Construction prediction method and device based on incomplete information

Abstract

The invention relates to the technical field of seismic exploration, in particular to a structure prediction method and device based on incomplete information. The method comprises the following steps: randomly extracting discrete points in an existing structural map of a target work area, and constructing a pseudo well sample database; constructing a prediction model of regional structure and lithology information based on a pseudo well sample database, an existing structural map, and core data and surface road head data of a target work area pseudo well; constructing an incomplete information scatter diagram of the target work area on the basis of the core data of the real well of the target work area, the earth surface road head data information table and the lithology information; and based on the incomplete information scatter diagram of the target work area and the prediction model of the regional structure and lithology information, obtaining a structure prediction map and attribute information of the target work area. According to limited incomplete information such as a logging core and a field geological survey sample, a regional structure map and attributes are predicted, and an important basis is provided for structure prediction of a low-quality data region and a region without three-dimensional seismic data.

Description

Technical Field

[0001] This invention relates to the field of seismic exploration technology, and in particular to a structural prediction method and apparatus based on incomplete information. Background Technology

[0002] Structural maps of oil and gas fields indicate the top and bottom of underground oil and gas reservoirs and the structural morphology of the formations. They are crucial base maps for new well design, reserve calculation, development plan formulation, and dynamic analysis in oil and gas field exploration and development. Structural maps are typically drawn manually after processing and interpreting seismic data. However, in some areas, due to topographical limitations, there is a lack of 3D seismic data or the data quality is extremely poor, making it difficult to draw structural maps. How to conduct preliminary structural mapping work in these complex areas has become a significant challenge. Summary of the Invention

[0003] To address the aforementioned problems, and specifically for areas lacking 3D seismic data or with extremely poor data quality, thus lacking a basis for structural mapping, this invention aims to provide a structural prediction method and apparatus based on incomplete information. The purpose is to predict regional structural maps and attributes based on limited and incomplete information such as well log core samples and field geological survey samples. To achieve the above objective, this invention provides the following technical solution:

[0004] In a first aspect of the invention, a method for constructing predictions based on incomplete information is provided, the method comprising:

[0005] Randomly select discrete points from the existing structural maps of the target work area to construct a pseudo-well sample database;

[0006] Based on the pseudo-well sample database, existing structural maps, core data from pseudo-wells in the target work area, and surface road head data, a predictive model for regional tectonic and lithological information is constructed.

[0007] Based on the core data and surface road data information table and lithological information of the actual wells in the target work area, an incomplete information scatter plot of the target work area is constructed.

[0008] Based on the incomplete information scatter plot of the target work area and the prediction model of regional tectonic and lithological information, the structural prediction map and attribute information of the target work area are obtained.

[0009] Furthermore, the pseudo-well sample database includes the horizontal coordinate information, vertical coordinate information, and depth information of each discrete point.

[0010] Furthermore, based on the pseudo-well sample database, existing structural maps, core data from pseudo-wells in the target work area, and surface road data, a predictive model for regional tectonics and lithology is constructed, including:

[0011] A deep neural network is trained based on a pseudo-well sample database and existing structural maps to construct a predictive model that can predict regional structural maps from discrete points.

[0012] Based on the core data and surface roadhead data of the pseudo wells in the target work area, a coordinate axis information table of the core and outcrop data of the pseudo wells in the target work area is established.

[0013] The coordinate axis information table of core and outcrop data of the pseudo-well in the target work area is used to train the prediction model that can predict regional structural maps from discrete points, so as to construct a prediction model of regional structural and lithological information.

[0014] Furthermore, the coordinate axis information table of core and outcrop data of the pseudo well in the target work area includes the x and y axis coordinates, lithological information and depth information of each core of the pseudo well.

[0015] Furthermore, the target work area's actual well coordinate information table includes the actual well's x and y axis coordinates, lithological information, and depth information.

[0016] Furthermore, based on the incomplete information scatter plot of the target work area and the prediction model of regional tectonic and lithological information, the structural prediction map and attribute information of the target work area are obtained, including:

[0017] The incomplete information scatter plot of the target work area is input into the prediction model of regional tectonics and lithology information, and the structural prediction map and attribute information of the target work area are output and obtained.

[0018] In a second aspect of the invention, a predictive construction apparatus based on incomplete information is provided, the apparatus comprising,

[0019] The first construction unit is used to randomly extract discrete points from the existing construction graph to construct a pseudo-well sample database;

[0020] The second building unit is used to construct a predictive model of regional tectonic and lithological information based on the pseudo-well sample database, existing structural maps, core data from pseudo-wells, and surface road data.

[0021] The third building unit is used to construct an incomplete information scatter plot of the target work area based on the actual well coordinate axis information table and lithological information of the target work area;

[0022] The acquisition unit is used to acquire the structural prediction map and attribute information of the target work area based on the incomplete information scatter plot and the prediction model of regional tectonic and lithological information of the target work area.

