A method, device, medium and product for dynamically predicting the production of a fractured oil and gas well

Abstract

The present application discloses a method, device, medium and product for dynamically predicting the production of a fractured oil and gas well, relating to the field of oil and gas field development. First, geological parameter data and production dynamic data of different types of reservoirs in different oil and gas reservoirs are obtained; the geological parameter data is converted into image data; an unsupervised clustering method is used to classify the image data; a data set is constructed based on the image data and classification numbers to train a deep learning model, and after the training is completed, it is used as a geological parameter classification model to obtain the predicted classification number of the image data to be predicted and retrieve the target image data and the image to be corrected accordingly. The image to be corrected is selected from the Gram angular field image generated from the production dynamic data; the difference image between the image data to be predicted and the target image data is calculated to correct the image to be corrected; the corrected image is inversely converted into the predicted production dynamic data, which can greatly reduce the calculation time required for production dynamic prediction and reduce the prediction cost.

Description

Technical Field

[0001] The present application relates to the technical field of oil and gas field development, and particularly to a method, device, medium and product for dynamically predicting the production of fractured oil and gas wells. Background Art

[0002] Tight oil and gas reservoirs have low permeability and low productivity in conventional wells. Generally, hydraulic fracturing technology is required to obtain economic productivity. Accurately predicting the production dynamics of fractured oil and gas wells is the basis for formulating reliable development plans and accurately evaluating development benefits. At present, the production dynamics prediction of fractured oil and gas wells is mostly based on reservoir numerical simulation, which must rely on expensive numerical simulation software, and the simulation calculation is time-consuming and laborious. Summary of the Invention

[0003] The purpose of the present application is to provide a method, device, medium and product for dynamically predicting the production of fractured oil and gas wells, so as to greatly reduce the calculation time required for production dynamics prediction and reduce the prediction cost.

[0004] To achieve the above object, the present application provides the following solutions.

[0005] In a first aspect, the present application provides a method for dynamically predicting the production of fractured oil and gas wells, including:

[0006] Obtaining multiple groups of geological parameter data and corresponding production dynamic data of different types of reservoirs in different oil and gas reservoirs; the production dynamic data is production data at different times;

[0007] Converting each group of geological parameter data into corresponding image data;

[0008] Using an unsupervised clustering method to classify the image data, optimizing the classification results, and numbering the optimized clusters as the final classification;

[0009] Constructing a data set based on the image data and the corresponding classification numbers, training a deep learning model, and using it as a geological parameter classification model after training; the input of the geological parameter classification model is image data, and the output is the corresponding classification number;

[0010] Processing the production dynamic data using Gramian angular field to generate a Gramian angular field image;

[0011] Correspondingly storing the Gramian angular field image and the image data converted from the corresponding geological parameter data to construct a database;

[0012] Converting the geological parameter data to be predicted into image data to be predicted, inputting it into the geological parameter classification model for classification, and obtaining a predicted classification number;

[0013] Retrieve, from the database based on the predicted classification number, the image data with the highest structural similarity under the same classification number as the target image data;

[0014] Calculate the difference image between the image data to be predicted and the target image data using the structural similarity method;

[0015] Take the Gram angular field image corresponding to the target image data in the database as the image to be corrected, and correct the image to be corrected based on the difference image to obtain the corrected image;

[0016] Inverse-transform the corrected image into the predicted production dynamic data.

[0017] Optionally, the obtaining of multiple groups of geological parameter data for different types of reservoirs in different oil and gas reservoirs specifically includes:

[0018] If the reservoir type is an oil reservoir, each group of geological parameter data obtained includes thickness, porosity, permeability, formation pressure, oil-gas-water saturation, oil-gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, and relative permeability curve data; PVT data includes pressure, volume, and temperature data;

[0019] If the reservoir type is a tight gas reservoir, each group of geological parameter data obtained includes thickness, porosity, permeability, formation pressure, gas-water saturation, gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, and relative permeability curve data;

[0020] If the reservoir type is a shale gas reservoir or a coalbed methane reservoir, each group of geological parameter data obtained includes thickness, porosity, permeability, formation pressure, gas-water saturation, gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, relative permeability curve, Langmuir volume, and Langmuir pressure data.

