Multi-modal combined imaging method and platform for shale nanopore three-dimensional reconstruction and connectivity quantitative evaluation

Abstract

The invention belongs to the technical field of multi-modal combined imaging, and discloses a multi-modal combined imaging method and platform for shale nano-pore three-dimensional reconstruction and connectivity quantitative evaluation, and aims to realize shale organic pore and inorganic pore structure three-dimensional and quantitative characterization based on FIB-SEM and X-CT. High-resolution imaging and chemical imaging technologies are used for revealing the occurrence state of kerogen in mineral particles or at the boundaries, the mineral surface catalyzes organic matter to generate hydrocarbon, and the pore evolution mechanism is regulated and controlled.

Description

Technical Field

[0001] This invention belongs to the field of multimodal imaging technology, and in particular relates to a multimodal imaging method and platform for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores. Background Technology

[0002] Compared to conventional oil and gas, unconventional oil and gas resources, represented by shale oil and gas, have become a crucial direction for theoretical and technological innovation in the global oil and gas industry due to their existence within extremely fine pores and complex pore networks. Efficient exploration and development of shale oil and gas plays a key role in promoting energy structure transformation. However, the micro- and nano-pore structures of shale reservoirs are complex and diverse, and their geometry, size distribution, and connectivity directly affect the occurrence state, seepage behavior, and enrichment conditions of fluids. Especially in organic-rich shale, organic and inorganic pores are typically tightly interwoven at the nanometer to micrometer scale. During the thermal maturation of organic matter, the pore network evolves, accompanied by mineral dissolution, recrystallization, and organic-inorganic interface interactions. These interface processes significantly influence pore formation and connectivity. The pore system includes not only isolated pores but also complex branched pore-throat structures with significant differences in connectivity, making it difficult for traditional two-dimensional characterization methods to fully reflect its three-dimensional spatial characteristics.

[0003] Currently, the characterization of shale pore structure mainly employs methods such as scanning electron microscopy (SEM), nitrogen adsorption, and high-pressure mercury intrusion porosimetry. These techniques can reveal two-dimensional information about pore morphology and macroscopic pore parameters, but they have significant limitations in spatial scale, three-dimensional reconstruction, and connectivity evaluation. For example, SEM and traditional two-dimensional microscopic observation methods can only provide planar images of slices, failing to accurately reflect the three-dimensional spatial distribution and connectivity paths of pores. While porosity measurement techniques such as nitrogen adsorption and mercury intrusion porosimetry can quantitatively obtain pore volume parameters, their ability to characterize the complexity of pore shape and pore-crack structure information is limited. Consequently, it is difficult to accurately characterize the microscopic mechanisms such as the spatial coupling relationship between organic and inorganic pores, the spatial topological characteristics of the pore network, and the connectivity of dominant seepage channels.

[0004] In recent years, micro / nano-scale imaging and three-dimensional digital core technologies, such as micro / nano-X-ray computed tomography (micro / nano-CT) and focused ion beam scanning electron microscopy (FIB-SEM), have been introduced to attempt to reveal the spatial network characteristics of shale pore structure from a three-dimensional perspective. FIB-SEM, through nanoscale continuous slice imaging of samples, can obtain high-resolution two-dimensional image sequences, which can be used for three-dimensional reconstruction of nanoscale pore structures. Based on digital rock technology, pore parameters can be extracted and network analysis performed. In some studies, this has been used to construct pore network models and statistically analyze information such as pore size distribution and connectivity parameters.

[0005] Despite the progress made by the aforementioned 3D imaging technologies in shale porosity characterization, existing technologies still have significant shortcomings. On the one hand, a single imaging method cannot simultaneously satisfy the contradiction between large field-of-view representativeness and nanometer-level resolution: for example, conventionally sized FIB-SEM reconstruction models can usually only cover a very small volume, and when sample heterogeneity is significant, their pore parameters are insufficient to represent the overall reservoir, making it difficult to fully reflect the true pore network structure. ([Frontiers][2]) On the other hand, existing imaging and analysis workflows for pore connectivity evaluation are still limited to simplified geometric parameter statistics or connected domain analysis based on a single scale, failing to effectively combine nanometer- to micrometer-level multi-scale pore information for comprehensive quantitative evaluation of connectivity. Furthermore, issues such as image registration, spatial fusion, and noise and artifact handling between different imaging modalities remain key technical challenges in achieving high-precision 3D reconstruction and network extraction, and these problems have not yet been systematically solved in existing technologies.

[0006] Therefore, although traditional imaging and 3D reconstruction techniques can reveal the local characteristics of shale pores to some extent, they cannot clearly reveal the spatial distribution of the 3D structure of organic and inorganic pores in shale and their true connectivity characteristics. Furthermore, a mature and quantifiable multimodal joint imaging and connectivity quantitative evaluation system has not yet been established. These technical deficiencies restrict the accurate understanding of the microscopic pore network structure of shale reservoirs and the laws governing hydrocarbon occurrence / migration, as well as the ability to predict shale using digital rock physics. Summary of the Invention

[0007] To address the problems existing in the prior art, this invention provides a multimodal joint imaging method and platform for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores.

