A High-Quality Lithofacies Identification Method for Storm Genesis Based on Multi-Source Data Integration

Abstract

This invention belongs to the field of geological sedimentology, specifically a method for identifying high-quality lithofacies of storm formation based on multi-source data integration. A: Data Acquisition and Preprocessing; S1: Data Acquisition and Confirmation, i.e., using a convolutional neural network (CNN) to automatically identify "Bouma-like sequence" features in the core based on core samples, outcrops, well logging curves, and well logging data, outputting storm depositional sample information; S2: Using deep learning (U-Net) combined with sedimentary structures to automatically classify lithofacies in the core images; S3: Using time series analysis (LSTM or Transformer) to automatically identify "box-shaped" (storm dam body) and "bell-shaped, finger-shaped" (storm dam side edge) curve features. This invention establishes a high-quality lithofacies identification model by integrating multi-source data, greatly improving the accuracy and efficiency of identification, and providing a scientific basis for evaluating oil and gas reservoirs in storm depositional environments.

Description

Technical Field

[0001] This invention relates to the field of geological sedimentology, and in particular to a method for identifying high-quality lithofacies of storm origin based on comprehensive identification of multi-source data. Background Technology

[0002] Storm deposits are an important type of sedimentation in semi-deep lacustrine environments. The high-quality facies formed by storm deposits typically have high porosity and permeability, and are an important component of oil and gas reservoirs.

[0003] Traditional identification methods primarily rely on core observations of oil-bearing levels and well logging characteristics. However, these methods struggle to comprehensively reflect the high-quality lithofacies features indicated by complex characteristics in storm-originating sediments, such as oscillating water flows and vertical "Bouma-like sequence" sedimentary sequences. Multi-source data, including sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index, can provide more comprehensive information on rock characteristics. However, a systematic method for integrating these data into the identification of storm-originating high-quality lithofacies is currently lacking. Therefore, we propose a multi-source data-based method for identifying storm-originating high-quality lithofacies to address these issues. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a method for identifying high-quality lithofacies of storm formation based on comprehensive identification of multi-source data.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A method for identifying high-quality lithofacies of storm genesis based on multi-source data integration includes the following steps:

[0007] A: Data Acquisition and Preprocessing

[0008] S1: Data acquisition and confirmation, i.e., based on core samples, field outcrops, logging curves, and well logging data, use convolutional neural networks (CNN) to automatically identify "Bouma-like sequence" features in the core samples and output storm deposition sample information;

[0009] S2: Automatic lithofacies classification of core images is performed using deep learning (U-Net) combined with sedimentary structures;

[0010] S3: Use time series analysis (LSTM or Transformer) to automatically identify the curve characteristics of "box-shaped" (main body of storm dam) and "bell-shaped, finger-shaped" (side edge of storm dam);

[0011] S4: Obtain data on the porosity and permeability of rocks through physical property testing;

[0012] S5: Obtain the pore distribution characteristics of the rock through mercury porosimetry, including pore radius and mercury removal efficiency;

[0013] S6: Obtain the grain size distribution characteristics of rocks through grain size analysis, including average grain size, sorting coefficient, and skewness;

[0014] S7: Obtain reservoir quality differences, including porosity, from thin section analysis parameters;

[0015] B: Establishment of a high-quality lithofacies identification parameter system for storm formation

[0016] S8: Feature importance analysis (random forest) is used to automatically select the best 7 parameters and establish a parameter system for identifying high-quality lithofacies of storm sediments, namely sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index.

[0017] C: Construction of a comprehensive recognition model

[0018] S9: Based on assigning values ​​to each parameter, construct a pairwise comparison matrix of each element in the criterion layer to determine the relative weight of each criterion element;

[0019] S10: The formula for calculating the prediction coefficient of high-quality lithofacies of storm sediments based on sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index, i.e., the comprehensive score;

[0020] S11: Calculate the comprehensive score and use unsupervised learning (K-means) to classify the quality lithofacies grade.

[0021] Preferably, a pairwise comparison matrix is ​​constructed for each element to determine the relative weight of each criterion element.

[0022] Preferably, the fully automatic parameter calculation outputs the product Ai of the elements in the i-th row of the matrix and its fourth root ai.

