Method and device for predicting water invasion channel of dual-medium reservoir water-driven gas reservoir

Abstract

The invention discloses a water invasion channel prediction method and device for a dual-medium reservoir water-driven gas reservoir, and belongs to the technical field of gas reservoir description and development. The method comprises the following steps: S1, establishing a dual-medium model by applying well-to-seismic interpretation data and dynamic data; s2, calculating a crack equivalent permeability model and a matrix permeability model; s3, carrying out gridding description on the heterogeneity characterization parameters of the matrix and the crack; s4, describing the heterogeneity degree between the geologic characteristic parameter grids of the matrix and the fracture; s5, calculating an average probability attribute body of the matrix and the crack; and S6, fusing the average probability attribute bodies of the matrix and the crack into a water invasion channel distribution model. According to the method, the water invasion channel in the gas reservoir range can be accurately predicted, the influence of human factors is reduced, the geometrical morphology and the spatial configuration relation of the water invasion channel are described, and visual, fine, high-calculation-efficiency and comprehensive evaluation parameter prediction of the double-medium reservoir water invasion channel is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of gas reservoir description and development, and relates to a method for predicting water invasion channels, specifically a method and device for predicting water invasion channels in a water drive gas reservoir with a dual - medium reservoir. Background Art

[0002] The biggest challenge in gas reservoir development is gas well waterlogging. Gas well waterlogging not only increases the cost of surface water treatment, but also shortens the stable production time of the gas reservoir, accelerates the production decline, reduces the gas reservoir recovery rate, thus causing huge resource waste and economic losses. An intuitive and refined gas reservoir water invasion channel prediction method can enable technicians to timely and accurately take targeted water - proof and water - control countermeasures, which plays a crucial role in the long - term stable production and efficient development of the gas reservoir.

[0003] Currently, the commonly used methods for identifying gas reservoir water invasion channels include tracer method, logging data identification method, well test data identification method, analytic hierarchy process, and geological modeling method. These methods all have their own technical limitations:

[0004] The tracer method is an engineering method that injects gas tracers into gas wells to monitor the connectivity between wells. The advantage of this method is that it can qualitatively understand the connectivity between production wells, but it cannot describe the reservoir connectivity details between wells and in the far - well zone.

[0005] The logging data identification method is a method that obtains single - well reservoir physical property parameters through comprehensive interpretation of logging curves to judge the connectivity relationship between wells. This method can quantitatively describe the permeability changes of different intervals in the well, and can achieve a relatively refined description of the water invasion channels between two wells with a small well spacing, but it cannot be identified in areas without well control.

[0006] The well test data identification method is a method that interprets reservoir characteristic parameters through formation pressure data obtained by methods such as pressure build - up well test and interference well test to judge whether there are water invasion channels or seepage barriers between wells. This method can relatively accurately describe the reservoir changes within the well - controlled range, but it also cannot make a judgment on areas without well control.

[0007] The analytic hierarchy process is to decompose the factors related to water invasion channel identification into several levels according to different attributes by establishing a hierarchical structure model, and then obtain the final decision result through processes such as pairwise comparison matrix and consistency test. The advantage of this method is simplicity and speed, but it is greatly affected by human factors and can only be used as an auxiliary method for qualitative description.

[0008] The conventional geological modeling method comprehensively applies dynamic and static data of gas reservoirs to establish a three-dimensional geological model, uses the permeability distribution as a discrimination basis, and describes the planar distribution and spatial configuration relationship of water invasion channels. To a certain extent, this method can achieve the purpose of intuitive and refined prediction of water invasion channels. However, its general research object is a single-medium reservoir, and the discrimination criterion is single, so it cannot achieve the goals of refinement, high calculation efficiency, and comprehensive evaluation parameters for predicting water invasion channels in dual-medium reservoirs. Summary of the Invention

[0009] The purpose of the present invention is to provide a method and device for predicting water invasion channels in a water drive gas reservoir in a dual-medium reservoir, so as to solve the problem that the conventional geological modeling method cannot achieve refinement, high calculation efficiency, and comprehensive evaluation parameters for predicting water invasion channels in dual-medium reservoirs.

