Data and physical dual-driven water drive dynamic fast prediction method

Abstract

This invention relates to the field of oilfield development technology, specifically to a rapid prediction method for water-driven dynamics driven by both data and physical parameters. The method includes the following steps: collecting geological and development-related physical parameters of the target water-drive reservoir; establishing a water-driven dynamics prediction model; calculating water-driven dynamics indices for the target well area based on the obtained physical parameters and the model; and calculating the overall coefficient of determination R of the obtained water-driven dynamics indices. 2 , will R 2 Compare with the set threshold, if R 2 If R ≥ the threshold, then the established water-driven prediction model is used for prediction; if R 2 If the value is less than the threshold, adjust the parameters of the water-driven model and repeat the above steps until R < 0. 2 ≥Threshold. The method described in this invention establishes a prediction model based on relevant physical parameters of water-drive reservoirs and corrects the calculation accuracy of the obtained prediction model, realizing dual-driven prediction based on physics and data, thereby improving the accuracy and speed of prediction.

Description

Technical Field

[0001] This invention relates to the field of oilfield development technology, specifically to a rapid prediction method for water-driven states driven by both data and physical factors. Background Technology

[0002] Water-driven dynamic prediction is a crucial aspect of oilfield development, significantly impacting production and operation. Existing methods primarily include numerical simulation based on physical models and statistical methods. While numerical simulation offers high accuracy, it is computationally intensive, requiring substantial resources and time. Statistical methods, though fast, suffer from lower accuracy, failing to meet the demands of practical production. Therefore, developing a rapid and accurate water-driven dynamic prediction method remains a critical challenge in oilfield development.

[0003] Patent CN201910997691.X discloses a rapid prediction method for the secondary enrichment rate of remaining oil in water-driven reservoirs. This method first selects well areas for the application of secondary enrichment technology based on reservoir development characteristics. Then, it comprehensively determines the key parameter values ​​affecting the secondary enrichment rate of remaining oil in the selected well areas by integrating seismic data, core test data, logging data, well test data, and development dynamic data. Finally, it predicts the secondary enrichment rate of remaining oil. This method can predict the secondary enrichment rate of remaining oil, but it cannot predict water-driven indicators such as well production and water cut.

[0004] Patent CN111810101A discloses a method and apparatus for dynamic analysis of water-driven reservoirs. This method first collects data such as cumulative oil production at different times in the reservoir, then plots a data point graph showing the relationship between the logarithm of the apparent cumulative oil production and the logarithm of the cumulative water production, and performs linear regression to obtain the characteristic curve relationship of the water-driven reservoir. This method belongs to a data-driven dynamic prediction method for development, capable of predicting the future water-driven state under stable injection and production parameters in the well area, but unable to predict the water-driven state under different injection and production parameter settings in the future.

[0005] Patent CN202011046642.7 discloses a method and system for calculating development indicators of water-driven oil reservoirs. This method first collects reservoir geological characteristics and development dynamic data, and then iteratively calculates the development indicators corresponding to each time step based on the principle of material balance. While the theory of this method is clear and explicit, errors often occur in the process of collecting reservoir geological parameters, leading to discrepancies between the calculated results and the actual situation. Summary of the Invention

[0006] To address the aforementioned problems, the main objective of this invention is to provide a rapid prediction method for water-driven reservoirs driven by both data and physical parameters. The method described in this invention establishes a prediction model based on relevant physical parameters of water-driven reservoirs and corrects the calculation accuracy of the obtained prediction model, achieving dual-driven prediction based on both physical parameters and data, thereby improving both the accuracy and speed of prediction.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, this invention provides a rapid prediction method for water-driven dynamics driven by both data and physical parameters. The method includes the following steps: collecting geological and development-related physical parameters of the target water-drive reservoir; establishing a water-driven dynamics prediction model; calculating water-driven dynamics indices for the target well area based on the obtained physical parameters and the model; and calculating the overall coefficient of determination R0 of the obtained water-driven dynamics indices. 2 , will R 2 Compare with the set threshold, if R 2 If R ≥ the threshold, then the established water-driven prediction model is used for prediction; if R 2 If the value is less than the threshold, adjust the parameters of the water-driven model and repeat the above steps until R... 2 ≥ Threshold.

