A design optimization method and device for injection-production well pattern

Abstract

The application discloses a design optimization method and device of injection-production well pattern, relates to the technical field of well pattern deployment optimization in oil and gas field development engineering, and aims to automatically optimize well positions according to three-dimensional distribution of remaining oil, so as to improve recovery efficiency. The main technical scheme is as follows: determining the remaining oil saturation field of a specified area according to a three-dimensional geological model; identifying a remaining oil enrichment area based on a spatial form parameter and a connectivity index in the remaining oil saturation field; calculating a mining potential index and a mining risk grade of each remaining oil enrichment area according to remaining oil reserves, displacement efficiency and interwell pressure and flow of each remaining oil enrichment area; taking the maximization of recovery efficiency and the equalization of production capacity as optimization targets, taking the mining potential index as a recovery efficiency weight factor and the mining risk grade as a risk constraint condition, and generating a target injection-production well pattern by using a multi-objective optimization algorithm.

Description

Technical Field

[0001] This application relates to the field of well network deployment optimization technology in oil and gas field development engineering, and in particular to a method and apparatus for designing and optimizing injection and production well networks. Background Technology

[0002] As global oil and gas resource exploitation enters its middle and late stages, the proportion of high water-cut oilfields and oil reservoirs with complex geological conditions has increased significantly. In old oilfields in eastern China, the average recovery rate is insufficient, and the remaining oil reserves are high, making it urgent to improve the recovery rate through precision well placement technology.

[0003] Currently, regular well network layout patterns (such as square and rhomboid inverse nine-point methods) still dominate existing technologies. These patterns are based on uniform grid division and deploy injection-production well networks with fixed well and row spacing. The core principle is to achieve a balanced distribution of displacement pressure through regular well placement. However, this well network layout pattern does not consider the three-dimensional distribution characteristics of remaining oil, leading to the widespread existence of "dead oil zones" between wells. Furthermore, once the injection-production relationship is fixed, it is difficult to optimize it in real time based on production dynamics. In addition, in pursuit of uniform displacement, the production capacity of local high-yield areas may be sacrificed. Therefore, existing technologies lack a dynamic response mechanism and cannot automatically optimize well locations based on the three-dimensional distribution of remaining oil, resulting in a bottleneck in recovery rate improvement. Summary of the Invention

[0004] In view of the above problems, this application provides a method and apparatus for designing and optimizing injection-production well networks. The main purpose is to automatically optimize well locations based on the three-dimensional distribution of remaining oil in order to improve the recovery rate.

[0005] To solve the above-mentioned technical problems, this application proposes the following solution:

[0006] Firstly, this application provides a method for designing and optimizing an injection-production well network, the method comprising:

[0007] The remaining oil saturation field of a specified area is determined based on a three-dimensional geological model, which is constructed based on seismic data, well logging data, and production dynamic data of the specified area.

[0008] Residual oil enrichment areas are identified based on the spatial morphological parameters and connectivity index in the residual oil saturation field.

[0009] Obtain the displacement efficiency, inter-well pressure, and flow rate in the specified area during the injection and production process;

[0010] Based on the remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate of each remaining oil-rich zone, calculate the exploitation potential index and exploitation risk level of each remaining oil-rich zone;

[0011] Maximizing recovery rate and balancing production capacity are the optimization objectives. The extraction potential index is used as the recovery rate weighting factor, and the extraction risk level is used as the risk constraint. A multi-objective optimization algorithm is used to generate the target injection-production well network.

[0012] Secondly, this application provides a device for designing and optimizing injection-production well patterns, the device comprising:

[0013] The determination unit is used to determine the remaining oil saturation field of a specified area based on a three-dimensional geological model, wherein the three-dimensional geological model is constructed based on seismic data, well logging data and production dynamic data of the specified area;

[0014] The identification unit is used to identify the remaining oil enrichment area based on the spatial morphological parameters and connectivity index in the remaining oil saturation field obtained by the determining unit.

[0015] The acquisition unit is used to acquire the displacement efficiency, inter-well pressure, and flow rate of the specified area during the injection and production process.

[0016] The calculation unit is used to calculate the exploitation potential index and exploitation risk level of each of the remaining oil-rich areas based on the remaining oil reserves of each of the remaining oil-rich areas obtained by the identification unit, the displacement efficiency obtained by the acquisition unit, and the inter-well pressure and flow rate.

[0017] The optimization unit is used to maximize the recovery rate and balance the production capacity as optimization objectives. The exploitation potential index obtained by the calculation unit is used as the recovery rate weighting factor, and the exploitation risk level is used as the risk constraint condition. A multi-objective optimization algorithm is used to generate the target injection-production well network.

[0018] To achieve the above objectives, according to a third aspect of this application, a storage medium is provided, the storage medium including a stored program, wherein, when the program is executed, the device where the storage medium is located is controlled to perform the injection-production well pattern design optimization method of the first aspect.

[0019] To achieve the above objectives, according to a fourth aspect of this application, a processor is provided for running a program, wherein the program executes the injection-production well pattern design optimization method of the first aspect above.

[0020] Using the above technical solution, this application provides a method and apparatus for designing and optimizing an injection-production well network. First, the remaining oil saturation field of a designated area is determined based on a three-dimensional geological model. The three-dimensional geological model is constructed based on seismic data, well logging data, and production dynamic data of the designated area. Then, the remaining oil-rich areas are identified based on the spatial morphological parameters and connectivity index in the remaining oil saturation field. Next, the displacement efficiency, inter-well pressure, and flow rate of the designated area during the injection-production process are obtained. Then, based on the remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate of each remaining oil-rich area, the exploitation potential index and exploitation risk level of each remaining oil-rich area are calculated. Finally, the optimization objectives are to maximize the recovery rate and balance the production capacity. The exploitation potential index is used as the recovery rate weighting factor, and the exploitation risk level is used as the risk constraint. A multi-objective optimization algorithm is used to generate the target injection-production well network. The technical solution provided in this application determines the remaining oil saturation field of a specified area by combining a three-dimensional geological model, and identifies remaining oil-rich areas based on this, combined with spatial morphology and connectivity. This allows for more precise location of remaining oil, avoiding resource waste caused by inaccurate well placement. By calculating the exploitation potential index and exploitation risk level of each remaining oil-rich area, a multi-objective optimization algorithm is used to generate a target injection-production well network. This approach allows for real-time adjustment of the well network layout based on the latest production dynamic data, improving adaptability and flexibility. With maximizing recovery rate and balancing production capacity as optimization objectives, it ensures overall development benefits while also focusing on the full development of local high-efficiency areas, thereby maximizing overall production capacity. Compared to traditional regular well network layout patterns, this approach demonstrates significant advantages in improving recovery rate, optimizing resource allocation, and enhancing the ability to respond to dynamic changes, and is particularly suitable for the efficient tapping of remaining oil potential in heterogeneous reservoirs.

