A method, device and medium for characterizing a carbonate rock weathering crust reservoir

Abstract

The present application relates to the field of oil exploration, in particular to a carbonate rock weathering crust reservoir characterization method, equipment and medium. The method comprises fault probability body calculation and calibration; according to the fault development mode of the research area, the fault sample label definition is carried out in the seismic profile; the AI basic model is optimized by using the transfer learning technology to obtain the AI fault prediction model; the label definition and model optimization are carried out on the reservoir body to obtain the AI bead reservoir prediction model; the fault probability body and the bead reservoir obtained by AI image recognition are used to carry out three-dimensional space stereoscopic sculpture. The present application is based on AI image recognition to predict carbonate rock weathering crust reservoir, the prediction effect is improved, the influence of geological factors such as karst boundary and river channel on the prediction result is reduced, finally, based on the AI prediction data body, three-dimensional stereoscopic sculpture technology is used to realize accurate characterization of three-dimensional space characteristics of carbonate rock weathering crust reservoir.

Description

Technical Field

[0001] This invention relates to the field of petroleum exploration, specifically to a method, equipment, and medium for characterizing carbonate weathering crust reservoirs. Background Technology

[0002] A carbonate weathering crust reservoir refers to an underground carbonate rock layer that accumulates oil and gas, formed through long-term weathering. Carbonate rocks are rocks primarily composed of calcium carbonate, including limestone and dolomite. When carbonate strata are eroded and dissolved by the atmosphere, groundwater, and chemicals, a thin weathering crust forms. This weathering crust typically consists of gypsum, clay, clay minerals, and other carbonate rock remnants. The abundance of pores and fractures within the weathering crust makes it a potential oil and gas reservoir. Oil and gas stored in carbonate weathering crust reservoirs are commonly referred to as weathering crust reservoir oil and gas.

[0003] Accurate and efficient three-dimensional characterization of fracture and fracture-cavity reservoirs beneath the weathering crust of carbonate rocks is a key focus and challenge in current research.

[0004] Currently, the main methods for identifying and predicting fractured solutions in carbonate weathering crust reservoirs are: using gradient structure tensor properties to identify the boundaries of fractured solutions, and combining fracture detection properties such as fine coherence and maximum likelihood, as well as wave impedance inversion (phase-controlled wave impedance inversion) to describe the internal structural characteristics of fractured solutions.

[0005] Fine coherence is a concept in seismic exploration used to describe the variation of coherence in seismic data across different frequency ranges and time windows. Seismic coherence refers to the degree of similarity between two different seismic records. In seismic exploration, analyzing the coherence of seismic data can reveal information about subsurface structures, such as stratigraphic interfaces and fault locations. Fine coherence analysis further focuses on and studies the coherence of seismic data, observing and describing it in a more granular manner. Fine coherence analysis typically involves specific calculations and representations of coherence across different frequency ranges to reveal different characteristics and frequency responses of subsurface structures.

[0006] Traditional fracture detection attributes, such as coherence bodies, show good predictive performance for fractures in both planar and cross-sectional views, but they cannot effectively characterize spatial structural features. Furthermore, they are sensitive to identifying non-fault geological bodies such as karst boundaries and channels in carbonate weathering crust reservoirs, and suffer from issues like incoherent axis artifacts and unclear, indistinct fault identification. Gradient structure tensor attributes are also sensitive to other non-fault-controlled reservoirs, such as carbonate weathering crust reservoirs. The erosion zone of the carbonate weathering crust is influenced by the overlying strata, forming a strong continuous reflection at the T74 surface, which manifests as a continuous sheet-like pseudo-reservoir in three-dimensional space.

[0007] Therefore, there is an urgent need for a method, equipment, and medium for characterizing carbonate rock weathering crust reservoirs. Summary of the Invention

[0008] To avoid the aforementioned problems in the existing technology, the present invention aims to provide a method, equipment and medium for characterizing carbonate weathering crust reservoirs, optimize the training model for predicting faults and reservoirs, reduce the interference of geological factors such as weathering crust karst boundaries and channels on the prediction of faults and reservoirs, improve the signal-to-noise ratio of the prediction results, and accurately and effectively characterize the changes in the spatial structural features of weathering crust reservoirs.

