history and basic reservoir characterization serve as primary input variables for the recommended analysis.

Feasibility of Waterflooding for a Carbonate Oil Field Through Whole-Field Simulation Studies

2015 ◽  
Vol 137 (6) ◽  
Author(s):  
Wenting Yue ◽  
John Yilin Wang
Keyword(s):  
Oil Recovery ◽  
Injection Rate ◽  
Carbonate Reservoir ◽  
Oil Field ◽  
Improved Oil Recovery ◽  
Smackover Formation ◽  
Artificial Lift ◽  
Production History ◽  
Current Production ◽  
The Impact

The carbonate oil field studied is a currently producing field in U.S., which is named “PSU” field to remain anonymity. Discovered in 1994 with wells on natural flow or through artificial lift, this field had produced 17.8 × 106 bbl of oil to date. It was noticed that gas oil ratio had increased in certain parts and oil production declined with time. This study was undertaken to better understand and optimize management and operation of this field. In this brief, we first reviewed the geology, petrophysical properties, and field production history of PSU field. We then evaluated current production histories with decline curve analysis, developed a numerical reservoir model through matching production and pressure data, then carried out parametric studies to investigate the impact of injection rate, injection locations, and timing of injection, and finally developed optimized improved oil recovery (OIR) methods based on ultimate oil recovery and economics. This brief provides an addition to the list of carbonate fields available in the petroleum literature and also improved understandings of Smackover formation and similar analogous fields. By documenting key features of carbonated oil field performances, we help petroleum engineers, researchers, and students understand carbonate reservoir performances.

scholarly journals An integrated Shannon Entropy and reference ideal method for the selection of enhanced oil recovery pilot areas based on an unsupervised machine learning algorithm

2021 ◽  
Vol 76 ◽  
pp. 82
Author(s):  
S. Mahdia Motahhari ◽  
Mehdi Rafizadeh ◽  
S. Mahmoud Reza Pishvaie ◽  
Mohammad Ahmadi
Keyword(s):  
Machine Learning ◽  
Enhanced Oil Recovery ◽  
Shannon Entropy ◽  
Oil Recovery ◽  
Oil Field ◽  
Unsupervised Machine Learning ◽  
Main Challenge ◽  
Ideal Method ◽  
Selection Of ◽  
Economic Criteria

Pilot-scale enhanced oil recovery in hydrocarbon field development is often implemented to reduce investment risk due to geological uncertainties. Selection of the pilot area is important, since the result will be extended to the full field. The main challenge in choosing a pilot region is the absence of a systematic and quantitative method. In this paper, we present a novel quantitative and systematic method composed of reservoir-geology and operational-economic criteria where a cluster analysis is utilized as an unsupervised machine learning method. A field of study will be subdivided into pilot candidate areas, and the optimized pilot size is calculated using the economic objective function. Subsequently, the corresponding Covariance (COV) matrix is computed for the simulated 3-D reservoir quality maps in the areas. The areas are optimally clustered to select the dominant cluster. The operational-economic criteria could be applied for decision making as well as the proximity of each area to the center of dominant cluster as a geological-reservoir criterion. Ultimately, the Shannon entropy weighting and the reference ideal method are applied to compute the pilot opportunity index in each area. The proposed method was employed for a pilot study on an oil field in south west Iran.

Sign in / Sign up

Export Citation Format

Share Document