Petroleum Engineering Problems Research Articles

Review of deep learning algorithms in molecular simulations and perspective applications on petroleum engineering

In the last few decades, deep learning (DL) has afforded solutions to macroscopic problems in petroleum engineering, but mechanistic problems at the microscale have not benefited from it. Mechanism studies have been the strong demands for the emerging projects, such as the gas storage and hydrate production, and for some problems encountered in the storage process, which are common found as the chemical interaction between injected gas and mineral, and the formation of hydrate. Emerging advances in DL technology enable solving molecular dynamics (MD) with quantum accuracy. The conventional quantum chemical method is computational expensive, whereas the classical MD method cannot guarantee high accuracy because of its empirical force field parameters. With the help of the DL force field, precision at the quantum chemistry level can be achieved in MD. Moreover, the DL force field promotes the computational speed compared with first-principles calculations. In this review, the basic knowledge of the molecular force field and deep neural network (DNN) is first introduced. Then, three representative open-source packages relevant to the DL force field are introduced. As the most common components in the development of oil and gas reservoirs, water and methane are studied from the aspects of computational efficiency and chemical reaction respectively, providing the foundation of oil and gas researches. However, in the oil and gas problems, the complex molecular topo structures and various element types set a high challenge for the DL techniques in MD. Regarding the computational efficiency, it needs improvement via GPU and parallel accelerations to compete with classical MD. Even with such difficulties, the booming of this technique in the area of petroleum engineering can be predictable.

Corrosion law of cement paste under hydrogen sulfide conditions in natural gas wells

Downhole anticorrosion in sour gas (H2S or CO2) wells is one of the technical problems in petroleum engineering, and the corrosion law of cement paste, which is the “first barrier of the wellbore,” needs to be focused on. Aiming at the problems existing in the current research on the corrosion of underground cement paste, a curing method for interfacial corrosion is proposed. X-ray diffraction and scanning electron microscopy are used to investigate the corrosion mechanism of cement paste cured under hydrogen sulfide (H2S) conditions in natural gas wells. Experimental results showed that the corrosion depth of cement paste is proportional to the partial pressure of H2S and the corrosion time, and the compressive strength of cement paste after corrosion is inversely proportional to the H2S partial pressure value and the corrosion time. Due to the gradual enrichment, accumulation, and migration of products after the cement paste is corroded by H2S, the cement paste forms a relatively stable dense layer or corrosion transition zone. The porosity and permeability of cement paste after corrosion increased with corrosion time, showing the characteristics of first increasing and then decreasing and finally making it more difficult for the corrosive medium to enter the interior of the cement paste. It is an important method stable corrosion transition zone forms as soon as possible, which is important to maintain the long-term sealing and chemical integrity of the cement sheaths in natural gas wells containing H2S.

Intelligent modeling with physics-informed machine learning for petroleum engineering problems

The advancement in big data and artificial intelligence has enabled a novel exploration mode for the study of petroleum engineering. Unlike theory-based solution methods, the data-driven intelligent approaches demonstrate superior flexibility, computational efficiency and accuracy for dealing with complex multi-scale, and multi-physics problems. However, these intelligent models often disregard physical laws in pursuit of error minimization, which leads to certain uncertainties. Therefore, physics-informed machine learning approaches have been developed based on data, guided by physics, and supported by machine learning models. This study summarizes four embedding mechanisms for introducing physical information into machine learning models, including input databased embedding, model architecture-based embedding, loss function-based embedding, and model optimization-based embedding mechanism. These “data + physics” dualdriven intelligent models not only exhibit higher prediction accuracy while adhering to physic laws, but also accelerate the convergence to improve computational efficiency. This paradigm will facilitate the guide developments in solving petroleum engineering problems toward a more comprehensive and efficient direction. Cited as: Xie, C., Du, S., Wang, J., Lao, J., Song, H. Intelligent modeling with physics-informed machine learning for petroleum engineering problems. Advances in Geo-Energy Research, 2023, 8(2): 71-75. https://doi.org/10.46690/ager.2023.05.01

