Optimization method of segmented injection and production parameters based on plane flow unit delineation

Abstract

The present invention relates to a method for optimizing segmented injection and production parameters of a water-drive oil reservoir, and in particular to a method for optimizing segmented injection and production parameters of a water-drive oil reservoir based on planar flow unit delineation. The method comprises the following steps: step 1. collecting geological and production data; step 2. establishing an initial description model of the production dynamics of a water-drive oil reservoir; step 3. performing automatic historical fitting of reservoir parameters, correcting the description model, and establishing an accurate description model of the production dynamics of a water-drive oil reservoir; step 4. utilizing the accurate description model of the water-drive development dynamics, with the aid of an optimization algorithm, and taking production optimization as the goal, obtaining an optimal injection and production parameter scheme. The method steps of the present invention are simple to operate, and on the premise of being able to accurately reflect the injection and production relationship, can quickly realize accurate optimization of the injection and production parameters of a water-drive oil reservoir, thereby overcoming existing deficiencies.

Description

Technical Field

[0001] The invention relates to a method for optimizing segmented injection and production parameters of a water drive oil reservoir, and in particular to a method for optimizing segmented injection and production parameters of a water drive oil reservoir based on plane flow unit delineation. Background Art

[0002] Old water-driven oil fields, represented by Shengli Oilfield, have many oil-bearing strata. With the continuous development, the reservoir heterogeneity has further intensified, the development has become increasingly prominent, the crude oil production has shown a clear downward trend, and the development effect has deteriorated year by year. The main reason is that due to the differences in the physical properties of the reservoirs in each small layer in the stratum, their water absorption capacity is different. When water is injected, most of the water injected into the oil layer is absorbed by the high permeability layer with a small thickness. The water injection absorption profile is very uneven, the water injection utilization rate is low, and the invalid water circulation is serious. Reasonable adjustment of injection and production parameters can effectively improve the water drive effect and achieve stable oil and water control.

[0003] Chinese patent application CN109872007A discloses a multi-objective optimization method for reservoir injection and production parameters based on a support vector machine proxy model. The method generates a certain number of injection and production schemes as samples through numerical simulation software; inputs the samples into the least squares support vector machine to form a substitute model. If there is no convergence in this process, a particle swarm algorithm is used to find the optimal solution to achieve convergence; after generating an initial injection and production parameter population of size M based on the least squares support vector machine, a suitable objective function, pareto level and crowding distance are selected, and then the non-dominated sorting multi-objective optimization genetic algorithm NSGA-Ⅱ with an elite strategy is used to optimize the injection and production parameters to obtain a pareto solution set; according to the requirements of the objective function, the optimal reservoir injection and production parameters are found in the pareto solution set. Although the invention realizes the application of the method for multi-objective optimization design of reservoir injection and production parameters, it improves efficiency while ensuring the prediction accuracy of the proxy model.

[0004] Chinese patent application CN110924908A discloses a method for determining injection and production parameters of a water-drive oil reservoir and a computer-readable storage medium, the method comprising: the step of establishing a streamline simulation model: establishing a streamline simulation model of a target oil reservoir; the step of generating injection and production parameters: randomly generating injection and production parameters under the constraints of injection and production parameters; the step of determining the streamline oil displacement capacity value: predicting the instantaneous flow field distribution of the streamline simulation model with injection and production parameters through a streamline simulator, and extracting streamline characteristic parameters; determining the streamline oil displacement capacity value of the injection and production parameters based on the streamline characteristic parameters; and the step of evaluating the injection and production parameters: judging whether the streamline oil displacement capacity value reaches the optimization target relative to the initial value of the streamline oil displacement capacity based on the optimization objective function; wherein, if the optimization target is reached, determining the injection and production parameters as output injection and production parameters; otherwise, taking the streamline oil displacement capacity value as the initial value of the streamline oil displacement capacity, and returning to the step of generating the injection and production parameters. Rapid determination of the injection and production parameters is achieved.

