Against the backdrop of the accelerated development of the new energy system and the deepening integration of multiple energy sources, traditional dispatch and control modes oriented toward a single carrier and a single network can no longer meet the coordinated operational requirements of the substance-energy network. To address this challenge, this study systematically investigates the paradigm and key technologies of intelligent dispatch and control for the substance-energy network, focusing on the strong coupling among multiple energy and material flows, the dynamic evolution across multiple time scales, and the pronounced uncertainties on both the supply side and the demand side. First, an integrated dispatch and control framework for the substance-energy network is proposed. The framework consists of five layers, namely the resources and infrastructure layer, the sensing and communication layer, the prediction-evaluation layer, the coordinated decision-making layer, and the feedback-evolution layer. This framework reflects the overall logic from physical foundation to digital mapping, from state cognition to coordinated decision-making, and from control execution to feedback evolution, thereby forming a unified architecture for the coordinated perception, analysis, decision-making and regulation of multi-network integrated systems. Second, three fundamental modules are identified and refined, including real-time monitoring, early warning and emergency support; global coordinated dispatch and control and optimal allocation of green energy; and a full-chain metering and delivery system. These modules together form a closed-loop dispatch and control framework covering perception, decision-making and evaluation. The objectives are to establish a system-wide perception and early warning framework, maintain critical energy supply and support rapid recovery under extreme scenarios, realize coordinated operation across multiple networks, regions and carriers under normal conditions, promote the efficient allocation and coordinated accommodation of green energy on a wider scale, and build a clear and traceable correspondence between physical flows and value settlement, thereby supporting refined operation, market-oriented trading and low-carbon governance of the substance-energy network. On this basis, a closed-loop technical chain is further constructed, consisting of resource and demand forecasting, state perception and estimation, unified-chain simulation and modeling, and dispatch decision-making and active control. The corresponding key technologies of each link are further summarized, including forecasting methods under multiple uncertainties, multimodal sensing and state estimation, unified modeling and simulation of coupled energy-material processes, and coordinated dispatch and active control under complex operating conditions. This study aims to provide a reference for the construction of the integrated dispatch and control system and the development of related key technologies for the substance-energy network.

Against the backdrop of the emerging “substance-energy network”, the coupling among multi-energy systems is becoming increasingly strong. As a key hub linking the oil and gas supply side with the power system, natural gas pipeline networks play an important role in flexible energy transfer and cross-domain peak shaving. With the growing penetration of upstream renewable energy and the increasing fluctuations of downstream gas demand, the dispatch and operation of natural gas pipeline networks have shifted from steady-state analysis and day-level adjustment to dynamic control over refined temporal scales. Around this transition, related studies have developed along two closely coupled lines. The first is control-oriented model construction, which transforms high-dimensional nonlinear transient models into low-dimensional dynamic models suitable for online feedback control while preserving key physical characteristics such as pressure-wave propagation, linepack variation, compressor effects, and boundary disturbances. The second is model-based control design, which involves using models to predict system states and express control constraints under equipment operation limitations and highly uncertain source-load fluctuations, so as to achieve flexible system operation. This paper systematically reviews representative studies in this field from the perspectives of model construction and control design. On the modeling side, it focuses on the development of transfer-function models and state-space models in lumped-parameter form, as well as AI-enabled data-driven models, and analyzes the trade-offs among physical interpretability, transient accuracy, online computational efficiency, and control applicability. On the control side, it summarizes the research trajectory from open-loop operational optimization to predictive feedback control and direct feedback control. On this basis, the paper further identifies key obstacles in existing theoretical studies and engineering deployment, including the modeling and compensation of propagation delays and the non-ideal characteristics of practical actuators. Finally, it summarizes three major tensions: the tension between model accuracy and control applicability, the tension between the strengths of individual algorithms and the pursuit of multiple objectives, and the tension between theoretical assumptions and practical engineering deviations. Future research directions are then discussed, including adaptive low-dimensional modeling for multiple operating conditions, deep integration of mechanism-based and data-driven methods, intelligent control methods, and composite control architectures. This paper is expected to provide useful references for future system-level intelligent control of natural gas pipeline networks and to support the construction of a new ecosystem characterized by the symbiotic integration of substance and intelligence.

With the accelerated development of China’s nationwide integrated natural gas pipeline network, natural gas pipeline networks are increasingly characterized by multiple sources, multiple sinks, multiple loops, and large-scale complex topologies, imposing higher requirements on both the accuracy and computational efficiency of transient simulation. To address this challenge, this study proposes a Nets-Link-based acceleration method for transient simulation of natural gas pipeline networks. The network is spatially decomposed according to its geographical distribution and topological characteristics, while a depth-first search algorithm is employed to identify connected components and their associated attributes. A transient hydraulic model is established using an implicit central finite-difference scheme and solved by the Newton-Raphson method. At each time layer, a hierarchical iterative strategy is adopted. In the inner iteration, the operating parameters of nodes and pipelines in each subdomain are solved independently. In the outer iteration, a low-dimensional approximate Jacobian matrix for the interfaces is constructed based on the Schur complement concept, and the interface variables are iteratively corrected and updated using the Broyden formula until interface consistency among all subdomains is achieved. The solutions of all subdomains are then assembled to obtain the solution of the entire network. Results from a 24 h transient simulation case show that, compared with the TGNET benchmark, the mean relative errors of pressure and flow rate do not exceed 0.14% and 1.43%, respectively. Compared with the global iterative solution, partitioning the network into four to eight subdomains reduces the computational time by approximately 68%-78%, while the gain in computational efficiency exhibits diminishing marginal returns as the number of subdomains increases. The proposed method can significantly reduce the solution scale and computational complexity of transient simulation for large-scale and complex natural gas pipeline networks while maintaining high computational accuracy, thereby providing an efficient simulation tool for network operational analysis and cross-regional natural gas resource allocation.

