Method and device for evaluating helium resources in a compact reservoir

Abstract

The application discloses a kind of compact reservoir helium resource evaluation method and device, the method comprises: obtaining the single well data and fracture data of well in study area;Determine reservoir quality index;Determine fracture density index;According to fracture data, determine effective fracture opening index;Effective fracture opening index is used to characterize the fluid flow capacity of fracture;Reservoir quality index, fracture density index and effective fracture opening index are input into the prediction model trained in advance, and the output is helium fracture coupling index and helium content;According to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of study area, the reservoir is dynamically meshed using clustering algorithm;Using small face element cumulative volume method, according to single well data, determine the geological resources of each dynamic grid, can accurately depict the heterogeneity of reservoir, improve the accuracy and efficiency of compact reservoir resource, especially helium resource evaluation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of oil and gas associated resource exploration, and particularly relates to a method and device for evaluating helium resources in tight reservoirs. BACKGROUND

[0002] This section is intended to provide background or context to the present application embodiments. The description herein does not constitute admission of prior art.

[0003] At present, helium resources have become a scarce resource. Through the development of helium resource analysis, the helium content in the to-be-tested or to-be-drilled traps is predicted, the helium resource potential and distribution in key areas are found out, the favorable drilling targets are optimized, the new high-quality reserves are discovered, and the energy security is ensured, which has become an important research goal for those skilled in the art.

[0004] Those skilled in the art believe that helium and natural gas resources have similarities in migration and accumulation, and there are almost consistent trap accumulation elements. The helium system, like the natural gas system, needs source rock, primary migration and secondary migration, reservoir, trap and preservation condition and other accumulation elements. Although helium and natural gas accumulate in the same trap, it is believed from the comparison of the helium content in the natural gas of the study area and the distribution trend of helium in part of the gas reservoirs that there is no direct correlation between the distribution of natural gas reserves and helium reserves, but the helium content in the natural gas reservoir is a key parameter reflecting the helium and natural gas reserves, and also determines the size of helium reserves and resource abundance. However, due to the special accumulation characteristics of helium, the conventional natural gas resource favorable distribution area prediction method has been difficult to apply to the optimization of helium favorable enrichment area, and the mechanism of various geological factors controlling the accumulation and distribution of helium resources is not clear at present. In addition, the helium content in the natural gas is currently obtained by collecting gas samples from discovered gas fields or drilling wells, and using chromatograph or mass spectrometer for quantitative detection.

[0005] The current helium reserves and resource analysis method of tight sandstone gas includes interpolation method, static grid division and the like. However, the existing technology has the following defects:

[0006] Limitations of interpolation method: the traditional Kriging method and inverse distance weighted method ignore the physical mechanism of helium migration, and the prediction error is significant in the fracture development area, and it is difficult to depict the heterogeneity of the reservoir.

[0007] Static grid division: the traditional method divides the reservoir into static and uniform grids, and this uniform grid division method leads to insufficient resolution of high potential area (fracture dense zone) and missing of local enrichment area.

[0008] Single characterization method: the traditional tight reservoir analysis method only uses single factor or double factors of porosity and permeability for evaluation, and it is difficult to realize multi-parameter fusion such as fracture opening, connectivity and helium content change. SUMMARY

[0009] The embodiment of the present application provides a method for evaluating helium resources in a compact reservoir, which is used for accurately describing the heterogeneity of the reservoir and improving the accuracy and efficiency of the analysis of the helium resources in the compact reservoir, and the method comprises the following steps:

[0010] Obtain single-well data and fracture data of wells in a research area; the research area is a compact reservoir block;

[0011] Determine a reservoir quality index according to the single-well data;

[0012] Determine a fracture density index according to the single-well data and a preset weight; the fracture density index is used for representing the development degree of fractures in the reservoir;

[0013] Determine an effective fracture opening index according to the fracture data; the effective fracture opening index is used for representing the fluid flow capacity of the fractures;

[0014] Input the reservoir quality index, the fracture density index and the effective fracture opening index into a pre-trained prediction model, and output a helium fracture coupling index and a helium content; the prediction model is obtained by training an XGBoost machine learning model by using historical or simulated reservoir quality indexes, fracture density indexes and effective fracture opening indexes;

[0015] According to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the research area, perform dynamic grid division on the reservoir by using a clustering algorithm;

[0016] Determine the geological resource quantity of each dynamic grid according to the single-well data by using a small face element cumulative volume method.

[0017] The embodiment of the present application also provides a device for evaluating helium resources in a compact reservoir, which is used for accurately describing the heterogeneity of the reservoir and improving the accuracy and efficiency of the analysis of the helium resources in the compact reservoir, and the device comprises the following steps:

[0018] An acquisition module is configured to acquire single-well data and fracture data of wells in a research area; the research area is a compact reservoir block;

[0019] A reservoir quality index determination module is configured to determine a reservoir quality index according to the single-well data;

[0020] A fracture density index determination module is configured to determine a fracture density index according to the single-well data and a preset weight; the fracture density index is used for representing the development degree of fractures in the reservoir;

[0021] An effective fracture opening index determination module is configured to determine an effective fracture opening index according to the fracture data; the effective fracture opening index is used for representing the fluid flow capacity of the fractures;

[0022] The helium fracture coupling index determination module is configured to input the reservoir quality index, the fracture density index and the effective fracture opening index into a pre-trained prediction model, and output the helium fracture coupling index and the helium content.

[0023] The dynamic grid division module is configured to divide the reservoir into dynamic grids by using a clustering algorithm according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the study area.

[0024] The geological resource quantity determination module is configured to determine the geological resource quantity of each dynamic grid by using a small face element cumulative volume method according to the single-well data.