[0005] Chinese patent application CN110439515A discloses a method for optimizing injection and production parameters, comprising: step 1, based on the production dynamic data of the injection wells and the production wells in the block, obtaining the dynamic well-to-well connectivity coefficient between the injection wells and the production wells in the block; step 2, constructing an injection and production parameter optimization model with the maximum oil production of the production wells in the block as the objective function; step 3, bringing the water content of the production wells in the block and the dynamic well-to-well connectivity coefficient into the injection and production parameter optimization model to obtain the optimal injection amount of the injection wells in the block. The present invention optimizes the monthly water injection amount of each water injection well in the oil field by constructing an injection and production parameter optimization model based on the dynamic well-to-well connectivity coefficient, taking the maximum oil production of the reservoir in the current block as the optimization target, and forming a dynamic optimization and control method to maximize the water injection efficiency, which can achieve balanced displacement of injected water and improve the final recovery rate.

[0006] The optimization problem of reservoir production parameters belongs to a large-scale optimization problem. In recent years, many scholars have used optimization algorithms to couple with reservoir numerical simulators to achieve the optimization of reservoir problems. Since the solution process needs to be combined with the reservoir numerical simulator, it involves the solution of large matrices, which is very cumbersome and the accuracy of the solution cannot be guaranteed. At the same time, this method is not very practical for engineering and technical personnel. The main reasons are that, first, not all reservoirs have established reservoir numerical models; second, the historical fitting work in the early stage of reservoir numerical simulation will take up a lot of time and energy, and the reservoir numerical simulation requires high professionalism and experience of operators. Summary of the invention

[0007] The main purpose of the present invention is to provide a segmented injection-production parameter optimization method based on planar flow unit delineation. The method steps of the present invention are simple to operate. Under the premise of being able to accurately reflect the injection-production relationship, it can quickly realize the accurate optimization of the injection-production parameters of water-displacement oil reservoirs, thus overcoming the existing shortcomings.

[0008] To achieve the above object, the present invention adopts the following technical solutions:

[0009] The present invention provides a segmented injection and production parameter optimization method based on planar flow unit delineation, which comprises the following steps:

[0010] Step 1. Collect geological and production data;

[0011] Step 2. Establish an initial description model of water-flooding reservoir production performance;

[0012] Step 3. Perform automatic history matching on reservoir parameters, modify the description model, and establish an accurate description model of water-drive reservoir production dynamics;

[0013] Step 4. Use the water drive development dynamic accurate description model, with the help of optimization algorithm, with the goal of optimal production, to obtain the optimal injection and production parameter plan.

[0014] Preferably, the geological data include reservoir physical property parameters, and the reservoir physical property parameters include porosity, permeability, reservoir thickness, saturation, pressure, oil-water density, oil-water viscosity, phase permeability curve and reservoir rock compression coefficient.

[0015] Preferably, the production data include oil and water well locations, vertical stratification conditions and historical production data.

[0016] Preferably, in step 2, the reservoir is delineated both vertically and horizontally according to the target reservoir's production and injection conditions, well location relationships, geological development parameters, etc., to obtain a series of single-layer inter-well flow units, and the production dynamics within the units are calculated using the two-phase oil recovery theory, thereby establishing an initial description model for water drive development dynamics.

[0017] Further preferably, the step of dual delineation of the target reservoir in longitudinal and horizontal directions also includes two cyclic iterations of longitudinal delineation and horizontal delineation.

[0018] Further preferably, vertically, the working system of each well in each layer is delineated according to the geological development parameters such as the production / injection amount of each well and each layer section and the seepage resistance of each layer; horizontally, different inter-well connected flow units are delineated on the plane according to the well position relationship and seepage resistance within the well group, and the production dynamics within the unit are calculated using the two-phase oil recovery theory.

