A quantitative evaluation method and device for fault sealing performance

Abstract

The present invention provides a method and device for quantitatively evaluating the fault sealing performance. The method includes: receiving the fault sealing characterization parameters of a target work area; quantitatively evaluating the fault sealing performance in the target work area according to the fault sealing characterization parameters and a pre-generated sealing performance evaluation model, where the sealing performance evaluation model is obtained by training and optimizing an initial model generated based on the fault sealing characterization parameters. The method and device for quantitatively evaluating the fault sealing performance provided by the present invention utilize the concept of machine learning, adopt the decision tree algorithm, and establish a quantitative evaluation model of fault sealing based on a non-linear correlation on the basis of comprehensively considering multiple factors affecting fault sealing, filling the blank of the quantitative calculation method for fault sealing, and the results are accurate and effective, thereby providing a more scientific, objective, more applicable and more extensive evaluation method for the quantitative evaluation of fault sealing.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas exploration, in particular to the technical field of fault sealing evaluation related to oil and gas reservoirs, and specifically relates to a quantitative evaluation method and device for fault sealing performance. Background Art

[0002] It can be understood that in oil and gas exploration, faults are both important conduits in the process of oil and gas migration and important blocking conditions in the process of oil and gas accumulation. The sealing effect of faults on oil and gas determines the scale and prospective resource potential of related oil and gas reservoirs. According to statistics, more than 70% of the discovered oil and gas reservoirs worldwide have been proven to be related to faults. Therefore, since the birth of the oil and gas industry, the research and evaluation of fault sealing have received extensive attention. In most cases, fault-depressed basins are dominant in oil and gas-rich basins, with wide distribution, large quantity and complex nature of faults. In this case, the quantitative evaluation of fault sealing performance is more critical for oil exploration and development. The sealing mechanism of faults is relatively complex, and the sealing of faults has been proven to be related to factors such as fault dip angle, fault throw, fault scale, lithologic contact relationship on both sides of the fault, physical properties of fault rocks, and diagenesis of fault rocks.

[0003] Quantitative evaluation of fault sealing is a petroleum geology research method for evaluating the sealing performance of faults on oil and gas by selecting appropriate quantitative characterization parameters according to the fault sealing mechanism. In the prior art, the fault sealing evaluation method generally observes the lithologic docking relationship on both sides of the fault through the study of core, logging curves and field outcrop data to qualitatively evaluate the sealing characteristics of the fault. With the in-depth understanding of the fault sealing mechanism, some scholars have successively constructed characterization parameters such as fault net-to-gross ratio, shale smear factor of fault, fault shale ratio, and normal pressure on the fault plane, and calculated various characterization parameters through linear quantitative calculation formulas based on linear regression analysis and statistical principles, so as to achieve the purpose of quantitatively evaluating fault sealing.

[0004] With the development and improvement of the theory of oil and gas exploration and development, the current evaluation of fault sealing mostly focuses on the quantitative evaluation method based on characterization parameters. Among them, the shale smear factor method of fault, the fault shale ratio method and the normal pressure method on the fault plane are relatively commonly used methods. For example, "Quantitative Evaluation Method for Lateral Sealing of Faults", "Evolution Characteristics of Vertical Sealing of Vertical Fractures in Budate Group in Beier Sag", "Shale Smear Effect and Its Control on Fault Sealing - Taking the Area of Well Shixi 11 - Well Shixi 6 as an Example", and "Quantitative Fault Seal Prediction" all use the above methods to evaluate fault sealing. The above methods have the following deficiencies:

[0005] (1) The fault sealing performance is jointly controlled by multiple factors, and the current calculation methods only quantitatively evaluate one or two influencing factors of fault sealing, and the evaluation results often deviate from exploration practices;

[0006] (2) The calculation of the quantitative characterization parameters of fault sealing is based on the linear regression method. If the linear relationship between the parameters and the characteristic values is not obvious or non-linear characteristics appear during the calculation process, such methods are difficult to be applied in practice. Summary of the Invention

[0007] Aiming at the problems in the prior art, the quantitative evaluation method and device for fault sealing performance provided by the present invention utilize the concept of machine learning and adopt the decision tree algorithm to establish a quantitative evaluation model of fault sealing based on non-linear correlation on the basis of comprehensively considering multiple influencing factors of fault sealing, filling the gap in the quantitative calculation method of fault sealing, and the results are accurate and effective, thus providing a more scientific, objective, more applicable and more extensive evaluation method for the quantitative evaluation of fault sealing.

[0008] In the first aspect, the present invention provides a quantitative evaluation method for fault sealing performance, including:

[0009] Receiving the fault sealing characterization parameters of the target work area;

[0010] Quantitatively evaluating the fault sealing performance in the target work area according to the fault sealing characterization parameters and the pre-generated sealing evaluation model.

[0011] In an embodiment, the steps of generating the sealing evaluation model include:

[0012] Using the decision tree algorithm to generate an initial model of the sealing evaluation model according to the fault sealing characterization parameters;

[0013] Training the initial model with pre-generated training data to generate a transition model;

[0014] Optimizing the transition model according to the number of layers, the amount of node data, and the amount of branch data generated by the nodes in the transition model to generate the sealing evaluation model.