Carbon capture, utilization and storage (CCUS) is a key pathway to deep decarbonization in China’s industrial sector amid the global energy transition and China’s carbon peaking and carbon neutrality goals. Supercritical carbon dioxide (sCO₂) pipelines are important infrastructure for large-scale CCUS deployment because of their high transport density and potential economic advantages. However, sCO₂ pipeline technology in China remains at an early stage of development. Phase instability, corrosion and leakage risks pose challenges to long-term safe operation and wider CCUS deployment. This review examines the development of sCO₂ pipelines and advances in phase behavior control, controlled depressurization, leakage prevention, fracture propagation control and corrosion protection. Large-scale, long-distance sCO₂ pipelines in China remain at the demonstration stage. The Jilin Petrochemical-Jilin Oilfield CO₂ pipeline, which is under construction and is designed for supercritical/dense-phase transport, is expected to provide an engineering setting for evaluating phase control, key equipment and safety measures in long-distance, high-pressure CO₂ transport. Nevertheless, transient multiphase flow, fracture propagation and corrosion under conditions involving complex terrain and multicomponent CO₂ streams require further investigation. Future work should combine full-scale demonstrations with technological development to strengthen safety measures under interacting risk factors. Digital technologies should also be integrated throughout pipeline design, construction, operation and maintenance, alongside the development of standards for sCO₂ transport. These efforts would support the development of integrated, digitally managed sCO₂ pipeline networks and the wider deployment of CCUS.

To address the future demand for hydrogen transportation within natural gas pipeline networks, it is essential to conduct economic research that integrates the operational characteristics of hydrogen-blended natural gas pipelines. This study investigates the operational characteristics and economic performance of such pipelines under multiple modes and factors. Steady-state simulations are coupled with economic indicators, including net present value and rate of return, to evaluate the economic viability of two hydrogen delivery modes. One mode is direct hydrogen blending, and the other is hydrogen separation followed by sales. Both modes are examined under two throughput conditions, namely a constant standard volumetric flow rate and a constant total heating value. Results show that the pressure drop decreases as the hydrogen blending ratio increases under the constant standard volumetric flow condition, whereas it increases under the constant total heating value condition. Meanwhile, the standard volume and total heating value of the pipeline inventory decrease under both conditions. The rate of return under the constant total heating value condition is superior to that under the constant standard volumetric flow condition, and the hydrogen separation and sales mode outperforms the direct blending mode economically. The upstream-downstream hydrogen price differential significantly impacts the profitability of pipeline hydrogen blending. Under the constant total heating value condition, the system exhibits greater tolerance to hydrogen price volatility. In the planning of hydrogen-blended natural gas pipelines, the two operating conditions (constant standard volumetric flow rate and constant total heating value) and the two operational modes (direct blending via an open service platform, and hydrogen separation and sales from an independent operator perspective) can all serve as useful references for designing operational schemes for hydrogen-blended transmission pipelines.

Hydrogen-blended natural gas transportation is a key pathway for the large-scale utilization of hydrogen energy. However, the hydrogen environment may induce material performance degradation, increasing the failure risk of defective pipelines, necessitating the development of efficient and accurate assessment methods. To rapidly and accurately predict the failure pressure of defective hydrogen-blended natural gas pipelines, this study proposes a method that integrates finite element simulation and deep learning. First, a three-dimensional finite element model of X80 pipeline steel with external corrosion defects was established using ABAQUS, and the effect of the hydrogen environment on material properties was represented by introducing stress-strain curves under different hydrogen blending ratios. The model was validated for mesh independence and experimental comparison to ensure its reliability and accuracy. Subsequently, 401 cases with different hydrogen blending ratios, defect sizes, and pipeline geometries were simulated to construct a failure pressure database, providing sample support for the deep learning model. The results show that defect depth and length have a significant negative correlation with failure pressure, defect width has a relatively weak and nonlinear effect, and the hydrogen blending ratio has a negligible impact on failure pressure. The deep learning model was optimized through hyperparameter tuning, determining the network structure with 3 hidden layers, 350 neurons, and the ReLU activation function with a learning rate of 0.0001. After 6 random validation tests, the predicted results highly matched the finite element calculations, with an average relative error of approximately 0.765% and a maximum error of less than 4.02%, demonstrating good generalization capability. In terms of computational efficiency, deep learning significantly outperforms finite element analysis, requiring only about 0.01 s to predict a single sample, greatly reducing the evaluation cycle. The study shows that the established finite element-deep learning integration framework provides a reliable tool for the efficient and accurate prediction of failure pressures in hydrogen-blended natural gas defective pipelines.

Gathering pipelines are an integral part of oil and gas gathering and transportation systems, and their safe and stable operation is essential to the reliability of the oil and gas supply chain. Leakage monitoring plays a critical role in maintaining operational safety by identifying potential leaks and providing early warning of associated risks. In practice, however, fluctuations in equipment operating conditions, changes in weather conditions, and human activities near the pipeline can introduce strong interference into monitoring systems. Such interference can lead to frequent false alarms and undermine alarm credibility, thereby compromising subsequent risk assessments. An alarm therefore needs to be interpreted in light of the operating conditions and external influences present when it is generated.To address this challenge, this study proposes a leakage risk monitoring method based on multisource information fusion for gathering pipelines subject to strong interference. Operational scenarios are first analyzed to identify relevant interference factors. A Bayesian network is then developed to integrate information on equipment operating conditions, weather conditions, human activities, and alarms from multiple monitoring sources. Using the available evidence, the network estimates the posterior probability of actual leakage given the observed alarms and interference conditions. Leakage is then defined as the initiating event in an event tree model for quantitative consequence assessment. Its occurrence probability is set to the posterior leakage probability inferred from the Bayesian network. A relative risk index combines the inferred leakage probability with the assessed consequences to provide an integrated measure of leakage risk.The method is evaluated through a case study of a pipeline in Zichang, Yan’an, Shaanxi Province, China, using regional land-use information and weather statistics. Different interference scenarios are compared using the same observed alarm evidence. Specifically, only an acoustic monitoring alarm is observed, while the assumed leak orifice diameter and the conditions used for consequence calculations remain unchanged across the scenarios. As interference factors are progressively combined, the inferred leakage probability decreases from 93.63% to 75.49%. The corresponding relative risk index decreases from 1.4606 to 1.1776, representing a reduction of 19.376% relative to the baseline scenario.The case study illustrates how the method assesses the credibility of leakage alarms under strong interference and incorporates the resulting probability estimates into consequence-based risk assessment. By explicitly accounting for uncertainty about whether an alarm indicates an actual leak, the method reduces the influence of false alarms on risk assessment and provides risk estimates that better reflect the effects of interference under practical operating conditions. The proposed approach provides a basis for operational safety management, early warning of leakage risks, and emergency decision-making for gathering pipelines operating under strong interference conditions.