[0021] Optionally, the converting of each group of geological parameter data into corresponding image data specifically includes:

[0022] According to the number of types of geological parameter data included in each group of geological parameter data Determine the width and height of the image data to be converted; and , is a fixed padding to make an integer;

[0023] Normalize each geological parameter data included in each group of geological parameter data, and normalize the value range to between [0 - 1];

[0024] The normalized data is mapped to the range of [0 - 255], which is expressed as the gray value of the grayscale image;

[0025] The mapped gray values are arranged from top to bottom and from left to right, and reconstructed into rows columns of a two-dimensional array. Each position in the two-dimensional array represents a pixel position, and the gray value at each position represents the gray value of the pixel. The positions without gray values are filled with zeros, thus forming a grayscale image as the corresponding image data.

[0026] Optionally, the unsupervised clustering method is used to classify the image data, and the classification result is optimized. The optimized clusters are numbered as the final classification, specifically including:

[0027] The Gaussian kernel function is used to calculate the similarity between every two pieces of image data, and a similarity matrix is constructed;

[0028] The k-nearest neighbor graph is used to construct a similarity graph corresponding to the similarity matrix, and a Laplacian matrix is constructed based on the similarity graph;

[0029] The Laplacian matrix is eigen-decomposed, and the first eigenvectors of the Laplacian matrix are calculated. These eigenvectors are used as columns to form matrix U. Each row in matrix U is a dimensional row vector, and there are a total of row vectors;

[0030] These row vectors are clustered using the K-means clustering method to obtain the distribution of clusters, and each cluster represents a classification;

[0031] Based on the DB index, the clusters included in the classification result are optimized, and the optimized clusters are numbered as the final classification.

[0032] Optionally, the Gramian angular field is used to process the production dynamic data to generate a Gramian angular field image, specifically including:

[0033] The production dynamic data is represented in tuple form as ; where each pair of values refers to the production data at time is ; , is the number of sampling points;

[0034] The production data is scaled to the range of to obtain the scaled production data ;

[0035] Represent the scaled production data in the polar coordinate system , obtaining The polar angle in the polar coordinate system and the polar radius ;

[0036] Based on the polar angle and the polar radius Generate a Gram angle field image.

[0037] Optionally, the method for calculating the difference image between the image data to be predicted and the target image data using the structural similarity method specifically includes:

[0038] Calculate the structural similarity data between the image data to be predicted and the target image data using the structural similarity method;

[0039] Convert the structural similarity data into corresponding image data as the difference image.

[0040] Optionally, the method for correcting the image to be corrected based on the difference image to obtain the corrected image specifically includes:

[0041] Divide the image to be corrected into pixel blocks;

[0042] Based on the gray value of the corresponding pixel point in the difference image of size Adjust the RGB channel color values of the corresponding pixel blocks in the

[0043] pixel blocks to obtain the corrected image.

[0044] In a second aspect, the present application provides a computer device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, and the processor executes the computer program to implement the method for dynamically predicting the production of a fractured oil and gas well.

[0045] In a third aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method for dynamically predicting the production of a fractured oil and gas well is implemented.

[0046] In a fourth aspect, the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method for dynamically predicting the production of a fractured oil and gas well is implemented.

[0047] A method, device, medium and product for dynamically predicting the production of a fractured oil and gas well provided by the present application pre-establishes a data set composed of image data and corresponding classification numbers, trains a deep learning model by using a deep learning method, and can quickly classify the geological parameter data to be predicted by using the trained geological parameter classification model; then, based on the predicted classification number, the target image data can be quickly retrieved from the database; the image to be corrected is corrected by using the difference image between the image data to be predicted and the target image data, and the corrected image is inversely transformed to obtain the predicted production dynamic data. The method of the present application does not need to rely on expensive numerical simulation software for reservoir numerical simulation, so it greatly reduces the calculation time required for production dynamic prediction and reduces the prediction cost. Brief Description of the Drawings

