Water production probability prediction method for tight gas sandstone gas reservoir

Abstract

The invention provides a water production probability prediction method for a tight gas sandstone gas reservoir, and relates to the technical field of oil and gas resource exploration, and the method comprises the steps: obtaining the pore structure data of the gas reservoir based on a mercury injection test, and the pore structure data comprises the porosity, the permeability, the displacement pressure, the median pressure and the pore throat structure; classifying and summarizing the gas reservoirs according to the pore structure data through a clustering analysis algorithm to obtain a plurality of gas reservoir groups; matching the sub-water production prediction model corresponding to the gas reservoir group, and predicting the sub-water production probability of the gas reservoir group; and calculating the total water production probability of the gas reservoir according to the sub-water production probability based on a weighted average algorithm. According to the method, the water production condition of the tight gas sandstone gas reservoir can be objectively, scientifically and effectively predicted, the risk of oil-gas exploration is reduced, the success rate of oil-gas drilling is increased, and the method has wide applicability.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas resource exploration, and particularly to a method for predicting the water production probability of a tight gas sandstone gas reservoir. Background Art

[0002] As an important field of oil and gas exploration and development, the tight gas sandstone gas reservoir has unique reservoir characteristics, such as low porosity, low permeability, and complex gas-water distribution, which bring great challenges to the effective development of oil and gas resources. Such reservoirs play an increasingly important role in the global energy structure. Especially in some areas with complex geological structures and diverse sedimentary processes, such as the Suxi area, tight sandstone gas reservoirs are often accompanied by significant water production. However, water production not only affects the production and recovery rate of oil and gas wells, but also may increase the production cost and environmental risk. Therefore, accurately predicting the water production probability of a tight gas sandstone gas reservoir is of great significance for optimizing the exploitation plan and improving the utilization rate of oil and gas resources.

[0003] The tight gas sandstone gas reservoir has characteristics such as low porosity and low permeability in geology. At present, the variation law of complex gas-water distribution has not been solved at the microscopic mechanism, which leads to difficulties in quantitative prediction and comprehensive evaluation of the tight gas sandstone gas reservoir, seriously affecting the exploration and development effects of such gas reservoirs. And the existing technology cannot accurately predict the water production situation of the tight gas sandstone gas reservoir, which may greatly reduce the success rate and economic benefits of oil and gas exploration.

[0004] Therefore, it is necessary to provide a method for predicting the water production probability of a tight gas sandstone gas reservoir to solve the above technical problems. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a method for predicting the water production probability of a tight gas sandstone gas reservoir to solve the problems that the gas-water distribution of the tight gas sandstone gas reservoir is complex, it is difficult to carry out scientific and objective quantitative prediction and effective comprehensive evaluation, and at the same time, the water production situation of the tight gas sandstone gas reservoir cannot be accurately predicted, affecting the success rate and economic benefits of oil and gas exploration.

[0006] A method for predicting the water production probability of a tight gas sandstone gas reservoir provided by the present invention includes:

[0007] Based on the mercury intrusion test, obtain the pore structure data of the gas reservoir, and the pore structure data includes porosity, permeability, displacement pressure, median pressure, and pore throat structure;

[0008] Classify and summarize the gas reservoir according to the pore structure data by the clustering analysis algorithm to obtain multiple gas reservoir groups;

[0009] Match the corresponding sub-water production prediction models of the gas reservoir groups, and predict the sub-water production probabilities of the gas reservoir groups;

[0010] Calculate the total water production probability of the gas reservoir according to the sub-production water probability based on the weighted average algorithm.

[0011] Preferably, the pore structure data of the gas reservoir is obtained based on the mercury intrusion test, and the pore structure data includes porosity, permeability, displacement pressure, median pressure, and pore throat structure, specifically including:

[0012] Calculate the porosity of the gas reservoir based on the mass-volume method, and the calculation formula for the porosity is:

[0013]

[0014] In the formula, θ represents the porosity of the gas reservoir; V q represents the pore volume of the gas reservoir; V represents the total volume of the gas reservoir;

[0015] Determine the permeability H of the gas reservoir based on Darcy's law;

[0016] Based on the mercury intrusion test, determine the pressure required for mercury to start entering the gas reservoir as the displacement pressure P 1 , and the pressure required for mercury to enter half of the pores in the gas reservoir as the median pressure P 2 ;

[0017] Analyze the size distribution of the pores and throats in the gas reservoir to generate the pore throat structure D;

[0018] Obtain the porosity θ, permeability H, displacement pressure P 1 , median pressure P 2 and pore throat structure G of the gas reservoir according to the mercury intrusion test.