[0021] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0022] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0023] Figure 1 A flowchart illustrating a method for designing and optimizing an injection-production well network according to an embodiment of this application is shown.

[0024] Figure 2This application provides a flowchart of another injection-production well pattern design optimization method according to an embodiment of the present application.

[0025] Figure 3 This paper shows a block diagram of a design optimization device for injection-production well networks provided in an embodiment of this application;

[0026] Figure 4 A block diagram of another injection-production well network design optimization device provided in an embodiment of this application is shown. Detailed Implementation

[0027] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0028] Currently, regular well network layout patterns (such as square and rhomboid inverse nine-point methods) still dominate existing technologies. These patterns are based on uniform grid division and deploy injection-production well networks with fixed well and row spacing. The core principle is to achieve a balanced distribution of displacement pressure through regular well layout. For example, in the early stages of low-permeability reservoir development, a 300m×300m square inverse nine-point method well network is commonly used to establish a displacement system through uniform water injection. However, this well network layout pattern does not consider the three-dimensional distribution characteristics of remaining oil, resulting in the widespread existence of "dead oil zones" between wells. Furthermore, once the injection-production relationship is fixed, it is difficult to optimize it in real time based on production dynamics. In addition, in pursuit of uniform displacement, the production capacity of local high-yield areas may be sacrificed. Therefore, existing technologies lack a dynamic response mechanism and cannot automatically optimize well locations based on the three-dimensional distribution of remaining oil, leading to a bottleneck in recovery rate improvement.

[0029] Therefore, this application provides a method for designing and optimizing injection-production well patterns. This method can automatically optimize well locations based on the three-dimensional distribution of remaining oil to improve recovery rate. The specific execution steps are as follows: Figure 1 As shown, it includes:

[0030] 101. Determine the remaining oil saturation field of a specified area based on the three-dimensional geological model.

[0031] The three-dimensional geological model is constructed based on seismic data, well logging data, and production dynamic data of a specified area.

[0032] In this step, the designated area refers to the specific reservoir or geological unit for which residual oil saturation field modeling and subsequent well placement optimization are required. Its definition and scope must be determined based on actual oilfield development needs, geological conditions, and data coverage. After determining the designated area, seismic data, well logging data, and production dynamic data can be collected separately. The data is then normalized to eliminate data differences between different wells. Based on the spatial variability of seismic and well logging data, a three-dimensional geological model is generated using the sequential Gaussian simulation method. This model must include reservoir heterogeneity (such as planar, interlayer, and intralayer heterogeneity) and interlayer distribution. The model is dynamically validated using numerical simulation software, inputting historical production data to invert the residual oil distribution. The output is the residual oil saturation field of the designated area, for example, labeling high-oil-cut areas, low-oil-cut areas, and dead oil areas.

[0033] 102. Identify residual oil enrichment areas based on spatial morphological parameters and connectivity indices in the residual oil saturation field.

[0034] In this step, spatial morphological parameters include, but are not limited to: remaining oil volume: calculating the remaining oil reserves in each region through three-dimensional saturation field integration; shape factor: describing the geometric complexity of the remaining oil body (e.g., aspect ratio, compactness); directionality: analyzing the principal axis direction of the remaining oil body (related to sedimentary microfacies or tectonic trends); and distribution density: the concentration of remaining oil per unit volume (reflecting the degree of enrichment). Connectivity indices include, but are not limited to: number of connected paths: counting the number of connected channels within the remaining oil body and with other regions; and minimum spanning tree length: calculating the shortest connected path length within the remaining oil body using graph theory algorithms (e.g., Prim's algorithm). Specifically, a sample library can be constructed based on the remaining oil-rich areas already labeled in historical reservoir models. For example, an oil saturation threshold, such as 0.5, can be set, defining oil areas above this threshold as remaining oil-rich areas. Random perturbations are applied to the spatial morphological parameters to enhance model robustness. Deep learning (e.g., convolutional neural networks) or traditional machine learning (e.g., random forests) are used to train the oil area identification model. By inputting the parameters of the current remaining oil saturation field, the initial enrichment region and its confidence probability are output, and the final remaining oil enrichment region is determined by the confidence probability.

[0035] In addition, to ensure accuracy, the relationship between remaining oil volume and total remaining reserves, as well as between minimum spanning tree length and diagonal length, can be used for further determination. This embodiment does not limit this approach.

[0036] 103. Obtain the displacement efficiency, inter-well pressure, and flow rate in a specified area during the injection and production process.

[0037] In this step, displacement efficiency measures the proportion of crude oil displaced during water injection development relative to the crude oil content within the displacement area. It is a crucial indicator for evaluating oilfield development effectiveness, reflecting the efficiency with which injected water displaces crude oil in the reservoir. A higher value indicates more effective displacement of crude oil from the reservoir by the injected water. Specifically, it can be determined by combining experimental data, production dynamic data, and numerical simulations. For example, it can be calculated using the produced oil volume and injected water volume. Data is acquired through real-time flow meters and pressure sensors between injection and production wells, and the displacement efficiency distribution is inverted using numerical simulations (e.g., displacement efficiency >70% in high-permeability areas and <30% in low-permeability areas). For pressure data, bottomhole flowing pressure and saturation pressure between injection and production wells can be collected to calculate the production pressure differential. Anomalies in pressure gradients (e.g., high-pressure areas may be areas of remaining oil enrichment) are analyzed. For flow data, instantaneous flow rates in injection and production wells are recorded using flow meters to calculate the fluid exchange rate between wells. The fluid displacement capacity is determined by combining the mobility ratio.

[0038] 104. Based on the remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate of each remaining oil-rich zone, calculate the exploitation potential index and exploitation risk level of each remaining oil-rich zone.