[0025] Compared with the technical scheme for evaluating the helium resources in the dense reservoir in the prior art, the single-well data and the fracture data of the wells in the study area are obtained, the study area is a dense reservoir block, the reservoir quality index is determined according to the single-well data, the fracture density index is determined according to the single-well data and a preset weight, the fracture density index is used to represent the fracture development degree of the reservoir, the effective fracture opening index is determined according to the fracture data, the effective fracture opening index is used to represent the fluid flow capacity of the fractures, the reservoir quality index, the fracture density index and the effective fracture opening index are input into a pre-trained prediction model, and the helium fracture coupling index and the helium content are output, the prediction model is obtained by training an XGBoost machine learning model by using the historical or simulated reservoir quality index, fracture density index and effective fracture opening index, the reservoir is divided into dynamic grids by using a clustering algorithm according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the study area, and the geological resource quantity of each dynamic grid is determined by using a small face element cumulative volume method according to the single-well data, so that the heterogeneity of the reservoir can be accurately described, and the accuracy and efficiency of the analysis of the resources in the dense reservoir, especially the helium resources, are improved. BRIEF DESCRIPTION OF DRAWINGS

[0026] In order to more clearly illustrate the technical scheme in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort. In the drawings:

[0027] Figure 1 The flowchart of the method for evaluating the helium resources in the dense reservoir provided in the embodiments of the present application;

[0028] Figure 2 The box plot of the thickness of the dense reservoir provided in the embodiments of the present application;

[0029] Figure 3 A helium fracture coupling index feature importance diagram provided in an embodiment of the present application;

[0030] Figure 4 A visual dynamic panel grid diagram provided in an embodiment of the present application;

[0031] Figure 5 A helium content distribution grid diagram provided in an embodiment of the present application;

[0032] Figure 6 A schematic diagram of a compact reservoir helium resource evaluation device provided in an embodiment of the present application;

[0033] Figure 7 A schematic diagram of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, further detailed description will be made to the embodiments of the present application in combination with the drawings. Herein, the illustrative embodiments of the present application and the description thereof are used to explain the present application, but not as a limitation to the present application.

[0035] The acquisition, storage, use, processing and the like of data in the technical scheme of the present application all conform to the relevant provisions of laws and regulations.

[0036] The term "and / or" in the present application is merely used to describe an associated relationship, which means that there can be three relationships, for example, A and / or B can mean that A exists alone, A and B exist together, and B exists alone. In addition, the term "at least one" in the present application means any one of a plurality of or any combination of at least two of a plurality of, for example, including at least one of A, B and C can mean including any one or more elements selected from the set consisting of A, B and C.

[0037] In the description of the present application, "including", "containing", "having", "comprising" and the like are all open terms, which means including but not limited to. The description of the terms "one embodiment", "one specific embodiment", "some embodiments", "for example" and the like means that the specific features, structures or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present application, the illustrative description of the above terms does not necessarily mean the same embodiment or example. Moreover, the specific features, structures or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. The order of the steps involved in the embodiments is used to illustrate the embodiments of the present application, and the order of the steps is not limited, which can be adjusted as needed.

[0038] As a kind of rare gas, helium has the following specialities in its accumulation mechanism:

[0039] 1. Weak source accumulation characteristics: Helium has a slow generation rate, a wide distribution of source rocks and a long accumulation period, and its sources include uranium-rich sedimentary layers and faulted basement;

[0040] 2. Complex transport system: Helium migration channels include faults, formation water and other gases;

[0041] 3. Coexistence dependence: Helium cannot accumulate alone and needs to rely on the coexistence of carrier gases such as nitrogen, carbon dioxide and methane.

[0042] Helium-containing gas fields (reservoirs) can be divided into the following types according to the coexistence relationship between helium and carrier gas: homogenous and homogenously stored gas fields (reservoirs); heterogenous and homogenously stored gas fields (reservoirs); heterogenous and heterogenously stored gas fields (reservoirs); homogenous and heterogenously stored gas fields (reservoirs); among them, heterogenous and homogenously stored gas fields (reservoirs) are the most widely distributed, covering conventional natural gas fields, unconventional tight gas reservoirs, volcanic rock gas reservoirs, non-hydrocarbon gas reservoirs (such as carbon dioxide, nitrogen), water-soluble gas reservoirs, helium-containing coal seam gas reservoirs and shale gas reservoirs, etc.

[0043] As an important natural gas resource, unconventional tight gas has the characteristics of low permeability, strong heterogeneity and dense lithology in helium-containing tight reservoirs. In tight reservoirs, helium reserves account for 40% of all discovered helium reserves, and the resource potential is huge, but helium resource evaluation and development are difficult. The core characteristics include low porosity and low permeability, near-source accumulation and strong heterogeneity, which are the key basis for helium resource evaluation and efficient development plan design.

[0044] Tight reservoirs have three characteristics in accumulation and occurrence: first, near-source accumulation, source and reservoir are integrated or adjacent, and natural gas usually comes from adjacent source rocks with short migration distance. Second, continuous accumulation without clear trap boundaries, gas is widely distributed in low-permeability reservoirs, forming a "continuous gas reservoir". Third, special occurrence state, high proportion of adsorbed gas, natural gas (especially methane) exists in adsorbed state on the surface of micropores, and needs to be produced by desorption under reduced pressure. Free gas and adsorbed gas coexist, and free gas may exist in fractures and larger pores, but overall, adsorbed gas is dominant (accounting for 30% to 60%). Fourth, helium enrichment particularity. After helium from the outside enters the tight gas reservoir, due to the tight and heterogeneous nature of the reservoir, the helium content and helium distribution also have heterogeneity, which brings adjustment for quantitative and level resource evaluation and favorable area optimization.

[0045] The current helium reserve and resource analysis method for tight sandstone gas has many defects, and it is urgent to develop an intelligent resource analysis method for helium-containing tight reservoirs that can integrate geological rules and data-driven models, dynamically optimize grid division and integrate multiple parameters.

[0046] In view of the problems in the prior art, the embodiments of the present application aim to provide a helium-bearing dense reservoir helium resource evaluation method based on the symbiotic relationship of helium carrier gas and the reservoir heterogeneity. It is particularly suitable for predicting the three-dimensional spatial distribution of helium content in low-porosity and low-permeability, strongly heterogeneous reservoirs and calculating the reserves or resource quantity in stages.