[0019] Preferably, classify and summarize the gas reservoir according to the pore structure data through the clustering analysis algorithm to obtain multiple gas reservoir groups, specifically including:

[0020] Use the pore structure data corresponding to the gas reservoir as the classification standard, and randomly select k pieces of the pore structure data as the initial cluster centers;

[0021] For the pore structure data x i , calculate the distance between the pore structure data x i and the k initial cluster centers, and assign the pore structure data x i to the cluster C j corresponding to the nearest initial cluster center μ j , and the calculation formula for the distance is:

[0022]

[0023] In the formula, d(x i , μ n ) represents the distance between the pore structure data x i and the initial cluster center μ n ; represents the value of the pore structure data x i on the d-th dimension; represents the value of the initial cluster center μ n on the d-th dimension; D represents the dimension of the pore structure data;

[0024] Recalculate the center of the cluster C j according to the partitioning result of the cluster C j :

[0025]

[0026] In the formula, μ j represents the center of the cluster C j ; |C j | represents the number of pore structure data in the cluster C j ; x represents the pore structure data belonging to the cluster C j ;

[0027] Repeat the above steps until the center of the cluster C j no longer changes or reaches the preset number of iterations;

[0028] The objective of the clustering analysis algorithm is to minimize the sum of squared errors corresponding to the classification result of the gas reservoir, and the formula for calculating the sum of squared errors is:

[0029]

[0030] In the formula, E represents the sum of squared errors corresponding to the classification result of the gas reservoir; k represents the number of clusters; C j represents the j-th cluster; x represents the pore structure data belonging to the cluster C j ; μ j represents the center of the cluster C j ;

[0031] Based on the clustering analysis algorithm, divide the pore structure data into k clusters, that is, classification categories, and the similarity of the pore structure data within the cluster is the highest, and the similarity of the pore structure data between the clusters is the lowest. Classify and summarize the gas reservoirs based on the classification categories to obtain multiple gas reservoir groups.

[0032] Preferably, match the sub-production water prediction model corresponding to the gas reservoir group and predict the sub-production water probability of the gas reservoir group, specifically including:

[0033] For the gas reservoir group with high permeability, matching the seepage theory model to predict the water production probability of the gas reservoir group includes the following steps:

[0034] Calculate the fluid flow velocity in the gas reservoir group based on Darcy's law:

[0035]

[0036] In the formula, Q represents the flow rate of the fluid in the gas storage group, that is, the flow velocity; K represents the permeability of the gas storage group; A represents the cross-sectional area when the fluid flows in the gas storage group; ΔP represents the fluid pressure difference in the gas storage group; represents the fluid viscosity in the gas storage group; L represents the length of the gas storage group;

[0037] After obtaining the fluid flow velocity, combine the fluid pressure distribution and flow state in the gas storage group to obtain the water production probability of the gas reservoir group with high permeability.

[0038] Preferably, for the gas reservoir group with low permeability, matching the numerical simulation model to predict the water production probability of the gas reservoir group includes the following steps:

[0039] Divide the gas reservoir group into multiple grids based on the finite difference method, and use the difference equation to represent the momentum equation and momentum equation when the fluid flows in the gas reservoir group in each grid, generating a set of discretized algebraic equations. By solving the algebraic equations, obtain the flow state of the fluid in the gas reservoir group and predict the water production probability of the gas reservoir group;

[0040] The seepage equation for the pressure distribution and flow velocity of the fluid when flowing in the gas reservoir group is:

[0041]

[0042] In the formula, represents the change rate of the fluid pressure in the gas reservoir group with respect to time t; K x and K y represent the permeability of the gas storage group in the x-direction and y-direction; and represent the second-order partial derivatives of the fluid pressure in the gas storage group in the x-direction and y-direction, that is, the change rate of the fluid pressure in the gas storage group in space; μ represents the porosity of the porous medium in the gas storage group; q represents the source term of the gas storage group;

[0043] Discretize the seepage equation using the finite difference method to obtain the discretized algebraic equation:

[0044]

[0045] In the formula, and represent the fluid pressure values at the grid points (e,f) at time steps b + 1 and b; Δt represents the time step; Δx and Δy represent the grid sizes in the x and y directions; K x and K y represent the permeabilities of the gas storage group in the x and y directions; μ represents the porosity of the porous medium in the gas storage group; and represent the fluid pressure values at the grid points (e + 1,f) and (e - 1,f) at time step b; and represent the fluid pressure values at the grid points (e,f + 1) and (e,f - 1) at time step b; represents the source term value at the grid point (e,f) at time step b;

[0046] After obtaining the discretized algebraic equation, based on the preset boundary conditions and preset initial conditions, an iterative method is used to solve the discretized algebraic equation to obtain the numerical solutions of the pressure distribution and flow velocity during the fluid flow in the gas reservoir group, and predict the water production probability of the gas reservoir group.

[0047] Preferably, the preset boundary conditions are the pressure distribution and flow velocity during the fluid flow on the boundary of the gas reservoir group; the preset initial conditions are the fluid pressure distribution of the gas reservoir group at the initial time step.

[0048] Preferably, calculating the total water production probability of the gas reservoir according to the water production probability of the sub - production water, specifically includes:

[0049] According to the water production probability of the gas reservoir group, the weighted average algorithm is used to predict the total water production probability of the entire gas reservoir:

[0050]

[0051] In the formula, ZCS represents the total water production probability of the entire gas reservoir; FCS m represents the water production probability of the m - th gas reservoir group; W m represents the weight of the m - th gas reservoir group; M represents the total number of gas reservoir groups; ∑ represents the summation symbol.

