Determination method, device and equipment of thickened CO2 flooding development scheme, medium and product

Abstract

The invention provides a method, device and equipment for determining a thickening CO2 flooding development scheme, a medium and a product, and relates to the technical field of oil and gas development. The method comprises the steps of obtaining a plurality of initial development schemes of a target oil reservoir, wherein the initial development schemes are obtained by permutation and combination of a plurality of preset values corresponding to a plurality of development characteristic parameters; inputting each initial development scheme into a pre-trained development dynamic prediction learning model to obtain development dynamic data corresponding to each initial development scheme; adopting a preset multi-target mathematical model to calculate a plurality of index values corresponding to each initial development scheme; determining a candidate development scheme set by adopting a multi-objective collaborative optimization algorithm according to the plurality of index values corresponding to each initial development scheme; and determining a thickening CO2 flooding target development scheme of the target oil reservoir based on the preset weight threshold values corresponding to the indexes. According to the method, through fusion of the preset multi-target mathematical model and the multi-target collaborative optimization algorithm, the optimization efficiency and precision of the thickening CO2 flooding development scheme are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of oil and gas development technology, and in particular to a method, apparatus, equipment, medium and product for determining a thickening CO2 flooding development scheme. Background Technology

[0002] As a core pillar of the global energy supply system and a fundamental chemical raw material for industrial production, the stable and efficient development of petroleum plays an irreplaceable role in ensuring energy security and economic development. Since most oil fields have gradually entered the high water-cut development stage, traditional waterflooding technology, limited by its own mechanisms, can only extract a small amount of geological reserves, leaving a large amount of residual oil underground. Therefore, there is a huge potential and urgent need to improve oil recovery. CO2, with its unique advantages such as low density, easy solubility in crude oil, and the ability to achieve geological sequestration, has proven to be a promising new oil displacement medium. However, CO2 flooding generally suffers from high development costs, significant process risks, and susceptibility to gas channeling. Field practice shows that introducing thickening compounds to increase the apparent viscosity of supercritical carbon dioxide can significantly improve the gas-oil mobility ratio and expand the reservoir swept volume.

[0003] In the development of thickeners, clean and environmentally friendly polymer thickeners have been proven through numerical simulation to improve oil recovery by more than 20% by switching to a thickening CO2-water alternating flooding mode in the later stages of waterflooding development in heterogeneous reservoirs. In the study of oil displacement mechanisms, numerical simulation and visualization experiments have clarified that the balance between viscous forces and gravity is key to improving gravity over-coverage during CO2 flooding. Particularly for thick, immiscible reservoir flooding scenarios, the study reveals the inherent law that increasing gas injection rate exacerbates gravity over-coverage and thus affects recovery. Regarding the application of intelligent optimization technology, a deep learning-assisted optimization framework for continuous CO2 flooding considering gas channeling has been proposed. A surrogate model is established based on a graph attention-long short-term memory neural network, and a multi-objective particle swarm optimization algorithm is used to achieve coordinated optimization of well location deployment and gas injection rate, effectively improving the computational efficiency of scheme design.

[0004] However, thickened CO2 flooding development schemes involve multiple objectives, and related technologies mostly focus on optimizing a single objective. They have failed to build a multi-objective synergistic optimization system that takes into account improving recovery rate, reducing thickener consumption, and improving CO2 geological storage efficiency, resulting in poor matching between development schemes and actual working conditions. Summary of the Invention

[0005] This application provides a method, apparatus, equipment, medium, and product for determining a thickening CO2 drive development scheme, which solves the technical problems of efficiency, accuracy, and engineering applicability of traditional development scheme design methods.

[0006] In a first aspect, this application provides a method for determining a thickening CO2 flooding development scheme, the method comprising:

[0007] Multiple initial development schemes for the target reservoir are obtained. Each initial development scheme is obtained by arranging and combining multiple preset values ​​corresponding to multiple development characteristic parameters. The development characteristic parameters include injection timing, thickener concentration, water injection rate, gas injection rate, half-cycle length, and bottom flow pressure of the oil well.

[0008] Each of the initial development schemes is input into a pre-trained development dynamics prediction learning model to obtain development dynamics data corresponding to each of the initial development schemes. The development dynamics data includes oil production rate and gas-oil ratio.

[0009] A preset multi-objective mathematical model is used to calculate multiple index values ​​corresponding to each of the initial development schemes based on the development dynamic data and preset values ​​of each development characteristic parameter. The multiple index values ​​include cumulative oil production, CO2 sequestration, and cumulative thickener usage.

[0010] A set of candidate development schemes is determined by using a multi-objective collaborative optimization algorithm based on the multiple indicator values ​​corresponding to each initial development scheme.

[0011] Based on the preset weight thresholds corresponding to each of the aforementioned indicators and the values ​​of each indicator corresponding to each candidate development scheme, the thickening CO2 flooding target development scheme corresponding to the target reservoir is determined.

[0012] In one possible design, after acquiring multiple initial development schemes for the target reservoir, the following is also included:

[0013] Acquire performance data of multiple candidate thickeners under multiple reservoir conditions and historical data of the target reservoir during the waterflooding stage;

[0014] The initial component numerical simulation model was calibrated based on the performance data of each thickener and the historical data of the target reservoir to obtain the target component numerical simulation model;

[0015] The Latin hypercube uniform sampling method is used to extract multiple first development schemes from multiple initial development schemes, and each first development scheme is input into the target component numerical simulation model to obtain the development dynamic data corresponding to each first development scheme.

[0016] The development dynamic data corresponding to each first development scheme and the preset values ​​corresponding to multiple development feature parameters are used as training sample sets to train the development dynamic prediction learning model, so as to obtain the pre-trained development dynamic prediction learning model.

[0017] In one possible design, the formula corresponding to the preset multi-objective mathematical model includes:

[0018]

[0019]

[0020]

[0021]

[0022] In one possible design, the multi-objective collaborative optimization algorithm is an improved CLMEA algorithm. The step of determining the candidate development scheme set using the multi-objective collaborative optimization algorithm based on multiple index values ​​corresponding to each initial development scheme includes:

[0023] The steps are executed iteratively in the order of classifier-assisted hierarchical learning pre-screening, hypervolume-based non-dominated search, and sparse target space local search until the maximum number of iterations is reached to obtain a set of candidate development solutions.

[0024] In one possible design, determining the thickening CO2 flooding target development scheme for the target reservoir based on the preset weight thresholds corresponding to each of the aforementioned indicators and the indicator values ​​corresponding to each candidate development scheme includes:

[0025] Obtain the preset weight coefficients corresponding to each indicator;

[0026] Based on the values ​​of each indicator and the preset weight coefficients of each indicator for each candidate development scheme, the comprehensive evaluation value of each candidate development scheme is calculated.

[0027] The candidate solution with the highest comprehensive evaluation value and whose values ​​for each indicator are all within the corresponding preset threshold range is selected as the target development solution.

[0028] In one possible design, after determining the thickening CO2 flooding target development scheme corresponding to the target reservoir, the method further includes:

[0029] Obtain multiple production index values ​​during the development process of the target reservoir;

[0030] In response to any production indicator value exceeding the corresponding preset threshold range, a pre-trained development dynamic prediction learning model and a multi-objective collaborative optimization algorithm are used to adjust the values ​​corresponding to each development feature data.