[0047] The embodiments of the present application aim to solve the problems of the limitations of existing interpolation methods, low resolution of traditional static grid division, and single dense reservoir characterization method. According to the helium accumulation characteristics of dense reservoirs and the symbiotic relationship with carrier gas, input single well data and fracture data, data preprocessing and missing value processing, reservoir heterogeneity evaluation RQI, fracture density index FDI evaluation, effective fracture opening index FEI, helium fracture coupling index HFCI for predicting helium content, dynamic cell grid division are carried out. On this basis, the helium geological resource quantity is calculated in stages, and the resource evaluation map is drawn. This method uses a multi-factor prediction model to accurately calculate the helium resource abundance of different target areas, which helps to analyze the helium resource potential and distribution, optimizes the favorable enrichment area, and ensures the development and use of helium resources. The technical scheme of the present application completely covers the whole process of data input, algorithm design and result output, and has both theoretical innovation and industrial practicality, and can be widely applied to helium resource evaluation of unconventional reservoirs such as tight sandstone gas reservoirs, tight carbonate gas reservoirs and shale gas reservoirs.

[0048] Figure 1 A flowchart of a dense reservoir helium resource evaluation method provided in the embodiments of the present application is shown in Figure 1 The method can include the following steps:

[0049] Step 101, obtaining single well data and fracture data of wells in the study area; the study area is a dense reservoir block;

[0050] Step 102, determining the reservoir quality index according to the single well data;

[0051] Step 103, determining the fracture density index according to the single well data and the preset weight; the fracture density index is used to characterize the fracture development degree of the reservoir;

[0052] Step 104, determining the effective fracture opening index according to the fracture data; the effective fracture opening index is used to characterize the fluid flow capacity of the fracture;

[0053] Step 105, inputting the reservoir quality index, the fracture density index and the effective fracture opening index into a pre-trained prediction model to output the helium fracture coupling index and the helium content; the prediction model is obtained by training an XGBoost machine learning model using historical or simulated reservoir quality index, fracture density index and effective fracture opening index;

[0054] Step 106, according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the study area, the clustering algorithm is used to divide the reservoir into dynamic grids;

[0055] Step 107, using the small face cumulative volume method, the geological resource of each dynamic grid is determined according to the single well data.

[0056] The embodiment of the present application provides a compact reservoir helium resource evaluation method, which can accurately depict the heterogeneity of the reservoir and improve the accuracy and efficiency of the analysis of the compact reservoir resources, especially the helium resources, through the above steps.

[0057] The embodiment of the present application is based on the symbiotic relationship between the helium source and the carrier gas helium, and combines the helium accumulation characteristics in the compact gas reservoir, so that the hierarchical calculation and three-dimensional visualization of the helium reserves or resource amount are realized through data cleaning and missing value processing, reservoir heterogeneity evaluation, helium fracture coupling evaluation and dynamic face grid division.

[0058] In one embodiment, the single well data includes one or any combination of the following: gas-bearing area, effective thickness of gas layer, porosity, effective porosity of gas layer, permeability, gas saturation, gas volume coefficient, helium content, acoustic time difference, density and natural gamma.

[0059] The single well data of the compact sandstone helium-containing block in the study area is input. The single well data can include: reservoir thickness (H, unit: m), porosity (φ, unit: %), effective porosity of gas layer, permeability (K, unit: mD), gas saturation (Sg, unit: %), gas volume coefficient (Bg, dimensionless), helium content (CHe, unit: %), acoustic time difference, density, natural gamma.

[0060] In one embodiment, the fracture data includes one or any combination of the following: fracture opening, fracture length, connectivity factor and total number of fractures; the connectivity factor represents the connectivity between fractures.

[0061] The fracture data of the compact sandstone helium-containing block in the study area is input. The fracture data can include: fracture density, fracture opening, fracture length, connectivity factor, and total number of fractures.

[0062] In one embodiment, after step 101, a data preprocessing step can also be included.

[0063] Firstly, the 3σ principle is used to eliminate abnormal data exceeding the mean value ± 3 times the standard deviation, which is suitable for the case that the data approximately obeys the normal distribution. The mean value μ and the standard deviation σ of the data are calculated, and the formula is as follows:

[0064]

[0065] Where x is the original data value; μ is the mean of the data; σ is the standard deviation of the data; i is the number of data points.

[0066] Secondly, determine the range of normal data: according to the 3σ principle, the normal data should fall within the range of ± 3 times the standard deviation, and the data points not within the range of [μ-3σ, μ+3σ] need to be removed.

[0067] Then the outliers and missing values can be processed:

[0068] The box plot method is used to identify outliers of reservoir thickness and gas saturation. The box plot consists of five key parameters: the first quartile Q1: the 25% quantile of data points. The third quartile Q3: the 75% quantile of data points. The interquartile range (IQR): IQR = Q3-Q1; Upper boundary: Q3+1.5×IQR; Lower boundary: Q1-1.5×IQR.

[0069] Q1, Q3, upper boundary and lower boundary are calculated for reservoir thickness and gas saturation data respectively; the box plot is used to visualize data distribution and mark outliers, and data points beyond the upper and lower boundaries are extracted.

[0070] The processing method is: first, if the outlier is data error or noise, it can be directly deleted. Second, replace the outlier with the median, homogeneous or other statistical value. Third, if the outlier is part of the real data (such as special geological phenomena), it can be retained and analyzed separately.

[0071] Linear interpolation, Kriging interpolation, and log transformation + Kriging interpolation are used to fill in missing values for well logging curves. For data with less missing values and smooth trend, linear interpolation can be used. For example, porosity is usually related to depth and lithology, and the data changes smoothly, so linear interpolation can be used. Gas saturation is usually closely related to depth and reservoir properties, with non-uniform distribution and spatial correlation characteristics, so Kriging interpolation can be used. Permeability usually follows a lognormal distribution, so take the logarithm of the data first, and then perform Kriging interpolation.

[0072] In one embodiment, the helium content can not be preprocessed, and the missing value filling can be performed in the step of outputting the helium fracture coupling index. That is, when there is a missing value of helium content in single well data, the reservoir quality index, the fracture density index, and the effective fracture opening index are input into a pre-trained prediction model to output the helium fracture coupling index and the helium content, which can include: inputting the reservoir quality index, the fracture density index, and the effective fracture opening index into a pre-trained prediction model to output the helium fracture coupling index and the missing value of the helium content.

