Drilling position determination method, drilling position determination device, medium and electronic equipment

Abstract

The present application provides a method for determining a borehole position, a device for determining a borehole position, a medium, and an electronic device. The method for determining a borehole position includes: obtaining a first set of spatial point coordinates; processing the first set of spatial point coordinates through a first neural network model to obtain a first predicted formation attribute corresponding to the voxel and a first probability value corresponding to the first predicted formation attribute; generating a first formation prediction model and a first uncertainty model based on the first predicted formation attribute and the first probability value; and obtaining the positions of several first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model. Compared with relying on the engineer's cognition to determine the borehole position, the method for determining a borehole position can improve the rationality of the borehole position determination, and the method for determining a borehole position can correspondingly reduce labor costs, thereby reducing survey costs.

Description

Technical Field

[0001] The present application belongs to the field of engineering geological survey and relates to a method for determining a drilling position, and in particular to a method for determining a drilling position, a drilling position determining device, a medium and an electronic device. Background Art

[0002] Geological survey and geological interpretation are key to engineering construction. Proper geological work is fundamental to ensuring the safe construction and long-term operation of projects, and even the safety of people's lives and property. However, the geological bodies in engineering areas are often highly anisotropic and the geological conditions are complex, making the engineering geological interpretation process full of uncertainty.

[0003] The process of geological survey is essentially a process of reducing the uncertainty of geological cognition. It usually requires geological engineers to pre-interpret the geological conditions based on known information and focus on surveying areas with obvious uncertainty. Geological survey and geological interpretation are constantly promoting each other and gradually improving in this iterative process.

[0004] As a means of geological survey, drilling survey usually requires determining the drilling location. The current drilling location determination process relies on engineers' professional knowledge and spatial imagination and guesswork. There are problems such as unreasonable drilling location determination due to insufficient cognition and increased survey costs. Therefore, the current drilling location determination method has the problems of low rationality of drilling location determination and increased survey costs. Summary of the Invention

[0005] The purpose of this application is to provide a drilling position determination method, a drilling position determination device, a medium and an electronic device, which are used to solve the problems of low rationality of drilling position determination and increased survey costs in current drilling position determination methods.

[0006] In a first aspect, the present application provides a method for determining a drilling position, comprising: obtaining a first set of spatial point coordinates, wherein the first spatial point coordinates in the first set of spatial point coordinates are the spatial point coordinates of the centroid of a voxel in a voxel set of a three-dimensional engineering area; processing the first set of spatial point coordinates through a first neural network model to obtain a first predicted formation attribute corresponding to the voxel and a first probability value corresponding to the first predicted formation attribute; generating a first formation prediction model and a first uncertainty model based on the first predicted formation attribute and the first probability value, the first formation prediction model being used to represent the first predicted formation attribute in the three-dimensional engineering area, and the first uncertainty model being used to represent the uncertainty of the first predicted formation attribute in the three-dimensional process area; obtaining the positions of several first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model.

[0007] In the borehole position determination method, the first spatial point coordinate set is processed by a first neural network model to obtain the first predicted formation attribute corresponding to the voxel and the first probability value corresponding to the first predicted formation attribute, and the positions of several first boreholes in the three-dimensional engineering area are obtained based on the first formation prediction model and the first uncertainty model. Compared with relying on the engineer's cognition to determine the borehole position, the rationality of the borehole position determination can be improved, and the borehole position determination method can correspondingly reduce labor costs, thereby reducing survey costs.

[0008] In one embodiment of the present application, the method for obtaining the first neural network model includes: obtaining a second spatial point coordinate set of several second boreholes in the three-dimensional engineering area and a first real stratigraphic attribute corresponding to the second spatial point coordinates in the second spatial point coordinate set, the horizontal coordinate value and the vertical coordinate value of the second spatial point coordinate are the same horizontal coordinate value and the same vertical coordinate value, and the height coordinate value of the second spatial point coordinate is in the range of [a, b], a represents the bottom elevation of the second borehole, and b represents the surface elevation of the second borehole; based on the second spatial point coordinate set and the first real stratigraphic attribute, the initialized neural network model is trained by a back propagation algorithm to obtain the first neural network model.