[0008] This invention is achieved as follows: a multimodal joint imaging method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores includes:

[0009] Step 1, Image Acquisition;

[0010] Step 2, 3D reconstruction: Combining FIB-SEM technology with CT digital core reconstruction technology, the shale sample is digitally reconstructed to form a 3D digital model;

[0011] Step 3, Connectivity and Quantitative Evaluation:

[0012] The porosity and average pore size parameters of the sample were obtained by characterizing it using a 3D digital model, and the pore connectivity was analyzed.

[0013] Furthermore, the image acquisition:

[0014] In the images after X-CT scan, based on the region of interest, SEM-Maps technology is used to perform continuous and partially overlapping ultra-high precision small-area scans in a certain scanning order, and then computer technology is used to stitch them together to form a two-dimensional scan image with a large field of view and high resolution.

[0015] High-resolution two-dimensional images were obtained using FIB-SEM, such as those acquired during the analysis of pyrolytic and non-pyrolytic oil shale samples from the Green River Shale and the Longmaxi Formation shale in the southern Sichuan Basin.

[0016] Furthermore, the connectivity and quantitative evaluation:

[0017] This study analyzes the development mechanism of organic-inorganic nanopores, compares the effects of different shale gas development, and explains the differences using the number and volume ratio of organic pores. It also analyzes the indirect influence of quartz and clay minerals on the development and distribution of nanopores, the characteristics of inorganic and organic pores, and the protective effect of the proportion of silica in heterogeneous shale on organic pores.

[0018] Another objective of this invention is to provide a multimodal joint imaging platform for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores, comprising:

[0019] The image acquisition module is used to perform continuous and partially overlapping ultra-high precision small-area scanning in a certain scanning order using SEM-Maps technology, and then use computer technology to stitch together a two-dimensional scan image with a large field of view and high resolution.

[0020] The 3D reconstruction module is used to combine FIB-SEM technology with digital core reconstruction technology to complete the digital reconstruction of shale samples and form a 3D digital model.

[0021] The connectivity and quantitative evaluation module is used to characterize samples using 3D digital models, obtain sample porosity and average pore size parameters, and analyze pore connectivity.

[0022] Another object of the present invention is to provide a computer device comprising a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the steps of the multimodal joint imaging method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores.

[0023] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the multimodal joint imaging method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores.

[0024] Another objective of this invention is to provide an information data processing terminal, which is used to realize a multimodal joint imaging platform for three-dimensional reconstruction and quantitative evaluation of the connectivity of shale nanopores.

[0025] Based on the above technical solutions and the technical problems solved, the advantages and positive effects of the technical solution to be protected by this invention are as follows:

[0026] The core challenge addressed by this invention lies in the fact that shale reservoirs are characterized by small pore size, complex pore types, and the combination of organic matter and minerals, resulting in high heterogeneity. Furthermore, the precise characterization of organic and inorganic pores presents significant challenges, and single imaging or single-parameter characterization struggles to simultaneously consider both field of view and resolution. This makes it difficult to simultaneously, quantitatively, and reproducibly analyze the organic matter occurrence state, pore network connectivity, and organic-inorganic interface mechanisms. Simultaneously, pore evolution is controlled by thermal maturation, mineral catalysis, and fluid interactions. Traditional two-dimensional imaging analysis and statistics often fail to distinguish between isolated pores and interconnected pore networks, and it is also difficult to establish an integrated correlation between chemical composition, mechanical properties, and pore structure. Consequently, there is a lack of verifiable structural evidence to explain the discrepancies between reservoir evaluation and development.

[0027] To address the aforementioned issues, this invention employs X-ray tomography to provide overall structural constraints, utilizes SEM mosaicking to obtain high-resolution two-dimensional structural images under a large field of view, and then combines FIB-SEM continuous slicing to obtain nanoscale three-dimensional details. Through multimodal registration and fusion, a unified coordinate system fusion dataset is established, and finally, a three-dimensional voxel model is formed by digital core reconstruction. Based on this, connected component identification and quantitative parameter calculation are performed, fundamentally solving the problems of structural omissions and scale distortion caused by the contradiction between resolution and field of view. This enables traceable and comparable accurate representation of organic pores, inorganic pores, and mineral skeletons in the same three-dimensional space.

[0028] In terms of overcoming difficulties, the key breakthrough of this invention lies in elevating "structural visualization" from a single image display to "cross-modal consistent modeling and computable evaluation." On the one hand, multimodal data achieves spatial alignment through coarse registration driven by mutual information and fine registration constrained by deformation field, avoiding mismatches caused by differences in grayscale mechanisms between different modes and ensuring the geometric consistency of 3D reconstruction. On the other hand, through automatic segmentation and voxelization of pore phase and mineral phase, nanopores are transformed from two-dimensional qualitative judgments into three-dimensional quantitative computational objects, which can stably output parameters such as porosity, pore size distribution, and maximum connected domain volume fraction, and further distinguish between "isolated pores" and "connected pore networks," thereby directly addressing the pain point of difficulty in quantifying "effective pore contribution" in shale development. More importantly, this three-dimensional model provides a verifiable platform for multi-scale evaluation of the development mechanisms of organic and inorganic pores in different types of shale reservoirs. It enables the phenomena observed during the research and development process to be repeatedly verified in the form of structural parameters. For example, it can explain the development differences in different blocks by the number of organic pores and the spatial volume fraction, assess the indirect control of quartz and clay on pore development by the correlation between mineral phase distribution and pore connectivity, and further characterize the protective effect of the siliceous rigid framework on the organic pores against compaction and collapse.