[0019] Further preferably, the plane flow unit delineation of a single layer segment is described using the flow unit area A, the well spacing d, and the delineation angle θ; in an injection and production well group, there are m oil wells with the water well as the center, and the delineation angle of the flow unit corresponding to the i-th oil well is:

[0020] θ i =2πα i

[0021] Where α i is the demarcation coefficient of the i-th oil well, and its solution formula is:

[0022]

[0023] Where pi is the pressure difference between the injection well and the i-th oil well, MPa; m is the total number of production wells; Ri is the degree of production in the direction of the i-th oil well.

[0024] Further preferably, for a multi-layer reservoir, due to heterogeneity, the liquid volume of each layer also needs to be delineated, and the liquid volume of each layer is delineated according to the following formula:

[0025]

[0026] Where Ri is the demarcation coefficient of the i-th layer:

[0027]

[0028] Qi is the liquid volume in the i-th layer, Ki is the permeability of the i-th layer, Ai is the cross-sectional area of ​​the i-th layer, Δp is the injection-production pressure difference, Gi is the starting pressure of the i-th layer, a is the unit conversion coefficient, d is the well spacing, λro is the oil phase mobility, and λrw is the water phase mobility.

[0029] Preferably, in step 3, the CMA-ES algorithm is used to generate N sets of reservoir parameters, and the water-drive reservoir production dynamic initial description model is used for historical fitting, and the model reservoir parameters are continuously corrected and updated to obtain the optimal reservoir parameters, and further obtain the water-drive reservoir production dynamic accurate description model.

[0030] Preferably, in step 4, N sets of injection-production plans are generated using the MCS algorithm, and the production performance accurate description model of the water-drive reservoir is used to predict the production performance results to obtain the optimal injection-production plan.

[0031] Compared with the prior art, the present invention has the following advantages:

[0032] The present invention provides a simple and fast method for optimizing injection and production parameters of water-drive oil reservoirs. The present invention transforms the history matching problem into an optimal solution problem by establishing a dynamic initial model of water-drive oil reservoir production, thereby forming an automatic history matching technology, which can quickly and accurately correct the development dynamic description model, and realizes rapid and accurate optimization of injection and production parameters with the goal of optimal production through an optimization algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0034] Figure 1 It is a flow chart of a segmented injection-production parameter optimization method based on planar flow unit delineation according to a specific embodiment of the present invention;

[0035] Figure 2 A schematic diagram of delineating a flow unit according to a specific embodiment of the present invention;

[0036] Figure 3 This is a flowchart of automatic history matching according to a specific embodiment of the present invention;

[0037] Figure 4 This is a flow chart of injection-production parameter optimization according to a specific embodiment of the present invention;

[0038] Figure 5A graph of a predicted oil production rate curve before history matching and an actual oil production rate curve according to a specific embodiment of the present invention;

[0039] Figure 6 A graph showing a predicted oil production rate curve and an actual oil production rate curve after historical matching according to a specific embodiment of the present invention;

[0040] Figure 7 This is a comparison diagram of the effects of using the optimal injection and production parameters according to a specific embodiment of the present invention, A: changes in each oil well before and after optimization; B: changes in each water well before and after optimization. DETAILED DESCRIPTION

[0041] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0042] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations and / or combinations thereof.

[0043] Example 1 Segmented injection and production parameter optimization method based on planar flow unit delineation

[0044] like Figure 1 As shown, the method comprises the following steps:

[0045] Step 1. Collect geological and production data:

[0046] The reservoir physical property parameters include porosity, permeability, reservoir thickness, saturation, pressure, oil-water density, oil-water viscosity, phase permeability curve and reservoir rock compressibility coefficient.

[0047] The production data include oil and water well locations, vertical stratification and historical production data.

[0048] Step 2. Establish an initial description model of water-flooding reservoir production performance:

[0049] According to the target reservoir's production and injection conditions, well location relationships, geological development parameters, etc., the reservoir is delineated both vertically and horizontally to obtain a series of single-layer inter-well flow units. The two-phase flooding theory is used to calculate the production dynamics within the unit, thereby establishing an initial description model for water flooding development dynamics.