[0009] In one embodiment of the present application, a method for obtaining the positions of several first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model includes: step S1, based on the first formation prediction model and the first uncertainty model, obtaining a third spatial coordinate point set of several third boreholes in the three-dimensional engineering area and a second real formation attribute corresponding to the third spatial point coordinates in the third spatial coordinate point set, the horizontal coordinate value and the vertical coordinate value of the third spatial point coordinate are the same horizontal coordinate value and the same vertical coordinate value, and the height coordinate value of the third spatial point coordinate is in the range of [c, d], c represents the bottom elevation of the third borehole, and d represents the surface elevation of the third borehole; step S2, based on the third spatial point coordinate set, the second real formation attribute, the second spatial point coordinate set and the first real formation attribute, the first neural network model is trained to obtain a second neural network Model, if the average cross entropy value of the first neural network model during the training process is greater than the preset cross entropy value, continue to execute steps S3-S4 and go to step S1 after step S4 is completed, the first formation prediction model in step S1 is the second formation prediction model, the first uncertainty model is the second uncertainty model, and the first neural network model in step S2 is the second neural network model; if the average cross entropy value is not greater than the preset cross entropy value, the third spatial coordinate point set of the third borehole is the position of the first borehole in the three-dimensional engineering area; step S3, process the first spatial point coordinate set through the second neural network model to obtain the second predicted formation attribute corresponding to the voxel and the second probability value corresponding to the second predicted formation attribute; step S4, generate a second formation prediction model and a second uncertainty model based on the second predicted formation attribute and the second probability value.

[0010] In one embodiment of the present application, the implementation method of generating a second formation prediction model and a second uncertainty model based on the second predicted formation attributes and the second probability value includes: generating the second formation prediction model based on the second predicted formation attributes, the second formation prediction model being a stereoscopic image of the three-dimensional engineering area, the color of the voxels in the second formation prediction model being determined by their corresponding second predicted formation attributes, and each second predicted formation attribute having a corresponding color; generating the second uncertainty model based on the second probability value, the second uncertainty model including the uncertainty distribution of the three-dimensional engineering area in the horizontal direction.

[0011] In one embodiment of the present application, a method for obtaining a third spatial coordinate point set of several third boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model includes: obtaining a first area and a second area in the three-dimensional process area based on the first formation prediction model and the first uncertainty model, the first area being an area affecting the stability of the three-dimensional engineering area, the stability being determined by the first predicted formation attributes, and the second area being an area with greater uncertainty in the second uncertainty model; obtaining the third spatial coordinate point set of the third borehole based on the first area and the second area, the position of the third spatial coordinate point set in the three-dimensional engineering area being in the first area or the second area.

[0012] In one embodiment of the present application, the structure of the initialized neural network model is determined by the data volume of the second borehole spatial coordinate set and the first real formation attribute, the prior distribution of the parameters of the initialized neural network model is Gaussian distribution or gamma distribution, the number of neurons in the input layer of the initialized neural network model is 3, and the number of neurons in the output layer of the initialized neural network model is the same as the number of formations in the three-dimensional engineering area.

[0013] In one embodiment of the present application, the average cross entropy value is expressed as:

[0014]

[0015] Where L represents the average cross entropy value, M represents the number of formation attribute categories, c is a value in the interval [1, M], and y ic It is expressed as a sign function, which means that if the true stratum attribute category corresponding to the training sample i is the same as the stratum attribute category corresponding to c, then it takes 1, otherwise it takes 0, p ic It represents the probability that the predicted formation attribute corresponding to the training sample i belongs to the formation attribute category corresponding to c, N represents the number of training samples, the predicted formation attribute is the first predicted formation attribute or the second predicted formation attribute, and the real formation attribute is the first real formation attribute or the second real formation predicted attribute.

[0016] In a second aspect, the present application provides a drilling position determination device, which includes: an engineering information acquisition module for acquiring a first set of spatial point coordinates, wherein the first spatial point coordinates in the first set of spatial point coordinates are the spatial point coordinates of the centroid of the voxel in the voxel set of the three-dimensional engineering area; a network model processing module for processing the first set of spatial point coordinates through a first neural network model to obtain a first predicted formation attribute corresponding to the voxel and a first probability value corresponding to the first predicted formation attribute; a formation model generation module for generating a first formation prediction model and a first uncertainty model based on the first predicted formation attribute and the first probability value, the first formation prediction model being used to represent the first predicted formation attribute in the three-dimensional engineering area, and the first uncertainty model being used to represent the uncertainty of the first predicted formation attribute in the three-dimensional process area; a drilling position determination module for acquiring the positions of several first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model.

[0017] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the drilling position determination method described in any one of the first aspects of the present application.

[0018] In a fourth aspect, the present application provides an electronic device comprising: a memory storing a computer program; and a processor communicating with the memory, for executing the drilling position determination method described in any one of the first aspects of the present application when the computer program is called.

