Oil layer subdivision layer sand body prediction method, device and system and storage medium

Abstract

The invention discloses an oil layer subdivision layer sand body prediction method, device and system and a storage medium. The method comprises the steps that S1, historical seismic data and historical logging data of a preset area are acquired; s2, obtaining a sand body prediction model according to the historical seismic data and the historical logging data; wherein the sand body prediction model is a multi-dimensional parameter prediction model based on a selective kernel convolutional neural network SK-CNN; and S3, inputting the seismic data of the target area into the sand body prediction model, and predicting the oil sand body distribution of the target area through a preset equivalent weight matrix. By adopting the technical scheme of the invention, the precision of predicting the oil sand body distribution is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of oil reservoir sand body prediction, and in particular relates to a method and device, system, and storage medium for predicting oil reservoir sand bodies by fine layers. Background Art

[0002] Due to lithological changes, some sandstone layers with relatively high permeability are separated by some layers with low permeability into several independent sand layer individuals that are not connected to each other, which are called sand bodies. The sand bodies in the oil reservoir are called oil sand bodies. In lithologic oil reservoirs, oil sand bodies are the smallest oil-bearing units in the underground oil reservoir and are also the most basic units controlling the movement of oil and water. In oilfield geological work and oilfield development work, how to effectively and accurately predict oil sand bodies and their boundaries and improve the drilling encounter rate of oil layers is the key to oil reservoir development. Currently, the more commonly used methods for predicting the distribution of oil sand bodies mainly include two types. One is based on well logging data, the cross-well interpolation algorithm. Well logging data can intuitively reflect the lithology, physical properties, and oil and gas bearing characteristics and changes of the reservoir. However, in oilfields with fewer wells and complex reservoir distributions, the reliability and accuracy of using cross-well interpolation to predict the distribution of oil sand bodies are not high. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a method and device, system, and storage medium for predicting oil reservoir sand bodies by fine layers.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for predicting oil reservoir sand bodies by fine layers includes:

[0006] Step S1: Obtain historical seismic data and historical well logging data of a preset area;

[0007] Step S2: Obtain a sand body prediction model according to the historical seismic data and historical well logging data; wherein, the sand body prediction model is a multi-dimensional parameter prediction model based on the selective kernel convolutional neural network SK-CNN;

[0008] Step S3: Input the seismic data of the target area into the sand body prediction model, and predict the distribution of oil sand bodies in the target area through a preset equivalent weight matrix.

[0009] Preferably, preprocess the historical seismic data and historical well logging data, including data cleaning to eliminate the influence of the original data dimension on the results.

[0010] Preferably, the multi-dimensional parameter prediction model includes: thirteen Conv2D convolutional layers, ten BN normalization layers, ten ReLU activation layers, a Concatenate splicing layer, six Add summation layers, a MaxPooling2D max pooling layer, a CBS module, and an SPPF module.

[0011] Preferably, step S2 includes:

[0012] Construct a sample sand body set according to historical seismic data and historical logging data;

[0013] Extract the specific seismic attributes and specific logging attributes of each sample sand body in the sample sand body set, and construct a sample attribute set according to the specific seismic attributes and specific logging attributes;

[0014] Determine an equivalent weight matrix according to the sample attribute set;

[0015] Obtain a sand body prediction model for predicting the distribution of oil sand bodies in a preset area according to the equivalent weight matrix.

[0016] The present invention also provides an oil reservoir sub-layer sand body prediction device, including:

[0017] A first processing module, configured to obtain historical seismic data and historical logging data of a preset area;

[0018] A second processing module, configured to obtain a sand body prediction model according to historical seismic data and historical logging data; wherein, the sand body prediction model is a multi-dimensional parameter prediction model based on a selective kernel convolutional neural network SK-CNN;

[0019] A third processing module, configured to input the seismic data of the target area into the sand body prediction model, and predict the distribution of oil sand bodies in the target area through a preset equivalent weight matrix.

[0020] Preferably, the first processing module is further configured to preprocess the historical seismic data and historical logging data, including data cleaning, to eliminate the influence of the original dimension of the data on the result.

[0021] Preferably, the multi-dimensional parameter prediction model includes: thirteen Conv2D convolutional layers, ten BN normalization layers, ten ReLU activation layers, a Concatenate splicing layer, six Add summation layers, a MaxPooling2D max pooling layer, a CBS module, and an SPPF module.

[0022] Preferably, the second processing module includes:

[0023] A first processing unit, configured to construct a sample sand body set according to historical seismic data and historical logging data;

[0024] A second processing unit, configured to extract the specific seismic attributes and specific logging attributes of each sample sand body in the sample sand body set, and construct a sample attribute set according to the specific seismic attributes and specific logging attributes;

[0025] A third processing unit, configured to determine an equivalent weight matrix according to the sample attribute set;

[0026] A fourth processing unit, configured to obtain a sand body prediction model for predicting the distribution of oil sand bodies in a preset area according to an equivalent weight matrix.