[0050] The double delineation steps of the target reservoir, both vertically and horizontally, also include two iterative links of vertical delineation and horizontal delineation: vertically, the working system of each well in each layer is delineated according to geological development parameters such as the production / injection volume of each well and each layer section and the seepage resistance of each layer; horizontally, different inter-well connected flow units are delineated on the plane according to the well position relationship and seepage resistance within the well group, and the production dynamics within the unit are calculated using the two-phase oil recovery theory.

[0051] The plane flow unit delineation of a single layer segment is described by the flow unit area A, the well spacing d, and the delineation angle θ. In an injection-production well group, there are m oil wells with the water well as the center, and the delineation angle of the flow unit corresponding to the i-th oil well is:

[0052] θ i =2πα i

[0053] Where α i is the demarcation coefficient of the i-th oil well, and its solution formula is:

[0054]

[0055] Where p i is the pressure difference between the injection well and the i-th oil well, MPa; m is the total number of production wells; R i is the recovery degree in the direction of the i-th oil well.

[0056] The flow conditions within each plane demarcated unit can be calculated using the Buckley-Leverett oil-water two-phase displacement theory.

[0057] For multi-layer reservoirs, due to heterogeneity, the liquid volume of each layer also needs to be delineated. The liquid volume of each layer is delineated according to the following formula:

[0058]

[0059] Where Ri is the demarcation coefficient of the i-th layer:

[0060]

[0061] Qi is the liquid volume in the i-th layer, Ki is the permeability of the i-th layer, Ai is the cross-sectional area of ​​the i-th layer, Δp is the injection-production pressure difference, Gi is the starting pressure of the i-th layer, a is the unit conversion coefficient, d is the well spacing, λro is the oil phase mobility, and λrw is the water phase mobility.

[0062] Step 3. Perform automatic history matching on reservoir parameters, modify the description model, and establish an accurate description model of water-drive reservoir production dynamics:

[0063] The CMA-ES algorithm is used to generate N sets of reservoir parameters, and the water-drive reservoir production dynamic initial description model is used for historical fitting. The model reservoir parameters are continuously corrected and updated to obtain the optimal reservoir parameters, and further obtain the water-drive reservoir production dynamic accurate description model.

[0064] Step 4: Use water flooding to develop a dynamic and accurate description model, and with the help of optimization algorithms, obtain the optimal injection and production parameter solution with the goal of optimal production:

[0065] The MCS algorithm is used to generate N sets of injection and production plans, and the production performance accurate description model of water drive reservoir is used to predict the production performance results to obtain the optimal injection and production plan.

[0066] Example 2: Segmented injection and production parameter optimization method based on planar flow unit delineation

[0067] like Figure 1 As shown, the method comprises the following steps:

[0068] Step 1. Collect geological and production data:

[0069] The reservoir physical property parameters include porosity, permeability, reservoir thickness, saturation, pressure, oil-water density, oil-water viscosity, phase permeability curve and reservoir rock compressibility coefficient.

[0070] The production data include oil and water well locations, vertical stratification and historical production data.

[0071] Step 2. Establish an initial description model of water-flooding reservoir production performance:

[0072] According to the target reservoir's production and injection conditions, well location relationships, geological development parameters, etc., the reservoir is delineated both vertically and horizontally to obtain a series of single-layer inter-well flow units. The two-phase flooding theory is used to calculate the production dynamics within the unit, thereby establishing an initial description model for water flooding development dynamics.

[0073] The double delineation steps of the target reservoir, both vertically and horizontally, also include two iterative links of vertical delineation and horizontal delineation: vertically, the working system of each well in each layer is delineated according to geological development parameters such as the production / injection volume of each well and each layer section and the seepage resistance of each layer; horizontally, different inter-well connected flow units are delineated on the plane according to the well position relationship and seepage resistance within the well group, and the production dynamics within the unit are calculated using the two-phase oil recovery theory.