[0019] As described above, the drilling position determination method, drilling position determination device, medium, and electronic device described in this application have beneficial effects:

[0020] In the borehole position determination method, the first spatial point coordinate set is processed by a first neural network model to obtain the first predicted formation attribute corresponding to the voxel and the first probability value corresponding to the first predicted formation attribute, and the positions of several first boreholes in the three-dimensional engineering area are obtained based on the first formation prediction model and the first uncertainty model. Compared with relying on the engineer's cognition to determine the borehole position, the rationality of the borehole position determination can be improved, and the borehole position determination method can correspondingly reduce labor costs, thereby reducing survey costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 Shown is a schematic diagram of the hardware structure for running the drilling position determination method according to an embodiment of the present application.

[0022] Figure 2Shown is a flow chart of the drilling position determination method described in an embodiment of the present application.

[0023] Figure 3 Shown is a flowchart of the method for obtaining the first neural network model in an embodiment of the present application.

[0024] Figure 4 Shown is a flowchart of an implementation method of the present application for obtaining the positions of several first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model.

[0025] Figure 5 Shown is a flowchart of an implementation method of the present application for generating a second formation prediction model and a second uncertainty model based on the second predicted formation attribute and the second probability value.

[0026] Figure 6 Shown is a flowchart of an implementation method of the present application for obtaining the position of the first borehole in the three-dimensional engineering area based on the second formation prediction model and the second uncertainty model.

[0027] Figure 7 Shown is a structural schematic diagram of the drilling position determination device described in an embodiment of the present application.

[0028] Component number description

[0029] 10 Electronic devices

[0030] 110 Memory

[0031] 120 processors

[0032] 1210 CPU

[0033] 1220 Neural Network Processor

[0034] 12210 Neural Network Implementation Engine

[0035] 12220 dedicated hardware circuit

[0036] 122210 matrix calculation unit

[0037] 122220 vector computing unit

[0038] 700 Drilling Position Determination Device

[0039] 710 Engineering Information Acquisition Module

[0040] 720 Network Model Processing Module

[0041] 730 Formation Model Generation Module

[0042] 740 Drilling Position Determination Module

[0043] Steps S11-S14

[0044] Steps S21-S22

[0045] Steps S1-S4

[0046] Steps S31-S32

[0047] Steps S41-S42 DETAILED DESCRIPTION

[0048] The following describes the embodiments of the present application through specific examples. Those skilled in the art can easily understand the other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0049] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. Therefore, the illustrations only show components related to the present application and are not drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component can be changed at will, and the component layout type may also be more complicated.

[0050] The technical solutions in the embodiments of the present application are described in detail below in conjunction with the drawings in the embodiments of the present application.

[0051] The drilling position determination method provided in the embodiment of the present application can be run in electronic devices such as computers. Figure 1 For example, Figure 1 This is a hardware block diagram of an electronic device implementing the drilling location determination method. The electronic device 10 includes a memory 110 and a processor 120. The processor 120 may be a central processing unit 1210 or a dedicated neural network processor 1220. The neural network processor 1220 includes a neural network implementation engine 12210 and dedicated hardware circuits 12220. The dedicated hardware circuits 12220 include a matrix calculation unit 122210 and a vector calculation unit 122220.

[0052] Optionally, the neural network processor 12220 is a processor that uses a dedicated hardware circuit 12220 to perform neural network calculations, and the dedicated hardware circuit 12220 is an integrated circuit for performing neural network calculations and includes a matrix calculation unit 122210 and a vector calculation unit 122220 for hardware execution of vector-matrix multiplication.

[0053] Optionally, the neural network implementation engine 12210 is used to generate instructions for execution by the dedicated hardware circuit 12220. When the instructions are executed by the dedicated hardware circuit 12220, the dedicated hardware circuit 12220 performs operations specified by the neural network to generate a neural network output from the received neural network input.

[0054] like Figure 2 As shown, this embodiment provides a method for determining a drilling position, which can be implemented by a processor of a computer device. The method includes:

[0055] S11 , obtaining a first spatial point coordinate set, where the first spatial point coordinates in the first spatial point coordinate set are spatial point coordinates of a body centroid in a voxel set of a three-dimensional engineering area.

[0056] Optionally, the three-dimensional engineering area can be the engineering area to be surveyed, and the voxel set can refer to a series of small cubic units divided according to the three-dimensional engineering area. The cubic units can enclose the three-dimensional engineering area, and the length, width and height of the cubic units can be flexibly set according to actual conditions. In this embodiment, the length, width and height of the cubic units are set to 1m, 1m and 0.15m respectively.

[0057] S12: Process the first spatial point coordinate set through a first neural network model to obtain a first predicted stratum attribute corresponding to the voxel and a first probability value corresponding to the first predicted stratum attribute.

