Methods, systems, equipment, and storage media for CO2 enhanced gas recovery and its sequestration.

Abstract

The present application relates to the technical field of carbon dioxide capture, utilization and geological storage, and particularly to a method, system, device and storage medium for CO2 enhanced gas reservoir recovery and storage, the method comprising: S100, acquiring dynamic parameters of a developed target gas reservoir in a current state, and analyzing key factors; S200, constructing a potential prediction model for CO2 enhanced recovery of the target gas reservoir, and generating a CO2 displacement recovery enhancement range of the target gas reservoir; S300, analyzing whether the CO2 displacement recovery enhancement range of the target gas reservoir is greater than an economic production range; S400, constructing a basic numerical simulation model of an actual development well group of the target gas reservoir, and generating an economic net present value and a CO2 geological storage amount; S500, generating a conversion timing and corresponding simulation engineering parameters by using a Bayesian adaptive direct search algorithm, and calculating a target function; and S600, generating optimal engineering parameters. By using the present application, the CO2 recovery rate can be improved, the effective storage amount can be maximized, the optimization cost can be reduced, and the optimization efficiency can be improved.

Description

Technical Field

[0001] This invention relates to the field of carbon dioxide capture, utilization and geological storage technology, and particularly to... Methods, systems, equipment, and storage media for enhancing gas reservoir recovery and storage. Background Technology

[0002] Traditional methods for predicting the development potential of gas reservoirs often employ numerical simulations to model the production dynamics of the reservoir and obtain its development potential. However, these methods require specialized component simulators, which are typically expensive and computationally inefficient, making it difficult to quickly make decisions regarding the conversion of reservoirs during actual development. Existing gas reservoirs Drive parameter optimization is mostly a single-objective optimization of a single process, and has not yet been realized. An optimized design for coordinated gas expulsion and storage is needed. Therefore, there is an urgent need for a method that can simultaneously achieve... Methods to improve recovery rate and maximize effective storage, while reducing optimization costs and improving optimization efficiency. Summary of the Invention

[0003] This invention provides Methods, systems, equipment, and storage media for enhancing gas reservoir recovery and storage can simultaneously achieve… The goal is to improve the recovery rate and maximize the effective storage capacity, while reducing optimization costs and improving optimization efficiency.

[0004] The basic solution provided by this invention is as follows:

[0005] Methods for enhancing gas reservoir recovery and storage include the following steps:

[0006] S100: Obtain dynamic parameters of the developed target gas reservoir in its current state. These dynamic parameters include geological parameters, development parameters, and engineering parameters. Analyze the factors affecting the target gas reservoir. Key factors for oil recovery rate;

[0007] S200, construct the target gas reservoir based on the aforementioned key factors. A potential prediction model for enhanced oil recovery; using the potential prediction model, target gas reservoirs are generated. The extent to which recovery rate was increased;

[0008] S300, Analyze the target gas reservoir Does the increase in recovery rate exceed the economic recovery rate? If so, then adopt... If not, then the current development method will be used for development.

[0009] S400, based on the aforementioned key factors, construct a basic numerical simulation model of the actual development well group of the target gas reservoir; through the basic numerical simulation model, simulate injection... Natural gas extraction and geological storage generate economic net present value and Geological reserves;

[0010] S500 uses a Bayesian adaptive direct search algorithm to generate the conversion timing and corresponding simulation engineering parameters, and calculates the objective function based on the simulation engineering parameters;

[0011] S600 analyzes the objective function corresponding to each simulated engineering parameter and generates the optimal engineering parameters.

[0012] Furthermore, the geological parameters include one or more of the following: rock porosity, permeability, oil / gas / water saturation, wettability, reservoir depth, oil / gas / water interface depth, effective thickness, and degree of heterogeneity; the development parameters include one or more of the following: formation temperature, formation pressure, recovery rate, daily gas production, water cut, and remaining reserves; the engineering parameters include one or more of the following: well location coordinates, perforation location, gas production rate, and bottom hole flowing pressure for each individual well.