[0031] Secondly, this application provides an apparatus for determining a thickening CO2 flooding development scheme, the apparatus comprising:

[0032] The acquisition module is used to acquire multiple initial development schemes for the target reservoir. The initial development schemes are obtained by arranging and combining multiple preset values ​​corresponding to multiple development characteristic parameters. The development characteristic parameters include the timing of the transfer injection, CO2 thickener concentration, water injection rate, gas injection rate, half-cycle length, and bottom flow pressure of the oil well.

[0033] The input module is used to input each of the initial development schemes into a pre-trained development dynamic prediction learning model to obtain development dynamic data corresponding to each of the initial development schemes. The development dynamic data includes oil production rate and gas-oil ratio.

[0034] The calculation module is used to calculate multiple index values ​​corresponding to each of the initial development schemes based on the development dynamic data corresponding to each of the initial development schemes and the preset values ​​corresponding to each development characteristic parameter using a preset multi-objective mathematical model. The multiple indexes include cumulative oil production, CO2 storage amount and cumulative thickener usage.

[0035] The determination module is used to determine the set of candidate development schemes based on multiple indicator values ​​corresponding to each initial development scheme using a multi-objective collaborative optimization algorithm.

[0036] The determining module is further configured to determine the thickening CO2 flooding target development scheme corresponding to the target reservoir based on the preset weight thresholds corresponding to each of the indicators and the indicator values ​​corresponding to each candidate development scheme.

[0037] Thirdly, this application provides a device for determining a thickening CO2 drive development scheme, the device comprising: a processor, and a memory communicatively connected to the processor;

[0038] The memory stores computer-executed instructions;

[0039] The processor executes computer execution instructions stored in the memory to implement the method as described in any of the first aspects.

[0040] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in any of the first aspects above.

[0041] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method described in any of the first aspects above.

[0042] The thickening CO2 drive development scheme determination method, apparatus, equipment, medium, and product provided in this application generate multiple initial development schemes by arranging and combining preset values ​​of development characteristic parameters, covering various possible combinations of development strategies. This provides a rich sample base for subsequent screening and optimization, increasing the likelihood of finding the optimal development scheme. The pre-trained development dynamic prediction learning model can quickly predict a large number of initial development schemes without performing complex numerical simulation calculations, rapidly acquiring development dynamic data, saving computation time and cost, and accelerating the entire development scheme determination process. The preset multi-objective mathematical model can comprehensively evaluate each initial development scheme from the perspective of multiple key indicators. By calculating specific indicator values, the performance of different development schemes is quantified, facilitating direct comparison and analysis of each scheme. This provides clear optimization objectives and directions for subsequent optimization algorithms, helping to select better development schemes. The multi-objective collaborative optimization algorithm can balance and optimize among multiple conflicting objectives, finding a set of non-dominated solutions, i.e., a candidate development scheme set. These schemes have good comprehensive performance on different indicators, avoiding the local optimum problem that may be caused by single-objective optimization, and improving the overall quality of the development schemes. By setting threshold weights for indicators, the importance of different indicators can be flexibly adjusted according to the actual situation of the target reservoir and the development goals, ensuring that the final target development plan meets the actual development needs. Attached Figure Description

[0043] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0044] Figure 1 An application scenario diagram illustrating the method for determining a thickening CO2 drive development scheme according to an embodiment of this application;

[0045] Figure 2 A flowchart illustrating a method for determining a thickening CO2 drive development scheme according to an embodiment of this application;

[0046] Figure 3 A flowchart illustrating a method for determining a thickening CO2 drive development scheme according to another embodiment of this application;

[0047] Figure 4 A schematic diagram of the structure of a device for determining a thickening CO2 drive development scheme according to an embodiment of this application;

[0048] Figure 5 This is a schematic diagram of the structure of a device for determining a thickening CO2 drive development scheme according to an embodiment of this application.

[0049] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0050] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0051] It should be noted that certain software, components, and models may be mentioned in the embodiments of this application. These should be considered as exemplary and are intended only to illustrate the feasibility of implementing the technical solution of this application. However, they do not mean that the applicant has used or necessarily used the solution.

[0052] To clearly understand the technical solution of this application, the solutions of the prior art will be described in detail first.

[0053] Petroleum occupies a core position in global energy supply and is an indispensable chemical raw material for industrial production. Its stable and efficient development is of great significance to energy security and economic development. Currently, most oilfields have entered the high water-cut development stage. Traditional waterflooding technology, limited by its mechanism, can only extract a small amount of geological reserves from the reservoir, leaving a large amount of residual oil underground. Therefore, there is huge potential and an extremely urgent need to improve oil recovery. CO2, due to its low density, easy solubility in crude oil, and geological storage capabilities, has become a highly promising new oil displacement medium. However, CO2 flooding suffers from high development costs, significant process risks, and susceptibility to gas channeling. Field practice has shown that introducing thickeners to increase the apparent viscosity of supercritical carbon dioxide can improve the gas-oil mobility ratio and expand the reservoir swept volume. In the development of thickeners, clean and environmentally friendly polymer thickeners, based on numerical simulation predictions, can further increase oil recovery by more than 20% by switching to thickening CO2-water alternating flooding in the later stages of waterflooding in heterogeneous reservoirs. The study of oil displacement mechanisms, through numerical simulation and visualization experiments, clarified that the balance between viscosity and gravity is key to improving gravity overlap. It also revealed the relationship between gas injection rate, gravity overlap degree, and recovery rate in thick, immiscible oil reservoirs. Regarding intelligent optimization technology, a deep learning-assisted optimization framework for continuous CO2 flooding considering gas channeling was proposed. A surrogate model was established to collaboratively optimize well location and gas injection rate, improving computational efficiency. However, thickening CO2 flooding development schemes exhibit significant multi-objective characteristics, and related technologies mostly focus on single-objective optimization, failing to construct a multi-objective collaborative optimization system, resulting in poor matching between development schemes and actual operating conditions.

[0054] Therefore, when facing technical problems in existing technologies, considering that multiple key factors in the development process will affect the development effect, multiple preset values ​​are set for multiple development characteristic parameters, and multiple initial development schemes for the target reservoir are generated by permutation and combination. This covers various possible parameter combinations, avoids missing potential effective development schemes, and provides a rich sample basis for subsequent screening and optimization. In order to quickly evaluate the potential effect of each initial development scheme, a pre-trained development dynamic prediction learning model can be used to simulate the development dynamic process of the reservoir under different combinations of development characteristic parameters. Cumulative oil production directly reflects the ability of the development scheme to improve oil recovery, CO2 sequestration reflects the contribution of the development scheme to environmental protection, and the cumulative amount of thickener used is closely related to development costs. Therefore, a preset multi-objective mathematical model is adopted. Based on the development dynamic data corresponding to each initial development scheme and the preset values ​​corresponding to each development characteristic parameter, multiple index values ​​are calculated to comprehensively and objectively reflect the advantages and disadvantages of the development schemes and avoid the one-sidedness of single index evaluation. Since there are multiple conflicting objective indicators, in order to find a balance among these objectives and achieve multi-objective synergistic optimization, the multiple index values ​​corresponding to each initial development scheme are input into the multi-objective synergistic optimization algorithm. This ensures that the selected candidate development schemes exhibit good overall performance across different indicators. In actual development, different development objectives may have varying degrees of importance. Therefore, based on preset weight thresholds for each indicator and the corresponding indicator values ​​for each candidate development scheme, a weighted calculation method is used to determine the thickening CO2 flooding target development scheme for the target reservoir. The weights of different indicators can be flexibly adjusted according to actual development needs, thereby formulating a development scheme that conforms to the actual situation.