[0015] In an embodiment, the fault sealing characterization parameters include: fault dip angle, fault throw, extension length, fault gouge ratio, microfracture surface density, and carbonate cement content; the method for generating the training data includes:

[0016] Determining the fault sealing property according to the single-well production curve of the fault hanging wall and footwall and / or well logging interpretation and / or gas chromatography fingerprint;

[0017] Generate the training data according to the multiple fault sealing properties and their corresponding fault sealing characterization parameters.

[0018] In one embodiment, tuning the transition model according to the number of layers, the amount of node data, and the Gini coefficient of the branch data generated by the nodes in the transition model to generate the sealing evaluation model includes:

[0019] Taking the number of layers of the transition model as the maximum value, decreasing it by one layer successively, and performing an accuracy test after each decrease to determine the optimal number of layers;

[0020] Determine the total amount of data in the training data and the maximum value of the amount of node data in the transition model;

[0021] Determine the minimum value of the amount of node data according to the total amount of the training data and the maximum value of the amount of node data;

[0022] Taking a preset threshold as the maximum value, increasing the Gini coefficient successively, and performing an accuracy check after each increase to determine the optimal Gini coefficient.

[0023] In a second aspect, the present invention provides a quantitative evaluation device for fault sealing performance, and the device includes:

[0024] A parameter receiving module, configured to receive the fault sealing characterization parameters of the target work area;

[0025] A sealing quantitative evaluation module, configured to quantitatively evaluate the fault sealing performance in the target work area according to the fault sealing characterization parameters and a pre-generated sealing evaluation model.

[0026] In one embodiment, the quantitative evaluation device for fault sealing performance further includes: an evaluation model generation module, configured to generate the sealing evaluation model, and the evaluation model generation module includes:

[0027] An initial model generation unit, configured to use a decision tree algorithm to generate an initial model of the sealing evaluation model according to the fault sealing characterization parameters;

[0028] A transition model generation unit, configured to train the initial model using pre-generated training data to generate a transition model;

[0029] An evaluation model generation unit, configured to tune the transition model according to the number of layers, the amount of node data, and the amount of branch data generated by the nodes in the transition model to generate the sealing evaluation model.

[0030] In one embodiment, the quantitative evaluation device for fault sealing performance further includes: a training data generation module, configured to generate the training data, and the training data generation module includes:

[0031] A sealing property determination unit for determining the fault sealing property according to the single-well productivity curves of the hanging wall and footwall of the fault and / or well logging interpretation and / or gas chromatography fingerprint;

[0032] A training data generation unit for generating the training data according to multiple fault sealing properties and their corresponding fault sealing characterization parameters;

[0033] The fault sealing characterization parameters include: fault dip angle, fault throw, extension length, fault gouge ratio, microfracture surface density, and carbonate cement content.

[0034] In one embodiment, the evaluation model generation unit includes:

[0035] An optimal layer number determination unit for taking the layer number of the transition model as the maximum value, decreasing by one layer successively, and performing accuracy tests after each decrease to determine the optimal layer number;

[0036] A data quantity determination unit for determining the total amount of data in the training data and the maximum value of the node data quantity in the transition model;

[0037] A minimum value determination unit for determining the minimum value of the node data quantity according to the total amount of the training data and the maximum value of the node data quantity;

[0038] An optimal coefficient determination unit for taking a preset threshold as the maximum value, increasing the Gini coefficient successively, and performing accuracy checks after each increase to determine the optimal Gini coefficient.

[0039] In a third aspect, the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the method for quantitatively evaluating the fault sealing performance are implemented.

[0040] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for quantitatively evaluating the fault sealing performance are implemented.

[0041] As can be seen from the above description, the quantitative evaluation method and device for fault sealing performance provided by the embodiments of the present invention first receive the fault sealing characterization parameters of the target work area; then quantitatively evaluate the fault sealing performance in the target work area according to the fault sealing characterization parameters and the pre-generated sealing evaluation model, and the sealing evaluation model is obtained by training and optimizing the initial model generated based on the fault sealing characterization parameters. The present invention fills the gap in the quantitative evaluation method for fault sealing in petroleum geological evaluation, and can quantitatively evaluate the fault sealing with complex mechanisms more objectively and efficiently, thereby laying a foundation for formulating exploration and development strategies for fault-type oil and gas reservoirs and evaluating the potential of oil and gas resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0043] Figure 1 The first structural schematic diagram of the quantitative evaluation system for fault sealing performance according to the embodiment of the present application;

[0044] Figure 2 The second structural schematic diagram of the quantitative evaluation system for fault sealing performance according to the embodiment of the present application;

[0045] Figure 3 The flowchart of the quantitative evaluation method for fault sealing performance in the embodiment of the present invention Figure 1 ;

[0046] Figure 4 The flowchart of the quantitative evaluation method for fault sealing performance in the embodiment of the present invention Figure 2 ;

[0047] Figure 5 The flowchart of step 300 in the embodiment of the present invention;

[0048] Figure 6 The flowchart of the quantitative evaluation method for fault sealing performance in the embodiment of the present invention Figure 3 ;

[0049] Figure 7 The flowchart of step 400 in the embodiment of the present invention;