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0049] Figure 1 It is a schematic flow chart of a method for dynamically predicting the production of a fractured oil and gas well according to the present application;

[0050] Figure 2 It is a schematic diagram of the structural similarity data between the image data to be predicted and the target image data in an exemplary embodiment;

[0051] Figure 3 It is a schematic diagram of dividing the image to be corrected into pixel blocks in an exemplary embodiment. Detailed Embodiments

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0053] The present application proposes a method, device, medium and product for dynamically predicting the production of a fractured oil and gas well, aiming to greatly reduce the calculation time required for production dynamic prediction and reduce the prediction cost.

[0054] In order to make the above objects, features and advantages of the present application more obvious and understandable, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0055] In an exemplary embodiment, as Figure 1 shown, a method for dynamically predicting the production of a fractured oil and gas well is provided, including the following steps 1 to 11.

[0056] Step 1: Obtain multiple sets of geological parameter data of different types of reservoirs in different oil and gas reservoirs and the corresponding production dynamic data.

[0057] In the data collection stage, for different types of reservoirs in different oil and gas reservoirs, multiple sets of geological parameter data and the corresponding production dynamic data are collected. The types of reservoirs include oil reservoirs, tight gas reservoirs, shale gas reservoirs, coalbed methane reservoirs, etc.

[0058] Specifically, if the reservoir type is an oil reservoir, each set of geological parameter data obtained includes data such as thickness, porosity, permeability, formation pressure, oil-gas-water saturation, oil-gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, and relative permeability curve. PVT data includes pressure, volume, and temperature data; where P represents the pressure of the oil and gas reservoir (pressure), V represents volume (volume), and T represents temperature (temperature).

[0059] If the reservoir type is a tight gas reservoir, each set of geological parameter data obtained includes data such as thickness, porosity, permeability, formation pressure, gas-water saturation, gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, and relative permeability curve.

[0060] If the reservoir type is a shale gas reservoir or a coalbed methane reservoir, each set of geological parameter data obtained includes data such as thickness, porosity, permeability, formation pressure, gas-water saturation, gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, relative permeability curve, Langmuir volume, and Langmuir pressure.

[0061] The production dynamic data refers to the production data at different times (time), that is, the time series of production data. Each set of geological parameter data corresponds to a set of production dynamic data.

[0062] Step 2: Convert each set of geological parameter data into corresponding image data.

[0063] Perform overall processing on each set of geological parameter data in multiple sets of geological parameter data, and convert each set of geological parameter data into a corresponding image data. Denote the number of types of geological parameter data included in each set of geological parameter data as , that is to say, each set of geological parameter data includes geological parameter data. Based on the number of types determine the width and height of the image data to be converted, so as to Reshape the geological parameter data, and satisfy the following formula:

[0064] (1);

[0065] where is a fixed padding to make an integer.

[0066] For the deep learning model to be used in step 4, to ensure efficiency and performance during training and inference, make the convolutional kernel and stride consistent. It is necessary to make the side lengths of the reshaped image data equal, that is and also need to satisfy the following formula:

[0067] (2).

[0068] Determine the width and height After that, normalize each geological parameter data contained in each group of geological parameter data, and normalize its value range to between [0 - 1]. Then map the normalized data to the range of [0 - 255], which is represented as the gray value of the grayscale image.

[0069] Arrange the mapped gray values in the order from top to bottom and from left to right, and reconstruct them into a row column two-dimensional array. Each position in the two-dimensional array represents a pixel point position, and the gray value of each position represents the gray value of the pixel point, thus forming a grayscale image as the corresponding image data. In the row column two-dimensional array, fill the positions without gray values with zeros. That is to say, there are positions that need to be filled with zeros.

[0070] Step 3: Use an unsupervised clustering method to classify the image data, optimize the classification results, and number the optimized clusters as the final classification.