[0023] Furthermore, the steps performed by the second building block include:

[0024] A deep neural network is trained based on a pseudo-well sample database and existing structural maps to construct a predictive model that can predict regional structural maps from discrete points.

[0025] Based on the core data and surface roadhead data of the pseudo wells in the target work area, a coordinate axis information table of the core and outcrop data of the pseudo wells in the target work area is established.

[0026] The coordinate axis information table of core and outcrop data of the pseudo-well in the target work area is used to train the prediction model that can predict regional structural maps from discrete points, so as to construct a prediction model of regional structural and lithological information.

[0027] In a third aspect of the invention, an electronic device is also provided, the electronic device comprising at least one processor and at least one memory, the memory being data-connected to the processor, wherein,

[0028] The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method described above.

[0029] In a fourth aspect of the invention, a computer-storeable medium is also provided, characterized in that the storage medium stores computer instructions, which, when executed by a processor, specifically perform the steps in the method described above.

[0030] In a fifth aspect of the invention, a computer program product is also provided, comprising computer instructions, characterized in that, when the computer instructions are executed by a processor, they specifically perform the steps in the method described above.

[0031] The technical effects and advantages of this invention are as follows:

[0032] Based on limited and incomplete information such as well logging cores and field geological survey samples, this invention predicts regional structural maps and attributes, providing an important basis for structural prediction in areas with low-quality data and areas without 3D seismic data. It can reduce dependence on the quality of seismic data, achieve automated mapping, and reduce human error.

[0033] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description and the drawings. Attached Figure Description

[0034] Figure 1 This is a flowchart of the prediction method based on incomplete information in a specific embodiment of the present invention;

[0035] Figure 2This is a diagram of a construction prediction device based on incomplete information in a specific embodiment of the present invention;

[0036] Figure 3 This is a block diagram of an electronic device according to an embodiment of the present invention.

[0037] Figure 4 This is earthquake data for a certain region in a specific embodiment of the present invention;

[0038] Figure 5a These are logging core data from specific embodiments of the present invention;

[0039] Figure 5b This is a specific embodiment of the present invention. Figure 5a Enlarged view of well logging core data;

[0040] Figure 6a This refers to surface rock outcrop data in a specific embodiment of the present invention;

[0041] Figure 6b This is a specific embodiment of the present invention. Figure 6a Enlarged view of surface rock outcrops;

[0042] Figure 7 This is a scatter plot of incomplete information in a specific embodiment of the present invention;

[0043] Figure 8 This is a construction prediction graph in a specific embodiment of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] To address the shortcomings of existing technologies, this invention discloses a prediction method based on incomplete information, such as... Figure 1 As shown, the method includes,

[0046] Step 1: Randomly select discrete points from the existing construction graph to build a pseudo-well sample database;

[0047] Step 2: Based on the pseudo-well sample database, existing structural maps, core data, and surface road data, construct a predictive model for regional tectonic and lithological information;

[0048] Step 3: Based on the actual well coordinate axis information table and lithological information of the target work area, construct an incomplete information scatter plot of the target work area;

[0049] Step 4: Based on the incomplete information scatter plot of the target work area and the prediction model of regional tectonic and lithological information, obtain the structural prediction map and attribute information of the target work area.

[0050] In a specific embodiment of the present invention, the specific operation steps for step 1 are as follows:

[0051] From a large number of existing structural maps, discrete points are randomly selected. These discrete points contain horizontal coordinate information, vertical coordinate information, and discrete point depth information. These numerous discrete points serve as pseudo-well points, constituting pseudo-well sample data. Because this sample data is extracted from a large number of existing structural maps, it forms a corresponding relationship with the existing structural maps. This data serves as an intelligent prediction tag library, a type of tag library that has not been established by predecessors.

[0052] The pseudo-well sample database includes the x-coordinate information, y-coordinate information, and depth information of each discrete point.

[0053] In a specific embodiment of the present invention, for step 2, a predictive model for regional tectonics and lithology information is constructed based on the pseudo-well sample database, existing structural maps, core data, and surface road data. The specific operational steps include:

[0054] Step 201: Train a deep neural network based on a pseudo-well sample database and existing structural maps to build a prediction model that can predict regional structural maps from discrete points;

[0055] Step 202: Based on the core data and surface outcrop data of the pseudo-well, establish a coordinate axis information table for the core and outcrop data of the pseudo-well; the coordinate axis information table for the core and outcrop data of the pseudo-well includes the x and y axis coordinates, lithological information and depth information of each core.

[0056] Step 203: Use the coordinate axis information table of the core and outcrop data of the pseudo-well to train the prediction model that predicts the regional structural map from discrete points, and construct a prediction model of regional structural and lithological information.

