Data-driven approaches and AI algorithms are promising enough to be relied on even more than physics-based methods; their main feed is data which is the fundamental element of each phenomenon. These algorithms learn from data and unveil unseen patterns out of it. The petroleum industry as a realm where huge volumes of data are generated every second is of great interest to this new technology. As the oil and gas industry is in the transition phase to oilfield digitization, there has been an increased drive to integrate data-driven modeling and machine learning algorithms in different petroleum engineering challenges. ML has been widely used in different areas of the industry. Many extensive studies have been devoted to exploring AI applicability in various disciplines of this industry; however, lack of two main features is noticeable. Most of the research is either not practical enough to be applicable in real-field challenges or limited to a specific problem and not generalizable. Attention must be given to data itself and the way it is classified and stored. Although there are sheer volumes of data coming from different disciplines, they reside in departmental silos and are not accessible by consumers. In order to derive as much insight as possible out of data, the data needs to be stored in a centralized repository from where the data can be readily consumed by different applications.
The application of artificial intelligence (AI) has become inevitable in the petroleum industry. In drilling and completion engineering, AI is regarded as a transformative technology that can lower costs and significantly improve drilling efficiency (DE). In recent years, numerous studies have focused on intelligent algorithms and their application. Advanced technologies, such as digital twins and physics-guided neural networks, are expected to play roles in drilling and completion engineering. However, many challenges remain to be addressed, such as the automatic processing of multi-source and multi-scale data. Additionally, in intelligent drilling and completion, methods for the fusion of data-driven and physicsbased models, few-sample learning, uncertainty modeling, and the interpretability and transferability of intelligent algorithms are research frontiers. Based on intelligent application scenarios, this study comprehensively reviews the research status of intelligent drilling and completion and discusses key research areas in the future. This study aims to enhance the berthing of AI techniques in drilling and completion engineering.
Despite the advances that have been made in renewable energy over the past decade, crude oil or petroleum remains one of the most important energy resources to the world. Petroleum production presents many challenging issues, such as the destabilization of complex oil–water emulsions, fouling phenomena on pipelines and other facilities, and water treatment. These problems are influenced by the molecular forces at the oil/water/solid/gas interfaces involved in relevant processes. Herein, we present an overview of recent advances on probing the interfacial forces in several petroleum production processes (e.g., bitumen extraction, emulsion stabilization and destabilization, fouling and antifouling phenomena, and water treatment) by applying nanomechanical measurement technologies such as a surface forces apparatus (SFA) and an atomic force microscope (AFM). The interaction forces between bitumen and mineral solids or air bubbles in the surrounding fluid media determine the bitumen liberation and flotation efficiency in oil sands production. The stability of complex oil/water emulsions is governed by the forces between emulsion drops and particularly between interface-active species (e.g., asphaltenes). Various oil components (e.g., asphaltenes) and emulsion drops interact with different substrate surfaces (e.g., pipelines or membranes), influencing fouling phenomena, oil–water separation, and wastewater treatment. Quantifying these intermolecular and interfacial forces has advanced the mechanistic understanding of these interfacial interactions, facilitating the development of advanced materials and technologies to solve relevant challenging issues and improve petroleum production processes. Remaining challenges and suggestions on future research directions in the field are also presented.
The commercial exploitation of unconventional petroleum resources (e.g., shale oil/gas and tight oil/gas) has drastically changed the global energy structure within the past two decades. Sweet-spot intervals (areas), the most prolific unconventional hydrocarbon resources, generally consist of extraordinarily high organic matter (EHOM) deposits or closely associated sandstones/carbonate rocks. The formation of sweet-spot intervals (areas) is fundamentally controlled by their depositional and subsequent diagenetic settings, which result from the coupled sedimentation of global or regional geological events, such as tectonic activity, sea level (lake level) fluctuations, climate change, bottom water anoxia, volcanic activity, biotic mass extinction or radiation, and gravity flows during a certain geological period. Black shales with EHOM content and their associated high-quality reservoir rocks deposited by the coupling of major geological events provide not only a prerequisite for massive hydrocarbon generation but also abundant hydrocarbon storage space. The Ordovician–Silurian Wufeng–Longmaxi shale of the Sichuan Basin, Devonian Marcellus shale of the Appalachian Basin, Devonian–Carboniferous Bakken Formation of the Williston Basin, and Triassic Yanchang Formation of the Ordos Basin are four typical unconventional hydrocarbon systems selected as case studies herein. In each case, the formation of sweet-spot intervals for unconventional hydrocarbon resources was controlled by the coupled sedimentation of different global or regional geological events, collectively resulting in a favorable environment for the production, preservation, and accumulation of organic matter, as well as for the generation, migration, accumulation, and exploitation of hydrocarbons. Unconventional petroleum sedimentology, which focuses on coupled sedimentation during dramatic environmental changes driven by major geological events, is key to improve the understanding of the formation and distribution of sweet-spot intervals (areas) in unconventional petroleum systems.