[0073] In one embodiment, according to single well data, a reservoir quality index is determined, including: according to porosity, permeability and a preset conversion coefficient, determining the reservoir quality index.

[0074] Reservoir heterogeneity refers to the non-uniformity of lithology, physical properties (such as porosity and permeability) and fluid distribution in space, which has an important influence on helium content and resource distribution. Reservoirs with strong heterogeneity may lead to complex helium flow, large changes in helium content and large differences in resource abundance, so evaluating reservoir heterogeneity is the key to optimizing helium development schemes. Reservoir quality index (RQI) is an important parameter for evaluating the physical properties of dense reservoir rocks and their influence on fluid flow. RQI combines permeability and porosity, which can more comprehensively evaluate the fluid conductivity of the reservoir. The higher the RQI, the stronger the fluid flow capacity of the reservoir, and the greater the development potential. According to the RQI value, the reservoir can be divided into high-quality reservoirs and low-quality reservoirs, providing a scientific basis for the development of dense reservoir gas reservoirs.

[0075] The calculation formula is:

[0076]

[0077] Where K is the permeability (unit: millidarcy, mD); φ is the porosity (unit: decimal, dimensionless); 0.0314 is a preset conversion coefficient for unit conversion to microns (μm).

[0078] In one embodiment, according to single well data and a preset weight, a fracture density index is determined, including: according to acoustic travel time and a corresponding first preset weight, density and a corresponding second preset weight, natural gamma and a preset third preset weight, determining the fracture density index.

[0079] Fracture density index (FDI) is an important parameter for evaluating the degree of fracture development in a reservoir, and is usually used for the evaluation of fractured reservoirs. The calculation method of FDI can be adjusted according to the specific research area and data characteristics. The common method is based on logging data, and the calculation formula is as follows:

[0080] FDI = w1 × acoustic travel time + w2 × density + w3 × natural gamma

[0081] Where w1 is the first preset weight; w2 is the second preset weight; w3 is the third preset weight, determined according to actual data and research objectives. Acoustic travel time, density and natural gamma need to be normalized, for example, using Min-Max normalization.

[0082] In one embodiment, according to fracture data, an effective fracture opening index is determined, including: according to fracture opening, fracture length, connectivity factor and total number of fractures, determining the effective fracture opening index.

[0083] Effective Fracture Aperuture Index (FEI) is a parameter representing the fluid flow capacity in the fracture system, which is usually calculated based on the fracture aperture, length and connectivity, and the formula is:

[0084]

[0085] where b i is the aperture of the i-th fracture (unit, mm); L i is the length of the i-th fracture (unit, m); c i is the connectivity factor of the i-th fracture (0 to 1, 1 represents complete connectivity); and n is the total number of fractures.

[0086] In one embodiment, in step 105, a helium fracture coupling index (HFCI) is needed to predict the helium content, and the HFCI is used to predict the helium content. The HFCI represents the nonlinear correlation strength between fracture characteristics and helium content, and the XGBoost machine learning model is used, the input is FDI, FEI and RQI, the output is helium fracture coupling index and helium content (C He ), and the missing helium content is supplemented.

[0087] The performance indicators of the output model include mean squared error (MSE): which measures the difference between the predicted value and the true value. The coefficient of determination (R2) measures the ability of the model to explain the variation of the data.

[0088] The data set is divided into features (X) and target variables (Y), and standardized, and XGBoost is used for regression to generate a feature importance chart.

[0089] In one embodiment, according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the study area, a clustering algorithm is used to dynamically divide the reservoir into a plurality of dynamic surface element grids, including: dividing the reservoir into a plurality of dynamic surface element grids according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the study area; wherein different dynamic surface element grids have different helium fracture coupling indexes; and using a K-Means clustering algorithm to cluster the plurality of dynamic surface element grids to determine the final dynamic surface element grid.

[0090] In order to represent the heterogeneity of helium distribution in the dense reservoir, the reservoir is divided into a plurality of dynamic surface element grids according to the helium fracture coupling index value, and each grid has a different helium fracture coupling index value. This method combines learning (K-Means clustering algorithm) and visualization technology, and provides a scientific basis for the development of the dense reservoir.

[0091] First, input the spatial coordinates (X, Y, Z) of the reservoir and the HFCI value of each coordinate point. According to the HFCI value, the reservoir is divided into multiple dynamic cell grids, and the K-Means clustering algorithm is used to partition the HFCI.

[0092] The reservoir is divided into several dynamic cell grids, the K-Means clustering algorithm is used to cluster the coordinate points, the average value of HFCI of each grid is calculated, and the output result is obtained.

[0093] In one embodiment, it further includes: taking the average value of the helium fracture coupling index of the multiple dynamic cell grids included in each final dynamic cell grid, and determining the average value as the helium fracture coupling index of the final dynamic cell grid; using a scatter plot to visualize the dynamic cell grid and the corresponding helium fracture coupling index, and / or the final dynamic cell grid and the corresponding helium fracture coupling index.

[0094] The dynamic cell grid division result is visualized using a three-dimensional scatter plot, each grid is represented by a different color, and the HFCI average value of each grid is displayed in the legend, which reflects the degree of helium fracture coupling in the area.

[0095] In one embodiment, the small cell cumulative volume method is used to determine the geological resource quantity of each dynamic grid based on single well data, including: using the small cell cumulative volume method to determine the geological resource quantity of each dynamic grid based on the gas-bearing area, effective gas layer thickness, effective gas layer porosity, gas saturation, gas volume coefficient and helium content.

[0096] The reservoir in the study area is divided into several grid units, and the reserve calculation parameters of each grid are determined as follows: the small cell cumulative volume method is used to calculate the geological resource quantity (GIIP), and the formula is as follows:

[0097] GIIP = Σ (A x H x φ' x Sg / Bgi x C He );

[0098] Where: GIIP is the helium reserve or resource quantity (10 8 m 3 ); A is the gas-bearing area (km 2 ); H is the effective gas layer thickness (m); φ' is the effective gas layer porosity (decimal, %); S g is the gas saturation (decimal, %); B gi is the gas volume coefficient (dimensionless); C He is the (grid) helium content (decimal, %).

