Fracture form prediction method

Abstract

The invention provides a fracture form prediction method, and the method comprises the steps: constructing a database based on geological parameters, engineering parameters and fracture form parameters corresponding to at least one oil well in a first preset time period; wherein any oil well comprises at least one fracture formed based on a volume fracturing technique. And acquiring logging curve parameters corresponding to at least one oil well. And based on the logging curve parameter corresponding to any oil well, carrying out expansion operation on the geological parameter corresponding to any oil well in the first preset time period so as to update the database. And training a preset model based on the updated database to obtain a target model. And acquiring geological parameters and engineering parameters corresponding to the to-be-predicted oil well in the second preset time period. And inputting the geological parameters and the engineering parameters corresponding to the to-be-predicted oil well in the second preset time period into the target model, and determining the output of the target model as the fracture form parameters corresponding to the to-be-predicted oil well in the second preset time period, so as to improve the accuracy of predicting the fracture form of the to-be-predicted oil well.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unconventional oil and gas resource exploitation, and particularly relates to a fracture morphology prediction method. BACKGROUND

[0002] Unconventional oil and gas resources have huge recoverable reserves, and are usually located in complex underground environments with low porosity, low permeability, strong heterogeneity, etc. Volume fracturing technology is usually used to exploit unconventional oil and gas resources. The core of volume fracturing technology is to form a suitable fracture morphology. The fracture morphology not only affects the connectivity of the reservoir, but also affects the fluid mobility in the reservoir. Therefore, it is of great significance to quickly and accurately predict the fracture morphology for exploiting unconventional oil and gas resources.

[0003] At present, the fracture morphology to be fractured can be predicted by existing fracturing software based on the fitting of geological engineering parameters. However, this process requires a lot of time, and the prediction results vary from person to person, with low accuracy. Therefore, how to improve the accuracy of fracture morphology prediction is a problem to be solved. SUMMARY

[0004] The technical problem to be solved by the present application is to provide a fracture morphology prediction method, which can improve the accuracy of fracture morphology prediction.

[0005] The technical solution of the present application to solve the above technical problem is as follows:

[0006] In a first aspect, the present application provides a fracture morphology prediction method. In the method, a database is constructed based on at least one oil well corresponding to geological parameters, engineering parameters and fracture morphology parameters in a first preset time period. For any one of the at least one oil well, any one of the oil wells includes at least one fracture formed based on volume fracturing technology. The geological parameters are associated with the geological characteristics of the corresponding fracture position in the first preset time period. The engineering parameters are associated with the volume fracturing technology for forming the corresponding fracture in the first preset time period. The fracture morphology parameters are associated with the size of the corresponding fracture in the first preset time period. Then, the logging curve parameters corresponding to the at least one oil well are obtained. The logging curve parameters corresponding to any one of the oil wells include the geological parameters, engineering parameters and fracture morphology parameters corresponding to any one of the oil wells collected based on the whole process of volume fracturing of any one of the oil wells by volume fracturing technology. For any one of the oil wells, the geological parameters corresponding to any one of the oil wells in the first preset time period are expanded based on the logging curve parameters corresponding to any one of the oil wells, so as to update the database. A preset model is trained based on the updated database to obtain a target model. The geological parameters and engineering parameters corresponding to a to-be-predicted oil well in a second preset time period are obtained. The geological parameters and engineering parameters corresponding to the to-be-predicted oil well in the second preset time period are input into the target model, and the output of the target model is determined as the fracture morphology parameters corresponding to the to-be-predicted oil well in the second preset time period.

[0007] Based on the above technical solution, the present invention can be further improved as follows.

[0008] Furthermore, the geological parameters include at least one of the following: gamma parameter, resistivity parameter, permeability parameter, and porosity parameter at the corresponding fracture location within the corresponding time period. The engineering parameters include at least one of the following: the fracturing pressure parameter, the injection volume parameter, and the drainage volume parameter of the corresponding fracture within the corresponding time period. The fracture morphology parameters include the length, width, and height parameters of the corresponding fracture within the corresponding time period.

[0009] Furthermore, based on the location of any fracture in at least one fracture corresponding to any oil well within a first preset time period, the geological parameters corresponding to any fracture within the first preset time period are obtained from the logging curve parameters corresponding to any oil well. Based on the geological parameters corresponding to each fracture included in any oil well within the first preset time period, the geological parameters corresponding to any oil well within the first preset time period are expanded to update the database.

