A precise assessment of tight gas operational efficiency is critical for investment decisions in unconventional reservoir development. However, quantifying production efficiency remains challenging due to the complex relationships between geological and operational factors. This study proposes a novel data-driven framework for predicting tight gas productivity, effectively integrating computing algorithms, machine learning algorithms, feature selection, production prediction and fracturing parameter optimization. A dataset of 3146 horizontal wells from the Montney tight gas field was used to train six machine learning models, aiming to identify the most significant factors. Results indicate that fluid-injection volumes, burial depth, number of stages, Young’s modulus, formation pressure, saturation, sandstone thickness and total organic carbon are the key variables for tight gas production. The Random Forest-based model achieved the highest accuracy of 88.6%. Case studies for the test demonstrate well that gas production could be nearly doubled by increasing fracturing fluid injection by 97.5%. This work provides evidence-based recommendations to refine development strategies and maximize reservoir performance.

The mechanisms underlying formation energy depletion after initial fracturing and post-refracturing production decline in shale oil horizontal wells remain poorly understood. This study proposes a novel numerical simulation framework for refracturing processes based on a three-dimensional fully coupled hydromechanical model. By dynamically reconfiguring the in situ stress field through integration of production data from initial fracturing stages, our approach enables precise control over fracture propagation trajectories and intensities, thereby enhancing reservoir stimulation volume (RSV) and residual oil recovery. The implementation of fully coupled hydromechanical simulation reveals two critical findings: (1) the 70 m fracture half-length generated during initial fracturing fails to access residual oil-rich zones due to insufficient fracture network complexity; (2) a 3–5° stress reorientation combined with reservoir repressurization before refracturing significantly improves fracture network interconnectivity. Field validation demonstrates that refracturing extends fracture half-lengths to 97–154 m (38–120% increase) and amplifies RSV by 125% compared to initial operations. The developed seepage–stress coupling methodology establishes a theoretical foundation for optimizing repeated fracturing designs in unconventional reservoirs, providing critical insights into residual oil mobilization through engineered stress field manipulation.

As oil resources continue to be depleted, traditional extraction technologies face significant challenges, and improving remaining oil recovery has become a critical issue in the development of low-permeability oil fields. The study first establishes a stratigraphic framework through a fine stratigraphic correlation of key wells, followed by a comprehensive characterization of the internal arrangement of sand bodies, with a detailed analysis of the distribution of interlayers and the complexity of sand body stacking patterns. Based on field data, including tracer monitoring and water absorption profiles, the distribution of remaining oil is predicted. The results reveal that sandy debris flow deposition plays a key role in the enrichment of remaining oil, particularly considering how its internal structure impacts reservoir connectivity and waterflood efficiency. The sedimentary pattern in the study area is dominated by debris flow deposition, with poor continuity of sand bodies; over 90% of individual sand bodies encountered during horizontal well drilling have a width of less than 60 m. Sand body stacking types are classified into overlapping, splicing, and isolated types, and different stacking types significantly influence reservoir waterflood efficiency and the distribution of remaining oil. The innovation of this study lies not only in the microstructural investigation of sandy debris flow but also in the comprehensive characterization of its strong heterogeneity. By analyzing the internal configuration of sand bodies, sedimentary microfacies, and physical property distributions, this study uncovers how these factors influence the distribution of remaining oil. This approach provides valuable insights for accurately predicting and enhancing remaining oil potential.

The application of CO2-enhanced oil recovery (EOR) in in the highly heterogeneous lacustrine shales is challenged by severe gas channeling through natural fractures, poor sweep efficiency, and low net carbon sequestration by the conventional operation mode. To address the issues, this study proposes a Closed-Loop CO2 Recycling (CLCR) technology, in which the produced gas (including CO2 and light hydrocarbons) is captured, compressed, and reinjected to offset the total CO2 injection amount. A compositional cross-scale numerical model is developed that implicitly couples flow in the shale matrix, bedding fractures, and hydraulic fractures, with pressure-driven and diffusion-controlled mass transfer among the continua. The model is validated against field history data from a Gulong shale, which is the most typical shale in China, with a matching accuracy over 85%. Based on a 20-year simulation, three metrics are evaluated including oil replacement ratio (ORR), CO2 sequestration efficiency (CSE), and oil production multiplier (OPM). The results show that an injection rate of 50 t/d in a 1650 m horizontal well gives the highest OPM of 2.39, while early CO2 injection without prior depletion is confirmed essential to fracture conductivity preservation and gas breakthrough delay. By CLCR, the CSE increases from 58% to over 99% for all recycling triggers. The reinjection starting when the produced gas contains 5% CO2 in mole fraction achieves the highest ORR of 0.152 t oil/t CO2 and a very competitive OPM of 2.612. Early recycling preserves light hydrocarbons in the reinjected gas, which improves oil-gas compatibility, strengthens the displacement efficiency, and enhances oil mobilization in the tight matrix. CLCR enables nearly complete CO2 sequestration while increasing oil production, offering a viable pathway toward carbon-neutral or even zero-carbon-emission shale oil development.

Seismic inversion is vital for reservoir characterization but faces significant challenges in complex fluvial-deltaic systems due to strong heterogeneity and thin-bedded formations. Current methods, including convolution-based, geostatistical, and artificial neural network (ANN) approaches, are often limited by wavelet stationarity assumptions, spatial uncertainty, and physical implausibility. This study develops a novel artificial intelligence (AI) seismic inversion algorithm that integrates a convolutional physical model with data-driven learning to overcome these drawbacks. The proposed physics-guided hybrid model employs a multi-wavelet inversion framework, incorporating 8–10 spatially variable wavelets per inversion cell to account for lateral wavelet variability. These physically constrained inversion candidates are then intelligently fused using a computationally efficient neural network, which maintains a 3.4% training error and 4.7% validation accuracy. This integrated approach achieves remarkable improvements: a 30% enhancement in vertical resolution enabling 1–3m thin-bed detection, a 40% improvement in lateral continuity (with correlation coefficients increasing from 0.85), and 70% better noise suppression. Application in a complex fluvial-deltaic system covering 7.2 km ² with 80 wells confirmed the method’s robustness, delivering over 80% accuracy in sandbody prediction while significantly reducing geologically implausible results.

The WieZhou12 oilfield (also known as WZ12 oilfield, the same below) is in urgent need of development using large-scale volumetric fracturing technology since it is a typical complex fault-block oilfield with low porosity, low permeability, and no natural production capacity. To study the fracturing measures with surfactants in offshore oilfields like WZ12, the surfactant fracturing fluid types were experimentally selected based on their effect of decreasing interfacial tension and enhancing matrix wettability. The water cut law and oil displacement efficiency in displacement experiments were also analyzed, according to surfactant type and fluid characteristics. Next, using the numerical simulation software CMG, the study completed the integrated simulation of volumetric fracturing in the “injection–soaking–flowback” process. Finally, some critical parameters were optimized for the block model, including the quantity of injected fluids, the soaking time, and the rate of fluid flowback. The results showed that the most suitable surfactant was 0.5% ammonium lauryl polyether sulfate (ALES), which had a low interfacial tension of 1.7 × 10⁻² mN/m, a contact angle of 20.071° with the core, and a 52% oil displacement efficiency. From the simulations, the suggested production parameters for energy storage fracturing are as follows: a daily injection volume of 600 m³/d, a soaking time of 25 days post fracturing, and a fluid production rate of 270 m³/d. The findings of this study establish a significant theoretical foundation for optimizing surfactant type and provide construction advice for the integrated measure of fracturing, well shut-in, and production in offshore oilfields.