[0010] To achieve the above purpose, the technical solution adopted by the present invention is:

[0011] A method for predicting water invasion channels in a water drive gas reservoir in a dual-medium reservoir includes the following steps carried out in sequence:

[0012] S1. Jointly apply well-seismic interpretation data and dynamic data to establish a matrix-fracture dual-medium model;

[0013] S2. Calculate the fracture equivalent permeability model through the Oda method and calculate the matrix permeability model through the pore-permeability relationship formula for stratification and microfacies respectively;

[0014] S3. Quantitatively describe the heterogeneity characterization parameters of the matrix permeability model and the fracture equivalent permeability model in a grid manner respectively;

[0015] S4. Describe the degree of heterogeneity between the geological characteristic parameter grids of the matrix and fractures respectively;

[0016] S5. Set the weights of the geological characteristic parameters respectively, and calculate the average probability attribute bodies of the matrix and fractures respectively;

[0017] S6. Integrate the average probability attribute bodies of the matrix and fractures into the water invasion channel distribution model with the maximum probability, and the prediction of the water invasion channel can be completed;

[0018] The heterogeneity characterization parameters include the breakthrough coefficient, range difference, and coefficient of variation; the geological characteristic parameters include permeability and heterogeneity characterization parameters.

[0019] As a limitation, the step S1 specifically includes:

[0020] S11. Jointly apply the seismic interpretation data and well logging interpretation data of the study area to establish a stratigraphic framework model;

[0021] S12. Using the phased seismic inversion data volume matching the dynamic data as a constraint, applying multi-well variogram analysis, and using the sequential indicator simulation method to establish a facies model; using the Gaussian random method with the facies model as a constraint, and using the co-kriging method for deterministic simulation with the facies-controlled inversion porosity plan view as a plane constraint, and combining the two to calculate the matrix porosity model;

[0022] S13. Using the imaging logging interpretation conclusion as a basic parameter, classifying the fractures according to the occurrence and sub-layers, using the fracture prediction plan view and the fault distance volume as constraints, and randomly simulating the fracture intensity model; using the fracture intensity model as a constraint, and randomly simulating to establish the DFN model of various fractures.

[0023] As a further limitation, the step S2 specifically includes:

[0024] S21. Establishing cross-plots of porosity and permeability for different sedimentary microfacies of each sub-layer respectively, and fitting the relationship curve between porosity and permeability to obtain a power function:

[0025] K = a·φ b

[0026] In the formula, K is the permeability; φ is the porosity; a is the power function coefficient; b is the power function exponent

[0027] Calculating the permeability of different sedimentary microfacies of each sub-layer in the matrix porosity model using the power function to obtain the matrix permeability model;

[0028] S22. Interpreting the fracture aperture through imaging logging data, and calculating the fracture sheet permeability according to the fracture plane flow theory formula. The calculation formula is as follows:

[0029] K f = b 2 / 12

[0030] In the formula, K f is the fracture sheet permeability; b is the fracture aperture

[0031] Calculating the fracture equivalent permeability model using the Oda method in the DFN model.

[0032] As another limitation, the step S3 specifically includes performing the following operations on the matrix permeability model and the fracture equivalent permeability model respectively:

[0033] S31. Statistically calculating the average permeability within the entire grid range, and then using the following formula to calculate the permeability breakthrough coefficient of each grid to complete the quantitative description of the breakthrough coefficient gridding:

[0034] T K = K n / K mean

[0035] In the formula, T k is the breakthrough coefficient of a single grid; K n is the permeability of a single grid; K mean is the average permeability within the entire grid range;

[0036] S32. Statistically analyze the minimum permeability within the entire grid range, and then use the following formula to calculate the range of each grid to complete the quantitative description of range gridding:

[0037] J K = K n / K min

[0038] In the formula, J k is the range of a single grid; K n is the permeability of a single grid; K min is the minimum permeability within the entire grid range;

[0039] S33. Statistically analyze the average permeability within the entire grid range, and then use the following formula to calculate the coefficient of variation of each grid to complete the quantitative description of coefficient-of-variation gridding:

[0040] V K =(K n -K mean ) / K mean

[0041] In the formula, V k is the coefficient of variation of a single grid; K n is the permeability of a single grid; K mean is the average permeability within the entire grid range.

[0042] As the third limitation, the step S4 specifically includes:

[0043] Classify the geological feature parameter values into homogeneous, heterogeneous, and strongly heterogeneous according to the frequency distribution histogram of each geological feature parameter, and respectively describe the degree of heterogeneity between the geological feature parameter grids of the matrix and the fractures.