[0009] To achieve the above objectives, the present invention provides the following technical solution: a method for characterizing carbonate weathering crust reservoirs, comprising the following steps:

[0010] S1: Calculate and calibrate the fault probability volume using a pre-trained AI basic model;

[0011] S2: Based on the fault development pattern of the study area, fault sample labels are defined in the seismic profile;

[0012] S3: Based on the tomographic samples defined in step S2, learn and train, and use transfer learning technology to optimize the AI ​​basic model to obtain the AI ​​tomographic prediction model;

[0013] S4: Recalculate the fault probability volume using the AI ​​fault prediction model;

[0014] S5: Define labels and optimize models for the storage group to obtain the AI ​​beaded storage group prediction model, and make predictions for the storage group;

[0015] S6: Based on the fault probability volume and beaded reservoir prediction results obtained from AI image recognition, three-dimensional spatial sculpting is carried out to accurately characterize the spatial features of carbonate weathering crust reservoirs.

[0016] The present invention is further configured such that step S1 specifically involves processing the original seismic data using an AI basic model pre-trained based on forward modeling synthetic data to calculate the fault probability volume; subsequently, calibration analysis is performed on the calculation results, and the reliability is judged by comparing them with the actual situation.

[0017] The present invention is further configured such that, in step S2, if the initial calculation result of the fault probability volume in step S1 differs from the actual calibration rate, i.e. the rate of agreement is lower than the preset value, then it is necessary to combine the regional fault development characteristics and compare them with the pre-training results to construct fault sample labels.

[0018] The present invention is further configured such that, in step S2, based on the expert’s professional geological knowledge of the fault development pattern in the actual work area, the number of samples is increased and improved, and the sample label profiles are evenly spaced, thereby improving the accuracy of the training model.

[0019] The present invention is further configured such that step S5 specifically involves defining the labels of reservoirs based on experts' professional geological understanding of the actual work area reservoirs, optimizing the training model based on the defined string of reservoirs using transfer learning technology, obtaining an AI string of reservoirs prediction model, and achieving effective prediction of string of reservoirs.

[0020] The present invention is further configured such that, in step S6, the spatial characteristics of the carbonate weathering crust reservoir include fault distribution and the spatial morphology of beaded reservoirs.

[0021] The present invention is further configured such that the spatial morphology of the beaded reservoir is a spatial distribution feature consisting of a series of small blocks or bead-shaped weathering residues perpendicular to the layered reservoir formed after the carbonate rock layers have undergone weathering during the geological tectonic evolution process.

[0022] The present invention also includes an electronic device, the electronic device comprising:

[0023] Memory, which stores executable instructions;

[0024] A processor that executes the executable instructions in the memory to implement the above-described method for characterizing carbonate weathering crust reservoirs.

[0025] A computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for characterizing carbonate weathering crust reservoirs.

[0026] It should be noted that three-dimensional spatial sculpting can effectively distinguish, describe, and characterize the spatial distribution, thickness, structure, and lithology of carbonate weathering crust reservoirs. This technology can accurately reconstruct the spatial morphology of reservoirs, including the top and bottom interfaces and heterogeneous characteristics of the medium, such as lithological variations and reservoir pore networks. Through three-dimensional sculpting technology, more comprehensive and accurate reservoir models can be obtained, providing an important geological foundation for the exploration and development of carbonate weathering crust reservoirs.

[0027] The "beaded space" characteristic arises from the heterogeneity of carbonate rock weathering; some areas are affected by weathering, while others retain relatively intact rock properties. Therefore, under weathering, carbonate reservoirs form a series of randomly distributed small, bead-like remnants with distinct reservoir gaps between them. The impact of the beaded space characteristic on carbonate weathering crust reservoirs is mainly reflected in two aspects:

[0028] 1. Pore characteristics: The beaded spatial characteristics lead to the formation of many small pores in the reservoir. These pores are usually randomly distributed, vary in size, and have poor connectivity. Due to the complexity of pore formation and distribution, the porosity and permeability of the reservoir are usually low.

[0029] 2. Pore Connectivity: The beaded spatial characteristics of the reservoir create obstacles between pores, such as weathered remnants. These obstacles restrict the pore connectivity of the reservoir. The presence of these obstacles reduces the effective pore volume of the reservoir and affects the flow properties of oil and gas.