[0058] Optionally, the first neural network model is a trained neural network model, the first predicted formation attribute refers to the formation attribute predicted by the first neural network model, each of the first spatial point coordinates has a corresponding first predicted formation attribute, which is used to represent the first predicted formation attribute of the voxel corresponding to the first spatial point coordinate. The formation attribute may refer to lithology or formation structure characteristics, etc. The first probability value may be the probability value that the formation attribute predicted by the first neural network model is the first predicted formation attribute.

[0059] Optionally, since the coordinates of the first spatial point are the coordinates of the centroid of the body, one spatial point coordinate corresponds one to one voxel, and the first predicted stratum attribute corresponding to the voxel can be obtained by processing the corresponding first spatial point coordinates by the first neural network model.

[0060] Optionally, the first neural network model may be a Bayesian neural network model.

[0061] S13. Generate a first formation prediction model and a first uncertainty model based on the first predicted formation attributes and the first probability value. The first formation prediction model is used to represent the first predicted formation attributes in the three-dimensional engineering area, and the first uncertainty model is used to represent the uncertainty of the first predicted formation attributes in the three-dimensional process area.

[0062] Optionally, the first formation prediction model is a first stereoscopic image of the three-dimensional engineering area, the first stereoscopic image includes the first predicted formation attributes, and the first predicted formation attributes can be represented by color in the first stereoscopic image.

[0063] Optionally, the first uncertainty model is a second stereoscopic image of the three-dimensional engineering area, and the second stereoscopic image contains the uncertainty of the first predicted formation properties in the horizontal distribution direction of the second stereoscopic image. The uncertainty of the first predicted formation properties can be represented by color in the horizontal distribution direction of the second stereoscopic image.

[0064] S14: Acquire positions of a plurality of first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model.

[0065] Optionally, the position of the first borehole may be the borehole position finally determined in the borehole position determination method.

[0066] According to the above description, the drilling position determination method described in this embodiment includes: obtaining a first spatial point coordinate set, the first spatial point coordinate in the first spatial point coordinate set is the spatial point coordinate of the centroid of the voxel in the voxel set of the three-dimensional engineering area; processing the first spatial point coordinate set through a first neural network model to obtain the first predicted formation attribute corresponding to the voxel and the first probability value corresponding to the first predicted formation attribute; generating a first formation prediction model and a first uncertainty model based on the first predicted formation attribute and the first probability value, the first formation prediction model is used to represent the first predicted formation attribute in the three-dimensional engineering area, and the first uncertainty model is used to represent the uncertainty of the first predicted formation attribute in the three-dimensional process area; obtaining the positions of several first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model.

[0067] In the borehole position determination method, the first spatial point coordinate set is processed by a first neural network model to obtain the first predicted formation attribute corresponding to the voxel and the first probability value corresponding to the first predicted formation attribute, and the positions of several first boreholes in the three-dimensional engineering area are obtained based on the first formation prediction model and the first uncertainty model. Compared with relying on the engineer's cognition to determine the borehole position, the rationality of the borehole position determination can be improved, and the borehole position determination method can correspondingly reduce labor costs, thereby reducing survey costs.

[0068] like Figure 3 As shown, this embodiment provides a method for obtaining the first neural network model, including:

[0069] S21, obtaining a second spatial point coordinate set of several second boreholes in the three-dimensional engineering area and a first real stratigraphic attribute corresponding to the second spatial point coordinates in the second spatial point coordinate set, the horizontal coordinate value and the vertical coordinate value of the second spatial point coordinate are the same horizontal coordinate value and the same vertical coordinate value, and the height coordinate value of the second spatial point coordinate is in the range of [a, b], where a represents the bottom elevation of the second borehole, and b represents the surface elevation of the second borehole.

[0070] Optionally, the first borehole and the second borehole have different locations in the three-dimensional engineering area. The second spatial point coordinate set is used to represent the coordinate position set of several points in the second borehole in the three-dimensional engineering area. For example, borehole 1 has three points b1, b2, and b3. The coordinates at b1 can be expressed as (4m, 4m, 0.75m), the coordinates at b2 can be expressed as (4m, 4m, 1m), and the coordinates at b3 can be expressed as (4m, 4m, 1.25m). The bottom elevation of the second borehole can refer to the deepest elevation value of the second borehole in the three-dimensional engineering area. The surface elevation of the second borehole can refer to the height of the second borehole on the surface in the three-dimensional engineering area, that is, the height of the second borehole mouth from the ground.

[0071] Optionally, the first real formation attribute may be an manually marked formation attribute corresponding to the second spatial point coordinate, and each second spatial point coordinate has a corresponding first real formation attribute. The first real formation attribute may refer to manually marked lithology or formation structure characteristics, etc.