[0057] The coordinate axis information table of the core and outcrop data of the pseudo well is randomly divided into a training set and a validation set. The training set is used to train the prediction model that predicts the regional structural map from discrete points to obtain the trained prediction model.

[0058] The trained model is validated using a validation set. When the validation set error converges and is comparable to the training set error, it is considered to have convergence capability. At this point, the trained prediction model is a model capable of predicting regional tectonic and lithological information, i.e., a prediction model that acquires regional tectonic and lithological information.

[0059] In a specific embodiment of the present invention, for step 3: constructing an incomplete information scatter plot of the target work area based on the actual well coordinate axis information table and lithological information of the target work area, the specific execution steps are as follows:

[0060] Select the target work area, collect core data and surface roadhead data of the actual wells in the target work area, establish a coordinate axis information table of the core and outcrop data of the actual wells in the target work area, and use the x, y axis coordinates, depth information and lithology information in the coordinate axis information table of the core and outcrop data of the actual wells to establish an incomplete information scatter plot.

[0061] In a specific embodiment of the present invention, step 4: obtaining the structural prediction map and attribute information of the target work area based on the incomplete information scatter plot of the target work area and the prediction model of regional structure and lithology information, includes inputting the incomplete information scatter plot of the target work area into the prediction model of regional structure and lithology information, outputting and obtaining the structural prediction map and attribute information of the target work area, wherein the attribute information includes lithology information and physical property information.

[0062] The target work area's real well coordinate axis information table includes the real well's x and y axis coordinates, lithological information, and depth information.

[0063] This invention also discloses a construction prediction device based on incomplete information, such as... Figure 2 As shown, the device includes,

[0064] The first construction unit is used to randomly extract discrete points from the existing construction graph to construct a pseudo-well sample database;

[0065] The second building unit is used to construct a predictive model of regional tectonic and lithological information based on the pseudo-well sample database, existing structural maps, core data and surface road data.

[0066] The third building unit is used to construct an incomplete information scatter plot of the target work area based on the actual well coordinate axis information table and lithological information of the target work area;

[0067] The acquisition unit is used to acquire the structural prediction map and attribute information of the target work area based on the incomplete information scatter plot and the prediction model of regional tectonic and lithological information of the target work area.

[0068] In one specific embodiment of the present invention, the steps performed by the second building unit include:

[0069] A deep neural network is trained based on a pseudo-well sample database and existing structural maps to construct a predictive model that can predict regional structural maps from discrete points.

[0070] Based on core data and surface outcrop data, establish a coordinate axis information table for core and outcrop data;

[0071] The coordinate axis information table of the core and outcrop data is used to train the prediction model that predicts regional tectonic maps from discrete points, thereby constructing a prediction model for regional tectonic and lithological information.

[0072] Regarding the apparatus in the above embodiments, the specific manner in which each unit performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0073] Based on the above disclosure, the present invention also provides an electronic device. For example... Figure 3 As shown, the electronic device of this disclosure includes at least one processor electrically connected to the present invention and at least one memory electrically connected to the processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the method steps as executed by the controller above.

[0074] An embodiment of the present invention also provides a storable medium storing computer instructions, which, when executed by a processor, are specifically executed according to the steps in the method described in the above embodiment.

[0075] An embodiment of the present invention also provides a computer program product, including computer instructions, which, when executed by a processor, specifically follow the steps in the method described in the above embodiment.

[0076] The following will provide further explanation with reference to specific embodiments.

[0077] like Figure 4 The seismic data for a certain work area shown in the figure is chaotic and makes it difficult to discern structural information. The present invention is used to predict the structural map of this work area, and the specific steps are as follows:

[0078] First, discrete points are randomly selected from a large number of existing structural maps. These discrete points contain horizontal coordinate information, vertical coordinate information, and discrete point depth information. These large number of discrete points serve as pseudo-well points, forming pseudo-well sample data. Since these sample data are extracted from a large number of existing structural maps, they form a corresponding relationship with the existing structural maps. This data serves as an intelligent prediction label library.

[0079] Second, using pseudo-well sample data and the existing large number of structural maps as input and output, a deep neural network is trained so that the network has the function of predicting regional structural maps from discrete pseudo-well points, that is, obtaining a prediction model that can predict regional structural maps from discrete points.

[0080] Third, collect core data and surface road data, such as Figures 5a-6bAs shown, a coordinate axis information table for core and outcrop data of a pseudo-well is established. This information table includes the x and y axis coordinates, lithological information, and depth information of each core.

[0081] Fourth, using the coordinate axis information table of core and outcrop data from the pseudo-well, we continue to train the prediction model that can predict regional structural maps from discrete points, so that the prediction model that can predict regional structural and lithological information has the ability to predict regional structural and lithological information, that is, the prediction model that can acquire regional structural and lithological information.