[0013] S100 includes:

[0014] The geological parameters, development parameters, and engineering parameters of each region are combined to generate several parameter combinations, and corresponding conceptual numerical models are constructed for each parameter combination.

[0015] Using the aforementioned conceptual numerical model, annotations are performed respectively. Development and non-payment The developed simulation generates annotations. Development and recovery rate and uninjected Develop and improve the recovery rate; inject Development recovery rate minus uninjected Develop the recovery rate and generate the conceptual numerical model. Driven development and recovery rate range;

[0016] Calculate each dynamic parameter and The correlation coefficient of the recovery rate of the oil recovery is calculated using the following formula:

[0017]

[0018] In the formula, The correlation coefficient is given by X1 and X2, where X1 and X2 are variables, DX represents variance, Cov represents covariance, and E represents expectation.

[0019] Calculate each dynamic parameter and The Pearson coefficient for the recovery rate of oil recovery is calculated using the following formula:

[0020]

[0021] In the formula, Indicates the standard deviation of each variable;

[0022] Dynamic parameters whose corresponding correlation coefficients are less than a preset correlation coefficient and whose corresponding Pearson coefficients are greater than a preset Pearson coefficient are selected as those affecting the target gas reservoir. Key factors for oil recovery rate;

[0023] Based on the single-factor analysis method, the value ranges of each key factor are generated.

[0024] Furthermore, the S200 includes:

[0025] Within the range of values ​​for each key factor, several combinations of key parameters are generated, and numerical simulation models corresponding to each combination of key parameters are constructed.

[0026] Calculate the numerical simulation models The extent to which oil recovery is improved; using the aforementioned key factors as inputs to the input layer. The increase in oil recovery rate is used as the output of the output layer. Machine learning models are trained and evaluation matrices are calculated for each model. The machine learning models include one or more of artificial neural networks, gradient boosting decision trees, extreme gradient boosting decision trees, and lightweight gradient boosting machines. The evaluation matrices include mean absolute error, mean relative error, root mean square error, and coefficient of determination.

[0027] The formula for calculating the evaluation matrix is ​​as follows:

[0028]

[0029]

[0030]

[0031]

[0032] In the formula, MAE is the mean absolute error, MRE is the mean relative error, RMSE is the root mean square error, and R0 is the mean square error. 2 As the coefficient of determination, For calculation using a numerical simulation model The extent to which recovery rate was increased Predicted by machine learning models The extent to which recovery rate was increased All calculated The average increase in oil recovery rate n The number of samples;

[0033] Based on the evaluation matrices of each machine learning model, a machine learning model is selected, and the target gas reservoir is constructed based on the selected machine learning model. A predictive model for the potential to enhance oil recovery.

[0034] Furthermore, the S300 also includes:

[0035] The economic extraction margin is calculated using the following formula:

[0036]

[0037] In the formula, For the first k Year The injection cost For the first k Year The amount of injection, N t To evaluate time, N p Geological reserves, For the first k Annual natural gas sales price, To increase the economic growth rate.

[0038] Furthermore, the S400 includes:

[0039] Based on key factors, a basic numerical simulation model of the actual development well group of the target gas reservoir is constructed, and production dynamic history is fitted.

[0040] The natural gas extraction method in the basic numerical simulation model was adjusted to injection. ;injection Monitoring During the process, the maximum formation pressure in the basic numerical simulation model is analyzed to determine if the maximum formation pressure equals a pressure threshold. If so, the injection process is stopped. ;

[0041] Computational injection The economic net present value (NPV) during the process is calculated using the following formula:

[0042]

[0043] In the formula, NPV For economic net present value, , The first i Koujing in the k Annual gas and water production The selling price of oil. The cost of treating the produced water, For gas injection costs, N t To evaluate time, N p For the number of production wells, N i For the number of injection wells, b The discount rate is... For the kth year The amount injected;

[0044] Stop injection Subsequently, the geological storage process is simulated using the aforementioned basic numerical simulation model to obtain the tectonic storage capacity, dissolution storage capacity, residual phase storage capacity, and mineralization storage capacity, and the final storage capacity of the target gas reservoir is calculated. Geological reserves;

[0045] The The formula for calculating geological reserves is as follows:

[0046]

[0047] In the formula, N s For the final target gas reservoir Geological reserves N c To construct the storage quantity, N g This refers to the amount of residual phase stored. N a This refers to the amount dissolved and sealed. N m This refers to the amount of mineralized sequestration.