To investigate the dynamic mechanisms of risk factors in petrochemical product transportation and thereby formulate risk control strategies, this study takes petrochemical product transportation risk as its research object. Based on 86 typical accident cases from 2018 to 2024, the Wuli-Shili-Renli (WSR, i.e., physical-logical-human) methodology is adopted to systematically identify 11 key risk factors across three dimensions. The Analytic Network Process (ANP) is employed to calculate the weights of these factors and clarify their mutual influencing relationships, and System Dynamics (SD) is used to construct a simulation model of the risk level of petrochemical product transportation, so as to explore the effects of different strategies on risk levels. The results show that intervention strategies across all three dimensions—physical, logical, and human—can effectively reduce the risk level of petrochemical product transportation. Among them, increased investment in the human dimension has the most significant effect on risk control; among all influencing factors, the intensity of government supervision has the greatest impact on the risk level of petrochemical product transportation; the optimal intervention in the physical dimension is increasing investment in emergency facilities, in the logical dimension is strengthening government supervision, and in the human dimension is improving the operational skills of workers. The conclusions provide a decision-making reference for the safety management of petrochemical product transportation: it is recommended that the government improve regulations and standards and strengthen supervision, that enterprises rationally allocate safety investments across the three dimensions, and that they focus on personnel education, training, and skill enhancement to reduce the risk of the petrochemical product transportation system.

In surface gathering and transportation systems for ultra-high water-cut oilfields, gathering without hot water mixing have emerged as an effective measure for energy saving and carbon reduction. However, the gelation characteristics of crude oil with high water content lead to complex and unpredictable wellhead pressure evolution, further constraining the low-carbon synergistic operation of the entire extraction-to-surface gathering process. Based on static and dynamic operational data from actual production wells in a certain oilfield, this paper considers the impact of time-domain nonlinear shifts on wellhead pressure evolution patterns. The Dynamic Time Warping algorithm (DTW) is employed to identify temporal characteristics of wellhead pressure in crude oil gathering pipelines without water mixed. Feature expansion reconstruction and mutual information-based dimensionality reduction are then applied to construct a nonlinear reconstructed dataset. Subsequently, by adopting a Mixture of Experts (MoE) model as the intelligent prediction framework, with Random Forest Regression (RFR), Neural Network Regression (NNR), and Gradient Boosting Regression (GBR) selected as sub-expert models, a DTW-MoE wellhead pressure prediction model was established, aiming to reveal the influencing factors of wellhead pressure in gathering pipelines and rank their degrees of impact. Results indicate that the four optimal path matches in the DTW-MoE wellhead pressure prediction model are ξ₁-GBR-72, ξ₂-GBR-6, ξ₃-RFR-3 and ξ₄-NNR-6. Particularly in the high pressure section where the wellhead pressure exceeds 0.5 MPa, the prediction errors are all controlled within 10%. The DTW algorithm identified four clusters with distinct temporal characteristics. Cluster ξ₁ is primarily governed by the synergistic effects of the production radius and liquids production rate on wellhead pressure, with corresponding composite SHAP values of 0.0386 and 0.0225 respectively. For Clusters ξ₂, ξ₃ and ξ₄, the key influencing factors are the production radius, liquids production rate, and pipe diameter, though their relative impacts vary significantly. The findings of these studies provide valuable insights for energy conservation and emission reduction oversight in oilfield gathering and transportation systems, particularly in the context of digital intelligence empowerment and green transformation.

Key equipment in oil and gas production systems, including pipelines, pressure vessels, and subsea production facilities, operates for extended periods under extremely harsh environmental conditions characterized by high temperature, high pressure, and severe corrosion. These extreme conditions pose significant risks to equipment safety, operational efficiency, and the overall reliability of the production process. Consequently, their design, inspection, and maintenance activities throughout the entire lifecycle are highly dependent on comprehensive equipment integrity management standards. However, since multi-source standards are often formulated by different organizations, significant discrepancies exist in their document structures, professional terminology definitions, and information granularity. This inherent heterogeneity leads to fragmented standard knowledge and ambiguous associations between clauses, which significantly increases the difficulty of information retrieval for engineers and makes it extremely challenging to support consistent decision-making across different equipment types, ultimately leading to inefficiencies in maintenance planning. To address these challenges, this study applies Graph Retrieval-Augmented Generation (GraphRAG) technology to construct a knowledge graph specifically for oil and gas equipment integrity standards. Furthermore, it develops an intelligent retrieval and reasoning method tailored for practical engineering problems, thereby achieving unified organization of cross-standard knowledge, explicit semantic association, and full traceability of evidence. The specific research workflow includes three well-defined steps. First, multi-source standard texts are digitized, with layout normalization and semantic segmentation applied to transform the heterogeneous raw data into a structured corpus. Second, core elements—such as equipment, components, materials, failure modes, inspection methods, and maintenance strategies—along with their intrinsic relationships are systematically extracted. Based on this extraction, a comprehensive semantic network with an “equipment-component-risk-event-strategy” backbone is constructed. Furthermore, entity alignment and fusion techniques are applied to resolve conflicts and eliminate redundant information, ensuring the consistency of the knowledge base. Finally, cross-document clauses are aggregated along the pathways of the knowledge graph. By leveraging sophisticated reasoning algorithms, the system outputs precise reasoning conclusions accompanied by their corresponding supporting clauses, providing clear and reliable evidence. Experimental results demonstrate that the constructed corpus covers more than one hundred standard documents, yielding approximately fifty thousand semantic fragments and over one hundred thousand entity-relation pairs. In comprehensive evaluation tests, the system achieves an accuracy of 87.3% in multi-hop question-answering tasks, and the evidence traceability coverage reaches a high level of 91.5%. These findings indicate that the proposed method effectively improves the efficiency of standard knowledge integration and enhances reasoning transparency. Ultimately, this research provides robust technical support for equipment integrity assessment, scientific inspection plan selection, and the strategic formulation of maintenance plans in complex industrial environments.