[0044] As the fourth limitation, the step S5 specifically includes:

[0045] S51. Calibrate the discrete permeability logging curve of a single well with the core experimental permeability, and use the following formula to respectively verify the correlation coefficient between the calibrated permeability logging curve and each heterogeneity characterization parameter of the matrix:

[0046]

[0047] In the formula, U kis the logging permeability value of single - well discretization; H mk is a certain matrix heterogeneity characterization parameter; Cov(U k , H mk ) is the covariance of U k and H mk ; Var[U k is the variance of U k ; Var[H mk is the variance of H mk ;

[0048] S52. Verify the correlation coefficients between the fracture intensity curve of single - well discretization and each fracture heterogeneity characterization parameter respectively with the following formula:

[0049]

[0050] In the formula, I f is the fracture intensity value of single - well discretization; H fk is a certain fracture heterogeneity characterization parameter; Cov(I f , H fk ) is the covariance of I f and H fk ; Var[I f is the variance of I f ; Var[H fk is the variance of H fk ;

[0051] S53. Set the weights of each geological characteristic parameter respectively: The weight of permeability is 1;

[0052] For the heterogeneity characterization parameter: If the correlation coefficient r = 1, the weight of this heterogeneity characterization parameter is 1; If 0.5 ≤ r < 1, the weight of this heterogeneity characterization parameter is a value greater than 0.5 and less than 1; If 0 < r ≤ 0.5, the weight of this heterogeneity characterization parameter is a value greater than 0 and less than 0.5; If r = 0, this heterogeneity characterization parameter cannot be used to establish the average probability property body;

[0053] S54. Calculate the average probability property bodies of the matrix and fractures according to the following two formulas:

[0054]

[0055] In the formula, is the weighted average value of a single grid; K n is the permeability of a single grid; f1 is the weight of K n ; T k is the breakthrough coefficient of a single grid; f2 is the weight of T k ; J k is the range of a single grid; f3 is the weight of Jk Weight; V k Is the coefficient of variation of a single grid; f4 is V k Weight;

[0056]

[0057] In the formula, P i Is the average probability of a single grid; Is the weighted average of a single grid; Is the largest weighted average among all grids statistically.

[0058] As the fifth limitation, the step S6 specifically includes:

[0059] Set the weights of the matrix average probability attribute body and the fracture average probability attribute body, and fuse the average probability attribute bodies of the matrix and the fracture into the water invasion channel distribution model with the maximum probability according to the following formula, and the prediction of the water invasion channel can be completed:

[0060]

[0061] In the formula, Is the water invasion channel probability value of a single grid; P im Is the average probability of a single matrix grid; y1 is P im Weight; P if Is the average probability of a single fracture grid; y2 is P if Weight.

[0062] The present invention also provides a water invasion channel prediction device for a water drive gas reservoir in a dual - porosity reservoir. The prediction device includes a memory and a processor, as well as a computer program stored on the memory and running on the processor. The processor is coupled to the memory, and when the processor executes the computer program, it implements the above - mentioned water invasion channel prediction method for the water drive gas reservoir in the dual - porosity reservoir.

[0063] Due to the adoption of the above - mentioned technical solution, compared with the prior art, the technical progress achieved by the present invention is as follows:

[0064] ① A water invasion channel prediction method and device for a water drive gas reservoir in a dual - porosity reservoir provided by the present invention improve the defects of the conventional geological modeling method. It can not only accurately judge the water invasion channels or seepage barriers within the well - controlled range, but also predict the water invasion channels or the areas to be water - invaded outside the well - controlled range and within the gas reservoir range. Moreover, it reduces the influence of human factors, depicts the geometric shape and spatial configuration relationship of the water invasion channels, and realizes an intuitive, fine, high - computational - efficiency and comprehensive - evaluation - parameter prediction of the water invasion channels in the dual - porosity reservoir;

[0065] ② The water invasion channel prediction method and device for a dual - medium reservoir water - drive gas reservoir provided by the present invention jointly apply well - seismic interpretation data and dynamic data to establish a refined dual - medium model. When establishing the matrix permeability model, the pore - permeability relationship formula for different intervals and facies belts is adopted to further improve the refinement degree of the model. The fracture equivalent permeability model calculated by the Oda method further improves the calculation efficiency of the model, making the prediction of water invasion channels in dual - medium reservoirs more targeted and reliable.

[0066] ③ The water invasion channel prediction method and device for a dual - medium reservoir water - drive gas reservoir provided by the present invention respectively conduct grid - based quantitative evaluation on the heterogeneity characterization parameters of the matrix and fractures, and respectively establish the average probability volumes of the matrix and fractures through the weighted average method, and then fuse them into the maximum probability water invasion channel distribution model, achieving the purpose of comprehensively, intuitively, and refinedly predicting the geometric shape and spatial configuration relationship of the water invasion channels with comprehensive evaluation parameters.