[0052] Preferably, the weight of the gas reservoir group for the entire gas reservoir is determined according to the area, volume, thickness and reservoir quality of the gas reservoir group.

[0053] Compared with the related technology, a method for predicting the water production probability of a tight gas sandstone gas reservoir provided by the present invention has the following beneficial effects:

[0054] The present invention obtains pore structure data of a gas reservoir based on mercury intrusion tests, and the pore structure data includes porosity, permeability, displacement pressure, median pressure, and pore throat structure; classifies and summarizes the gas reservoirs according to the pore structure data through a clustering analysis algorithm to obtain multiple gas reservoir groups; matches the corresponding water production prediction models for the gas reservoir groups to predict the water production probability of the gas reservoir groups; and calculates the total water production probability of the gas reservoir based on the weighted average algorithm. By comprehensively analyzing the microscopic physical property characteristics and type attributes of tight gas sandstone gas reservoirs, the present invention can scientifically and objectively predict the water production probability of tight gas sandstone gas reservoirs, avoiding subjectivity and uncertainty in traditional methods; at the same time, for different types of reservoirs, the present invention uses different calculation models for prediction, which can improve the accuracy and effectiveness of prediction; and by comprehensively considering the water production probabilities of each reservoir, the prediction result of the entire reservoir can be obtained, further improving the reliability of prediction. The method of the present invention is not only applicable to tight gas sandstone gas reservoirs, but also can be extended to other types of unconventional gas reservoirs, providing technical support for the oil and gas resource exploration field. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 is a flowchart of a method for predicting the water production probability of a tight gas sandstone gas reservoir according to the present invention;

[0056] Figure 2 is a flowchart of calculating the total water production probability of a method for predicting the water production probability of a tight gas sandstone gas reservoir according to the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0057] The present invention will be further described below with reference to the drawings and embodiments.

[0058] Embodiment 1

[0059] As Figure 1 shown, a method for predicting the water production probability of a tight gas sandstone gas reservoir, the method comprising:

[0060] S1, obtaining pore structure data of a gas reservoir based on mercury intrusion tests, and the pore structure data includes porosity, permeability, displacement pressure, median pressure, and pore throat structure;

[0061] Among them, in the technical field of oil and gas resource exploration, accurately evaluating the pore structure characteristics of tight gas sandstone gas reservoirs is crucial for subsequent analysis of fluid flow behavior, prediction of production capacity, and formulation of development strategies. Mercury intrusion tests are widely used to obtain pore structure data of gas reservoirs. In this test, high-pressure mercury is injected into the sample, and by using the non-wetting property of mercury to pores, different-sized pore spaces are gradually filled, so that the key parameters of porosity, permeability, displacement pressure, median pressure, and pore throat structure of tight gas sandstone gas reservoirs can be accurately measured.

[0062] Specifically, the porosity of tight gas sandstone gas reservoirs reflects the proportion of pore volume in the total volume of the reservoir and is the basis for evaluating the reservoir's storage capacity; the permeability reflects the ease with which fluids can pass through the pore network and is directly related to subsequent production efficiency; the displacement pressure and median pressure respectively represent the pore pressures in the pore system that are the most difficult and on average the most difficult to be filled with fluids; the pore throat structure reflects the morphology of the connecting channels between pores and directly affects the fluid flow path and efficiency.

[0063] S2. Classify and summarize the gas reservoirs according to the pore structure data through a clustering analysis algorithm to obtain multiple gas reservoir groups.

[0064] It can be understood that after obtaining the pore structure data of the gas reservoirs, in order to further reveal the heterogeneity of the gas reservoirs and optimize the development strategy, the present invention uses a clustering analysis algorithm to deeply analyze and classify the pore structure data. Based on the inherent similarity of the data, the clustering analysis algorithm can group gas reservoirs with similar pore structure characteristics into one category to form multiple gas reservoir groups.

[0065] Through the above method, the complexity of subsequent analysis can be simplified, and precise development strategies can be formulated for specific types of reservoirs.

[0066] S3. Match the sub-production water prediction models corresponding to the gas reservoir groups and predict the sub-production water probabilities of the gas reservoir groups.

[0067] It should be noted that for each gas reservoir group, corresponding sub-production water prediction models can be matched. These models are constructed based on historical production data, geostatistical methods, and machine learning algorithms, and can comprehensively consider geological, engineering, and hydrodynamic factors to predict the sub-production water probabilities of each reservoir group.

[0068] Through the above method, the sub-production water probabilities of the gas reservoir groups can be accurately predicted, which is convenient for subsequent evaluation of the water flooding risk of the reservoir, optimization of the completion design of production wells, and formulation of optimal water control measures.

[0069] S4. Calculate the total production water probability of the gas reservoir based on the weighted average algorithm according to the sub-production water probabilities.

[0070] In practical applications, in order to obtain the total production water probability of the entire gas reservoir, the present invention uses a weighted average algorithm to integrate and calculate the sub-production water probabilities of each gas reservoir group. By comprehensively considering the reserve weights of different gas reservoir groups, the weighted average algorithm can ensure that the prediction of the total production water probability is more in line with the actual situation, thereby improving the accuracy and scientificity of gas reservoir evaluation and providing data support for subsequent oil and gas field development planning.

