Method, device, equipment and storage medium for predicting production of tight sandstone oil and gas reservoir

Abstract

The embodiment of the application provides a method, device and equipment for predicting the yield of a compact sandstone oil and gas reservoir and a storage medium, the method comprising: collecting single-well data of a target oil and gas field to obtain a first data set, the single-well data comprising geological parameters, engineering parameters and yield data of a single well; determining geological main control parameters and engineering main control parameters that affect the yield data; performing normalization processing on the geological main control parameters and the engineering main control parameters in the first data set; using a k-nearest neighbor algorithm to calculate the yield data of a well with the closest attributes to each well in a training set; using an XGBoost algorithm to establish a yield prediction model by using the data in the training set; and inputting related parameters of a new well to be predicted into the yield prediction model to obtain a yield prediction result of the new well to be predicted. The embodiment of the application can improve the prediction accuracy of the model, thereby realizing accurate prediction of the yield of a new well, providing a reliable basis for planning and deployment of the yield of an oil field new well, and helping efficient development of an oil and gas reservoir.

Description

Technical Field

[0001] The present application relates to the field of oil and gas exploration technology, and specifically to a method, device, equipment and storage medium for predicting the production of tight sandstone oil and gas reservoirs. Background Art

[0002] Recoverable reserves are the material foundation for oil and gas field development and one of the primary indicators for evaluating development effectiveness. Existing methods for predicting recoverable reserves primarily include reservoir numerical simulation, typical curve methods, and machine learning. Numerical reservoir simulation, based on reservoir geology and actual development conditions, establishes a mathematical model describing the flow patterns within the reservoir and uses computers to obtain numerical solutions to investigate these patterns. Essentially, numerical reservoir simulation replicates actual reservoirs using mathematical, geological, physical, and computer methods, and its foundational theory is Darcy's flow law. The typical curve method assumes that wells within a block or field have similar production patterns, thereby predicting the production of new wells based on the production patterns of existing wells. Machine learning methods include fully connected neural networks (FCNNs), random forests (RFs), and support vector machines (SVMs). These methods automatically learn the production patterns of large amounts of dynamic and static data from individual wells, thereby generating a single-well production prediction model.

[0003] In the process of implementing this application, the applicant discovered that the prior art has at least the following problems:

[0004] (1) Production prediction methods based on numerical simulation methods have a limited range of parameters, too many theoretical assumptions, and are unable to describe the high-dimensional, nonlinear relationship between geological and engineering parameters and production, resulting in low prediction accuracy. The seepage mechanism of tight sandstone gas reservoirs is complex, and numerical simulation methods are difficult to accurately simulate the gas reservoir laws;

[0005] (2) The production prediction method based on the typical curve method cannot take into account the impact of complex geological and engineering conditions on production. It can only make a rough estimate of the production of new wells based on the production patterns of old wells in the block or oil field;

[0006] (3) Most of the current yield prediction methods based on machine learning only consider the impact of geological or engineering parameters on yield. However, in actual production, the impact of the coupling effect of geological and engineering parameters on yield must be considered. It is difficult to accurately predict yield by only considering geological or engineering factors.

[0007] (4) Current machine learning-based yield prediction methods, such as linear regression, random forest, and SVR, have a single algorithm model structure and limited fitting capabilities, resulting in poor prediction results when used alone. A composite machine learning model needs to be established to improve prediction accuracy.

[0008] (5) Today's yield prediction methods based on machine learning contain many hyperparameters. Adjusting the model hyperparameters is time-consuming and labor-intensive, and has a certain impact on the performance of the final model.

[0009] It should be pointed out that the information disclosed in the background technology section of this application is only intended to deepen the understanding of the general background technology of this application, and should not be regarded as an admission or any form of implication that the information constitutes prior art already known to those skilled in the art. Summary of the Invention

[0010] In view of this, the present application provides a method, device, equipment and storage medium for predicting the production of tight sandstone oil and gas reservoirs, so as to solve the problems existing in the prior art.