[0074] like Figure 2 As shown in the figure, the plane flow unit delineation of a single layer segment is described by the flow unit area A, the well spacing d, and the delineation angle θ; in an injection and production well group, there are m oil wells with the water well as the center, and the delineation angle of the flow unit corresponding to the i-th oil well is:

[0075] θ i =2πα i

[0076] Where α i is the demarcation coefficient of the i-th oil well, and its solution formula is:

[0077]

[0078] Where p i is the pressure difference between the injection well and the i-th oil well, MPa; m is the total number of production wells; R i is the recovery degree in the direction of the i-th oil well.

[0079] The flow conditions within each plane demarcated unit can be calculated using the Buckley-Leverett oil-water two-phase displacement theory.

[0080] For multi-layer reservoirs, due to heterogeneity, the liquid volume of each layer also needs to be delineated. The liquid volume of each layer is delineated according to the following formula:

[0081]

[0082] Where Ri is the demarcation coefficient of the i-th layer:

[0083]

[0084] Qi is the liquid volume in the i-th layer, Ki is the permeability of the i-th layer, Ai is the cross-sectional area of ​​the i-th layer, Δp is the injection-production pressure difference, Gi is the starting pressure of the i-th layer, a is the unit conversion coefficient, d is the well spacing, λro is the oil phase mobility, and λrw is the water phase mobility.

[0085] Step 3. Perform automatic history matching on reservoir parameters, modify the description model, and establish an accurate description model of water-drive reservoir production dynamics:

[0086] Taking reservoir parameters as the adjustment object, a mathematical method for automatic history fitting problem was established. The porosity, permeability, effective thickness and control area in each subdivision unit were optimized to correct the mathematical model for water drive dynamic description, in order to minimize the gap between the calculation results of the reservoir production dynamic description model and the actual reservoir.

[0087] Objective function:

[0088]

[0089] Constraints:

[0090]

[0091] As attached Figure 3As shown in the figure, the CMA-ES adaptive intelligent optimization algorithm is used to minimize the objective function value, and the reservoir production dynamic description model is called to predict the production dynamic results. The model reservoir parameters are continuously corrected and updated until the convergence conditions are reached to obtain the optimal model reservoir parameters, and the reservoir production dynamic description model that can accurately describe the production dynamics of water-flooded reservoirs is obtained. In the iteration step k, CMA-ES first samples λ individuals to form a population according to the following formula

[0092]

[0093] Where N(…,…) is a random vector of multivariate normal distribution; m k is the mean vector; C k is the covariance matrix; σ is the step size factor.

[0094] During the optimization process, the covariance matrix is ​​updated according to the following formula:

[0095]

[0096] In the formula, It becomes an evolutionary path to characterize the optimization direction.

[0097] It also needs to be updated in each iteration step. The update formula is as follows:

[0098]

[0099] In the formula, c c is a constant between 0 and 1; Used to describe reorganization.

[0100] Step size factor σ k+1 The determination formula is as follows

[0101]

[0102] in, For the conjugate search path, determine as follows

[0103]

[0104] By adaptively scaling the covariance matrix through the step size factor, a good convergence speed can be achieved.

[0105] Step 4. Use the water flooding development dynamic accurate description model, with the help of optimization algorithm, aiming at optimal production, to obtain the optimal injection and production parameter scheme:

[0106] Based on the corrected water-flooding reservoir production dynamic description model, different injection-production plans are generated with the help of the MCS multi-level intelligent optimization algorithm. The reservoir production dynamic description model is called to predict the production dynamic results, and the injection-production production parameters are continuously corrected until the convergence conditions are reached to obtain the optimal injection-production plan.

[0107] Objective function: Maximum cumulative oil production from t0 to t1 in the future

[0108]

[0109] Optimization variables: production parameters Q of each well and each layer. m n Represents the production / injection volume of the mth layer of the nth well.