[0050] Figure 8 The schematic diagram of gas chromatography fingerprint characteristics of the hanging wall and footwall of the fault in the embodiment of the present invention under the condition of fault sealing;

[0051] Figure 9 Schematic diagram of gas chromatography fingerprint characteristics of the hanging wall and footwall of a fault in the embodiment of the present invention under the condition of fault connection;

[0052] Figure 10 Schematic flow chart of step 303 in the embodiment of the present invention;

[0053] Figure 11 Schematic flow chart of the method for quantitatively evaluating the fault sealing performance in a specific application example of the present invention;

[0054] Figure 12 Schematic flow chart of the process of constructing a decision tree evaluation model in a specific application example of the present invention;

[0055] Figure 13 Schematic diagram of the decision tree model for evaluating fault sealing in Area X based on machine learning theory in a specific application example of the present invention;

[0056] Figure 14 Program steps and corresponding codes for constructing and tuning parameters of the decision tree model in a specific application example of the present invention;

[0057] Figure 15 Schematic composition of the device for quantitatively evaluating fault sealing performance in the embodiment of the present invention Figure 1 ;

[0058] Figure 16 Schematic composition of the device for quantitatively evaluating fault sealing performance in the embodiment of the present invention Figure 2 ;

[0059] Figure 17 Schematic diagram of the composition of the evaluation model generation module 30 in the embodiment of the present invention;

[0060] Figure 18 Schematic composition of the device for quantitatively evaluating fault sealing performance in the embodiment of the present invention Figure 3 ;

[0061] Figure 19 Schematic diagram of the composition of the training data generation module 40 in the embodiment of the present invention;

[0062] Figure 20 Schematic diagram of the composition of the evaluation model generation unit 303 in the embodiment of the present invention;

[0063] Figure 21 Schematic diagram of the structure of the electronic device in the embodiment of the present invention. Detailed implementation manner

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0065] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0066] It should be noted that the terms "comprising" and "having" in the specification and claims of this application and any variations thereof in the above accompanying drawings are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0067] This application also provides a quantitative evaluation system for fault sealing performance. Refer to Figure 1 , this system can be a server A1, and the server A1 can be communicatively connected to multiple fault sealing property characterization parameter measuring instruments B1. The server A1 can also be communicatively connected to multiple databases respectively, or as Figure 2 shown, these databases can also be set in the server A1. Among them, the fault sealing property characterization parameter B1 is used to obtain the fault dip angle, fault throw, extension length, fault gouge ratio, microfracture surface density, and carbonate cement content of the target block. After receiving the fault dip angle, fault throw, extension length, fault gouge ratio, microfracture surface density, and carbonate cement content, the server A1 quantitatively evaluates the fault sealing performance in the target work area.

[0068] It can be understood that the client C1 can include a smart phone, a tablet electronic device, a network set-top box, a portable computer, a desktop computer, a personal digital assistant (PDA), a vehicle-mounted device, a smart wearable device, etc. Among them, the smart wearable device can include smart glasses, smart watches, smart bracelets, etc.

[0069] In practical applications, the part for quantitatively evaluating the fault sealing performance can be executed on the server A1 side as described above, that is, in the architecture as shown in Figure 1 or Figure 2 shown, or all operations can be completed in the client C1 device. Specifically, it can be selected according to the processing capacity of the client device and the limitations of the user usage scenario, etc. The present application does not limit this. If all operations are completed in the client device, the client device may further include a processor for performing operations such as processing the quantitative evaluation of the fault sealing performance.

[0070] The above-mentioned client C1 device may have a communication module (i.e., communication unit) and can communicate with a remote server to achieve data transmission with the server. The server may include a server for quantitatively evaluating the sealing performance of faults in the target work area, and in other implementation scenarios, it may also include a server of an intermediate platform, such as a server of a third-party server platform having a communication link with the server for quantitatively evaluating the fault sealing performance. The server may include a single computer device, or may include a server cluster composed of multiple servers, or a server structure of a distributed device.

[0071] Any suitable network protocol can be used for communication between the server and the client device, including network protocols not yet developed on the filing date of the present application. The network protocol may, for example, include TCP / IP protocol, UDP / IP protocol, HTTP protocol, HTTPS protocol, etc. Of course, the network protocol may also include, for example, the RPC protocol (Remote Procedure Call Protocol, remote procedure call protocol) and the REST protocol (Representational State Transfer, representational state transfer protocol) used on top of the above protocols.

[0072] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0073] An embodiment of the present invention provides a specific implementation manner of a method for quantitatively evaluating the fault sealing performance. Refer to Figure 3 , and the method specifically includes the following contents:

[0074] Step 100: Receive the fault sealing characterization parameters of the target work area.

[0075] Preferably, the fault sealing characterization parameters in step 100 include: quantitative characterization parameters such as fault dip angle, fault throw, extension length, shale gouge ratio (SGR), microfracture surface density, carbonate cement content, etc.

[0076] Step 200: Quantitatively evaluate the fault sealing performance in the target work area according to the fault sealing characterization parameters and a pre-generated sealing evaluation model, where the sealing evaluation model is obtained by training and optimizing an initial model generated based on the fault sealing characterization parameters.

