A method, device, equipment and medium for evaluating gas-containing properties of tight reservoirs

Abstract

The present invention discloses a method, device, equipment and storage medium for evaluating the gas-bearing performance of a tight reservoir. The method comprises: obtaining control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determining the gas-bearing characteristics of the tight reservoir to be evaluated according to the control factor data; determining the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and a pre-trained gas-bearing performance prediction model; wherein the gas-bearing performance prediction model comprises a plurality of K nearest neighbor sub-models, and the number of K nearest neighbor sub-models in the gas-bearing performance prediction model is determined based on the test results of a preset number of candidate gas-bearing performance prediction models. This technical solution solves the problems of low efficiency and poor accuracy in evaluating the gas-bearing performance of tight reservoirs, and a fast and accurate evaluation of the gas-bearing performance of tight reservoirs can be achieved through the gas-bearing performance prediction model.

Description

Technical Field

[0001] The present invention relates to the technical field of reservoir evaluation, and in particular to a method, device, electronic equipment and storage medium for evaluating the gas-containing properties of a dense reservoir. Background Art

[0002] Tight sandstone gas refers to natural gas resources enriched in low-permeability to ultra-low permeability tight sandstone reservoirs. It is an important part of unconventional natural gas exploration and development, and is of great strategic significance to improving the energy structure and ensuring source security.

[0003] At present, the existing methods for evaluating the gas-bearing properties of tight reservoirs mainly include geological risk probability evaluation, fuzzy comprehensive evaluation, multiple regression evaluation, Bayesian and evidence weight evaluation, Mahalanobis distance and support vector machine evaluation, and oil and gas accumulation value evaluation. Among them, geological risk probability evaluation and fuzzy comprehensive evaluation mainly evaluate the gas-bearing properties of tight reservoirs through artificial experience, which are highly subjective and have a high degree of human interference, and are not suitable for actual exploration work. Multiple regression evaluation, Bayesian and evidence weight evaluation, Mahalanobis distance and support vector machine evaluation, and oil and gas accumulation value evaluation methods all quantitatively describe the gas-bearing properties based on data mining and comprehensive analysis of actual oil and gas exploration results. They have high requirements on data sample conditions and require continuous variables, and the gas-bearing performance evaluation effect is not ideal.

[0004] Therefore, there is an urgent need for an evaluation method for the gas-bearing properties of tight reservoirs to achieve a high-accuracy evaluation of the gas-bearing properties of tight reservoirs. Summary of the invention

[0005] The present invention provides a method, device, equipment and storage medium for evaluating the gas-containing performance of a tight reservoir, so as to solve the problems of low efficiency and poor accuracy in evaluating the gas-containing performance of a tight reservoir. Through a gas-containing performance prediction model, a fast and accurate evaluation of the gas-containing performance of a tight reservoir can be achieved.

[0006] According to one aspect of the present invention, a method for evaluating gas-bearing properties of a tight reservoir is provided, the method comprising:

[0007] Acquiring control factor data of gas-bearing properties of the tight reservoir to be evaluated in the target area, and determining the gas-bearing characteristics of the tight reservoir to be evaluated based on the control factor data;

[0008] Determine the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and the pre-trained gas-bearing performance prediction model;

[0009] The gas-bearing performance prediction model includes a plurality of K nearest neighbor sub-models, and the number of K nearest neighbor sub-models in the gas-bearing performance prediction model is determined based on the test results of a preset number of candidate gas-bearing performance prediction models.

[0010] According to another aspect of the present invention, a device for evaluating gas-bearing properties of a tight reservoir is provided, the device comprising:

[0011] A gas-bearing characteristic determination module is used to obtain control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determine the gas-bearing characteristics of the tight reservoir to be evaluated based on the control factor data;

[0012] An evaluation result determination module is used to determine the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and the pre-trained gas-bearing performance prediction model;

[0013] The gas-bearing performance prediction model includes a plurality of K nearest neighbor sub-models, and the number of K nearest neighbor sub-models in the gas-bearing performance prediction model is determined based on the test results of a preset number of candidate gas-bearing performance prediction models.

[0014] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0015] at least one processor; and

[0016] a memory communicatively connected to the at least one processor; wherein,

[0017] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for evaluating the gas-containing properties of a tight reservoir described in any embodiment of the present invention.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for evaluating the gas content performance of a tight reservoir as described in any embodiment of the present invention when executed.

[0019] The technical solution of the embodiment of the present invention obtains the control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determines the gas-bearing characteristics of the tight reservoir to be evaluated according to the control factor data; and determines the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and the pre-trained gas-bearing performance prediction model. This technical solution solves the problems of low efficiency and poor accuracy in the evaluation of the gas-bearing performance of the tight reservoir, and can realize fast and accurate evaluation of the gas-bearing performance of the tight reservoir through the gas-bearing performance prediction model.

[0020] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 This is a flow chart of a method for evaluating gas-containing properties of a tight reservoir provided in accordance with Embodiment 1 of the present invention;

[0023] Figure 2 This is a flow chart of a method for evaluating gas-bearing properties of a tight reservoir provided in accordance with Embodiment 2 of the present invention;

[0024] Figure 3 2 is a schematic structural diagram of a gas-containing property evaluation device for a tight reservoir provided in accordance with a third embodiment of the present invention;

[0025] Figure 4 It is a schematic diagram of the structure of an electronic device for implementing the method for evaluating the gas-containing properties of a dense reservoir according to an embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. The acquisition, storage, use, processing, etc. of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0028] Embodiment 1

[0029] Figure 1 A flowchart of a method for evaluating the gas-bearing performance of a tight reservoir is provided for the first embodiment of the present invention. This embodiment is applicable to the evaluation scenario of the gas-bearing performance of a tight reservoir. The method can be performed by a device for evaluating the gas-bearing performance of a tight reservoir. The device can be implemented in the form of hardware and / or software, and the device can be configured in an electronic device. Figure 1 As shown, the method includes:

[0030] S110, obtaining control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determining the gas-bearing characteristics of the tight reservoir to be evaluated according to the control factor data.