[0099] Through dynamic cell grid division and volume method calculation, the distribution and geological resource quantity of helium in the reservoir can be quantified, providing a scientific basis for the coordinated development of natural gas and helium.

[0100] The following is described through a specific embodiment.

[0101] A certain research area is a dense sand helium-containing block with an area of 10.8km 2 . Well 1 obtains industrial gas flow in the target layer, with daily gas production of 109,000 square meters, and the measured helium content is 0.079%-0.098%, which confirms that the area is a fault-controlled dense sand helium-containing reservoir.

[0102] S1, input single well data and fracture data.

[0103] Single well data: reservoir thickness (H, unit: m), porosity (φ, unit: %), permeability (K, unit: mD), gas saturation (Sg, unit: %), gas volume coefficient (Bg, dimensionless), helium content (CHe, unit: %), acoustic time, density, natural gamma.

[0104] Fracture data, including fracture density 0.2-5 / m 2 , fracture length, connectivity factor 1.5, total number of fractures 25, helium content 0.05%-0.6%.

[0105] S2, pre-process the data, first use the 3σ principle to eliminate abnormal data exceeding ±3 times the standard deviation of the mean. Among them, 5 gas volume coefficients are obtained by measurement, which are 1.5, 0.2, 0.85, 0.78, 0.82, μ=0.83, σ=0.4606, then the abnormal values are 1.5 and 0.2, which can be directly deleted.

[0106] S3, process outliers and missing values.

[0107] Figure 2 The dense reservoir thickness box plot provided in the embodiment of the application is shown in Figure 2 , the upper quartile, lower quartile, median, minimum value and maximum value of the reservoir thickness and gas saturation data are calculated respectively; the box plot is used to visualize the data distribution, and the outliers are marked, and the data points exceeding the minimum value and the maximum value are extracted. Figure 2 The "whisker" represents the data within 1.5 times the interquartile range (the difference between the 1st and 3rd quartiles), which can also be used to display the highest and lowest points in the data. The identification result: the box plot shows that the outliers of the reservoir thickness are 120m, and the outliers of the gas saturation are 1.20, as shown in Table 1.

[0108] Table 1

[0109] Serial number Pre-treatment thickness m Pre-treatment thickness m Pre-treatment gas porosity Post-treatment gas porosity 1 120 0.65 0.65 2 45.2 45.2 1.20 3 55.6 55.6 0.58 0.58 4 48.9 48.9 0.62 0.62 5 52.3 52.3 0.59 0.59

[0110] The missing values of the logging curves are filled by linear interpolation, Kriging interpolation, and log transformation + Kriging interpolation.

[0111] S4, reservoir heterogeneity RQI evaluation.

[0112] The RQI calculation formula is:

[0113]

[0114] wherein K is the permeability (unit: millidarcy, mD); φ is the porosity (unit: decimal, dimensionless); 0.0314 is a conversion coefficient, used to unify the unit to microns (μm).

[0115] The calculation results of the research area are shown in Table 2.

[0116] Table 2

[0117] Depth Porosity (%) Permeability (mD) RQI 3601.2 0.09 8 0.30 3601.7 0.089 4.6 0.23 3602.2 0.088 6.7 0.27 3602.7 0.077 3.6 0.21 3603.2 0.081 0.5 0.08 3603.7 0.065 5.6 0.29 3604.2 0.086 4.2 0.22 3604.7 0.073 1.4 0.14 3605.2 0.082 0.6 0.08 3605.7 0.083 0.6 0.08

[0118] S5. Fracture density index FDI evaluation.

[0119] The common calculation method of FDI is to use acoustic travel time, density and natural gamma logging parameters, which need to be normalized, for example, using Min-Max normalization. The implementation process is as follows:

[0120] First, use Min-Max normalization to map the acoustic travel time, density and natural gamma data to the [0, 1] interval. The method is described above in the Min-Max normalization method.

[0121] Second, calculate FDI: assuming that w1, w2 and w3 are 0.4, 0.3 and 0.3 respectively, then the FDI calculation formula is:

[0122] FDI = 0.4 x acoustic travel time + 0.3 x density + 0.3 x natural gamma;

[0123]

[0124] Then FDI = 0.33 x 0.4 + 0.67 x 0.3 + 0.67 x 0.3 = 0.534.

[0125] The calculated FDI values are shown in Table 3.

[0126] Table 3

[0127] Depth Acoustic time difference Density (g / cm3) Natural gamma Fracture density FDI 3601.2 75 2.45 60 5 0.534 3602.2 80 2.40 55 8 0.600 3603.2 70 2.50 65 3 0.467 3604.2 85 2.35 50 10 0.667 3605.2 78 2.42 58 6 0.567

[0128] S6. Effective fracture opening index FEI evaluation.

[0129] For the fracture summation at a depth of 3601.2 m, the FEI calculation

[0130]

[0131] Similarly, the FEI of the other 4 depth points can be obtained, as shown in Table 4 below.

[0132] The total effective fracture opening index of the 5 fractures is 0.05+0.1536+0.0126+0.25+0.0734=0.5396. It shows that the effective fracture opening index of the fracture system, in which the contribution of fracture 4 is the largest because of the largest opening and full opening, and the contribution of fracture 3 is the smallest because of the smallest opening and lower connectivity.

[0133] Table 4

[0134] Depth m Fracture opening (mm) Fracture length (m) Connectivity factor Effective fracture opening index FEI 3601.2 0.5 2.0 0.8 0.0500 3602.2 0.8 3.0 0.9 0.1536 3603.2 0.3 1.5 0.7 0.0126 3604.2 1.0 4.0 1.0 0.2500 3605.2 0.6 2.5 0.85 0.0734

[0135] S7. Predicting helium content using helium-fracture coupling index HFCI.

[0136] Using the XGBoost machine learning model, the input is FDI, FEI, RQI, and the output is helium-fracture coupling index (HFCI) and helium content (C He ), and supplement the missing helium content.

[0137] The performance indicators of the output model include MSE: measure the difference between the predicted value and the true value. R 2 Measure the ability of the model to explain the variation of the data.