[0009] Furthermore, the collection of geological and development-related physical parameters of the target water-drive reservoir includes: collecting oil-water viscosity, relative permeability curves, and number of sub-layers in the target well area; collecting the perforation history of each well, historical oil production rate, water production rate, and water injection rate of each well; and collecting the well location coordinates, permeability, porosity, and sub-layer thickness of each well in each sub-layer.

[0010] Furthermore, if data on oil-water viscosity, phase permeability curves, permeability, porosity, and sublayer thickness cannot be collected or are partially missing, an estimate is performed.

[0011] Furthermore, establishing a water-driven prediction model includes the following steps: determining the injection-production relationship of the target well area based on the well location coordinates of each well in each sub-layer and the perforation history of each well; for oil and water wells with injection-production relationships, delineating injection-production control units, determining the geological parameters of each injection-production control unit, allocating the injection-production fluid volume of each well to its corresponding injection-production units, and for each injection-production unit, calculating the cumulative oil production of each injection-production unit based on reservoir engineering theory, and establishing the resulting water-driven prediction model.

[0012] Furthermore, the injection-production relationship of the target well area is determined according to the following formula:

[0013]

[0014] In the formula, Let t represent the injection-production connectivity between well i and well j in layer l at time t, where 0 represents an injection-production relationship and 1 represents no injection-production relationship. and These represent the perforation status of wells i and j at time t in layer l, where 0 represents no perforation and 1 represents perforation; d max This represents the well spacing threshold; wells with a spacing exceeding this threshold cannot establish an injection-production relationship. i,l and x j,l Let x and y be the x coordinates of well i and well j at time t in layer l, respectively. i,l and y j,l Let y be the y coordinates of well i and well j at time t in the l-th layer; and These are the injection and production indicators for wells i and j at time t, respectively. 0 represents well shutdown, 1 represents water injection, and -1 represents production.

[0015] Furthermore, the geological parameters include the average permeability, average porosity, average effective thickness, and unit pore volume of the injection-production control unit. The calculation formulas for each parameter are as follows:

[0016] k i,j,l =(k i,l +k j,l ) / 2

[0017] φ i,j,l =(φ i,l +φ j,l ) / 2

[0018] h i,j,l =(h i,l +h j,l ) / 2

[0019] V i,j,l =A i,j,l h i,j,l φ i,j,l

[0020] In the formula, k i,j,l Let k be the average permeability of the injection-production unit formed by wells i and j in the lth layer; i,l and k j,l The permeability of wells i and j in the l-th layer are respectively; φ i,j,l φ represents the average porosity of the injection-production unit formed by wells i and j in the lth layer; i,l and φ j,l The porosity of wells i and j in the l-th layer are respectively; h i,j,l h represents the average effective thickness of the injection-production unit formed by wells i and j in the lth layer. i,l and h j,l The effective thicknesses of wells i and j in the l-th layer are respectively; V i,j,l Let A be the pore volume of the injection-production unit formed by wells i and j in the lth layer; i,j,lLet be the control area of ​​the injection-production unit formed by wells i and j in the l-th layer.

[0021] Furthermore, the injection and production fluid volumes of each well are allocated to their respective injection and production units, with the specific allocation rules as follows:

[0022]

[0023] In the formula, Q i,j,l Q represents the amount of injection and production fluid split from well i in the l-th layer towards well j; i J represents the injection and production volume of well i; J represents the total number of wells with injection and production relationships with well i; and L represents the total number of layers.

[0024] Furthermore, the cumulative oil production of each injection and production unit is obtained according to the following formula:

[0025]

[0026] In the formula, W i,j,l μ represents the cumulative water injection volume of well i in the direction from well j in layer l; o and μ w ρ represents the viscosity of oil and water, respectively; c and d are characteristic parameters of the phase permeability curve; S wc The bound water saturation; N i,j,l The cumulative oil production of well i in the direction from well j in layer l.