[0029] The resulting innovative technological effect is reflected in the "integrated quantitative characterization and interpretable correlation of non-destructive multidimensional information." This invention not only obtains pore geometry but also integrates high-resolution electron microscopy (SEM), atomic force microscopy (AFM), and chemical imaging information within the same framework. This enables the synergistic characterization of organic matter morphology, pore structure, and mechanical properties, clearly revealing the occurrence state of kerogen within mineral particles or at their boundaries. Furthermore, by combining the differences in chemical composition and structural connectivity characteristics of the interface region, it can quantitatively support the judgment that mineral surfaces catalyze organic matter to generate hydrocarbons and regulate pore evolution. It also transforms the influence of chemical bond coupling and intermolecular forces at the organic-inorganic interface on hydrocarbon retention and migration into observable and calculable structural evidence. Compared to traditional two-dimensional statistical or single-modal analysis, this invention achieves cross-scale pore connectivity and calculable evaluation while ensuring non-destructive and high resolution. This shifts reservoir evaluation from empirical judgment to structure-driven quantitative interpretation, significantly improving the objectivity, reproducibility, and engineering applicability of shale oil and gas reservoir evaluation.

[0030] (1) The expected benefits and commercial value of the technical solution of this invention after transformation are as follows:

[0031] Expected benefits: After the technology transfer, relying on the existing technology and equipment in this laboratory, the risks of reservoir evaluation in shale oil and gas exploration and development can be reduced, and the efficiency of shale oil and gas exploration and development can be improved, thereby increasing revenue for the oil field.

[0032] Commercial Value: This invention will solve the problem of shale oil and gas reservoir evaluation, and will be widely applied in shale oil and gas exploration and development. It will promote large-scale exploration and development of shale oil and gas in deep and complex structural areas, fill the gaps in related technologies and understanding in China, and improve the overall understanding of shale oil and gas reservoirs in my country.

[0033] (2) The technical solution of the present invention solves the technical problem that people have long wanted to solve but have never been able to solve: the problem of multi-scale three-dimensional characterization and connectivity analysis of organic and inorganic pores in shale oil and gas reservoirs. Attached Figure Description

[0034] Figure 1 This is a flowchart of a multimodal joint imaging method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores provided in an embodiment of the present invention.

[0035] Figure 2 This is a block diagram of the multimodal joint imaging platform for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores provided in this embodiment of the invention.

[0036] Figure 3 These are the resolutions and main functions of the different imaging and analysis technologies provided in the embodiments of the present invention.

[0037] Figure 4 It is the effective observation scale of the imaging device provided in the embodiments of the present invention (FEI Corporation, 2013).

[0038] Figure 5 The elemental composition of the Tasmanite cysts in the Woodford Shale, as analyzed by scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) according to the embodiments of the present invention is shown in (a, b) as Tasmanite cysts observed in samples pyrolyzed at 340°C; (c, d) as iron-bearing minerals detected in samples pyrolyzed at 300°C; and (e, f) as Tasmanite structures filled with silica in the original samples.

[0039] Figure 6 The following are scanning electron microscope images of organic matter in the Cambrian Qiongzhusi Formation shale of the Sichuan Basin provided in this embodiment of the invention: A. Algolium, a carbonaceous fossil, with the upper semi-circular mineral being organic matter, and the twisted part on the right also being algolium. The discovery of algolium indicates that the rock had life activity in the environment during the early diagenesis and became a carbon-based fossil; B. Thick platy algolium; C. Club-shaped algolium; D. Energy dispersive spectroscopy analysis of club-shaped algolium.

[0040] Figure 7 This is a three-dimensional morphology of shale surface under an atomic force microscope provided in an embodiment of the present invention.

[0041] Figure 8These are transmission electron microscope images of ultrathin sections of shale provided in embodiments of the present invention.

[0042] Figure 9 The following are micron-CT characterization sample pore network and three-dimensional connected body model diagrams (segmented superimposed two-dimensional grayscale images of core) provided in the embodiments of the present invention: a) segmented display model of pore network; b) segmented display model of pore connectivity.

[0043] Figure 10 The red box shows the selection of the FIB experimental area using FESEM (Field Emission Scanning Electron Microscope) provided in this embodiment of the invention.

[0044] Figure 11 This is a description of FIB three-dimensional image reconstruction and image analysis provided in the embodiments of the present invention. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0046] The shale nanopore three-dimensional reconstruction and connectivity quantitative evaluation platform of the present invention realizes fine three-dimensional reconstruction of the nanoscale pore structure inside shale and quantitative calculation of connectivity parameters through the collaborative processing of multimodal image data. Its overall workflow consists of four stages: image acquisition, data processing, analysis and calculation, and data storage.

[0047] First, the image acquisition unit acquires low-resolution volumetric data, high-resolution two-dimensional surface images, and continuous two-dimensional slice images, respectively. The low-resolution volumetric data provides a reference for the overall spatial structure and scale of the sample, the high-resolution two-dimensional surface images characterize the fine morphological features of the pores, and the continuous two-dimensional slice images reflect the continuity of the pore structure along the depth direction. These multi-source heterogeneous data differ in spatial resolution, imaging mechanism, and viewing angle; therefore, a unified data processing workflow is required for their fusion and utilization.