Ultra-deep wells are generally characterized by high temperature, high pressure, and strong reservoir heterogeneity. The physical properties of gas-water two-phase fluids within wellbores vary dynamically with formation temperature and pressure, leading to highly nonlinear responses in production profile logging measurements. Conventional gas-water two-phase interpretation models, such as the slip model and drift-flux model, are developed based on laboratory experiments under surface ambient temperature and pressure. These models fail to accurately describe fluid production behavior under downhole high-temperature and high pressure conditions, accompanied by large interpretation errors and limited applicability, which restricts the fine dynamic evaluation of ultra deep gas reservoirs flow logging as a single constraint, lacking collaborative regulation of multi-logging data including temperature, pressure and density. Moreover, these methods ignore the differences in resolution and measurement reliability among various logging instruments, which readily results in distorted identification of bottom water and inaccurate productivity interpretation. To address the above limitations, this study proposes an interpretation method for gas-water two-phase production profiles of high temperature and high pressure ultra deep well that integrates multi-source logging constraints with the Seagull Optimization Algorithm(SOA), the model embeds high temperature and high pressure gas-water physical properties dynamically calculated from measured downhole temperature-pressure at each depth into temperature logging, pressure logging and density logging response models, and introduces the gas-phase Joule-Thomson thermal effect to characterize the thermal behavior of high pressure gas in wellbore, which overcomes the systematic deviation caused by fixed physical parameters under surface ambient temperature and pressure adopted in traditional models. Taking the gas and water volumetric flow of each perforated interval as unknown variables to be solved, theoretical response models for temperature, pressure and logging and density are separately constructed based on wellbore energy conservation and multiphase momentum balance principles. A high-dimensional nonlinear objective function is further established by integrating the resolution and reliability coefficients as an equality constraint. The seagull optimization algorithm(SOA) is adopted to implement iterative computation, enabling simultaneous solution of gas and water flow for all perforated intervals. Validation using field production profile logging data form actual ultra deep wells demonstrates that the relative errors between the total gas and water production calculated by the proposed method and surface metering data are only 5.23% and 6.06%, respectively, which fully meets the processing specifications of production logging data. The proposed method can provide reliable technical support for quantitative productivity evaluation, water shutoff and profile control, as well as long term dynamic monitoring of ultra deep oil and gas reservoirs.

Natural fractures strongly affect flow capacity and development performance in tight sandstone reservoirs. In the Chang 8 Member of the southwestern Yishan Slope, Ordos Basin, thin sandstone-mudstone interbeds, strong fracture heterogeneity, fracture filling, and borehole conditions complicate conventional logging responses, making single-log thresholds unsuitable for continuous fracture identification. Based on core observations, interpretations of electrical imaging logs, and conventional well logs, this study analyzed the development characteristics and logging-response differences of natural fractures in the Chang 8 Member and developed a Neighborhood Reference-Constrained LogFormer (NRC-LogFormer) model for fracture identification. The model uses local neighborhood responses within each well as references and jointly captures center-neighborhood differences, multiscale perturbations, and sequential context, thereby improving the identification of weak fracture responses and reducing misidentification caused by lithologic changes and borehole disturbances. The results show that structural fractures and bedding-parallel/lithologic-interface fractures dominate the study area, with high-angle fractures being prevalent. Fractures with dips of 70°~90° account for 72.4%, and those with dips greater than 60° account for 82.3%. Fractures are more developed in Chang 8₁ than in Chang 8₂ and mainly occur in sandstone and fine sandstone, with their distribution jointly influenced by tectonic stress, sand-body distribution, and lithologic interfaces. Compared with stable non-fracture backgrounds within the same well, intervals containing interpreted fracture points generally exhibit lower sonic transit time, higher compensated density, lower compensated neutron, and differentiated resistivity responses, characterized by lower deep induction resistivity and eight-laterolog resistivity, but higher formation resistivity. These response patterns may be related to host lithology, fracture filling, mud-filtrate invasion, and measurement-volume effects. Caliper, gamma ray, and spontaneous potential provide limited independent discrimination, indicating that fracture-related logging responses are multivariate and non-unique. The model achieved precision, recall, and F₁ values of 0.875, 0.863, and 0.869, respectively, outperforming support vector machine (SVM), extreme gradient boosting (XGBoost), one-dimensional convolutional neural network, and Transformer models. Applications to representative wells showed that high-probability fracture intervals agreed well with electrical imaging-log interpretations and microresistivity circumferential imaging dynamic images. Prediction uncertainty can further assist in identifying suspected fracture responses and intervals requiring focused review. The proposed method provides support for fracture-interval screening, logging-interpretation review, and reservoir evaluation in tight sandstone reservoirs where electrical imaging-log data are limited.