Stabilizing global climate change to within 1.5 °C requires a reduction in greenhouse gas emissions, with a primary focus on carbon dioxide (CO2) emissions. CO2 flooding in oilfields has recently been recognized as an important way to reduce CO2 emissions by storing CO2 in oil reservoirs. This work proposes an advanced CO2 enhanced oil recovery (EOR) method—namely, storage-driven CO2 EOR—whose main target is to realize net-zero or even negative CO2 emissions by sequestrating the maximum possible amount of CO2 in oil reservoirs while accomplishing the maximum possible oil recovery. Here, dimethyl ether (DME) is employed as an efficient agent in assisting conventional CO2 EOR for oil recovery while enhancing CO2 sequestration in reservoirs. The results show that DME improves the solubility of CO2 in in situ oil, which is beneficial for the solubility trapping of CO2 storage; furthermore, the presence of DME inhibits the “escape” of lighter hydrocarbons from crude oil due to the CO2 extraction effect, which is critical for sustainable oil recovery. Storage-driven CO2 EOR is superior to conventional CO2 EOR in improving sweeping efficiency, especially during the late oil production period. This work demonstrates that storage-driven CO2 EOR exhibits higher oil-in-place (OIP) recovery than conventional CO2 EOR. Moreover, the amount of sequestrated CO2 in storage-driven CO2 EOR exceeds the amount of emissions from burning the produced oil; that is, the sequestrated CO2 offsets not only current emissions but also past CO2 emissions. By altering developing scenarios, such as water alternating storage-driven CO2 EOR, more CO2 sequestration and higher oil recovery can be achieved. This work demonstrates the potential utilization of DME as an efficient additive to CO2 for enhancing oil recovery while improving CO2 storage in oil reservoirs.
Deep coal seams are one of the world's most widespread deposits for carbon dioxide (CO2) disposal and are generally located near large point sources of CO2 emissions. The injection of CO2 into coal seams has great potential to sequester CO2 while simultaneously enhancing coalbed methane (CO2-ECBM) recovery. Pilot tests of CO2-ECBM have been conducted in coal seams worldwide with favorable early results. However, one of the main technical barriers in coal seams needs to be resolved: Injecting CO2 reduces coal permeability and well injectivity. Here, using in situ synchrotron X-ray microtomography, we provide the first observational evidence that injecting nitrogen (N2) can reverse much of this lost permeability by reopening fractures that have closed due to coal swelling induced by CO2 adsorption. Our findings support the notion that injecting minimally treated flue gas—a mixture of mainly N2 and CO2—is an attractive alternative for ECBM recovery instead of pure CO2 injection in deep coal seams. Firstly, flue gas produced by power plants could be directly injected after particulate removal, thus avoiding high CO2-separation costs. Secondly, the presence of N2 makes it possible to maintain a sufficiently high level of coal permeability. These results suggest that flue-gas ECBM for deep coal seams may provide a promising path toward net-zero emissions from coal mines.