[0048] Furthermore, the S500 includes:

[0049] The Bayesian adaptive direct search algorithm is used to generate the conversion timing and corresponding simulation engineering parameters. The Bayesian adaptive direct search algorithm continuously updates the iteration points by alternating between searching and polling. The search process includes searching for regions where the Gaussian value is higher than the first preset value and the Gaussian value is lower than the preset value, and iteratively selecting new evaluation points. The polling process includes evaluating the deployment in any direction.

[0050] The transition timing includes when the target gas reservoir is converted to injection using existing development methods. The timing of gas reservoir development is quantified by the average formation pressure during the development of the target gas reservoir. Timing of driver development transition;

[0051] The simulation engineering parameters include one or more of the following: injection rate, injection pressure, total gas injection volume, and gas production rate;

[0052] Calculate the implementation based on the simulated engineering parameters. The objective function for maximizing gas recovery and storage capacity is defined by the following formula:

[0053]

[0054] In the formula: F ( x Let ) be the objective function. The weight of the net present value. for The weight of geological reserves.

[0055] Furthermore, in S600, the cumulative number of operations in S500 is obtained, and it is determined whether the cumulative number of operations is greater than the preset maximum number of updates. If not, S500 is returned to be executed. If so, the objective function corresponding to each simulation engineering parameter is analyzed, and the optimal engineering parameters are generated.

[0056] The second basic solution provided by this invention:

[0057] A system for enhancing gas reservoir recovery and storage includes a key factor analysis module and a potential prediction model construction module. The module includes a potential extraction analysis module, a basic numerical simulation model construction module, an objective function calculation module, and an objective function analysis module.

[0058] The key factor analysis module is used to obtain dynamic parameters of the developed target gas reservoir in its current state. These dynamic parameters include geological parameters, development parameters, and engineering parameters, and analyze the factors affecting the target gas reservoir. Key factors for oil recovery rate;

[0059] The potential prediction model construction module is used to construct the target gas reservoir based on the key factors. A potential prediction model for enhanced oil recovery; using the potential prediction model, target gas reservoirs are generated. The extent to which recovery rate was increased;

[0060] The The extraction potential analysis module is used to analyze the target gas reservoir. Does the increase in recovery rate exceed the economic recovery rate? If so, then adopt... If not, then the current development method will be used for development.

[0061] The basic numerical simulation model construction module is used to construct a basic numerical simulation model of the actual development well group of the target gas reservoir based on the key factors; and to simulate injection using the basic numerical simulation model. Natural gas extraction and geological storage generate economic net present value and Geological reserves;

[0062] The objective function calculation module is used to generate the transformation timing and corresponding simulation engineering parameters using the Bayesian adaptive direct search algorithm, and to calculate the objective function based on the simulation engineering parameters.

[0063] The objective function analysis module is used to analyze the objective function corresponding to each simulated engineering parameter and generate the optimal engineering parameters.

[0064] The third basic solution provided by this invention:

[0065] An apparatus for enhancing gas reservoir recovery and its sequestration includes a processor and a memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the aforementioned functions. Methods to improve gas reservoir recovery and storage.

[0066] The fourth basic solution provided by this invention:

[0067] A storage medium for enhancing gas reservoir recovery and its sequestration, used to store computer-executable instructions, which, when executed, achieve the above-mentioned goals. Methods to improve gas reservoir recovery and storage.