[0067] ④ The water invasion channel prediction method and device for a dual - medium reservoir water - drive gas reservoir provided by the present invention can be used for predicting the water invasion channels in dual - medium reservoir water - drive gas reservoirs, and have high practical value and broad application prospects for preventing water, controlling water, ensuring the stable production of gas reservoirs, and improving the recovery rate of gas reservoirs.

[0068] The present invention improves the defects of the conventional geological modeling method, can accurately predict the water invasion channels or seepage barriers near the wellbore, can also predict the water invasion channels or the areas to be water - invaded in the whole gas reservoir, reduces the influence of human factors, depicts the geometric shape and spatial configuration relationship of the water invasion channels, and realizes the intuitive, refined, high - calculation - efficiency, and comprehensive - evaluation - parameter prediction of the water invasion channels in dual - medium reservoirs. Description of the Drawings

[0069] Figure 1 It is the flow chart of the water invasion channel prediction method for a dual - medium reservoir water - drive gas reservoir in Embodiment 1;

[0070] Figure 2 It is the intersection diagram of porosity and permeability of different sedimentary micro - facies of each small layer in Embodiment 1;

[0071] Figure 3 It is the matrix permeability model in Embodiment 1;

[0072] Figure 4 It is the fracture equivalent permeability model in Embodiment 1;

[0073] Figure 5 It is the grid - based quantitative description of the heterogeneity characterization parameters of the matrix and fractures in Embodiment 1, where Figure 5 a is the matrix range, Figure 5 b is the matrix breakthrough coefficient, Figure 5 c is the matrix variation coefficient,Figure 5 d is the crack range difference, Figure 5 e is the crack breakthrough coefficient, Figure 5 f is the crack variation coefficient;

[0074] Figure 6 is the average probability attribute body of the matrix in Example 1;

[0075] Figure 7 is the average probability attribute body of the cracks in Example 1;

[0076] Figure 8 is the water invasion channel distribution model of the dual - medium reservoir water - drive gas reservoir in Example 1. Specific implementation manners

[0077] The present invention will be further described in detail below through specific examples. It should be understood that the described examples are only used to explain the present invention and do not limit the present invention.

[0078] Example 1 A method for predicting water invasion channels of a dual - medium reservoir water - drive gas reservoir

[0079] This example discloses a method for predicting water invasion channels of a dual - medium reservoir water - drive gas reservoir. Its flow block diagram is as Figure 1 shown. In this example, the water invasion channels of a certain carbonate gas reservoir are predicted. The carbonate gas reservoir is located in Central Asia, with sedimentary characteristics of reef - bank body deposition, tectonic characteristics of faulted anticline structure, reservoir type of fracture - porosity type, complex internal connectivity relationship of the gas reservoir. Analyzing from formation pressure, productivity, and water production conditions, there are large differences between wells. It is very difficult to finely describe the reservoir heterogeneity characteristics and depict the water invasion channels by conventional methods. This example specifically includes the following steps carried out in sequence:

[0080] S1. Jointly apply well - seismic interpretation data and dynamic data to establish a refined matrix - fracture dual - medium model

[0081] The matrix - fracture dual - medium model includes two parts: the matrix attribute model and the fracture attribute model. Among them, the matrix attribute model part includes a facies model and a matrix porosity model; the fracture attribute model part includes a fracture intensity model and a DFN model.

[0082] S11. Jointly apply seismic interpretation data and well - logging interpretation data of the study area to establish a stratigraphic framework model;

[0083] S12. For the matrix attribute model part: Using the facies - controlled seismic inversion data volume matched with dynamic data as a constraint, apply multi - well variogram analysis, and use the sequential indicator simulation method to establish a facies model; Using the Gaussian random method with the facies model as a constraint, and using the co - kriging method for deterministic simulation with the facies - controlled inversion porosity plan view as a plane constraint, and combining the two to calculate the matrix porosity model;

[0084] S13. Fracture property model part: Using the interpretation results of imaging logging as the basic parameters, classify fractures according to their occurrence and sub-layers, and randomly simulate the fracture intensity model with the fracture prediction plan view and fault distance volume as constraints; using the fracture intensity model as a constraint, randomly simulate and establish the DFN model of various fractures.