We present a framework that couples a high-fidelity compositional reservoir simulator with Bayesian optimization (BO) for injection well scheduling optimization in geological carbon sequestration. This work represents one of the first at tempts to apply BO and high-fidelity physics models to geological carbon storage. The implicit parallel accurate reservoir simulator (IPARS) is utilized to accurately capture the underlying physical processes during CO2 sequestration. IPARS provides a framework for several flow and mechanics models and thus supports both stand-alone and coupled simulations. In this work, we use the compositional flow module to simulate the geological carbon storage process. The compositional flow model, which includes a hysteretic three-phase relative permeability model, accounts for three major CO2 trapping mechanisms: structural trapping, residual gas trapping, and solubility trapping. Furthermore, IPARS is coupled to the International Business Machines (IBM) Corporation Bayesian Optimization Accelerator (BOA) for parallel optimizations of CO2 injection strategies during field-scale CO2 sequestration. BO builds a probabilistic surrogate for the objective function using a Bayesian machine learning algorithm—the Gaussian process regression, and then uses an acquisition function that leverages the uncertainty in the surrogate to decide where to sample. The IBM BOA addresses the three weaknesses of standard BO that limits its scalability in that IBM BOA supports parallel (batch) executions, scales better for high-dimensional problems, and is more robust to initializations. We demonstrate these merits by applying the algorithm in the optimization of the CO2 injection schedule in the Cranfield site in Mississippi, USA, using field data. The optimized injection schedule achieves 16% more gas storage volume and 56% less water/surfactant usage compared with the baseline. The performance of BO is compared with that of a genetic algorithm (GA) and a covariance matrix adaptation (CMA)-evolution strategy (ES). The results demonstrate the superior performance of BO, in that it achieves a competitive objective function value with over 60% fewer forward model evaluations.
Unconventional oil and gas resources have become the most important and realistic field for increasing China's domestic oil and gas reserves and production. At present, the production scale does not match the massive amount of resources and the rapid growth of proven geological reserves. The challenges of technology, cost, management, and methodology restrict large-scale and economic development. Based on successful practices, a ″one engine with six gears″ system engineering methodology is put forward, which includes life-cycle management, overall synergy, interdisciplinary cross-service integration, marketoriented operation, socialized support, digitalized management, and low-carbon and green development. The methodology has been proved to be effective in multiple unconventional oil and gas national demonstration areas, including the Jimusar continental shale oil demonstration area. Disruptive views are introduced—
namely, that unconventional oil and gas do not necessarily yield a low return, nor do they necessarily have a low recovery factor. A determination to achieve economic benefit must be a pervasive underlying goal for managers and experts. Return and recovery factors, as primary focuses, must be adhered to during China's development of unconventional oil and gas. The required methodology transformation
includes a revolution in management systems to significantly decrease cost and increase production, resulting in technological innovation.
Many properties of natural fractures are uncertain, such as their spatial distribution, petrophysical properties, and fluid flow performance. Bayesian theorem provides a framework to quantify the uncertainty in geological modeling and flow simulation, and hence to support reservoir performance predictions. The application of Bayesian methods to fractured reservoirs has mostly been limited to synthetic cases. In field applications, however, one of the main problems is that the Bayesian prior is falsified, because it fails to predict past reservoir production data. In this paper, we show how a global sensitivity analysis (GSA) can be used to identify why the prior is falsified. We then employ an approximate Bayesian computation (ABC) method combined with a tree-based surrogate model to match the production history. We apply these two approaches to a complex fractured oil and gas reservoir where all uncertainties are jointly considered, including the petrophysical properties, rock physics properties, fluid properties, discrete fracture parameters, and dynamics of pressure and transmissibility. We successfully identify several reasons for the falsification. The results show that the methods we propose are effective in quantifying uncertainty in the modeling and flow simulation of a fractured reservoir. The uncertainties of key parameters, such as fracture aperture and fault conductivity, are reduced.
The efficient exploration and development of unconventional oil and gas are critical for increasing the self-sufficiency of oil and gas supplies in China. However, such operations continue to face serious problems (e.g., borehole collapse, loss, and high friction), and associated formation damage can severely impact well completion rates, increase costs, and reduce efficiencies. Water-based drilling fluids possess certain advantages over oil-based drilling fluids (OBDFs) and may offer lasting solutions to resolve the aforementioned issues. However, a significant breakthrough with this material has not yet been made, and major technical problems continue to hinder the economic and large-scale development of unconventional oil and gas. Here, the international frontier external method, which only improves drilling fluid inhibition and lubricity, is expanded into an internal–external technique that improves the overall wellbore quality during drilling. Bionic technologies are introduced into the chemical material synthesis process to imitate the activity of life. A novel drilling and completion fluid technique was developed to improve wellbore quality during drilling and safeguard formation integrity. Macroscopic and microscopic analyses indicated that in terms of wellbore stability, lubricity, and formation protection, this approach could outperform methods that use typical OBDFs. The proposed method also achieves a classification upgrade from environmentally protective drilling fluid to an ecologically friendly drilling fluid. The developed technology was verified in more than 1000 unconventional oil and gas wells in China, and the results indicate significant alleviation of the formation damage attributed to borehole collapse, loss, and high friction. It has been recognized as an effective core technology for exploiting unconventional oil and gas resources. This study introduces a novel research direction for formation protection technology and demonstrates that observations and learning from the natural world can provide an inexhaustible source of ideas and inspire the creation of original materials, technologies, and theories for petroleum engineering.