[0077] Specifically, a machine learning method is adopted, and after comprehensively and preferably selecting multiple fault sealing performance characterization parameters, the relationship between these multiple fault sealing performance characterization parameters and their corresponding fault sealing properties is established. This relationship may be non-linear or not obvious, thereby solving the technical pain point in the prior art that only one or two parameters can be used to characterize (and must be an obvious characterization relationship) the fault sealing properties.

[0078] As can be seen from the above description, the quantitative evaluation method for fault sealing performance provided by the embodiments of the present invention first receives the fault sealing characterization parameters of the target work area; then quantitatively evaluates the fault sealing performance in the target work area according to the fault sealing characterization parameters and a pre-generated sealing evaluation model. The present invention fills the gap in the quantitative evaluation method for fault sealing in petroleum geological evaluation, and can more objectively and efficiently conduct quantitative evaluation on the fault sealing with complex mechanisms, thereby laying a foundation for formulating exploration and development strategies for fault-related oil and gas reservoirs and evaluating the potential of oil and gas resources.

[0079] In one embodiment, referring to Figure 4 , the quantitative evaluation method for fault sealing performance further includes: Step 300: Generate the sealing evaluation model. Then, referring to Figure 5 , Step 300 further includes:

[0080] Step 301: Use the decision tree algorithm to generate an initial model of the sealing evaluation model according to the fault sealing characterization parameters;

[0081] It can be understood that a decision tree is a prediction model; it represents a mapping relationship between object attributes and object values. Each node in the tree represents a certain object, each bifurcation path represents a certain possible attribute value, and each leaf node corresponds to the value of the object represented by the path experienced from the root node to this leaf node. A decision tree has only a single output. If there are multiple outputs, independent decision trees can be established to handle different outputs.

[0082] Step 302: Use the pre-generated training data to train the initial model to generate a transition model;

[0083] In one embodiment, the obtained training data can also be divided into two parts. One part is used to train the initial model, and the other part is used to test the model. When implementing step 302, every time (or several times) of training, the trained model is verified by using the test model. When the accuracy rate meets the expectation, the trained model at this time is the transition model.

[0084] Step 303: Optimize the transition model according to the number of layers, the amount of node data, and the amount of branch data generated by nodes in the transition model to generate the sealing evaluation model.

[0085] It can be understood that when training based on the initial model, the training end condition is often the preset correct rate. However, due to the complexity of the model, the volume of training data, and other parameters, it is often impossible to reach the preset correct rate after a long time of training or it takes a very long time. Therefore, the present application optimizes the transition model from the following three aspects:

[0086] (1) The maximum depth of the decision tree.

[0087] (2) The minimum value of the internal data volume of each node.

[0088] (3) The minimum value of the data volume allowed for each node to continue generating branches.

[0089] In one embodiment, the fault sealing characterization parameters include: fault dip angle, fault throw, extension length, fault gouge ratio, microfracture surface density, and carbonate cement content;

[0090] Parameters such as fault dip angle, fault throw, and fault extension length can be obtained from data such as outcrop profiles in the field, seismic profiles, and reserve reports. The fault gouge ratio (SGR) can be calculated by formula 1:

[0091]

[0092] where H i is the formation thickness (m) of the i-th layer, S i is the shale content (%) of the i-th layer, and T is the fault throw (m).

[0093] The microfracture surface density (D sf ) is defined as the ratio of the total length of microfractures and the view area in the photomicrograph of the cast thin section, and its calculation method is as formula 2:

[0094]

[0095] where D sf is the microfracture surface density (cm / cm 2 ), A s is the view area (cm2 ),L i is the length (cm) of the i-th microcrack in the field of view.

[0096] The carbonate cement content can be quantitatively obtained by the image processing software of the casting thin section microscope.

[0097] In one embodiment, referring to Figure 6 , the quantitative evaluation method of fault sealing performance further includes: Step 400: Generate the training data. Referring to Figure 7 , Step 400 further includes:

[0098] Step 401: Determine the fault sealing property according to the single-well productivity curve of the hanging wall and footwall of the fault and / or well logging interpretation and / or gas chromatography fingerprint;

[0099] Specifically, if the single-well productivity curves of the hanging wall and footwall of the fault are similar, it proves that the fault sealing property is fault opening (or connected), otherwise it proves that the fault sealing property is fault closing.

[0100] If the well logging interpretations of the reservoirs in the hanging wall and footwall of the fault are both oil layers or gas layers, it proves that the fault is open. If the well logging interpretation of the hanging wall is an oil layer or a gas layer and the well logging interpretation of the footwall is a water layer, it proves that the fault is closed. Preferably, the fault sealing property assignment method is: if the fault is open, it is assigned 1, and if the fault is closed, it is assigned 0.

[0101] Referring to Figure 8 and Figure 9 , when the shapes of the gas chromatography fingerprints of the hanging wall and footwall of the fault are similar, the fault sealing property is connected, otherwise the fault sealing property is closed.

[0102] Step 402: Generate the training data according to multiple fault sealing properties and their corresponding fault sealing characterization parameters.

[0103] It can be understood that a single training data in the training data includes the following mapping relationship: the relationship between the fault sealing property (closed or open) of a certain fault and the fault dip angle, fault throw, extension length, fault gouge ratio, microcrack surface density, and carbonate cement content of the fault.