[0085] S2. Calculate the fracture equivalent permeability model through the Oda method and calculate the matrix permeability model through the pore-permeability relationship formula for stratification and microfacies

[0086] S21. Establish cross plots (as shown in Figure 2 ) for the porosity and permeability of different sedimentary microfacies in each sub-layer, fit the relationship curve between porosity and permeability according to the cross plot, and obtain the power function:

[0087] K = aφ b

[0088] In the formula, K is the permeability, mD; φ is the porosity, decimal; a is the power function coefficient; b is the power function exponent

[0089] In this embodiment, the relevant power functions are shown in Table 1:

[0090] Table 1 Power function table of pore-permeability relationship for stratification and sedimentary microfacies

[0091]

[0092] In the matrix porosity model, use the above power function to calculate the permeability of different sedimentary microfacies in each sub-layer to obtain the matrix permeability model, as shown in Figure 3 ;

[0093] S22. Interpret the fracture aperture through imaging logging data, and calculate the fracture sheet permeability according to the fracture plane flow theory formula. The calculation formula is as follows:

[0094] K f = b 2 / 12

[0095] In the formula, K f is the fracture sheet permeability, mD; b is the fracture aperture, μm

[0096] In the DFN model, use the Oda method to calculate the fracture equivalent permeability model, as shown in Figure 4 ;

[0097] S3. Carry out grid-based quantitative description of the heterogeneity characterization parameters of the matrix permeability model and the fracture equivalent permeability model

[0098] Perform the following operations on the matrix permeability model and the fracture equivalent permeability model respectively:

[0099] S31. Statistically calculate the average permeability within the entire grid range, and then use the following formula to calculate the permeability breakthrough coefficient for each grid to complete the quantitative description of the breakthrough coefficient grid:

[0100] T K = K n / K mean

[0101] In the formula, T k is the breakthrough coefficient of a single grid, dimensionless; K n is the permeability of a single grid, mD; K mean is the average permeability within the entire grid range, mD;

[0102] S32. Statistically calculate the minimum permeability within the entire grid range, and then use the following formula to calculate the range for each grid to complete the quantitative description of the range grid:

[0103] J K = K n / K min

[0104] In the formula, J k is the range of a single grid, dimensionless; K n is the permeability of a single grid, mD; K min is the minimum permeability within the entire grid range, mD;

[0105] S33. Statistically calculate the average permeability within the entire grid range, and then use the following formula to calculate the coefficient of variation for each grid to complete the quantitative description of the coefficient of variation grid:

[0106] V K = (K n - K mean ) / K mean

[0107] In the formula, V k is the coefficient of variation of a single grid, dimensionless; K n is the permeability of a single grid, mD; K mean is the average permeability within the entire grid range, mD

[0108] The quantitative descriptions of the range, breakthrough coefficient, and coefficient of variation grids for the matrix permeability model and the fracture equivalent permeability model are as Figure 5 shown.

[0109] S4. Describe the degree of heterogeneity between the geological characteristic parameter grids of the matrix and fractures respectively

[0110] According to the frequency distribution histogram of each geological characteristic parameter, the geological characteristic parameter values ​​are divided into three categories: homogeneous, heterogeneous and strongly heterogeneous, describing the degree of heterogeneity between the geological characteristic parameter grids of the matrix and fractures, as shown in Table 2 and Table 3:

[0111] Table 2 Heterogeneity of matrix geological characteristic parameters

[0112] Geological characteristic parameters Homogeneous Heterogeneous Strongly heterogeneous Permeability (mD) ≤5 >5 and ≤50 >50 Advance coefficient ≤1 >1 and ≤5 >5 Range ≤10 >10 and ≤100 >100 Coefficient of variation ≤1 >1 and ≤5 >5

[0113] Table 3 Heterogeneity of fracture geological characteristic parameters

[0114]

[0115]

[0116] S5. Set the weights of geological characteristic parameters respectively, and calculate the average probability attribute volume of matrix and fracture respectively.