[0031] This solution can be executed by electronic devices such as computers and servers, and the electronic devices can obtain the control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area through the oil and gas exploration platform. It can be understood that the target area can be a research area of ​​an oil and gas reservoir, and the tight reservoir to be evaluated can be a tight reservoir that needs to be evaluated for gas-bearing performance. The control factor data of the gas-bearing performance can be data characterizing the conditions of the tight reservoir, and specifically, the control factor data can include at least one of the hydrocarbon generation intensity, relative structural height, reservoir physical properties, and distance from the adjusted fracture.

[0032] The electronic device can determine the control factor data of the gas-bearing performance of the tight reservoir to be evaluated based on the well logging data, porosity data, permeability data, hydrocarbon generation intensity data of the source rock, and the top surface structural height contour map of the target layer in the target area. Specifically, the electronic device can obtain the well logging interpretation reservoir data of the target layer of the tight sandstone gas reservoir in the target area, such as the well location name, well location coordinates, reservoir top depth and reservoir bottom depth of each well logging interpretation reservoir. After obtaining the basic data such as the well location name, well location coordinates, reservoir top depth and reservoir bottom depth of each well logging, the electronic device can determine the source rock layer adjacent to the tight reservoir, and obtain the distribution law of the hydrocarbon generation intensity of the source rock layer on the plane through the hydrocarbon generation potential method, basin simulation method and other methods, and then obtain the hydrocarbon generation intensity data of the well logging interpretation reservoir of each target layer of the tight sandstone gas reservoir. It is easy to understand that the hydrocarbon generation intensity refers to the hydrocarbon generation amount per unit area of ​​a set of source rocks, which can effectively characterize the hydrocarbon source sufficiency of the tight sandstone gas reservoir.

[0033] Tight sandstone gas reservoirs are mainly distributed in relatively low structural areas. The relative structural height can characterize the relative structural height of a set of tight reservoirs, and then evaluate the favorable degree of tight reservoir development. According to the structural height value line diagram of the top surface of the target layer in the target area, the electronic equipment can identify the structural height of each target layer at each well location, and determine the maximum and minimum structural heights that match each target layer at each well location. The calculation formula for the relative structural height of the tight reservoir can be expressed as:

[0034]

[0035] Among them, H rgi It indicates the relative structural height of the tight reservoir in the target layer, H indicates the structural height of the well in the target layer, and H min Indicates the minimum structural height that the well location matches in the target layer, H max Indicates the maximum structural height matched by the well location in the target layer.

[0036] It can be understood that porosity and permeability can effectively characterize the storage space, seepage capacity and self-sealing property of tight reservoirs. The porosity and permeability of the target layer logging interpretation reservoirs of each tight sandstone gas reservoir can be obtained through core test data or logging interpretation data. The fracture activity after the formation of tight sandstone gas reservoirs will destroy the tight sandstone gas reservoirs. These fractures are called adjustment fractures. The distance between the tight sandstone gas reservoir and the adjustment fracture can effectively characterize the later preservation conditions of the tight sandstone reservoir. According to the distribution map of the adjustment fractures of the target layer in the target area, the electronic equipment can identify the adjustment fractures that were active after the tight sandstone gas accumulation period, find the midpoint of the adjustment fracture closest to each well location and calculate the distance, so as to obtain the distance data of the target layer logging interpretation reservoir from the adjustment fracture of each tight sandstone gas reservoir.

[0037] After obtaining the control factor data, the electronic device can perform preprocessing operations such as data cleaning, normalization and standardization on the control factor data, and generate the gas-containing characteristics of the tight reservoir to be evaluated based on the preprocessed control factor data, for example, generating a one-dimensional feature vector based on the control factor data of each well location in each target layer.

[0038] S120, determining a gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and a pre-trained gas-bearing performance prediction model.

[0039] After obtaining the gas-bearing characteristics of the tight reservoir to be evaluated, the gas-bearing characteristics are input into the gas-bearing performance prediction model, and the electronic device can determine the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the output data of the gas-bearing performance prediction model. The gas-bearing performance prediction model includes a plurality of K nearest neighbor sub-models, and the number of K nearest neighbor sub-models in the gas-bearing performance prediction model is determined based on the test results of a preset number of candidate gas-bearing performance prediction models. The gas-bearing performance evaluation result may include evaluation levels such as oil and gas, low oil and gas, etc.

[0040] It can be understood that the gas-bearing performance prediction model can be a multi-random K-nearest neighbor joint prediction model, that is, based on the gas-bearing performance prediction results output by multiple K-nearest neighbor sub-models, the gas-bearing performance evaluation results of the tight reservoir to be evaluated are determined. The number of K-nearest neighbor sub-models in the gas-bearing performance prediction model can be used as a parameter of the gas-bearing performance prediction model. The electronic device can obtain multiple candidate gas-bearing performance prediction models by setting different parameters, and determine the gas-bearing performance prediction model according to the test results of each candidate gas-bearing performance prediction model.

[0041] The technical solution of the embodiment of the present invention obtains the control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determines the gas-bearing characteristics of the tight reservoir to be evaluated according to the control factor data; and determines the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and the pre-trained gas-bearing performance prediction model. This technical solution solves the problems of low efficiency and poor accuracy in the evaluation of the gas-bearing performance of the tight reservoir, and can realize fast and accurate evaluation of the gas-bearing performance of the tight reservoir through the gas-bearing performance prediction model.

[0042] Embodiment 2

[0043] Figure 2 This is a flow chart of a method for evaluating the gas-bearing performance of a tight reservoir provided in the second embodiment of the present invention. This embodiment refines the training process of the gas-bearing performance prediction model based on the above embodiment. Figure 2 As shown, the method includes:

[0044] S210, obtaining a set of K nearest neighbor sub-model quantity values, a set of K nearest neighbor coefficient values, and multiple groups of control factor data on gas-bearing properties of evaluated tight reservoirs.