Based on the latest developments in the well logging industry, this paper systematically sorts out breakthroughs in key technologies and equipment, distinctive features of industry development, and future evolution trends, providing a reference for technological innovation in the field and the efficient development of oil and gas resources. The results indicate that international oilfield service giants are spearheading the industrial transformation centered on cable-free operation, front-end sensing and intelligent collaboration, whereas Chinese enterprises have prioritized the independence and controllability of core technologies, achieving a transition from follower to leader in such fields as ultra-high temperature and ultra-high pressure equipment, intelligent applications, and offshore and ultra-deep well operations. The well logging industry is currently characterized by five distinctive features: (1) Cable-free logging technology has advanced from “concept verification” to “large-scale application”, and autonomous operation has become a new benchmark for medium-to-high-end equipment. (2) AI technology has penetrated the entire “acquisition-interpretation-decision-making” chain, realizing dual enhancements in operational efficiency and evaluation accuracy. (3) Adaptability to extreme operating conditions has been markedly strengthened, with continuous upgrades in equipment temperature and pressure resistance indicators, tackling the bottlenecks of “unreachable logging” and “inaccurate measurement”. (4) Bit-end and forward-looking sensing technologies have filled industry gaps. (5) Green and low-carbon technologies have undergone iterative upgrading, where cable-free designs, near-bit controlled-source LWD technologies, and high-efficiency operation processes drastically reduce carbon emissions and environmental pollution, lowering energy consumption while improving measurement accuracy. In terms of application expansion, well logging technologies have extended to the new energy sector, with growing applications in scenarios including CCUS carbon sequestration monitoring, geothermal resource exploration, and offshore oilfield potential tapping, facilitating the coordinated development of oil and gas development and low-carbon transformation.

In ultra-deep carbonate exploration, shear-wave transit time (DTS) is a petrophysical parameter of considerable engineering significance. However, dipole sonic logging—the primary means of acquiring shear-wave data—is often excluded from conventional logging suites because of cost, time, and downhole risks, making full-interval coverage difficult. Therefore, predicting DTS from conventional logs has substantial practical value. In the ultra-deep fractured-vuggy carbonate reservoirs of a large superimposed basin in western China, which experienced multiple phases of karst modification and fault control, compressional-wave transit time (AC) and DTS lack a stable single-valued mapping, with AC-DTS Pearson correlation coefficients across the six study wells ranging from 0.42 to 0.96. Consequently, conventional models generalize poorly, with high within-well accuracy but amplified errors when extrapolated to unseen wells. To address this issue, six conventional logs (AC, DEN, GR, CNL, RD, and RS) are utilized as inputs to construct a deep sequence model, CNN-BiLSTM-Attention, integrating one-dimensional convolution, bidirectional long short-term memory networks, and an attention mechanism. The Castagna relation is incorporated as a soft rock-physics prior, a fixed dual-neural-network fusion is adopted as the main model, and domain-adaptive gating is introduced as an exploratory scheme to mitigate cross-well response drift. The evaluation incorporates the coefficient of determination R², root-mean-square error RMSE, mean absolute error MAE, mean absolute percentage error MAPE, Pearson correlation coefficient, and Kling-Gupta efficiency KGE, utilizing leave-one-well-out (LOWO) cross-validation as the core measure of extrapolation to unseen wells. Across the 9533 pooled test samples, the model attains a pooled coefficient of determination Pooled R² of 0.818, an RMSE of 9.79 μs/ft, MAE of 4.69 μs/ft, MAPE of 3.88%, and Pearson and KGE values of 0.918 and 0.756, respectively. Under the stricter LOWO protocol, the median and mean R² reach 0.516 and 0.565, distinctly outperforming the physics-only baseline derived solely from the Castagna formula, for which the LOWO median R² is merely 0.204. Error propagation analysis indicates that DTS prediction errors are transferred to derived geomechanical parameters (dynamic Young's modulus, Poisson's ratio, and normalized brittleness index) that rely on DTS, with well-level error correlation coefficients of 0.87~0.95. This demonstrates that high-DTS intervals and zones with strongly developed vugs and fractures are not only challenging for shear-wave prediction but also represent areas of concentrated uncertainty for engineering parameter evaluation. Meanwhile, evaluating R² in isolation can be misleading due to the small-denominator effect of narrow dynamic ranges, necessitating joint evaluation with absolute error and KGE. Overall, the workflow stably recovers the principal DTS trend in most typical wells, providing a physics-guided and cross-well-transferable approach to acquiring shear-wave transit time for ultra-deep fractured-vuggy carbonate intervals lacking shear-wave data.

Natural shale samples usually exhibit complex mineral compositions, variable depositional environments, and significant differences in thermal maturity, making it difficult to independently identify the controlling effects of organic matter maturation on rock physical properties. In this study, an artificial simulation approach was adopted to prepare three types of artificial shale samples with different thermal evolution stages while maintaining consistent mineral compositions and organic matter contents. Scanning electron microscopy (SEM), nuclear magnetic resonance (NMR), and low-temperature nitrogen adsorption experiments were conducted to characterize the porosity and microscopic pore structures of the samples at different thermal evolution stages. Ultrasonic velocity measurements were performed along three directions under confining pressure conditions to investigate the influence of organic matter structural evolution on pore characteristics and elastic responses, as well as their interrelationships. The experimental results indicate that, with increasing thermal maturity, the porosity of the samples first increases and then slightly decreases. The mature sample exhibits the highest total porosity, whereas the over-mature sample shows a slight reduction in total porosity. With increasing thermal maturity, organic-matter micropores gradually develop, resulting in an increasing proportion of micropores in the samples. Under low confining pressure conditions, the mature sample exhibits the most significant variations in both P-wave and S-wave velocities. The velocity differences among samples with different thermal evolution stages gradually decrease with increasing confining pressure. In the 45° direction, noticeable differences in P-wave and S-wave velocities are observed among the three maturity stages, whereas the velocities of mature and over-mature samples become nearly identical under high confining pressure conditions. Thermal evolution modifies the microscopic structure of organic matter-hosted pores within the samples, resulting in enhanced heterogeneity of the mature samples under low confining pressure conditions. At high confining pressures, the closure of micropores in mature and over-mature samples leads to nearly identical anisotropy parameters. This experimental approach provides an effective method for understanding the influence of organic matter thermal evolution on the elastic properties of natural shale reservoirs.