[0039] In this step, after obtaining the remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate of each remaining oil-rich zone, the exploitation potential and risk of each zone can be analyzed. For exploitation potential, a weighted system can be pre-constructed, adjusting parameter weights according to reservoir type (e.g., low permeability, complex fault blocks). For example, remaining oil reserves (0.4), displacement efficiency (0.3), and flowability factor (0.3). The core purpose of the flowability factor is to assess the ease of fluid flow in the reservoir, thus providing key parameters for subsequent exploitation potential index calculation and risk level assessment. This flowability factor is determined using inter-well pressure and flow rate. By combining the remaining oil reserves, displacement efficiency, and flowability factor with their respective weighting coefficients, the exploitation risk index of each remaining oil-rich zone can be determined. To address extraction risks, a three-dimensional risk assessment matrix can be pre-constructed, including pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient. The pressure risk coefficient is used to assess the impact of formation pressure anomalies (such as overpressure or underpressure) on development, the geological risk coefficient is used to consider reservoir heterogeneity and interlayer distribution, and the disturbance risk coefficient is used to analyze the degree of fluid interference between injection and production wells (such as crossflow risk). Similarly, a corresponding weighting system can be set, which can be dynamically adjusted according to the oilfield geological conditions. By combining the pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient with their respective authority coefficients, the extraction risk level of each remaining oil-rich area can be determined.

[0040] 105. Taking the maximization of recovery rate and the equalization of production capacity as optimization objectives, the exploitation potential index as the weighting factor of recovery rate and the exploitation risk level as the risk constraint, a multi-objective optimization algorithm is used to generate the target injection-production well network.

[0041] In this step, a genetic algorithm is used to pre-encode the well location coordinates, well type, etc. For example, the well location coordinates (x, y, z) and well type (water injection well / oil production well) are encoded into binary strings. For example, 110101... indicates that a well is an oil production well located at coordinates (1000, 2000, -1500).

[0042] Fitness function: Maximizing recovery: f1 = Well-controlled reserves = ∑(SPI (Well-controlled reserves). Production equalization: f2 = min(maximum single-well production). Minimum single-well production). Overall fitness: F = αf1 + βf2, where α + β = 1, adjusted according to the development stage (e.g., emphasizing α in the early stage and β in the later stage). Specifically, multiple initial injection-production well networks (e.g., 100) are randomly generated, and the fitness score of each initial injection-production well network is calculated. A new generation of population is generated through crossover and mutation operations. Risk constraints are introduced: well networks with R > Rthreshold are eliminated. The iteration terminates when the fitness score does not change significantly for 5 consecutive generations. The target injection-production well network is imported into a 3D geological model (e.g., Petrel, Skyline platform), and the matching degree between the well locations and the geological model is calibrated, such as checking whether the well spacing meets the minimum spanning tree length requirement, and numerical simulation is run to verify the effect of the well network, such as predicting the recovery rate improvement and water cut changes. An interactive 3D scene is constructed using Skyline technology to achieve visualization display. Specific functions include, but are not limited to: dynamic rendering: real-time display of the remaining oil saturation field and well network layout; multi-view linkage: profile maps, contour maps and 3D models are updated synchronously; risk warning: high-risk areas are highlighted in red, etc.

[0043] Based on the above Figure 1As can be seen from the implementation method, the injection-production well network design optimization method provided in this application determines the remaining oil saturation field of a specified area by combining a three-dimensional geological model, and identifies the remaining oil-rich areas based on this, combined with spatial morphology and connectivity. This allows for more accurate location of the remaining oil, avoiding resource waste caused by inaccurate well placement. By calculating the exploitation potential index and exploitation risk level of each remaining oil-rich area, and using this as a basis, a multi-objective optimization algorithm is adopted to generate the target injection-production well network. This method allows for real-time adjustment of the well network layout based on the latest production dynamic data, improving adaptability and flexibility. With maximizing recovery rate and balancing production capacity as optimization objectives, it ensures overall development benefits while also focusing on the full development of local high-efficiency areas, thereby maximizing overall production capacity. Compared with the traditional regular well network layout mode, it shows significant advantages in improving recovery rate, optimizing resource allocation, and enhancing the ability to cope with dynamic changes, and is particularly suitable for the efficient tapping of remaining oil potential in heterogeneous reservoirs.

[0044] Furthermore, the preferred embodiments of this application are based on the above... Figure 1 Based on this, a detailed explanation of the process for designing and optimizing injection-production well patterns is provided, including the specific steps as follows: Figure 2 As shown, it includes:

[0045] 201. Acquire wide-azimuth 3D seismic data, high-resolution well logging data, and minute-level production dynamic data for a specified area.

[0046] In this step, high-precision seismic data for a designated area is acquired using wide-azimuth 3D seismic exploration techniques (such as OBN and 4D seismic). Noise and multiple interference are eliminated through techniques such as deconvolution, static correction, and velocity analysis. Key strata (such as the top and bottom interfaces of oil layers) and faults are identified using seismic attributes (such as amplitude, frequency, and phase). Seismic attribute volumes (such as coherence volume, curvature volume, and AVO attributes) are extracted to assist in reservoir prediction, resulting in wide-azimuth 3D seismic data. Reservoir parameters are acquired using high-resolution logging tools (such as array induction logging and nuclear magnetic resonance logging) in directional and horizontal wells. This logging data covers all wells in the designated area, including key parameters such as porosity, permeability, saturation, and lithology. Environmental corrections (such as wellbore influence and mud invasion correction) are applied to the logging curves. Lithology (such as sandstone and mudstone) is classified using cross-plot methods (such as sonic-density cross-plots) or neural network algorithms. Simultaneously, calculation models for parameters such as porosity and permeability are calibrated by comparing logging and core data to obtain high-resolution logging data. The SCADA system is used to monitor the production data of water injection wells and oil production wells in real time, including water injection wells: water injection volume and water injection pressure; and oil production wells: oil production volume, water cut, and bottom hole flow pressure. The data is sampled at a frequency of minutes (e.g., once every minute) to ensure the timeliness of dynamic data. Abnormal data is filtered through statistical analysis, and missing data is filled by linear interpolation or kriging interpolation. At the same time, key indicators such as injection-production ratio and pressure gradient are calculated to obtain minute-level dynamic production data.