Towards automatic and accurate core-log processing

The analysis of rocks plays an important role in geological and petroleum-engineering problems. In these tasks, core (i.e., core-log) is a crucial element since it provides underlying information on the geophysical properties of the area. Thereby, works often leverage core data to assign the correct category from well-logs. Unfortunately, processing core is laborious and time-consuming; hence, its analysis takes long periods. For example, for a single well, a human could have to adjust thousands of core data. Besides, due to its nature, log analysis must handle noise and missing data. Given these issues, our goal is to propose an automatic (e.g., without any human intervention) and accurate core processing using the gamma-ray. To achieve this goal and additionally demonstrate the most promising pattern recognition strategies, we assess the effectiveness of several models to diagnostic the core-log. Such models include simple regressions, ensembles, gradient boosting, recurrence models and the relatively recent Transformer network. Our comprehensive evaluation is different from existing works on geoscience tasks, which evaluate a small number of models or variations of a single model. Particularly, we evaluate more than 200 unique models (i.e., models with different hyperparameters) and observe that deep learning techniques outperform other techniques by a large margin. Furthermore, we compare the compromise between predictive ability and computational cost of several models – a reliable information for real-time lithology, which is often overlooked by previous works. According to our results, we can effectively replace the manual (i.e., by a human expert – geologist) diagnostic of core by pattern recognition methods, mainly by the Transformer model, as it aligns the cores in accordance (i.e., in the same direction) with the geologist. More specifically, given the core and gamma-ray as input, the Transformer model outputs an adjusted core obtaining an R2 of 93,63, which indicates a fine-grained adjustment. We empirically demonstrate that the success behind Transform is its effectiveness in expressing large sequences of core and well-log. On the other hand, other models such as RNN, LSTM and GRU meet collapse when processing large sequences of core-log. We further confirm that Transformers are among the top-performance models and often surpass recurrence models on several geoscience applications such as lithology and log-shape classification, and prediction of oil production. Regarding the first task, the models successfully classify different facies categories such as coarse sandstone, medium sandstone, fine sandstone, siltstone, dolomite, limestone and mudstone. In these tasks, Transformer outperforms the widely-employed LSTM by up to 14 percentage points. Our empirical observations encourage the exploration of multiple models and suggest Transformers as strong baselines for future research on geoscience tasks. In this direction, we release all data and trained models used throughout the work. To the best of our knowledge, we are the first study exploring Transformer models on geoscience applications.

Multi-objective optimization of petroleum engineering problems using a hybrid workflow: Combination of particle swarm optimization, fuzzy logic, imperialist competitive algorithm and response surface methodology

Optimization techniques are used to find the strategies for chemically enhanced oil recovery in a sandstone reservoir. This research develops a multi-objective optimization methodology by combining experimental design methods and artificial intelligence techniques. The capability of this hybrid artificial intelligence methodology is evaluated in the optimal design of control variables to achieve the highest performance of a surfactant/polymer injection project into a sandstone reservoir. In the first step, a two-level full factorial design is used to screen initial control variables. Thereafter a response surface methodology (RSM) is employed to optimize the RF and NPV of a CEOR application. The neuro-simulation technique provides the required outputs for screening and RSM designs. The performance of network is improved using the imperialist competitive algorithm (ICA). Having precise fitness functions, multi-attribute optimization was performed using particle swarm optimization (PSO) and fuzzy logic (FL). This paper discusses the advantages of different perspectives over single-objective approaches. Using the RF-objective PSO algorithm, RF exceeded 64% of original oil in place (OOIP), while the profit of project slumped to $5.90 MM. On the other hand, NPV-attribute PSO increased NPV to $8.48 MM. Meanwhile, RF, as the technical success of the project, plunged to less than 53% OOIP. However, the proposed multi-objective algorithm increased RF to 57% OOIP with NPV of $8.11 MM, solving the trade-off between technical and economic terms. The results of this study indicate the efficacy of proposed hybrid workflow for multi-attribute decision-making of CEOR field implementation.