During the in situ conversion of oil shale, the highly cross-linked structure and complex reaction pathways of kerogen restrict the formation of light products and are often accompanied by carbon skeleton rearrangement and the accumulation of heavy components. To address the difficulty of achieving both efficient heat transfer and chemical upgrading using the single convective medium, this study constructed the multi-component thermal fluid system composed of 60% steam (H₂O), 30% oxygen (O₂), and 10% carbon dioxide (CO₂), and investigated its influence on the non-isothermal evolution behavior of Fushun oil shale kerogen using ReaxFF reactive molecular dynamics simulations. The results show that the multi-component thermal fluid significantly promotes the formation of lighter products. The yields of gaseous hydrocarbons (C₁-C₄) and light oil (C₅-C₁₃) increased from 29.7% and 26.3% to 35.7% and 30.5%, respectively, while the yields of heavy components (C₁₄-C₄₀) and semi-coke (C₄₀₊) decreased from 29.5% and 12.6% to 26.5% and 5.9%, respectively. Reaction site and bond-breaking analyses indicate that the multi-component thermal fluid promotes the early cleavage of weak bonds such as bridged ether bonds and aliphatic C-C bonds, shifting the cracking pathway from side-chain-dominated cleavage to the synergistic cleavage of both the carbon skeleton and side chains. Reaction network analysis further shows that ethylene serves as a core hub in the reaction network during kerogen evolution. In the multi-component thermal fluid system, steam and oxygen are located near the core region of the reaction network, whereas CO₂ mainly plays a peripheral regulatory role, reflecting an overall synergistic mechanism of “oxidation-hydrogen donation-regulation.” This study systematically reveals, at the molecular scale, the synergistic roles of different components in the multi-component thermal fluid and their regulatory effects on kerogen evolution behavior and product distribution, providing theoretical guidance for efficient upgrading during oil shale in situ conversion.

Organic-rich shales are dark-coloured, fine-grained sedimentary rocks characterised by high total organic carbon (TOC) content. They comprise three distinct types of hydrocarbon resources: shale gas, shale oil, and oil shale. In 2025, China’s total domestic oil and gas production reached 420 million tonnes of oil equivalent (toe), of which organic-rich shales alone contributed approximately 30 million toe. This substantial share underscores the considerable resource potential of organic-rich shale formations and their growing importance in China’s energy portfolio.Based on a systematic review of domestic and international literature concerning the efficient and environmentally responsible (green) exploitation of organic-rich shales, this paper provides a comprehensive synthesis of the theories and technological advances relevant to the extraction of shale gas, medium to highmaturity shale oil, medium to lowmaturity shale oil, and oil shale. The main findings are summarised as follows.(1) The theoretical framework for the exploration and development of organic-rich shales has undergone a threestage evolution: initially treating organic-rich shales merely as source rocks, then recognising them as reservoir rocks, and most recently creating “artificial hydrocarbon reservoirs” through proactive engineering interventions. Today, a systematic theoretical system has been firmly established. Current research and development efforts are expanding into new frontiers, including deepburied formations, highefficiency green extraction, and intelligent exploration and development. The overall trajectory is clearly moving towards lower thermal maturity, greater environmental compatibility, higher intelligence, and finer operational precision.(2) In China, shale gas extraction has now matured into an industrialscale technological framework centred on horizontal drilling coupled with volumetric fracturing. China has become the largest shale gas producer outside North America. Future advances will focus on more precise reservoir characterisation and production forecasting, adaptation to increasingly complex geological conditions, and the pursuit of safer and cleaner extraction processes.(3) For medium to highmaturity shale oil in China, inherent disadvantages—including unfavourable gas-oil ratios, poor lateral continuity of oil layers, and suboptimal crude oil properties—pose substantial challenges to technically and economically viable extraction. Consequently, ongoing and future research will continue to prioritise strategies for achieving largescale, sustainable production and for improving overall economic feasibility.(4) For medium to lowmaturity shale oil and oil shale, insitu underground conversion is recognised as the most effective and technically feasible approach. Nevertheless, most insitu conversion technologies remain at the pilottesting stage, and a long path lies ahead before economically viable and efficient development can be realised. Future research will remain firmly focused on advancing insitu conversion technologies, accelerating the transition from laboratoryscale experiments to largescale, commercially viable field operations.