[0045] It can be understood that the gas-containing performance prediction model may include multiple parameters such as the number of K nearest neighbor sub-models and the K nearest neighbor coefficients. The electronic device can select the optimal parameter combination from different parameter combinations to ensure that the prediction effect of the gas-containing performance prediction model is the best. Specifically, the electronic device can pre-acquire the value range of the number of K nearest neighbor sub-models and the value range of the K nearest neighbor coefficients. Among them, the number of K nearest neighbor sub-models and the K nearest neighbor coefficients are both non-zero natural numbers. The value set of the number of K nearest neighbor sub-models may include all selectable values ​​of the number of K nearest neighbor sub-models within its value range, and the value set of the K nearest neighbor coefficients may include all selectable values ​​of the K nearest neighbor coefficients within its value range.

[0046] S220. Determine at least two parameter combinations according to the K nearest neighbor sub-model quantity value set and the K nearest neighbor coefficient value set.

[0047] The electronic device may combine elements of the K nearest neighbor submodel quantity value set with elements of the K nearest neighbor coefficient value set to obtain multiple parameter combinations, wherein each parameter combination includes a K nearest neighbor submodel quantity and a K nearest neighbor coefficient.

[0048] S230, dividing each group of control factor data into samples according to the gas-bearing properties matched by each group of control factor data, and determining a sample set matched by each parameter combination.

[0049] It can be understood that the electronic device can obtain the production capacity data of the well logging interpretation reservoir of each target layer of each tight sandstone gas reservoir, such as the cumulative production of each gas layer, the daily production of each gas layer, the thickness of the gas layer and other data. According to the production capacity data of the well logging interpretation reservoir of each target layer of the tight sandstone gas reservoir, the electronic device can determine to evaluate each tight reservoir and obtain the gas-containing performance of the well logging interpretation reservoir of each target layer of the tight sandstone gas reservoir. Each well logging interpretation reservoir of the target layer of the tight sandstone gas reservoir can correspond to a set of control factor data and a gas-containing performance. Among them, the gas-containing performance can be used as a label of the control factor data of the well logging interpretation reservoir of the target layer of the same tight sandstone gas reservoir, and the gas-containing performance can be oil-containing or low oil-gas-containing.

[0050] According to the gas-containing properties matched by each group of control factor data, the electronic device can divide each group of control factor data into samples to determine the number of samples of each type of gas-containing properties. According to the number of samples of each type of gas-containing properties, the sample set matched by each parameter combination is determined. Specifically, the sample set matched by each parameter combination can be determined according to the number of K nearest neighbor sub-models in each parameter combination.

[0051] In this solution, optionally, the gas-containing property includes a first gas-containing property and a second gas-containing property;

[0052] According to the gas-bearing properties matched by each group of control factor data, each group of control factor data is divided into samples to determine the sample set matched by each parameter combination, including:

[0053] Determine a first sample set and a second sample set according to the gas-bearing properties matched by each group of control factor data; wherein the first sample set includes each group of control factor data whose gas-bearing properties are the first gas-bearing properties, and the second sample set includes each group of control factor data whose gas-bearing properties are the second gas-bearing properties;

[0054] According to the first sample set and the second sample set, a sample set matching each parameter combination is determined.

[0055] It can be understood that the first gas-bearing property may be oil and gas-bearing, and the second gas-bearing property may be low oil and gas-bearing. The electronic device may divide the data set consisting of the control factor data and its labels according to the labels of each group of control factor data, that is, the matching gas-bearing properties, to obtain a first sample set and a second sample set. The first sample set includes each group of control factor data whose gas-bearing properties are the first gas-bearing properties, and the second sample set includes each group of control factor data whose gas-bearing properties are the second gas-bearing properties. Based on the first sample set, the second sample set, and the number of K-nearest neighbor sub-models in each parameter combination, the electronic device may determine the sample set that matches each parameter combination.

[0056] In this embodiment, optionally, determining a sample set matching each parameter combination according to the first sample set and the second sample set includes:

[0057] If the number of first samples in the first sample set is greater than the number of second samples in the second sample set, determining the first sample set to which each parameter combination matches according to the second sample number;

[0058] Determining a sample set matching each parameter combination according to the first sample set and the second sample set matching each parameter combination;

[0059] or,

[0060] If the number of first samples in the first sample set is less than the number of second samples in the second sample set, determining the first sample set for which each parameter combination matches according to the number of first samples;

[0061] According to the second sample set matched by each parameter combination and the first sample set, a sample set matched by each parameter combination is determined.

[0062] It is easy to understand that in the data set consisting of the control factor data and its labels, the number of first samples in the first sample set and the number of second samples in the second sample set may be equal or unequal. If they are equal, the union of the first sample set and the second sample set can be used as the sample set for matching each parameter combination. If they are not equal, a sample set with a larger number of samples is determined in the first sample set and the second sample set, all samples in the sample set with a smaller number of samples are retained, and a sample subset with the same number of samples as the sample set with a smaller number of samples is randomly selected from the sample set with a larger number of samples, and the union of the sample subset and the sample set with a smaller number of samples is used as the sample set for matching each parameter combination.

[0063] It should be noted that, when the number of samples in the first sample set and the second sample set are not equal, the sample set matched by each parameter combination may be one or more, and the sample sets matched by each parameter combination may be the same or different. For example, the electronic device may use the number of K nearest neighbor sub-models in each parameter combination as the random number, and randomly select multiple sample subsets with the same number of samples as the sample set with a smaller number of samples from the sample set with a larger number of samples according to the random number, and use the union of the sample subsets obtained randomly each time and the sample set with a smaller number of samples as the sample sets matched by each K nearest neighbor sub-model.

[0064] This solution can maximize the utilization of sample data, ensure the reliability and scenario applicability of the gas-bearing performance prediction model, and avoid the impact of abnormal data on model test results.