[0011] In a first aspect, an embodiment of the present application provides a method for predicting the production of tight sandstone oil and gas reservoirs, comprising:

[0012] Collecting single-well data of a target oil and gas field to obtain a first data set, wherein the single-well data includes geological parameters, engineering parameters, and production data of the single well;

[0013] Determining, from the geological parameters and the engineering parameters, geological master control parameters and engineering master control parameters that affect the production data;

[0014] Normalizing the geological control parameters and engineering control parameters in the first data set to obtain a second data set, wherein the single well data in the second data set includes normalized geological control parameters, normalized engineering control parameters, and production data;

[0015] Calculating, using a k-nearest neighbor algorithm, the production data of the well whose attribute is closest to that of each well in a training set, wherein the training set belongs to the second data set;

[0016] Using the XGBoost algorithm, a production prediction model is established using the normalized geological main control parameters, the normalized engineering main control parameters, the production data of the well with the closest attributes, and the production data in the training set;

[0017] The geological main control parameters, engineering main control parameters and production data of the well with the closest attributes of the new well to be predicted are input into the production prediction model to obtain the production prediction result of the new well to be predicted.

[0018] In a possible implementation, the method further includes:

[0019] Calculating, using a k-nearest neighbor algorithm, the production data of the well whose attributes are closest to each well in a test set, wherein the test set belongs to the second data set;

[0020] The production prediction model is tested using the normalized geological main control parameters, normalized engineering main control parameters, production data of the well with the closest attributes, and production data in the test set.

[0021] In a possible implementation, the method further includes:

[0022] Optimizing hyperparameters of the yield prediction model using a Bayesian optimization algorithm;

[0023] The Bayesian optimization algorithm is used to optimize and adjust the tree depth, learning rate, number of sub-models, solution method and / or regularization term weight of the k-nearest neighbor algorithm.

[0024] In a possible implementation, the geological parameter includes: at least one of water saturation, porosity, brittleness index, and average fracture pressure;

[0025] The engineering parameters include at least one of: fracturing operation displacement, average section length, number of sections, horizontal section length, proppant dosage per section, minimum pump stop pressure, maximum pump stop pressure, nitrogen volume pumped into each section, average operation pressure, and azimuth angle;

[0026] The production data includes: single well ultimate recoverable reserves EUR.

[0027] In a possible implementation, the single well data further includes location parameters, which include at least one of bottom hole longitude, bottom hole latitude, and an average depth of a development layer.

[0028] In a possible implementation, determining the geological master control parameters and the engineering master control parameters that affect the production data from the geological parameters and the engineering parameters respectively includes:

[0029] The XGBoost algorithm is applied to calculate the degree of influence of the geological parameters and the engineering parameters on the production data, and the geological main control parameters and the engineering main control parameters that affect the production data are determined from the geological parameters and the engineering parameters according to the degree of influence.

[0030] In a possible implementation, determining, from among the geological parameters and the engineering parameters, the geological master control parameters and the engineering master control parameters that affect the production data according to the degree of influence includes:

[0031] The geological parameters and engineering parameters ranked in the top 50% of the influence degree are regarded as the geological main control parameters and engineering main control parameters affecting the production data.

[0032] In a second aspect, an embodiment of the present application provides a tight sandstone oil and gas reservoir production prediction device, comprising:

[0033] A data collection module is used to collect single-well data of the target oil and gas field to obtain a first data set, wherein the single-well data includes geological parameters, engineering parameters and production data of the single well;

[0034] a main control parameter determination module, configured to determine, from the geological parameters and the engineering parameters, geological main control parameters and engineering main control parameters that affect the production data;

[0035] a normalization processing module, configured to perform normalization processing on the geological main control parameters and the engineering main control parameters in the first data set to obtain a second data set, wherein the single well data in the second data set includes normalized geological main control parameters, normalized engineering main control parameters, and production data;

[0036] a production data calculation module for the well with the closest attribute, configured to calculate the production data of the well with the closest attribute for each well in the training set using a k-nearest neighbor algorithm, wherein the training set belongs to the second data set;

[0037] A production prediction model building module is used to establish a production prediction model using the normalized geological main control parameters, normalized engineering main control parameters, production data of the well with the closest attributes, and production data in the training set using the XGBoost algorithm;

[0038] The new well production prediction module is used to input the geological main control parameters, engineering main control parameters and production data of the well with the closest attributes of the new well to be predicted into the production prediction model to obtain the production prediction result of the new well to be predicted.