[0071] For multiple groups of image data converted from multiple groups of geological parameter data, use the method of spectral clustering + K-means clustering for classification. Spectral clustering transforms the clustering problem into a graph partitioning problem by using the graph structure of the data, while K-means clustering groups the data by iteratively optimizing the positions of the cluster centers. By obtaining the low-dimensional feature representation through spectral clustering and then performing K-means clustering, it can effectively handle geological parameter data with complex shapes and improve the clustering effect. The specific steps of step 3 include the following steps 3.1 to 3.5.

[0072] Step 3.1: Calculate the similarity between every two image data using the Gaussian kernel function and construct a similarity matrix.

[0073] Take each piece of image data in the multiple pieces of image data as a sample, and calculate the similarity between the samples using the Gaussian kernel function, which can reduce the similarity of samples that are far apart to close to zero, thus highlighting the local structure. The calculation formula for the Gaussian kernel similarity is:

[0074] (3);

[0075] where, and represent two different samples; is the Euclidean distance between samples and ; is a parameter that controls the attenuation rate of the similarity. is the similarity matrix the -th column element.

[0076] The calculation formula for the Euclidean distance is:

[0077] (4);

[0078] where and respectively represent the gray values of the -th pixel positions in samples .

[0079] Step 3.2: Use the k-nearest neighbor graph to construct a similarity graph corresponding to the similarity matrix, and construct a Laplacian matrix based on the similarity graph.

[0080] According to the calculated similarity matrix construct a similarity graph, which is constructed using the k-nearest neighbor graph. Specifically, take each element in the similarity matrix as a node, and the constructed similarity graph models the local neighborhood relationship between the nodes. Take the connection between two nodes as an edge, and the construction result is a directed graph. Since the nearest neighbor relationship is asymmetric, after connecting all the nodes, for each node in the directed graph, find its k neighbor nodes with the highest structural similarity, and then only retain the similarities of these neighbor nodes, and set other similarities to 0 to construct a similarity graph. After constructing the similarity graph, assign weights to the edges according to the similarity of the nodes, and calculate the degree matrix D and the adjacency matrix W of the nodes. Then the Laplacian matrix formula is L = D - W. Where L is the Laplacian matrix.

[0081] Step 3.3: Perform eigenvalue decomposition on the Laplacian matrix L, and calculate the first eigenvectors of the Laplacian matrix. Use these eigenvectors as columns to form matrix U. Each row in matrix U is a -dimensional row vector, and there are a total of row vectors.

[0082] Step 3.4: Use the K-means clustering method to cluster these row vectors, classify all the image data converted from geological parameter data, and obtain the distribution of clusters. Each cluster represents a new classification.

[0083] Step 3.5: Optimize the clusters included in the classification result based on the DB index, and number the optimized clusters as the final classification.

[0084] Set the sample quantity threshold for each classification (cluster) to 5. When the number of samples in a certain cluster is less than or equal to 5, use the following formula to calculate its DB index:

[0085] (5);

[0086] Where, is the number of clusters, that is, the number of clusters in the classification result; and respectively represent two different clusters; is the average radius of cluster , defined as the average distance from all points in the cluster to the centroid of the cluster; is the average radius of cluster , defined as the average distance from all points in the cluster to the centroid of the cluster; and respectively represent the centroids of cluster and cluster ; is and the distance between; means that for each cluster , find the cluster with the largest difference compared to it. is the calculated DB index. The closer its value is to 0, the higher the compactness and better the separability of the cluster. Conversely, it means that the compactness of the cluster is low and the separability is poor.

[0087] In this application, the compactness and separability of the cluster are evaluated using the DB index. When is less than or equal to 0.3, it is considered that the compactness is high and the separability is good. When it is greater than 0.3, it is considered that the compactness is low and the separability is poor. When the number of samples in a certain cluster is less than or equal to 5 and When it is less than or equal to 0.3, it indicates that the cluster is too small, with low compactness and poor separability, and is considered data noise, and the cluster needs to be deleted. After deleting all clusters that do not meet the requirements, the optimized clusters are numbered as the final classification. For example, after deleting all clusters that do not meet the requirements, there are still ten optimized clusters left, and these ten clusters are numbered as C1 to C10 respectively.