[0072] Optionally, the second spatial point coordinates and the first real formation attributes corresponding to the second spatial point coordinates can be represented by one coordinate. For example, the coordinates at b1 can be expressed as (4m, 4m, 0.75m, D0), where D0 represents the first real formation attributes at b1.

[0073] S22: Based on the second spatial point coordinate set and the first real formation attributes, the initialized neural network model is trained by a back propagation algorithm to obtain the first neural network model.

[0074] Optionally, the initialized neural network model parameters have a total of 10 layers, the number of neurons in each hidden layer ranges from 16 to 128, the activation function can be a LeakyRelu activation function, and the optimizer can be an Adam optimizer.

[0075] Optionally, the structure of the initialized neural network model is determined by the data volume of the second borehole spatial coordinate set and the first real formation attribute, the prior distribution of the parameters of the initialized neural network model is Gaussian distribution or gamma distribution, the number of neurons in the input layer of the initialized neural network model is 3, and the number of neurons in the output layer of the initialized neural network model is the same as the number of formations in the three-dimensional engineering area.

[0076] Optionally, the implementation method of training the initialized neural network model by a back-propagation algorithm to obtain the first neural network model includes: estimating the posterior distribution probability of the parameters of the initialized neural network model by a Markov chain Monte Carlo method.

[0077] Optionally, the back-propagation algorithm fine-tunes the weights and biases in the initialized neural network model by calculating a loss function to improve the performance and accuracy of the initialized neural network model.

[0078] like Figure 4 As shown, this embodiment provides a method for obtaining positions of a plurality of first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model, including:

[0079] Step S1, based on the first formation prediction model and the first uncertainty model, obtain a third spatial coordinate point set of several third boreholes in the three-dimensional engineering area and a second real formation attribute corresponding to the third spatial point coordinates in the third spatial coordinate point set, the horizontal coordinate value and the vertical coordinate value of the third spatial point coordinate are the same horizontal coordinate value and the same vertical coordinate value, and the height coordinate value of the third spatial point coordinate is in the range of [c, d], c represents the bottom elevation of the third borehole, and d represents the surface elevation of the third borehole.

[0080] Optionally, the second real formation attribute may be an artificially marked formation attribute corresponding to the third spatial point coordinate, and each third spatial point coordinate has a corresponding second real formation attribute.

[0081] Optionally, the position of the third borehole may be different from the positions of the second borehole and the first borehole, and the third spatial coordinate point set is used to represent a coordinate position set of several points in the third borehole in the three-dimensional engineering area.

[0082] Step S2: Based on the third spatial point coordinate set, the second real formation attribute, the second spatial point coordinate set and the first real formation attribute, the first neural network model is trained to obtain a second neural network model. If the average cross-entropy value of the first neural network model during the training process is greater than the preset cross-entropy value, steps S3-S4 are continued to be executed and after step S4 is completed, step S1 is executed. The first formation prediction model in step S1 is the second formation prediction model, the first uncertainty model is the second uncertainty model, and the first neural network model in step S2 is the second neural network model. If the average cross-entropy value is not greater than the preset cross-entropy value, the third spatial coordinate point set of the third borehole is the position of the first borehole in the three-dimensional engineering area.

[0083] Optionally, the average cross entropy value can be expressed as:

[0084]

[0085] Wherein, L represents the average cross entropy value, M represents the number of formation attribute categories, and c is a value in the interval [1, M]. The corresponding predicted formation attribute category can be represented by c. For example, predicted formation attribute category 1 can be represented by c=1, and predicted formation attribute category 2 can be represented by c=2. Predicted formation attribute category 2 is the predicted formation attribute category corresponding to c=2. ic It is expressed as a sign function, which means that if the true stratum attribute category corresponding to the training sample i is the same as the stratum attribute category corresponding to c, then it takes 1, otherwise it takes 0, p ic represents the probability that the predicted stratigraphic attribute corresponding to the training sample i belongs to the stratigraphic attribute category corresponding to c, N represents the number of training samples, the predicted stratigraphic attribute is the first predicted stratigraphic attribute or the second predicted stratigraphic attribute, and the real stratigraphic attribute is the first real stratigraphic attribute or the second real stratigraphic predicted attribute. The second spatial point coordinate and its corresponding first real stratigraphic attribute or the third spatial point coordinate and its corresponding second real stratigraphic attribute can both be used as a training sample, and the predicted stratigraphic attribute category is the category of the predicted stratigraphic attribute. The category of the predicted stratigraphic attribute can be flexibly divided according to actual conditions. Similarly, the real stratigraphic attribute category is the category of the real stratigraphic attribute. In addition, the predicted stratigraphic attribute corresponding to the training sample i is the stratigraphic attribute output by the neural network model after predicting the training sample i, and the real stratigraphic attribute corresponding to the training sample i can be a manually labeled stratigraphic attribute.