[0065] S240, determining the test results of the candidate gas-bearing performance prediction models matched by each parameter combination according to the sample sets matched by each parameter combination, and determining the target parameter combination according to each test result.

[0066] After obtaining the sample sets matching each parameter combination, the electronic device can train each K nearest neighbor sub-model based on the K nearest neighbor algorithm according to the sample sets matching each parameter combination, and perform model testing after each training to obtain the test results of the candidate gas-bearing performance prediction model matching each parameter combination. According to the test results of the candidate gas-bearing performance prediction model matching each parameter combination, the target parameter combination is determined in each parameter combination.

[0067] In a feasible solution, optionally, determining the test results of the candidate gas-bearing performance prediction model matched by each parameter combination according to the sample set matched by each parameter combination includes:

[0068] Determine each parameter combination as the current parameter combination in sequence;

[0069] Determine a test sample from samples that have not been tested in the sample set that matches the current parameter combination;

[0070] According to the current parameter combination, based on the K nearest neighbor algorithm, the other samples in the sample set matching the current parameter combination except the test sample are trained to obtain a candidate gas-bearing performance prediction model matching the test sample;

[0071] Determine the gas-bearing performance prediction result of the test sample in this iteration according to the test sample and the candidate gas-bearing performance prediction model matched by the test sample;

[0072] Return to the execution to determine the test sample in the samples that have not been tested in the sample set that matches the current parameter combination, until there are no untested samples in the sample set that matches the current parameter combination, and output the gas-containing property prediction result of each sample in the sample set;

[0073] According to the gas-bearing performance prediction results of each sample, the test results of the candidate gas-bearing performance prediction model matched by the current parameter combination are determined.

[0074] In this solution, the electronic device can sequentially determine each parameter combination as the current parameter combination, and randomly select one or more samples from the untested samples in the sample set matched by each K nearest neighbor sub-model as test samples. According to the K nearest neighbor coefficient in the current parameter combination, based on the K nearest neighbor algorithm, the other samples in the sample set matched by each K nearest neighbor sub-model except the test sample are trained to obtain the test object of the test sample, that is, the candidate gas content performance prediction model corresponding to the test sample. Among them, the candidate gas content performance prediction model includes each K nearest neighbor sub-model matched by the current parameter combination. The test sample is input into the candidate gas content performance prediction model, and according to the output data of the candidate gas content performance prediction model, the electronic device can determine the gas content performance prediction result of the test sample.

[0075] After the test sample is tested, the electronic device may return to determine the test sample from the samples that have not been tested in the sample set matched by the current parameter combination, until there are no untested samples in the sample set matched by each K nearest neighbor sub-model, and output the gas-containing property prediction result of each sample. Based on the gas-containing property prediction result of each sample, the electronic device may determine the test result of the candidate gas-containing property prediction model matched by the current parameter combination.

[0076] On the basis of the above scheme, optionally, the candidate gas-bearing property prediction model matched by the current parameter combination includes the K nearest neighbor sub-models of the number of K nearest neighbor sub-models in the current parameter combination;

[0077] The step of determining the gas-containing property prediction result of the test sample in this iteration according to the test sample and the candidate gas-containing property prediction model matched by the test sample includes:

[0078] The test samples matched by each K nearest neighbor sub-model are input into the K nearest neighbor sub-model to determine the gas-bearing performance prediction results of the test samples matched by each K nearest neighbor sub-model in this iteration.

[0079] It can be understood that the electronic device can generate gas-containing characteristics corresponding to the test samples based on the test samples matched by each K nearest neighbor sub-model, input the gas-containing characteristics matched by each K nearest neighbor sub-model into the K nearest neighbor sub-model, and determine the gas-containing performance prediction results output by each K nearest neighbor sub-model, that is, the gas-containing performance prediction results of the test samples matched by each K nearest neighbor sub-model in this iteration.

[0080] Based on the above embodiment, optionally, the outputting of the gas-containing property prediction result of each sample includes:

[0081] Determine a first gas-containing property quantity and a second gas-containing property quantity of each sample in each iteration of the gas-containing property prediction result;

[0082] The gas-containing property prediction result of each sample is determined according to the first gas-containing property quantity and the second gas-containing property quantity.

[0083] After obtaining the gas-containing performance prediction results output by each K-nearest neighbor sub-model in each iteration, the electronic device can count each gas-containing performance prediction result to obtain a first gas-containing performance quantity and a second gas-containing performance quantity. It can be understood that in each iterative training of each K-nearest neighbor sub-model, there may be multiple gas-containing performance prediction results for each sample, and the electronic device can use the gas-containing performance prediction result that appears most frequently in the gas-containing performance prediction results as the gas-containing performance prediction result of the sample. Specifically, the electronic device can compare the first gas-containing performance quantity and the second gas-containing performance quantity of each sample. If the first gas-containing performance quantity is greater than the second gas-containing performance quantity, the first gas-containing performance is used as the gas-containing performance prediction result of the sample. If the second gas-containing performance quantity is greater than the first gas-containing performance quantity, the second gas-containing performance is used as the gas-containing performance prediction result of the sample.

[0084] In this solution, optionally, the test result of the candidate gas-bearing performance prediction model matched by the current parameter combination is determined according to the gas-bearing performance prediction result of each sample, including:

[0085] The gas-bearing property prediction result of each sample is compared with the gas-bearing property matched by each sample, and the prediction accuracy of the candidate gas-bearing property prediction model matched by the current parameter combination is determined according to the comparison result;

[0086] Determining the target parameter combination according to each test result includes:

[0087] The target parameter combination is determined according to the prediction accuracy of the candidate gas-bearing performance prediction model matched with each parameter combination.