Developments and applications of the CFD-DEM method in particle–fluid numerical simulation in petroleum engineering: A review

Particle and fluid flow is a general problem existing in a series of engineering activities during oil and gas development. Many efforts have been made to investigate the migration, deposition, and plugging characteristics of particles carried by the fluid medium. The discrete element method (DEM) is used for the particle phase, the computational fluid dynamic (CFD) is applied for the fluid phase and many coupling schemes are introduced for interactions between phases. The CFD-DEM studies have been conducted to help mechanism explanation and flow prediction under various work conditions. This work reviews the progress in CFD-DEM coupling simulations about oil and gas development by summarizing the models and relevant applications. Different coupling schemes and their applicable petroleum engineering problems are reviewed. The needs for future research are also discussed. This will be helpful for researchers who are new to particle flow numerical simulation in the CFD-DEM method to shorten the learning cycle and improve the efficiency of problem solving.

Sign-preserving second-order IMPEC time discretization and its application in compressible miscible displacement with Darcy-Forchheimer models

In this paper, we construct sign-preserving second-order IMplicit Pressure Explicit Concentration (IMPEC) time methods for generalized coupled non-Darcy flow and transport problems in petroleum engineering, extending the algorithm given in [10] which is only applicable to Darcy flows. We use interior penalty discontinuous Galerkin (IPDG) methods for spatial discretization, and develop bound-preserving technique to obtain physically relevant numerical approximations. The sign-preserving second-order IMPEC method is an important innovation. The method is based on the framework of the second-order strong-stability-preserving Runge-Kutta (SSP-RK2) method. The basic idea is to treat the pressure equation implicitly and the concentration equation explicitly so as to obtain a first-order time integration. Then we introduce a correction stage to compensate the accuracy, maintaining the physical bounds of the numerical cell averages. Unfortunately, the above algorithm is not applicable to non-Darcy problems. There are two main difficulties. Firstly, since the velocity equation is nonlinear and time-independent, all variables in the equation must be calculated at the same time level. However, the treatment in [10] will yield an extremely complicated algorithm and significantly large computational cost, and some iterations whose convergence may not be available if the solutions are not smooth. In our scheme, we linearize the velocity equation and the numerical solutions are not at the same time level, leading to first-order accurate solutions. Therefore, we adopt a completely different approach from [10] to derive the correction stage, keeping the physical bounds of the numerical solutions. Secondly, though with the new correction stage, it is still not easy to solve velocity equations in some stages. In this paper, we construct a direct solver for the velocity equation to save computational cost. Numerical experiments will be given to demonstrate that the improved sign-preserving second-order IMPEC scheme can reduce the computational cost significantly compared with explicit schemes if the diffusion coefficient D is small in the concentration equation. The proposed method also yields much larger CFL number compared with first-order IMPEC schemes. Moreover, the effectiveness of the bound-preserving technique will also be verified.

An integrated framework based on deep learning algorithm for optimizing thermochemical production in heavy oil reservoirs

The development of heavy oil reservoirs with active edge and bottom water is one of the most challenging problems in petroleum engineering. In response to the limited thermal recovery of these reservoirs, a multi-phase and multi-component numerical simulation model for thermal and chemical recovery is proposed. An edge-water assisted chemical flooding (EAC flooding) is proposed, which can improve oil displacement efficiency and sweep efficiency by rational utilization of edge-water energy when compounding multi-component chemical system. Then, a deep reinforcement learning algorithm is proposed to predict dynamic production parameters and determine the optimal working system to maximize the oil recovery according to the above mathematical model. The deep reinforcement learning (DRL) model can predict the dynamic production curves according to given states with optimal strategy. At the same time, the proposed model can determine the best conversion timing from cyclic steam stimulation to EAC flooding. Finally, the DRL model can automatically obtain the optimal working system, effectively improving the oil recovery while considering the economic benefits. Thus, the DRL model can solve traditional numerical simulation's time-consuming and labor-intensive challenges and accurately give the optimal working system for developing heavy oil reservoirs with edge water in the field.