After more than 70 years of evolution, great achievements have been made in machine translation. Especially in recent years, translation quality has been greatly improved with the emergence of neural machine translation (NMT). In this article, we first review the history of machine translation from rule-based machine translation to example-based machine translation and statistical machine translation. We then introduce NMT in more detail, including the basic framework and the current dominant framework, Transformer, as well as multilingual translation models to deal with the data sparseness problem. In addition, we introduce cutting-edge simultaneous translation methods that achieve a balance between translation quality and latency. We then describe various products and applications of machine translation. At the end of this article, we briefly discuss challenges and future research directions in this field.
Tin phosphides are attractive anode materials for ultrafast lithium-ion batteries (LIBs) because of their ultrahigh Li-ion diffusion capability and large theoretical-specific capacity. However, difficulties in synthesis and large size enabling electrochemical irreversibility impede their applications. Herein, an in situ catalytic phosphorization strategy is developed to synthesize SnP/CoP hetero-nanocrystals within reduced graphene oxide (rGO)-coated carbon frameworks, in which the SnP relative formation energy is significantly decreased according to density functional theory (DFT) calculations. The optimized hybrids exhibit ultrafast charge/discharge capability (260 mA·h·g−1 at 50 A·g−1) without capacity fading (645 mA·h·g−1 at 2 A·g−1) through 1500 cycles. The lithiation/delithiation mechanism is disclosed, showing that the 4.0 nm sized SnP/CoP nanocrystals possess a very high reversibility and that the previously formed metallic Co of CoP at a relatively high potential accelerates the subsequent reaction kinetics of SnP, hence endowing them with ultrafast charge/discharge capability, which is further verified by the relative dynamic current density distributions according to the finite element analysis.Graphical abstractDownload : Download high-res image (191KB)Download : Download full-size image
Egg custard is a common dish on the dining table and exhibits a uniform porous structure after freeze-drying. The protein within egg custard is a rich source of carbon and nitrogen, and the custard's unique microstructure and adjustable electrical properties make it a potential porous carbon precursor. Herein, nitrogen in situ doped porous carbons (NPCs) and potassium-carbonate-modified NPCs (PNPCs) are obtained through a simple gelation and carbonization process using egg white as the raw material. The unique morphologies of the porous carbon are inherited from the protein and include fibrous clusters, honeycomb holes, and a grooved skeleton. Their excellent impedance matching and effective internal loss make the obtained porous carbons good candidates for lightweight electromagnetic (EM) wave absorbers without the need to dope with metal elements. As a representative porous carbon, PNPC10-700 has multiple structures, including fibrous clusters, honeycomb holes, and a porous skeleton. Moreover, it achieves a maximum reflection loss value of −66.15 dB (with a thickness of 3.77 mm) and a broad effective absorption bandwidth of 5.82 GHz (from 12.18 to 18.00 GHz, with a thickness of 2.5 mm), which surpasses the reported values in most of the literature. Thus, gelation combined with the further carbonization of egg white (protein) is a new method for designing the morphology and EM properties of porous carbon absorbers.