[0039] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0040] processor;

[0041] Memory;

[0042] and a computer program, wherein the computer program is stored in the memory, and the computer program includes instructions, which, when executed by the processor, enable the electronic device to perform any one of the methods described in the first aspect.

[0043] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the methods described in the first aspect.

[0044] The technical solutions provided by the embodiments of the present application have at least the following technical effects:

[0045] (1) By adding the production data of the wells with the closest attributes to the data set, the correlation between the input parameter group and the output parameter in the production prediction process is improved, thereby improving the model prediction accuracy;

[0046] (2) By combining supervised and unsupervised learning, the relationship between tight sandstone production data and parameter groups is learned, and a nonlinear mapping relationship between production and its parameter groups is established;

[0047] (3) The Bayesian hyperparameter optimization method is used to optimize the model structure and comprehensively improve the prediction accuracy of the model, thereby achieving accurate prediction of the production of new wells, providing a reliable basis for the planning and deployment of the production of new wells in oil fields, and facilitating the efficient development of oil and gas reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0049] Figure 1 A schematic flow chart of a method for predicting the production of tight sandstone oil and gas reservoirs provided in an embodiment of the present application;

[0050] Figure 2 A technical framework diagram of a method for predicting production of tight sandstone oil and gas reservoirs provided in an embodiment of the present application;

[0051] Figure 3 A schematic diagram of a data set provided in an embodiment of the present application;

[0052] Figure 4 A schematic diagram of parameter influence ranking provided in an embodiment of the present application;

[0053] Figure 5 A single well production prediction plan provided in an embodiment of the present application;

[0054] Figure 6 This is a structural block diagram of a tight sandstone oil and gas reservoir production prediction device provided in this application;

[0055] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0056] In order to better understand the technical solution of the present application, the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0057] It should be clear that the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0058] The terms used in the embodiments of the present application are for the purpose of describing specific embodiments only and are not intended to limit the present application. The singular forms "a", "an", "the" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise.

[0059] It should be understood that the term "and / or" as used herein simply describes a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A alone, A and B together, or B alone. Furthermore, the character " / " in this document generally indicates an "or" relationship between the associated objects.

[0060] In response to the technical problems existing in the prior art, the embodiments of the present application provide a method for predicting the production of tight sandstone oil and gas reservoirs. By combining supervised and unsupervised learning, the relationship between tight sandstone production data and parameter groups is learned, and a nonlinear mapping relationship between production and its parameter groups is established. The model structure is optimized using the Bayesian hyperparameter optimization method to comprehensively improve the model's prediction accuracy, thereby achieving accurate prediction of new well production, providing a reliable basis for planning and deploying new well production in oil fields, and facilitating the efficient development of oil and gas reservoirs. A detailed description is provided below in conjunction with specific implementation methods.

[0061] See also Figure 1 , which is a flow chart of a method for predicting the production of tight sandstone oil and gas reservoirs provided in an embodiment of the present application; see Figure 2 , is a technical framework diagram of a method for predicting the production of tight sandstone oil and gas reservoirs provided in the embodiment of this application. Figure 1 Combined with Figure 2 As shown in FIG, the method for predicting the production of tight sandstone oil and gas reservoirs mainly includes the following steps.

[0062] Step S101: collecting single-well data of a target oil and gas field to obtain a first data set, where the single-well data includes geological parameters, engineering parameters, and production data of the single well.