[0068] The principles and advantages of this invention are as follows:

[0069] Based on historical data, we analyze the geological, development, and engineering parameters that influence the target gas reservoir. Key factors for oil recovery and the construction of target gas reservoirs. A potential prediction model for enhanced oil recovery is used to obtain the target gas reservoir. The extent to which the recovery rate is improved; the determination of the target gas reservoir. Does the increase in recovery rate exceed the economic recovery rate? If so, it indicates that the optimization scheme is effective and can be adopted. The driver is developed to ensure optimization results; otherwise, it indicates that it is being used. Developing the gas reservoir by driving the flow could actually reduce recovery and storage capacity, so it's crucial to terminate the simulation promptly to improve optimization efficiency. Based on these key factors, a basic numerical simulation model of the actual development well group for the target gas reservoir should be constructed to simulate injection. Natural gas extraction and geological storage, analysis of injection sites The actual benefits after conversion are then assessed. A Bayesian adaptive direct search algorithm is then used to generate the conversion timing and corresponding simulation engineering parameters. Based on these parameters, the objective function is calculated to obtain the optimal conversion timing. Once the cumulative number of calculations reaches the preset maximum update count, the objective functions corresponding to each calculation process are integrated, and the simulation engineering parameters corresponding to the optimal objective function are selected as the best engineering parameters. Using the above scheme, from the conversion (unnoted) Note The simulation considered multiple aspects, including effectiveness, optimal conversion timing, and maximizing conversion benefits, to ensure the optimal conversion scheme for the gas reservoir. The effectiveness of the driving measures, and the ability to achieve them simultaneously. The goal is to improve the recovery rate and maximize the effective storage capacity, while reducing optimization costs and improving optimization efficiency. Attached Figure Description

[0070] Figure 1 This is an embodiment of the present invention. A flowchart of methods for enhancing gas reservoir recovery and its storage.

[0071] Figure 2 This is an embodiment of the present invention. A planar schematic diagram of the conceptual numerical model in methods for enhancing gas reservoir recovery and storage.

[0072] Figure 3 This is an embodiment of the present invention. A schematic diagram simulating the recovery rate range in methods for enhancing gas reservoir recovery and its storage.

[0073] Figure 4 This is an embodiment of the present invention. Correlation thermograms in methods to enhance gas reservoir recovery and storage.

[0074] Figure 5 This is an embodiment of the present invention. A schematic diagram illustrating the calculation results of the Pearson coefficient in methods for enhancing gas reservoir recovery and storage.

[0075] Figure 6 This is an embodiment of the present invention. Artificial neural networks in methods to enhance gas reservoir recovery and storage Cross-plot of predicted and actual values ​​of driving potential.

[0076] Figure 7 This is an embodiment of the present invention. Gradient boosting decision tree as a method to enhance gas reservoir recovery and storage Cross-plot of potential forecasts and actual values.

[0077] Figure 8 This is an embodiment of the present invention. Ultimate Gradient Boosting Decision Tree: Methods for Enhancing Gas Reservoir Recovery and Storage Cross-plot of potential forecasts and actual values.

[0078] Figure 9 This is an embodiment of the present invention. Lightweight gradient lift machine is a method for enhancing gas reservoir recovery and storage. Cross-plot of potential forecasts and actual values.

[0079] Figure 10 This is an embodiment of the present invention. A schematic diagram of a numerical simulation model of an actual well group in methods to improve gas reservoir recovery and storage.

[0080] Figure 11 This is an embodiment of the present invention. A schematic diagram of historical fitting of cumulative gas production and water cut in actual well groups in methods for improving gas reservoir recovery and storage.

[0081] Figure 12 This is an embodiment of the present invention. A dynamic diagram illustrating the objective function optimization in methods for improving gas reservoir recovery and storage. Detailed Implementation

[0082] The following detailed description illustrates the specific implementation method:

[0083] Example 1:

[0084] Methods to enhance gas reservoir recovery and storage, such as Figure 1 As shown, it includes the following steps:

[0085] S100: Obtain the dynamic parameters of the developed target gas reservoir in its current state, and analyze the factors affecting the target gas reservoir. Key factors for recovery rate.