[0023] The preferred formula for calculating the weights corresponding to the output parameters is as follows:

[0024]

[0025] Preferably, the weights corresponding to the output sand-to-soil ratio are ε1 = 0.34; the weights corresponding to the output gamma value are ε2 = 0.24; the weights corresponding to the output porosity are ε3 = 0.24; the weights corresponding to the output component maturity are ε4 = 0.072; the weights corresponding to the output mercury removal efficiency are ε5 = 0.072; the weights corresponding to the output particle size sorting coefficient are ε6 = 0.021; and the weights corresponding to the output bioturbation index are ε7 = 0.015.

[0026] Preferably, in S8, the analytic hierarchy process (AHP) is used. First, a corresponding scale is assigned to each parameter. Based on the assigned values ​​for each parameter, the weights are calculated, i.e., sand-to-soil ratio (SR) = 4, gamma value (GR) = 4, and porosity = 4. —4, composition maturity (R) —2, mercury removal efficiency (Hg) —2, particle size sorting coefficient (Sc) —1, bio-disturbance index (Bio) —1 / 2.

[0027] Preferably, the formula for calculating the storm-deposited high-quality lithofacies prediction coefficient based on the seven parameters—sand-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index—is as follows:

[0028]

[0029] Preferably, based on the Q value, and using the comprehensive evaluation standard for fine-grained sedimentary reservoirs, combined with unsupervised learning to automatically classify and rate the storm sedimentary dominant facies, the evaluation level is divided into three levels: Q>50 is dominant facies, Q between 40-50 is intermediate facies, and Q<40 is poor facies.

[0030] Preferably, the construction of the comprehensive identification model also includes AI model verification and optimization, that is, inputting eight long samples from the same study area to verify the accuracy of the high-quality lithofacies comprehensive scoring model.

[0031] The beneficial effects of this invention are:

[0032] (1) By integrating multi-source data such as sand-to-land ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, particle size sorting coefficient, and bioturbation index, the accuracy and reliability of identifying high-quality lithofacies of storm formation have been improved.

[0033] (2) A systematic high-quality lithofacies identification model was established, providing a scientific basis for the evaluation of oil and gas reservoirs in storm depositional environments.

[0034] (3) The method is simple and easy to implement, and is applicable to the identification of high-quality lithology in different types of reservoirs. Attached Figure Description

[0035] Figure 1 This is a flowchart of the storm formation high-quality lithofacies identification method based on multi-source data comprehensive identification proposed in this invention;

[0036] Figure 2 This is a lithofacies core image of the storm formation high-quality lithofacies identification method based on multi-source data comprehensive identification proposed in this invention.

[0037] Figure 3 This is a bio-disturbance map of the storm formation high-quality lithofacies identification method based on multi-source data comprehensive identification proposed in this invention.

[0038] Figure 4 This is a schematic diagram of an ideal sedimentary profile and a storm depositional sequence in the study area for the storm formation high-quality lithofacies identification method based on multi-source data comprehensive identification proposed in this invention. Detailed Implementation

[0039] The technical solution of the present invention will now be clearly and completely described with reference to specific embodiments. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0040] Reference Figure 1-4 A high-quality lithofacies identification method for storm formation based on multi-source data comprehensive identification includes the following steps:

[0041] A: Data Acquisition and Preprocessing

[0042] S1: Data acquisition and confirmation, i.e., based on core samples, field outcrops, logging curves, and well logging data, use convolutional neural networks (CNN) to automatically identify "Bouma-like sequence" features in the core samples and output storm deposition sample information;

[0043] S2: Automatic lithofacies classification of core images is performed using deep learning (U-Net) combined with sedimentary structures;

[0044] S3: Use time series analysis (LSTM or Transformer) to automatically identify the curve characteristics of "box-shaped" (main body of storm dam) and "bell-shaped, finger-shaped" (side edge of storm dam);

[0045] S4: Obtain data on the porosity and permeability of rocks through physical property testing;

[0046] S5: Obtain the pore distribution characteristics of the rock through mercury porosimetry, including pore radius and mercury removal efficiency;

[0047] S6: Obtain the grain size distribution characteristics of rocks through grain size analysis, including average grain size, sorting coefficient, and skewness;

[0048] S7: Obtain reservoir quality differences, including porosity, from thin section analysis parameters;

[0049] B: Establishment of a high-quality lithofacies identification parameter system for storm formation

[0050] S8: Feature importance analysis (random forest) is used to automatically select the best 7 parameters and establish a parameter system for identifying high-quality lithofacies of storm sediments, namely sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index.