[0138] First, generate a simulated data set containing 1000 samples, the features include FDI, FCI, RQI, and the target variable is HFCI, as shown in Table 5 below.

[0139] Table 5

[0140] Depth m FDI FEI RQI HFCI C He ]]> 3601.2 0.3745 0.7319 0.5987 0.8123 0.08123 3602.2 0.1560 0.0581 0.8662 0.2345 Supplement 3603.2 0.6011 0.6011 0.7081 0.9012 0.09012 3604.2 0.7081 0.1562 0.9507 0.7890 0.07890 3605.2 0.8862 0.9507 0.3745 0.6543 Supplement

[0141] Figure 3 For the feature importance diagram of the helium-fracture coupling index provided in the embodiments of the present application, the data set is divided into features (X) and target variables (Y), and standardized, and regression is performed using XGBoost, and the feature importance diagram is generated as shown in Figure 3 .

[0142] FDI: the highest importance, indicating that the fracture density has the greatest impact on the degree of control of helium distribution.

[0143] RQI: moderate importance, indicating that the reservoir quality parameter also has a certain influence on the helium distribution.

[0144] FEI: Lowest importance, indicating that the effective fracture aperture has little impact on helium distribution. Results analysis: MSE is 0.0001, R² is 0.9783, indicating high model prediction accuracy.

[0145] Mean_HFCI: Indicates the mean value of the helium crack coupling index.

[0146] Grid: Different dynamic face grids.

[0147] S8. Dynamic surface mesh generation.

[0148] The reservoir is divided into multiple dynamic surface grids based on the helium fracture coupling index (HFCI) value, with each grid having a different HFCI value. This method combines machine learning (K-Means clustering algorithm) and visualization techniques, providing a scientific basis for the development of tight reservoirs.

[0149] First, input the spatial coordinates (X, Y, Z) of the reservoir and the HFCI value for each coordinate point. Based on the HFCI values, the reservoir is divided into multiple dynamic element grids, and the K-Means clustering algorithm is used to partition the HFCI.

[0150] The reservoir in the study area was divided into 5 dynamic surface grids. The K-Means clustering algorithm was used to cluster the coordinate points, and the average HFCI of each grid was calculated. The output results are shown in Table 6.

[0151] Table 6

[0152]

[0153]

[0154] Figure 4 The visualized dynamic surface mesh diagram provided in the embodiments of the present invention, such as Figure 4 As shown, the dynamic surface mesh is visualized: the dynamic surface mesh division results (Grid0 to Grid4) are visualized using a 3D scatter plot. Each grid is represented by a different color. The legend shows the average HFCI of each grid, which reflects the degree of helium fracture coupling in the region.

[0155] S9. Calculation of Helium Geological Resources by Level.

[0156] Figure 5 The helium content distribution grid diagram provided in the embodiments of the present invention is as follows: Figure 5 As shown, the reservoir in the study area is divided into 5 grids, and the parameters of each grid are shown in Table 7 below. The gas volume factor and the geological resources of each grid are calculated.

[0157] Table 7

[0158]

[0159] The geological resource quantity (GIIP) is calculated by using the small facet cumulative volume method, and the first grid resource quantity is calculated as follows:

[0160] GIIP5=A*H*phi'*Sg / Bgi*C He

[0161] =3.0*10*0.05*0.60 / 0.0041*0.08%

[0162] =17.56098

[0163] The total geological resource quantity GIIP of the five units is Σ (A*H*phi'*Sg / Bgi*C He )

[0164] =17.56098+24.04286+18.8160+10.29767+27.10400

[0165] =97.82151

[0166] Wherein, GIIP5 is the fifth grid helium reserve or resource quantity (10 4 m 3 );A is the gas-bearing area (km 2 );H is the effective thickness of the gas layer (m);phi' is the effective porosity of the gas layer (decimal, %);S g is the gas saturation (decimal, %);B gi is the gas volume coefficient (dimensionless quantity);C He is the (grid) helium content (decimal, %)。

[0167] The results show that the total helium geological resource quantity of the five units is 97.82151*104, wherein the helium reserve of the fifth grid is the largest, and the helium reserve of the fourth grid is the smallest. Through dynamic facet grid division and volume method calculation, the distribution and geological resource quantity of helium in the reservoir can be quantified, and a scientific basis is provided for the coordinated development of natural gas and helium.

[0168] The output results of the embodiment of the present application are: a helium resource distribution map and a reserve classification report.

[0169] The exploration implementation effect of the embodiment of the present application is: accurately positioning the helium enrichment target area, reducing the drilling risk by more than 30%, as shown in the following table 8.

[0170] The development implementation effect of the embodiment of the present application is: the average helium reserve of a single well in the I type area is improved by 3 times, which proves the effectiveness of the fracture index.

[0171] Table 8

[0172] Serial number Index Traditional method This method 1 Data fusion Single fracture density index Multi-parameter coupling + machine learning 2 Helium content prediction and reserve calculation error 23％ 6.5％ 3 Calculation time 10 hours 18 minutes 4 Rich helium area prediction recognition rate Qualitative description 94％ 5 Practical effect Qualitative description Reduce risk by more than 30%

[0173] The embodiment of the present application also provides a device for evaluating helium resources in a tight reservoir, which has a similar principle to the method for evaluating helium resources in a tight reservoir, and will not be described here.

[0174] Figure 6 A schematic diagram of the device for evaluating helium resources in a tight reservoir provided in the embodiment of the present application is shown in FIG. 1, which can include: Figure 5

[0175] The acquisition module 601 is configured to acquire single-well data and fracture data of a well in a study area; the study area is a tight reservoir block.

[0176] The reservoir quality index determination module 602 is configured to determine a reservoir quality index according to the single-well data.

[0177] The fracture density index determination module 603 is configured to determine a fracture density index according to the single-well data and a preset weight; the fracture density index is used to represent a fracture development degree of the reservoir.

[0178] The effective fracture opening index determination module 604 is configured to determine an effective fracture opening index according to the fracture data; the effective fracture opening index is used to represent a fluid flow capacity of the fracture.