[0082] Fifth, select specific work areas and establish a coordinate axis information table for core and outcrop data. Using the x and y axis coordinates, lithological information, and depth information from this table, create an incomplete information scatter plot, such as... Figure 7 As shown.

[0083] Sixth, the incomplete information scatter plot is used as input to the prediction model of regional tectonic and lithological information obtained in the fourth step of training, and the output is a tectonic map and attribute information, such as... Figure 8 As shown, structural maps can clearly represent the nature of structural phenomena such as folds, faults, magmatic activity, and volcanic structures, as well as the age and distribution of strata, such as the distribution characteristics of strata in anticlines and synclines, and maps of geological structural features during a specific geological period. They eliminate the influence of later tectonic processes and can accurately reflect the structural pattern of a particular geological period. Figure 7 The incomplete information scatter plot was successfully used to form a structural prediction map, providing an important basis for structural prediction in areas with low-quality data and areas without 3D seismic data.

[0084] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method of constructing a prognosis based on incomplete information, characterized by, The method comprises, randomly extracting discrete points in an existing structure map of a target work area to construct a pseudo-well sample database; based on the pseudo-well sample database, the existing structure map, core data and surface outcrop data of the pseudo-well in the target work area, a prediction model of regional structure and lithology information is constructed; based on the core data and surface outcrop data information table and lithology information of the true well in the target work area, an incomplete information scatter plot of the target work area is constructed; based on the incomplete information scatter plot of the target work area and the prediction model of regional structure and lithology information, a structure prediction map and attribute information of the target work area are obtained.

2. The method of claim 1, wherein, The pseudo-well sample database comprises horizontal coordinate information, vertical coordinate information and discrete point depth information corresponding to each discrete point.

3. The method of claim 1, wherein, Based on the pseudo-well sample database, the existing structure map, the core data and the surface outcrop data of the pseudo-well in the target work area, a prediction model of regional structure and lithology information is constructed, which comprises: training a deep neural network based on the pseudo-well sample database and the existing structure map to construct a prediction model for predicting a regional structure map from discrete points; based on the core data and surface outcrop data of the pseudo-well in the target work area, a core and outcrop data coordinate axis information table of the pseudo-well in the target work area is established; training the core and outcrop data coordinate axis information table of the pseudo-well in the target work area on the prediction model for predicting a regional structure map from discrete points to construct a prediction model of regional structure and lithology information.

4. The structure prediction method based on incomplete information according to claim 3, wherein The core and outcrop data coordinate axis information table of the pseudo-well in the target work area comprises x and y axis coordinates, lithology information and depth information of each core of the pseudo-well.

5. The structure prediction method based on incomplete information according to claim 1, wherein The true well coordinate axis information table of the target work area comprises x and y axis coordinates, lithology information and depth information of the true well.

6. The method of claim 1-5, wherein, Based on the incomplete information scatter plot of the target work area and the prediction model of regional structure and lithology information, a structure prediction map and attribute information of the target work area are obtained, which comprises: inputting the incomplete information scatter plot of the target work area into the prediction model of regional structure and lithology information, and outputting and obtaining a structure prediction map and attribute information of the target work area.

7. A predictive device based on incomplete information, characterized in that, The device comprises, a first construction unit configured to randomly extract discrete points in an existing structure map to construct a pseudo-well sample database; a second construction unit configured to construct a prediction model of regional structure and lithology information based on the pseudo-well sample database, the existing structure map, core data and surface outcrop data of the pseudo-well; a third construction unit configured to construct an incomplete information scatter plot of the target work area based on the true well coordinate axis information table and lithology information of the target work area; an acquisition unit configured to obtain a structure prediction map and attribute information of the target work area based on the incomplete information scatter plot of the target work area and the prediction model of regional structure and lithology information.

8. The construction prognosis apparatus based on incomplete information according to claim 7, characterized by, The steps performed by the second construction unit comprise: training a deep neural network based on the pseudo-well sample database and the existing structure map to construct a prediction model for predicting a regional structure map from discrete points; Based on the core data and surface outcrop data of the target work area pseudo well, a core and outcrop data coordinate axis information table of the target work area pseudo well is established; The core and outcrop data coordinate axis information table of the target work area pseudo well is trained on the prediction model with the prediction area structure map constructed from discrete points to construct a prediction model of regional structure and lithology information. 9.An electronic device, comprising at least one processor and at least one memory data-connected with the processor, wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-6.

10. A computer storable medium, characterized by The computer instructions stored on the storable medium are executed by the processor to specifically perform the steps in the method of any one of claims 1-6.

11. A computer program product comprising computer instructions, characterized in that, The computer instructions are executed by the processor to specifically perform the steps in the method of any one of claims 1-6.

Images