[0117] S51. The permeability logging curve discretized by the core experiment is corrected, and the correlation coefficients between the corrected permeability logging curve and various matrix heterogeneity characterization parameters (progression coefficient, range and coefficient of variation) are verified by the following formulas:

[0118]

[0119] Where U k is the discretized logging permeability value of a single well, mD; H mk Cov(U k , H mk ) is U k With H mk The covariance of Var[U k ] is U k The variance of Var[H mk ] is H mk The variance of

[0120] S52. Use the following formulas to verify the correlation coefficients between the single well discretized fracture intensity curve and the fracture heterogeneity characterization parameters (progression coefficient, range and coefficient of variation):

[0121]

[0122] In the formula, I f is the discretized fracture strength value of a single well, dimensionless; H fk is the penetration coefficient, range or coefficient of variation of the crack; Cov(I f , H fk ) is I f With H fkCovariance of; Var[I f ; Var[I f is the variance of I fk ; Var[H fk is the variance of H

[0123] S53. Set the weights of each geological feature parameter respectively: the weight of permeability is 1;

[0124] Heterogeneity characterization parameter: If the correlation coefficient r = 1, it proves 100% positive correlation, and the weight of this heterogeneity characterization parameter is 1; if 0.5 ≤ r < 1, the weight of this heterogeneity characterization parameter can be set to a value greater than 0.5 and less than 1; if 0 < r ≤ 0.5, the weight of this heterogeneity characterization parameter can be set to a value less than 0.5 and greater than 0; if r = 0, it proves completely uncorrelated, and this heterogeneity characterization parameter cannot be used to establish the average probability attribute body;

[0125] In this embodiment, the weights of each heterogeneity characterization parameter are set according to Table 4 and Table 5 respectively;

[0126] Table 4 Weight setting table of matrix heterogeneity characterization parameters

[0127] Heterogeneity characterization parameter Range Advance coefficient Coefficient of variation Weight 0.6 0.5 0.2

[0128] Table 5 Weight setting table of fracture heterogeneity characterization parameters

[0129] Heterogeneity characterization parameter Range Advance coefficient Coefficient of variation Weight 0.8 0.6 0.4

[0130] S54. Calculate the average probability attribute body of the matrix according to the following two formulas, and calculate the average probability attribute body of the fracture in the same way:

[0131]

[0132] In the formula, is the weighted average value of a single grid, mD; K n is the permeability of a single grid, mD; f1 is the weight of K n ; T k is the breakthrough coefficient of a single grid, dimensionless; f2 is the weight of T k ; J k is the range of a single grid, dimensionless; f3 is the weight of J k ; V k is the coefficient of variation of a single grid, dimensionless; f4 is the weight of V k ;

[0133]

[0134] In the formula, P i is the average probability of a single grid; is the weighted average of a single grid; is the maximum weighted average among all grids

[0135] The average probability attribute volume of the matrix and the average probability attribute volume of the fracture are as Figure 6 、 Figure 7 shown.

[0136] S6. Integrate the average probability attribute volumes of the matrix and the fracture into a maximum probability water invasion channel distribution model

[0137] Set the weights of the average probability attribute volume of the matrix and the average probability attribute volume of the fracture, which can be adjusted accordingly according to the actual development degree of the matrix and fracture in the gas reservoir reservoir and the influence degree on gas-water seepage. The weight of the higher development degree or greater influence degree is set larger.

[0138] In this embodiment, both the matrix and the fracture of the gas reservoir reservoir are relatively developed, and the gas-water seepage channels include both the matrix and the fracture, belonging to the dual-permeability channel mode. After analysis, the weights of the average probability attribute volume of the matrix and the average probability attribute volume of the fracture are both set to 1. According to the following formula, the average probability attribute volumes of the matrix and the fracture are integrated into a maximum probability water invasion channel distribution model, and the prediction of the water invasion channel can be completed:

[0139]

[0140] In the formula, is the water invasion channel probability value of a single grid, in decimals; P im is the average probability of a single grid of the matrix, in decimals; y1 is the weight of P im ; P if is the average probability of a single grid of the fracture, in decimals; y2 is the weight of P if

[0141] The water invasion channel model of the dual-porosity reservoir water drive gas reservoir predicted in this embodiment is as Figure 8 shown.

[0142] Perform numerical simulation on this water invasion channel model, compare the calculation results with the production performance. The historical fitting results show that: the fitting error of the formation pressure is 0.47%, there is no fitting error in the daily gas production, the fitting error of the wellhead oil pressure is 4.28%, and the fitting error of the daily water production is 4.5%. It shows that the water invasion channel model of the dual-porosity reservoir water drive gas reservoir has high prediction accuracy, and this prediction method is reliable, achieving the purpose of comprehensively, intuitively and finely predicting the geometric shape and spatial configuration relationship of the water invasion channel with evaluation parameters.

[0143] If there are new wells drilled in the study area, the water invasion channel distribution model can also be further improved by the above method in combination with the new well data.