[0010] Furthermore, based on linear interpolation and the geological parameters corresponding to each fracture in any oil well within the first preset time period, the geological parameters corresponding to any oil well within the first preset time period are expanded to update the database.

[0011] Furthermore, the updated database, including geological and engineering parameters, undergoes standardization processing to obtain standardized geological and engineering parameters. Standardization is used to unify the dimensions of all geological and engineering parameters included in the updated database. Based on the standardized geological and engineering parameters, and the fracture morphology parameters included in the updated database, a pre-defined model is trained to obtain the target model.

[0012] Furthermore, the standardized geological and engineering parameters corresponding to any oil well in the updated database are input into the target model to obtain the predicted fracture morphology parameters for any oil well output by the target model. Based on the difference between the predicted fracture morphology parameters for any oil well and the fracture morphology parameters for any oil well in the database, the model parameters of the preset model are adjusted. The target model is obtained based on the preset convergence condition achieved through training the preset model.

[0013] Furthermore, the sampling length for each well in at least one oil well is determined. The sampling length includes at least one sampling point. Based on the sampling length, a first time series is established. The first time series includes at least one time point, each time point corresponding to one of the at least one sampling point. Different labels are assigned to different geological parameters. For any geological parameter to be standardized, based on the sampling time of that geological parameter, the label corresponding to that geological parameter is filled into the corresponding time point position in the first time series. The first time series filled with the labels corresponding to each geological parameter to be standardized is determined as the standardized geological parameter.

[0014] Furthermore, a second time series is established based on the sampling length. The second time series includes at least one time point, and each time point corresponds to one of the at least one sampling points. For any engineering parameter to be standardized, the parameter is filled into each time point position included in the second time series. The second time series filled with the engineering parameters to be standardized is determined as the standardized engineering parameter.

[0015] Furthermore, the pre-defined model includes a Long Short-Term Memory (LSTM) network layer, a fully connected layer, and an output layer. The LSM layer receives standardized geological and engineering parameters, encodes the received geological parameters according to a pre-defined encoding rule, and inputs the encoded geological parameters and the received engineering parameters into the fully connected layer. The fully connected layer, based on the input engineering parameters and the encoded geological parameters, obtains fracture morphology parameters corresponding to the input engineering parameters and the encoded geological parameters. The output layer outputs the fracture morphology parameters obtained by the fully connected layer.

[0016] In a second aspect, the present invention provides an electronic device, comprising: a memory and one or more processors; the memory and the processors are coupled; wherein the memory stores computer program code, the computer program code including computer instructions, and when the computer instructions are executed by the processor, the electronic device performs a crack morphology prediction method as described in any of the first aspects above.

[0017] Thirdly, the present invention provides a computer-readable storage medium including computer instructions that, when executed on an electronic device, cause the electronic device to perform a crack morphology prediction method as described in any of the first aspects above.

[0018] Fourthly, the present invention provides a computer program product that, when run on a computer, causes the computer to execute a crack morphology prediction method as described in any of the first aspects above.

[0019] The beneficial effects of this invention are as follows: This invention matches discrete fracturing segment data with continuous logging data using linear interpolation, expanding the data sample size and making the establishment of the preset model more accurate. Through training the preset model, the resulting target model can quickly predict fracture morphology, and based on the uniqueness of the input, its output is unique, achieving a quantitative characterization of fracture morphology. Furthermore, this invention can improve the accuracy of fracture morphology prediction, which is beneficial for optimizing fracturing schemes and has significant implications for unconventional oil and gas field development. Attached Figure Description

[0020] Figure 1 A schematic flowchart of a crack morphology prediction method provided by the present invention;

[0021] Figure 2 This is a schematic diagram of a pre-defined model provided by the present invention. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. In the description of this application, unless otherwise stated, " / " indicates that the objects before and after are in an "or" relationship. For example, A / B can represent A or B. "And / or" in this application is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone, where A and B can be singular or plural. Furthermore, in the description of this application, unless otherwise stated, "multiple" refers to two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple. Furthermore, to facilitate a clear description of the technical solutions in the embodiments of this application, the terms "first" and "second" are used in the embodiments of this application to distinguish identical or similar items with substantially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and that "first" and "second" are not necessarily different. Meanwhile, in the embodiments of this application, words such as "exemplary" or "for example" are used to indicate that something is being used as an example, illustration, or description. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes.