A new vapor-liquid-asphaltene three-phase equilibrium computation algorithm based on the free-asphaltene assumption

Asphaltene precipitation is one of the major flow-assurance problems in petroleum engineering and can occur at different stages of oil production due to pressure/temperature changes or compositional changes. It can cause formation damage in the reservoir and choking of oil tubing in the wellbore, thus leading to the reduction of production rates. Accurately predicting under what conditions asphaltene precipitation is a prerequisite to the effective mitigation of the flow assurance problems related to asphaltene precipitation. In this work, a new three-phase vapor-liquid-asphaltene precipitation algorithm is developed based on Li and Li's work (Li and Li, 2019) using the free-asphaltene assumption, in which the asphaltene phase is considered as a liquid phase that only contains asphaltene component (Nascimento et al., 2018). The algorithm is validated by comparing the calculation results with the asphaltene precipitation data of five different oil samples. The tuned parameters for oil sample 3 and oil sample 4 are then used to predict asphaltene precipitation with different gas injections. The calculation results agree well with the experimental data. It indicates that the three-phase VLA equilibrium calculation algorithm can make reliable predictions of the PT and Px phase diagrams with different gas injections as well as the asphaltene precipitation amount. The calculation results yielded by the free-asphaltene assumption are also compared with the results yielded by the solid assumption and the liquid assumption. The algorithm developed using the free-asphaltene assumption is shown to be able to perform slightly better than the algorithm developed using the solid assumption, and perform as well as the algorithm developed using the liquid assumption.

Multi-Objective Optimization of Petroleum Engineering Problems Using a Hybrid Workflow: Combination of Combination of Particle Swarm Optimization, Fuzzy Logic, Imperialist Competitive Algorithm and Response Surface Methodology

Multi-Objective Optimization of Petroleum Engineering Problems Using a Hybrid Workflow: Combination of Combination of Particle Swarm Optimization, Fuzzy Logic, Imperialist Competitive Algorithm and Response Surface Methodology

Undrained cylindrical cavity expansion/contraction in stiff clays using a two‐surface plasticity model

AbstractThis paper presents a semi‐analytical solution of undrained cylindrical cavity expansion/contraction problem using a two‐surface plasticity model for natural stiff clays from the view point of poromechanics. The present solution reduces the classical boundary problem into a group of seven first‐order ordinary differential equations via an auxiliary independent variable ξ under undrained conditions. It strictly follows the premise of constant soil mass (CSM). Meanwhile, the solution for the constant soil volume (CSV) problem is used for comparison. It shows that CSM assumption is of paramount importance in developing solutions under high stress condition, while CSV induces non‐negligible errors in effective stress and pore pressure distributions. Based on the premise of CSM, extensive parametric studies on cavity expansion/contraction are undertaken to investigate the effects of the model key parameters on the stress components distribution, excess pore pressure distribution, and cavity expansion curves. The present solution could be useful while dealing with geotechnical and petroleum engineering problems involving stiff clays or soft claystones. The application in Praclay Gallery excavation in natural Boom clay well demonstrates the validity of the present solution.

Identification of non-Darcian flow effect in double-porosity fractured aquifer based on multi-well pumping test