[0047] 202. Based on wide-azimuth 3D seismic data, high-resolution well logging data, and minute-level production dynamic data, a 3D geological model of a specified area is constructed through sequential Gaussian simulation.

[0048] In this step, after obtaining wide-azimuth 3D seismic data, high-resolution logging data, and minute-level production dynamic data, a 3D geological model of the specified area can be constructed using sequential Gaussian simulation. Using reservoir numerical simulation software such as Eclipse or CMG, the following parameters are input: reservoir parameters: porosity, permeability (from the 3D geological model); fluid parameters: oil-water viscosity, relative permeability curves; production history: production dynamic data of injection and production wells (such as water injection volume and oil production volume).

[0049] 203. Based on numerical simulation, calculate the remaining oil saturation distribution of the three-dimensional geological model and generate the remaining oil saturation field of the specified area.

[0050] In this step, parameters such as permeability and relative permeability are adjusted to ensure that the simulation results match actual production data (such as water content and pressure) with an error ≤10%. During the simulation, the oil saturation of each grid is recorded, ultimately generating a residual oil saturation distribution. Minute-level dynamic production data (such as water injection volume and pressure) are input into the numerical simulator in real time to dynamically correct the residual oil saturation distribution.

[0051] Based on the detailed implementation of steps 201-203, by fusing wide-azimuth 3D seismic data, high-resolution well logging data, and minute-level production dynamic data, the realism and resolution of the 3D geological model are significantly improved, laying the foundation for accurate modeling of the remaining oil saturation field, realizing accurate characterization of the remaining oil distribution, effectively avoiding the risk of misjudgment caused by single or lagging data in traditional methods, and improving the accuracy and reliability of remaining oil identification.

[0052] 204. Obtain the spatial morphological parameters and connectivity index in the remaining oil saturation field.

[0053] Among them, the spatial morphology parameters include at least the remaining oil volume, shape factor, orientation, and distribution density, and the connectivity index includes at least the number of connected paths and the minimum spanning tree length.

[0054] In this step, the spatial morphology parameters include at least the remaining oil volume, shape factor, orientation, and distribution density. The connectivity index includes at least the number of connected paths and the minimum spanning tree length. The remaining oil volume characterizes the total volume of remaining oil within a certain region of the remaining oil saturation field. The shape factor characterizes the geometric complexity (e.g., irregularity) of the remaining oil enrichment area; a smaller value indicates a more spherical enrichment area (easier to extract), while a larger value indicates a more complex shape (higher extraction difficulty). The orientation characterizes the main direction of the remaining oil enrichment area (e.g., along the extension direction of the river sand body), reflecting the correlation between the distribution of remaining oil and geological structures (e.g., faults, rivers). The distribution density characterizes the distribution density of the remaining oil volume per unit area. The number of connected paths characterizes the number of connected paths from injection wells to production wells within the enrichment area; a higher number of paths indicates stronger connectivity (greater water injection development potential). The minimum spanning tree length characterizes the total length of the minimum spanning tree formed by all connected nodes (grids) within the enrichment area; a shorter minimum spanning tree length indicates more compact connectivity within the enrichment area (higher development efficiency).

[0055] 205. Input the spatial morphology parameters and connectivity index into the pre-trained oil area identification model to obtain the initial remaining oil enrichment area in the remaining oil saturation field and the confidence probability of each initial remaining oil enrichment area.

[0056] Among them, the oilfield identification model is trained based on samples obtained by marking the remaining oil-rich areas in the historical reservoir model and applying random perturbations to the spatial morphological parameters.

[0057] In this step, historical reservoir models are pre-collected, and labeled remaining oil-rich areas (positive samples) and non-rich areas (negative samples) are extracted from these models. Random perturbations (e.g., ROV ± 10%, SF ± 0.1) are applied to the spatial morphology parameters to generate diverse samples. Parameters are standardized (e.g., Min-Max normalization), and training / test sets are divided (8:2 ratio). Random forests, support vector machines (SVMs), or deep learning models (e.g., CNNs) can be used to train the oilfield identification model. The input data for this oilfield identification model is a remaining oil saturation field containing spatial morphology parameters and connectivity indices; the output data is the initial remaining oil-rich areas in the remaining oil saturation field and the confidence probability of each initial remaining oil-rich area.

[0058] 206. If the confidence probability is greater than or equal to the preset confidence probability threshold, and the remaining oil volume is greater than the preset volume or the minimum spanning tree length is less than the preset spanning tree length, then the initial remaining oil enrichment region is taken as the final remaining oil enrichment region.

[0059] The preset volume is the total remaining reserves at a first ratio, and the preset spanning tree length is the diagonal length of the initial remaining oil enrichment zone at a second ratio.

[0060] In this step, the extracted spatial morphological parameters and connectivity index are input into the pre-trained oil region identification model, which outputs initial enriched regions and their confidence probabilities. The confidence probability of each initial remaining oil enriched region is compared with a confidence probability threshold (0.7) to select the initial remaining oil enriched regions with higher confidence. Based on this, the remaining oil volume of each initial remaining oil enriched region is compared with a preset volume, and the minimum spanning tree length is compared with a preset spanning tree length. If the remaining oil volume is greater than the preset volume or the minimum spanning tree length is less than the preset spanning tree length, then the initial remaining oil enriched region is taken as the final remaining oil enriched region.

[0061] Regarding the determination of the preset volume and preset spanning tree length, specifically, the first ratio is set to k1, and the second ratio is set to k2. Preset volume = k1 × total remaining reserves, preset spanning tree length = k2 × diagonal length of the initial remaining oil-rich zone. The settings of parameters k1 and k2 need to be combined with reservoir engineering principles and actual development needs; for example, k1 = 0.5%, k2 = 10%.