[0055] Figure 1 This is an application scenario diagram of the method for determining a thickening CO2 drive development scheme provided in an embodiment of this application, such as... Figure 1 As shown in the diagram, the application scenario of the method for determining the thickening CO2 drive development scheme provided in this embodiment includes: a server 101 and a terminal device 102. The device for determining the thickening CO2 drive development scheme is integrated into the server 101.

[0056] Specifically, when a development plan needs to be determined, the user can initiate a determination request on the terminal device 102. This request may include the identifier of the target reservoir for which the development plan needs to be determined, as well as the preset value ranges corresponding to multiple development characteristic parameters of the target reservoir. After receiving the determination request from the terminal device 102, the server 101 selects values ​​from the preset value ranges corresponding to the multiple development characteristic parameters of the target reservoir and arranges them in a permutation and combination to obtain multiple initial development plans. Each initial development plan includes a set of values ​​for the development characteristic parameters. Then, the server 101 inputs each initial development plan into a pre-trained development dynamic prediction learning model to obtain development dynamic data corresponding to each initial development plan. Then, a preset multi-objective mathematical model is used to calculate multiple index values ​​corresponding to each initial development plan based on the development dynamic data and the preset values ​​corresponding to each development characteristic parameter. Based on the multiple index values ​​corresponding to each initial development plan, a multi-objective collaborative optimization algorithm is used to determine a set of candidate development plans. Finally, based on the preset weight thresholds corresponding to each index and the index values ​​corresponding to each candidate development plan, the thickening CO2 flooding target development plan corresponding to the target reservoir is determined. Server 101 sends the determined target development plan to terminal device 102 so that users can mine according to the target development plan.

[0057] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0058] Figure 2 A flowchart illustrating a method for determining a thickening CO2 drive development scheme according to an embodiment of this application is shown below. Figure 2 As shown, the execution subject of this embodiment is a device for determining a thickening CO2 drive development scheme. This device can be implemented through a computer program, or through a medium storing the relevant computer program, such as a USB flash drive and / or optical disc, or it can be integrated into a device for determining a thickening CO2 drive development scheme, such as a server. The method for determining a thickening CO2 drive development scheme provided in this embodiment includes the following steps:

[0059] Step 201: Obtain multiple initial development schemes for the target reservoir. The initial development schemes are obtained by arranging and combining multiple preset values ​​corresponding to multiple development characteristic parameters. The development characteristic parameters include the timing of transfer injection, thickener concentration, water injection rate, gas injection rate, half-cycle length, and bottom flow pressure of the oil well.

[0060] Among them, the target reservoir refers to a specific oil and gas reservoir for which the development plan needs to be optimized.

[0061] The initial development scheme refers to a set of unoptimized candidate development schemes generated based on the fundamental parameters of the target reservoir, through the permutation and combination of development characteristic parameters and their corresponding preset values. Each initial development scheme corresponds to a complete set of development characteristic parameters.

[0062] Among them, the timing of the switch to injection refers to the specific time point when the reservoir is switched from conventional water injection to thickening CO2 flooding after the initial stage of reservoir development.

[0063] The preset values ​​refer to the reasonable values ​​that are pre-set for each development characteristic parameter, taking into account the geological conditions of the target reservoir, engineering experience, and numerical simulation results. For example, the thickener concentration may be 0.1%, 0.3%, 0.5%, or 0.7%.

[0064] Understandably, one can pre-determine the preset value range of each development characteristic parameter by referring to similar reservoir CO2 flooding development cases, numerical simulation sensitivity analysis, and combining engineering constraints, and then select values ​​from them, such as selecting 10 values ​​evenly within the value range.

[0065] Optionally, the preset value ranges of each development characteristic parameter corresponding to the target reservoir can be sent to the thickening CO2 flooding development scheme determination device in a determined manner, or the preset value ranges of each development characteristic parameter corresponding to each reservoir can be stored in advance according to the reservoir identifier. This embodiment does not limit this.

[0066] Specifically, when determining the development plan, values ​​are assigned to various development characteristic parameters. For example, if the preset range for thickener concentration is 0.1% to 1%, the preset values ​​could be 0.1%, 0.2%, 0.3%, 0.4%, etc. After determining the preset values ​​for each development characteristic parameter, they are arranged and combined, such as using the full factorial combination method to determine the initial development plan.

[0067] For example, initial development scheme 1: timing of injection, water content 20%; thickener concentration, 0.1%; injection speed, 100m. 3 / d; Injection rate, 80 m 3 / d; half-cycle length, 3 months; wellbore bottom flow pressure, 5MPa.

[0068] Optionally, after generating multiple initial development schemes, they can be validated, including eliminating schemes that do not meet engineering constraints, such as a scheme with a bottom hole pressure of 11 MPa, which exceeds the casing compressive strength of 10 MPa and is therefore directly eliminated; and merging duplicate schemes, such as those with a water injection rate of 250 m... 3 The gas injection rate at / d must be 200 m / s. 3 If / d, then adjust the combinational logic.

[0069] Step 202: Input each initial development scheme into the pre-trained development dynamic prediction learning model to obtain the development dynamic data corresponding to each initial development scheme. The development dynamic data includes oil production rate and gas-oil ratio.

[0070] Among them, the pre-trained development dynamic prediction learning model refers to the prediction model trained by machine learning algorithms based on historical reservoir development data.

[0071] Specifically, after inputting each initial development plan into the pre-trained development dynamic prediction learning model, the model first preprocesses the data included in each initial development plan, including standardization and normalization. Then, according to the learned mapping relationship between development feature parameters and development dynamic data, each initial development plan is processed in batches or in stages, and the predicted development dynamic data is bound to the corresponding initial development plan before being output.

[0072] Among them, dynamic data can be output in units of wells.

[0073] Step 203: Using a preset multi-objective mathematical model, based on the development dynamic data corresponding to each initial development scheme and the preset values ​​corresponding to each development characteristic parameter, calculate the values ​​of multiple indicators corresponding to each initial development scheme. The multiple indicators include cumulative oil production, CO2 sequestration, and cumulative thickener usage.

[0074] The pre-defined multi-objective mathematical model refers to a pre-constructed set of objective functions that simultaneously quantifies multiple core development objectives. The model takes dynamic development data and development characteristic parameters as input, and through explicit calculation formulas, outputs quantitative indicators that can be directly used for scheme comparison.

[0075] Cumulative oil production refers to the total mass or volume of crude oil produced by oil wells within a forecast period, such as 5 years.