Shale reservoirs are characterized by low porosity and low permeability, rendering conventional development methods ineffective. CO₂ injection technology can simultaneously enhance hydrocarbon recovery and achieve geological sequestration. However, mineral dissolution and precipitation induced by CO₂ injection significantly alter pore structure and seepage capacity, yet the underlying microscopic mechanisms remain insufficiently understood. Based on real shale digital rock, this study constructs a pore-scale reactive transport model using the lattice Boltzmann method to systematically simulate the spatiotemporal evolution of dissolution and precipitation during CO₂-mineral reactions, and analyzes the regulatory effects of injection rate and reaction intensity on mineral reaction mechanisms and permeability evolution. The results indicate that after CO₂ injection, the shale exhibits a cyclic evolution characterized by alternating dissolution, precipitation, and re-dissolution, with permeability showing periodic fluctuations. The formation of preferential flow pathways is the key factor enabling fluid breakthrough. Dissolution and precipitation jointly control pore structure evolution, with a clear positive feedback mechanism between them: dissolution enhances fluid flow and reactant supply, while precipitation locally blocks throats and redirects flow paths. Injection rate and reaction intensity significantly regulate the advancement distance of the reaction front, the degree of precipitation clogging, and the development of preferential channels. Higher injection rates expand the reaction sweep range and facilitate channel breakthrough, whereas lower rates lead to more uniform front advancement. Higher reaction intensities enhance mineral dissolution near the inlet and prolong precipitation blocking periods, while lower intensities result in more frequent but smaller-amplitude dissolution-precipitation cycles. This study reveals the microscopic mechanisms of pore structure dynamic evolution driven by mineral reactions following CO₂ injection, clarifies the influence patterns of injection parameters on seepage capacity variation, and provides a theoretical basis for optimizing injection-production parameters and preventing formation damage during CO₂ injection in shale reservoirs.

With the large-scale development of deep coal gas resources in China, conventional hydraulic fracturing technologies have achieved initial breakthroughs; however, they still face a common industry challenge characterized by high initial production followed by rapid decline. Considering the unique reservoir characteristics of deep coal-measure gas, including adsorption dominance, dual-porosity media, and low permeability, this paper systematically analyzes the limitations of directly applying large-scale volumetric fracturing techniques, originally developed for shale gas, to deep coal gas reservoirs. A novel fracturing paradigm, Dilation-Governed Volume Fracturing (DGVF) is proposed, with the core objectives of matrix activation, flow field reconstruction, and long-term stable production. Rooted in the coupled stress-strain-permeability behavior of coal, this paradigm achieves matrix damage dilation by controlling fracture net pressure. It forms a new microfracture network without inducing structural failure and establishes an efficient and continuous flow path, encompassing the matrix, microfractures, cleats, hydraulic fractures, and wellbore. This paper details the theoretical framework, technical approach, and engineering practices of DGVF. The focus is on four key technologies: design of maximum flow capacity, natural fracture activation, coordinated regulation of injection rate and pressure, and intelligent control of production pressure drawdown. Based on experimental data from representative blocks in the Ordos Basin and Qinshui Basin, the paper demonstrates the significant effects of DGVF in enhancing single-well recovery and extending the stable production period. Finally, the paper discusses potential future developments for DGVF, providing theoretical insights and technical reserves for the efficient development of deep coal gas in China.

The fracturing of offshore low-permeability oilfields started relatively late. Due to the unique operational environment and high costs, the fracturing process requires efficient and accurate monitoring methods to achieve effective fracture characterization, reasonable effect evaluation, and ensure precise fracturing with maximized production-to-investment ratios. Currently, various monitoring technologies have been applied in offshore oilfields, including autonomous array acoustic logging technology and tracer production profile monitoring technology. Additionally, monitoring techniques such as micro-seismic, fiber-optic, water hammer pressure wave, and active sonar are still in the exploratory and research stages. Based on an analysis of the characteristics and current status of offshore low-permeability oilfields and fracturing, this paper reviews the research and application progress of various monitoring technologies. It also analyzes future development trends from multiple perspectives, including monitoring methods, hardware products, indoor experimental theoretical research, field tests, and intelligence, addressing the needs of offshore fracturing. This study provides valuable insights for advancing the development of offshore low-permeability oilfield technologies.

Addressing the issue of sand plugging susceptibility in the “acid fracturing + sand addition” composite stimulation technology applied to tight carbonate reservoir, particularly the degradation mechanism of fracturing fluid proppant-carrying performance under acidic fluid environments, this paper establishes a systematic and refined experimental evaluation methodology along with an optimization strategy. The effects of formation water, reservoir rock minerals, and residual acid from acid-rock reactions on the rheological properties of guar gum fracturing fluid were analyzed. Experimental results reveal that the residual acid following the reaction between the pre-flush acid and carbonate rocks is the primary factor causing fracturing fluid gel breaking; as little as 2 vol% residual acid leads to severe gel breaking. Formation water is the second most significant factor, with 10 vol% formation water intrusion inducing substantial gel breaking. Carbonate rock cuttings also cause a notable reduction in fluid viscosity: at a cuttings mass fraction of 10%, the viscosity decreases by approximately 60%. A comprehensive optimization strategy combining chemical modification of the fluid and physical isolation via pump sequencing is proposed, providing a solid experimental basis for the efficient stimulation of tight carbonate reservoirs.