[0063] In one possible implementation, geological parameters include: at least one of water saturation, porosity, brittleness index and average fracture pressure; engineering parameters include: fracturing operation displacement, average section length, number of sections, horizontal section length, proppant dosage per section, minimum pump-off pressure, maximum pump-off pressure, nitrogen volume pumped into each section, average operation pressure and at least one of azimuth angle; production data include: ultimate recoverable reserves EUR of a single well. In addition, in addition to the above data, single well data may also include location parameters, wherein the location parameters include at least one of bottom hole longitude, bottom hole latitude and average depth of the development layer. For example, the first data set is such as Figure 3 shown.

[0064] In one possible implementation, after obtaining the first dataset, the data in the first dataset can be divided into a training set, a validation set, and a test set according to a certain ratio. Of course, the division of the training set, validation set, and test set can also be performed in subsequent steps, and this embodiment of the present application does not impose specific limitations on this.

[0065] Step S102: determining geological master control parameters and engineering master control parameters that affect production data from among geological parameters and engineering parameters.

[0066] In one possible implementation, the XGBoost algorithm is used to calculate the degree of influence of geological and engineering parameters on yield data. Based on the degree of influence, the primary geological and engineering control parameters that influence yield data are identified. Specifically, the top 50% of the geological and engineering parameters ranked by influence can be used as the primary geological and engineering control parameters that influence yield data.

[0067] For example, the influence degree of geological parameters and engineering parameters is ranked as follows: Figure 4 As shown in the figure, the first 50% (9) of the parameters are used as the main control parameters. Specifically, the geological main control parameters include water saturation, porosity, brittleness index, and average fracture pressure; the engineering main control parameters include fracturing operation flow rate, segment length, number of segments, horizontal segment length, and proppant dosage per segment.

[0068] In one possible implementation, after determining the geological main control parameters and the engineering main control parameters, only the geological main control parameters in the geological parameters and the engineering main control parameters in the engineering parameters can be retained in the first data set, and other geological parameters and engineering parameters can be deleted to update the data in the first data set.

[0069] Step S103: normalizing the geological main control parameters and engineering main control parameters in the first data set to obtain a second data set. The single well data in the second data set includes normalized geological main control parameters, normalized engineering main control parameters and production data.

[0070] In one possible implementation, the data for each attribute in the training set can be normalized using the maximum and minimum values ​​of each attribute data in the geological and engineering control parameters. This yields the normalized geological and engineering control parameters for the training set. The validation and test sets are then subjected to the same normalization using the maximum and minimum values ​​of each attribute data in the training set to obtain the corresponding normalized geological and engineering control parameters. In a specific implementation, the normalization formula is shown in Formula 1.

[0071] Formula 1:

[0072]

[0073] Among them, X is the value of a certain attribute data, X min is the minimum value of a certain attribute data, X max is the maximum value of a certain attribute data, X norm It is the normalized value of a certain attribute data.

[0074] It is understood that after obtaining the normalized values, they are used in subsequent calculations. Therefore, the normalized geological control parameters and normalized engineering control parameters can be used to update the geological control parameters and engineering control parameters in the first dataset, respectively. For ease of distinction, the dataset containing the normalized geological control parameters, normalized engineering control parameters, and production data is referred to as the second dataset.

[0075] Step S104: Using the k-nearest neighbor algorithm, calculate the production data of the well whose attributes are closest to each well in the training set.

[0076] That is, for each well in the training set, a new attribute is added: the production data of the well whose attribute is closest to that of the well. For example, if the attribute of well A in the training set is closest to that of well B, then the production data of well B will be the production data of the well whose attribute is closest to that of well A.

[0077] In practice, the k-nearest neighbor (KNN) algorithm can be used, with the Euclidean distance function used as the method for calculating the proximity of attributes between wells. Since the KNN algorithm averages the production data of the K wells whose attributes are closest to the well being tested, setting K to 1 calculates the production data of the well whose attributes are closest to the well being tested.

[0078] Step S105: Using the XGBoost algorithm, the normalized geological control parameters, normalized engineering control parameters, production data of the well with the closest attributes, and production data in the training set are used to establish a production prediction model.