Well testing in double-porosity fractured aquifers or oil and gas reservoirs is one of the long-lasting research problems in subsurface hydrology and petroleum engineering, where the double-porosity implies that the media of concern can be approximated as a two interrelated continuums (fracture network and rock matrix) with two distinctively different porosities. However, most of those studies in double-porosity fractured aquifers only concern Darcian flow in the fractured continuum. In this study, we will expand such studies from Darcian flow regime to non-Darcian flow regime, which is most likely to be the case for the field well testing site reported in this investigation. New solutions based on power law and a linearization approximation are obtained for hydraulic heads in Laplace domain and subsequently inverted to give spatiotemporal distributions in real-time domain. The non-Darcian flow model in single porosity media is obtained by setting the leakage coefficient (C) to 0 or the storage coefficient of the matrix (Sm) to 0, where C is the rate of fluid transfer between the fracture and rock matrix. Parameter analysis is conducted using dimensionless formats. A larger dimensionless quasi conductivity coefficient KqD, a larger drawdown in the early stage, and a smaller drawdown in the late stage. The dimensionless parameters CD and ϕ influence the drawdown in transitional state, and the values get larger, the drawdown in transitional state drop smaller. A pumping test conducted in Huangtun, Anhui province of China has been applied to test the non-Darcian flow effect in a double-porosity aquifer, in which a particle swarm optimization (PSO) algorithm will be applied here to seek the optimal hydraulic parameters. As the result shows, the observed drawdown data fit well with the new model in the pumping stage. The predicted recovery drawdown curve with the calculated parameters of the new model also performs well with the field drawdown data, which supports the applicability of the new model to interpret the field data.

An improved vapor-liquid-asphaltene three-phase equilibrium computation algorithm

Asphaltene precipitation is one of the major flow-assurance problems in petroleum engineering. Asphaltene precipitation in the reservoirs can lead to formation damage, while asphaltene precipitation in the production tubing can reduce the production rate and even plug the production tubing. It may occur to the naturally produced oil experiencing pressure/temperature changes or during gas injection processes (e.g., CO2 flooding) for enhanced oil recovery. It is a prerequisite to accurately predicting under what conditions asphaltene precipitation will occur and how much precipitates if it occurs in order to better predict and control flow assurance problems due to asphaltene precipitation. In this work, a robust three-phase vapor-liquid-asphaltene precipitation algorithm is developed. To make the algorithm more robust and effective, the new algorithm is built based upon two previously proven works: the asphaltene precipitation model proposed by Kohse et al. (1993) which considers the effects of pressure, temperature and gas injection on asphaltene precipitation, and the initialization method and algorithm developed by Li and Li (2019) which does not consider the effect of temperature on asphaltene precipitation. The algorithm is validated using the asphaltene precipitation data measured by Jamaluddin et al. (2000 and 2002) and Gonzalez et al. (2004). As for the first two oil samples (Jamaluddin et al., 2000 and 2002), both the onset conditions and the amounts of asphaltene precipitation are computed with the consideration of varied pressure-temperature conditions and the injection of different gases. The pressure-temperature (PT) phase diagrams and the pressure-composition (Px) phase diagrams are drawn to illustrate the calculation results. Our algorithm is shown to be converged at every test point involved in the three-phase equilibrium calculations for constructing the PT and Px phase diagrams. After being calibrated using the measured asphaltene onset data for the second and third oil samples, the three-phase VLS equilibrium calculation algorithm is shown to be capable of making reliable predictions of the PT phase diagrams with gas injection and the precipitated asphaltene amounts versus injectant concentration.

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

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.

An Integrated Framework Based on Deep Reinforcement Learning Algorithm for Optimizing Thermal and Chemical Production in Edge-Water Heavy Oil Reservoirs

The development of heavy oil reservoirs with active edge and bottom water is one of the most challenging problems in petroleum engineering. In response to the limited thermal recovery of these reservoirs, a multi-phase and multi-component numerical simulation model for thermal and chemical recovery is proposed. An edge-water assisted chemical flooding (EAC flooding) is proposed, which can improve oil displacement efficiency and sweep efficiency by rational utilization of edge-water energy when compounding multicomponent chemical system. Then, a deep reinforcement learning algorithm, ActorCritic structure, is proposed to predict dynamic production parameters and determine the optimal working system to maximize the oil recovery according to the above mathematical model. The deep reinforcement learning (DRL) model can predict the dynamic production curves and estimate the maximum oil recovery. At the same time, the proposed model can determine the best conversion timing from cyclic steam stimulation to EAC flooding. Finally, the DRL model can automatically obtain the optimal working system, effectively improving the oil recovery while considering the economic benefits. Thus, the DRL model can solve traditional numerical simulation’s time-consuming and labor-intensive challenges and accurately give the optimal working system for developing heavy oil reservoirs with edge water in the field.