Proppant transport and placement within hydraulic fractures directly determine effective fracture conductivity, stimulated reservoir volume utilization, and long-term post-fracturing productivity, making them critical factors governing the success of hydraulic fracturing in unconventional reservoirs. During hydraulic fracturing, proppant migration is a complex solid-liquid two-phase transport process involving particle settling, suspension, deposition, re-transport, and redistribution, which is jointly controlled by fluid properties, particle characteristics, fracture geometry, and operating conditions. In recent years, research on proppant transport has progressively evolved from investigations of particle settling and proppant bank development in idealized single fractures toward the characterization of heterogeneous placement behavior, multiscale simulation, and rapid prediction under complex fracture networks, rough-walled fractures, and dynamically propagating fracture conditions. Following the framework of experimental observation-mechanistic understanding-numerical simulation-intelligent prediction, this paper systematically reviews recent advances in experimental methods, key physical mechanisms, numerical simulation approaches, and data-driven prediction techniques for proppant transport and placement in hydraulic fractures. Experimental studies have established fundamental understandings of particle settling, proppant bed formation, equilibrium height evolution, and transport front migration in smooth single fractures. With increasing attention to realistic reservoir conditions, experimental systems have been extended to complex fracture networks, rough fracture surfaces, and reconstructed fractures with dynamic loading conditions, revealing the significant influence of fracture junction flow partitioning, rough-wall effects, heterogeneous aperture distribution, and local geometric constraints on proppant placement. Numerical simulation methods have further improved the understanding of proppant transport mechanisms beyond the limitations of experimental observations. Existing models can generally be classified into Eulerian-Eulerian and Eulerian-Lagrangian approaches. Eulerian-Eulerian methods, including mixture and two-fluid models, provide advantages in computational efficiency and engineering-scale prediction, whereas Eulerian-Lagrangian methods, including CFD-DEM, DDPM, and MP-PIC, retain particle-scale information and enable more detailed analysis of particle interactions, non-uniform placement, and local bridging or plugging mechanisms. However, achieving a balance between physical accuracy, computational efficiency, and engineering applicability remains a major challenge, particularly for dynamic fracture propagation, rough fracture surfaces, non-spherical particles, and multiscale fracture systems. In addition, intelligent prediction methods based on mechanism-data fusion have recently emerged as promising approaches for accelerating high-fidelity simulations and improving real-time decision-making. Surrogate models, physics-informed neural networks, and data assimilation techniques provide new possibilities for rapid prediction, parameter optimization, and dynamic model updating. Nevertheless, these approaches still face limitations related to data availability, physical consistency, and generalization capability under complex field conditions, and they cannot yet replace physics-based models. Current challenges mainly include insufficient characterization of realistic fracture geometries, incomplete coupling among dynamic fracture propagation, proppant transport, and fluid leakoff processes, limited representation of rough-wall effects and non-spherical particle behavior, and the lack of reliable field-scale validation data. Future research should integrate high-fidelity experiments, advanced numerical simulations, physics-constrained surrogate models, and field data assimilation to establish a unified multiscale and multiphysics framework, thereby improving the prediction capability of proppant placement and supporting hydraulic fracturing design optimization and real-time operational decision-making.

During the production of deepwater gas fields, it is usually necessary to predict a series of key production parameters according to future production allocation plans, namely future gas demand. These parameters include downhole pressure and temperature, wellhead pressure and temperature, gas production rate, oil production rate, and water production rate. Accurate prediction of these parameters can assist engineers in rapidly evaluating existing production allocation schemes and formulating new ones, thereby providing direct support for production decision-making in deepwater gas fields. However, dynamic production parameter prediction for deepwater gas wells still faces two major challenges. On the one hand, as gas field development continues, reservoir and wellbore conditions continuously evolve, and the statistical characteristics of key production parameters as well as the coupled response relationships among multiple parameters exhibit non-stationary changes at different development stages. Such time-varying shifts significantly affect the generalization capability and long-term adaptability of prediction models. On the other hand, changes in production parameters are constrained by future production allocation plans. Therefore, effectively extracting and modeling the correlations between production allocation plans and multidimensional production parameters remains a key factor limiting further improvement in prediction performance. To address the above problems, this paper proposes X-NLinear, a dynamic production parameter prediction model for gas wells based on multidimensional temporal feature fusion. The model inherits the lightweight linear structure of NLinear and its dynamic normalization mechanism based on sample-specific statistical characteristics, thereby reducing the influence of non-stationary changes in production parameters on prediction results. On this basis, a cross-variable feature extraction module is further designed to integrate future production allocation plans with multiple types of production parameter information, capturing the dynamic influence of gas demand changes on pressure, temperature, and oil, gas, and water production rates. Based on real production data from deepwater gas wells, X-NLinear is compared with representative time-series forecasting models. The experimental results show that X-NLinear improves the accuracy and stability of multi-parameter joint prediction for deepwater gas wells while maintaining computational efficiency. In addition, the cross-variable feature extraction module further reduces prediction errors in terms of mean absolute error and symmetric mean absolute percentage error. The proposed method can provide technical support for rapid evaluation of production allocation schemes and production optimization in deepwater gas fields.

Methane emission sources across oil and gas systems are widely dispersed, and episodic releases are common. Observations obtained at different spatial and temporal scales are therefore difficult to reconcile. Effective mitigation depends on connecting emission detection, emissions quantification and accounting, and mitigation implementation. This review examines publications indexed in the Web of Science Core Collection from 2015 to 2024. It combines a bibliometric sample of 5,600 papers with a thematic review. The analysis identifies annual trends and thematic structures and compares major technical pathways, applicability boundaries, and industry practices in monitoring, accounting, and governance. The findings show that monitoring has developed from equipment inspections and point-source detection into a multiscale system integrating ground-based, mobile, airborne, satellite, and continuous observations. Its functions have expanded from emission detection to regional screening, facility-level attribution, and on-site response. Emissions quantification and accounting have progressed from emission-factor and activity-data approaches to facility- and process-based accounting, supply-chain assessment, and measurement-informed inventories. Accounting outputs are no longer limited to total emissions but increasingly include emission intensity and methane footprints, with uncertainty ranges also being reported more explicitly. Governance has moved from individual engineering measures toward coordination between engineering mitigation and policy instruments, with monitoring, reporting and verification mechanisms linking implementation responsibilities to performance feedback. Further analysis shows that mitigation outcomes are also constrained by interfaces among these stages. Incorporating monitoring evidence into accounting requires scale conversion, source attribution, and adequate temporal representativeness. The accounting process also requires consistent system boundaries and appropriate uncertainty propagation. Applying accounting results to governance requires alignment among data-quality requirements, criteria for recognizing reported outcomes, and responsibility rules. Oil and gas companies should therefore use facility registers to connect anomaly detection, field confirmation, repair actions, post-repair measurements, annual inventories, and information disclosure. Regulators and verification bodies should further clarify requirements for boundary adjustments and outcome verification. Effective oil and gas methane mitigation requires a continuous mechanism linking emission detection, emissions quantification and accounting, governance responses, and outcome verification. This mechanism provides a traceable evidence base for corporate asset management, industry regulation, and supply-chain methane mitigation.