[0086] Optionally, the average cross entropy value can be flexibly set according to actual conditions. During the loop process of S1-S4, multiple different "third space point coordinate sets" and "second neural network models" may be obtained. During the loop process, the training data of the first neural network model includes multiple different "third space point coordinate sets".

[0087] Step S3: Processing the first spatial point coordinate set through the second neural network model to obtain a second predicted stratum attribute corresponding to the voxel and a second probability value corresponding to the second predicted stratum attribute.

[0088] Optionally, the second predicted formation attribute may refer to the formation attribute predicted by the second neural network model. The second probability value may be a probability value that the formation attribute predicted by the second neural network model is the second predicted formation attribute.

[0089] Step S4: generating a second formation prediction model and a second uncertainty model based on the second predicted formation attribute and the second probability value.

[0090] According to the above description, by training the first neural network model with an expanded training data set to obtain the second neural network model, the accuracy of the neural network model finally obtained can be improved, thereby improving the rationality of drilling position determination.

[0091] like Figure 5 As shown, this embodiment provides a method for generating a second formation prediction model and a second uncertainty model based on the second predicted formation attribute and the second probability value, including:

[0092] S41, generating the second formation prediction model based on the second predicted formation attributes, wherein the second formation prediction model is a stereoscopic image of the three-dimensional engineering area, and the color of the voxels in the second formation prediction model is determined by the corresponding second predicted formation attributes, and each second predicted formation attribute has a corresponding color.

[0093] S42: Generate a second uncertainty model based on the second probability value, where the second uncertainty model includes uncertainty distribution of the three-dimensional engineering area in the horizontal direction.

[0094] Optionally, the uncertainty distribution of the three-dimensional engineering area in the horizontal direction may be the uncertainty distribution on the top surface of the three-dimensional engineering area. Since the three-dimensional engineering area contains multiple points where the horizontal projections of the body centroid are the same, the method for generating the second uncertainty model based on the second probability value includes: obtaining a set of body centroids in the three-dimensional engineering area where the horizontal projections of the body centroid are the same; and based on the set of body centroids, obtaining a cumulative value of the second probability values ​​corresponding to each voxel in a selected voxel set, the cumulative value being used to represent the uncertainty distribution of the three-dimensional engineering area in the horizontal direction. Each body centroid in the set of body centroids corresponds to a voxel, and the set formed by these voxels is the selected voxel set. The horizontal projection points of the body centroids may be the projection points of the body centroids on the top surface of the three-dimensional engineering area, and each set of body centroids corresponds to a cumulative value. The cumulative value can be represented on the top surface of the three-dimensional engineering area to represent the uncertainty distribution of the three-dimensional engineering area in the horizontal direction. The three-dimensional engineering area can be considered a closed three-dimensional image.

[0095] like Figure 6 As shown, this embodiment provides a method for obtaining a third spatial coordinate point set of a plurality of third boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model, including:

[0096] S51, based on the first formation prediction model and the first uncertainty model, obtain the first area and the second area in the three-dimensional process area, the first area is the area that affects the stability of the three-dimensional engineering area, and the stability is determined by the first predicted formation attribute, and the second area is the area with greater uncertainty in the second uncertainty model.

[0097] S52: Based on the first area and the second area, obtain the third spatial coordinate point set of the third borehole, where the position of the third spatial coordinate point set in the three-dimensional engineering area is within the first area or the second area.

[0098] The protection scope of the drilling position determination method described in the embodiment of the present application is not limited to the execution order of the steps listed in this embodiment. All solutions implemented by adding, reducing, or replacing steps in the existing technology based on the principles of the present application are included in the protection scope of the present application.

[0099] like Figure 7 As shown, this embodiment provides a drilling position determination device 700, and the drilling position determination device 700 includes:

[0100] The engineering information acquisition module 710 is configured to acquire a first set of spatial point coordinates, wherein the first spatial point coordinates in the first set of spatial point coordinates are the spatial point coordinates of the centroid of a body in a voxel set of a three-dimensional engineering area.

[0101] The network model processing module 720 is used to process the first spatial point coordinate set through a first neural network model to obtain a first predicted stratum attribute corresponding to the voxel and a first probability value corresponding to the first predicted stratum attribute.

[0102] The formation model generation module 730 is used to generate a first formation prediction model and a first uncertainty model based on the first predicted formation attributes and the first probability value, wherein the first formation prediction model is used to represent the first predicted formation attributes within the three-dimensional engineering area, and the first uncertainty model is used to represent the uncertainty of the first predicted formation attributes within the three-dimensional process area.

[0103] The borehole position determination module 740 is configured to obtain positions of a plurality of first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model.