[0027] The present invention further provides an oil reservoir fine-layer sand body prediction system, including: a memory and a processor, where a computer program run by the processor is stored on the memory, and the computer program executes an oil reservoir fine-layer sand body prediction method when run by the processor.

[0028] The present invention further provides a storage medium, on which a computer program is stored, and the computer program executes an oil reservoir fine-layer sand body prediction method when running.

[0029] The present invention obtains a sand body prediction model according to historical seismic data and historical logging data; wherein, the sand body prediction model is a multi-dimensional parameter prediction model based on a selective kernel convolutional neural network (SK-CNN); the seismic data of a target area is input into the sand body prediction model, and the distribution of oil sand bodies in the target area is predicted through a preset equivalent weight matrix. By adopting the technical solution of the present invention, the accuracy of predicting the distribution of oil sand bodies is improved. Description of the Drawings

[0030] 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 use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0031] Figure 1 It is a flowchart of the oil reservoir fine-layer sand body prediction method according to the embodiment of the present invention. Detailed Embodiments

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0033] To make the above objects, features, and advantages of the present invention more obvious and understandable, the present invention will be further described in detail below with reference to the drawings and specific embodiments.

[0034] Embodiment 1:

[0035] As Figure 1 shown, the embodiment of the present invention provides an oil reservoir fine-layer sand body prediction method, including:

[0036] Step S1: Obtain the historical seismic data and historical logging data of a preset area;

[0037] Step S2: Obtain a sand body prediction model based on the historical seismic data and historical logging data; wherein, the sand body prediction model is a multi-dimensional parameter prediction model based on the Selective Kernel Convolutional Neural Network (SK-CNN);

[0038] Step S3: Input the seismic data of the target area into the sand body prediction model, and predict the distribution of oil sand bodies in the target area through a preset equivalent weight matrix.

[0039] As an implementation manner of an embodiment of the present invention, preprocess the historical seismic data and historical logging data, including data cleaning to eliminate the influence of the original dimension of the data on the result, specifically including:

[0040] Screen out the garbled data and out-of-range data from the historical seismic data and historical logging data to obtain the screened index data;

[0041] Perform a vectorization operation on the screened index data to obtain a screened index vector set, and add spatio-temporal vectors to each screened index vector in the screened index vector set to obtain a spatio-temporal vector set;

[0042] Perform feature clustering on the spatio-temporal vector set to obtain a clustering center data set;

[0043] Fill in each null value data in the screened index data according to the clustering center data set.

[0044] As an implementation manner of an embodiment of the present invention, the multi-dimensional parameter prediction model includes: thirteen Conv2D convolutional layers, ten BN normalization layers, ten ReLU activation layers, a Concatenate splicing layer, six Add summation layers, a MaxPooling2D max pooling layer, a CBS module, and an SPPF module. Specifically:

[0045] The first Conv2D convolutional layer, the first BN normalization layer, and the first ReLU activation layer are connected in sequence; the output end of the first ReLU activation layer is respectively connected to the input ends of the second Conv2D convolutional layer, the third Conv2D convolutional layer, the fourth Conv2D convolutional layer, the fifth Conv2D convolutional layer, the sixth Conv2D convolutional layer, the seventh Conv2D convolutional layer, the eighth Conv2D convolutional layer, the ninth Conv2D convolutional layer, the tenth Conv2D convolutional layer, the eleventh Conv2D convolutional layer, the input end of the first Add summation layer, and the input end of the fourth Add summation layer; the output end of the first Add summation layer is connected to the input end of the fourth Add summation layer;

[0046] The second Conv2D convolutional layer, the second BN normalization layer, the second ReLU activation layer, the twelfth Conv2D convolutional layer, and the first Add summation layer are connected in sequence; the third Conv2D convolutional layer, the third BN normalization layer, the third ReLU activation layer, and the Concatenate splicing layer are connected in sequence; the fourth Conv2D convolutional layer, the fourth BN normalization layer, the fourth ReLU activation layer, and the Concatenate splicing layer are connected in sequence; the output end of the Concatenate splicing layer is connected to the input end of the fourth Add summation layer;

[0047] The fifth Conv2D convolutional layer, the fifth BN normalization layer, the fifth ReLU activation layer, and the second Add summation layer are connected in sequence; the sixth Conv2D convolutional layer, the sixth BN normalization layer, the sixth ReLU activation layer, and the second Add summation layer are connected in sequence; the seventh Conv2D convolutional layer, the seventh BN normalization layer, the seventh ReLU activation layer, and the second Add summation layer are connected in sequence;