[0086] The dynamic parameters include geological parameters, development parameters, and engineering parameters. The geological parameters include one or more of the following: rock porosity, permeability, oil / gas / water saturation, wettability, reservoir depth, oil / gas / water interface depth, effective thickness, and degree of heterogeneity. In this application, the symbol " / " means "or". The development parameters include one or more of the following: formation temperature, formation pressure, recovery rate, daily gas production, water cut, and remaining reserves. The engineering parameters include one or more of the following: well location coordinates, perforation location, gas production rate, and bottom hole flowing pressure. In this embodiment, the geological parameters, development parameters, and engineering parameters all include all of the aforementioned parameters.

[0087] S100 includes:

[0088] Geological parameters, development parameters, and engineering parameters from various regions are combined to generate several parameter combinations, and corresponding conceptual numerical models are constructed for each parameter combination. In this embodiment, geological parameters, development parameters, and engineering parameters from historical data are randomly combined.

[0089] Using the aforementioned conceptual numerical model, annotations are performed respectively. Development and non-payment The developed simulation generates annotations. Development and recovery rate and uninjected Develop and improve the recovery rate; inject Development recovery rate minus uninjected Develop the recovery rate and generate the conceptual numerical model. Driven development and recovery rate range;

[0090] Calculate each dynamic parameter and The correlation coefficient of the recovery rate of the oil recovery is calculated using the following formula:

[0091]

[0092] In the formula, The correlation coefficient is given by X1 and X2, where X1 and X2 are variables, DX represents variance, Cov represents covariance, and E represents expectation.

[0093] Calculate each dynamic parameter and The Pearson coefficient for the recovery rate of oil recovery is calculated using the following formula:

[0094]

[0095] In the formula, This represents the standard deviation of each variable;

[0096] Dynamic parameters whose corresponding correlation coefficients are less than a preset correlation coefficient and whose corresponding Pearson coefficients are greater than a preset Pearson coefficient are selected as those affecting the target gas reservoir. Key factors for oil recovery rate;

[0097] Based on the single-factor analysis method, the value ranges of each key factor are generated.

[0098] S200, construct the target gas reservoir based on the aforementioned key factors. A potential prediction model for enhanced oil recovery; using the potential prediction model, target gas reservoirs are generated. The extent to which the recovery rate was increased.

[0099] S200 includes:

[0100] Within the range of values ​​for each key factor, several combinations of key parameters are generated, and numerical simulation models corresponding to each combination of key parameters are constructed.

[0101] Calculate the numerical simulation models The extent to which oil recovery is improved; using the aforementioned key factors as inputs to the input layer. The increase in oil recovery rate is used as the output of the output layer. The machine learning model is trained and the evaluation matrix of each machine learning model is calculated. The machine learning model includes one or more of artificial neural networks, gradient boosting decision trees, extreme gradient boosting decision trees, and lightweight gradient boosting machines.

[0102] In this embodiment, the machine learning models to be screened include artificial neural networks, gradient boosting decision trees, extreme gradient boosting decision trees, and lightweight gradient boosting machines; the evaluation matrix includes mean absolute error, mean relative error, root mean square error, and coefficient of determination.

[0103] The formula for calculating the evaluation matrix is ​​as follows:

[0104]

[0105]

[0106]

[0107]

[0108] In the formula, MAE is the mean absolute error, MRE is the mean relative error, RMSE is the root mean square error, and R0 is the mean square error. 2 As the coefficient of determination, For calculation using a numerical simulation model The extent to which recovery rate was increased Predicted by machine learning models The extent to which recovery rate was increased All calculated The average increase in recovery rate n The number of samples;

[0109] Based on the evaluation matrices of each machine learning model, a machine learning model is selected, and the target gas reservoir is constructed based on the selected machine learning model. A predictive model for the potential to enhance oil recovery.

[0110] S300, Analyze the target gas reservoir Whether the increase in recovery rate exceeds the economic recovery rate determines whether the target gas reservoir is suitable. Driven development; if so, then adopt... If the driver is not used, then the current development method will be used for development.