As Industry Changes, So Does Petroleum Engineering Education

As the oil and gas industry becomes increasingly complex, the requirements for petroleum engineers—and petroleum engineering education—are changing. “Petroleum engineering jobs in the future are likely to be smaller in number and radically different from those of today,” wrote Nathan Meehan, president of Gaffney, Cline and Associates and 2016 SPE President, in paper SPE•194746, The End of Petroleum Engineering As We Know It. “The next generation of petroleum engineers will have to address demands for sustainability, lower carbon intensity, and needs for radical productivity improvements, which only artificial intelligence (AI) and digital can drive. This suggests that we will need to revisit university education for petroleum engineers and all aspects of career development and•training.” Petroleum engineering education reflects the E&P industry it serves. Training of competent graduates with both the domain and digital knowledge to immediately contribute when they join the industry has become more complicated as expectations change. What are petroleum engineering schools doing to adapt to these changes, and how are they helping their students navigate a future career in the oil and gas industry? Balancing Theory and Practical Application A survey by Ryder Scott Petroleum Consultants identified the need to modify the skills and knowledge currently taught in academic institutions during undergraduate study. Ryder Scott surveyed clients who were industry managers or supervisors with direct experience with newly minted petroleum engineering graduates (fewer than 5 years of experience). The survey asked participants—engineering managers in the oil and gas reserves sector—their opinions regarding the preparedness of recent graduates entering the workforce (Fig. 1). Generally, the respondents were satisfied with the amount of theory being taught, but not with the amount of practical application of petroleum engineering principles. The consensus was that curricula should be kept the same in terms of courses, but modernized to include the practical application of new technologies. This presents a significant challenge for engineering education programs whose already-full curricula to meet requirements of the Accreditation Board of Engineering and Technology (ABET) leave little room for additional coursework, labs, or field work. The current requirement for a BS degree in petroleum engineering at most schools is 130-plus credit hours. Yet, in addition to fundamentals, today’s graduates are expected to be fluent in data analytics, machine learning, and data sciences and to understand concepts such as cyber security and physical security. In addition, many are expected to use their subsurface engineering skills to plan and design carbon sequestration solutions. “We teach students the fundamentals of petroleum engineering,” said Jeff Spath, SPE 2014 President and head of the Texas A&M University Petroleum Engineering Department. “But more importantly, we teach them to solve petroleum engineering problems.”

A Correlation for Estimating the Biot Coefficient

SummaryThe objective of this paper is to develop an easy-to-use correlation for estimating the Biot coefficient. This is important because the Biot coefficient plays an important role in solving many practical petroleum-engineering problems, including, for example, the design of hydraulic-fracturing jobs and the estimation of in-situ closure stress on proppant.The procedure for developing the proposed empirical correlation uses data from various lithologies including limestone, sandstone, shale, marble, and granite. Thus, the correlation has application in conventional and unconventional petroleum reservoirs. The use of the correlation requires knowledge of permeability and porosity, the data commonly available in petroleum engineering (on the other hand, the Biot coefficient data are almost never available). The ratios of permeability to porosity (commonly known as process or delivery speed) and pore-throat radii (rp35) are entered for estimating the Biot coefficient from the correlation proposed in this paper. The correlation is useful in those cases where sophisticated experimental work needed for estimating the Biot poroelastic coefficient is not available.Testing against various data sets indicates that the proposed correlation provides reasonable results. In the past, methods with different complexity levels have been used for estimating the Biot coefficient. These have included, for example, (1) a method that requires the knowledge of bulk modulus of the rock mineral and bulk modulus of the skeleton with no fluids in it, parameters that are not usually available for petroleum reservoirs; (2) a method that is based on knowledge of only porosity; (3) a method that is based on the knowledge of only permeability; and (4) an approach that simply assumes that the Biot coefficient is equal to 1.0 or some other number. The proposed correlation falls somewhere in the middle. It is not as simple as saying that the Biot coefficient is equal to unity or saying that it depends only on porosity, or only on permeability. On the other hand, it is not as complex as requiring sophisticated laboratory work of the type mentioned in (1) above.The novelty of this work is the development of an original easy-to-use correlation for estimating the Biot coefficient in conventional and unconventional (tight and shale) reservoirs on the basis of knowledge of the ratio of permeability to porosity (k/ϕ) and the pore-throat radius (rp35). The correlation is developed in such a way that it also has application for estimating the Biot coefficient in the case of unconsolidated petroleum reservoirs and oil sands. The overall approach allows the integration of geomechanics with flow units, geology, petrophysics, and reservoir engineering (RE).