[0104] Preferably, the training data can be divided into two parts, one part for training (training data set) and one part for testing (testing data set). For each training data set and testing data set divided by the random sampling method, a corresponding decision tree fault sealing evaluation model can be established, and the programs are similar. Therefore, a large number of decision tree models can be constructed in a short time by changing the training data set and testing data set for optimization.

[0105] In one embodiment, referring to Figure 10 , Step 303 includes:

[0106] Step 3031: Take the number of layers of the transition model as the maximum value, decrease by one layer successively, and perform accuracy tests after each decrease to determine the optimal number of layers.

[0107] Maximum depth of decision tree (D max )): Specifically, it is the number of layers of the decision tree model, which is a key parameter determining the learning depth and generalization ability of the decision tree, and is the most prioritized adjustment parameter during the parameter tuning process. Specifically, the adjustment idea of the maximum depth of the decision tree in the present invention is as follows:

[0108] ① Take the decision tree depth before parameter tuning as the maximum value, and perform test accuracy tests by decreasing one layer each time.

[0109] ② The minimum value is not less than 4, that is, 4 ≤ D max ≤ the decision tree model depth before parameter tuning.

[0110] Step 3032: Determine the total amount of data in the training data and the maximum value of the node data volume in the transition model.

[0111] Step 3033: Determine the minimum value of the node data volume according to the total amount of the training data and the maximum value of the node data volume.

[0112] In Step 3032 and Step 3033, the minimum value of the internal data volume of each node (S min )): Specifically, it is the value of the data volume contained inside each branch node of the decision tree model. If this value is too small, it proves that the data features contained in this node are mostly individual case features and do not have strong generalization ability, and should be discarded. The decision tree node is the basic unit constituting the decision tree, and its adjustment has an important impact on the prediction effect and accuracy of the decision tree. The adjustment idea of the minimum value of the internal data volume of the decision tree node is as follows:

[0113] ① Determine the number of data S a in the dataset participating in training, and the maximum value S max of the internal data volume of the decision tree node before parameter tuning;

[0114] ② The minimum value of the internal data volume of each node should be greater than one-twentieth (rounded) of the total amount of data participating in training and less than one-half of the maximum value of the internal data volume of the decision tree node before parameter tuning, that is, 0.05×S a < S min < 0.5×S max .

[0115] Step 3034: Take the preset threshold as the maximum value, increase the Gini coefficient successively, and perform accuracy checks after each increase to determine the optimal Gini coefficient.

[0116] The minimum Gini coefficient of the amount of data that each node allows to continue generating branches (Gini min ): The Gini coefficient is a parameter value used to measure the effectiveness of each node. The higher the Gini coefficient, the more effective the node is proven. Nodes with too low Gini coefficient values are not only disadvantageous to the generalization ability of the decision tree, but also have a negative impact on the accuracy of the model. After the decision tree nodes are restricted by the maximum depth of the decision tree and the amount of data inside the nodes, the limitation of the Gini coefficient of the decision nodes can further enhance the generalization ability and accuracy of the decision tree model. The adjustment idea of the minimum Gini coefficient in the present invention is as follows:

[0117] ① Take the minimum (0) value of the Gini coefficient as the lower limit value and the maximum (0.5) value of the Gini coefficient as the upper limit value for parameter adjustment; that is, 0 < Gini min <0.5

[0118] ② Adjust the parameters incrementally by 0.01 each time, compare the test accuracy after parameter adjustment, and find the optimal lower limit parameter of the Gini coefficient.

[0119] It should be noted that steps 3031 to 3034 are executed sequentially in order.

[0120] As can be seen from the above description, the embodiments of the present invention propose a quantitative evaluation method for fault sealing based on machine learning theory; first, a data set composed of fault dip angle, fault throw, fault extension length, fault SGR value, carbonate cement content, microfracture development degree and other fault sealing quantitative characterization parameters and fault opening and closing states is established. Then, a decision tree algorithm is used to establish multiple decision tree evaluation models for identifying fault sealing based on non-linear regression, and the existing data set is used to test its accuracy, and the decision model is preferably selected from them. Finally, the preferred model is adjusted and corrected to improve the model, so that its evaluation accuracy is improved, and new data is used to test its reliability to verify the applicability of the evaluation model. The present invention fills the blank of the quantitative evaluation method for fault sealing in petroleum geological evaluation, and quantitatively evaluates the fault sealing with complex mechanisms more objectively and efficiently, laying a foundation for formulating exploration and development strategies for fault-type oil and gas reservoirs and evaluating the potential of oil and gas resources.

[0121] To further illustrate the present solution, the present invention also takes Work Area X as an example to provide a specific application example of the quantitative evaluation method for fault sealing performance. The specific application example specifically includes the following content. See Figure 11 .

[0122] S1: Construct a database for characterizing fault sealing characteristics, including quantitative characterization parameters such as fault dip angle, fault throw, extension length, shale gouge ratio (SGR), microfracture surface density, and carbonate cement content.