[0071] In the specific implementation process, the pore structure data of the gas reservoir are obtained based on the mercury intrusion test, and the pore structure data include porosity, permeability, displacement pressure, median pressure, and pore throat structure, specifically including:

[0072] Calculate the porosity of the gas reservoir based on the mass - volume method, and the calculation formula for the porosity is:

[0073]

[0074] In the formula, θ represents the porosity of the gas reservoir; V q represents the pore volume of the gas reservoir; V represents the total volume of the gas reservoir;

[0075] Determine the permeability H of the gas reservoir based on Darcy's law;

[0076] Based on the mercury intrusion test, determine the pressure required for mercury to start entering the gas reservoir as the displacement pressure P 1 , and the pressure required for mercury to enter half of the pores in the gas reservoir as the median pressure P 2 ;

[0077] Analyze the size distribution of the pores and throats in the gas reservoir to generate the pore throat structure D;

[0078] According to the mercury intrusion test, obtain the porosity θ, permeability H, displacement pressure P 1 , median pressure P 2 and pore throat structure G of the gas reservoir.

[0079] In practical applications, porosity, as one of the important indicators for measuring the storage capacity of tight gas sandstone gas reservoirs, is usually calculated using the mass - volume method. The calculation formula for porosity is: Among them, θ represents the porosity of the gas reservoir, which is a dimensionless ratio value and can reflect the percentage of the pore space in the reservoir accounting for the total volume; V q represents the pore volume of the gas reservoir, that is, the total volume of all connected pores in the reservoir; V represents the total volume of the gas reservoir, including the volume of pores, rock skeletons, and all other components; the permeability H represents the ease of fluid passing through the reservoir rock, and it is determined based on Darcy's law.

[0080] It should be noted that Darcy's law describes the linear relationship between the flow velocity and the pressure gradient when fluid flows in a porous medium, and it is the basis for evaluating the fluid conduction performance of the reservoir.

[0081] In order to obtain the microscopic structural characteristics of the gas reservoir, the present invention uses the mercury intrusion test to conduct an in - depth analysis of the gas reservoir. Specifically, first, the minimum pressure required for mercury to start penetrating into the reservoir can be measured, that is, the displacement pressure P 1, and the pressure when the reservoir is half filled with mercury, i.e., the median pressure P 2 , which can further reflect the pore connectivity and pore size distribution of the reservoir. The lower the displacement pressure, the better the pore connectivity of the reservoir; the median pressure is related to the average pore size of the reservoir and is used to predict the fluid flow characteristics subsequently.

[0082] Furthermore, by analyzing the mercury injection test data, the pore-throat structure G of the gas reservoir can be generated. The pore-throat structure can reflect the size distribution of pores and throats connecting the pores in the reservoir, facilitating the analysis of the migration mechanism of fluids in the reservoir.

[0083] Calculate the porosity θ by the mass-volume method, determine the permeability H in combination with Darcy's law, and the displacement pressure P obtained from the mercury injection test 1 , the median pressure P 2 and the pore-throat structure G, which can comprehensively and deeply analyze the physical properties of the gas reservoir and provide a scientific basis for subsequent oil and gas field development.

[0084] The gas reservoirs are classified and summarized according to the pore structure data by the clustering analysis algorithm to obtain multiple gas reservoir groups, specifically including:

[0085] Take the pore structure data corresponding to the gas reservoir as the classification criterion, and randomly select k pieces of the pore structure data as the initial cluster centers;

[0086] For the pore structure data x i , calculate the distances between the pore structure data x i and the k initial cluster centers, and assign the pore structure data x i to the cluster C j corresponding to the nearest initial cluster center μ j , and the calculation formula for the distance is:

[0087]

[0088] In the formula, d(x i ,μ n ) represents the distance between the pore structure data x i and the initial cluster center μ n ; represents the value of the pore structure data x i in the d-th dimension; represents the value of the initial cluster center μ n in the d-th dimension; D represents the dimension of the pore structure data;

[0089] Recalculate the center of the cluster C j according to the partitioning result of the cluster C j :

[0090]

[0091] wherein, μ j represents the center of cluster C j ; |C j | represents the number of pore structure data in cluster C j ; x represents the pore structure data belonging to cluster C j .

[0092] Repeat the above steps until the center of the cluster C j no longer changes or reaches a preset number of iterations;

[0093] The objective of the clustering analysis algorithm is to minimize the sum of squared errors corresponding to the classification results of the gas reservoir, and the calculation formula for the sum of squared errors is:

[0094]

[0095] wherein, E represents the sum of squared errors corresponding to the classification results of the gas reservoir; k represents the number of clusters; C j represents the jth cluster; x represents the pore structure data belonging to cluster C j ; μ j represents the center of cluster C j ;

[0096] Based on the clustering analysis algorithm, the pore structure data is divided into k clusters, that is, classification categories, and the similarity of the pore structure data within the clusters is the highest, and the similarity of the pore structure data between the clusters is the lowest. Based on the classification categories, the gas reservoirs are classified and summarized to obtain multiple gas reservoir groups.