[0027] Furthermore, the water-driven state indicators include liquid production, oil production, water cut, and water injection volume, and the overall coefficient of determination R of the water-driven state indicators is... 2 It is obtained from the following formula:

[0028]

[0029] In the formula, R 2 The coefficient of determination for the whole population; These are the coefficients of determination for fluid production, oil production, water cut, and water injection, respectively; FLPR, FOPR, FWCT, and FWIR are the actual fluid production, oil production, water cut, and water injection in the well area, respectively. These are the well production, oil production, water cut, and water injection volume calculated using a physical-driven water-driven state calculation method, respectively; MSE is the mean square error calculation function; and VAR is the standard deviation calculation function.

[0030] In a second aspect, the present invention provides an apparatus for a data- and physical dual-driven rapid prediction method for water-driven states, the apparatus comprising: at least one processor and a memory communicatively connected to the at least one processor; wherein the memory stores instructions which are executed by the at least one processor to cause the at least one processor to implement the data- and physical dual-driven rapid prediction method for water-driven states as described in the first aspect above.

[0031] In a third aspect, the present invention provides a readable storage medium storing computer-executable instructions for causing a computer to execute the data- and physical dual-driven rapid prediction method for water-driven states as described in the first aspect above.

[0032] Compared with the prior art, the present invention has the following technical advantages:

[0033] The method described in this invention combines the advantages of both data-driven and physics-driven approaches, effectively improving prediction accuracy and applicability without requiring complex numerical simulation calculations. Attached Figure Description

[0034] Figure 1 A schematic diagram showing the amount of fluid split from each well to each injection and production unit;

[0035] Figure 2 This is a comparison chart of the calculated oil production and the actual oil production in Example 2;

[0036] Figure 3 This is a comparison chart of the liquid production volume obtained in Example 2 and the actual liquid production volume;

[0037] Figure 4 This is a comparison chart of the moisture content obtained in Example 2 and the actual moisture content;

[0038] Figure 5 This is a comparison chart of the water injection volume obtained in Example 2 and the actual water injection volume;

[0039] Figure 6 This is a comparison chart of the moisture content obtained after adjusting the model parameters in Example 2 and the actual moisture content;

[0040] Figure 7 This is a comparison chart of the oil production obtained after adjusting the model parameters in Example 2 and the actual oil production.

[0041] Figure 8 This is a graph showing the predicted oil production trend obtained from the model in Example 2. Detailed Implementation

[0042] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0043] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, and / or combinations thereof.

[0044] To enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention will be described in detail below with reference to specific embodiments.

[0045] Example 1

[0046] A fast prediction method for water-driven states driven by both data and physics includes the following steps:

[0047] Step 1: Collect geological and development parameters of water-drive oil reservoirs:

[0048] (1) For the target well area, collect the oil-water viscosity, relative permeability curve, and number of small layers in the well area;

[0049] (2) Collect the perforation history of each well and the historical oil production rate / water production rate / water injection rate of each well;

[0050] (3) Collect the well location coordinates, permeability, porosity, and thickness of each sub-layer for each well;

[0051] If data on oil-water viscosity, relative permeability curves, permeability, porosity, and sublayer thickness cannot be collected or are partially missing, an estimate can be given based on experience or data from surrounding wells. These uncertain data points can be automatically corrected in subsequent steps.

[0052] Step 2: Physics-driven water-driven state calculation:

[0053] (1) Based on the well location coordinates of each well in each sub-layer and the perforation history of each well, determine the injection-production relationship of the target well area. The specific judgment rules are as follows:

[0054]

[0055] In the formula, Let t represent the injection-production connectivity between well i and well j in layer l at time t, where 0 represents an injection-production relationship and 1 represents no injection-production relationship. and These represent the perforation status of wells i and j at time t in layer l, where 0 represents no perforation and 1 represents perforation; d max This represents the well spacing threshold; wells with a spacing exceeding this threshold cannot establish an injection-production relationship. i,l and x j,l Let x and y be the x coordinates of well i and well j at time t in layer l, respectively. i,l and y j,l Let y be the y coordinates of well i and well j at time t in the l-th layer; and These are the injection and production indicators for wells i and j at time t, respectively. 0 represents well shutdown, 1 represents water injection, and -1 represents production.