[0079] In one possible implementation, the k-nearest neighbor algorithm can also be used to calculate the production data of the well with the attributes closest to each well in the test set; then, the production prediction model is tested using the normalized geological main control parameters, normalized engineering main control parameters, the production data of the well with the attributes closest to it, and the production data in the test set. Based on the test results, it is determined whether the production prediction model needs to be further trained to achieve the ideal prediction effect.

[0080] Step S106: Input the geological main control parameters, engineering main control parameters, and production data of the well with the closest attributes of the new well to be predicted into the production prediction model to obtain the production prediction result of the new well to be predicted.

[0081] In a specific implementation, the geological and engineering control parameters of the new well to be predicted can be normalized using the above-mentioned normalization method. Furthermore, since only old wells have production data, the well with the properties closest to the new well to be predicted should be an old well.

[0082] It can be understood that after the geological main control parameters, engineering main control parameters, and production data of the well with the closest attributes of the new well to be predicted are input into the production prediction model, the production prediction model can output the production data of the new well to be predicted, that is, obtain the production prediction result. After obtaining the production prediction result, a single well production prediction plan can also be drawn based on the production prediction result, such as Figure 5 shown.

[0083] In one possible implementation, in order to obtain better prediction results, the Bayesian optimization algorithm can also be used to optimize the hyperparameters of the yield prediction model; the Bayesian optimization algorithm can be used to optimize and adjust the tree depth, learning rate, number of sub-models, solution method and / or regularization term weight of the k-nearest neighbor algorithm. Among them, during Bayesian optimization, several sets of hyperparameters will be randomly used for trial calculations. The number of trial calculations can be set manually, for example, the number of trial calculations can be set to 10; then, the actual Bayesian optimization will be performed. After the trial calculation is completed, each Bayesian optimization will refer to the results of the previous calculation, that is, the performance of the model on the validation set, so as to select the hyperparameters, batch training number and iterative training number that should be used for the next calculation.

[0084] In the specific implementation, based on the same training set, three methods, namely KNN, XGBoost, and typical curve, were used to establish a single well production prediction model for the reservoir. The prediction accuracy of each optimized model was evaluated using the same untrained test set data. The evaluation results are shown in Table 1.

[0085] Table 1:

[0086] Model R2 Prediction accuracy This application method 0.82 81.8％ KNN 0.63 69％ XGBoost 0.59 64％ Typical curve 0.43 49％

[0087] The above results show that the prediction effect of the prediction method provided by the embodiment of the present application is much better than the traditional method.

[0088] In summary, the technical solutions provided by the embodiments of the present application have at least the following technical effects:

[0089] (1) By adding the production data of the wells with the closest attributes to the data set, the correlation between the input parameter group and the output parameter in the production prediction process is improved, thereby improving the model prediction accuracy;

[0090] (2) By combining supervised and unsupervised learning, the relationship between tight sandstone production data and parameter groups is learned, and a nonlinear mapping relationship between production and its parameter groups is established;

[0091] (3) The Bayesian hyperparameter optimization method is used to optimize the model structure and comprehensively improve the prediction accuracy of the model, thereby achieving accurate prediction of the production of new wells, providing a reliable basis for the planning and deployment of the production of new wells in oil fields, and facilitating the efficient development of oil and gas reservoirs.

[0092] Corresponding to the above embodiment, the present application also provides a tight sandstone oil and gas reservoir production prediction device.

[0093] See also Figure 6 , is a structural block diagram of a tight sandstone oil and gas reservoir production prediction device provided in this application. Figure 6 As shown, it mainly includes the following modules.