[0088] After obtaining the gas-containing performance prediction results of each sample, each sample may correspond to a gas-containing performance prediction result. The electronic device may compare the gas-containing performance prediction result of each sample with the gas-containing performance matched by the sample. If the gas-containing performance prediction result is the same as the sample label (the gas-containing performance matched by the sample), it is determined that the prediction of the sample is correct. If the gas-containing performance prediction result is different from the sample label (the gas-containing performance matched by the sample), it is determined that the prediction of the sample is wrong. Based on the number of samples predicted correctly and the total number of samples, the electronic device may calculate the prediction accuracy of the candidate gas-containing performance prediction model matched by the current parameter combination.

[0089] After obtaining the prediction accuracy of the candidate gas-containing property prediction models matched by each parameter combination, the electronic device may select the parameter combination corresponding to the candidate gas-containing property prediction model with the highest prediction accuracy as the target parameter combination.

[0090] S250, determining a gas-bearing performance prediction model according to candidate gas-bearing performance prediction models matched by target parameter combination.

[0091] The electronic device can use the candidate gas-bearing performance prediction model matched with the target parameter combination as the gas-bearing performance prediction model for evaluating the gas-bearing performance of the tight reservoir to be evaluated.

[0092] S260, obtaining control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determining the gas-bearing characteristics of the tight reservoir to be evaluated based on the control factor data.

[0093] S270, determining a gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and a pre-trained gas-bearing performance prediction model.

[0094] The technical solution of the embodiment of the present invention obtains the control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determines the gas-bearing characteristics of the tight reservoir to be evaluated according to the control factor data; and determines the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and the pre-trained gas-bearing performance prediction model. This technical solution solves the problems of low efficiency and poor accuracy in the evaluation of the gas-bearing performance of the tight reservoir, and can realize fast and accurate evaluation of the gas-bearing performance of the tight reservoir through the gas-bearing performance prediction model.

[0095] Specific application scenario 1

[0096] This application scenario is a specific embodiment based on the above embodiment. In this solution, the gas content performance evaluation of the tight reservoir includes the following tasks:

[0097] (1) Constructing a tight reservoir condition dataset

[0098] Generally, according to the summary results of the main controlling factors of gas content in tight reservoirs recognized by scholars in this field, the data that can characterize the conditions of tight reservoirs include at least four aspects: the hydrocarbon generation intensity of the source rock adjacent to the tight reservoir, the relative structural height of the tight reservoir, the physical properties of the tight reservoir, and the distance of the tight reservoir from the adjusted fault. To construct the above four indicators, it is necessary to collect original data, including well logging data, porosity data, permeability data, hydrocarbon generation intensity data of the source rock, and the contour map of the top structural height of the target formation.

[0099] a. Obtaining logging interpretation reservoir data of the target layer of tight sandstone gas reservoir in the study area

[0100] The logging interpretation reservoir data of the target layer of the tight sandstone gas reservoir in the study area are obtained, including the well location name, well location coordinates, reservoir top depth and reservoir bottom depth of each logging interpretation reservoir.

[0101] b. Obtaining hydrocarbon generation intensity data of source rocks near tight reservoirs

[0102] Hydrocarbon generation intensity refers to the amount of hydrocarbon generated per unit area of ​​a set of source rocks, which can effectively characterize the hydrocarbon sufficiency of tight sandstone gas reservoirs. First, the source rock layers adjacent to the tight reservoir are determined, and then the distribution law of the hydrocarbon generation intensity of the above source rock layers on the plane is obtained through methods such as hydrocarbon generation potential method and basin simulation method, and then the hydrocarbon generation intensity data of the reservoir interpreted by well logging of the target layer of each tight sandstone gas reservoir is obtained.

[0103] c. Obtaining relative structural height data of tight reservoirs

[0104] Tight sandstone gas reservoirs are mainly distributed in relatively low structural parts. The relative structural height can characterize the relative structural height of a set of tight reservoirs, and then evaluate the favorable degree of tight reservoir development. According to the structural height value line diagram of the top surface of the target layer in the study area, the structural height of each target layer at each well position and the local maximum structural height and minimum structural height of its target layer are identified. The relative structural height of the tight reservoir is the structural height of each well position in the target layer minus the minimum structural height of the target layer divided by the maximum structural height of the target layer minus the minimum structural height. According to the above calculation method, the relative structural height data of the reservoir of the target layer of each tight sandstone gas reservoir is obtained.

[0105] d. Obtaining tight reservoir physical property data

[0106] Porosity and permeability can effectively characterize the storage space, seepage capacity and self-sealing property of tight reservoirs. The porosity and permeability data of target layers of tight sandstone gas reservoirs can be obtained by actual core testing or logging interpretation data.

[0107] e. Obtain the distance data of tight reservoir and adjusted fracture

[0108] The fault activity after the formation of tight sandstone gas reservoirs will destroy the tight sandstone gas reservoirs. These faults are called adjustment faults. The distance between the tight sandstone gas reservoir and the adjustment fault can effectively characterize the later preservation conditions of the tight sandstone reservoir. According to the distribution map of the adjustment faults in the target layer of the study area, the adjustment faults active after the formation of tight sandstone gas are identified, the midpoint of the adjustment fault closest to each well location is found and the size is calculated, and finally the distance data of the reservoir from the adjustment fault is obtained from the well logging interpretation of the target layer of each tight sandstone gas reservoir.

[0109] (2) Classification of gas content

[0110] First, the production capacity data of the target layers of each tight sandstone gas reservoir are collected through logging interpretation, including the cumulative production of each gas layer, the daily production of each gas layer, the thickness of the gas layer and other indicators that can describe the production capacity of the gas layer.

[0111] Secondly, according to the production capacity of each gas layer in the study area, it is divided into two categories: oil and gas and low oil and gas.

[0112] (3) Establishing a training dataset

[0113] Firstly, the hydrocarbon generation intensity of the source rock adjacent to the reservoir interpreted by logging in the target layer of each tight sandstone gas reservoir, the relative structural height of the tight reservoir, the physical properties of the tight reservoir, and the distance of the tight reservoir from the adjusted fault are obtained as attributes and normalized. Then, the test result type of each test layer is set as a label to establish a data set.