[0097] It can be understood that the present invention effectively classifies gas reservoirs according to pore structure data by using a clustering analysis algorithm, and then through a mathematical model and iterative calculation, the accurate division of gas reservoir characteristics can be realized, and multiple gas reservoir groups with different characteristics can be obtained.

[0098] Specifically, first, the pore structure data corresponding to the gas reservoir can be used as the classification basis. The pore structure data includes multi-dimensional information such as pore size, shape, and connectivity, and is used to evaluate the storage capacity and fluid mobility of the reservoir. Randomly select k pore structure data points as the initial cluster centers, and these center points will be used as the basis for cluster division in the subsequent iterative process. For each pore structure data point x i , calculate its distance from the k initial cluster centers based on the Euclidean distance formula, and assign each data point x i to the cluster C j corresponding to the initial cluster center μ j to which it is closest. According to cluster Cj Based on the current partitioning result, recalculate the center μ of each cluster j ′ , so as to ensure that the cluster center can accurately reflect the overall characteristics of the data within the cluster, demonstrating the self - adaptability and optimization ability of the clustering algorithm. Repeat the above steps until the convergence condition is reached, that is, the position of the cluster center no longer changes significantly or the number of iterations reaches the preset upper limit, thereby ensuring the stability and accuracy of the clustering result.

[0099] The goal of the clustering analysis algorithm is to minimize the sum of squared errors E corresponding to the gas reservoir classification result, in order to find the optimal cluster partitioning scheme, making the data points within the same cluster as similar as possible, while the data points between different clusters are as different as possible.

[0100] Based on the results of the clustering analysis algorithm, the pore structure data can be divided into k clusters, that is, classification categories. The pore structure data within these clusters have a high degree of similarity, while the similarity of the pore structure data between clusters is relatively low. Based on these classification categories, the gas reservoirs can be classified and summarized to obtain multiple gas reservoir groups with different pore structure characteristics, which is convenient for subsequent reservoir evaluation, oil and gas resource potential assessment, and the formulation of exploitation strategies.

[0101] As Figure 2 shown, matching the sub - water production prediction model corresponding to the gas reservoir group to predict the sub - water production probability of the gas reservoir group specifically includes:

[0102] For the gas reservoir group with high permeability, matching the seepage theory model to predict the sub - water production probability of the gas reservoir group includes the following steps:

[0103] Calculate the fluid flow velocity in the gas reservoir group based on Darcy's law:

[0104]

[0105] In the formula, Q represents the flow rate of the fluid in the gas storage group, that is, the flow velocity; K represents the permeability of the gas storage group; A represents the cross - sectional area when the fluid flows in the gas storage group; ΔP represents the fluid pressure difference in the gas storage group; represents the fluid viscosity in the gas storage group; L represents the length of the gas storage group;

[0106] After obtaining the fluid flow velocity, combined with the fluid pressure distribution and flow state in the gas storage group, obtain the sub - water production probability of the gas reservoir group with high permeability.

[0107] Among them, by matching the sub - water production prediction model suitable for a specific gas reservoir group, the sub - water production probability of this gas reservoir group can be accurately predicted.

[0108] For gas reservoir groups with high permeability, a percolation theory model can be used for prediction. First, based on Darcy's law, the fluid flow velocity inside the gas reservoir group can be calculated. Darcy's law is the basic principle that describes the proportional relationship between the fluid flow velocity in a porous medium and its hydraulic gradient. Its mathematical expression is: In this formula, Q represents the flow rate of the fluid in the gas reservoir group, that is, the flow velocity, which is a key parameter to measure the ability of the fluid to pass through the medium; K represents the permeability of the gas reservoir group, which reflects the ability of the medium to allow the fluid to pass through; A represents the cross-sectional area of fluid flow, which is directly related to the size of the fluid flow channel; ΔP represents the fluid pressure difference in the gas reservoir group, which is the main driving force for fluid flow; represents the fluid viscosity, which affects the resistance of fluid flow; L represents the length of the gas reservoir group, which is the path length of fluid flow.

[0109] After obtaining the fluid flow velocity, it is necessary to comprehensively consider the fluid pressure distribution and flow state inside the gas reservoir group. The non-uniformity of the fluid pressure distribution may lead to an increase in the complexity of fluid flow, while the flow state directly affects the ability of the fluid to carry water. Through advanced data analysis techniques and numerical simulation methods, the water production probability of the high-permeability gas reservoir group can be further analyzed and calculated.