Dry reforming of ethane (DRE) has received significant attention because of its potential to produce chemical raw materials and reduce carbon emissions. Herein, a composition-induced strong metal–support interaction (SMSI) effect over FeNi/Al–Ce–O catalysts is revealed via X-ray photoelectron spectroscopy (XPS), H2-temperature programmed reduction (TPR), and energy dispersive X-ray spectroscopy (EDS) elemental mapping. The introduction of Al into Al–Ce–O supports significantly influences the dispersion of surface active components and improves the catalytic performance for DRE over supported FeNi catalysts due to enhancement of the SMSI effect. The catalytic properties, for example, C2H6 and CO2 conversion, CO selectivity and yield, and turnover frequencies (TOFs), of supported FeNi catalysts first increase and then decrease with increasing Al content, following the same trend as the theoretical effective surface area (TESA) of the corresponding catalysts. The FeNi/Ce–Al0.5 catalyst, with 50% Al content, exhibits the best DRE performance under steady-state conditions at 873 K. As observed by with in situ Fourier transform infrared spectroscopy (FTIR) analysis, the introduction of Al not only increases the content of surface Ce3+ and oxygen vacancies but also promotes the dispersion of surface active components, which further alters the catalytic properties for DRE over supported FeNi catalysts.
Inspired by the tremendous achievements of meta-learning in various fields, this paper proposes the local quadratic embedding learning (LQEL) algorithm for regression problems based on metric learning and neural networks (NNs). First, Mahalanobis metric learning is improved by optimizing the global consistency of the metrics between instances in the input and output space. Then, we further prove that the improved metric learning problem is equivalent to a convex programming problem by relaxing the constraints. Based on the hypothesis of local quadratic interpolation, the algorithm introduces two lightweight NNs; one is used to learn the coefficient matrix in the local quadratic model, and the other is implemented for weight assignment for the prediction results obtained from different local neighbors. Finally, the two sub-models are embedded in a unified regression framework, and the parameters are learned by means of a stochastic gradient descent (SGD) algorithm. The proposed algorithm can make full use of the information implied in target labels to find more reliable reference instances. Moreover, it prevents the model degradation caused by sensor drift and unmeasurable variables by modeling variable differences with the LQEL algorithm. Simulation results on multiple benchmark datasets and two practical industrial applications show that the proposed method outperforms several popular regression methods.
Tunnel seismic detection methods are effective for obtaining the geological structure around the tunnel face, which is critical for safe construction and disaster mitigation in tunnel engineering. However, there is often a lack of accuracy in the acquired geological information and physical properties ahead of the tunnel face in the current tunnel seismic detection methods. Thus, we apply a frequency-domain acoustic full-waveform inversion (FWI) method to obtain high-resolution results for the tunnel structure. We discuss the influence of the frequency group selection strategy and the tunnel observation system settings regarding the inversion results and determine the structural imaging and physical property parameter inversion of abnormal geological bodies ahead of the tunnel face. Based on the conventional strategies of frequency-domain acoustic FWI, we propose a frequency group selection strategy that combines a low-frequency selection covering the vertical wavenumber and a high-frequency selection of anti-aliasing. This strategy can effectively obtain the spatial structure and physical parameters of the geology ahead of the tunnel face and improve the inversion resolution. In addition, by linearly increasing the side length of the tunnel observation system, we share the influence of the length of the two sides of the observation systems of different tunnels on the inversion results. We determined that the inversion results are best when the side length is approximately five times the width of the tunnel face, and the influence of increasing the side observation length beyond this range on the inversion results can be ignored. Finally, based on this approach, we invert for the complex multi-stratum model, and an accurate structure and physical property parameters of the complex stratum ahead of the tunnel face are obtained, which verifies the feasibility of the proposed method.
This paper presents the background, scientific objectives, experimental design, and preliminary achievements of the Xin'anjiang nested experimental watershed (XAJ-NEW), implemented in 2017 in eastern China, which has a subtropical humid monsoon climate and a total area of 2674 km2. The scientific objectives of the XAJ-NEW include building a comprehensive, multiscale, and nested hydrometeorological monitoring and experimental program, strengthening the observation of the water cycle, discovering the spatiotemporal scaling effects of hydrological processes, and revealing the mechanisms controlling runoff generation and partitioning in a typical humid, hilly area. After two years of operation, preliminary results indicated scale-dependent variability in key hydrometeorological processes and variables such as precipitation, runoff, groundwater, and soil moisture. The effects of canopy interception and runoff partitioning between the surface and subsurface were also identified. Continuous operation of this program can further reveal the mechanisms controlling runoff generation and partitioning, discover the spatiotemporal scaling effects of hydrological processes, and understand the impacts of climate change on hydrological processes. These findings provide new insights into understanding multiscale hydrological processes and their responses to meteorological forcings, improving model parameterization schemes, and enhancing weather and climate forecast skills.