[0062] Based on the detailed implementation of steps 204-206, by introducing spatial morphology parameters (such as residual oil volume, shape factor, orientation, and distribution density), the distribution pattern of residual oil in three-dimensional space can be characterized from geometric and statistical perspectives. This helps to distinguish between continuous oil areas with "real development value" and inefficient, scattered areas. By introducing connectivity indices (such as the number of connected paths and the minimum spanning tree length), the degree of connectivity within the residual oil body or between adjacent areas can be reflected, determining whether there are good displacement paths and availability. This allows for a more scientific and accurate characterization of the physical properties of residual oil-rich areas. To avoid misjudgments caused by reliance on subjective experience in traditional methods, a pre-trained oilfield identification model is introduced to achieve intelligent identification, that is, a shift from manual delineation to automated identification, improving the consistency and repeatability of identification results. Furthermore, a confidence probability mechanism is introduced to facilitate quality assessment and screening of identification results. At the same time, two key screening conditions are set: confidence probability ≥ pre-set confidence probability threshold, and remaining oil volume ≥ k1 × total remaining reserves or minimum spanning tree length ≤ k2 × diagonal length. This dual threshold control, which ensures that the selected area has practical development value from both the perspectives of resource quantity and connectivity, ensures the reliability of identification results.

[0063] 207. Obtain the displacement efficiency, inter-well pressure, and flow rate of a specified area during the injection and production process.

[0064] This step combines the description of step 103 in the above method, and the same content will not be repeated here.

[0065] 208. Based on the remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate of each remaining oil-rich zone, calculate the exploitation potential index and exploitation risk level of each remaining oil-rich zone.

[0066] This step combines the description of step 104 in the above method, and the same content will not be repeated here.

[0067] Furthermore, the specific process for calculating the exploitation potential index and exploitation risk level of each remaining oil-rich zone based on its remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate is as follows: Determine the corresponding remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate based on the spatial location of each remaining oil-rich zone; calculate the flow capacity factor of each remaining oil-rich zone based on the inter-well pressure and flow rate; determine the first weight coefficient corresponding to each of the remaining oil reserves, displacement efficiency, and flow capacity factor based on a preset regional adaptability weight system; and, based on the first weight coefficient, combine the remaining oil reserves of each remaining oil-rich zone, ... The exploitation potential index of each remaining oil-rich area is determined by displacement efficiency and flow capacity factor. A three-dimensional risk assessment matrix is ​​determined for each remaining oil-rich area based on its remaining oil reserves, displacement efficiency, and flow capacity factor. This matrix includes pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient. Second weight coefficients are determined for each of the pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient based on a pre-set risk assessment weight system. The exploitation risk level of each remaining oil-rich area is determined based on these second weight coefficients, combined with the pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient.

[0068] In this step, the remaining oil saturation field is calculated using three-dimensional geological models (such as Petrel and GMSS) and numerical simulations (such as Eclipse and CMG), and the remaining oil volume of each remaining oil enrichment zone is extracted. The specific expression is:

[0069]

[0070] in, This represents the remaining oil saturation of the i-th grid. This represents the porosity of the i-th grid. This represents the mesh volume.

[0071] Displacement efficiency characterizes the efficiency of water injection in displacing crude oil and reflects the sweep capacity of water injection development on the remaining oil. The sweep efficiency coefficient between water injection wells and production wells is calculated by fitting the historical data of water injection development.

[0072] Pressure data is obtained through real-time monitoring using bottom-hole flow pressure sensors in injection and production wells. Flow rate data is then obtained through a SCADA system to acquire water injection and oil production rates. The flow capacity factor is calculated by comparing the pressure gradient between the injection and production wells with the fluid flow rate (water injection or oil production). A smaller flow capacity factor indicates better reservoir permeability and lower fluid flow resistance; a larger flow capacity factor indicates poorer reservoir permeability and the need for higher pressure to drive the fluid.

[0073] A regional adaptability weighting system is constructed using either expert scoring (oilfield development experts assigning importance scores to each parameter) or the analytic hierarchy process (analytic hierarchy process) (constructing a judgment matrix and calculating the weights of each parameter). For example, remaining oil reserves (0.4), displacement efficiency (0.3), and flow capacity factor (0.3) are weighted. The first weight coefficients for each of the remaining oil reserves, remaining oil reserves, and flow capacity factor are determined using the pre-defined regional adaptability weighting system. A weighted sum based on these first weight coefficients yields the development potential index.

[0074] A risk dimension is predefined, including pressure risk, geological risk, and disturbance risk. A three-dimensional risk assessment matrix is ​​constructed based on these three dimensions. Specifically, this matrix includes pressure risk coefficients, geological risk coefficients, and disturbance risk coefficients. The pressure risk coefficient indicates that excessively high pressure gradients between injection and production wells may lead to reservoir fracturing or water channeling; the geological risk coefficient indicates the impact of reservoir heterogeneity (such as permeability variation coefficient) on development; and the disturbance risk coefficient indicates the degree of disturbance to the target enrichment area caused by adjacent well network development.

[0075] The specific expression for the pressure risk coefficient is as follows:

[0076]

[0077] in, This is the reservoir safety pressure threshold (e.g., 10 MPa).

[0078] The specific expression for the geological risk coefficient is as follows:

[0079]

[0080] in, The standard deviation of penetration rate. This represents the average penetration rate.

[0081] The specific expression for the interference risk coefficient is:

[0082]

[0083] in, This represents the minimum distance between the enriched area and the adjacent well. For a safe distance (e.g., 50 meters).

[0084] A risk assessment weighting system can be constructed for the pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient. For example, pressure risk reflects development safety (e.g., 0.4), geological risk reflects reservoir stability (e.g., 0.3), and disturbance risk reflects development complexity (e.g., 0.3). By determining the second weighting coefficients for each of the pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient through the preset risk assessment weighting system, and then performing a weighted sum based on these second weighting coefficients, the development risk level can be obtained.