[0023] Currently, existing fracturing software can be used to predict fracture morphology by fitting geological engineering parameters. However, this process is time-consuming, and the prediction results vary from person to person, are not unique, and have low accuracy. Furthermore, accurately matching the data of fractured sections or clusters with continuous logging curve data points is also a problem that needs to be solved.

[0024] To address the aforementioned problems, this invention provides a fracture morphology prediction method. First, linear interpolation is used to expand the data sample, and then an LSTM algorithm is employed to predict fracture morphology parameters. This invention plays a significant role in improving horizontal well oil and gas production capacity, enhancing reservoir development efficiency, and reducing engineering operation costs.

[0025] See Figure 1 The following is a flowchart illustrating a crack morphology prediction method provided by the present invention, including the following steps S101-S106:

[0026] S101: Construct a database based on the geological parameters, engineering parameters, and fracture morphology parameters of at least one oil well in the first preset time period.

[0027] Specifically, for any one of the at least one oil wells, each oil well includes at least one fracture formed using volumetric fracturing technology. The geological parameters, engineering parameters, and fracture morphology parameters corresponding to any oil well obtained in S101 can be predicted values ​​calculated based on the characteristics corresponding to that oil well, or they can be approximate values ​​obtained based on measurements from detection equipment.

[0028] In some embodiments, geological parameters are associated with the geological characteristics of the corresponding fracture location during a first preset time period. The geological parameters include at least one of the following: gamma parameter, resistivity parameter, permeability parameter, and porosity parameter for the corresponding fracture location during the corresponding time period.

[0029] In some embodiments, the engineering parameters are associated with the volumetric fracturing technique that forms the corresponding fracture in a first preset time period. The engineering parameters include at least one of the following: the fracturing pressure parameter for forming the corresponding fracture in the corresponding time period, the injection volume parameter, and the discharge volume parameter of the corresponding fracture in the corresponding time period.

[0030] In some embodiments, crack morphology parameters are associated with the size of the corresponding crack in a first preset time period. Crack morphology parameters include the length, width, and height of the corresponding crack in the corresponding time period.

[0031] S102: Obtain the logging curve parameters corresponding to at least one oil well.

[0032] The logging parameters for any given oil well include geological parameters, engineering parameters, and fracture morphology parameters collected throughout the entire process of volumetric fracturing of that oil well using volumetric fracturing technology. Alternatively, the aforementioned "entire process" can be understood as a third preset time period, which includes the first preset time period.

[0033] S103: For any oil well, based on the logging curve parameters corresponding to any oil well, perform an expansion operation on the geological parameters corresponding to any oil well in the first preset time period to update the database.

[0034] In some embodiments, the geological parameters corresponding to any fracture in at least one fracture corresponding to any oil well during the first preset time period can be obtained from the logging curve parameters corresponding to any oil well during the first preset time period based on the location of any fracture during the first preset time period. Based on the geological parameters corresponding to each fracture included in any oil well during the first preset time period, the geological parameters corresponding to any oil well during the first preset time period can be expanded to update the database.

[0035] In some embodiments, the geological parameters of any oil well corresponding to the first preset time period can be expanded based on linear interpolation and the geological parameters of each fracture included in any oil well corresponding to the first preset time period, so as to update the database.

[0036] For example, because the number of fracturing sections in an oil well is very limited, machine learning requires sufficient data samples, thus necessitating the expansion of the data samples in the database. Linear interpolation can be used to expand the geological and engineering parameters of each oil well's fracturing section. First, the geological parameters of the fracturing section can be expanded using well logging curve parameters. Then, linear interpolation can be used to convert the corresponding section data into fixed values ​​(e.g., 20 rows). Simultaneously, the engineering parameters of the corresponding section can be assigned to the depth of each sampling point to maintain consistency in the feature dimensions of the data.

[0037] The principle of linear interpolation is to assume a linear relationship between two known data points to estimate the value of unknown data points. Due to its simplicity, it is suitable for handling situations with few missing values ​​and relatively stable data changes. Therefore, when interpolation is very dense, the geological parameter characteristics exhibit similar and stable changes.