[0114] (4) Building a multi-random K-nearest neighbor joint prediction model

[0115] a. Determine the number n of K nearest neighbor sub-models included in the multi-random K nearest neighbor joint prediction model that needs to be built.

[0116] b. Determine the number of samples in the gas-bearing layer and low-gas-bearing layer categories, and define the sample with a large number as multi-class data, and vice versa as minority-class data. Randomly extract samples with the same number of minority-class data from the multi-class data to build a new data set. Repeat the above process n times to build n data sets, and each data set will be trained into an independent K nearest neighbor sub-model.

[0117] c. Determine the optimal proximity coefficient (K) of the above n K nearest neighbor sub-models. In this process, the distance formula for calculating any two sample points is:

[0118]

[0119] in, is the distance between the i-th sample point data and the j-th sample point data; GI i , RGI i , Por i 、Perm i 、DBFi are the hydrocarbon generation intensity of the tight sandstone adjacent to the source rock, the relative structural height of the tight reservoir, the physical properties of the tight reservoir, and the distance between the tight reservoir and the adjusted fault of the i-th sample point data; GI j , RGI j , Por j 、Perm j 、DBF j They are respectively the hydrocarbon generation intensity of the tight sandstone adjacent to the source rock of the jth sample point, the relative structural height of the tight reservoir, the physical properties of the tight reservoir, and the distance of the tight reservoir from the adjusted fault.

[0120] (5) Select the optimal parameters of the multi-random K nearest neighbor joint prediction model

[0121] a. Construct all possible parameter combinations of the model

[0122] The optimal parameters of the proposed multi-random K nearest neighbor joint prediction model include two parameters: the number of K nearest neighbor sub-models n and the optimal proximity coefficient K of each K nearest neighbor sub-model. In order to construct all possible parameter combinations of the model, first, the value range of the optimal proximity coefficient K is set [K min ,K max ], where K, K min and K max are all non-zero natural numbers, setting the value range of the number of sub-models n [n min ,n max ], where n, n min 、n max All are non-zero natural numbers. List all combinations of the two parameters K and n, substitute all combination parameters into the model and evaluate the performance of each prediction model, and finally find the parameter combination that makes the model perform best.

[0123] b. The specific method of substituting all the combined parameters into the model and evaluating the performance of each prediction model mentioned above is as follows:

[0124] b1. Select any of the above parameter combinations to deploy the model. Assume that the selected parameter combination is [Ki,nj], and then substitute the parameters into (4)-a.

[0125] b2. Randomly select a sample point data from the constructed data set as the test data and keep the remaining sample point data as training data.

[0126] b3. Substitute the training data into (4)-b, substitute the classification parameters of the sub-model into Ki and use the divided nj data sets to train nj sub-models.

[0127] b4. Substitute the test model attributes reserved in step b2 into the nj sub-models trained in step b3 to obtain nj prediction results. The prediction result with the highest frequency among the nj prediction results is taken as the final prediction result.

[0128] b5. Loop steps b2 to b4 so that all the data in the data set are used as test data, obtain prediction results, and compare the actual results of each test data to obtain the overall prediction accuracy of all test data, which is used as the accuracy of the prediction model constructed by the parameter combination selected in step b1.

[0129] b6. Loop b1 to b5 to obtain the prediction accuracy of the prediction models constructed by all parameter combinations. The parameter combination of the prediction model with the highest prediction accuracy is used as the optimal parameter of the multi-random K nearest neighbor joint prediction model.

[0130] (6) The optimal parameter model obtained in step (5) is used as a prediction model for predicting the gas content of unknown tight reservoirs.

[0131] Embodiment 3

[0132] Figure 3 This is a schematic diagram of the structure of a gas-bearing property evaluation device for a dense reservoir provided in Example 3 of the present invention. Figure 3 As shown, the device comprises:

[0133] The gas-bearing characteristic determination module 310 is used to obtain the control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determine the gas-bearing characteristics of the tight reservoir to be evaluated based on the control factor data;

[0134] An evaluation result determination module 320 is used to determine the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and the pre-trained gas-bearing performance prediction model;

[0135] The gas-bearing performance prediction model includes a plurality of K nearest neighbor sub-models, and the number of K nearest neighbor sub-models in the gas-bearing performance prediction model is determined based on the test results of a preset number of candidate gas-bearing performance prediction models.

[0136] In this solution, optionally, the device further includes a model training module, and the model training module includes:

[0137] A parameter value acquisition unit, used to acquire a K-nearest neighbor sub-model quantity value set, a K-nearest neighbor coefficient value set, and multiple groups of control factor data of gas-bearing properties of evaluated tight reservoirs;

[0138] A parameter combination determination unit, used to determine at least two parameter combinations according to a set of K nearest neighbor submodel quantity values ​​and a set of K nearest neighbor coefficient value values; wherein each parameter combination includes a K nearest neighbor submodel quantity and a K nearest neighbor coefficient;

[0139] A sample set determination unit is used to divide each group of control factor data into samples according to the gas-bearing properties matched by each group of control factor data, and determine the sample set matched by each parameter combination;

[0140] A target combination determination unit is used to determine the test results of the candidate gas-bearing performance prediction models matched by each parameter combination according to the sample set matched by each parameter combination, and determine the target parameter combination according to each test result;

[0141] The prediction model determination unit is used to determine the gas-bearing performance prediction model according to the candidate gas-bearing performance prediction models matched by the target parameter combination.

[0142] In a feasible solution, optionally, the gas-containing property includes a first gas-containing property and a second gas-containing property;

[0143] The sample set determination unit is specifically used to:

[0144] Determine a first sample set and a second sample set according to the gas-bearing properties matched by each group of control factor data; wherein the first sample set includes each group of control factor data whose gas-bearing properties are the first gas-bearing properties, and the second sample set includes each group of control factor data whose gas-bearing properties are the second gas-bearing properties;

[0145] According to the first sample set and the second sample set, a sample set matching each parameter combination is determined.