[0051] C: Construction of a comprehensive recognition model

[0052] S9: Based on assigning values ​​to each parameter, construct a pairwise comparison matrix of each element in the criterion layer to determine the relative weight of each criterion element;

[0053] S10: The formula for calculating the prediction coefficient of high-quality lithofacies of storm sediments based on sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index, i.e., the comprehensive score;

[0054] S11: Calculate the comprehensive score and use unsupervised learning (K-means) to classify the quality lithofacies grade.

[0055] In this embodiment, a pairwise comparison matrix of each element is constructed to determine the relative weight of each criterion element.

[0056] In this embodiment, the parameters are calculated automatically, and the product Ai of the elements in the i-th row of the matrix and its fourth root ai are output.

[0057] In this embodiment, the formula for calculating the weights corresponding to the output parameters is as follows:

[0058]

[0059] In this embodiment, the weights corresponding to the output sand-to-soil ratio are ε1 = 0.34; the weights corresponding to the output gamma value are ε2 = 0.24; the weights corresponding to the output porosity are ε3 = 0.24; the weights corresponding to the output component maturity are ε4 = 0.072; the weights corresponding to the output mercury removal efficiency are ε5 = 0.072; the weights corresponding to the output particle size sorting coefficient are ε6 = 0.021; and the weights corresponding to the output bioturbation index are ε7 = 0.015.

[0060] In this embodiment, in step S8, the analytic hierarchy process (AHP) is used to first assign a scale to each parameter. Based on the assigned values ​​for each parameter, weights are calculated: sand-to-soil ratio (SR) = 4, gamma value (GR) = 4, and porosity... —4, composition maturity (R) —2, mercury removal efficiency (Hg) —2, particle size sorting coefficient (Sc) —1, bio-disturbance index (Bio) —1 / 2.

[0061] In this embodiment, the formula for calculating the storm-deposited high-quality lithofacies prediction coefficient based on seven parameters—sand-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index—is as follows:

[0062]

[0063] In this embodiment, based on the Q value and the comprehensive evaluation standard for fine-grained sedimentary reservoirs, combined with unsupervised learning to automatically classify and rate the storm sedimentary dominant facies, the evaluation level is divided into three levels: Q>50 is dominant facies, Q between 40-50 is intermediate facies, and Q<40 is poor facies.

[0064] In this embodiment, the construction of the comprehensive identification model also includes AI model verification and optimization, that is, inputting eight long samples from the same study area to verify the accuracy of the high-quality lithofacies comprehensive scoring model.

[0065] Example 2

[0066] Reference Figure 1-4 A high-quality lithofacies identification method for storm formation based on multi-source data comprehensive identification includes the following steps:

[0067] Step S1: Data input, that is, based on well logging curves, well logging data, core scanning images, and high-definition photos of field outcrops (drone aerial photography + LiDAR 3D modeling), use convolutional neural networks (CNN) to automatically identify the "Bouma-like sequence" features in the core and output storm deposition sample information;

[0068] Step S2: Use deep learning (U-Net) combined with sedimentary structures to automatically classify lithofacies of core images;

[0069] Step S3: Use time series analysis (LSTM) to automatically identify the curve characteristics of "box-shaped" (main body of the storm dam) and "bell-shaped, finger-shaped" (side edge of the storm dam);

[0070] Step S4: Data input, namely, high-quality lithofacies identification parameters such as sand-to-soil ratio, porosity, permeability, pore size, mercury removal efficiency, average particle size, sorting coefficient, compositional maturity, porosity, and bioturbation density.

[0071] Step S5: Use feature importance analysis (random forest) to automatically select the 7 optimal parameters and establish a parameter system for identifying high-quality lithofacies of storm sediments;

[0072] Step 6: Using the Analytic Hierarchy Process (AHP), first assign a scale to each parameter, then establish a hierarchical relationship matrix based on the assigned values ​​for each parameter, and calculate the weights.