[0076] Among them, CO2 sequestration refers to the total mass or volume of CO2 captured and long-term stored in the reservoir during the prediction period.

[0077] The cumulative amount of thickener used refers to the total mass of thickener injected into the reservoir during the predicted period.

[0078] Specifically, the preset multi-objective mathematical model will calculate the values ​​of each indicator corresponding to each initial development scheme based on the development dynamic data corresponding to each initial development scheme and the preset values ​​corresponding to each development feature parameter, using a preset objective function calculation formula.

[0079] Step 204: Determine the set of candidate development schemes using a multi-objective collaborative optimization algorithm based on the multiple index values ​​corresponding to each initial development scheme.

[0080] Among them, the multi-objective collaborative optimization algorithm refers to an intelligent optimization algorithm designed for multiple mutually constraining optimization objectives. It is used to select the non-dominated Pareto front solution set from all initial development schemes, that is, the case where no scheme has all indicators that are better than another scheme, and form a set of candidate development schemes.

[0081] Understandably, the multi-objective collaborative optimization algorithm predefines the optimization priorities of each indicator, such as maximizing cumulative oil production; maximizing CO2 sequestration; and minimizing cumulative thickener usage.

[0082] Specifically, the multi-objective collaborative optimization algorithm iterates based on the index values ​​of each initial development scheme according to the preset execution logic until the preset number of iterations is reached and the constraints are met. After the algorithm converges, the schemes with a dominance count of 0 in all frontier layers are the final non-dominated solution set, which is the candidate development scheme set.

[0083] Step 205: Based on the preset weight thresholds corresponding to each indicator and the values ​​of each indicator corresponding to each candidate development scheme, determine the thickening CO2 flooding target development scheme corresponding to the target reservoir.

[0084] Among them, the preset weight threshold refers to the combination of the importance coefficient of each indicator and the qualified threshold that are set in advance based on the core requirements of the project.

[0085] Specifically, after determining the set of candidate development schemes, the preset weight thresholds corresponding to each indicator are obtained, and each candidate development scheme is calculated one by one to obtain the corresponding comprehensive score. The scheme with the highest comprehensive score is selected as the thickening CO2 flooding target development scheme corresponding to the target reservoir in descending order.

[0086] The method for determining thickening CO2-driven development schemes provided in this application generates multiple initial development schemes by arranging and combining preset values ​​of development characteristic parameters, covering various possible combinations of development strategies. This provides a rich sample base for subsequent screening and optimization, increasing the likelihood of finding the optimal development scheme. The pre-trained development dynamic prediction learning model can quickly predict a large number of initial development schemes without requiring complex numerical simulation calculations, rapidly acquiring development dynamic data, saving computation time and cost, and accelerating the entire development scheme determination process. The preset multi-objective mathematical model can comprehensively evaluate each initial development scheme from the perspective of multiple key indicators. By calculating specific indicator values, the performance of different development schemes is quantified, facilitating direct comparison and analysis of each scheme. This provides clear optimization goals and directions for subsequent optimization algorithms, helping to select better development schemes. The multi-objective collaborative optimization algorithm can balance and optimize among multiple conflicting objectives, finding a set of non-dominated solutions, i.e., a candidate development scheme set. These schemes exhibit good comprehensive performance on different indicators, avoiding the local optimum problem that may be caused by single-objective optimization, and improving the overall quality of the development schemes. By setting threshold weights for indicators, the importance of different indicators can be flexibly adjusted according to the actual situation of the target reservoir and the development goals, ensuring that the final target development plan meets the actual development needs.

[0087] As an optional implementation, after obtaining multiple initial development schemes for the target reservoir based on the above embodiments, the method further includes:

[0088] Acquire performance data of multiple candidate thickeners under multiple reservoir conditions and historical data of the target reservoir during the waterflooding stage;

[0089] The initial component numerical simulation model was calibrated based on the performance data of each thickener and the historical data of the target reservoir to obtain the target component numerical simulation model;

[0090] The Latin hypercube uniform sampling method was used to extract multiple first development schemes from multiple initial development schemes, and each first development scheme was input into the target component numerical simulation model to obtain the development dynamic data corresponding to each first development scheme.

[0091] The development dynamic data corresponding to each first development scheme and the preset values ​​corresponding to multiple development feature parameters are used as training sample sets to train the development dynamic prediction learning model, so as to obtain the pre-trained development dynamic prediction learning model.

[0092] Among them, the thickeners to be selected are those with strong solubility, good dispersion stability and excellent thickening performance, which are selected by conducting experiments such as Fourier transform infrared spectroscopy, supercritical CO2 visualization solubility test and viscosity-enhancing effect test in advance for the conditions of the target reservoir.

[0093] Among them, performance data refers to the pre-determined relationship between the viscosity and temperature of each candidate thickener under different reservoir conditions, such as different reservoir temperatures, pressures, and salinity conditions, as well as the relationship between the viscosity and concentration of the thickener.

[0094] Among them, the historical data of the target reservoir during the water drive stage refers to the data generated by the target reservoir during the water drive stage, such as pressure, production, water cut, etc.

[0095] The target component numerical simulation model refers to a high-precision numerical simulation model obtained by adjusting the parameters of the initial component numerical simulation model using thickener performance data and water drive historical data.

[0096] The first development scheme refers to a subset of schemes selected from the initial development schemes through Latin hypercube uniform sampling, which serves as input samples for training and developing the dynamic prediction learning model.

[0097] Specifically, after obtaining performance data of multiple candidate thickeners under various reservoir conditions and historical data of the target reservoir during the water-drive stage, the model predictions and historical actual values ​​were compared by adjusting sensitive parameters such as the oil-water relative permeability endpoint and bound water saturation. A trial-and-error method or automatic history fitting tool was used to gradually adjust the calibration parameters until the error met the requirements, thus completing the historical fitting of the target reservoir's water-driven state data. Then, fluid phase experiments such as reservoir fluid pseudo-component splitting, single flash evaporation, constant composition expansion, multi-stage contact differential release, and long core displacement were fitted to describe the dissolution, extraction, and diffusion behavior of supercritical CO2 under the target reservoir conditions. Finally, based on the performance data of each thickener, the calibration parameters were adjusted by fitting the experimental data to describe the mechanism by which the thickeners improve the gas-oil mobility ratio, resulting in a numerical simulation model of the target component.

[0098] Furthermore, using the Latin hypercube uniform sampling method, the preset value range of each parameter is divided into n equally probable intervals. One value is randomly selected from each interval for each parameter, and the values ​​of different parameters are combined to form the first development scheme. The development characteristic parameter values ​​of each first development scheme are input into the target component numerical simulation model according to the model requirements. Numerical simulations are run sequentially for all first development schemes. The development dynamic time series data of each scheme are extracted from the simulation results. A correspondence table of first development scheme number, development characteristic parameters, and development dynamic data is formed, which serves as the core data of the training sample set.

[0099] Furthermore, the six development feature parameters of the first development scheme are used as input features, and the dynamic development data is used as output features to train a dynamic prediction learning model, such as a recurrent neural network LSTM or GRU. The sample set can be randomly split into 80%, 10%, and 10% portions for training, validation, and testing, respectively.