[0048] Subsequently, the data processing unit performs scale normalization and spatial registration operations on the acquired multimodal image data, aligning data from different sources within a unified coordinate system. This process includes scaling pixel dimensions and correcting spatial positions through feature matching or geometric constraints, ensuring consistent representation of different modal data at the same spatial location. After registration, the multimodal data undergoes fusion processing to comprehensively utilize the overall structural information of the low-resolution volumetric data with the detailed information of the high-resolution image, improving the completeness and accuracy of pore structure representation.

[0049] Based on the fused data, the data processing unit performs pore phase segmentation to distinguish pores from the matrix in the image, obtaining the spatial distribution of pores. The segmentation model is constructed using supervised learning and periodically updated using manually labeled samples to adapt to differences in the composition of different shale samples and imaging conditions, improving the stability and accuracy of the segmentation results. Based on the spatial information of pores obtained from the segmentation, a three-dimensional pore structure model is further constructed to achieve a visual representation of shale nanopores in three-dimensional space.

[0050] After obtaining the three-dimensional pore structure model, the analysis unit performs connectivity analysis and structural parameter calculations on the model to identify the spatial connectivity relationships between different pores. Based on this, it calculates key indicators such as porosity, pore size distribution, and connectivity parameters to quantitatively evaluate the pore structure characteristics and fluid migration potential of shale reservoirs. These parameters reflect the spatial complexity of the pore network and its impact on reservoir performance.

[0051] Finally, the storage unit uniformly stores the original image data, the fused intermediate data, and the final reconstructed 3D model and parameter results to support result traceability, comparative analysis, and subsequent applications. Through the above collaborative workflow, this invention realizes a complete technical chain from multimodal image acquisition to 3D pore structure reconstruction and then to quantitative connectivity evaluation, improving the accuracy and reliability of shale nanopore characterization and providing effective technical support for shale reservoir evaluation.

[0052] like Figure 1 As shown in the figure, the multimodal joint imaging method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores provided by the present invention includes the following steps:

[0053] S101, Image Acquisition;

[0054] S102, 3D Reconstruction: Combining FIB-SEM technology with digital core reconstruction technology, the shale sample is digitally reconstructed to form a 3D digital model;

[0055] S103, Connectivity and Quantitative Evaluation:

[0056] The porosity and average pore size parameters of the sample were obtained by characterizing it using a 3D digital model, and the pore connectivity was analyzed.

[0057] The multimodal joint imaging method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores described in this invention works on the following principle: by spatially registering and fusing micro-nano-scale imaging data with different resolutions and imaging mechanisms, a three-dimensional digital model that truly reflects the structural characteristics of shale pores is constructed. Based on this model, quantitative characterization of pore geometric parameters and connectivity characteristics is achieved, thereby overcoming the problems of limited resolution, insufficient information dimension, and difficulty in accurately characterizing connectivity of single imaging methods.

[0058] First, in step S101, raw image data of the shale sample under multi-scale imaging conditions are acquired to provide basic information for subsequent reconstruction. Specifically, FIB-SEM technology is used for layer-by-layer cutting and high-resolution scanning imaging of the sample, obtaining two-dimensional sequential images of pore morphology and solid skeleton distribution at the nanoscale. Simultaneously, digital core imaging data is introduced to supplement information on larger-scale pore structure and spatial continuity. Through the coordinated acquisition of these multi-modal images, full-scale coverage of the shale pore structure from the nanoscale to the microscale is achieved.

[0059] Secondly, in step S102, three-dimensional reconstruction processing is performed based on the acquired two-dimensional sequence images. Specifically, the FIB-SEM sequence images are denoised, contrast-enhanced, and thresholded to distinguish between the porous phase and the mineral matrix phase; then, registration and stacking are performed according to the spatial order between slices to reconstruct the three-dimensional geometric structure of the nanopores. Simultaneously, the larger-scale pore structure obtained from the digital core reconstruction is scale-matched and fused with the aforementioned nanoscale three-dimensional model to construct a unified three-dimensional digital core model. This three-dimensional model can simultaneously reflect the spatial morphology, size distribution, and interconnections of the pores, thereby realistically restoring the shale pore network structure.

[0060] Finally, in step S103, a quantitative analysis of the pore structure is performed based on a three-dimensional digital model. The overall pore volume fraction is calculated using voxel statistics to obtain porosity parameters, and the average pore diameter and pore size distribution characteristics are calculated using the equivalent sphere diameter or maximum inscribed sphere method. Simultaneously, connected component analysis or skeleton extraction algorithms are used to identify the connecting paths between pores, evaluating the connectivity, continuity, and distribution of dominant flow channels in the pore network. This achieves a comprehensive quantitative evaluation of the geometric characteristics and connectivity properties of the shale nanoporous structure, providing reliable structural foundation data for the analysis of shale reservoir performance and seepage behavior.

[0061] Image acquisition provided by embodiments of the present invention:

[0062] In the images after X-CT scan, based on the region of interest, SEM-Maps technology is used to perform continuous and partially overlapping ultra-high precision small-area scans in a certain scanning order, and then computer technology is used to stitch them together to form a two-dimensional scan image with a large field of view and high resolution.