[0073] Step 7: Automatically generate the calculation formula Q for the comprehensive rating prediction coefficient of storm sedimentary facies;

[0074] Step 8: Calculate the Q-value of all 6-segment samples using batch data processing (Python automated script), and automatically classify and rate them using unsupervised learning (K-means);

[0075] Step 9: AI model verification and optimization, i.e., inputting 8 long samples from the same study area to verify the accuracy of the high-quality lithofacies comprehensive scoring model.

[0076] The specific process of the Analytic Hierarchy Process (AHP):

[0077] Process 1: First, input seven evaluation indicators: sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, particle size separation coefficient, and bioturbation index. Automatic scaling optimization (Bayesian optimization) adjusts the initial scale of each parameter, generating the corresponding scale for each indicator: sand-to-soil ratio (SR) - 4, gamma value (GR) - 4, porosity... —4, composition maturity (R) —2, mercury removal efficiency (Hg) —2, particle size sorting coefficient (Sc) —1, bio-disturbance index (Bio) —1 / 2;

[0078] Step 2: Construct a pairwise comparison matrix for each element and determine the relative weight of each criterion element;

[0079]

[0080]

[0081] in, Indicates the importance of a parameter to another parameter.

[0082] Process 3: Fully automatic parameter calculation, outputting the product Ai of the elements in the i-th row of the matrix and its fourth root ai:

[0083] A1 = 512;

[0084] A2 = 128;

[0085] A3 = 128;

[0086] A4 = 1;

[0087] A5 = 1;

[0088] A6 = 1 / 128;

[0089] A7 = 1 / 16384;

[0090] Step 4: Output the weights corresponding to the parameters according to Equation 1:

[0091]

[0092] The weights corresponding to the output sand ratio are ε1 = 0.34; the weights corresponding to the output gamma value are ε2 = 0.24; the weights corresponding to the output porosity are ε3 = 0.24; the weights corresponding to the output composition maturity are ε4 = 0.072; the weights corresponding to the output mercury removal efficiency are ε5 = 0.072; the weights corresponding to the output particle size sorting coefficient are ε6 = 0.021; and the weights corresponding to the output bioturbation index are ε7 = 0.015.

[0093] The formula for calculating the prediction coefficient of high-quality lithofacies storm sediments based on seven parameters—sand-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index—is as follows:

[0094]

[0095] Step 5: Input six sections of sandstone samples from various lithofacies in the study area, namely massive bedding sandstone, massive bedding sandstone containing mudstone rifts, mound-shaped cross-bedding sandstone, wavy cross-bedding sandstone, wave-like rippled bedding sandstone, deformed bedding sandstone, and parallel bedding sandstone. Take multiple samples for each lithofacies and input the sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index.

[0096] Step 6: Input the identification indicators into the prediction coefficient formula to obtain different calculation results. Average the prediction coefficients of each lithofacies to obtain Q. 块状 =57.73; Q 撕裂屑块状 =54.95; Q 丘状 =51.2; Q 液状 =46.1; Q 浪成 =44.3; Q 平形 =28.9; Q 变形 =37.3.

[0097] Step 7: Based on the Q value, and using the comprehensive evaluation standard for fine-grained sedimentary reservoirs, combined with unsupervised learning (K-means) to automatically classify and rate the dominant lithofacies of storm sediments, the evaluation level is divided into 3 levels: Q>50 is dominant lithofacies, Q between 40-50 is intermediate lithofacies, and Q<40 is poor lithofacies.

[0098] Step 8: Use sandstone samples from the same study area (Table 2) to verify the accuracy of the high-quality lithofacies comprehensive scoring model and output the model's generalization ability assessment (if the accuracy rate is >80%, the model passes the verification).

[0099] Step 9: Average the prediction coefficients for each lithofacies to obtain Q. 块状 =57.29; Q 撕裂屑块状 =53.71; Q 丘状 =49.82; Q 液状=45.96; Q 浪成 =44.37; Q 平形 =26.82; Q 变形 =38.23, the accuracy rate after generalization ability assessment is 85.7%, which passes the validation.

[0100] This invention improves the accuracy and reliability of identifying high-quality lithofacies from storm formations by integrating multi-source data such as sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index. It establishes a systematic high-quality lithofacies identification model, providing a scientific basis for evaluating oil and gas reservoirs in storm depositional environments. The method is simple and easy to implement, and is applicable to the identification of high-quality lithofacies in different types of reservoirs.