[0030] In summary, the beneficial effects of the above-mentioned technical solution of the present invention are as follows:

[0031] This invention uses AI image recognition to predict carbonate weathering crust reservoirs, improving the signal-to-noise ratio and reducing the impact of geological factors such as karst boundaries and river channels on the prediction results. Finally, based on the AI ​​prediction data, 3D sculpting technology is used to accurately characterize the three-dimensional spatial features of carbonate weathering crust reservoirs. By combining intelligent fracture and bead recognition with conventional ant-like body attributes for 3D sculpting, the relationship between faults and reservoirs is better depicted. The coupling relationship between strike-slip faults and karst processes in high-yield wells is more clearly defined, which is beneficial for summarizing high-yield patterns and improving the success rate of well deployment. Attached Figure Description

[0032] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1 This is a flowchart of the AI ​​tomography prediction model established according to the present invention;

[0034] Figure 2 This is the process of establishing label definitions during fault prediction according to the present invention;

[0035] Figure 3 This is a schematic diagram of sample labels and training models in the fault prediction process according to the present invention;

[0036] Figure 4 This is a schematic diagram of sample labels and training models in the storage group prediction process according to the present invention;

[0037] Figure 5 This is a comparison between images obtained by finely coherently extracting T74 layer attributes and extracting T74 layer attributes based on AI image recognition fracture prediction.

[0038] Figure 6 This invention provides a comparison chart of the predicted energy body extracted from the beads and the AI ​​intelligent recognition of the beads.

[0039] Figure 7 This is a sculpted schematic diagram obtained by three-dimensional spatial sculpting based on the prediction results of faults and reservoir masses in the weathered crust reservoir space according to the present invention, which is based on AI image recognition. Detailed Implementation

[0040] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Based on the embodiments of the present invention, other similar embodiments obtained by those skilled in the art without creative effort should all fall within the scope of protection of the present invention.

[0041] Furthermore, the directional terms mentioned in the following embodiments, such as "up," "down," "left," and "right," are only for reference to the accompanying drawings. Therefore, the directional terms used are for illustrative purposes and not for limiting the invention.

[0042] The present invention will be further described below with reference to the accompanying drawings and preferred embodiments.

[0043] Example 1:

[0044] like Figure 1 As shown, this is a preferred embodiment of the present invention: 1. A method for characterizing carbonate weathering crust reservoirs, comprising the following steps:

[0045] S1: Calculate and calibrate the fault probability volume using a pre-trained AI basic model;

[0046] Using an AI-based model pre-trained on forward modeling synthetic data, the raw seismic data is processed to calculate the fault probability volume; then, the calculation results are calibrated and analyzed, and compared with the actual situation.

[0047] S2: Based on the fault development pattern of the study area, fault sample labels are defined in the seismic profile;

[0048] The process of creating tags is as follows: Figure 2 As shown, the label image is first defined, then the image grayscale is processed and dimensionality reduction feature decomposition is performed; then the image is upgraded in dimensionality for feature reconstruction, and finally the image features are extracted.

[0049] If the initial calculation result of the fault probability volume matches the actual calibration result less than the preset value, then the fault sample label is constructed by combining the regional fault development characteristics with the pre-training results.

[0050] Based on experts' professional geological understanding of the fault development patterns in the actual work area, the number of samples was increased and improved, and the sample label profiles were ensured to be evenly spaced, thereby improving the accuracy of the training model.

[0051] The matching rate is the degree of agreement between the fault location, morphology, and other characteristics predicted by the fault probability volume and the actual fault situation obtained when calibration is performed through drilling, logging, and other means.

[0052] S3: As Figure 3 As shown, Figure 3 Figure A in the diagram is a schematic diagram of sample labels during the fault prediction process. Figure 3 Figure B in the diagram illustrates the training of the model during the fault prediction process. Based on the fault samples defined in step S2, learning and training are performed, and transfer learning techniques are used to optimize the AI ​​base model, resulting in the AI ​​fault prediction model.

[0053] S4: Recalculate the fault probability volume using the optimized AI fault prediction model; further improve the accuracy and reliability of fault prediction.