[0100] For example, the training set is input into the model, and the error between the predicted value and the label value is minimized through the backpropagation algorithm, with the mean squared error (MSE) selected as the loss function. Every 10 training iterations, the model accuracy is evaluated using the validation set. If the validation set error no longer decreases, methods such as early stopping and dropout are used to suppress overfitting. Hyperparameters are adjusted, such as increasing the number of neurons in the hidden layer and decreasing the learning rate, until the validation set error reaches the preset requirement, such as an oil production rate prediction error ≤10%. The model accuracy is finally evaluated using the test set. If the test set error meets the requirement, such as an oil-gas ratio prediction error ≤15%, then the training is complete. If it does not meet the requirement, the number of samples is increased, or the model algorithm is adjusted.

[0101] The method for determining the CO2 flooding development scheme provided in this application addresses the significant performance differences of different thickeners under varying reservoir conditions. By combining historical waterflooding data from actual reservoirs, model parameters can be adjusted more precisely, making the numerical simulation model of the target component more closely reflect the actual conditions of the target reservoir. The Latin hypercubic uniform sampling method is used to extract the first development scheme from multiple initial development schemes. This ensures that the extracted samples are uniformly distributed across the parameter space, allowing for better coverage of the entire parameter space with a smaller sample size, reducing the number of simulations and improving screening efficiency. The development dynamics data corresponding to each first development scheme output by the numerical simulation model of the target component, along with preset values ​​for development characteristic parameters, are used as the training sample set to train the development dynamics prediction learning model. This enables the model to learn development dynamics patterns that better match the characteristics of the target reservoir, thereby providing more accurate results in subsequent predictions and providing strong support for the formulation of development schemes.

[0102] As an optional implementation, based on the above embodiments, the formula corresponding to the preset multi-objective mathematical model includes:

[0103]

[0104]

[0105]

[0106]

[0107] Specifically, the cumulative oil production of a single well = the integral of the average oil production rate of that well over time within the prediction period (0~t); the total cumulative oil production = the total oil production of all producing wells (1~N). p The sum of the cumulative oil production of a single well (1~N); if calculated in discrete time steps, such as monthly, the integral can be replaced by summation. The amount of thickener used in a single well = the product of the cumulative CO2 injection volume of that well and the thickener concentration; the total cumulative thickener used = the total amount of thickener used in all injection wells (1~N). w The sum of thickener usage per well is calculated as follows: CO2 sequestration is the difference between the total mass of injected CO2 and the total mass of produced CO2, which is the sum of the integrals of the gas injection rate of each injection well over time within the prediction period minus the sum of the integrals of the gas production rate of each production well over time within the prediction period. The calculation results of the three indicators are correlated with the corresponding initial development scheme to form structured data output.

[0108] Optionally, the preset multi-objective mathematical model also includes constraints, such as the single-well gas injection rate not exceeding the equipment's maximum injection capacity; the thickener concentration being within an effective range to avoid ineffectiveness due to excessively low concentration or reservoir blockage due to excessively high concentration; and meeting the minimum storage requirements for environmental assessments. After calculating the values ​​of each indicator, the constraints are used to verify them, and invalid results are eliminated. If the thickener dosage exceeds the budget threshold, the scheme is deemed invalid and eliminated.

[0109] The method for determining the thickening CO2 flooding development scheme provided in this application embodiment directly reflects the improved oil recovery rate of the development scheme in terms of cumulative oil production; the cumulative amount of thickener used is closely related to the development cost; and the CO2 sequestration reflects the contribution of the development scheme to environmental protection. Through these three objective functions, the development effect of the target reservoir can be comprehensively evaluated from multiple key dimensions such as economy, cost, and environmental protection, avoiding the one-sidedness of evaluation by a single indicator.

[0110] As an optional implementation, based on the above embodiments, the multi-objective collaborative optimization algorithm is an improved CLMEA algorithm. The multi-objective collaborative optimization algorithm determines the set of candidate development schemes based on multiple index values ​​corresponding to each initial development scheme, including:

[0111] The steps are executed iteratively in the order of classifier-assisted hierarchical learning pre-screening, hypervolume-based non-dominated search, and sparse target space local search until the maximum number of iterations is reached to obtain a set of candidate development solutions.

[0112] Among them, the improved CLMEA algorithm refers to the efficient selection of non-dominated solutions with the best overall performance, i.e., candidate development schemes, through three iterative steps: pre-screening, non-dominated search, and local search.

[0113] Specifically, a small number of known high-quality solutions are first selected, such as previously screened non-dominated solutions and the optimal solutions for similar reservoirs, as positive samples. Schemes with significantly poor performance, such as those with oil production below the economic baseline or those whose storage capacity does not meet environmental protection requirements, are selected as negative samples. The index values ​​of these samples are input into a classifier to train a model that judges the index values ​​and their quality levels. The three index values ​​of all initial development schemes are then input into the trained classifier, which assigns a quality level to each scheme based on the learned patterns. Poor-quality schemes are directly eliminated, while high-quality and medium-quality schemes are retained, forming a streamlined scheme pool. Subsequent searches are only performed on schemes in this pool, significantly reducing computational load. After each iteration, newly screened non-dominated solutions are added as positive samples to update the classifier's judgment model, improving the accuracy of the next round of pre-screening.

[0114] Furthermore, for all solutions in the simplified solution pool, the indicator values ​​are compared pairwise to determine the dominance relationship. If solution A's oil production ≥ solution B, its storage volume ≥ solution B, and its usage ≤ solution B, and at least one of these conditions is strictly met, then A dominates B, and B is a dominated solution. The dominated solution is temporarily stored and marked. All dominated solutions are removed, and the remaining solutions not dominated by any solution form the initial non-dominated solution set. The total hypervolume of the initial non-dominated solution set is calculated, and then the contribution value of each solution to the total hypervolume is calculated. The solutions with the highest hypervolume contribution values ​​(top 80%~90%) are retained, and redundant solutions with small contribution values ​​are removed to obtain the optimized non-dominated solution set, ensuring that this set is both high-quality and free of redundancy.

[0115] Furthermore, the optimized set of non-dominated solutions is mapped onto the three-dimensional target space. The number of solutions within a certain range around each solution, such as oil production ±500 tons, storage capacity ±1000 tons, and usage ±10 tons, is calculated to identify sparse regions. High-quality non-dominated solutions around these sparse regions are selected as initial points. The development characteristic parameters corresponding to the initial points are slightly adjusted, such as the gas injection rate ±5m³ / s. 3 Using the following parameters: d, thickener concentration ±0.05%, transfer timing ±2% water content, several new solutions are generated. The three indices of each new solution are calculated to determine if it is a non-dominated solution. If it is a non-dominated solution, it is added to the non-dominated solution set, and the total hypervolume is recalculated; if it is a dominated solution, it is directly discarded. The above steps are repeated until there are no obvious sparse regions in the target space, or the contribution of newly added non-dominated solutions to the hypervolume is lower than a preset threshold, at which point the local search stops.