[0063] High-resolution two-dimensional images were obtained using FIB-SEM, such as those acquired during the analysis of pyrolytic and non-pyrolytic oil shale samples from the Green River Shale and the Longmaxi Formation shale in the southern Sichuan Basin.

[0064] The connectivity and quantitative evaluation provided by the embodiments of the present invention:

[0065] This study analyzes the development mechanism of organic-inorganic nanopores, compares the effects of different shale gas development, and explains the differences using the number and volume ratio of organic pores. It also analyzes the indirect influence of quartz and clay minerals on the development and distribution of nanopores, the characteristics of inorganic and organic pores, and the protective effect of the proportion of silica in heterogeneous shale on organic pores.

[0066] This invention provides a multimodal joint imaging method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores. Its working principle is to compensate for the limitations of single imaging technology in terms of resolution, field of view and imaging depth by collaboratively acquiring and fusing multi-scale and multimodal imaging data, thereby realizing the true three-dimensional reconstruction and quantitative characterization of the connectivity of shale nanopore structure.

[0067] First, in the image acquisition stage, X-CT is used to perform non-destructive scanning of the shale sample to obtain information on the overall internal structure and density distribution of the sample, and based on this, a region of interest (ROI) containing representative pore structures is selected. Within the ROI, SEM-Maps technology is used to perform continuous and partially overlapping high-resolution scanning of small areas according to a preset scanning path. Multiple local two-dimensional images are then stitched together using an image stitching algorithm to form a two-dimensional structural image with both a large field of view and high resolution, thereby ensuring nanometer-level resolution while avoiding the limited field of view problem of traditional high-resolution microscopy. Simultaneously, for key areas, FIB-SEM layer-by-layer cutting and imaging is further employed to obtain a continuous two-dimensional slice sequence with true depth information, providing a high-precision data foundation for subsequent three-dimensional reconstruction.

[0068] Secondly, in the 3D reconstruction stage, the acquired multi-source 2D image data undergoes unified registration, denoising, grayscale correction, and segmentation. Based on the spatial correspondence between images, voxel-level stacking reconstruction is performed to form a 3D digital model of shale nanopores. By combining the nanoscale depth information provided by FIB-SEM with digital core reconstruction algorithms, the spatial structure of pores, mineral matrix, and organic matter components is restored, thereby obtaining a 3D digital core model that accurately reflects the pore geometry, scale distribution, and spatial connectivity.

[0069] Finally, in the connectivity and quantitative evaluation stage, the pore system is quantitatively analyzed based on the constructed three-dimensional digital model. Parameters such as porosity, average pore size, pore volume fraction, pore size distribution function, and number of connected channels are extracted. The connectivity characteristics of the pore network are calculated through graph theory analysis or seepage models. Combined with organic matter distribution information, the spatial distribution differences between organic and inorganic pores and their impact on connectivity are analyzed. By comparing the pore structure parameters of different shale samples (such as pyrolyzed and unpyrolyzed samples, and samples from different formations), the regulatory effects of organic matter thermal evolution, quartz and clay mineral content, and silica component ratio on the development and preservation of nanopores are revealed, thereby explaining the differences in the storage performance and development effects of different shale gas species.

[0070] This invention achieves the transformation from "two-dimensional observation" to "three-dimensional quantitative understanding" of shale nanoporous structure through the collaborative acquisition, cross-scale fusion reconstruction, and quantitative analysis of multimodal imaging data. It provides a high-precision, quantifiable, and repeatable methodological basis for the characterization of shale reservoir microstructure and the evaluation of its development potential.

[0071] like Figure 2 As shown, this invention utilizes multimodal joint imaging and digital reconstruction technology to perform cross-scale, multi-source information fusion modeling of shale nanoporous structures, thereby achieving high-precision reconstruction of the three-dimensional pore structure and quantitative evaluation of its connectivity. Its core principle lies in fully leveraging the complementary advantages of different imaging technologies in spatial resolution, field of view, and imaging mechanism. Through unified coordinate registration and data fusion, structural information scattered across multiple two-dimensional and three-dimensional images is transformed into a consistent three-dimensional pore structure model.

[0072] In the image acquisition stage, low-resolution volumetric data of the entire sample is first obtained using X-ray tomography to determine the positional relationships of the bedding structure, mineral distribution, and pore-rich regions. Subsequently, high-resolution two-dimensional surface images are acquired for the region of interest using a scanning electron microscope (SEM) stitching method. Continuous two-dimensional slice images are then obtained through a combined focused ion beam (FIB) and SEM imaging method, thus comprehensively recording the true morphology of the pores and mineral framework at the micrometer to nanometer scale. By employing partially overlapping scans and stitching, the observation field is expanded while maintaining nanometer-level resolution, avoiding scale distortion and structural omissions caused by single imaging methods.

[0073] During the data processing stage, scale normalization and spatial registration are performed on data from different imaging modalities to align various images within a unified coordinate system. By constructing a multimodal fusion dataset, the overall structural constraints provided by volumetric data are organically combined with the detailed information provided by 2D slices, providing highly complete input data for subsequent 3D modeling. Subsequently, automatic segmentation of porosity and mineral facies is performed based on the fused data, eliminating the uncertainty of manual identification and ensuring consistency and repeatability in the porosity structure identification process.