[0179] The helium fracture coupling index determination module 605 is configured to input the reservoir quality index, the fracture density index and the effective fracture opening index into a pre-trained prediction model, and output a helium fracture coupling index and a helium content; the prediction model is obtained by training an XGBoost machine learning model using historical or simulated reservoir quality indexes, fracture density indexes and effective fracture opening indexes.

[0180] The dynamic grid division module 606 is configured to perform dynamic grid division on the reservoir by using a clustering algorithm according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the study area.

[0181] The geological resource quantity determination module 607 is configured to determine a geological resource quantity of each dynamic grid according to the single-well data by using a small-face-element cumulative volume method.

[0182] In one embodiment, the single-well data includes one or any combination of a gas-bearing area, an effective thickness of a gas layer, porosity, effective porosity of a gas layer, permeability, gas saturation, a gas volume coefficient, a helium content, acoustic time difference, density and natural gamma.

[0183] In one embodiment, the helium fracture coupling index determination module 605 is specifically configured to:

[0184] ​The reservoir quality index, the fracture density index and the effective fracture opening index are input into the pre-trained prediction model, and the helium fracture coupling index and the helium content missing value are output.

[0185] In one embodiment, the reservoir quality index determination module 602 is specifically configured to:

[0186] The reservoir quality index is determined according to the porosity, the permeability and the preset conversion coefficient.

[0187] In one embodiment, the fracture density index determination module 603 is specifically configured to:

[0188] The fracture density index is determined according to the acoustic travel time and the corresponding first preset weight, the density and the corresponding second preset weight, and the natural gamma and the preset third preset weight.

[0189] In one embodiment, the fracture data includes one or any combination of the fracture opening, the fracture length, the connectivity factor and the total number of fractures, and the connectivity factor represents the connectivity between fractures.

[0190] In one embodiment, the effective fracture opening index determination module 604 is specifically configured to:

[0191] The effective fracture opening index is determined according to the fracture opening, the fracture length, the connectivity factor and the total number of fractures.

[0192] In one embodiment, the dynamic grid division module 606 is specifically configured to:

[0193] The reservoir is divided into a plurality of dynamic surface element grids according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the study area, wherein different dynamic surface element grids have different helium fracture coupling indexes.

[0194] The K-Means clustering algorithm is used to cluster the plurality of dynamic surface element grids to determine the final dynamic surface element grid.

[0195] In one embodiment, the dense reservoir helium resource evaluation device further includes a visualization module configured to:

[0196] The helium fracture coupling indexes of the plurality of dynamic surface element grids included in each final dynamic surface element grid are averaged to determine the helium fracture coupling index of the final dynamic surface element grid.

[0197] The dynamic surface element grids and the corresponding helium fracture coupling indexes, and / or the final dynamic surface element grids and the corresponding helium fracture coupling indexes are visualized using a scatter plot.

[0198] In one embodiment, the geological resource quantity determination module 607 is specifically configured to:

[0199] The small facet cumulative volume method is used to determine the geological resource of each dynamic grid according to the gas-bearing area, effective thickness of gas layer, effective porosity of gas layer, gas saturation, gas volume coefficient and helium content.

[0200] Compared with the technical scheme for evaluating helium resources in a tight reservoir in the prior art, the single well data and fracture data of the wells in the research area are obtained, the research area is a tight reservoir block, the reservoir quality index is determined according to the single well data, the fracture density index is determined according to the single well data and a preset weight, the fracture density index is used to represent the fracture development degree of the reservoir, the effective fracture opening index is determined according to the fracture data, the effective fracture opening index is used to represent the fluid flow capacity of the fracture, the reservoir quality index, the fracture density index and the effective fracture opening index are input into a pre-trained prediction model to output the helium fracture coupling index and the helium content, the prediction model is obtained by training an XGBoost machine learning model using historical or simulated reservoir quality index, fracture density index and effective fracture opening index, the reservoir is dynamically gridded by using a clustering algorithm according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the research area, and the geological resource of each dynamic grid is determined according to the single well data by using the small facet cumulative volume method, so that the heterogeneity of the reservoir can be accurately described, and the accuracy and efficiency of the analysis of the resources in the tight reservoir, in particular the helium resources, are improved.

[0201] The embodiments of the present application include the following technical features that are different from the prior art:

[0202] (1) Fine zoning:

[0203] Traditional methods usually divide the reservoir into static and uniform grids, while the dynamic facet grid dynamically adjusts the grid size and shape according to the reservoir characteristics (such as fracture density, connectivity, helium fracture coupling index, etc.), which is more consistent with the heterogeneity of the reservoir.

[0204] (2) Data-driven:

[0205] The machine learning algorithm (such as K-Means clustering) is used to divide the reservoir, which can better reflect the actual characteristics of the reservoir and improve the scientificity and accuracy of the zoning.

[0206] (3) Multi-parameter fusion:

[0207] The dynamic facet grid division not only considers a single parameter (such as porosity or gas saturation), but also integrates multiple geological and engineering parameters (such as fracture opening, connectivity, helium content, etc.), making the zoning result more representative.

[0208] By implementing the present application, the following effects can be achieved:

[0209] 1. Precision improvement: helium content prediction mean absolute error (MAE) ≤0.05%, reserve calculation error <8% (traditional method error >20%).

[0210] 2. Efficiency optimization: calculation time is shortened from traditional hours to 15 minutes, supporting real-time dynamic update.

[0211] 3. Industrial value: three-dimensional helium distribution map and reserve / resource quantity classification map are generated, and high-abundance target area (He%>0.1%) is accurately located, which can be integrated into geological software to directly output exploration and development scheme.

[0212] The embodiment of the present application also provides a computer device, Figure 7 The computer device 700 includes a memory 710, a processor 720, and a computer program 730 stored in the memory 710 and capable of running on the processor 720, and the processor 720 implements the above-mentioned compact reservoir helium resource evaluation method when executing the computer program 730.

[0213] The embodiment of the present application also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above-mentioned compact reservoir helium resource evaluation method.

[0214] The embodiment of the present application also provides a computer program product, which includes a computer program, and the computer program is executed by a processor to implement the above-mentioned compact reservoir helium resource evaluation method.

[0215] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can adopt a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt a computer program product in the form of being implemented on one or more computer usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer usable program code.