[0088] Step 4: Construct a data set based on the image data and the corresponding classification numbers, and train the deep learning model. After training is completed, it is used as a geological parameter classification model.

[0089] Take the image data converted from the geological parameter data as the input variable, and the corresponding classification number of the geological parameter data as the output variable to construct a data set, and store the data set in the database.

[0090] In an exemplary embodiment, the MobileNet V2 convolutional neural network model is selected as the deep learning model for training. The MobileNet V2 convolutional neural network model uses an inverted residual structure in its structure and a linear layer in the last layer, and has the advantages of higher accuracy and smaller model size compared to other deep learning models.

[0091] Randomly divide the entire data set into a training set and a validation set in a ratio of 7:3, train and validate the deep learning model, and use it as a geological parameter classification model after training. The input of the geological parameter classification model is image data, and the output is the corresponding classification number.

[0092] Step 5: Process the production dynamic data using the Gramian angular field to generate a Gramian angular field image.

[0093] Process the production dynamic data using the Gramian angular field, and then convert the production dynamic data into a Gramian angular field image or a Gramian angular difference field image.

[0094] Step 5 specifically includes the following steps 5.1 to 5.4.

[0095] Step 5.1: Represent the production dynamic data in the form of a tuple as ; where each pair of values refers to the production data at time is ; , is the number of sampling points. Then create a time correlation matrix for each pair of values , specifically, make the time correspond one-to-one with the production data , and then represent it in the form of a matrix.

[0096] Step 5.2: Scale the production data to to obtain the scaled production data , and the calculation formula is as follows:

[0097] (6);

[0098] where represents the set of all production data in the time correlation matrix ; and represent the maximum and minimum values in the set respectively. represents the production data corresponding to time , represents the result of scaling to .

[0099] Step 5.3: Represent the scaled production data in the polar coordinate system to obtain the polar angle and the polar radius in the polar coordinate system, and the formula is as follows:

[0100] (7);

[0101] where is the set of all ; is a constant factor for the span of the regularized polar coordinate system.

[0102] Step 5.4: Generate a Gram angular field image and a Gram angular difference field image based on the polar angle and the polar radius .

[0103] Based on the polar angle and the polar radius , a Gram angular field image and a Gram angular difference field image can be drawn, and the values of the corresponding Gram angular field and Gram angular difference field can be calculated according to the following formula:

[0104] (8);

[0105] where and represent adjacent polar angles; represents traversing each value in the set ; and represent the calculated values of the Gram angular field and the Gram angular difference field respectively.

[0106] Among them The calculation method is as follows:

[0107] (9).

[0108] Both the generated Gram angular field image and Gram angular difference field image are RGB color images. In this application, the Gram angular field image is taken as an example for detailed description. Of course, in practical applications, it is also possible to choose to convert the production dynamic data into a Gram angular difference field image for subsequent processing, and the processing method is the same as that of the Gram angular field image.

[0109] Step 6: Correspondingly store the Gram angular field image and the image data converted from the corresponding geological parameter data to build a database.

[0110] Traverse all the geological parameter data already stored in the database, and convert the production dynamic data corresponding to each group of geological parameter data into a Gram angular field image or a Gram angular difference field image according to the processing method in Step 5. After the conversion is completed, store the geological parameter data, image data, production dynamic data, and the Gram angular field image or Gram angular difference field image in one-to-one correspondence.

[0111] Step 7: Convert the geological parameter data to be predicted into image data to be predicted, and input it into the geological parameter classification model for classification to obtain a predicted classification number.

[0112] The geological parameter data to be predicted refers to the geological parameter data that needs to be predicted for a certain reservoir of the current oil and gas reservoir. Convert the geological parameter data to be predicted into the corresponding image data according to the processing method in Step 2, which is called the image data to be predicted. Input the image data to be predicted into the geological parameter classification model trained in Step 4 for classification to obtain the classification number of the classification to which the geological parameter data to be predicted belongs, which is called the predicted classification number.