[0110] For the gas reservoir group with low permeability, a numerical simulation model is matched to predict the water production probability of the gas reservoir group, including the following steps:

[0111] Based on the finite difference method, the gas reservoir group is divided into multiple grids, and in each grid, the momentum equation and momentum equation when the fluid flows in the gas reservoir group are expressed by difference equations, generating a set of discretized algebraic equations. By solving the algebraic equations, the flow state of the fluid in the gas reservoir group is obtained, and the water production probability of the gas reservoir group is predicted;

[0112] The seepage equation for the pressure distribution and flow velocity when the fluid flows in the gas reservoir group is:

[0113]

[0114] In the formula, represents the change rate of the fluid pressure in the gas reservoir group with respect to time t; K x and K y represent the permeability of the gas storage group in the x-direction and y-direction; and represent the second-order partial derivatives of the fluid pressure in the gas storage group in the x-direction and y-direction, that is, the change rate of the fluid pressure in the gas storage group in space; μ represents the porosity of the porous medium in the gas storage group; q represents the source term of the gas storage group;

[0115] The seepage equation is discretized using the finite difference method to obtain a discretized algebraic equation:

[0116]

[0117] In the formula, and represent the fluid pressure values at the grid point (e, f) at time steps b + 1 and b; Δt represents the time step; Δx and Δy represent the grid sizes in the x and y directions; K x and K y represent the permeabilities of the gas storage group in the x and y directions; μ represents the porosity of the porous medium in the gas storage group; and represent the fluid pressure values at the grid points (e + 1, f) and (e - 1, f) at time step b; and represent the fluid pressure values at the grid points (e, f + 1) and (e, f - 1) at time step b; represents the source term value at the grid point (e, f) at time step b;

[0118] After obtaining the discretized algebraic equation, based on the preset boundary conditions and preset initial conditions, the discretized algebraic equation is solved using an iterative method to obtain the numerical solutions of the pressure distribution and flow velocity during fluid flow in the gas reservoir group, and to predict the water production probability of each part of the gas reservoir group.

[0119] The preset boundary conditions are the pressure distribution and flow velocity during fluid flow on the boundary of the gas reservoir group; the preset initial conditions are the fluid pressure distribution of the gas reservoir group at the initial time step.

[0120] In practical applications, for a gas reservoir group with low permeability, a numerical simulation model is used to predict its water production probability of each part. First, the complex gas reservoir group system can be divided into multiple fine grid cells by the finite difference method, so that the behavior of fluid flow can be simulated inside each grid. This process uses difference equations to approximately represent the momentum equation and mass conservation equation during fluid flow, and then generates a set of discretized algebraic equations.

[0121] When constructing a numerical model, a seepage equation can be used to describe the flow state of fluid in the gas reservoir group. This equation can comprehensively consider the change rate of fluid pressure with time and the influence of permeabilities in different directions on fluid flow. The expression of this seepage equation is: Where, represents the change rate of fluid pressure with time; K x and K y represent the permeabilities of the gas reservoir group in the x and y directions respectively, and these parameters reflect the degree of obstruction of the medium to fluid flow; and are the second-order partial derivatives of the fluid pressure in the x and y directions, which describe the spatial distribution change of the fluid pressure; μ represents the porosity of the porous medium, which determines the size of the space where the fluid can flow in the medium; q represents the source term, which represents the rate of fluid injection or production.

[0122] To solve this seepage equation, the finite difference method can be used for discretization, converting the continuous equation into a discretized algebraic equation. The expression of the discretized algebraic equation is: After obtaining the discretized algebraic equation, it is necessary to set preset boundary conditions and preset initial conditions to ensure the accuracy and reliability of the numerical simulation. The preset boundary conditions usually include the known values of the fluid pressure on the boundary or the flow rate conditions on the boundary, while the preset initial conditions include the distribution of the fluid pressure at the beginning of the simulation. Based on these conditions, the iterative method can be used to solve the discretized algebraic equation. Through repeated iterative calculations, gradually approach the real fluid flow state, and finally obtain the numerical solutions of the pressure distribution and flow velocity when the fluid flows in the gas reservoir group, and then the water production probability of the low-permeability gas reservoir group can be predicted.

[0123] The total water production probability of the gas reservoir is calculated based on the weighted average algorithm according to the water production probability of each sub-reservoir group, specifically including:

[0124] According to the water production probability of the gas reservoir group, the weighted average algorithm is used to predict the total water production probability of the entire gas reservoir:

[0125]

[0126] In the formula, ZCS represents the total water production probability of the entire gas reservoir; FCS m represents the water production probability of the m-th gas reservoir group; W m represents the weight of the m-th gas reservoir group; M represents the total number of gas reservoir groups; ∑ represents the summation symbol.

[0127] Determine the weight of each gas reservoir group for the entire gas reservoir according to the area, volume, thickness and reservoir quality of the gas reservoir group.

[0128] It can be understood that based on the weighted average algorithm, according to the water production probability of each gas reservoir group, the total water production probability of the entire gas reservoir can be calculated. First, according to the previous geological exploration, logging data analysis and laboratory test results, the water production probability FCS of each gas reservoir group can be obtained m , and this water production probability FCS m is a direct indicator reflecting the fluid production capacity of each reservoir group. Subsequently, the weight W of each gas reservoir group in the overall gas reservoir can be determined m, the weight W m is determined based on a series of professional parameters, including the area, volume, thickness of the gas reservoir group, and reservoir quality. Specifically, the area and volume reflect the spatial scale of the reservoir, the thickness is related to the continuity of the reservoir, and the reservoir quality covers key attributes such as porosity, permeability, and fluid saturation. These parameters together determine the storage capacity of the reservoir and the potential for fluid production.