[0085] Based on the detailed implementation of this step, the corresponding reserves, displacement efficiency, inter-well pressure, and flow rate are determined by the spatial location of each remaining oil-rich area. This achieves effective linkage between spatial information and dynamic development parameters, improving the accuracy of data matching and providing a precise input basis for subsequent potential and risk assessment. The introduction of a flow capacity factor quantifies fluid migration capacity, more realistically reflecting the availability and permeability of fluids within the reservoir, compensating for the shortcomings of relying solely on reserves or saturation, and enhancing the comprehensiveness of the assessment. This helps identify areas with "high reserves but difficult to utilize," optimizing development priorities. Through multi-factor weighted fusion, quantitative evaluation of exploitation potential is achieved, shifting from qualitative judgment to quantitative assessment. The first weight can be flexibly adjusted according to different reservoir types, enhancing the method's adaptability and providing a scientific basis for well network optimization, supporting the goal of maximizing recovery. This is achieved by constructing a three-factor system including pressure risk, geological risk, and disturbance risk. The multidimensional risk assessment matrix makes risk assessment more comprehensive, covering key issues such as formation pressure anomalies, reservoir heterogeneity, and injection-production interference. It supports the upgrade from single risk factors to systemic risk awareness, which is conducive to avoiding high-risk areas during development and ensuring project safety. By adopting a risk assessment weight system, the rationality of risk classification is improved, avoiding biases caused by subjective experience judgments. Different oilfields can set weights according to actual geological conditions, improving the flexibility of the method. It provides structured input for risk constraints in multi-objective optimization, namely, quantitative characterization and intelligent identification of the development potential and risks of remaining oil-rich areas, providing a solid decision-making basis for subsequent well network optimization.

[0086] 209. Taking the maximization of recovery rate and the equalization of production capacity as optimization objectives, the exploitation potential index as the recovery rate weighting factor and the exploitation risk level as the risk constraint, a multi-objective optimization algorithm is used to generate the target injection-production well network.

[0087] This step combines the description of step 105 in the above method, and the same content will not be repeated here.

[0088] Furthermore, taking maximizing recovery rate and balancing production capacity as optimization objectives, and using the extraction potential index as the recovery rate weighting factor and the extraction risk level as the risk constraint, the specific execution process of generating the target injection-production well network using a multi-objective optimization algorithm is as follows: Well location coordinates and well types are encoded as chromosomes, and multiple initial populations are randomly constructed using a genetic algorithm. Each initial population corresponds to an initial injection-production well network, and each initial injection-production well network represents a combination of well location coordinates and well types. A multi-objective fitness function for maximizing recovery rate and balancing production capacity is constructed based on the recovery rate weighting factor and risk constraint. The multi-objective fitness function is used to solve for each initial injection-production well network to obtain the fitness score of each initial injection-production well network. The initial injection-production well network with the highest fitness score is selected as the target injection-production well network.

[0089] In this step, the specific construction process is as follows, targeting the two optimization objectives of maximizing recovery rate and balancing production capacity:

[0090] Objective 1 is to maximize oil recovery, with the extraction potential index (SPI) as its weighting factor, reflecting the development value of enriched areas (high SPI areas are prioritized for development). Its objective function is:

[0091]

[0092] in, This represents the exploitation potential index of the i-th enrichment zone. This represents the remaining oil reserves in the i-th enriched region. This represents the total remaining oil reserves in all enriched areas.

[0093] Objective two is to achieve production capacity leveling. Its objective function is:

[0094]

[0095] in, This indicates the production capacity (oil output or water injection) of the j-th well. This represents the average productivity of all wells.

[0096] Constraints:

[0097] Mining risk level ( ) must be lower than a preset threshold (e.g. <0.5). The expression is:

[0098]

[0099] in, This indicates the risk threshold.

[0100] In this step, each chromosome represents a well pattern scheme, which includes well location coordinates and well types. Well location coordinates are represented by three-dimensional coordinates (x, y, z), while well types are represented by 0 for water injection wells and 1 for oil production wells. Well location coordinates are represented by real numbers, and well types are represented by binary bits.

[0101] Well location coordinates are randomly generated based on the reservoir extent, and well types are randomly assigned (the ratio of water injection wells to production wells is set empirically, such as 1:2). The initial population size is set (P=100).

[0102] Combining the objectives of maximizing recovery rate and balancing production capacity, and introducing a risk penalty:

[0103]

[0104] in, , Indicates the target weight coefficient. This indicates the risk penalty coefficient (e.g., 10).

[0105] Based on genetic operations and optimization iterations, such as crossover and mutation, the chromosome with the highest fitness score is selected from the final population as the target injection-production well network.

[0106] Based on the detailed implementation of this step, well location coordinates and well types are encoded as chromosomes to construct an initial population, enabling parametric modeling of the well network layout. This facilitates rapid iterative solution by the algorithm, and the chromosome structure is flexible, supporting mixed deployment of various well types (such as water injection wells, oil production wells, and horizontal wells). An objective function is established with maximizing recovery rate as the goal and balanced production capacity as the constraint. Recovery rate weighting factors and risk levels are introduced as control parameters, supporting adjustments to the optimization focus at different development stages. The introduction of weighting factors and risk levels makes the model more interpretable and flexible. Multi-objective collaborative optimization avoids the "unbalanced" development caused by a single objective. Multiple initial well networks are evaluated and screened using a fitness function, realizing the transformation from "experience-based well placement" to "data-driven + algorithm-driven" approaches. This improves the efficiency and consistency of well placement scheme formulation, reduces deviations caused by insufficient human experience or subjective judgment, and selects the optimal well network as the target injection-production well network based on the fitness score. This ensures that the final output well network is the optimal scheme in terms of comprehensive performance under the current geological conditions and risk constraints. By introducing genetic algorithms and multi-objective optimization mechanisms, the automatic generation and intelligent optimization of injection-production well networks have been achieved. This not only improves the efficiency and scientific nature of well placement, but also provides strong technical support for achieving "precise development, balanced development, and safe development" under complex reservoir conditions.

[0107] Furthermore, as a response to the above Figure 1-2The implementation of the method embodiment shown in this application provides a device for designing and optimizing injection-production well networks. This device is used to automatically optimize well locations based on the three-dimensional distribution of remaining oil to improve recovery rate. The embodiment of this device corresponds to the aforementioned method embodiment. For ease of reading, this embodiment will not repeat the details of the aforementioned method embodiment, but it should be clear that the device in this embodiment can correspondingly implement all the contents of the aforementioned method embodiment. Specifically, as shown... Figure 3 As shown, the device includes:

[0108] The determination unit 31 is used to determine the remaining oil saturation field of a specified area based on a three-dimensional geological model, wherein the three-dimensional geological model is constructed based on seismic data, well logging data and production dynamic data of the specified area;

[0109] The identification unit 32 is used to identify the remaining oil enrichment area based on the spatial morphological parameters and connectivity index in the remaining oil saturation field obtained by the determining unit 31.