[0110]

[0111] Constraints:

[0112] Production and injection volume of each well and each layer q ij Cannot exceed the set upper limit q ij max and the lower limit q ij min :

[0113]

[0114] The bottom hole pressure p corresponding to the production and injection volume of each well and each layer ij Cannot exceed the set upper limit p ij max and the lower limit p ij min :

[0115]

[0116] The total liquid production of each oil well is constant:

[0117]

[0118] The total water injection volume of each well is constant:

[0119]

[0120] The MCS multi-level coordinated search algorithm is used to call the reservoir production dynamic description model to predict the production dynamic results under different injection and production parameter combinations. The injection and production parameters are continuously corrected and updated until the convergence conditions are reached to obtain the injection and production parameters corresponding to the optimal production. The solution process is shown in the attached figure. Figure 4 shown.

[0121] MCS specifically includes the following steps: 1) Initialization. In the initialization step, MCS splits the initial search space into a series of subspaces (called "boxes") through a pre-set initialization list. 2) Scanning. All boxes are sorted according to the size of the level, included in the scan list, and the boxes with the smallest objective function value at each level are marked. The marked boxes will enter the next splitting step for further splitting. The new boxes generated by the splitting are added to the scan list, and the scan list is updated. 3) Split. For the boxes that need to be split, determine whether the box has been split enough times. If it has been split enough times, use the sorting splitting method to further split along the direction where the box has been split the least times. Otherwise, use the expected splitting method, first estimate the expectation of the box to obtain more solutions when splitting in each direction, and then split along the direction with the largest expectation. 4) Shopping cart. After the scan is completed, the base points of all boxes that have reached the maximum level are placed in the shopping cart according to the size of the objective function value. For a point in the shopping basket, determine whether it is in a new local minimum area. If so, mark the point as the initial point for the next local search. Otherwise, if the point is in the same local minimum area as a point already in the shopping basket, the one with the better objective function value is retained. 5) Local search. The updated point in the shopping basket is used as the initial point to start the local search. The local search combines the ideas of coordinated search and sequential quadratic programming. First, a local quadratic form is constructed through a series of coordinated searches. Then, the search direction and search step size are determined according to the sequential quadratic programming. After the search is completed, the quadratic form is updated and iterated.

[0122] Example 3

[0123] Taking a block of Shengli Oilfield as an example, the method described in Example 2 is used to optimize injection and production parameters.

[0124] The relevant data of 9 wells in a block of Shengli Oilfield were collected, and the predicted oil production rate change curve of Well 1 was compared with the actual oil production rate change curve of the oil well. Figure 5 As shown in the figure, it can be seen that the currently predicted oil production rate curve is significantly different from the actual oil production rate curve, and automatic historical fitting of reservoir parameters is required.

[0125] The history matching was carried out by referring to the automatic history matching method of reservoir parameters described in Example 2 above. Through more than 9,000 times of multi-layer reservoir development dynamic prediction under different reservoir parameter combinations, the reservoir parameter combination that best matches the actual reservoir development dynamic curve was found. Under this reservoir parameter combination, the oil production rate change curve of oil well 1 is compared with the actual oil production rate change curve of the oil well, as shown in the figure below. Figure 6 As shown in the figure, it can be seen that the current predicted oil production rate curve is in good agreement with the actual oil production rate curve. At this time, it is believed that this set of reservoir parameter combinations can reflect the actual reservoir and can accurately predict future dynamics.

[0126] Based on the automatic historical matching of reservoir parameters, injection and production parameter optimization is carried out to optimize the injection and production parameter scheme with the goal of maximizing the cumulative oil production in the next five years. The constraint condition is set as the production well fluid volume range of 0-60m 3 / d, injection well fluid volume range 0-80m 3 / d. The total liquid volume is consistent with the original plan. Through more than 6,000 dynamic predictions of multi-layer reservoir development under different injection and production parameter combinations, the injection and production parameter plan with the largest cumulative oil production in the next five years is found. Figure 7 As shown in Figure 2, the optimal injection-production parameter scheme for this block is obtained.