[0104] In the drilling position determination device 700 provided in this embodiment, the engineering information acquisition module 710, the network model processing module 720, the formation model generation module 730 and the drilling position determination module 740 are connected with the drilling position determination device 700. Figure 2Steps S11 to S14 of the drilling position determination method correspond to each other and are not described in detail here.

[0105] In the several embodiments provided in this application, it should be understood that the disclosed devices or methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of modules / units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0106] The modules / units described as separate components may or may not be physically separate, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present application. For example, the functional modules / units in the various embodiments of the present application may be integrated into a processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into a single module / unit.

[0107] Those skilled in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0108] This embodiment provides an electronic device, which includes a memory storing a computer program; a processor connected to the memory for executing the computer program when the computer program is called; Figure 2 The method for determining the drilling position is shown.

[0109] The embodiment of the present application also provides a computer-readable storage medium. Those skilled in the art will understand that all or part of the steps in the method for implementing the above embodiment can be completed by instructing the processor through a program, and the program can be stored in a computer-readable storage medium, and the storage medium is a non-transitory medium, such as a random access memory, a read-only memory, a flash memory, a hard disk, a solid-state drive, a magnetic tape, a floppy disk, an optical disc, and any combination thereof. The above storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that includes one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a digital video disc (DVD)), or a semiconductor medium (for example, a solid-state drive (SSD)), etc.

[0110] The embodiment of the present application may also provide a computer program product, the computer program product including one or more computer instructions. When the computer instructions are loaded and executed on a computing device, the process or function described in the embodiment of the present application is generated in whole or in part. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer or data center to another website, computer or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method.

[0111] When the computer program product is executed by a computer, the computer executes the method described in the above method embodiment. The computer program product can be a software installation package. When the above method is needed, the computer program product can be downloaded and executed on the computer.

[0112] The descriptions of the processes or structures corresponding to the above figures have different emphases. For parts that are not described in detail in a certain process or structure, please refer to the relevant descriptions of other processes or structures.

[0113] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical concepts disclosed in this application shall be covered by the claims of this application.

Claims

1. A method for determining a drilling position, characterized in that: The drilling position determination method comprises: Acquire a first spatial point coordinate set, where the first spatial point coordinates in the first spatial point coordinate set are spatial point coordinates of a body centroid in a voxel set of a three-dimensional engineering area; Processing the first spatial point coordinate set through a first neural network model to obtain a first predicted stratum attribute corresponding to the voxel and a first probability value corresponding to the first predicted stratum attribute; generating a first formation prediction model and a first uncertainty model based on the first predicted formation attribute and the first probability value, wherein the first formation prediction model is used to represent the first predicted formation attribute in the three-dimensional engineering area, and the first uncertainty model is used to represent the uncertainty of the first predicted formation attribute in the three-dimensional process area; Obtaining positions of a plurality of first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model; The method for obtaining the first neural network model includes: obtaining a second spatial point coordinate set of several second boreholes in the three-dimensional engineering area and a first real stratigraphic attribute corresponding to the second spatial point coordinates in the second spatial point coordinate set, the horizontal coordinate value and the vertical coordinate value of the second spatial point coordinate are the same horizontal coordinate value and the same vertical coordinate value, and the height coordinate value of the second spatial point coordinate is in the range of [a, b], a represents the bottom elevation of the second borehole, and b represents the surface elevation of the second borehole; based on the second spatial point coordinate set and the first real stratigraphic attribute, the initialized neural network model is trained by a back propagation algorithm to obtain the first neural network model.

2. The method for determining a drilling position according to claim 1, wherein: The method for obtaining positions of a plurality of first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model includes: Step S1: Based on the first formation prediction model and the first uncertainty model, a third spatial coordinate point set of a plurality of third boreholes in the three-dimensional engineering area and second real formation attributes corresponding to the third spatial point coordinates in the third spatial coordinate point set are obtained, wherein the abscissa value and the ordinate value of the third spatial point coordinate are the same abscissa value and the same ordinate value, and the height coordinate value of the third spatial point coordinate is in the range of [c, d], where c represents the bottom elevation of the third borehole, and d represents the surface elevation of the third borehole; Step S2: Based on the third spatial point coordinate set, the second real formation attribute, the second spatial point coordinate set, and the first real formation attribute, the first neural network model is trained to obtain a second neural network model; if the average cross entropy value of the first neural network model during the training process is greater than the preset cross entropy value, then continue to execute steps S3-S4 and go to step S1 after step S4 is completed. The first formation prediction model in step S1 is the second formation prediction model, the first uncertainty model is the second uncertainty model, and the first neural network model in step S2 is the second neural network model. If the average cross entropy value is not greater than the preset cross entropy value, the third spatial coordinate point set of the third borehole is the position of the first borehole in the three-dimensional engineering area; Step S3, processing the first spatial point coordinate set by the second neural network model to obtain a second predicted stratum attribute corresponding to the voxel and a second probability value corresponding to the second predicted stratum attribute; Step S4: generating a second formation prediction model and a second uncertainty model based on the second predicted formation attribute and the second probability value.