[0111] The economic extraction margin is calculated using the following formula:

[0112]

[0113] In the formula, For the first k Year The injection cost For the first k Year The amount of injection, N t To evaluate time, N p Geological reserves, For the first k Annual natural gas sales price, To increase the economic growth rate.

[0114] S400, based on the aforementioned key factors, construct a basic numerical simulation model of the actual development well group of the target gas reservoir; through the basic numerical simulation model, simulate injection... Natural gas extraction and geological storage generate economic net present value and Geological reserves.

[0115] The S400 includes:

[0116] Based on key factors, a basic numerical simulation model of the actual development well group of the target gas reservoir is constructed, and production dynamic history is fitted.

[0117] The natural gas extraction method in the basic numerical simulation model was adjusted to injection. ;injection Monitoring During the process, the maximum formation pressure in the basic numerical simulation model is analyzed to determine if the maximum formation pressure equals a pressure threshold. If so, the injection process is stopped. Otherwise, injection is urgently needed. In this embodiment, the pressure threshold is 80% of the formation rock fracture pressure.

[0118] Computational injection The economic net present value (NPV) during the process is calculated using the following formula:

[0119]

[0120] In the formula, NPV For economic net present value, , The first i Koujing in thek Annual gas and water production, in t·a -1 and m 3 ·a -1 ; The selling price of oil is expressed in yuan per ton. -1 ; The cost of treating the produced water is expressed in yuan per ton. -1 ; The cost of gas injection is expressed in yuan per cubic meter. -3 ; N t The evaluation time is in years. N p This refers to the number of producing wells, expressed in wells. N i The number of injection wells, in units of wells; b The discount rate is expressed in % (%).

[0121] Stop injection Subsequently, the geological storage process is simulated using the aforementioned basic numerical simulation model to obtain the tectonic storage capacity, dissolution storage capacity, residual phase storage capacity, and mineralization storage capacity, and the final storage capacity of the target gas reservoir is calculated. Geological reserves, the final amount of the target gas reservoir Geological reserves are the sum of tectonic reserves, dissolution reserves, residual phase reserves, and mineralization reserves;

[0122] The The formula for calculating geological reserves is as follows:

[0123]

[0124] In the formula, N s For the final target gas reservoir Geological reserves N c To construct the storage quantity, N g This refers to the amount of residual phase stored. N a This refers to the amount dissolved and sealed. N m This refers to the amount of mineralized sequestration.

[0125] S500 uses a Bayesian adaptive direct search algorithm to generate conversion opportunities and corresponding simulation engineering parameters, and calculates the objective function based on the simulation engineering parameters. In this embodiment, several conversion opportunities are generated, and the objective function is calculated based on the corresponding simulation engineering parameters. Then, based on the objective function under each conversion opportunity, the optimal conversion opportunity is analyzed. The specific method is as follows.

[0126] The S500 includes:

[0127] The Bayesian adaptive direct search algorithm is used to generate the conversion timing and corresponding simulation engineering parameters. The Bayesian adaptive direct search algorithm continuously updates the iteration points by alternating between searching and polling. The search process includes searching for regions where the Gaussian value is higher than the first preset value and the Gaussian value is lower than the preset value, and iteratively selecting new evaluation points. The polling process includes evaluating the deployment in any direction.

[0128] The transition timing includes when the target gas reservoir is transitioned to injection using existing development methods. The timing of gas reservoir development is quantified by the average formation pressure during the development of the target gas reservoir. Timing of driver development transition;

[0129] The simulation engineering parameters include one or more of injection rate, injection pressure, total gas injection volume, and gas production rate; in this embodiment, the simulation engineering parameters include injection rate, injection pressure, total gas injection volume, and gas production rate.