[0144] Embodiment 2 A water invasion channel prediction device for a dual-porosity reservoir water drive gas reservoir​

[0145] This embodiment discloses a water invasion channel prediction device for a dual - medium reservoir water - drive gas reservoir. The prediction device includes a memory and a processor, as well as a computer program stored on the memory and running on the processor. The processor is coupled to the memory, and when the processor executes the computer program, it implements the water invasion channel prediction method for the dual - medium reservoir water - drive gas reservoir of the present invention.

[0146] The memory is used to store non - transient computer - readable instructions. Specifically, the memory may include one or more computer program products, and these computer program products may include various forms of computer - readable storage media, such as volatile memory and / or non - volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non - volatile memory may include, for example, read - only memory (ROM), hard disk, flash memory, etc.

[0147] The processor may be a central processing unit (CPU) or other forms of processing units with data - processing capabilities and / or instruction - execution capabilities, and can control other components in the electronic device to perform desired functions. The processor is used to run the computer - readable instructions stored in the memory.

[0148] Those skilled in the art should understand that, in order to solve the technical problem of how to obtain good user - experience effects, this embodiment may also include well - known structures such as communication buses, interfaces, etc., and these well - known structures should also be included in the protection scope of this disclosure.

[0149] For the detailed description of this embodiment, reference may be made to the corresponding descriptions in the foregoing embodiments, and details will not be repeated here.

Claims

1. A method for predicting water invasion channels in a water drive gas reservoir with a dual-medium reservoir, characterized in that, It includes the following steps carried out sequentially: S1. Jointly apply well-seismic interpretation data and dynamic data to establish a matrix-fracture dual-porosity model; S2. Calculate the fracture equivalent permeability model by the Oda method and the matrix permeability model by the pore-permeability relationship formula for each layer and microfacies respectively; S3. Quantitatively describe the heterogeneity characterization parameters of the matrix permeability model and the fracture equivalent permeability model in a grid manner; S4. Describe the degree of heterogeneity between the geological characteristic parameter grids of the matrix and fractures respectively; S5. Set the weights of the geological characteristic parameters respectively, and calculate the average probability attribute volumes of the matrix and fractures; S6. Fuse the average probability attribute volumes of the matrix and fractures into a maximum probability water invasion channel distribution model, and the prediction of the water invasion channel can be completed; The heterogeneity characterization parameters include the breakthrough coefficient, range difference, and coefficient of variation; the geological characteristic parameters include permeability and heterogeneity characterization parameters.

2. The water invasion channel prediction method for a dual-medium reservoir water drive gas reservoir according to claim 1, wherein The specific steps of S1 include: S11. Jointly apply the seismic interpretation data and logging interpretation data of the study area to establish a stratigraphic framework model; S12. Using the facies-controlled seismic inversion data volume matched with the dynamic data as a constraint, apply multi-well variogram analysis, and use the sequential indicator simulation method to establish a facies model; use the Gaussian random method with the facies model as a constraint, and use the co-kriging method for deterministic simulation with the facies-controlled inversion porosity plan as a plane constraint, and combine the two to calculate the matrix porosity model; S13. Using the imaging logging interpretation conclusion as a basic parameter, classify the fractures according to their occurrence and small layers, and use the fracture prediction plan and fault distance volume as constraints to randomly simulate the fracture intensity model; use the fracture intensity model as a constraint to randomly simulate and establish the DFN model of various fractures.

3. The water invasion channel prediction method for a dual-medium reservoir water drive gas reservoir according to claim 2, wherein The specific steps of S2 include: S21. Establish cross-plots of porosity and permeability for different sedimentary microfacies of each small layer respectively, and fit the relationship curve between porosity and permeability to obtain a power function: K = a·φ b In the formula, K is the permeability; φ is the porosity; a is the power function coefficient; b is the power function exponent Use the power function to calculate the permeability of different sedimentary microfacies of each small layer in the matrix porosity model to obtain the matrix permeability model; S22. Interpret the fracture aperture through imaging logging data, and calculate the fracture sheet permeability according to the fracture plane flow theory formula. The calculation formula is as follows: K f = b 2 / 12 where K f is the permeability of the fracture slice; b is the fracture aperture Use the Oda method in the DFN model to calculate the fracture equivalent permeability model.