[0038] Suppose we have two known data points (X1, Y1) and (X2, Y2), where Y is a function of X. We want to find the value of an unknown point (X, Y) located between these two points. Linear interpolation assumes that the relationship between Y and X is linear on the interval [X1, X2], i.e., it satisfies:

[0039]

[0040] Here, Y can be considered a geological parameter corresponding to a certain depth, and X is the corresponding depth. That is, X1 and X2 can represent different depths; Y1 and Y2 can represent different geological parameters. Furthermore, X1 corresponds to Y1, and X2 corresponds to Y2.

[0041] S104: Train the preset model based on the updated database to obtain the target model.

[0042] In some embodiments, the updated database, including geological parameters and engineering parameters, undergoes standardization processing to obtain standardized geological and engineering parameters. Standardization processing is used to unify the dimensions of the various geological parameters and engineering parameters included in the updated database. Based on the standardized geological and engineering parameters included in the updated database, and based on the fracture morphology parameters included in the updated database, a preset model is trained to obtain the target model.

[0043] In some embodiments, see Figure 2 The preset model includes a long short-term memory network layer (also referred to as an LSTM layer in this embodiment), a fully connected layer, and an output layer.

[0044] The LSTM layer is a special type of recurrent neural network (RNN) designed to solve the vanishing gradient problem of traditional RNNs. LSTM layers control the flow of information by introducing cell states and three gates (input gate, forget gate, and output gate). This allows LSTM layers to effectively remember long-term dependencies and important information, making them suitable for tasks such as time series prediction and natural language processing.

[0045] The LSTM layer receives standardized geological and engineering parameters, encodes the received geological parameters according to a preset encoding rule, and inputs the encoded geological parameters and the received engineering parameters into the fully connected layer. The fully connected layer, based on the input engineering parameters and the encoded geological parameters, obtains the corresponding fracture morphology parameters. The output layer outputs the fracture morphology parameters obtained by the fully connected layer.

[0046] In some embodiments, standardized geological and engineering parameters corresponding to any oil well in the updated database are input into the target model to obtain the predicted fracture morphology parameters corresponding to any oil well output by the target model. Based on the difference between the predicted fracture morphology parameters corresponding to any oil well and the fracture morphology parameters corresponding to any oil well in the database, the model parameters of the preset model are adjusted. Based on the preset model reaching the preset convergence condition, the target model is obtained.

[0047] For example, the data in the database can be divided into a training set and a test set. For instance, 80% of the sample data in the database can be designated as the training set, and 20% as the test set. Then, an LSTM layer or multiple LSTM layers can be constructed, combining the geological parameters and engineering data from the training set as input to the LSTM layer. The geological parameters are encoded after passing through the LSTM layer, representing the dynamic changes in the well logging data sequence. Before passing the data from the LSTM layer to the fully connected layer, the Dropout algorithm can be used to prevent overfitting of the neural network, enhancing the model's generalization ability. Afterward, the output of the LSTM layer (encoded geological parameters and engineering parameters) can be passed to the fully connected layer, thus mapping the LSTM layer output to the corresponding fracture attributes (fracture morphology parameters). The output format of the LSTM layer can be (batch_size, time_steps, lstm_hidden_size), and the output layer typically takes the output of the last time step (lstm_hidden_size), obtaining the final predicted value (fracture morphology parameters) through the fully connected layer. Then, mean squared error can be used as the regression loss function, and Adam can be selected as the optimizer. Based on the difference between the predicted and true values ​​of the output layer, the regression loss of crack morphology parameters is minimized. That is, the regression loss of crack length, width, and height is minimized to improve the accuracy of the target model.

[0048] In some embodiments, a sampling length is determined for each well in at least one oil well. The sampling length includes at least one sampling point. Based on the sampling length, a first time series is established. The first time series includes at least one time point, each time point corresponding to one of the at least one sampling point. Different labels are assigned to different geological parameters. For any geological parameter to be standardized, based on the sampling time of any geological parameter to be standardized, the label corresponding to that geological parameter is filled into the corresponding time point position in the first time series. The first time series filled with the labels corresponding to each geological parameter to be standardized is determined as the standardized geological parameters.

[0049] In some embodiments, a second time series is established based on the sampling length. The second time series includes at least one time point, and each time point corresponds to one of the at least one sampling points. For any engineering parameter to be standardized, the engineering parameter to be standardized is filled into each time point position included in the second time series. The second time series filled with each engineering parameter to be standardized is determined as the standardized engineering parameter.