[0110] The acquisition unit 33 is used to acquire the displacement efficiency, inter-well pressure and flow rate of the specified area during the injection and production process;

[0111] The calculation unit 34 is used to calculate the exploitation potential index and exploitation risk level of each of the remaining oil-rich areas based on the remaining oil reserves of each of the remaining oil-rich areas obtained by the identification unit 32, the displacement efficiency obtained by the acquisition unit 33, and the inter-well pressure and flow rate.

[0112] The optimization unit 35 is used to maximize the recovery rate and balance the production capacity as optimization objectives. It uses the mining potential index obtained by the calculation unit 34 as the recovery rate weighting factor and the mining risk level as the risk constraint condition, and uses a multi-objective optimization algorithm to generate the target injection-production well network.

[0113] Furthermore, such as Figure 4 As shown, the determining unit 31 includes:

[0114] The first acquisition module 311 is used to acquire wide-azimuth three-dimensional seismic data, high-resolution well logging data and minute-level production dynamic data of the specified area.

[0115] The first construction module 312 is used to construct the three-dimensional geological model of the specified area based on the wide-azimuth three-dimensional seismic data, the high-resolution well logging data and the minute-level production dynamic data obtained by the first acquisition module through sequential Gaussian simulation.

[0116] The first calculation module 313 is used to calculate the remaining oil saturation distribution of the three-dimensional geological model obtained by the first construction module 312 based on numerical simulation, and generate the remaining oil saturation field of the specified area.

[0117] Furthermore, such as Figure 4 As shown, the identification unit 32 includes:

[0118] The second acquisition module 321 is used to acquire the spatial morphological parameters and the connectivity index in the remaining oil saturation field. The spatial morphological parameters include at least the remaining oil volume, shape factor, directionality, and distribution density. The connectivity index includes at least the number of connected paths and the minimum spanning tree length.

[0119] The identification module 322 is used to input the spatial morphology parameters and the connectivity index obtained by the second acquisition module 321 into the pre-trained oilfield identification model to obtain the initial remaining oil enrichment area in the remaining oil saturation field and the confidence probability of each initial remaining oil enrichment area. The oilfield identification model is trained based on the samples obtained by marking the remaining oil enrichment area in the historical reservoir model and applying random perturbation to the spatial morphology parameters.

[0120] The first determining module 323 is used to determine the initial remaining oil enrichment region as the final remaining oil enrichment region if the confidence probability obtained by the identification module 322 is greater than or equal to a preset confidence probability threshold, and the remaining oil volume is greater than a preset volume or the minimum spanning tree length is less than a preset spanning tree length.

[0121] Furthermore, such as Figure 4 As shown, the preset volume is a first proportion of the total remaining reserves, and the preset spanning tree length is a second proportion of the diagonal length of the initial remaining oil enrichment zone.

[0122] Furthermore, such as Figure 4 As shown, the computing unit 34 includes:

[0123] The processing module 341 is used to determine the corresponding remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate according to the spatial location of each remaining oil enrichment zone, and to calculate the flow capacity factor of each remaining oil enrichment zone according to the inter-well pressure and flow rate.

[0124] The second determining module 342 is used to determine the first weight coefficients corresponding to the remaining oil reserves, the displacement efficiency and the flow capacity factor obtained by the processing module 341 based on a preset regional adaptive weighting system.

[0125] The third determining module 343 is used to determine the exploitation potential index of each of the remaining oil-rich areas based on the first weighting coefficient obtained by the second determining module 342, combined with the remaining oil reserves, the displacement efficiency and the flow capacity factor of each of the remaining oil-rich areas.

[0126] The fourth determining module 344 is used to determine a three-dimensional risk assessment matrix for each of the remaining oil-rich areas based on the remaining oil reserves, displacement efficiency, and flow capacity factor obtained by the processing module 341 for each of the remaining oil-rich areas. The three-dimensional risk assessment matrix includes a pressure risk coefficient, a geological risk coefficient, and a disturbance risk coefficient.

[0127] The fifth determining module 345 is used to determine the second weighting coefficients corresponding to the pressure risk coefficient, the geological risk coefficient, and the disturbance risk coefficient obtained by the fourth determining module 344 based on a preset risk assessment weighting system.

[0128] The sixth determining module 346 is used to determine the exploitation risk level of each of the remaining oil-rich areas based on the second weighting coefficient obtained by the fifth determining module 345, combined with the pressure risk coefficient, the geological risk coefficient and the disturbance risk coefficient.

[0129] Furthermore, such as Figure 4 As shown, the optimization unit 35 includes:

[0130] The second construction module 351 is used to encode well location coordinates and well type into chromosomes, and to randomly construct multiple initial populations using a genetic algorithm. Each initial population corresponds to an initial injection-production well network, and each initial injection-production well network represents a combination of well location coordinates and well type.

[0131] The third construction module 352 is used to construct a multi-objective fitness function for maximizing the recovery rate and equalizing the production capacity based on the recovery rate weighting factor and the risk constraint;

[0132] The second calculation module 353 is used to solve each of the initial injection-production well networks obtained by the second construction module 351 using the multi-objective fitness function obtained by the third construction module 352, and to obtain the fitness score of each initial injection-production well network.

[0133] The seventh determining module 354 is used to select the initial injection-production well network with the highest fitness score obtained by the second calculation module 353 as the target injection-production well network.

[0134] Furthermore, embodiments of this application also provide a storage medium for storing a computer program, wherein the computer program, when running, controls the device where the storage medium is located to execute the above-described... Figure 1-2 The method for designing and optimizing injection-production well networks described herein.

[0135] Furthermore, embodiments of this application also provide a processor for running a program, wherein the program executes the above-described... Figure 1-2 The method for designing and optimizing injection-production well networks described herein.

[0136] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0137] It is understood that the relevant features in the above methods and apparatus can be referenced interchangeably. Furthermore, the terms "first," "second," etc., in the above embodiments are used to distinguish between embodiments and do not represent the superiority or inferiority of any particular embodiment.

[0138] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0139] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, this application is not directed to any particular programming language. It should be understood that the content of this application described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of this application.