[0101] The above provides a detailed description of the storm formation high-quality lithofacies identification method based on multi-source data comprehensive identification provided by this invention. Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of these embodiments are merely for the purpose of helping to understand the method and core ideas of this invention. It should be noted that those skilled in the art can make various improvements and modifications to this invention without departing from its principles, and these improvements and modifications also fall within the protection scope of the claims of this invention.

Claims

1. A method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification, characterized in that, Includes the following steps: A: Data Acquisition and Preprocessing S1: Data acquisition and confirmation, i.e., based on core samples, field outcrops, logging curves, and well logging data, use convolutional neural networks to automatically identify "Bouma-like sequence" features in the core samples and output storm deposition sample information; S2: Use deep learning combined with sedimentary structures to automatically classify lithofacies in core images; S3: Automatically identify "box-shaped" and "bell-shaped, finger-shaped" curve features using time series analysis; S4: Obtain data on the porosity and permeability of rocks through physical property testing; S5: Obtain the pore distribution characteristics of the rock through mercury porosimetry, including pore radius and mercury removal efficiency; S6: Obtain the grain size distribution characteristics of rocks through grain size analysis, including average grain size, sorting coefficient, and skewness; S7: Obtain reservoir quality differences, including porosity, from thin section analysis parameters; B: Establishment of a high-quality lithofacies identification parameter system for storm formation S8: The optimal seven parameters are automatically selected by feature importance analysis to establish a high-quality lithofacies identification parameter system for storm sediments, namely sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index. C: Construction of a comprehensive recognition model S9: Based on assigning values ​​to each parameter, construct a pairwise comparison matrix of each element in the criterion layer to determine the relative weight of each criterion element; S10: The formula for calculating the prediction coefficient of high-quality lithofacies of storm sediments based on sand-to-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index, i.e., the comprehensive score; S11: Calculate the comprehensive score and use unsupervised learning to classify the quality lithofacies grade.

2. The method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification according to claim 1, characterized in that, Construct pairwise comparison matrices for each element and determine the relative weights of each criterion element.

3. The method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification according to claim 1, characterized in that, Fully automatic parameter calculation, outputting the product Ai of the elements in the i-th row of the matrix and its fourth root ai.

4. The method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification according to claim 1, characterized in that, The formula for calculating the weights corresponding to the output parameters:

5. The method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification according to claim 4, characterized in that, The weights corresponding to the output sand ratio are ε1 = 0.34; the weights corresponding to the output gamma value are ε2 = 0.24; the weights corresponding to the output porosity are ε3 = 0.24; the weights corresponding to the output composition maturity are ε4 = 0.072; the weights corresponding to the output mercury removal efficiency are ε5 = 0.072; the weights corresponding to the output particle size sorting coefficient are ε6 = 0.021; and the weights corresponding to the output bioturbation index are ε7 = 0.

015.

6. The method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification according to claim 5, characterized in that, In S8, the analytic hierarchy process (AHP) is used. First, a scale is assigned to each parameter. Based on the assigned values, weights are calculated: sand-to-soil ratio (SR) = 4, gamma value (GR) = 4, and porosity = 4. —4, composition maturity (R) —2, mercury removal efficiency (Hg) —2, particle size sorting coefficient (Sc) —1, bio-disturbance index (Bio) —1 / 2.

7. The method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification according to claim 1, characterized in that, The formula for calculating the prediction coefficient of high-quality lithofacies storm sediments based on seven parameters—sand-soil ratio, gamma value, porosity, compositional maturity, mercury removal efficiency, grain size sorting coefficient, and bioturbation index—is as follows:

8. The method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification according to claim 1, characterized in that, Based on the Q value, and using the comprehensive evaluation criteria for fine-grained sedimentary reservoirs, combined with unsupervised learning for automatic classification and rating, the evaluation level of the dominant facies of storm sedimentary rocks is divided into three levels: Q>50 is dominant facies, Q between 40-50 is intermediate facies, and Q<40 is poor facies.

9. The method for identifying high-quality lithofacies of storm formation based on multi-source data comprehensive identification according to claim 1, characterized in that, The construction of the comprehensive identification model also includes AI model verification and optimization, that is, inputting eight long samples from the same study area to verify the accuracy of the high-quality lithofacies comprehensive scoring model.

Images