[0123] S2: Determine the closed or open state of the fault through the single-well production capacity curves of the hanging wall and footwall of the fault, logging interpretation conclusions, gas chromatography fingerprints, etc., and assign values to the open or closed state of the fault.

[0124] S3: Use the random sampling method to divide the database established in step S1 into training data and test data, apply the decision tree algorithm in machine learning to the test data for model training, construct a decision tree to evaluate the fault sealing model, and use the test data set to test the model accuracy.

[0125] S4: According to the decision tree model and accuracy characteristics obtained in step S3, adjust the parameters of the model for calibration to improve the test accuracy of the model.

[0126] In step S3 and step S4, specifically, refer to Figure 12 , first collect the original data set and preprocess it. Then, divide the preprocessed data set into a training data set and a test data set by random sampling. Based on the training data set, construct a decision tree model and test the model. Finally, adjust the parameters of the model to generate a decision tree to evaluate the fault sealing model. Refer to Figure 13 .

[0127] The corresponding data input process and parameter adjustment and calibration process are as Figure 14 shown, where the bold part is the file or model setting value that needs to be input, and the italic underlined part is the parameter adjustment and calibration part, which can be adjusted according to the returned data characteristics and the actual situation of the data set.

[0128] S5: Use a new data set that does not belong to step S1 to conduct a reliability test on the accuracy of the evaluation model, and verify that the accuracy of the model meets the requirements of fault sealing evaluation.

[0129] Use 20 new data in area X to conduct a reliability test on this model. The results show that the accuracy of evaluating the fault sealing is 90%, as shown in Table 1:

[0130] Table 1

[0131] Test Point Serial Number Stratigraphic Horizon Depth Fault Opening and Closing Property Evaluation Model Result Evaluation Result 1 The First Member of Shahejie Formation 1642.1 Open Open Correct 2 The First Member of Shahejie Formation 1657.4 Open Closed Incorrect 3 The First Member of Shahejie Formation 1704.6 Open Open Correct 4 The First Member of Shahejie Formation 1733.0 Closed Closed Correct 5 The First Member of Shahejie Formation 1754.6 Open Open Correct 6 The First Member of Shahejie Formation 1840.5 Closed Closed Correct 7 The Second Member of Shahejie Formation 1996.2 Open Open Correct 8 The Second Member of Shahejie Formation 1997.1 Closed Closed Correct 9 The Second Member of Shahejie Formation 2032.3 Closed Closed Correct 10 The Second Member of Shahejie Formation 2144.9 Open Open Correct 11 The Second Member of Shahejie Formation 2154.7 Open Open Correct 12 The Third Member of Shahejie Formation 2743.2 Open Open Correct 13 The Third Member of Shahejie Formation 2800.9 Open Closed Incorrect 14 The Third Member of Shahejie Formation 2842.1 Open Open Correct 15 The Third Member of Shahejie Formation 2909.9 Closed Closed Correct 16 The Third Member of Shahejie Formation 3073.2 Closed Closed Correct 17 The Third Member of Shahejie Formation 3156.3 Closed Closed Correct 18 The Third Member of Shahejie Formation 3259.7 Open Open Correct 19 The Third Member of Shahejie Formation 3500.6 Closed Closed Correct 20 The Third Member of Shahejie Formation 3731.2 Open Open Correct

[0132] As can be seen from the above description, compared with the existing fault sealing evaluation methods, the quantitative evaluation method for fault sealing performance provided by the embodiments of the present invention has three beneficial effects:

[0133] (1) Introduce multiple factors affecting fault sealing into the construction process of the fault sealing evaluation model, making the evaluation model have wider applicability and higher reliability;

[0134] (2) Integrating the concepts of artificial intelligence and machine learning, introducing the decision tree algorithm into the process of establishing the evaluation model can fundamentally solve the defect that the previous methods cannot perform non-linear quantitative evaluation.

[0135] (3) Compared with the previous evaluation methods based on statistical calculations, this method can construct a large number of fault sealing evaluation models in a relatively short time, select the optimal model from them, and greatly improve the efficiency of model construction and selection.

[0136] Based on the same inventive concept, the embodiments of the present application also provide a quantitative evaluation device for fault sealing performance, which can be used to implement the method described in the above embodiments, as in the following embodiments. Since the principle of solving problems by the quantitative evaluation device for fault sealing performance is similar to that of the quantitative evaluation method for fault sealing performance, the implementation of the quantitative evaluation device for fault sealing performance can refer to the implementation of the quantitative evaluation method for fault sealing performance, and the repeated parts will not be described again. Hereinafter, the term "unit" or "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the systems described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0137] The embodiments of the present invention provide a specific implementation manner of a quantitative evaluation device for fault sealing performance that can implement the quantitative evaluation method for fault sealing performance. Refer to Figure 15 , and the quantitative evaluation device for fault sealing performance specifically includes the following:

[0138] A parameter receiving module 10, configured to receive the fault sealing characterization parameters of the target work area;

[0139] A sealing quantitative evaluation module 20, configured to quantitatively evaluate the fault sealing performance in the target work area according to the fault sealing characterization parameters and a pre-generated sealing evaluation model, where the sealing evaluation model is obtained by training and optimizing an initial model generated based on the fault sealing characterization parameters.