3. The method for determining a drilling position according to claim 2, wherein: The method for generating a second formation prediction model and a second uncertainty model based on the second predicted formation attribute and the second probability value includes: generating a second stratum prediction model based on the second predicted stratum attribute, the second stratum prediction model being a stereoscopic image of the three-dimensional engineering area, the color of a voxel in the second stratum prediction model being determined by its corresponding second predicted stratum attribute, and each second predicted stratum attribute having a corresponding color; A second uncertainty model is generated based on the second probability value, where the second uncertainty model includes uncertainty distribution of the three-dimensional engineering area in a horizontal direction.

4. The method for determining a drilling position according to claim 2, wherein: The method for obtaining a third spatial coordinate point set of a plurality of third boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model includes: Based on the first formation prediction model and the first uncertainty model, a first region and a second region in the three-dimensional process area are obtained, wherein the first region is a region affecting the stability of the three-dimensional engineering area, the stability being determined by the first predicted formation attribute, and the second region is a region with greater uncertainty in the second uncertainty model; Based on the first area and the second area, the third spatial coordinate point set of the third borehole is acquired, where the position of the third spatial coordinate point set in the three-dimensional engineering area is within the first area or the second area.

5. The method for determining a drilling position according to claim 2, wherein: The structure of the initialized neural network model is determined by the data volume of the second borehole spatial coordinate set and the first real formation attribute. The prior distribution of the parameters of the initialized neural network model is Gaussian distribution or gamma distribution. The number of neurons in the input layer of the initialized neural network model is 3, and the number of neurons in the output layer of the initialized neural network model is the same as the number of formations in the three-dimensional engineering area.

6. The method for determining a drilling position according to claim 2, wherein: The average cross entropy value is expressed as: Wherein, L represents the average cross entropy value, M represents the number of formation attribute categories, and c is a value in the interval [1, M]. It is expressed as a sign function, which means that if the true stratum attribute category corresponding to the training sample i is the same as the stratum attribute category corresponding to c, then it takes 1, otherwise it takes 0. It represents the probability that the predicted formation attribute corresponding to the training sample i belongs to the formation attribute category corresponding to c, N represents the number of training samples, the predicted formation attribute is the first predicted formation attribute or the second predicted formation attribute, and the real formation attribute is the first real formation attribute or the second real formation predicted attribute.

7. A drilling position determination device, characterized in that: The drilling position determining device comprises: A project information acquisition module, configured to acquire a first set of spatial point coordinates, wherein the first spatial point coordinates in the first set of spatial point coordinates are spatial point coordinates of a centroid of a body in a voxel set of a three-dimensional project area; a network model processing module, configured to process the first spatial point coordinate set using a first neural network model to obtain a first predicted stratum attribute corresponding to the voxel and a first probability value corresponding to the first predicted stratum attribute; a formation model generation module, configured to generate a first formation prediction model and a first uncertainty model based on the first predicted formation attribute and the first probability value, wherein the first formation prediction model is used to represent the first predicted formation attribute in the three-dimensional engineering area, and the first uncertainty model is used to represent the uncertainty of the first predicted formation attribute in the three-dimensional process area; a borehole position determination module, configured to obtain positions of a plurality of first boreholes in the three-dimensional engineering area based on the first formation prediction model and the first uncertainty model; The method for obtaining the first neural network model includes: obtaining a second spatial point coordinate set of several second boreholes in the three-dimensional engineering area and a first real stratigraphic attribute corresponding to the second spatial point coordinates in the second spatial point coordinate set, the horizontal coordinate value and the vertical coordinate value of the second spatial point coordinate are the same horizontal coordinate value and the same vertical coordinate value, and the height coordinate value of the second spatial point coordinate is in the range of [a, b], a represents the bottom elevation of the second borehole, and b represents the surface elevation of the second borehole; based on the second spatial point coordinate set and the first real stratigraphic attribute, the initialized neural network model is trained by a back propagation algorithm to obtain the first neural network model.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for determining the drilling position according to any one of claims 1 to 6 is implemented.

9. An electronic device, characterized in that: The electronic device comprises: a memory storing a computer program; A processor is communicatively connected to the memory and executes the drilling position determination method according to any one of claims 1 to 6 when calling the computer program.

Images