[0129] After determining the sub - water - production probabilities and weights of each gas reservoir group, the weighted average algorithm can be used to calculate the total water - production probability ZCS, and the corresponding formula is: where M represents the total number of gas reservoir groups; ∑ represents the summation over all gas reservoir groups. This calculation process not only considers the water - production characteristics of each reservoir group itself but also reflects the importance and contribution degree of each reservoir group in the overall gas reservoir through weight allocation, thereby ensuring the accuracy of the total water - production probability prediction.

[0130] It should be noted that the application of the weighted average algorithm also depends on the rationality of weight allocation. The determination of weights needs to be based on scientific geological analysis and data support, avoiding subjective speculation to ensure the objectivity and reliability of the prediction results. At the same time, as the development of oil and gas fields deepens, the reservoir characteristics may change. Therefore, the sub - water - production probabilities and weight data can be updated regularly to adjust the prediction model in a timely manner to ensure the timeliness and accuracy of the prediction results.

[0131] Through the introduction of the above embodiments, the water - production probability prediction method for tight gas sandstone gas reservoirs of the present invention obtains the pore structure data of the gas reservoir based on mercury intrusion tests, and the pore structure data includes porosity, permeability, displacement pressure, median pressure, and pore - throat structure; classifies and summarizes the gas reservoirs according to the pore structure data through the clustering analysis algorithm to obtain multiple gas reservoir groups; matches the corresponding sub - water - production prediction models for the gas reservoir groups to predict the sub - water - production probabilities of the gas reservoir groups; calculates the total water - production probability of the gas reservoir based on the sub - water - production probabilities using the weighted average algorithm. By comprehensively analyzing the microscopic physical properties and type attributes of tight gas sandstone gas reservoirs, the present invention can scientifically and objectively predict the water - production probability of tight gas sandstone gas reservoirs, avoiding subjectivity and uncertainty in traditional methods; at the same time, for different types of reservoirs, the present invention uses different calculation models for prediction, which can improve the accuracy and effectiveness of prediction; and by synthesizing the water - production probabilities of each reservoir, the prediction result of the entire reservoir can be obtained, further improving the reliability of the prediction. The method of the present invention is not only applicable to tight gas sandstone gas reservoirs but also can be extended to other types of unconventional gas reservoirs, providing technical support for the field of oil and gas resource exploration.

[0132] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing device generate means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0133] Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program. This program can be stored in a computer-readable storage medium, which includes read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM), or other optical disc memories, magnetic disk memories, tape memories, or any other medium that can be used to carry or store data and is computer-readable.

[0134] It should also be noted that the term "comprising", "including", or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, commodity, or device that includes a series of elements not only includes those elements but also includes other elements not explicitly listed, or further includes elements inherent to such process, method, commodity, or device. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of another identical element in the process, method, commodity, or device that includes the element.

Claims

1. A method for predicting water production probability of tight gas sandstone gas reservoirs, characterized in that: The method comprises: Acquiring pore structure data of the gas reservoir based on mercury injection test, wherein the pore structure data includes porosity, permeability, displacement pressure, median pressure and pore throat structure; Classifying and summarizing the gas reservoirs according to the pore structure data by using a cluster analysis algorithm to obtain a plurality of gas reservoir groups; Matching the water production prediction model corresponding to the gas reservoir group to predict the water production probability of the gas reservoir group; The total water production probability of the gas reservoir is calculated according to the partial water production probability based on a weighted average algorithm.

2. The method for predicting water production probability of a tight gas sandstone gas reservoir according to claim 1, characterized in that: The pore structure data of the gas reservoir is obtained based on the mercury injection test, and the pore structure data includes porosity, permeability, displacement pressure, median pressure and pore throat structure, specifically including: The porosity of the gas reservoir is calculated based on the mass volume method, and the calculation formula of the porosity is: Where, θ represents the porosity of the gas reservoir; V q represents the pore volume of the gas reservoir; V represents the total volume of the gas reservoir; Determine the permeability H of the gas reservoir based on Darcy's law; Based on the mercury injection test, the pressure required for mercury to start entering the gas reservoir is determined as the displacement pressure P1, and the pressure required for mercury to enter half of the pores in the gas reservoir is determined as the median pressure P2; Analyzing the size distribution of pores and throats of the gas reservoir to generate a pore-throat structure D; The porosity θ, permeability H, displacement pressure P1, median pressure P2 and pore throat structure G of the gas reservoir are obtained according to the mercury injection test.