[0140] In addition, the memory may include non-permanent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0141] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0142] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0143] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0144] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0145] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0146] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, like read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0147] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0148] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0149] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0150] The above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for designing and optimizing an injection-production well pattern, characterized in that, The method includes: Determine the remaining oil saturation field of a specified area based on a three-dimensional geological model; Identifying remaining oil-rich regions based on spatial morphological parameters and connectivity indices in the remaining oil saturation field includes: acquiring the spatial morphological parameters and connectivity indices in the remaining oil saturation field, wherein the spatial morphological parameters include at least remaining oil volume, shape factor, directionality, and distribution density, and the connectivity indices include at least the number of connected paths and the minimum spanning tree length; inputting the spatial morphological parameters and connectivity indices into a pre-trained oilfield identification model to obtain initial remaining oil-rich regions in the remaining oil saturation field and the confidence probability of each initial remaining oil-rich region, wherein the oilfield identification model is trained based on samples obtained by labeling the remaining oil-rich regions in a historical reservoir model and applying random perturbations to the spatial morphological parameters; if the confidence probability is greater than or equal to a preset confidence probability threshold, and the remaining oil volume is greater than a preset volume or the minimum spanning tree length is less than a preset spanning tree length, then the initial remaining oil-rich region is taken as the final remaining oil-rich region; Obtain the displacement efficiency, inter-well pressure, and flow rate in the specified area during the injection and production process; Based on the remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate of each remaining oil-rich zone, the exploitation potential index and exploitation risk level of each remaining oil-rich zone are calculated, including: determining the corresponding remaining oil reserves, displacement efficiency, and inter-well pressure and flow rate based on the spatial location of each remaining oil-rich zone, and calculating the flow capacity factor of each remaining oil-rich zone based on the inter-well pressure and flow rate; determining the first weight coefficient corresponding to each of the remaining oil reserves, displacement efficiency, and flow capacity factor based on a preset regional adaptability weight system; and, based on the first weight coefficient, combining the remaining oil reserves and displacement efficiency of each remaining oil-rich zone... The exploitation potential index of each remaining oil-rich zone is determined by the flow capacity factor; a three-dimensional risk assessment matrix of each remaining oil-rich zone is determined based on the remaining oil reserves, displacement efficiency, and flow capacity factor of each remaining oil-rich zone, the three-dimensional risk assessment matrix including pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient; a second weight coefficient corresponding to each of the pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient is determined based on a preset risk assessment weight system; and the exploitation risk level of each remaining oil-rich zone is determined based on the second weight coefficient, combined with the pressure risk coefficient, geological risk coefficient, and disturbance risk coefficient. Maximizing recovery rate and balancing production capacity are the optimization objectives. The extraction potential index is used as the recovery rate weighting factor, and the extraction risk level is used as the risk constraint. A multi-objective optimization algorithm is used to generate the target injection-production well network.

2. The method according to claim 1, characterized in that, The remaining oil saturation field of a specified area is determined based on a three-dimensional geological model, including: Acquire wide-azimuth 3D seismic data, high-resolution well logging data, and minute-level production dynamic data for the specified area; Based on the wide-azimuth 3D seismic data, the high-resolution well logging data, and the minute-level production dynamic data, a 3D geological model of the specified area is constructed through sequential Gaussian simulation. The remaining oil saturation distribution of the three-dimensional geological model is calculated based on numerical simulation, and the remaining oil saturation field of the specified area is generated.

3. The method according to claim 1, characterized in that, The preset volume is a first proportion of the total remaining reserves, and the preset spanning tree length is a second proportion of the diagonal length of the initial remaining oil enrichment zone.

4. The method according to any one of claims 1-3, characterized in that, Maximizing recovery rate and balancing production capacity are the optimization objectives. The exploitation potential index is used as the recovery rate weighting factor, and the exploitation risk level is used as the risk constraint. A multi-objective optimization algorithm is employed to generate the target injection-production well network, including: Well location coordinates and well type are encoded as chromosomes, and multiple initial populations are randomly constructed using a genetic algorithm. Each initial population corresponds to an initial injection-production well network, and each initial injection-production well network represents a combination of well location coordinates and well type. Based on the recovery rate weighting factor and the risk constraints, a multi-objective fitness function is constructed to maximize the recovery rate and balance the production capacity. The fitness score of each initial injection-production well network is obtained by solving the multi-objective fitness function. The initial injection-production well network with the highest fitness score is selected as the target injection-production well network.

5. A device for designing and optimizing injection-production well patterns, used to perform the method as described in any one of claims 1 to 4, characterized in that, The device includes: The determination unit is used to determine the remaining oil saturation field of a specified area based on a three-dimensional geological model, wherein the three-dimensional geological model is constructed based on seismic data, well logging data and production dynamic data of the specified area; The identification unit is used to identify the remaining oil enrichment area based on the spatial morphological parameters and connectivity index in the remaining oil saturation field obtained by the determining unit. The acquisition unit is used to acquire the displacement efficiency, inter-well pressure, and flow rate of the specified area during the injection and production process. The calculation unit is used to calculate the exploitation potential index and exploitation risk level of each of the remaining oil-rich areas based on the remaining oil reserves of each of the remaining oil-rich areas obtained by the identification unit, the displacement efficiency obtained by the acquisition unit, and the inter-well pressure and flow rate. The optimization unit is used to maximize the recovery rate and balance the production capacity as optimization objectives. The exploitation potential index obtained by the calculation unit is used as the recovery rate weighting factor, and the exploitation risk level is used as the risk constraint condition. A multi-objective optimization algorithm is used to generate the target injection-production well network.

6. The apparatus according to claim 5, characterized in that, The determining unit includes: The first acquisition module is used to acquire wide-azimuth three-dimensional seismic data, high-resolution well logging data, and minute-level production dynamic data of the specified area. The first construction module is used to construct the three-dimensional geological model of the specified area based on the wide-azimuth three-dimensional seismic data, the high-resolution well logging data and the minute-level production dynamic data obtained by the first acquisition module through sequential Gaussian simulation. The first calculation module is used to calculate the remaining oil saturation distribution of the three-dimensional geological model obtained by the first construction module based on numerical simulation, and generate the remaining oil saturation field of the specified area.

7. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is executed, it controls the device containing the storage medium to perform the injection-production well pattern design optimization method as described in any one of claims 1 to 4.

8. A processor, characterized in that, The processor is used to run a program, wherein the program executes the injection-production well pattern design optimization method as described in any one of claims 1 to 4.

Images