[0116] Furthermore, after completing one pre-screening, non-dominated search, and local search, the iteration count is recorded, and it is determined whether the non-dominated solution set has converged. If it has not converged and the maximum number of iterations has not been reached, the current non-dominated solution set is merged with the simplified solution pool, and the process returns to step one to re-execute the entire process; if it has converged or the maximum number of iterations has been reached, the iteration stops. The final non-dominated solution set is used as a candidate development solution set, containing the development characteristic parameter values, three index values, and the position information of each solution in the target space for each solution, providing a basis for subsequent final solution decisions.

[0117] The method for determining thickening CO2-driven development schemes provided in this application includes a classifier-assisted hierarchical learning pre-screening step, which enables rapid preliminary evaluation and classification of initial development schemes. By constructing a classifier, the algorithm can pre-screen out schemes that are clearly unsuitable or at a disadvantage based on some key features or preliminary calculation results. The non-dominated search based on hypervolume and the local search of the sparse target space are executed cyclically in a specific order, avoiding blind searches and redundant calculations, allowing the algorithm to more efficiently traverse the solution space and quickly find the set of candidate development schemes.

[0118] As an optional implementation, based on the above embodiments, a thickening CO2 flooding target development scheme corresponding to the target reservoir is determined based on the preset weight thresholds corresponding to each indicator and the indicator values ​​corresponding to each candidate development scheme, including:

[0119] Obtain the preset weight coefficients corresponding to each indicator;

[0120] Based on the values ​​of each indicator and the preset weight coefficients of each indicator for each candidate development scheme, the comprehensive evaluation value of each candidate development scheme is calculated.

[0121] The candidate solution with the highest comprehensive evaluation value and whose values ​​for each indicator are all within the corresponding preset threshold range is selected as the target development solution.

[0122] Among them, the preset weight coefficients refer to the pre-set importance coefficients of each indicator, such as cumulative oil production weight of 0.45, CO2 sequestration weight of 0.35, and thickener usage weight of 0.2.

[0123] The comprehensive evaluation value refers to the quantitative value obtained by weighting and summing according to the preset weight coefficients. The higher the value, the better the scheme performs in the multi-objective trade-off of high oil production, high storage, and low usage.

[0124] Among them, the preset threshold range is a rigid qualification standard set for each indicator, such as cumulative oil production ≥ 25,000 tons, CO2 storage ≥ 70,000 tons, and thickener usage ≤ 300 tons.

[0125] Specifically, after obtaining the preset weight coefficients for each indicator, for each candidate scheme, the comprehensive evaluation value is calculated one by one according to the formula: Comprehensive Evaluation Value = Standardized Cumulative Oil Production × Oil Production Weight + Standardized CO2 Sequestration Value × Sequestration Weight + Standardized Thickener Usage Value × Usage Weight. All candidate schemes are then sorted from highest to lowest comprehensive evaluation value. Following this order, the original indicator values ​​of each candidate scheme are checked to see if they all meet the preset thresholds. For example, if the top-ranked scheme has a cumulative oil production of 26,000 tons (≥25,000 tons), a sequestration volume of 72,000 tons (≥70,000 tons), and a thickener usage of 280 tons (≤300 tons), then it passes the threshold verification.

[0126] The method for determining a thickening CO2-driven development scheme provided in this application uses the weights of each indicator to reflect their importance and interrelationship within the overall development objective when calculating the comprehensive evaluation value. Different weight combinations lead to differences in the comprehensive evaluation value, thus demonstrating the varying degrees of influence of different indicators on the overall performance of the development scheme. Simultaneously, it requires that the values ​​of each indicator be within their corresponding preset threshold ranges, ensuring that the target development scheme not only achieves optimal overall performance but also meets certain feasibility and rationality requirements for each specific indicator. This dual-screening mechanism improves the reliability and operability of the target development scheme and reduces development risks.

[0127] As an optional implementation, after determining the thickening CO2 flooding target development scheme corresponding to the target reservoir based on the above embodiments, the method further includes:

[0128] Obtain multiple production index values ​​during the development process of the target reservoir;

[0129] In response to any production indicator value exceeding the corresponding preset threshold range, a pre-trained development dynamic prediction learning model and a multi-objective collaborative optimization algorithm are used to adjust the values ​​corresponding to each development feature data.

[0130] Among them, the production index values ​​refer to the quantitative parameters that reflect the development dynamics, collected in real time or periodically by on-site monitoring equipment during the development of thickened CO2 flooded reservoirs, such as oil production, formation pressure, and CO2 volume fraction in produced gas.

[0131] Specifically, during the development of the target reservoir, multiple production indicators are monitored in real time. These indicators are then compared to preset thresholds. If any indicator exceeds the specified range, a parameter adjustment process is immediately triggered. Current development characteristic data is extracted and used as initial values ​​to generate multiple candidate adjustment schemes within the adjustment constraints. These candidate schemes are input into a pre-trained development dynamic prediction learning model. The model outputs predicted production indicator values ​​for each scheme, providing data support for the optimization algorithm. The predicted production indicator values ​​are then input into a multi-objective collaborative optimization algorithm. This algorithm aims to achieve the target daily oil production, maximize CO2 sequestration, and minimize thickener usage. It calculates the comprehensive optimization value for each candidate scheme and selects non-dominated solutions. Based on these non-dominated solutions, the values ​​corresponding to each development characteristic data are adjusted, and development is restarted.

[0132] The method for determining a thickening CO2 flooding development scheme provided in this application acquires multiple production index values ​​during the development process of the target reservoir, enabling comprehensive real-time monitoring of the reservoir's development status. By acquiring these index values ​​in real time and comparing them with preset threshold ranges, any abnormal fluctuations or deviations from the normal range in any index can be detected promptly. By adjusting the development characteristic data in a timely manner, the reservoir development process can be kept stable and efficient, reducing production stoppages and reductions caused by abnormal index values.

[0133] Figure 3 A flowchart of a method for determining a thickening CO2 drive development scheme provided in another embodiment of this application is shown below. Figure 3 As shown, the method for determining the thickening CO2 drive development scheme provided in this embodiment includes the following steps:

[0134] Step 301: Obtain multiple initial development schemes for the target reservoir. The initial development schemes are obtained by arranging and combining multiple preset values ​​corresponding to multiple development characteristic parameters. The development characteristic parameters include the timing of transfer injection, thickener concentration, water injection rate, gas injection rate, half-cycle length, and bottom flow pressure of the oil well.

[0135] Step 302: Obtain performance data of multiple candidate thickeners under multiple reservoir conditions and historical data of the target reservoir during the water drive stage.

[0136] Step 303: Based on the performance data of each thickener and the historical data of the target reservoir, calibrate the initial component numerical simulation model to obtain the target component numerical simulation model.

[0137] Step 304: The Latin hypercube uniform sampling method is used to extract multiple first development schemes from multiple initial development schemes, and each first development scheme is input into the target component numerical simulation model to obtain the development dynamic data corresponding to each first development scheme.

[0138] Step 305: Use the development dynamic data corresponding to each first development scheme and the preset values ​​corresponding to multiple development feature parameters as training sample sets to train the development dynamic prediction learning model, so as to obtain the pre-trained development dynamic prediction learning model.