[0074] In the 3D reconstruction stage, the segmentation results are mapped to a 3D voxel model, allowing the pore, mineral, and interface structures to be expressed in space as discrete units. This enables a complete characterization of the geometry, spatial distribution, and interconnections of the pore network. This voxel model can not only be used for visualization but also serve as the structural basis for subsequent numerical analysis and seepage simulation.

[0075] In the connectivity and quantitative evaluation stage, different pore clusters are identified through three-dimensional connectivity analysis, and their volume fraction, pore size distribution, and connectivity are calculated, thereby quantitatively characterizing the effective permeability of the pore network. Based on the evaluation results, the differences in the number, size, and connectivity structure of organic and inorganic pores in different shale samples can be analyzed, and the intrinsic relationship between mineral composition, thermal evolution, and pore structure can be further revealed, providing a reliable structural basis for shale reservoir evaluation and development decisions.

[0076] Through the aforementioned collaborative mechanism, this invention realizes a closed-loop processing flow from multi-source imaging to structural modeling and then to connectivity evaluation, transforming the characterization of shale nanoporous structures from traditional two-dimensional qualitative analysis to three-dimensional quantitative analysis, thereby improving the accuracy, objectivity, and engineering applicability of shale reservoir evaluation.

[0077] Previous researchers have used scanning electron microscopy to identify that the identifiable organic micro-components in Early Paleozoic marine shale are mainly sapropelic components. Among them, bituminous bodies are mainly composed of nanoscale spherulites, and because their formation process is affected by thermal evolution and fluid transport, most of them do not have a fixed shape. As biogenic micro-components, algae not only retain obvious biological structural characteristics, but their size is mostly in the micrometer range. This characteristic is closely related to the original cell morphology of algae and the preservation conditions after deposition.

[0078] Shale porosity formation occurs throughout the entire hydrocarbon generation and expulsion process and is influenced by the type and maturity of organic matter. Previous studies using field emission scanning electron microscopy and Raman spectroscopy have shown that hydrogen-rich and lipid-rich organic matter generally develops porosity, while hydrogen-poor and carbon-rich organic matter has limited porosity development. This is because the heterogeneity of organic porosity development is closely related to the hydrocarbon generation and expulsion process. Hydrogen-rich and lipid-rich organic matter undergoes a continuous hydrocarbon generation process of initial liquid hydrocarbon generation, expulsion, and secondary pyrolysis, which not only promotes the formation of honeycomb-like and network-like primary pores in the kerogen matrix, but also results in a large number of nanoscale pores in the solid bitumen produced by its pyrolysis due to the retention of incompletely volatilized hydrocarbon components. In contrast, hydrogen-poor organic matter can only produce a small amount of gaseous hydrocarbons during thermal evolution and lacks the volume shrinkage effect caused by the filling and expulsion of liquid hydrocarbons, resulting in only scattered slit-like micropores with poor pore connectivity. This invention uses backscattered electron (BSE) and secondary electron (SE) scanning electron microscopy to analyze the microscopic characteristics of organic matter and minerals in the Cambrian Qiongzhusi Formation black shale reservoir in the Sichuan Basin from both planar and spatial perspectives. Figure 6 Studies have shown that shale minerals include quartz, albite and other alkaline feldspars, carbonate detrital minerals, pyrite and illite and other clay minerals. The organic matter is mainly composed of algae and bacteria, with morphologies including thick platy and rod-shaped forms. During the diagenetic evolution process, the hydrogen-oxygen ratio of the organic matter decreased, forming carbon-based fossils and developing a large number of micropores.

[0079] Example 1: Multimodal joint imaging and stitching acquisition

[0080] In this embodiment, the shale sample is first subjected to X-ray tomography to obtain low-resolution volumetric data of the entire sample, which is used to identify the main bedding structure and pore-rich regions within the sample. Based on this, the region of interest is determined according to the volumetric data, and the region is continuously scanned using a scanning electron microscope (SEM) stitching scanning method. Multiple high-resolution two-dimensional images with partial overlap are obtained according to a preset path, and a large-field-of-view two-dimensional image is generated using a stitching algorithm based on feature point matching and gray-level consistency constraints.

[0081] Simultaneously, a combined focused ion beam and scanning electron microscope imaging method is employed to slice the region of interest layer by layer and acquire continuous two-dimensional slice images. These slice images maintain a fixed interlayer spacing and spatial order. The data obtained in this manner are complementary in terms of spatial scale, resolution, and field of view, providing fundamental data conditions for subsequent multimodal registration and 3D reconstruction.

[0082] Example 2: Implementation of Multimodal Registration and Fusion

[0083] In this embodiment, scale normalization is performed on low-resolution volumetric data, high-resolution two-dimensional stitched images, and continuous slice images to unify data generated by different imaging methods into the same spatial coordinate system. First, a coarse registration is performed using an algorithm based on maximizing mutual information to align different modal data in their overall positions. Then, a fine registration is performed on local structures using an algorithm based on non-rigid deformation field optimization.

[0084] After registration, the multimodal data are fused according to their spatial relationships to construct a fused dataset containing density, grayscale, and structural boundary information. This fused dataset preserves both the overall structural outline and nanoscale pore details, providing high-information-density data input for subsequent pore segmentation and 3D reconstruction.