[0127] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A segmented injection and production parameter optimization method based on planar flow unit delineation, characterized in that: It includes the following steps: Step 1. Collect geological and production data; Step 2. Establish an initial description model of water-flooding reservoir production performance; Step 3. Perform automatic history matching on reservoir parameters, modify the description model, and establish an accurate description model of water-drive reservoir production dynamics; Step 4. Use the dynamic accurate description model of water flooding development and the optimization algorithm to automatically optimize the injection and production parameters with the goal of optimal production; In step 2, the reservoir is delineated vertically and horizontally according to the target reservoir's production and injection conditions, well location relationships, and geological development parameters, and a series of single-layer inter-well flow units are obtained. The production dynamics within the unit are calculated using the two-phase flooding theory, thereby establishing an initial description model for water flooding development dynamics. The plane flow unit delineation of a single layer segment is described by the flow unit area A, the well spacing d, and the delineation angle θ. In an injection-production well group, there are m oil wells with the water well as the center. The delineation angle of the flow unit corresponding to the i-th oil well is: i i =2pa i ; Where α i is the demarcation coefficient of the i-th oil well, and its solution formula is: Where pi is the pressure difference between the injection well and the i-th oil well, in MPa; m is the total number of production wells; Ri is the degree of production in the direction of the i-th oil well; For multi-layer reservoirs, due to heterogeneity, the liquid volume of each layer also needs to be delineated. The liquid volume of each layer is delineated according to the following formula: Where Ri is the demarcation coefficient of the i-th layer: Qi is the liquid volume of the i-th layer, Ki is the permeability of the i-th layer, Ai is the cross-sectional area of ​​the i-th layer, Δp is the injection-production pressure difference, Gi is the starting pressure of the i-th layer, a is the unit conversion coefficient, d is the well spacing, λro is the oil phase mobility, and λrw is the water phase mobility; In step 3, the CMA-ES adaptive intelligent optimization algorithm is used to minimize the objective function value, and the reservoir production dynamic description model is called to predict the production dynamic results. The model reservoir parameters are continuously corrected and updated until the convergence conditions are met to obtain the optimal model reservoir parameters, thereby obtaining a reservoir production dynamic description model that can accurately describe the production dynamics of the water drive reservoir. Objective function: Constraints:

2. The method according to claim 1, characterized in that: The geological data include reservoir physical property parameters, which include porosity, permeability, reservoir thickness, saturation, pressure, oil-water density, oil-water viscosity, phase permeability curve and reservoir rock compression coefficient.

3. The method according to claim 1, characterized in that: The production data include oil and water well locations, vertical stratification and historical production data.

4. The method according to claim 1, characterized in that: The step of dual delineation of the target reservoir in vertical and horizontal direction also includes two iterative cycles of vertical delineation and horizontal delineation.

5. The method according to claim 4, characterized in that: Vertically, the working system of each well in each layer is determined according to the production / injection volume of each well in each layer and the geological development parameters of the seepage resistance of each layer; On the plane, according to the well position relationship and seepage resistance within the well group, different inter-well connected flow units are delineated on the plane for each layer, and the production dynamics within the unit are calculated using the two-phase oil recovery theory.

6. The method according to claim 1, characterized in that: In step 3, the CMA-ES algorithm is used to generate N sets of reservoir parameters, and the water-drive reservoir production dynamic initial description model is used for historical fitting. The model reservoir parameters are continuously corrected and updated to obtain the optimal reservoir parameters, and further obtain the water-drive reservoir production dynamic accurate description model.

7. The method according to claim 1, characterized in that: In step 4, the MCS algorithm is used to generate N sets of injection and production plans, and the production performance accurate description model of water drive reservoir production is used to predict the production performance results to obtain the optimal injection and production plan.

Images