[0146] In this solution, optionally, the target combination determination unit includes:

[0147] A current combination determination subunit, used to sequentially determine each parameter combination as the current parameter combination;

[0148] A test sample determination subunit, used to determine a test sample from samples that have not been tested in the sample set that matches the current parameter combination;

[0149] The candidate model determination subunit is used to train other samples in the sample set matched by the current parameter combination except the test sample based on the K nearest neighbor algorithm according to the current parameter combination, so as to obtain the candidate gas-bearing property prediction model matched by the test sample;

[0150] A sample testing subunit, configured to determine a prediction result of the gas-bearing performance of the test sample in this iteration according to the test sample and a candidate gas-bearing performance prediction model matched by the test sample;

[0151] The prediction result determination subunit is used to return to determine the test samples in the samples that have not been tested in the sample set that matches the current parameter combination, until there are no untested samples in the sample set that matches the current parameter combination, and output the gas-containing property prediction results of each sample;

[0152] The test result determination subunit is used to determine the test result of the candidate gas-bearing performance prediction model matched by the current parameter combination according to the gas-bearing performance prediction result of each sample.

[0153] In a preferred solution, the candidate gas-bearing property prediction model matched by the current parameter combination includes K nearest neighbor sub-models of the number of K nearest neighbor sub-models in the current parameter combination;

[0154] The sample testing subunit is specifically used for:

[0155] The test samples matched by each K nearest neighbor sub-model are input into the K nearest neighbor sub-model to determine the gas-bearing performance prediction results of the test samples matched by each K nearest neighbor sub-model in this iteration.

[0156] Based on the above solution, optionally, the prediction result determination subunit is specifically used to:

[0157] Determine a first gas-containing property quantity and a second gas-containing property quantity of each sample in each iteration of the gas-containing property prediction result;

[0158] The gas-containing property prediction result of each sample is determined according to the first gas-containing property quantity and the second gas-containing property quantity.

[0159] In this embodiment, optionally, the test result determination subunit is specifically used to:

[0160] The gas-bearing property prediction result of each sample is compared with the gas-bearing property matched by each sample, and the prediction accuracy of the candidate gas-bearing property prediction model matched by the current parameter combination is determined according to the comparison result;

[0161] The target combination determination unit is specifically used to:

[0162] The target parameter combination is determined according to the prediction accuracy of the candidate gas-bearing performance prediction model matched with each parameter combination.

[0163] The gas-containing performance evaluation device for a tight reservoir provided in an embodiment of the present invention can execute the gas-containing performance evaluation method for a tight reservoir provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0164] Embodiment 4

[0165] Figure 4A schematic diagram of an electronic device 410 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0166] like Figure 4 As shown, the electronic device 410 includes at least one processor 411, and a memory connected to the at least one processor 411 in communication, such as a read-only memory (ROM) 412, a random access memory (RAM) 413, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 411 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 412 or the computer program loaded from the storage unit 418 to the random access memory (RAM) 413. In the RAM 413, various programs and data required for the operation of the electronic device 410 can also be stored. The processor 411, the ROM 412, and the RAM 413 are connected to each other via a bus 414. An input / output (I / O) interface 415 is also connected to the bus 414.

[0167] Multiple components in the electronic device 410 are connected to the I / O interface 415, including: an input unit 416, such as a keyboard, a mouse, etc.; an output unit 417, such as various types of displays, speakers, etc.; a storage unit 418, such as a disk, an optical disk, etc.; and a communication unit 419, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 419 allows the electronic device 410 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0168] The processor 411 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 411 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 411 executes the various methods and processes described above, such as a method for evaluating the gas-bearing properties of a tight reservoir.

[0169] In some embodiments, the gas-bearing performance evaluation method for a tight reservoir may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 418. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 410 via the ROM 412 and / or the communication unit 419. When the computer program is loaded into the RAM 413 and executed by the processor 411, one or more steps of the gas-bearing performance evaluation method for a tight reservoir described above may be performed. Alternatively, in other embodiments, the processor 411 may be configured to execute the gas-bearing performance evaluation method for a tight reservoir by any other appropriate means (e.g., by means of firmware).

[0170] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0171] The computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable tight reservoir gas-bearing property evaluation device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0172] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0173] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0174] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0175] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0176] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0177] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for evaluating the gas content of a tight reservoir, characterized in that: The method comprises: Acquire control factor data of gas-bearing performance of the tight reservoir to be evaluated in the target area, and determine the gas-bearing characteristics of the tight reservoir to be evaluated according to the control factor data; wherein the control factor data of gas-bearing performance is data characterizing the conditions of the tight reservoir, and the control factor data includes hydrocarbon generation intensity, relative structural height, reservoir physical properties, and distance from the adjustment fault; Determine the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and the pre-trained gas-bearing performance prediction model; the gas-bearing performance evaluation result includes the evaluation level of oil and gas and low oil and gas; Wherein, the gas-bearing performance prediction model includes a plurality of K nearest neighbor sub-models, and the number of K nearest neighbor sub-models in the gas-bearing performance prediction model is determined based on the test results of a preset number of candidate gas-bearing performance prediction models; The training process of the gas content performance prediction model includes: Obtaining a set of K nearest neighbor submodel quantity values, a set of K nearest neighbor coefficient value values, and multiple sets of control factor data of gas-bearing properties of evaluated tight reservoirs; Determine at least two parameter combinations according to a set of K nearest neighbor submodel quantity values ​​and a set of K nearest neighbor coefficient value values; wherein each parameter combination includes a K nearest neighbor submodel quantity and a K nearest neighbor coefficient; According to the gas-bearing properties matched by each group of control factor data, each group of control factor data is divided into samples to determine the sample set matched by each parameter combination; Determine each parameter combination as the current parameter combination in sequence; Determine a test sample from samples that have not been tested in the sample set that matches the current parameter combination; According to the current parameter combination, based on the K nearest neighbor algorithm, the other samples in the sample set matching the current parameter combination except the test sample are trained to obtain a candidate gas-bearing performance prediction model matching the test sample; Determine the gas-bearing performance prediction result of the test sample in this iteration according to the test sample and the candidate gas-bearing performance prediction model matched by the test sample; Return to the execution of determining the test sample in the samples that have not been tested in the sample set that matches the current parameter combination, until there are no untested samples in the sample set that matches the current parameter combination, and determine the first gas-containing property quantity and the second gas-containing property quantity of each sample in the gas-containing property prediction result of each iteration; Determining a gas-containing property prediction result of each sample according to the first gas-containing property quantity and the second gas-containing property quantity; The gas-bearing property prediction result of each sample is compared with the gas-bearing property matched by each sample, and the prediction accuracy of the candidate gas-bearing property prediction model matched by the current parameter combination is determined according to the comparison result; Determining a target parameter combination according to the prediction accuracy of the candidate gas-bearing performance prediction model matched by each of the parameter combinations; The gas-bearing performance prediction model is determined based on the candidate gas-bearing performance prediction models matched by the target parameter combination.