[0054] S5: Define labels and optimize models for the storage group to obtain the AI ​​beaded storage group prediction model, and make predictions for the storage group;

[0055] like Figure 4 As shown, Figure 4 Figure A in the diagram is a schematic diagram of sample labels during the storage group prediction process. Figure 4 Figure B in the diagram illustrates the training model during the reservoir prediction process. Based on experts' professional geological understanding of the reservoirs in the actual work area, reservoir labels are defined. Transfer learning techniques are then used to optimize the training model based on these defined string of reservoirs, resulting in an AI string of reservoirs prediction model that effectively predicts the formation of these reservoirs.

[0056] S6: Based on the fault probability volume and beaded reservoir prediction results obtained from AI image recognition, three-dimensional spatial sculpting is carried out to characterize the spatial features of carbonate weathering crust reservoirs. This provides an intuitive and comprehensive geological model for reservoir evaluation and development.

[0057] The spatial characteristics of the carbonate weathering crust reservoir include fault distribution and the spatial morphology of beaded reservoirs. The spatial morphology of beaded reservoirs is a spatial distribution feature composed of small blocks or bead-like weathering remnants perpendicular to the layered reservoir.

[0058] refer to Figures 5-7 , Figure 5 A comparison of images obtained by finely coherently extracting T74 layer attributes and extracting T74 layer attributes based on AI image recognition fracture prediction. Figure 6 Figure A in the figure shows the predicted energy body extracted from the beads according to the present invention. Figure 6 Image B in the diagram shows AI-powered intelligent bead recognition. Figure 7 Figure A in the figure is a sculpted schematic diagram obtained by three-dimensional spatial carving based on the prediction results of the ant body and reservoir body. Figure 7 Figure B in the figure is a sculpted schematic diagram obtained by three-dimensional spatial sculpting of the prediction results of the ant body, fault probability body, and reservoir body.

[0059] It can be seen that by using AI image recognition and integrating the rich professional knowledge of oilfield experts to optimize the training model for predicting fractures and reservoirs, the interference of geological factors such as weathered crust karst boundaries and river channels on the prediction of fractures and reservoirs is reduced, the signal-to-noise ratio of the prediction results is improved, and the changes in the spatial structure characteristics of weathered crust reservoirs are accurately and effectively characterized.

[0060] In the characterization of carbonate weathering crust reservoirs, three-dimensional spatial sculpting is a seismic data processing and interpretation technique. Based on a three-dimensional attribute model of seismic data, it integrates seismic data with a geological model through profile interpretation, seismic attribute analysis, and an understanding of geological concepts, thereby achieving a three-dimensional description and characterization of the reservoir. By combining conventional ant-body attributes with intelligent fault and bead-like identification, it better depicts the fault-reservoir relationship, clarifying the coupling relationship between strike-slip faults and karst processes in high-yield wells. This facilitates the summarization of high-yield patterns and improves the success rate of well deployment.

[0061] Three-dimensional spatial sculpting can effectively distinguish, describe, and characterize the spatial distribution, thickness, structure, and lithology of carbonate weathering crust reservoirs. This technology can accurately reconstruct the spatial morphology of reservoirs, including the top and bottom interfaces and heterogeneous characteristics of the medium, such as lithological variations and reservoir pore networks. Through 3D sculpting technology, more comprehensive and accurate reservoir models can be obtained, providing an important geological foundation for the exploration and development of carbonate weathering crust reservoirs.

[0062] The "beaded space" feature refers to the spatial distribution of a series of small, bead-like weathering remnants, almost perpendicular to the layered reservoir, formed during the geological evolution of carbonate strata through weathering. This beaded spatial feature is due to the heterogeneity of carbonate weathering; some areas are affected by weathering, while others retain relatively intact rock properties. Therefore, under weathering, carbonate reservoirs form a series of randomly distributed small, bead-like remnants with distinct reservoir gaps between them. The impact of the beaded spatial feature on carbonate weathering crust reservoirs is mainly reflected in two aspects:

[0063] 1. Pore characteristics: The beaded spatial characteristics lead to the formation of many small pores in the reservoir. These pores are usually randomly distributed, vary in size, and have poor connectivity. Due to the complexity of pore formation and distribution, the porosity and permeability of the reservoir are usually low.