[0085] Example 3: Pore Phase Segmentation Based on Deep Learning

[0086] In this embodiment, the fused multimodal data is input into a convolutional neural network model for segmentation of porous and mineral phases. The model is trained under supervised supervision using manually labeled samples and can automatically identify structural units such as organic pores, inorganic pores, and mineral frameworks, and output the corresponding classification results.

[0087] The segmentation results are stored in the form of voxel labels and used as the basic input for 3D reconstruction. This method avoids the inefficiency and subjectivity of manual layer-by-layer annotation, while ensuring the consistency and comparability of segmentation results across different samples.

[0088] Example 4: Construction of a 3D Voxel Model

[0089] In this embodiment, a three-dimensional voxel model is constructed based on the segmentation results. Each voxel corresponds to a spatial unit in the sample and carries a porosity or mineral attribute label. The voxel model can reflect the true distribution, morphology, and interrelationships of pores in three-dimensional space.

[0090] Once constructed, the voxel model is visualized and supports slicing in different directions and local magnification, which helps researchers intuitively analyze the pore network structure and spatial connectivity features.

[0091] Example 5: Pore Connectivity Analysis Method

[0092] In this embodiment, a three-dimensional connected component labeling algorithm is applied to the three-dimensional voxel model to identify different pore clusters and calculate the volume, surface area, and morphological parameters of each connected component. The overall connectivity of the pore network is evaluated by calculating the maximum connected component volume fraction.

[0093] Simultaneously, statistical parameters such as porosity, pore size distribution, and interpore spacing are calculated to compare and analyze the reservoir performance of different samples. The analytical results can be directly used for shale reservoir evaluation and development potential assessment.

[0094] Example 6: Platform-based hardware and software collaboration implementation

[0095] In this embodiment, the platform consists of an image acquisition unit, a data processing unit, an analysis unit, and a storage unit. The data processing unit includes a central processing unit and a graphics processing unit, which are used to undertake control scheduling tasks and large-scale parallel computing tasks, respectively.

[0096] The software includes an image stitching module, a registration module, a fusion module, a segmentation module, a 3D reconstruction module, and a connectivity analysis module. Each module is called sequentially through a data interface to form an automated processing flow, reducing manual intervention and improving processing efficiency and consistency.

[0097] Example 7: Model Update and Adaptive Optimization

[0098] In this embodiment, the segmentation model is periodically retrained and its parameters are updated using manually labeled data from newly collected samples, so that the model can gradually adapt to the structural differences of shale samples from different regions and with different maturity levels.

[0099] By introducing a model version management mechanism, the segmentation effects of different model versions can be compared and verified, ensuring that model updates do not reduce the processing accuracy of existing samples.

[0100] Example 8: Applicability of samples under different geological conditions

[0101] In this embodiment, the method was applied to shale samples from different regions and with different degrees of thermal evolution, including organic-rich shale and silica-rich shale. The results show that, under different mineral assemblages and pore types, the method can stably obtain the three-dimensional pore structure and connectivity parameters.

[0102] By comparing the results of different samples, the relationship between mineral composition, organic matter content and pore connectivity can be analyzed, providing a quantitative basis for reservoir evaluation.

[0103] Example 9: Multi-scale Co-analysis

[0104] In this embodiment, a three-dimensional model of nanoscale pores and a microscale structural model are analyzed together to study the connectivity between pores of different scales and their impact on fluid transport behavior.

[0105] By fusing cross-scale data, a continuous description of the structure from nanopores to microfractures can be achieved, improving our understanding of the multi-scale seepage characteristics of shale reservoirs.

[0106] Example 10: Engineering Application Scenarios

[0107] In this embodiment, the method is applied to the shale gas reservoir evaluation process to assist in determining the optimal fracturing sections and development scheme design. The connectivity evaluation results are used to determine the distribution of areas in the reservoir where effective seepage channels can form.

[0108] By comparing and verifying with production data, the evaluation results can provide a reliable structural basis for development decisions and improve the scientific nature and engineering applicability of reservoir evaluation.

[0109] The embodiments of this invention achieve high-precision characterization and quantitative evaluation of connectivity of shale nanoporous structure through multimodal joint imaging and three-dimensional digital reconstruction. Its technical effect has been verified by various experimental results, including scanning electron microscopy (SEM), focused ion beam electron microscopy (FIB-SEM), atomic force microscopy (AFM), transmission electron microscopy (TEM), energy dispersive spectroscopy (EDS), and statistics on organic matter content.

[0110] like Figure 7 As shown, SEM was used to observe the microstructure of the shale sample surface, which clearly revealed organic particles, irregular flaky minerals, and the nanoscale pore structures developed between them. Figure 6 A and Figure 6 B shows a large number of micropores and cracks developed inside organic matter particles and at the organic-inorganic interface at the 2μm and 5μm scales, respectively, indicating that the pores are mainly concentrated in the organic matter enrichment area and the mineral boundary. Figure 6 Combining local magnification with EDS point analysis results, the observed region was confirmed to be rich in elements such as C and O, further proving that the region is dominated by organic matter pores, providing a reliable basis for subsequent classification and statistics of organic and inorganic pores.