[0094] The data collection module 601 is used to collect single-well data of the target oil and gas field to obtain a first data set, wherein the single-well data includes geological parameters, engineering parameters and production data of the single well;

[0095] A main control parameter determination module 602 is used to determine the geological main control parameters and engineering main control parameters that affect the production data from the geological parameters and the engineering parameters;

[0096] A normalization processing module 603 is configured to perform normalization processing on the geological control parameters and engineering control parameters in the first data set to obtain a second data set, wherein the single well data in the second data set includes normalized geological control parameters, normalized engineering control parameters, and production data;

[0097] The production data calculation module 604 of the well with the closest attribute is used to calculate the production data of the well with the closest attribute for each well in the training set using the k-nearest neighbor algorithm, wherein the training set belongs to the second data set;

[0098] The production prediction model establishment module 605 is used to establish a production prediction model using the normalized geological control parameters, normalized engineering control parameters, production data of the well with the closest attributes, and production data in the training set using the XGBoost algorithm;

[0099] The new well production prediction module 606 is used to input the geological main control parameters, engineering main control parameters and production data of the well with the closest attributes of the new well to be predicted into the production prediction model to obtain the production prediction result of the new well to be predicted.

[0100] It should be pointed out that the specific contents involved in the embodiments of the present application can be found in the description of the above method embodiments. For the sake of brevity, they will not be repeated here.

[0101] Corresponding to the above embodiment, an embodiment of the present application further provides an electronic device.

[0102] See also Figure 7 , is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 7 As shown, the electronic device 700 may include: a processor 701, a memory 702, and a communication unit 703. These components communicate via one or more buses. Those skilled in the art will appreciate that the electronic device structure shown in the figure does not limit the embodiments of the present application. It may be a bus structure or a star structure, and may include more or fewer components than shown, or combine certain components, or arrange the components differently.

[0103] The communication unit 703 is used to establish a communication channel so that the electronic device can communicate with other devices.

[0104] The processor 701 is the control center of the electronic device. It uses various interfaces and lines to connect various parts of the entire electronic device. It runs or executes software programs and / or modules stored in the memory 702, and calls data stored in the memory to perform various functions of the electronic device and / or process data. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of multiple packaged ICs with the same or different functions. For example, the processor 701 can include only a central processing unit (CPU). In the embodiment of the present application, the CPU can be a single computing core or multiple computing cores.

[0105] The memory 702 is used to store execution instructions of the processor 701. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0106] When the execution instructions in the memory 702 are executed by the processor 701 , the electronic device 700 is enabled to execute part or all of the steps in the above method embodiments.

[0107] Corresponding to the above embodiment, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium may store a program, wherein, when the program is executed, the device containing the computer-readable storage medium may be controlled to perform some or all of the steps in the above method embodiment. In a specific implementation, the computer-readable storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0108] Corresponding to the above embodiment, an embodiment of the present application further provides a computer program product, which includes executable instructions. When the executable instructions are executed on a computer, the computer executes some or all of the steps in the above method embodiment.

[0109] In the embodiments of the present application, "at least one" refers to one or more, and "more" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and 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 and c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0110] Those skilled in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented using a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0111] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0112] In the several embodiments provided in this application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of this application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.

[0113] The above description is merely a specific embodiment of the present application. Any person skilled in the art may easily conceive of variations or substitutions within the technical scope disclosed in this application, and such variations or substitutions shall be within the scope of protection of this application. The scope of protection of this application shall be subject to the scope of protection of the claims.