[0216] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of the flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a machine that implements the flowcharts and / or block diagrams. Figure oneone or more processes and / or blocks Figure one an apparatus for performing the functions specified in the flowchart or multiple flows and / or blocks.

[0217] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flowchart or multiple flows and / or blocks. Figure one one or more processes and / or blocks Figure one an apparatus for performing the functions specified in the flowchart or multiple flows and / or blocks.

[0218] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flowchart or multiple flows and / or blocks. Figure one one or more processes and / or blocks Figure one an apparatus for performing the functions specified in the flowchart or multiple flows and / or blocks.

[0219] The above-described specific embodiments, the purpose, technical solutions and beneficial effects of the present application are further described in detail, it should be understood that the above-described is only a specific embodiment of the present application, and is not used to limit the protection scope of the present application, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A method for evaluating helium resources in a compact reservoir, characterized by, The application relates to a method for determining a helium gas fracture coupling index in a dense reservoir block. The method comprises the following steps: acquiring single-well data and fracture data of wells in a research area; the research area is a dense reservoir block; determining a reservoir quality index according to the single-well data; determining a fracture density index according to the single-well data and a preset weight; the fracture density index is used for representing a fracture development degree of the reservoir; determining an effective fracture opening index according to the fracture data; the effective fracture opening index is used for representing a fluid flow capacity of the fracture; inputting the reservoir quality index, the fracture density index and the effective fracture opening index into a pre-trained prediction model to output a helium gas fracture coupling index and a helium gas content; the prediction model is obtained by training an XGBoost machine learning model by using historical or simulated reservoir quality indexes, fracture density indexes and effective fracture opening indexes; performing dynamic grid division on the reservoir by using a clustering algorithm according to the helium gas fracture coupling index corresponding to each spatial coordinate point in the reservoir of the research area; 2. The method of claim 1, wherein, determining a geological resource quantity of each dynamic grid according to the single-well data by using a small face element cumulative volume method.

3. The method of claim 2, wherein, The single-well data comprises one or any combination of the following: a gas-bearing area, a gas layer effective thickness, a porosity, a gas layer effective porosity, a permeability, a gas saturation, a gas volume coefficient, a helium gas content, an acoustic time difference, a density and a natural gamma. When the single-well data has a missing value of the helium gas content, the method comprises the following steps:

4. The method of claim 2, wherein, inputting the reservoir quality index, the fracture density index and the effective fracture opening index into the pre-trained prediction model to output the helium gas fracture coupling index and the missing value of the helium gas content. The method for determining the reservoir quality index according to the single-well data comprises the following steps:

5. The method of claim 2, wherein, determining the reservoir quality index according to the porosity, the permeability and a preset conversion coefficient. The method for determining the fracture density index according to the single-well data and the preset weight comprises the following steps:

6. The method of claim 1, wherein, determining the fracture density index according to the acoustic time difference and a corresponding first preset weight, the density and a corresponding second preset weight, the natural gamma and a preset third preset weight.

7. The method of claim 6, wherein, The fracture data comprises one or any combination of the following: a fracture opening, a fracture length, a connectivity factor and a total number of fractures; the connectivity factor represents connectivity between fractures. The method for determining the effective fracture opening index according to the fracture data comprises the following steps:

8. The method of claim 1, wherein, determining the effective fracture opening index according to the fracture opening, the fracture length, the connectivity factor and the total number of fractures. The method for performing dynamic grid division on the reservoir by using the clustering algorithm according to the helium gas fracture coupling index corresponding to each spatial coordinate point in the reservoir of the research area comprises the following steps: dividing the reservoir into a plurality of dynamic face element grids according to the helium gas fracture coupling index corresponding to each spatial coordinate point in the reservoir of the research area; different dynamic face element grids have different helium gas fracture coupling indexes; 9. The method of claim 8, wherein, determining a final dynamic face element grid by clustering the plurality of dynamic face element grids by using a K-Means clustering algorithm. The method further comprises the following steps: calculating an average value of the helium gas fracture coupling indexes of the plurality of dynamic face element grids included in each final dynamic face element grid, and determining the calculated result as the helium gas fracture coupling index of the final dynamic face element grid. The dynamic panel grid and the corresponding helium fracture coupling index, and / or the final dynamic panel grid and the corresponding helium fracture coupling index are visualized using a scatter plot.

10. The method of claim 2, wherein, The geological resource of each dynamic grid is determined according to the single-well data by using a small-panel cumulative volume method, including: The geological resource of each dynamic grid is determined according to the gas-bearing area, effective thickness of gas layer, effective porosity of gas layer, gas saturation, gas volume coefficient and helium content by using a small-panel cumulative volume method.

11. A device for evaluating helium resources in a compact reservoir, characterized in that it comprises: The method comprises the following steps: An acquisition module is configured to acquire single-well data and fracture data of a well in a research area; the research area is a dense reservoir block; A reservoir quality index determination module is configured to determine a reservoir quality index according to the single-well data; A fracture density index determination module is configured to determine a fracture density index according to the single-well data and a preset weight; the fracture density index is used to represent the development degree of fractures in the reservoir; An effective fracture opening index determination module is configured to determine an effective fracture opening index according to the fracture data; the effective fracture opening index is used to represent the fluid flow capacity of the fractures; A helium fracture coupling index determination module is configured to input the reservoir quality index, the fracture density index and the effective fracture opening index into a pre-trained prediction model to output a helium fracture coupling index and a helium content; the prediction model is obtained by training an XGBoost machine learning model using historical or simulated reservoir quality indexes, fracture density indexes and effective fracture opening indexes; A dynamic grid division module is configured to divide the reservoir into dynamic grids by using a clustering algorithm according to the helium fracture coupling index corresponding to each spatial coordinate point in the reservoir of the research area; A geological resource determination module is configured to determine the geological resource of each dynamic grid according to the single-well data by using a small-panel cumulative volume method.

12. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the method of any one of claims 1 to 10 when executing the computer program.

13. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to implement the method of any one of claims 1 to 10.

14. A computer program product, characterised in that, The computer program product comprises a computer program, and the computer program is executed by the processor to implement the method of any one of claims 1 to 10.

Images