2. The method according to claim 1, characterized in that The gas-containing properties include a first gas-containing property and a second gas-containing property; According to the gas-bearing properties matched by each group of control factor data, each group of control factor data is divided into samples to determine the sample set matched by each parameter combination, including: Determine a first sample set and a second sample set according to the gas-bearing properties matched by each group of control factor data; wherein the first sample set includes each group of control factor data whose gas-bearing properties are the first gas-bearing properties, and the second sample set includes each group of control factor data whose gas-bearing properties are the second gas-bearing properties; According to the first sample set and the second sample set, a sample set matching each parameter combination is determined.

3. The method according to claim 1, characterized in that The candidate gas-bearing property prediction model matched by the current parameter combination includes the K nearest neighbor sub-models of the number of K nearest neighbor sub-models in the current parameter combination; Determining the gas-bearing performance prediction result of the test sample in this iteration according to the test sample and the candidate gas-bearing performance prediction model matched by the test sample, including: The test samples matched by each K nearest neighbor sub-model are input into the K nearest neighbor sub-model to determine the gas-bearing performance prediction results of the test samples matched by each K nearest neighbor sub-model in this iteration.

4. A device for evaluating the gas content of a dense reservoir, characterized in that: include: A gas-bearing characteristic determination module is used to obtain control factor data of the gas-bearing performance of the tight reservoir to be evaluated in the target area, and determine the gas-bearing characteristics of the tight reservoir to be evaluated based on the control factor data; wherein the control factor data of the gas-bearing performance is data characterizing the conditions of the tight reservoir, and the control factor data includes hydrocarbon generation intensity, relative structural height, reservoir physical properties, and distance from the adjustment fault; An evaluation result determination module is used to determine the gas-bearing performance evaluation result of the tight reservoir to be evaluated according to the gas-bearing characteristics and the pre-trained gas-bearing performance prediction model; the gas-bearing performance evaluation result includes oil-gas and low oil-gas evaluation levels; Wherein, the gas-bearing performance prediction model includes a plurality of K nearest neighbor sub-models, and the number of K nearest neighbor sub-models in the gas-bearing performance prediction model is determined based on the test results of a preset number of candidate gas-bearing performance prediction models; The device also includes a model training module, wherein the model training module includes a parameter value acquisition unit, a parameter combination determination unit, a sample set determination unit, a target combination determination unit, and a prediction model determination unit; The parameter value acquisition unit is used to acquire the K nearest neighbor submodel quantity value set, the K nearest neighbor coefficient value set and multiple groups of control factor data of gas-bearing properties of evaluated tight reservoirs; The parameter combination determination unit is used to determine at least two parameter combinations according to the K nearest neighbor submodel quantity value set and the K nearest neighbor coefficient value set; wherein each parameter combination includes a K nearest neighbor submodel quantity and a K nearest neighbor coefficient; The sample set determination unit is used to divide each group of control factor data into samples according to the gas-bearing properties matched by each group of control factor data, and determine the sample set matched by each parameter combination; The target combination determination unit includes a current combination determination subunit, a test sample determination subunit, a candidate model determination subunit, a sample test subunit, a prediction result determination subunit and a test result determination subunit; The current combination determination subunit is used to determine each parameter combination as the current parameter combination in sequence; The test sample determination subunit is used to determine the test sample from the samples that have not been tested in the sample set that matches the current parameter combination; The candidate model determination subunit is used to train other samples in the sample set matched by the current parameter combination except the test sample based on the K nearest neighbor algorithm according to the current parameter combination, so as to obtain the candidate gas-bearing property prediction model matched by the test sample; The sample testing subunit is used to determine the gas-containing performance prediction result of the test sample in this iteration according to the test sample and the candidate gas-containing performance prediction model matched by the test sample; The prediction result determination subunit is used to return to determine the test sample in the samples that have not been tested in the sample set that matches the current parameter combination, until there are no untested samples in the sample set that matches the current parameter combination, and determine the first gas-containing performance quantity and the second gas-containing performance quantity of each sample in the gas-containing performance prediction result of each iteration; determine the gas-containing performance prediction result of each sample according to the first gas-containing performance quantity and the second gas-containing performance quantity; The test result determination subunit is specifically used to: compare the gas-bearing performance prediction result of each sample with the gas-bearing performance matched by each sample, and determine the prediction accuracy of the candidate gas-bearing performance prediction model matched by the current parameter combination according to the comparison result; The target combination determination unit is specifically used to: determine the target parameter combination according to the prediction accuracy of the candidate gas-bearing performance prediction model matched by each parameter combination; The prediction model determination unit is used to determine the gas-bearing performance prediction model according to the candidate gas-bearing performance prediction models matched by the target parameter combination.

5. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for evaluating the gas-containing properties of a tight reservoir according to any one of claims 1 to 3.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method for evaluating the gas-containing properties of a tight reservoir according to any one of claims 1 to 3 when executed.

Images