[0139] Step 306: Input each initial development scheme into the pre-trained development dynamic prediction learning model to obtain the development dynamic data corresponding to each initial development scheme. The development dynamic data includes oil production rate and gas-oil ratio.

[0140] Step 307: Using a preset multi-objective mathematical model, based on the development dynamic data corresponding to each initial development scheme and the preset values ​​corresponding to each development characteristic parameter, calculate the values ​​of multiple indicators corresponding to each initial development scheme. The multiple indicators include cumulative oil production, CO2 sequestration, and cumulative thickener usage.

[0141] Step 308: Determine the set of candidate development schemes using a multi-objective collaborative optimization algorithm based on the multiple indicator values ​​corresponding to each initial development scheme.

[0142] Step 309: Obtain the preset weight coefficients corresponding to each indicator.

[0143] Step 310: Calculate the comprehensive evaluation value of each candidate development scheme based on the values ​​of each indicator and the preset weight coefficients of each indicator.

[0144] Step 311: The candidate solution with the highest comprehensive evaluation value and whose values ​​of each indicator are all within the corresponding preset threshold range is determined as the target development solution.

[0145] Step 312: Obtain multiple production index values ​​for the target reservoir development process.

[0146] Step 313: In response to any production indicator value exceeding the corresponding preset threshold range, the values ​​corresponding to each development feature data are adjusted using a pre-trained development dynamic prediction learning model and a multi-objective collaborative optimization algorithm.

[0147] In this embodiment, the implementation method and technical effect of steps 301-313 are similar to those of the corresponding solutions in the above embodiments, and will not be repeated here.

[0148] Figure 4 This is a schematic diagram of the structure of a device for determining a thickening CO2 drive development scheme according to an embodiment of this application, as shown below. Figure 4 As shown, the device for determining the thickening CO2 drive development scheme provided in this embodiment is located in the device for determining the thickening CO2 drive development scheme. The device 40 for determining the thickening CO2 drive development scheme provided in this embodiment includes: an acquisition module 41, an input module 42, a calculation module 43, and a determination module 44.

[0149] The system comprises the following modules: Acquisition module 41, which acquires multiple initial development schemes for the target reservoir. Each initial development scheme is obtained by arranging and combining multiple preset values ​​corresponding to multiple development characteristic parameters, including injection timing, CO2 thickener concentration, water injection rate, gas injection rate, half-cycle length, and bottomhole flow pressure. Input module 42, which inputs each initial development scheme into a pre-trained development dynamic prediction learning model to obtain development dynamic data corresponding to each initial development scheme. This development dynamic data includes oil production rate and gas-oil ratio. Calculation module 43, which uses a preset multi-objective mathematical model to calculate multiple index values ​​corresponding to each initial development scheme based on the development dynamic data and preset values ​​corresponding to each development characteristic parameter. These multiple indicators include cumulative oil production, CO2 sequestration, and cumulative thickener usage. Determination module 44, which uses a multi-objective collaborative optimization algorithm to determine a set of candidate development schemes based on the multiple index values ​​corresponding to each initial development scheme. Determination module 44 also determines the thickening CO2 flooding target development scheme for the target reservoir based on preset weight thresholds corresponding to each index and the index values ​​corresponding to each candidate development scheme.

[0150] The device for determining the thickening CO2 drive development scheme provided in this embodiment can perform... Figure 2 The methods provided in the embodiments are similar in their specific implementation principles and technical effects, and will not be described in detail here.

[0151] Optionally, the device for determining the thickening CO2 drive development scheme provided in this embodiment further includes a calibration module, an extraction module, and a training module.

[0152] Accordingly, the acquisition module 41 is also used to acquire performance data of multiple candidate thickeners under multiple reservoir conditions and historical data of the target reservoir during the water drive stage; the calibration module is used to calibrate the initial component numerical simulation model based on the performance data of each thickener and the historical data of the target reservoir to obtain the target component numerical simulation model; the extraction module is used to extract multiple first development schemes from multiple initial development schemes using the Latin hypercube uniform sampling method; the input module 42 is also used to input each first development scheme into the target component numerical simulation model to obtain the development dynamic data corresponding to each first development scheme; the training module is used to train the development dynamic prediction learning model using the development dynamic data corresponding to each first development scheme and the preset values ​​corresponding to multiple development feature parameters as training sample sets to obtain the pre-trained development dynamic prediction learning model.

[0153] Optionally, the formulas corresponding to the preset multi-objective mathematical model include:

[0154]

[0155]

[0156]

[0157]

[0158] Optionally, the multi-objective collaborative optimization algorithm is an improved CLMEA algorithm. The determination module 44 is specifically used to determine the candidate development scheme set by using the multi-objective collaborative optimization algorithm based on the multiple index values ​​corresponding to each initial development scheme: cyclically execute each step in the order of classifier-assisted hierarchical learning pre-screening, non-dominated search based on hypervolume, and local search in sparse target space until the maximum number of iterations is reached, so as to obtain the candidate development scheme set.

[0159] Optionally, the determining module 44, when determining the target development scheme for thickening CO2 flooding corresponding to the target reservoir based on the preset weight thresholds corresponding to each indicator and the values ​​of each indicator corresponding to each candidate development scheme, is specifically used for: obtaining the preset weight coefficients corresponding to each indicator; calculating the comprehensive evaluation value of each candidate development scheme based on the values ​​of each indicator corresponding to each candidate development scheme and the preset weight coefficients corresponding to each indicator; and determining the candidate scheme with the highest comprehensive evaluation value and all indicator values ​​within the corresponding preset threshold range as the target development scheme.

[0160] Optionally, the device for determining the thickening CO2 drive development scheme provided in this embodiment further includes an adjustment module.

[0161] Accordingly, the acquisition module 41 is also used to acquire multiple production index values ​​of the target reservoir development process; the adjustment module is used to adjust the values ​​corresponding to each development feature data by using a pre-trained development dynamic prediction learning model and a multi-objective collaborative optimization algorithm in response to any production index value exceeding the corresponding preset threshold range.

[0162] Figure 5 A schematic diagram of the structure of a device for determining a thickening CO2 drive development scheme according to an embodiment of this application is shown below. Figure 5 As shown, the device 50 for determining the thickening CO2 drive development scheme provided in this embodiment includes: a processor 51 and a memory 52 that is communicatively connected to the processor.

[0163] The memory 52 stores computer-executable instructions; the processor 51 executes the computer-executable instructions stored in the memory 52 to implement the method for determining the thickening CO2 drive development scheme provided in the above embodiment. Related explanations can be understood by referring to the relevant descriptions and effects corresponding to the steps in the accompanying drawings, and will not be elaborated upon here.

[0164] The program may include program code, which includes computer-executable instructions. Memory 52 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device.

[0165] In this embodiment, the processor 51 and the memory 52 are connected via a bus. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0166] This application also provides a computer-readable storage medium storing computer-executable instructions. When the controller executes the computer-executable instructions, it implements the various steps in the methods described above.

[0167] This application also provides a computer program product, including a computer program that, when executed by a controller, implements the various steps in the methods described above.