4. A method for predicting water invasion channels in a dual-medium reservoir water drive gas reservoir according to any one of claims 1-3, characterized in that, The specific steps of S3 include the following operations on the matrix permeability model and the fracture equivalent permeability model respectively: S31. Statistically calculate the average permeability value within the entire grid range, and then use the following formula to calculate the breakthrough coefficient of each grid to complete the quantitative description of the breakthrough coefficient in a grid manner: T K = K n / K mean where T k is the breakthrough coefficient of a single grid; K n is the permeability of a single grid; K mean is the average permeability within the entire grid range; S32. Statistically calculate the minimum permeability value within the entire grid range, and then use the following formula to calculate the range difference of each grid to complete the quantitative description of the range difference in a grid manner: J K = K n / K min Where, J k is the range of a single grid; K n is the permeability of a single grid; K min is the minimum value of the permeability within the entire grid range; S33. Statistically calculate the average permeability value within the entire grid range, and then use the following formula to calculate the coefficient of variation of each grid to complete the quantitative description of the coefficient of variation in a grid manner: V K = (K n - K mean ) / K mean Where, V k is the coefficient of variation of a single grid; K n is the permeability of a single grid; K mean is the average permeability within the entire grid range.

5. A method for predicting water invasion channels in a dual-medium reservoir water drive gas reservoir according to any one of claims 1-3, characterized in that, The specific steps of S4 include: According to the frequency distribution histograms of various geological characteristic parameters, the geological characteristic parameter values are divided into homogeneous, heterogeneous, and strongly heterogeneous, and the heterogeneity degrees between the geological characteristic parameter grids of the matrix and fractures are described respectively.

6. A method for predicting water invasion channels in a water drive gas reservoir with a dual-medium reservoir, according to any one of claims 1-3, characterized in that The specific steps of step S5 include: S51. Calibrate the single-well discretized permeability logging curve with the core experimental permeability, and use the following formula to verify the correlation coefficients between the calibrated permeability logging curve and each heterogeneity characterization parameter of the matrix respectively: where U k is the well log permeability value of single - well discretization; H mk is a characterization parameter of a certain matrix heterogeneity; Cov(U k , H mk ) is the covariance of U k and H mk ; Var[U k is the variance of U k , Var[H mk is the variance of H mk ; S52. Use the following formula to verify the correlation coefficients between the single-well discretized fracture intensity curve and each heterogeneity characterization parameter of the fractures respectively: Wherein, I f is the fracture intensity value of single well discretization; H fk is a characterization parameter of the heterogeneity of a certain fracture; Cov(I f , H fk ) is the covariance of I f and H fk ; Var[I f is the variance of I f , and Var[H fk is the variance of H fk ; S53. Set the weights of various geological characteristic parameters respectively: the weight of permeability is 1; Heterogeneity characterization parameter: If the correlation coefficient r = 1, the weight of this heterogeneity characterization parameter is 1; if 0.5 ≤ r < 1, the weight of this heterogeneity characterization parameter is a value greater than 0.5 and less than 1; if 0 < r ≤ 0.5, the weight of this heterogeneity characterization parameter is a value greater than 0 and less than 0.5; if r = 0, this heterogeneity characterization parameter cannot be used to establish the average probability attribute body; S54. Calculate the average probability attribute bodies of the matrix and fractures according to the following two formulas: In the formula, is the weighted average of a single grid; K n is the permeability of a single grid; f1 is the weight of K n ; T k is the breakthrough coefficient of a single grid; f2 is the weight of T k ; J k is the range of a single grid; f3 is the weight of J k ; V k is the coefficient of variation of a single grid; f4 is the weight of V k ; Where, P i is the average probability of a single grid; is the weighted average of a single grid; is the maximum weighted average among all grids.

7. A method for predicting water invasion channels in a water-drive gas reservoir with a dual-medium reservoir, according to any one of claims 1-3, characterized in that The specific steps of step S6 include: Set the weights of the average probability attribute body of the matrix and the average probability attribute body of the fractures, and fuse the average probability attribute bodies of the matrix and fractures into the maximum probability water invasion channel distribution model according to the following formula, and the prediction of the water invasion channel can be completed: In the formula, is the probability value of a single-grid water invasion channel; P im is the average probability of a single grid in the matrix; y1 is the weight of P im ; P if is the average probability of a single grid in the fracture; y2 is the weight of P if .

8. A water invasion channel prediction device for a water drive gas reservoir in a dual-medium reservoir, characterized in that The prediction device includes a memory and a processor, and a computer program stored on the memory and running on the processor. The processor is coupled to the memory, and when the processor executes the computer program, it implements the water invasion channel prediction method for the dual-porosity reservoir water drive gas reservoir according to any one of claims 1-7.

Images