[0140] In one embodiment, refer to Figure 16 , the quantitative evaluation device for fault sealing performance further includes: an evaluation model generation module 30, configured to generate the sealing evaluation model. Refer to Figure 17 , and the evaluation model generation module 30 includes:

[0141] An initial model generation unit 301, configured to use the decision tree algorithm to generate an initial model of the sealing evaluation model according to the fault sealing characterization parameters;

[0142] A transition model generation unit 302, configured to train the initial model using pre-generated training data to generate a transition model;

[0143] An evaluation model generation unit 303 is configured to optimize the transition model according to the number of layers, the amount of node data, and the amount of branch data generated by nodes in the transition model, so as to generate the sealing evaluation model.

[0144] In one embodiment, referring to Figure 18 , the quantitative evaluation device for fault sealing performance further includes: a training data generation module 40, configured to generate the training data, referring to Figure 19 , the training data generation module 40 includes:

[0145] A sealing property determination unit 401 is configured to determine the fault sealing property according to the single-well production capacity curve of the hanging wall and footwall of the fault and / or well logging interpretation and / or gas chromatography fingerprint;

[0146] A training data generation unit 402 is configured to generate the training data according to a plurality of fault sealing properties and their corresponding fault sealing characterization parameters;

[0147] The fault sealing characterization parameters include: fault dip angle, fault throw, extension length, fault gouge ratio, microfracture surface density, and carbonate cement content.

[0148] In one embodiment, referring to Figure 20 , the evaluation model generation unit 303 includes:

[0149] An optimal number of layers determination unit 3031 is configured to use the number of layers of the transition model as the maximum value, decrement by one layer successively, and perform an accuracy test after each decrement to determine the optimal number of layers;

[0150] A data quantity determination unit 3032 is configured to determine the total amount of data in the training data and the maximum value of the amount of node data in the transition model;

[0151] A minimum value determination unit 3033 is configured to determine the minimum value of the amount of node data according to the total amount of the training data and the maximum value of the amount of node data;

[0152] An optimal coefficient determination unit 3034 is configured to use a preset threshold as the maximum value, increment the Gini coefficient successively, and perform an accuracy check after each increment to determine the optimal Gini coefficient.

[0153] As can be seen from the above description, the quantitative evaluation device for fault sealing performance provided by the embodiments of the present invention first receives the fault sealing characterization parameters of the target work area; then quantitatively evaluates the fault sealing performance in the target work area according to the fault sealing characterization parameters and the pre-generated sealing evaluation model. The present invention fills the gap in the quantitative evaluation method for fault sealing in petroleum geology evaluation, and can quantitatively evaluate the fault sealing with complex mechanisms more objectively and efficiently, thus laying a foundation for formulating exploration and development strategies for fault-type oil and gas reservoirs and evaluating the potential of oil and gas resources.

[0154] The embodiments of the present application also provide a specific implementation manner of an electronic device that can implement all steps in the quantitative evaluation method for fault sealing performance in the above embodiments. See Figure 21 , and the electronic device specifically includes the following content:

[0155] A processor 1201, a memory 1202, a communication interface 1203, and a bus 1204;

[0156] Among them, the processor 1201, the memory 1202, and the communication interface 1203 communicate with each other through the bus 1204; the communication interface 1203 is used to implement information transmission between related devices such as the server-side device and the client-side device;

[0157] The processor 1201 is used to call the computer program in the memory 1202. When the processor executes the computer program, all steps in the quantitative evaluation method for fault sealing performance in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0158] Step 100: Receive the fault sealing characterization parameters of the target work area;

[0159] Step 200: Quantitatively evaluate the fault sealing performance in the target work area according to the fault sealing characterization parameters and the pre-generated sealing evaluation model, where the sealing evaluation model is obtained by training and optimizing the initial model generated based on the fault sealing characterization parameters.

[0160] The embodiments of the present application also provide a computer-readable storage medium that can implement all steps in the quantitative evaluation method for fault sealing performance in the above embodiments. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by the processor, all steps in the quantitative evaluation method for fault sealing performance in the above embodiments are implemented. For example, when the processor executes the computer program, the following steps are implemented:

[0161] Step 100: Receive the fault sealing characterization parameters of the target work area;

[0162] Step 200: Quantitatively evaluate the fault sealing performance in the target work area according to the fault sealing characterization parameters and a pre-generated sealing evaluation model, where the sealing evaluation model is obtained by training and optimizing an initial model generated based on the fault sealing characterization parameters.

[0163] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the hardware + program type embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the partial description of the method embodiments for the relevant parts.

[0164] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve the desired result. Additionally, the processes depicted in the figures do not necessarily require the specific order or sequential order shown to achieve the desired result. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0165] Although this application provides method operation steps such as in the embodiments or flowcharts, based on routine or non-creative labor, there can be more or fewer operation steps. The step sequences listed in the embodiments are only one way among many step execution sequences and do not represent the only execution sequence. When the actual device or client product is executed, it can be executed in the method sequence shown in the embodiments or the drawings or in parallel (such as in an environment with parallel processors or multi-threaded processing).