[0056] (2) For oil and water wells with an injection-production relationship, define injection-production control units and determine the geological parameters of each control unit, including the average permeability, average porosity, average effective thickness, and unit pore volume. The specific calculation formula is as follows:

[0057] k i,j,l =(k i,l +k j,l ) / 2

[0058] φ i,j,l =(φ i,l +φ j,l ) / 2

[0059] h i,j,l =(h i,l +h j,l ) / 2

[0060] V i,j,l =A i,j,l h i,j,l φ i,j,l

[0061] In the formula, k i,j,l Let k be the average permeability of the injection-production unit formed by wells i and j in the lth layer; i,l and k j,l The permeability of wells i and j in the l-th layer are respectively; φ i,j,l φ represents the average porosity of the injection-production unit formed by wells i and j in the lth layer; i,l and φ j,l The porosity of wells i and j in the l-th layer are respectively; h i,j,l h represents the average effective thickness of the injection-production unit formed by wells i and j in the lth layer. i,l and h j,l The effective thicknesses of wells i and j in the l-th layer are respectively; V i,j,l Let A be the pore volume of the injection-production unit formed by wells i and j in the lth layer; i,j,lLet be the control area of ​​the injection-production unit formed by wells i and j in the l-th layer.

[0062] (3) Considering the differences in geological development attributes of each injection-production unit, the injection and production volume of each well is allocated to its corresponding injection-production unit. The specific allocation rules are as follows:

[0063]

[0064] In the formula, Q i,j,l Q represents the amount of injection and production fluid split from well i in the l-th layer towards well j; i J represents the injection and production volume of well i; J represents the total number of wells with injection and production relationships with well i; and L represents the total number of layers.

[0065] (4) For each injection-production unit, based on reservoir engineering theory, the variation law of cumulative oil production with cumulative water injection volume is calculated. The specific calculation formula is as follows:

[0066]

[0067] In the formula, W i,j,l μ represents the cumulative water injection volume of well i in the direction from well j in layer l; o and μ w ρ represents the viscosity of oil and water, respectively; c and d are characteristic parameters of the phase permeability curve; S wc The bound water saturation; N i,j,l The cumulative oil production of well i in the direction from well j in layer l.

[0068] (5) Based on the cumulative water injection volume, cumulative oil production, and other water-driven dynamic indicators of each injection-production unit, calculate the overall water-driven dynamic indicators of the target well area, including well area fluid production, oil production, water cut, and water injection volume. The specific calculation formula is as follows:

[0069]

[0070] In the formula, The figures are the well production, oil production, water cut, and water injection volume calculated by the physical-driven water-driven state calculation method.

[0071] Step 3: Improve the accuracy of water-driven state calculations based on data-driven approaches:

[0072] By comparing the well production, oil production, water cut, and water injection volume calculated by the physical-driven water-driven dynamic calculation method with the actual well production, oil production, water cut, and water injection volume, the coefficient of determination for each indicator is calculated, and the average is taken to obtain the overall coefficient of determination. The specific calculation formula is as follows:

[0073]

[0074] In the formula, R2 The coefficient of determination for the whole population; , respectively, are the coefficients of determination for fluid production, oil production, water cut, and water injection; FLPR, FOPR, FWCT, and FWIR are the actual fluid production, oil production, water cut, and water injection in the well area, respectively; MSE is the mean square error calculation function; VAR is the standard deviation calculation function.

[0075] Step 4: Determine the coefficient of determination R. 2 If the set threshold is exceeded, it indicates that the accuracy of the physics-driven water-driven state calculation method meets the requirements and can be used for subsequent water-driven state prediction. The current oil-water viscosity, relative permeability curve, permeability, porosity, and sublayer thickness data should be saved. Otherwise, it indicates that the accuracy of the physics-driven water-driven state calculation method does not meet the requirements, and the next step is needed: adjusting the oil-water viscosity, relative permeability curve, permeability, porosity, or sublayer thickness data. Specific adjustment methods can include simulated annealing, particle swarm optimization, and genetic algorithms. Taking simulated annealing as an example, the specific adjustment formula is as follows:

[0076]

[0077]

[0078] c * =c + Δc

[0079] d * =d+Δd

[0080]