Indoor environmental quality (IEQ) significantly affects human health and wellbeing. Therefore, continuous IEQ monitoring and feedback is of great concern in both the industrial and academic communities. However, most existing studies only focus on developing sensors that cost-effectively promote IEQ measurement while ignoring interactions between the human side and IEQ monitoring. In this study, an intelligent IEQ monitoring and feedback system (IBEM) is developed. Firstly, the IBEM hardware instrument integrates air temperature, relative humidity, CO2, particulate matter with an aerodynamic diameter no greater than 2.5 μm (PM2.5), and illuminance sensors within a small device. The accuracy of this integrated device was tested through a co-location experiment with reference sensors; the device exhibited a strong correlation with the reference sensors, with a slight deviation (R2 > 0.97 and slopes between 1.01 and 1.05). Secondly, a wireless data transmission module, a cloud storage module, and graphical user interfaces (i.e., a web platform and mobile interface) were built to establish a pathway for dataflow and interactive feedback with the occupants of the indoor environments. Thus, the IEQ parameters can be continuously monitored with a high spatiotemporal resolution, interactive feedback can be induced, and synchronous data collection on occupant satisfaction and objective environmental parameters can be realized. IBEM has been widely applied in 131 buildings in 18 cities in China, with 1188 sample locations. Among these applications, we report on the targeted IEQ diagnoses of two individual buildings and the exploration of relationships between subjective and objective IEQ data in detail here. This work demonstrates the great value of IBEM in both industrial and academic research.
Replacing micro-reinforcing fibers with carbon nanotubes (CNTs) is beneficial for improving the impact properties of ultra-high performance concrete (UHPC); however, the weak wettability and dispersibility of CNTs and the weakly bonded interface between CNTs and UHPC limit their effectiveness as composites. Therefore, this study aims to enhance the reinforcement effect of CNTs on the impact properties of UHPC via functionalization. Unlike ordinary CNTs, functionalized CNTs with carboxyl or hydroxyl groups can break the Si–O–Ca–O–Si coordination bond in the C–S–H gel and form a new network in the UHPC matrix, effectively inhibiting the dislocation slip inside UHPC matrix. Furthermore, functionalized CNTs, particularly carboxyl-functionalized CNTs, control the crystallization process and microscopic morphology of the hydration products, significantly decreasing and even eliminating the width of the aggregate–matrix interface transition zone of the UHPC. Moreover, the functionalized CNTs further decrease the attraction of the negatively charged silicate tetrahedron to Ca2+ in the C–S–H gel, while modifying the pore structure (particularly the nanoscale pore structure) of UHPC, leading to the expansion of the intermediate C–S–H layer. The changes in the microstructures of UHPC brought about by the functionalized CNTs significantly enhance its dynamic compressive strength, peak strain, impact toughness, and impact dissipation energy at strain rates of 200–800 s−1. Impact performance of UHPC containing a small amount of carboxyl-functionalized CNTs (especially the short ones) is generally better than that of UHPC containing hydroxyl-functionalized and ordinary CNTs; it is even superior to that of UHPC with a high steel fiber content.
There has been a wealth of research that has examined the nature of rework in construction. Progress toward addressing the rework problem has been limited—it still plagues practice, adversely impacting a project's performance. Almost all rework studies have focused on determining its proximal or root causes and therefore have overlooked the conditions that result from its manifestation. In filling this void, this paper draws upon our previous empirical studies, amongst others, to provide a much-needed theoretical framing to understand better why rework occurs, what its consequences are, and how it can be mitigated during construction. The theoretical framing we derive from our review provides construction organizations and their projects with a realization that the journey to mitigating rework begins with creating an error-mastery culture comprising authentic leadership, psychological safety, an errormanagement orientation, and resilience. We suggest that, once an error-mastery culture is established within construction organizations and their projects, they will be better positioned to realize the benefits of the techniques, tools, and technologies espoused to address rework, such as the Last Planner® and building information modeling. We also provide directions for future research and identify implications for practice so that strides toward rework mitigation in construction can be made.