[0064] 2. Pore Connectivity: The beaded spatial characteristics of the reservoir create obstacles between pores, such as weathered remnants. These obstacles restrict the pore connectivity of the reservoir. The presence of these obstacles reduces the effective pore volume of the reservoir and affects the flow properties of oil and gas.

[0065] In summary, this invention uses AI image recognition to predict carbonate weathering crust reservoirs, improving the signal-to-noise ratio and reducing the impact of geological factors such as karst boundaries and river channels on the prediction results. Finally, based on the AI ​​prediction data, 3D sculpting technology is used to accurately characterize the three-dimensional spatial features of carbonate weathering crust reservoirs. By combining intelligent fracture and bead recognition with conventional ant-like body attributes for 3D sculpting, the relationship between faults and reservoirs is better depicted. The coupling relationship between strike-slip faults and karst processes in high-yield wells is more clearly defined, which is beneficial for summarizing high-yield patterns and improving the success rate of well deployment.

[0066] Example 2:

[0067] The present invention also includes an electronic device, the electronic device comprising:

[0068] Memory, which stores executable instructions;

[0069] A processor that executes the executable instructions in the memory to implement the above-described method for characterizing carbonate weathering crust reservoirs.

[0070] Example 3:

[0071] A computer-readable storage medium storing a computer program that, when executed by a processor, implements a method for characterizing carbonate weathering crust reservoirs.

[0072] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for characterizing carbonate weathering crust reservoirs, characterized in that, Includes the following steps: S1: Calculate and calibrate the fault probability volume using a pre-trained AI basic model; S2: Based on the fault development pattern of the study area, fault sample labels are defined in the seismic profile; S3: Based on the tomographic samples defined in step S2, learn and train, and use transfer learning technology to optimize the AI ​​basic model to obtain the AI ​​tomographic prediction model; S4: Recalculate the fault probability volume using the AI ​​fault prediction model; S5: Define labels and optimize models for the storage group to obtain the AI ​​beaded storage group prediction model, and make predictions for the storage group; S6: Based on the fault probability volume and beaded reservoir prediction results obtained from AI image recognition, three-dimensional spatial sculpting is carried out to characterize the spatial characteristics of carbonate weathering crust reservoirs.

2. The method for characterizing carbonate weathering crust reservoirs according to claim 1, characterized in that, Step S1 specifically involves using an AI-based model pre-trained on forward modeling synthetic data to process the original seismic data and calculate the fault probability volume; then, the calculation results are calibrated and analyzed, and compared with the actual situation.

3. The method for characterizing carbonate weathering crust reservoirs according to claim 2, characterized in that, Step S2 specifically involves the following steps: if the initial calculation result of the fault probability volume in step S1 has a lower match rate with the actual calibration than a preset value, then the fault sample label is constructed by combining the regional fault development characteristics with the pre-training results.

4. The method for characterizing carbonate weathering crust reservoirs according to claim 3, characterized in that, In step S2, based on the experts' professional geological understanding of the fault development patterns in the actual work area, the number of samples is increased and improved, and the sample label profiles are evenly spaced.

5. The method for characterizing carbonate weathering crust reservoirs according to claim 1, characterized in that, Step S5 specifically involves defining the labels of reservoirs based on experts' professional geological understanding of the actual work area reservoirs, optimizing the training model using transfer learning techniques based on the defined string of reservoirs, obtaining an AI string of reservoirs prediction model, and achieving effective prediction of string of reservoirs.

6. The method for characterizing carbonate weathering crust reservoirs according to claim 1, characterized in that, In step S6, the spatial characteristics of the carbonate weathering crust reservoir include fault distribution and the spatial morphology of beaded reservoirs.

7. The method for characterizing carbonate weathering crust reservoirs according to claim 6, characterized in that, The spatial morphology of beaded reservoirs is characterized by the spatial distribution of small blocks or bead-like weathered remnants perpendicular to the layered reservoir.

8. An electronic device, characterized in that, The electronic device includes: Memory, which stores executable instructions; A processor that executes the executable instructions in the memory to implement a method for characterizing a carbonate weathering crust reservoir according to any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements a method for characterizing carbonate weathering crust reservoirs according to any one of claims 1-7.

Images