[0081] In the formula, Δμ o and Δμ w Δc and Δd are the viscosity perturbation terms for oil and water, respectively; Δc and Δd are the characteristic parameter perturbation terms of the relative permeability curve; Δk i,j,l For wells i and j, the permeability perturbation term in the l-th layer is Δφ. i,j,l For wells i and j, the porosity perturbation term in the l-th layer is Δh. i,j,l Let be the effective thickness perturbation term for wells i and j in the l-th layer. and c* and d* are the adjusted viscosity of oil and water, respectively; c* and d* are the characteristic parameters of the adjusted phase permeability curve. The adjusted permeability of wells i and j in the l-th layer; The adjusted porosity of wells i and j in the l-th layer; The effective thickness of wells i and j after adjustment in the l-th layer.

[0082] Subsequently, the water-driven dynamic calculation method based on physical drive from step 2 was used to recalculate the well production, oil production, water cut, and water injection volume. These values ​​were then compared with the actual production, oil production, water cut, and water injection volume in the well area to obtain new coefficients of determination.

[0083] Example 2

[0084] The method for rapid prediction of water-driven states using both data and physical factors, as described in Example 1, includes the following steps:

[0085] 1. Collection of geological and development parameters for water-drive oil reservoirs:

[0086] 1-1: For the target well area, the oil-water viscosity of the well area was 10cp and 1cp, the relative permeability curves are shown in Table 1, and the number of sub-layers was 3;

[0087] Table 1

[0088]

[0089]

[0090] 1-2: Collect the perforation history of each well as shown in Table 2, and the history of oil production rate / water production rate / water injection rate of each well as shown in Table 3;

[0091] Table 2

[0092] 2020-01-01 W1 1，2，3 2020-01-01 W2 1，2，3 2020-01-01 W3 1，2，3 2020-01-01 W4 1，2，3 2020-01-01 W5 1，2，3

[0093] Table 3

[0094]

[0095] 1-3: Well location coordinates, permeability, porosity, and thickness of each sub-layer for each well are shown in Table 4.

[0096] Table 4

[0097]

[0098]

[0099] 1-4: If data on oil-water viscosity, relative permeability curves, permeability, porosity, and sublayer thickness cannot be collected or are partially missing, an estimate can be given based on experience or data from surrounding wells. These uncertain data can be automatically corrected in subsequent steps.

[0100] As can be seen from Table 4, permeability data for well W5 is missing. Therefore, the average permeability of surrounding wells is used as the estimated value for well W5 to complete the data. The complete data is shown in Table 5.

[0101] Table 5

[0102]

[0103] 2. Physics-driven water-driven state calculation:

[0104] 2-1: Based on the well location coordinates of each well in each sub-layer and the perforation history of each well, determine the injection-production relationship of the target well area. The specific judgment rules are as follows:

[0105]

[0106] In the formula, Let t represent the injection-production connectivity between well i and well j in layer l at time t, where 0 represents an injection-production relationship and 1 represents no injection-production relationship. and These represent the perforation status of wells i and j at time t in layer l, where 0 represents no perforation and 1 represents perforation; d max This represents the well spacing threshold; wells with a spacing exceeding this threshold cannot establish an injection-production relationship. i,l and x j,l Let x and y be the x coordinates of well i and well j at time t in layer l, respectively. i,l and y j,l Let y be the y coordinates of well i and well j at time t in the l-th layer; and These are the injection and production indicators for wells i and j at time t, respectively. 0 represents well shutdown, 1 represents water injection, and -1 represents production.

[0107] Taking t=2020-01 as an example, the injection-sampling relationship results of the first layer are shown in Table 6.

[0108] Table 6

[0109]

[0110] 2-2: For oil-water wells with an injection-production relationship, define injection-production control units and determine the geological parameters of each control unit, including the average permeability, average porosity, average effective thickness, and unit pore volume. The specific calculation formulas are as follows:

[0111] k i,j,l =(k i,l +k j,l ) / 2

[0112] φ i,j,l =(φ i,l +φ j,l ) / 2

[0113] h i,j,l =(h i,l +h j,l ) / 2

[0114] V i,j,l =A i,j,l h i,j,l φ i,j,l

[0115] In the formula, k i,j,l Let k be the average permeability of the injection-production unit formed by wells i and j in the lth layer; i,l and k j,l The permeability of wells i and j in the l-th layer are respectively; φ i,j,l φ represents the average porosity of the injection-production unit formed by wells i and j in the lth layer; i,l and φ j,l The porosity of wells i and j in the l-th layer are respectively; h i,j,l h represents the average effective thickness of the injection-production unit formed by wells i and j in the lth layer. i,l and h j,l The effective thicknesses of wells i and j in the l-th layer are respectively; V i,j,l Let A be the pore volume of the injection-production unit formed by wells i and j in the lth layer; i,j,l Let be the control area of ​​the injection-production unit formed by wells i and j in the l-th layer.

[0116] The calculated geological parameters for each injection-production unit in this well area are shown in Table 7.

[0117] Table 7

[0118] 5,1,1 100 0.2 2 9000 5,2,1 100 0.2 2 9000 5,3,1 100 0.2 2 9000 5,4,1 100 0.2 2 9000 5,1,2 200 0.2 2 9000 5,2,2 200 0.2 2 9000 5,3,2 200 0.2 2 9000 5,4,2 200 0.2 2 9000 5,1,3 300 0.2 2 9000 5,2,3 300 0.2 2 9000 5,3,3 300 0.2 2 9000 5,4,3 300 0.2 2 9000

[0119] 2-3: Considering the differences in geological development attributes of each injection-production unit, the injection and production volume of each well is allocated to its corresponding injection-production unit. The specific allocation rules are as follows:

[0120]

[0121] In the formula, Q i,j,l Q represents the amount of injection and production fluid split from well i in the l-th layer towards well j; i J represents the injection and production volume of well i; J represents the total number of wells with injection and production relationships with well i; and L represents the total number of layers.

[0122] Well W5 is a water injection well with an injection rate of 100m. 3 / d, Wells W1 to W4 are production wells with a production rate of 25m / d. 3 / d, the amount of fluid split from each well to each injection-production unit is calculated as follows: Figure 1 As shown.

[0123] 2-4: For each injection-production unit, based on reservoir engineering theory, the variation law of cumulative oil production with cumulative water injection volume is calculated. The specific calculation formula is as follows:

[0124]

[0125] In the formula, W i,j,l μ represents the cumulative water injection volume of well i in the direction from well j in layer l; o and μ w ρ represents the viscosity of oil and water, respectively; c and d are characteristic parameters of the phase permeability curve; S wc The bound water saturation; N i,j,l The cumulative oil production of well i in the direction from well j in layer l.

[0126] 2-5: Based on the cumulative water injection volume, cumulative oil production and other water-driven state indicators of each injection and production unit, calculate the overall water-driven state indicators of the target well area, including the well area production volume change curve, oil production volume change curve, water cut change curve and water injection volume change curve.

[0127] 3. Improve the accuracy of water-driven dynamic calculations based on data-driven approaches:

[0128] 3-1: Using the physical-driven water-driven state calculation method in step 2, calculate the water-driven state indices of the target well area, including the well area's fluid production, oil production, water cut, and water injection volume.

[0129] 3-2: Compare the well production, oil production, water cut, and water injection volume calculated by the physical-driven water-driven dynamic calculation method with the actual well production, oil production, water cut, and water injection volume. Calculate the coefficient of determination for each indicator and take the average to obtain the overall coefficient of determination. The specific calculation formula is as follows:

[0130]

[0131] In the formula, R 2 The coefficient of determination for the whole population; These are the coefficients of determination for fluid production, oil production, water cut, and water injection, respectively; FLPR, FOPR, FWCT, and FWIR are the actual fluid production, oil production, water cut, and water injection in the well area, respectively. These are the well production, oil production, water cut, and water injection volume calculated using a physical-driven water-driven state calculation method, respectively; MSE is the mean square error calculation function; and VAR is the standard deviation calculation function.

[0132] The values ​​are 1, 0.5504, 0.7477, and 1 respectively. R 2 It is 0.8245.

[0133] 3-3: Determine the coefficient of determination R 2If the set threshold is exceeded, it indicates that the accuracy of the physics-driven water-driven state calculation method meets the requirements and can be used for subsequent water-driven state prediction. The current oil-water viscosity, relative permeability curve, permeability, porosity, and sublayer thickness data should be saved. Otherwise, it indicates that the accuracy of the physics-driven water-driven state calculation method does not meet the requirements, and the next step needs to be performed.