[0048] The output end of the second Add summation layer is connected to the input end of the eleventh Conv2D convolutional layer; the output end of the eleventh Conv2D convolutional layer is connected to the input end of the third Add summation layer; the output end of the third Add summation layer is connected to the input end of the fourth Add summation layer; the output end of the fourth Add summation layer is connected to the input end of the MaxPooling2D max pooling layer;

[0049] The eighth Conv2D convolutional layer, the eighth BN normalization layer, the eighth ReLU activation layer, and the fifth Add summation layer are connected in sequence; the ninth Conv2D convolutional layer, the ninth BN normalization layer, the ninth ReLU activation layer, and the fifth Add summation layer are connected in sequence; the tenth Conv2D convolutional layer, the tenth BN normalization layer, the tenth ReLU activation layer, and the fifth Add summation layer are connected in sequence; the fifth Add summation layer, the thirteenth Conv2D convolutional layer, and the sixth Add summation layer are connected in sequence; the output end of the sixth Add summation layer is connected to the input end of the fourth Add summation layer;

[0050] The output end of the MaxPooling2D max pooling layer, the CBS module, and the SPPF module are connected in sequence;

[0051] Among them, the kernels of the first Conv2D convolutional layer and the second Conv2D convolutional layer are both 3×3, the kernel of the third Conv2D convolutional layer is 3×1, the kernel of the fourth Conv2D convolutional layer is 1×3, the kernel of the fifth Conv2D convolutional layer is 5×5, the kernel of the sixth Conv2D convolutional layer is 5×1, the kernel of the seventh Conv2D convolutional layer is 1×5, the kernel of the eighth Conv2D convolutional layer is 7×7, the kernel of the ninth Conv2D convolutional layer is 7×1, the kernel of the tenth Conv2D convolutional layer is 1×7, and the kernels of the eleventh Conv2D convolutional layer, the twelfth Conv2D convolutional layer, and the thirteenth Conv2D convolutional layer are all 1×1.

[0052] As an implementation manner of the embodiment of the present invention, step S2 includes:

[0053] According to the historical seismic data and historical logging data, obtain the sample sand bodies in the preset area that meet the preset conditions, and construct a sample sand body set according to the sample sand bodies. The sample sand body set includes at least one oil sand body and at least one non - oil sand body;

[0054] Extract the specific seismic attributes and specific logging attributes of each sample sand body in the sample sand body set, and construct a sample attribute set according to the specific seismic attributes and specific logging attributes;

[0055] Determine the equivalent weight matrix according to the sample attribute set;

[0056] Obtain a sand body prediction model for predicting the distribution of oil sand bodies in the preset area according to the equivalent weight matrix.

[0057] Further, determining the equivalent weight matrix according to the sample attribute set specifically includes: obtaining sensitive attributes that meet the second preset condition according to each attribute in the sample attribute set, and constructing a sensitive attribute set according to the sensitive attributes, where the sensitive attribute set includes at least one seismic sensitive attribute and at least one logging sensitive attribute;

[0058] Determining the equivalent weight matrix according to the sensitive attribute set.

[0059] As an implementation manner of an embodiment of the present invention, step S3 is specifically: obtaining the seismic attributes corresponding to the specific seismic attributes in the seismic data of the target area and the seismic sensitive attributes in the seismic data according to the seismic data of the target area, and combining the equivalent weight matrix to predict the distribution of oil sand bodies in the study area, where the seismic sensitive attributes correspond to the seismic sensitive attributes in the sensitive attribute set.

[0060] Embodiment 2:

[0061] The present invention also provides an oil reservoir fine-layer sand body prediction device, including:

[0062] A first processing module, configured to obtain historical seismic data and historical logging data of a preset area;

[0063] A second processing module, configured to obtain a sand body prediction model according to the historical seismic data and the historical logging data; wherein, the sand body prediction model is a multi-dimensional parameter prediction model based on the selective kernel convolutional neural network SK-CNN;

[0064] A third processing module, configured to input the seismic data of the target area into the sand body prediction model and predict the distribution of oil sand bodies in the target area through a preset equivalent weight matrix.

[0065] As an implementation manner of an embodiment of the present invention, the first processing module is further configured to preprocess the historical seismic data and the historical logging data, including data cleaning to eliminate the influence of the original dimension of the data on the result.

[0066] As an implementation manner of an embodiment of the present invention, the multi-dimensional parameter prediction model includes: thirteen Conv2D convolutional layers, ten BN normalization layers, ten ReLU activation layers, a Concatenate splicing layer, six Add summation layers, a MaxPooling2D max pooling layer, a CBS module, and an SPPF module.