3. The method for predicting water production probability of a tight gas sandstone gas reservoir according to claim 1, characterized in that: The gas reservoirs are classified and summarized according to the pore structure data by using a cluster analysis algorithm to obtain a plurality of gas reservoir groups, specifically including: Using the pore structure data corresponding to the gas reservoir as a classification standard, randomly selecting k pore structure data as initial cluster centers; For the pore structure data x i , calculate the pore structure data x i The distance from the k initial cluster centers is calculated, and the pore structure data xi is assigned to the initial cluster center μ that is closest to the pore structure data xi. j The corresponding cluster C j , the distance is calculated as follows: In the formula, d(x i , μ n ) represents the pore structure data x i With the initial cluster center μ n The distance between Represents the pore structure data x i The value in the dth dimension; represents the initial cluster center μ n The value in the dth dimension; D represents the dimension of the pore structure data; According to the cluster C j The partition result recalculates the cluster C j Center of: In the formula, μ j Represents cluster C j The center of |C j | represents cluster C j The number of mesoporous structure data; x indicates that it belongs to cluster C j Pore ​​structure data; Repeat the above steps until the cluster C j The center of does not change anymore or reaches the preset number of iterations; The goal of the cluster analysis algorithm is to minimize the total sum of square errors corresponding to the classification results of the gas reservoir. The calculation formula of the total sum of square errors is: Where E represents the total error sum of squares corresponding to the classification results of the gas reservoir; k represents the number of clusters; C j represents the jth cluster; x represents the cluster C j Pore ​​structure data of μ j Represents cluster C j the center of; Based on the cluster analysis algorithm, the pore structure data is divided into k clusters, i.e., classification categories, and the similarity of the pore structure data within the cluster is the highest, and the similarity of the pore structure data between the clusters is the lowest. Based on the classification categories, the gas reservoirs are classified and summarized to obtain multiple gas reservoir groupings.

4. The method for predicting water production probability of a tight gas sandstone gas reservoir according to claim 1, characterized in that: The matching of the water production prediction model corresponding to the gas reservoir group to predict the water production probability of the gas reservoir group specifically includes: For the gas reservoir group with high permeability, matching the permeability theoretical model to predict the probability of water production of the gas reservoir group includes the following steps: The fluid flow velocity in the gas reservoir group is calculated based on Darcy's law: In the formula, Q represents the flow rate of the fluid in the gas storage group, that is, the flow velocity; K represents the permeability of the gas storage group; A represents the cross-sectional area of ​​the fluid when it flows in the gas storage group; ΔP represents the fluid pressure difference in the gas storage group; represents the viscosity of the fluid in the gas storage group; L represents the length of the gas storage group; After the fluid flow velocity is obtained, the water production probability of the high permeability gas reservoir group is obtained in combination with the fluid pressure distribution and flow state in the gas storage group.

5. The method for predicting water production probability of a tight gas sandstone gas reservoir according to claim 4, characterized in that: For the gas reservoir group with low permeability, matching the numerical simulation model to predict the probability of water production of the gas reservoir group includes the following steps: Divide the gas reservoir group into a plurality of grids based on the finite difference method, and use a differential equation in each grid to represent the momentum equation and the momentum equation of the fluid flow in the gas reservoir group, and generate a set of discretized algebraic equations. By solving the algebraic equations, the flow state of the fluid in the gas reservoir group is obtained, and the probability of water production of the gas reservoir group is predicted; The pressure distribution and flow velocity of the fluid flowing in the gas reservoir group are as follows: In the formula, K represents the rate of change of fluid pressure in the gas reservoir group with time t; x and K y represents the permeability of the gas storage group in the x-direction and the y-direction; and represents the second-order partial derivative of the fluid pressure in the gas storage group in the x-direction and the y-direction, that is, the rate of change of the fluid pressure in the gas storage group in space; μ represents the porosity of the porous medium in the gas storage group; q represents the source term of the gas storage group; The finite difference method is used to discretize the seepage equation, and the discretized algebraic equation is obtained: In the formula, and represents the fluid pressure value of the grid point (e, f) at time steps b+1 and b; Δt represents the time step; Δx and Δy represent the grid size in the x and y directions; K x and K y represents the permeability of the gas storage group in the x-direction and the y-direction; μ represents the porosity of the porous medium in the gas storage group; and represents the fluid pressure value at grid points (e+1,f) and (e-1,f) at time step b; and represents the fluid pressure value at grid points (e,f+1) and (e,f-1) at time step b; represents the source value of the grid point (e,f) at time step b; After obtaining the discretized algebraic equation, the discretized algebraic equation is solved by an iterative method based on preset boundary conditions and preset initial conditions to obtain the numerical solution of the pressure distribution and flow velocity of the fluid flowing in the gas reservoir group, and predict the probability of water production of the gas reservoir group.

6. The method for predicting water production probability of a tight gas sandstone gas reservoir according to claim 5, characterized in that: The preset boundary condition is the pressure distribution and flow velocity of the fluid flowing on the boundary of the gas reservoir group; the preset initial condition is the fluid pressure distribution of the gas reservoir group at the initial time step.

7. The method for predicting water production probability of a tight gas sandstone gas reservoir according to claim 1, characterized in that: The calculating the total water production probability of the gas reservoir according to the partial water production probability based on the weighted average algorithm specifically includes: According to the water production probability of the gas reservoir group, the total water production probability of the entire gas reservoir is predicted using a weighted average algorithm: Where ZCS represents the total water production probability of the entire gas reservoir; FCS m represents the probability of water production of the mth gas reservoir group; W m represents the weight of the mth gas reservoir group; M represents the total number of gas reservoir groups; ∑ represents the summation symbol.

8. The method for predicting water production probability of a tight gas sandstone gas reservoir according to claim 7, characterized in that: The weight of the gas reservoir layer group to the entire gas reservoir is determined according to the area, volume, thickness and reservoir quality of the gas reservoir layer group.

Images