[0134] The threshold for R² is set to 0.9. Currently, R²... 2 The value is 0.8245, indicating that the accuracy of the physical-driven water-driven state calculation method does not meet the requirements, and it is necessary to proceed to the next step.

[0135] 3-4: Adjust the oil-water viscosity, relative permeability curve, permeability, porosity, and sublayer thickness data. Using the physical-driven water-driven state calculation method from step 2, recalculate the well area production, oil production, water cut, and water injection volume. Compare these figures with the actual production, oil production, water cut, and water injection volume in the well area to obtain new coefficients of determination. Return to step 3-3.

[0136] After adjusting the crude oil viscosity from 10 cp to 15 cp and recalculating the coefficient of determination, R was found to be... 2 It reached 0.95. Returning to step 3-3, we found that R... 2 Exceeding the set threshold indicates that the accuracy of the physics-driven water-driven state calculation method meets the requirements and can be used for subsequent water-driven state prediction.

[0137] 4. Set a new injection and production system for oil and water wells. Use the oil and water viscosity, relative permeability curve, permeability, porosity, and sublayer thickness data that meet the accuracy requirements after adjustment in step 3, as well as the water-driven state calculation method based on physical drive in step 2, to calculate water-driven state indicators such as well area production, oil production, water cut, and water injection.

[0138] Figure 8 This is a chart showing the predicted trend of oil production over the next 500 days.

[0139] Compared with purely physical-driven water-driven state prediction methods R 2 The value is 0.8245, which represents the R value of the fast prediction method for water-driven states driven by both data and physics in this invention. 2 The accuracy reached 0.95, effectively improving prediction precision. Compared to numerical simulation, this method takes 2.1 seconds to complete one calculation, while reservoir numerical simulation takes 11.5 seconds, demonstrating a significant improvement in computational efficiency.

[0140] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A rapid prediction method for water-driven states driven by both data and physics, characterized in that, It includes the following steps: Collect geological and development-related physical parameters of the target water-drive reservoir; A water-driven dynamic prediction model was established, and water-driven dynamic indices for the target well area were calculated based on the obtained physical parameters and the model. The overall coefficient of determination R of the obtained water-driven dynamic indices was calculated. 2 , will R 2 Compare with the set threshold, if R 2 If R ≥ the threshold, then the established water-driven prediction model is used for prediction; if R 2 If the value is less than the threshold, adjust the parameters of the water-driven model and repeat the above steps until R... 2 ≥ Threshold.

2. The rapid prediction method for water-driven states based on both data and physical factors according to claim 1, characterized in that, The collection of geological and development-related physical parameters of the target water-drive reservoir includes: collecting oil-water viscosity, relative permeability curves, and number of sub-layers in the target well area; collecting the perforation history, historical oil production rate, water production rate, and water injection rate of each well; and collecting the well location coordinates, permeability, porosity, and sub-layer thickness of each well in each sub-layer.

3. The rapid prediction method for water-driven states driven by both data and physics as described in claim 2, characterized in that, If data on oil-water viscosity, phase permeability curves, permeability, porosity, and sublayer thickness cannot be collected or are partially missing, then an estimate should be made.

4. The rapid prediction method for water-driven states driven by both data and physics as described in claim 1, characterized in that, Establishing a water-driven prediction model includes the following steps: Based on the well location coordinates of each well in each sub-layer and the perforation history of each well, the injection-production relationship of the target well area is determined. For oil and water wells with injection-production relationships, injection-production control units are delineated, the geological parameters of each injection-production control unit are determined, and the injection and production fluid volumes of each well are allocated to their corresponding injection-production units. For each injection-production unit, based on reservoir engineering theory, the cumulative oil production of each injection-production unit is calculated, and a water-driven dynamic model is established. In the formula, These represent the well production, oil production, water cut, and water injection rate calculated using a physics-driven water-driven state calculation method, respectively; Q is the injection-production rate vector; and g() is the water-driven state model.