Claims

1. A method for predicting the production of tight sandstone oil and gas reservoirs, characterized in that: include: Collecting single-well data of a target oil and gas field to obtain a first data set, wherein the single-well data includes geological parameters, engineering parameters, and production data of the single well; Determining, from among the geological parameters and the engineering parameters, the geological master controlling parameters and the engineering master controlling parameters that affect the production data, including: applying an XGBoost algorithm to calculate the degree of influence of the geological parameters and the engineering parameters on the production data, and determining, from among the geological parameters and the engineering parameters, the geological master controlling parameters and the engineering master controlling parameters that affect the production data based on the degree of influence, including: using the geological parameters and the engineering parameters that rank in the top 50% of the degree of influence as the geological master controlling parameters and the engineering master controlling parameters that affect the production data; Normalizing the geological control parameters and engineering control parameters in the first data set to obtain a second data set, wherein the single well data in the second data set includes normalized geological control parameters, normalized engineering control parameters, and production data; Calculating, using a k-nearest neighbor algorithm, the production data of the well whose attribute is closest to that of each well in a training set, wherein the training set belongs to the second data set; Establishing a production prediction model using the normalized geological control parameters, normalized engineering control parameters, production data of the well with the closest attributes, and production data in the training set using the XGBoost algorithm; including calculating the production data of the well with the closest attributes for each well in the test set using the k-nearest neighbor algorithm, where the test set belongs to the second data set; and testing the production prediction model using the normalized geological control parameters, normalized engineering control parameters, production data of the well with the closest attributes, and production data in the test set; The geological main control parameters, engineering main control parameters, and production data of the well with the closest attributes of the new well to be predicted are input into the production prediction model to obtain the production prediction result of the new well to be predicted; including using a Bayesian optimization algorithm to optimize the hyperparameters of the production prediction model; using a Bayesian optimization algorithm to optimize and adjust the tree depth, learning rate, number of sub-models, solution method and / or regularization term weight of the k-nearest neighbor algorithm.

2. The method according to claim 1, characterized in that The geological parameters include: at least one of water saturation, porosity, brittleness index and average fracture pressure; The engineering parameters include at least one of: fracturing operation displacement, average section length, number of sections, horizontal section length, proppant dosage per section, minimum pump stop pressure, maximum pump stop pressure, nitrogen volume pumped into each section, average operation pressure, and azimuth angle; The production data includes: single well ultimate recoverable reserves EUR.

3. The method according to claim 1, characterized in that The single well data further includes location parameters, which include at least one of bottom hole longitude, bottom hole latitude, and average depth of development layers.

4. A tight sandstone oil and gas reservoir production prediction device, characterized in that: include: A data collection module is used to collect single-well data of the target oil and gas field to obtain a first data set, wherein the single-well data includes geological parameters, engineering parameters and production data of the single well; a master control parameter determination module, configured to determine, from among the geological parameters and the engineering parameters, the geological master control parameters and the engineering master control parameters that affect the production data, including: applying an XGBoost algorithm to calculate the degree of influence of the geological parameters and the engineering parameters on the production data, and determining, from among the geological parameters and the engineering parameters, the geological master control parameters and the engineering master control parameters that affect the production data based on the degree of influence, including: using the geological parameters and the engineering parameters ranked in the top 50% of the degree of influence as the geological master control parameters and the engineering master control parameters that affect the production data; a normalization processing module, configured to perform normalization processing on the geological main control parameters and the engineering main control parameters in the first data set to obtain a second data set, wherein the single well data in the second data set includes normalized geological main control parameters, normalized engineering main control parameters, and production data; a production data calculation module for the well with the closest attribute, configured to calculate the production data of the well with the closest attribute for each well in the training set using a k-nearest neighbor algorithm, wherein the training set belongs to the second data set; a production prediction model establishment module, configured to establish a production prediction model using the normalized geological control parameters, normalized engineering control parameters, production data of the well with the closest attributes, and production data in the training set using the XGBoost algorithm; including calculating the production data of the well with the closest attributes for each well in the test set using the k-nearest neighbor algorithm, wherein the test set belongs to the second data set; and testing the production prediction model using the normalized geological control parameters, normalized engineering control parameters, production data of the well with the closest attributes, and production data in the test set; The new well production prediction module is used to input the geological main control parameters, engineering main control parameters, and production data of the well with the closest attributes of the new well to be predicted into the production prediction model to obtain the production prediction result of the new well to be predicted; including using the Bayesian optimization algorithm to optimize the hyperparameters of the production prediction model; using the Bayesian optimization algorithm to optimize the tree depth, learning rate, number of sub-models, solution method and / or regularization term weight of the k-nearest neighbor algorithm.

5. An electronic device, characterized in that: include: processor; Memory; and a computer program, wherein the computer program is stored in the memory, and the computer program includes instructions, which, when executed by the processor, enable the electronic device to perform the method according to any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the method according to any one of claims 1 to 3.

Images