[0168] The various embodiments described above in this application can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0169] The computer-executable instructions used to implement the methods of this application may be written in any combination of one or more programming languages. These computer-executable instructions may be provided to the processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the computer-executable instructions cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer-executable instructions may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a standalone software package, or entirely on a remote machine or electronic device.

[0170] In the context of this application, a computer-readable storage medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be a machine-readable signal medium or a machine-readable storage medium. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of computer-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), optical fibers, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. Alternatively, computer-readable storage media may include: resistive random access memory (RRAM), dynamic random access memory (DRAM), static random access memory (SRAM), enhanced dynamic random access memory (EDRAM), high-bandwidth memory (HBM), hybrid memory cube (HMC), and so on.

[0171] The systems and technologies described herein can be implemented in computing systems that include back-end components (e.g., as data electronic devices), or computing systems that include middleware components (e.g., application electronic devices), or computing systems that include front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0172] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps can be performed in other orders or simultaneously according to this application. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all optional embodiments, and the actions and modules involved are not necessarily essential to this application. In other words, the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps disclosed in this application can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this application can be achieved, and this is not limited herein.

[0173] It should be further noted that although the steps in the flowchart are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowchart may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0174] It should be understood that the above-described device embodiments are merely illustrative, and the device of this application can also be implemented in other ways. For example, the division of units / modules in the above embodiments is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple units, modules, or components may be combined, or integrated into another system, or some features may be ignored or not executed.

[0175] Furthermore, unless otherwise specified, the functional units / modules in the various embodiments of this application can be integrated into one unit / module, or each unit / module can exist physically separately, or two or more units / modules can be integrated together. The integrated units / modules described above can be implemented in hardware or as software program modules.

[0176] When an integrated unit / module is implemented in hardware, that hardware can be digital circuits, analog circuits, etc. The physical implementation of the hardware structure includes, but is not limited to, transistors, memristors, etc.

[0177] If the integrated unit / module is implemented as a software program module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to related technologies, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing computer-executable instructions, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0178] In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments. The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0179] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only.

[0180] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. Therefore, the specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the principles of this application should be included within the scope of protection of this application.

Claims

1. A method for determining a thickening CO2 flooding development scheme, characterized in that, The method includes: Multiple initial development schemes for the target reservoir are obtained. Each initial development scheme is obtained by arranging and combining multiple preset values ​​corresponding to multiple development characteristic parameters. The development characteristic parameters include injection timing, thickener concentration, water injection rate, gas injection rate, half-cycle length, and bottom flow pressure of the oil well. Each of the initial development schemes is input into a pre-trained development dynamics prediction learning model to obtain development dynamics data corresponding to each of the initial development schemes. The development dynamics data includes oil production rate and gas-oil ratio. A preset multi-objective mathematical model is used to calculate multiple index values ​​corresponding to each of the initial development schemes based on the development dynamic data and preset values ​​of each development characteristic parameter. The multiple index values ​​include cumulative oil production, CO2 sequestration, and cumulative thickener usage. A set of candidate development schemes is determined by using a multi-objective collaborative optimization algorithm based on the multiple indicator values ​​corresponding to each initial development scheme. Based on the preset weight thresholds corresponding to each of the aforementioned indicators and the values ​​of each indicator corresponding to each candidate development scheme, the thickening CO2 flooding target development scheme corresponding to the target reservoir is determined.

2. The method according to claim 1, characterized in that, After obtaining multiple initial development plans for the target reservoir, the method further includes: Acquire performance data of multiple candidate thickeners under multiple reservoir conditions and historical data of the target reservoir during the waterflooding stage; The initial component numerical simulation model was calibrated based on the performance data of each thickener and the historical data of the target reservoir to obtain the target component numerical simulation model; The Latin hypercube uniform sampling method is used to extract multiple first development schemes from multiple initial development schemes, and each first development scheme is input into the target component numerical simulation model to obtain the development dynamic data corresponding to each first development scheme. The development dynamic data corresponding to each first development scheme and the preset values ​​corresponding to multiple development feature parameters are used as training sample sets to train the development dynamic prediction learning model, so as to obtain the pre-trained development dynamic prediction learning model.

3. The method according to claim 1, characterized in that, The formulas corresponding to the preset multi-objective mathematical model include:

4. The method according to claim 1, characterized in that, The multi-objective collaborative optimization algorithm is an improved CLMEA algorithm. The step of determining the candidate development scheme set using the multi-objective collaborative optimization algorithm based on multiple index values ​​corresponding to each initial development scheme includes: The steps are executed iteratively in the order of classifier-assisted hierarchical learning pre-screening, hypervolume-based non-dominated search, and sparse target space local search until the maximum number of iterations is reached to obtain a set of candidate development solutions.

5. The method according to claim 1, characterized in that, The process of determining the thickening CO2 flooding target development scheme for the target reservoir based on the preset weight thresholds corresponding to each of the aforementioned indicators and the indicator values ​​corresponding to each candidate development scheme includes: Obtain the preset weight coefficients corresponding to each indicator; Based on the values ​​of each indicator and the preset weight coefficients of each indicator for each candidate development scheme, the comprehensive evaluation value of each candidate development scheme is calculated. The candidate solution with the highest comprehensive evaluation value and whose values ​​for each indicator are all within the corresponding preset threshold range is selected as the target development solution.

6. The method according to claim 1, characterized in that, After determining the target thickening CO2 flooding development scheme corresponding to the target reservoir, the following is also included: Obtain multiple production index values ​​during the development process of the target reservoir; In response to any production indicator value exceeding the corresponding preset threshold range, a pre-trained development dynamic prediction learning model and a multi-objective collaborative optimization algorithm are used to adjust the values ​​corresponding to each development feature data.

7. A device for determining a thickening CO2 flooding development scheme, characterized in that, The device includes: The acquisition module is used to acquire multiple initial development schemes for the target reservoir. The initial development schemes are obtained by arranging and combining multiple preset values ​​corresponding to multiple development characteristic parameters. The development characteristic parameters include the timing of the transfer injection, CO2 thickener concentration, water injection rate, gas injection rate, half-cycle length, and bottom flow pressure of the oil well. The input module is used to input each of the initial development schemes into a pre-trained development dynamic prediction learning model to obtain development dynamic data corresponding to each of the initial development schemes. The development dynamic data includes oil production rate and gas-oil ratio. The calculation module is used to calculate multiple index values ​​corresponding to each of the initial development schemes based on the development dynamic data corresponding to each of the initial development schemes and the preset values ​​corresponding to each development characteristic parameter using a preset multi-objective mathematical model. The multiple indexes include cumulative oil production, CO2 storage, and cumulative thickener usage. The determination module is used to determine the set of candidate development schemes based on multiple indicator values ​​corresponding to each initial development scheme using a multi-objective collaborative optimization algorithm. The determining module is further configured to determine the thickening CO2 flooding target development scheme corresponding to the target reservoir based on the preset weight thresholds corresponding to each of the indicators and the indicator values ​​corresponding to each candidate development scheme.

8. A device for determining a thickening CO2 flooding development scheme, characterized in that, The device includes: a processor, and a memory communicatively connected to the processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory to implement the method as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-6.

10. A computer program product comprising a computer program that, when executed by a processor, implements the method of any one of claims 1-6.

Images