[0113] Step 8: Based on the predicted classification number, retrieve the image data with the highest structural similarity under the same classification number from the database as the target image data.

[0114] Compare the image data to be predicted with all the image data under the predicted classification number in the database, and find the image data with the highest structural similarity as the target image data. The structural similarity calculation method is as follows:

[0115] (10);

[0116] Among them,

[0117] (11);

[0118] (12);

[0119] (13).

[0120] In the formula, and They represent the corresponding pixel points in the two image data to be compared, is the structural similarity of the two contrasting image data. , , Represent the comparison functions of brightness, contrast and structure respectively. , and It is a parameter used to adjust the importance of each part and is usually set to 1. and They are and The average brightness. and They are and The standard deviation of , indicating contrast. yes and , indicating the structural similarity between them. and They are all constants, used to avoid the denominator being zero, and are usually taken as . is the dynamic range of pixel values. is a constant, taking the value hour is 0.01, the value hour is 0.03. is also a constant, usually taken as .

[0121] Combining formulas (10) to (13), we can obtain:

[0122] (14).

[0123] Step 9: Use the structural similarity method to calculate the difference image between the image data to be predicted and the target image data.

[0124] Using the structural similarity method, by comparing all corresponding pixels in the image data to be predicted and the target image data, the structural similarity data between the image data to be predicted and the target image data is calculated using formula (14). Figure 2In an exemplary embodiment shown, the structural similarity between the pixel at the 4th row and 5th column in the image data to be predicted and the target image data is 0.223. The structural similarity data calculated for all pixels is mapped to the range [0 - 255] according to a processing method similar to that in step 2, represented as the gray value of a grayscale image, and then converted into corresponding image data, thereby constructing a difference image.

[0125] Step 10: Use the Gram angular field image corresponding to the target image data in the database as the image to be corrected, and correct the image to be corrected based on the difference image to obtain the corrected image.

[0126] Search the database to find the Gram angular field image of the production dynamic data corresponding to the target image data as the image to be corrected. Use the difference image data as the correction variable to correct the image to be corrected. The difference image contains the feature differences between the image data to be predicted and the target image data. Through the feature differences, the color of the corresponding pixel blocks in the image to be corrected is adjusted.

[0127] The specific steps of step 10 include the following steps 10.1 and 10.2.

[0128] Step 10.1: Divide the image to be corrected into pixel blocks.

[0129] Divide the image to be corrected according to the number of pixel blocks in the difference image so that the image to be corrected is also divided into pixel blocks. In an exemplary embodiment, , as Figure 3 shown, divide the image to be corrected into pixel blocks.

[0130] Step 10.2: Based on the gray value of the corresponding pixel in the difference image with a size of , adjust the RGB channel color values of the corresponding pixel blocks in pixel blocks to obtain the corrected image.

[0131] Denote the gray value of the pixel at the th row and th column in the difference image as . Among the pixel blocks obtained by dividing the image to be corrected, each pixel block further includes pixels. Then for a certain pixel block, the color adjustment is performed in the top - down and left - right directions, and the adjustment method is described as follows.

[0132] For example, for the pixel block at the th row and th column in the image to be corrected, for the For a pixel, set the adjustment value of the pixel at the first row and the first column to , set the adjustment value of the pixel at the th row and the th column to 0, and perform interpolation processing on the remaining pixels in the order from top to bottom and from left to right to obtain the adjustment values at different pixel positions, denoted as . Then, for a certain pixel, adjust its color value according to the following formula:

[0133] (15)

[0134] where , , are the color values of the red, green, and blue channels of a certain pixel in the image to be corrected, respectively; is the adjustment value of each channel at the position of this pixel. , , are the color values of the red, green, and blue channels after adjustment of this pixel, that is, the color values of the red, green, and blue channels of the corrected image at this pixel.

[0135] Perform color adjustment on each pixel block in the image to be corrected until all pixel blocks are corrected to obtain the corrected image.