[0050] For example, logging curve parameters have different sampling lengths in different dimensions. The maximum length of the fracturing segment in the training data is calculated, and then multiplied by the sampling interval to obtain the maximum time series. For instance, if the maximum sampling length is 50m, and 8 points are sampled per 1m (i.e., one sampling point per 0.125m), then the total number of sampling points included in the sampling length is 400. Engineering data (fluid volume, segment length, sand volume, sand ratio, etc.) have only one data point per segment.

[0051] Because different fracturing segments have different lengths, but the input data dimension needs to be kept uniform during training, the time-series data (geological parameters) and non-time-series data (engineering parameters) will be standardized separately. For geological parameters, a special labeling method can be used to fill the corresponding time series with special labels (e.g., -9999 or NaN) to mark the filling location. For engineering parameters, which can be used as static feature input, a repeated filling method is used to repeat the engineering data of each sample into the corresponding time series.

[0052] S105: Obtain the geological and engineering parameters of the oil well to be predicted in the second preset time period.

[0053] The second time period may be the same as or different from the first time period mentioned above. The third time period may or may not include the second time period.

[0054] S106: Input the geological and engineering parameters of the oil well to be predicted into the target model in the second preset time period, and determine the output of the target model as the fracture morphology parameters of the oil well to be predicted in the second preset time period.

[0055] In other words, given the geological and engineering parameters of any oil well at any given time period (past or future), this invention can predict the fracture morphology parameters of that oil well within that time period. It effectively solves the problem of predicting fracture morphology in unconventional oil and gas resources. This invention uncovers the true meaning behind the data, which is beneficial for optimizing the design of on-site fracturing schemes, rapidly and effectively improving oil and gas recovery rates, and contributing to increased reserves and reduced efficiency in oilfields.

[0056] In some solutions, multiple embodiments of this application can be combined, and the combined solution can be implemented. Optionally, some operations in the processes of each method embodiment may be combined, and / or the order of some operations may be changed. Furthermore, the execution order between the steps of each process is merely exemplary and does not constitute a limitation on the execution order between steps; other execution orders are also possible. It is not intended to indicate that the execution order is the only possible order in which these operations can be performed. Those skilled in the art will conceive of various ways to reorder the operations described herein. In addition, it should be noted that the process details involved in one embodiment of this document are similarly applicable to other embodiments, or different embodiments may be combined.

[0057] Furthermore, some steps in the method embodiments can be equivalently replaced with other possible steps. Alternatively, some steps in the method embodiments may be optional and can be deleted in certain use cases. Or, other possible steps may be added to the method embodiments. Moreover, the various method embodiments can be implemented individually or in combination.

[0058] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the model can be divided into different functional modules to complete all or part of the functions described above.

[0059] In the several embodiments provided in this application, it should be understood that the disclosed models and methods can be implemented in other ways. For example, the model embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another model, or some features may be ignored or not executed.

[0060] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0061] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, in essence, or the part that contributes, or all or part of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0062] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for predicting crack morphology, characterized in that, include: A database is constructed based on the geological parameters, engineering parameters, and fracture morphology parameters of at least one oil well corresponding to a first preset time period; for any one of the at least one oil wells, the oil well includes at least one fracture formed by volumetric fracturing technology; the geological parameters are associated with the geological characteristics of the corresponding fracture location in the first preset time period; the engineering parameters are associated with the volumetric fracturing technology that formed the corresponding fracture in the first preset time period; and the fracture morphology parameters are associated with the size of the corresponding fracture in the first preset time period. Obtain logging curve parameters corresponding to at least one oil well; the logging curve parameters corresponding to any oil well include geological parameters, engineering parameters, and fracture morphology parameters for any oil well collected throughout the entire process of volumetric fracturing based on volumetric fracturing technology. For any oil well, based on the logging curve parameters corresponding to any oil well, the geological parameters corresponding to any oil well in the first preset time period are expanded to update the database; The target model is obtained by training a pre-defined model based on the updated database; Obtain the geological and engineering parameters of the oil well to be predicted during the second preset time period; The geological and engineering parameters of the oil well to be predicted during the second preset time period are input into the target model, and the output of the target model is determined as the fracture morphology parameters of the oil well to be predicted during the second preset time period.