[0067] As an implementation manner of an embodiment of the present invention, the second processing module includes:

[0068] A first processing unit, configured to construct a sample sand body set according to the historical seismic data and the historical logging data;

[0069] A second processing unit, configured to extract specific seismic attributes and specific logging attributes of each sample sand body in the sample sand body set, and construct a sample attribute set according to the specific seismic attributes and specific logging attributes;

[0070] A third processing unit, configured to determine an equivalent weight matrix according to the sample attribute set;

[0071] A fourth processing unit, configured to obtain a sand body prediction model for predicting the distribution of oil sand bodies in a preset area according to the equivalent weight matrix.

[0072] Embodiment 3:

[0073] The present invention further provides an oil reservoir sand body prediction system for sub - layers, including: a memory and a processor, where a computer program run by the processor is stored on the memory, and the computer program, when run by the processor, executes the oil reservoir sand body prediction method for sub - layers.

[0074] Embodiment 4:

[0075] The present invention further provides a storage medium, where a computer program is stored on the storage medium, and the computer program, when running, executes the oil reservoir sand body prediction method for sub - layers.

[0076] The above - described embodiments are only descriptions of the preferred embodiments of the present invention, and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for predicting sand bodies in subdivided layers of an oil reservoir, characterized in that, Including: Step S1: Obtain historical seismic data and historical logging data of a preset area; Step S2: Obtain a sand body prediction model based on the historical seismic data and historical logging data; wherein, the sand body prediction model is a multi-dimensional parameter prediction model based on the Selective Kernel Convolutional Neural Network (SK-CNN); Step S3: Input the seismic data of the target area into the sand body prediction model, and predict the distribution of oil sand bodies in the target area through a preset equivalent weight matrix.

2. The method for predicting the sand body of the oil reservoir by fine layer division according to claim 1, wherein In Step S1, preprocess the historical seismic data and historical logging data, including data cleaning to eliminate the influence of the original dimension of the data on the results.

3. The method for predicting the sand body of a subdivided oil layer according to claim 2, wherein, The multi-dimensional parameter prediction model includes: thirteen Conv2D convolutional layers, ten BN normalization layers, ten ReLU activation layers, a Concatenate splicing layer, six Add summation layers, a MaxPooling2D max pooling layer, a CBS module, and an SPPF module.

4. The method for predicting the sand body of the oil reservoir by fine layer division according to claim 3, characterized in that, Step S2 includes: Construct a sample sand body set based on the historical seismic data and historical logging data; Extract the specific seismic attributes and specific logging attributes of each sample sand body in the sample sand body set, and construct a sample attribute set based on the specific seismic attributes and specific logging attributes; Determine an equivalent weight matrix based on the sample attribute set; Obtain a sand body prediction model for predicting the distribution of oil sand bodies in the preset area based on the equivalent weight matrix.

5. An apparatus for predicting sand bodies in subdivided oil layers, characterized in that, Including: A first processing module for obtaining historical seismic data and historical logging data of a preset area; A second processing module for obtaining a sand body prediction model based on the historical seismic data and historical logging data; wherein, the sand body prediction model is a multi-dimensional parameter prediction model based on the Selective Kernel Convolutional Neural Network (SK-CNN); A third processing module for inputting the seismic data of the target area into the sand body prediction model and predicting the distribution of oil sand bodies in the target area through a preset equivalent weight matrix.

6. The sand body prediction device for fine layer division of oil reservoir according to claim 5, characterized in that, The first processing module is also used for preprocessing the historical seismic data and historical logging data, including data cleaning to eliminate the influence of the original dimension of the data on the results.

7. The sand body prediction device for fine layering of oil reservoir according to claim 6, characterized in that, The multi-dimensional parameter prediction model includes: thirteen Conv2D convolutional layers, ten BN normalization layers, ten ReLU activation layers, a Concatenate splicing layer, six Add summation layers, a MaxPooling2D max pooling layer, a CBS module, and an SPPF module.

8. The sand body prediction device for fine layer division of oil reservoir according to claim 7, wherein, The second processing module includes: A first processing unit for constructing a sample sand body set based on the historical seismic data and historical logging data; A second processing unit for extracting the specific seismic attributes and specific logging attributes of each sample sand body in the sample sand body set and constructing a sample attribute set based on the specific seismic attributes and specific logging attributes; A third processing unit for determining an equivalent weight matrix based on the sample attribute set; A fourth processing unit for obtaining a sand body prediction model for predicting the distribution of oil sand bodies in the preset area based on the equivalent weight matrix.

9. An oil reservoir fine-layer sand body prediction system, characterized in that, Including: A memory and a processor, wherein a computer program is stored on the memory and is run by the processor, and the computer program, when run by the processor, executes the oil reservoir sub - layer sand body prediction method according to any one of claims 1 - 4.

10. A storage medium, characterized in that, A computer program is stored on the storage medium, and the computer program, when running, executes the oil reservoir sub - layer sand body prediction method according to any one of claims 1 - 4.

Images