[0136] Step 11: Inversely convert the corrected image into predicted production dynamic data.

[0137] Inverse convert the corrected image, that is, the corrected Gram angular field image or Gram angular difference field image, into the time series data of predicted production data through the reverse operation in Step 5, that is, obtain the production dynamic data predicted based on the geological parameter data to be predicted.

[0138] Specifically, first convert the corrected image into RGB color values, then perform normalization processing on it, obtain the polar angle through the inverse operation of trigonometric functions, obtain the polar radius through the size relationship between the corresponding time step and the production data, and then convert the polar coordinate data into rectangular coordinate data to obtain the tuple data of the production data and the corresponding moment. It should be noted that during the inverse conversion process, is the production data of the corresponding pixel block, and are determined by the pixel blocks with the darkest and lightest colors in the Gram angular field image or Gram angular difference field image.

[0139] The above-mentioned dynamic prediction method for the production of fractured oil and gas wells proposed in this application can greatly reduce the calculation time required for production prediction. Compared with the numerical simulation method, the calculation efficiency can be increased by more than two orders of magnitude, and all the required calculation methods are free, without the need to pay expensive software usage fees, which greatly reduces the prediction cost.

[0140] In an exemplary embodiment, this application further provides a computer device, which may be a server or a terminal. The computer device includes a processor, a memory, an input / output interface, and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with an external terminal through a network connection. The computer program, when executed by the processor, implements the dynamic prediction method for the production of fractured oil and gas wells.

[0141] In an exemplary embodiment, this application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the dynamic prediction method for the production of fractured oil and gas wells.

[0142] In an exemplary embodiment, this application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the dynamic prediction method for the production of fractured oil and gas wells.

[0143] Those of ordinary skill in the art can understand that all or part of the processes in the above-described method embodiments can be completed by hardware related to computer program instructions. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the method embodiments as described above. Among them, any reference to a memory or other medium provided in the embodiments of the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tapes, floppy disks, flash memories, optical memories, high-density embedded non-volatile memories, resistive random access memories (ReRAM), magnetoresistive random access memories (MRAM), ferroelectric random access memories (FRAM), phase change memories (PCM), graphene memories, etc. Volatile memories can include random access memory (RAM) or external cache memories, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0144] It should be noted that the information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of the relevant data need to comply with the relevant regulations.

[0145] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity in description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered to be within the scope described in this specification.

[0146] Specific examples are used in this article to elaborate on the principles and implementation manners of the present application. The descriptions of the above embodiments are only used to help understand the method and its core idea of the present application; at the same time, for those of ordinary skill in the art, according to the idea of the present application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for dynamically predicting the production of a fractured oil and gas well, characterized in that Including: Obtaining multiple groups of geological parameter data of different types of reservoirs in different oil and gas reservoirs and corresponding production dynamic data; The production dynamic data is the production data at different times; Converting each group of geological parameter data into corresponding image data; Using an unsupervised clustering method to classify the image data, optimizing the classification results, and numbering the optimized clusters as the final classification; Constructing a data set based on the image data and the corresponding classification numbers, training a deep learning model, and using it as a geological parameter classification model after training; the input of the geological parameter classification model is the image data, and the output is the corresponding classification number; Processing the production dynamic data using the Gram angular field to generate a Gram angular field image; Correspondingly storing the Gram angular field image and the image data converted from the corresponding geological parameter data to construct a database; Converting the geological parameter data to be predicted into image data to be predicted, inputting it into the geological parameter classification model for classification, and obtaining a predicted classification number; Based on the predicted classification number, retrieving the image data with the highest structural similarity under the same classification number from the database as the target image data; Calculating the difference image between the image data to be predicted and the target image data using the structural similarity method; Taking the Gram angular field image corresponding to the target image data in the database as the image to be corrected, and correcting the image to be corrected based on the difference image to obtain a corrected image; Inversely converting the corrected image into the predicted production dynamic data.