Comparative Analysis of Static and Dynamic Mechanical Behavior for Dry and Saturated Cement Mortar.

Deformational and breakage behaviors of concrete and cement mortar greatly influence various engineering structures, such as dams, river bridges, ports, tunnels, and offshore rig platforms. The mechanical and petrophysical properties are very sensitive to water content and are controlled by the liquid part in pore spaces to a large extent. The objective of this paper is to investigate the water saturation effect on the strength characteristics and deformability of cement mortar under two loading conditions, static and dynamic compression. A set of cement mortar samples was prepared and tested to study the mechanical behavior in dry and saturated states. The first part of the research incorporates the study of static mechanical properties for dry and brine-saturated cement mortar through uniaxial compressive strength tests (UCS). Second, drop-weight impact experiments were carried out to study the dynamic mechanical properties (impact resistance, deformation pattern, and fracture geometry) for dry and saturated cases. The comparative analysis revealed that water saturation caused substantial changes in compressive strength and other mechanical characteristics. Under static loading, water saturation caused a reduction in strength of 36%, and cement mortar tended to behave in a more ductile manner as compared to dry samples. On the contrary, under dynamic loading conditions, water saturation resulted in higher impact resistance and fracture toughness as compared to dry conditions. In addition, fractures could propagate to smaller depths as compared to dry case. The study will help resolve many civil, mining, and petroleum engineering problems where cement structures undergo static as well as dynamic compression, especially in a hydraulic environment where these structures interact with the water frequently. To the best of our knowledge, the effect of water saturation on the dynamic mechanical properties of cement mortar has not been well understood and reported in the literature.

Integrated Workflow Approach for Petroleum Engineering Problems

The author of a new book describes the challenges faced in making calculations for hydrocarbon exploration and extraction and how an integrated workflow approach can reduce uncertainties.

Efficient big data assimilation through sparse representation: A 3D benchmark case study in petroleum engineering.

Data assimilation is an important discipline in geosciences that aims to combine the information contents from both prior geophysical models and observational data (observations) to obtain improved model estimates. Ensemble-based methods are among the state-of-the-art assimilation algorithms in the data assimilation community. When applying ensemble-based methods to assimilate big geophysical data, substantial computational resources are needed in order to compute and/or store certain quantities (e.g., the Kalman-gain-type matrix), given both big model and data sizes. In addition, uncertainty quantification of observational data, e.g., in terms of estimating the observation error covariance matrix, also becomes computationally challenging, if not infeasible. To tackle the aforementioned challenges in the presence of big data, in a previous study, the authors proposed a wavelet-based sparse representation procedure for 2D seismic data assimilation problems (also known as history matching problems in petroleum engineering). In the current study, we extend the sparse representation procedure to 3D problems, as this is an important step towards real field case studies. To demonstrate the efficiency of the extended sparse representation procedure, we apply an ensemble-based seismic history matching framework with the extended sparse representation procedure to a 3D benchmark case, the Brugge field. In this benchmark case study, the total number of seismic data is in the order of . We show that the wavelet-based sparse representation procedure is extremely efficient in reducing the size of seismic data, while preserving the salient features of seismic data. Moreover, even with a substantial data-size reduction through sparse representation, the ensemble-based seismic history matching framework can still achieve good estimation accuracy.