5. The rapid prediction method for water-driven states driven by both data and physics as described in claim 4, characterized in that, The injection-production relationship of the target well area is determined based on the following formula: In the formula, Let t represent the injection-production connectivity between well i and well j in layer l at time t, where 0 represents an injection-production relationship and 1 represents no injection-production relationship. and The values ​​represent the perforation status of wells i and j at time t in layer l, where 0 represents no perforation and 1 represents perforation. d max This represents the well spacing threshold; wells with a spacing exceeding this threshold cannot establish an injection-production relationship. i,l and x j,l Let x and y be the x coordinates of well i and well j at time t in layer l, respectively. i,l and y j,l Let y be the y coordinates of well i and well j at time t in the l-th layer; and These are the injection and production indicators for wells i and j at time t, respectively. 0 represents well shutdown, 1 represents water injection, and -1 represents production.

6. The rapid prediction method for water-driven states driven by both data and physics as described in claim 4, characterized in that, The geological parameters include the average permeability, average porosity, average effective thickness, and unit pore volume of the injection-production control unit. The calculation formulas for each parameter are as follows: k i,j,l =(k i,l +k j,l ) / 2 f i,j,l =(φ i,l +φ j,l ) / 2 h i,j,l =(h i,l +h j,l ) / 2 IN i,j,l =A i,j,l h i,j,l φ i,j,l In the formula, k i,j,l Let k be the average permeability of the injection-production unit formed by wells i and j in the lth layer; i,l and k j,l The permeability of well i and well j in the l-th layer are respectively; φ i,j,l φ represents the average porosity of the injection-production unit formed by wells i and j in the lth layer; i,l and φ j,l The porosity of wells i and j in the l-th layer are respectively; h i,j,l h represents the average effective thickness of the injection-production unit formed by wells i and j in the lth layer. i,l and h j,l The effective thicknesses of wells i and j in the l-th layer are respectively; V i,j,l Let A be the pore volume of the injection-production unit formed by wells i and j in the lth layer; i,j,l Let be the control area of ​​the injection-production unit formed by wells i and j in the l-th layer.

7. The rapid prediction method for water-driven states driven by both data and physics as described in claim 4, characterized in that, The injection and production fluid volumes of each well are allocated to their respective injection and production units, and the specific allocation rules are as follows: In the formula, Q i,j,l Q represents the amount of injection and production fluid split from well i in the l-th layer towards well j; i J represents the injection and production volume of well i; J represents the total number of wells with injection and production relationships with well i; and L represents the total number of layers.

8. The rapid prediction method for water-driven states driven by both data and physics as described in claim 4, characterized in that, The cumulative oil production of each injection and production unit is obtained according to the following formula: In the formula, W i,j,l μ represents the cumulative water injection volume of well i in the direction from well j in layer l; o and μ w ρ represents the viscosity of oil and water, respectively; c and d are characteristic parameters of the phase permeability curve; S wc The bound water saturation; N i,j,l The cumulative oil production of well i in the direction from well j in layer l.

9. The rapid prediction method for water-driven states driven by both data and physics as described in claim 1, characterized in that, The water-driven dynamic indicators include liquid production, oil production, water cut, and water injection volume. The overall coefficient of determination R of the water-driven dynamic indicators is... 2 It is obtained from the following formula: In the formula, R 2 The coefficient of determination for the whole population; These are the coefficients of determination for fluid production, oil production, water cut, and water injection, respectively; FLPR, FOPR, FWCT, and FWIR are the actual fluid production, oil production, water cut, and water injection in the well area, respectively. These are the well production, oil production, water cut, and water injection volume calculated using a physical-driven water-driven state calculation method, respectively; MSE is the mean square error calculation function; and VAR is the standard deviation calculation function.

10. A device for a rapid prediction method of water-driven states driven by both data and physics, characterized in that, include: At least one processor, and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that are executed by the at least one processor to cause the at least one processor to implement the data and physical dual-driven water-driven state fast prediction method as described in any one of claims 1-9 when executing the instructions.

11. A readable storage medium, characterized in that, The readable storage medium stores computer-executable instructions for causing a computer to execute the data- and physical dual-driven rapid prediction method for water-driven states as described in any one of claims 1-7.

Images