The oil-based mud filtrate will invade the formation under the overbalanced pressure during drilling operations. As a result, alterations will occur to the nuclear magnetic resonance (NMR) response characteristics of the original formation, causing the relaxation time of the NMR T₂ spectrum of the free fluid part to move towards a slower relaxation time. Consequently, the subsequent interpretation and petrophysical evaluation will be heavily impacted. Therefore, the actual measured T₂ spectrum needs to be corrected for invasion. For this reason, considering the low-porosity and low-permeability of sandstone gas formations in the East China Sea as the research object, a new method to correct the incorrect shape of the NMR logging T₂ spectrum was proposed in three main steps. First, the differences in the morphology of the NMR logging T₂ spectrum between oil-based mud wells and water-based mud wells in adjacent wells were analyzed based on the NMR relaxation mechanism. Second, rocks were divided into four categories according to the pore structure, and the NMR logging T₂ spectrum was extracted using the multidimensional matrix method to establish the T₂ spectrum of water-based mud wells and oil-based mud wells. Finally, the correctness of the method was verified by two T₂ spectrum correction examples of oil-based mud wells in the study area. The results show that the corrected NMR T₂ spectrum eliminates the influence of oil-based mud filtrate and improves the accuracy of NMR logging for calculating permeability.

Tight sandstone reservoirs exhibit strong vertical heterogeneity and complex pore structures, challenging conventional permeability evaluation methods based on well-logging data. Although rising machine-learning (ML) techniques have demonstrated excellent accuracy for industrial applications, the physics and rationality within such a powerful “black box” remain less clear. Hence, reliable permeability prediction would benefit from an interpretable ML-based workflow that could reveal the controlling factors. To compare the models and examine the underlying features, 16 different ML submodels are tested after data preprocessing, feature selection, and hyperparameter optimization. By comparing the fitting accuracy and tuning time, the light gradient boosting machine optimized by the whale optimization algorithm, referred to as LGB-WOA, is determined to be the optimal model with the best fitting accuracy and relatively short tuning time. A field data application demonstrates that even in highly heterogeneous reservoir sections, the LGB-WOA model outperformed conventional petrophysical models by being the most consistent with reservoir permeability directly measured from the core samples (R2>0.6). The Shapley additive explanation values are then used to interpret the predictions of our LGB-WOA model. As expected, the porosity curve exhibits the highest feature importance among all input features, significantly contributing to permeability predictions. Conversely, a wellbore diameter and compensated neutron log contribute the least and need not be used for subsequent model improvements. These experiments and workflow provide a powerful method for accurately assessing the permeability in complex reservoirs and contribute to a broader understanding of the application of ML in reservoir characterization, paving the way for establishing more interpretable and reliable prediction models.

Core samples of the tight sandstone reservoir in the Bonan Oilfield were analyzed by using multiple petrophysical experimental techniques, then a multi-scale three-dimensional digital rock model was constructed. The pore structure parameters of tight sandstone and homogeneous Berea sandstone were compared. The electrical simulation method based on the digital rock model was utilized to quantitatively reveal the influence of five micro-pore structure parameters (pore size, throat size, pore-throat size, coordination number, and shape factor) on the rock’s electrical properties. In addition, the saturation of tight sandstone reservoirs was evaluated in combination with the three-component automatic mixed-connection conductivity model. The results show that the “non-Archie” phenomenon in sandstone is obvious, which is mainly caused by the small radius of the maximum connected pore throat and the complex structure of the pore throat. We noted that: with an increase in pore radius, throat radius, and coordination number, the formation factor decreases and tends to be stable; the pore-throat size increases and the formation factor decreases in the form of power function; the shape factor increases, and the formation factor increases; the larger the pore–throat ratio and shape factor, the greater the resistivity index; with an increase in coordination number, the resistivity index decreases; and the pore-throat size has no effect on the resistivity index. The calculation accuracy of oil saturation is improved by 6.54% by constructing the three-component automatic mixed-conductivity saturation model of tight sandstone.

Heterogeneous conglomeratic sandstone exhibits anisotropic physical properties, rendering a comprehensive analysis of its physical processes challenging with experimental measurements. Digital rock technology provides a visual and intuitive analysis of the microphysical processes in rocks, thereby aiding in scientific inquiry. Nevertheless, the multiscale characteristics of conglomeratic sandstone cannot be fully captured by a single-scale digital rock, thus limiting its ability to characterize the pore structure. Our work introduces a proposed workflow that uses multiscale digital rock fusion to investigate permeability anisotropy in heterogeneous rock. We use a cycle-consistent generative adversarial network to fuse computed tomography scan data of different resolutions, creating a large-scale and high-precision digital rock that comprehensively represents the conglomeratic sandstone pore structure. Subsequently, the digital rock is partitioned into multiple blocks, and the permeability of each block is simulated using a pore network. Finally, the total permeability of the sample is calculated by conducting an upscaling numerical simulation using the Darcy-Stokes equation. This process facilitates the analysis of the pore structure in conglomeratic sandstone and provides a step-by-step solution for permeability. From a multiscale perspective, this approach reveals that the anisotropy of permeability in conglomeratic sandstone stems from the layered distribution of grain sizes and differences in grain arrangement across different directions.