[0166] For convenience of description, when describing the above device, it is divided into various modules according to functions for separate description. Of course, when implementing the embodiments of this specification, the functions of each module can be implemented in the same or multiple software and / or hardware, or the modules implementing the same function can be realized by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the device or unit can be in electrical, mechanical or other forms.

[0167] Those skilled in the art also know that in addition to implementing the controller in the form of pure computer-readable program code, it is entirely possible to implement the same function by logically programming the method steps so that the controller is implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Therefore, such a controller can be considered a hardware component, and the devices included therein for implementing various functions can also be regarded as the structures within the hardware component. Or even, the devices for implementing various functions can be regarded as either software modules for implementing the method or structures within the hardware component.

[0168] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0169] Memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

[0170] Embodiments of this specification can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. Embodiments of this specification can also be practiced in a distributed computing environment where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0171] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and reference can be made to the partial description of the method embodiments for the relevant parts. In the description of this specification, the description with reference to terms such as "an embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the embodiments of this specification. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0172] The above are only the embodiments of the embodiments of this specification and are not used to limit the embodiments of this specification. For those skilled in the art, various changes and modifications can be made to the embodiments of this specification. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of this specification shall be included within the scope of the claims of the embodiments of this specification.

Claims

1. A quantitative evaluation method for fault sealing performance, characterized in that Including: Receiving the fault sealing characterization parameters of the target work area; Quantitatively evaluating the fault sealing performance in the target work area according to the fault sealing characterization parameters and a pre-generated sealing evaluation model, where the sealing evaluation model is obtained by training and optimizing an initial model generated based on the fault sealing characterization parameters; The steps of generating the sealing evaluation model include: Using a decision tree algorithm to generate an initial model of the sealing evaluation model according to the fault sealing characterization parameters; Using pre-generated training data to train the initial model to generate a transition model; Optimizing the transition model according to the number of layers, node data volume, and branch data volume generated by nodes in the transition model to generate the sealing evaluation model; Optimizing the transition model according to the Gini coefficients of the number of layers, node data volume, and branch data volume generated by nodes in the transition model to generate the sealing evaluation model, including: Taking the number of layers of the transition model as the maximum value, decreasing it by one layer successively, and performing accuracy tests after each decrease to determine the optimal number of layers; Determining the total amount of data in the training data and the maximum value of the node data volume in the transition model; Determining the minimum value of the node data volume according to the total amount of the training data and the maximum value of the node data volume; Taking a preset threshold as the maximum value, increasing the Gini coefficient successively, and performing accuracy checks after each increase to determine the optimal Gini coefficient.

2. The quantitative evaluation method according to claim 1, characterized in that, The fault sealing characterization parameters include: fault dip angle, fault throw, extension length, fault gouge ratio, microfracture surface density, and carbonate cement content; the method for generating the training data includes: Determining the fault sealing property according to the single-well production curve of the fault hanging wall and footwall and / or well logging interpretation and / or gas chromatography fingerprint; Generating the training data according to multiple fault sealing properties and their corresponding fault sealing characterization parameters.

3. A quantitative evaluation device for fault sealing performance, characterized in that Including: A parameter receiving module for receiving the fault sealing characterization parameters of the target work area; A sealing quantitative evaluation module for quantitatively evaluating the fault sealing performance in the target work area according to the fault sealing characterization parameters and a pre-generated sealing evaluation model, where the sealing evaluation model is obtained by training and optimizing an initial model generated based on the fault sealing characterization parameters; The quantitative evaluation device further includes: an evaluation model generation module for generating the sealing evaluation model, and the evaluation model generation module includes: An initial model generation unit for using a decision tree algorithm to generate an initial model of the sealing evaluation model according to the fault sealing characterization parameters; A transition model generation unit for using pre-generated training data to train the initial model to generate a transition model; An evaluation model generation unit for optimizing the transition model according to the number of layers, node data volume, and branch data volume generated by nodes in the transition model to generate the sealing evaluation model; The evaluation model generation unit includes: An optimal layer number determination unit, configured to use the layer number of the transition model as the maximum value, successively decrease by one layer, and perform accuracy tests after each decrease to determine the optimal layer number; A data quantity determination unit, configured to determine the total quantity of data in the training data and the maximum value of the node data quantity in the transition model; A minimum value determination unit, configured to determine the minimum value of the node data quantity according to the total quantity of the training data and the maximum value of the node data quantity; An optimal coefficient determination unit, configured to use a preset threshold as the maximum value, successively increase the Gini coefficient, and perform accuracy checks after each increase to determine the optimal Gini coefficient.

4. The quantitative evaluation device according to claim 3, wherein It further includes: A training data generation module, configured to generate the training data, and the training data generation module includes: A fault sealing property determination unit, configured to determine the fault sealing property according to the single-well productivity curves of the hanging wall and footwall of the fault and / or well logging interpretation and / or gas chromatography fingerprint; A training data generation unit, configured to generate the training data according to multiple fault sealing properties and their corresponding fault sealing characterization parameters; The fault sealing characterization parameters include: fault dip angle, fault throw, extension length, fault gouge ratio, microfracture surface density, and carbonate cement content.

5. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the quantitative evaluation method for fault sealing performance according to any one of claims 1 to 2.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the quantitative evaluation method for fault sealing performance according to any one of claims 1 to 2.

Images