2. The method for dynamically predicting the production of a fractured oil and gas well according to claim 1, wherein The obtaining of multiple groups of geological parameter data of different types of reservoirs in different oil and gas reservoirs specifically includes: If the reservoir type is an oil reservoir, each group of geological parameter data obtained includes thickness, porosity, permeability, formation pressure, oil-gas-water saturation, oil-gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, and relative permeability curve data; the PVT data includes pressure, volume, and temperature data; If the reservoir type is a tight gas reservoir, each group of geological parameter data obtained includes thickness, porosity, permeability, formation pressure, gas-water saturation, gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, and relative permeability curve data; If the reservoir type is a shale gas reservoir or a coalbed methane reservoir, each group of geological parameter data obtained includes thickness, porosity, permeability, formation pressure, gas-water saturation, gas-water PVT data, well spacing, fracture penetration ratio, fracture equivalent permeability, relative permeability curve, Langmuir volume, and Langmuir pressure data.

3. The method for dynamically predicting the production of a fractured oil and gas well according to claim 2, wherein The converting of each group of geological parameter data into corresponding image data specifically includes: According to the number of types of geological parameter data included in each group of geological parameter data Determine the width of the image data to be transformed and the height ; where and , is a fixed padding to make an integer; Normalizing each geological parameter data included in each group of geological parameter data, and normalizing the value range to between [0-1]; Mapping the data after normalization processing to the range of [0-255] and representing it as the gray value of a grayscale image; Arrange the mapped grayscale values from top to bottom and from left to right, and reconstruct them into row a two-dimensional array of columns. Each position in the two-dimensional array represents a pixel position, and the grayscale value at each position represents the grayscale value of the pixel. Fill zeros in the positions without grayscale values to form a grayscale image as the corresponding image data.

4. The fracturing oil and gas well production dynamic prediction method according to claim 3, characterized in that The using of an unsupervised clustering method to classify the image data, optimizing the classification results, and numbering the optimized clusters as the final classification specifically includes: Calculating the similarity between every two pieces of image data using a Gaussian kernel function to construct a similarity matrix; Construct a similarity graph corresponding to the similarity matrix using the k-nearest neighbor graph, and construct a Laplacian matrix based on the similarity graph; Perform eigen-decomposition on the Laplacian matrix and calculate the first eigenvectors of the Laplacian matrix. Use these eigenvectors as columns to form a matrix U. Each row in matrix U is a -dimensional row vector, and there are a total of row vectors; For these row vectors, the K-means clustering method is used for clustering to obtain the distribution of clusters, and each cluster represents a classification; Optimize the clusters included in the classification results based on the DB index, and number the optimized clusters as the final classification.

5. The fracturing oil and gas well production dynamic prediction method according to claim 1, characterized in that, The processing of the production dynamic data using the Gram angle field to generate a Gram angle field image specifically includes: The production dynamic data is represented in the form of a tuple as ; where each pair of values refers to the time and the production data at that time is ; , being the number of sampling points; Scale the production data to to obtain the scaled production data ; Represent the scaled production data in the polar coordinate system , obtaining the polar angle and the polar radius ; Based on the polar angle and the polar radius generate a Gram angle field image.

6. The dynamic prediction method for the production of a fractured oil and gas well according to claim 1, wherein The calculation of the difference image between the image data to be predicted and the target image data using the structural similarity method specifically includes: Calculate the structural similarity data between the image data to be predicted and the target image data using the structural similarity method; Convert the structural similarity data into corresponding image data as the difference image.

7. The method for dynamically predicting the production of a fractured oil and gas well according to claim 3, wherein The correction of the image to be corrected based on the difference image to obtain the corrected image specifically includes: Divide the image to be corrected into pixel blocks; Based on the grayscale value of the corresponding pixel in the difference image based on size, adjust the RGB channel color values of the corresponding pixel blocks in the pixel blocks to obtain the corrected image.

8. A computer device, comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the hydraulic fracturing oil and gas well production dynamic prediction method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the hydraulic fracturing oil and gas well production dynamic prediction method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the hydraulic fracturing oil and gas well production dynamic prediction method according to any one of claims 1 to 7.

Images