[0111] like Figure 7 As shown, the three-dimensional morphology of the sample surface was scanned by AFM to obtain the elevation distribution map at the nanoscale. Figure 7 A and Figure 7B represents different sample surfaces, each with a significantly different surface undulation height distribution. Sample B exhibits sharper protrusions and greater surface roughness, reflecting a more complex nanopore and microcrack structure. This invention utilizes this three-dimensional height information to correct two-dimensional imaging errors, improving the accuracy of pore volume fraction and surface fractal feature calculations.

[0112] like Figure 8 As shown, high-resolution imaging of layered siliceous and clay minerals in shale was performed using TEM, and EDS analysis confirmed that they are mainly composed of elements such as Si, Al, and C. Figure 10 Stable nanoscale interlayered pore structures are visible between organic matter interlayers and siliceous mineral lamellae, demonstrating that siliceous components protect and support the organic pores. This invention, through three-dimensional reconstruction and connectivity analysis, can identify the continuous spatial distribution characteristics of this type of interlayered pore, thereby revealing the positive influence of siliceous components on porosity preservation and connectivity in heterogeneous shale.

[0113] Figure 9 Micron-scale CT characterization of sample pore network and three-dimensional connected body model (segmented superimposed core two-dimensional grayscale images): a. Segmented display model of pore network; b. Segmented display model of pore connectivity;

[0114] Figure 10 Use FESEM (Field Emission Scanning Electron Microscope) to select the FIB experimental area (red box);

[0115] Figure 11 FIB 3D Image Reconstruction and Image Analysis Description.

[0116] This invention not only transforms the characterization of shale nanopores from two-dimensional to three-dimensional digital understanding, but also enables quantitative analysis of organic and inorganic pores and their interconnected networks, accurately revealing the intrinsic relationships between different mineral compositions, organic matter content, and pore structure. Through the mutual corroboration of multimodal evidence, this invention demonstrates significant technical effectiveness and reliability in the fine characterization of shale reservoir microstructure, connectivity evaluation, and reservoir performance interpretation, providing a solid data foundation and methodological support for shale gas resource evaluation and development.

[0117] It should be noted that embodiments of the present invention can be implemented in hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the above-described devices and methods can be implemented using computer-executable instructions and / or included in processor control code, for example, such code provided on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuitry such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of the above-described hardware circuitry and software, such as firmware.

[0118] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions, and improvements made by those skilled in the art within the scope of the technology disclosed in the present invention, and within the spirit and principles of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A multimodal joint imaging method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores, characterized in that, Includes the following steps: Step 1, Image Acquisition: X-ray tomography is performed on the shale sample to obtain low-resolution volume data. After determining the region of interest in the volume data, a high-resolution two-dimensional surface image is obtained by using a scanning electron microscope (SEM) stitching scanning method. Continuous two-dimensional slice images are obtained by using a combined focused ion beam and SEM slicing imaging method. Step 2, Multimodal registration and fusion: Spatial coordinate unification and scale matching are performed on the low-resolution volume data, high-resolution two-dimensional surface image and continuous two-dimensional slice image to form a multimodal fusion dataset; Step 3, 3D reconstruction: Construct a 3D voxel model of the shale pore structure based on the multimodal fusion dataset; Step 4, Connectivity and Quantitative Evaluation: Perform connected domain analysis on the three-dimensional voxel model to obtain porosity, pore size distribution parameters and connectivity parameters.

2. The method as described in claim 1, characterized in that, The scanning electron microscope stitching scanning method acquires partially overlapping two-dimensional images according to a preset scanning order, and forms a large-field-of-view, high-resolution two-dimensional image through a stitching algorithm based on feature point matching and grayscale consistency constraints.

3. The method as described in claim 1, characterized in that, The multimodal registration and fusion includes coarse registration based on maximizing mutual information and fine registration based on optimization of non-rigid deformation fields.

4. The method as described in claim 1, characterized in that, The three-dimensional voxel model is generated after classifying pores and mineral phases using a convolutional neural network segmentation model.

5. The method as described in claim 1, characterized in that, The connected component analysis employs a three-dimensional connected component labeling algorithm and calculates the maximum connected component volume fraction as a connectivity evaluation index.

6. A platform for three-dimensional reconstruction and quantitative evaluation of the connectivity of shale nanopores, implementing the method of any one of claims 1 to 5, characterized in that, include: The image acquisition unit is used to acquire low-resolution volume data, high-resolution two-dimensional surface images, and continuous two-dimensional slice images. The data processing unit is used to perform multimodal registration, fusion, and 3D reconstruction. Analysis unit, used to perform connected component analysis and output pore structure parameters; Storage units are used to store raw data, fused data, and reconstructed models.

7. The platform as described in claim 6, characterized in that, The data processing unit includes a central processing unit and a graphics processing unit, which are used to execute control logic and parallel computing tasks, respectively.

8. The platform as described in claim 6, characterized in that, The data processing unit includes an image stitching module, a registration module, a fusion module, a 3D reconstruction module, and a connectivity analysis module.

9. A data processing method for three-dimensional reconstruction and quantitative evaluation of connectivity of shale nanopores, characterized in that, include: Scale normalization and spatial registration are performed on multimodal image data; Pore ​​phase segmentation is performed on the registered data; A three-dimensional pore structure model is constructed based on the segmentation results; Based on the model, porosity, pore size distribution, and connectivity parameters are calculated.

10. The method as described in claim 9, characterized in that, The segmentation model is periodically updated using manually labeled samples through supervised learning.

Images