2. The method according to claim 1, characterized in that, The geological parameters include at least one of the following: gamma parameter, resistivity parameter, permeability parameter, and porosity parameter at the corresponding fracture location in the corresponding time period; the engineering parameters include at least one of the following: rupture pressure parameter, injection volume parameter, and drainage volume parameter of the corresponding fracture in the corresponding time period; the fracture morphology parameters include the length parameter, width parameter, and height parameter of the corresponding fracture in the corresponding time period.

3. The method according to claim 2, characterized in that, For any given oil well, based on the logging curve parameters corresponding to that oil well, the geological parameters of that oil well during the first preset time period are expanded to update the database, including: Based on the location of any fracture in at least one fracture corresponding to any oil well during the first preset time period, the geological parameters of any fracture during the first preset time period are obtained from the logging curve parameters corresponding to any oil well. Based on the geological parameters corresponding to each fracture in any oil well during the first preset time period, the geological parameters corresponding to any oil well during the first preset time period are expanded to update the database.

4. The method according to claim 3, characterized in that, The step of expanding the geological parameters of any oil well within the first preset time period based on the geological parameters of each fracture in the oil well during the first preset time period to update the database includes: Based on linear interpolation and the geological parameters corresponding to each fracture in any oil well during the first preset time period, the geological parameters corresponding to any oil well during the first preset time period are expanded to update the database.

5. The method according to claim 4, characterized in that, After updating the database by expanding the geological parameters of any oil well within the first preset time period based on the logging curve parameters corresponding to that oil well, the method further includes: The updated database includes geological parameters and engineering parameters, which are then standardized to obtain standardized geological parameters and engineering parameters. The standardization process is used to unify the dimensions of the geological parameters and engineering parameters included in the updated database. The process of training a preset model based on the updated database to obtain the target model includes: Based on the standardized geological and engineering parameters included in the updated database, and the fracture morphology parameters included in the updated database, the preset model is trained to obtain the target model.

6. The method according to claim 5, characterized in that, The process of training the preset model based on the standardized geological and engineering parameters included in the updated database, and the fracture morphology parameters included in the updated database, to obtain the target model, includes: The standardized geological and engineering parameters corresponding to any oil well in the updated database are input into the target model to obtain the predicted fracture morphology parameters corresponding to any oil well output by the target model. Based on the difference between the predicted fracture morphology parameters corresponding to any oil well and the fracture morphology parameters corresponding to any oil well included in the database, the model parameters of the preset model are adjusted. The target model is obtained by training the preset model to achieve the preset convergence condition.

7. The method according to claim 6, characterized in that, The standardization process includes: Determine the sampling length for each of the at least one oil well; the sampling length includes at least one sampling point; Based on the sampling length, a first time series is established; the first time series includes at least one time point, and each time point corresponds to one of the at least one sampling points. Set different labels for different geological parameters; For any geological parameter to be standardized, based on the sampling time of the geological parameter to be standardized, the label corresponding to the geological parameter to be standardized is filled into the corresponding time point position in the first time series; The first time series filled with labels corresponding to each geological parameter to be standardized is determined as the standardized geological parameter.

8. The method according to claim 7, characterized in that, The standardization process also includes: Based on the sampling length, a second time series is established; the second time series includes at least one time point, and each time point corresponds to one of the at least one sampling points. For any engineering parameter to be standardized, fill the engineering parameter to be standardized into each time point position included in the second time series; The second time series, filled with the engineering parameters to be standardized, is determined as the standardized engineering parameters.

9. The method according to claim 8, characterized in that, The preset model includes a long short-term memory network layer, a fully connected layer, and an output layer; wherein... The long short-term memory network layer is used to receive standardized geological parameters and engineering parameters, encode the received geological parameters based on a preset encoding rule, and input the encoded geological parameters and the received engineering parameters into the fully connected layer. The fully connected layer is used to obtain fracture morphology parameters corresponding to the input engineering parameters and the encoded geological parameters based on the input engineering parameters and the encoded geological parameters. The output layer is used to output the crack morphology parameters obtained by the fully connected layer.

10. An electronic device, characterized in that, include: A memory, one or more processors; the memory is coupled to the processors; wherein the memory stores computer program code, the computer program code including computer instructions, which, when executed by the processor, cause the electronic device to perform the crack morphology prediction method as described in any one of claims 1-9.

Images