[0130] Calculate the implementation based on the simulated engineering parameters. The objective function for maximizing gas recovery and storage capacity includes two sub-objectives and weights for each sub-objective; the sub-objectives include Maximizing the net present value of the gas displacement process Maximizing the amount of sequestration during geological sequestration; The net present value of the gas displacement process and During geological sequestration, the amount of sequestration is normalized using the arctangent function and substituted into the objective function;

[0131] Specifically, the formula for calculating the objective function is as follows:

[0132]

[0133] In the formula: F ( x Let ) be the objective function. The weight of the net present value. for The weight of geological reserves.

[0134] S600 analyzes the objective function corresponding to each simulated engineering parameter and generates the optimal engineering parameters.

[0135] In S600, the cumulative number of operations in S500 is obtained, and it is determined whether the cumulative number of operations is greater than the preset maximum number of updates. If not, S500 is returned to be executed. If so, the objective function corresponding to each simulation engineering parameter is analyzed, and the optimal engineering parameters are generated.

[0136] In this embodiment, the preset maximum number of updates is related to the set maximum number of simulations. Specifically, if the set maximum number of simulations has not been reached, the simulation engineering parameters will continue to be iteratively updated for calculation; if the set maximum number of simulations has been reached, the updating of the simulation engineering parameters will stop.

[0137] The system for enhancing gas reservoir recovery and storage employs the aforementioned Methods to improve gas reservoir recovery and storage include, specifically, a key factor analysis module and a potential prediction model construction module. The module includes a potential extraction analysis module, a basic numerical simulation model construction module, an objective function calculation module, and an objective function analysis module.

[0138] The key factor analysis module is used to obtain dynamic parameters of the developed target gas reservoir in its current state. These dynamic parameters include geological parameters, development parameters, and engineering parameters, and analyze the factors affecting the target gas reservoir. Key factors for oil recovery rate;

[0139] The potential prediction model construction module is used to construct the target gas reservoir based on the key factors. A potential prediction model for enhanced oil recovery; using the potential prediction model, target gas reservoirs are generated. The extent to which recovery rate was increased;

[0140] The The extraction potential analysis module is used to analyze the target gas reservoir. Does the increase in recovery rate exceed the economic recovery rate? If so, then adopt... If not, then the current development method will be used for development.

[0141] The basic numerical simulation model construction module is used to construct a basic numerical simulation model of the actual development well group of the target gas reservoir based on the key factors; and to simulate injection using the basic numerical simulation model. Natural gas extraction and geological storage generate economic net present value and Geological reserves;

[0142] The objective function calculation module is used to generate the transformation timing and corresponding simulation engineering parameters using the Bayesian adaptive direct search algorithm, and to calculate the objective function based on the simulation engineering parameters.

[0143] The objective function analysis module is used to analyze the objective function corresponding to each simulated engineering parameter and generate the optimal engineering parameters.

[0144] An apparatus for enhancing gas reservoir recovery and its sequestration includes a processor and a memory configured to store computer-executable instructions, which, when executed, cause the processor to perform the aforementioned functions. Methods to improve gas reservoir recovery and storage.

[0145] A storage medium for enhancing gas reservoir recovery and its sequestration, used to store computer-executable instructions, which, when executed, achieve the above-mentioned goals. Methods to improve gas reservoir recovery and storage.

[0146] Specifically, the above If methods for enhancing gas reservoir recovery and its sequestration are implemented as software functional units and sold or used as independent products, they can be stored in a readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a readable storage medium, and when executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms.

[0147] Readable storage media may include: any entity or device capable of carrying the computer program code, recording media, USB flash drive, portable hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0148] Example 2:

[0149] The basic principle of Example 2 is the same as that of Example 1. The difference is that in Example 2, the geological parameters include top depth, reservoir thickness, net-to-gross ratio, porosity, and permeability; the development parameters include formation temperature, formation pressure, recovery rate, daily gas production, water cut, and remaining reserves; and the engineering parameters include well spacing information.

[0150] To clearly illustrate the method, a specific practical application is presented below, using an actual depleted production block in a gas field as an example. The specific implementation process is as follows:

[0151] First, the geological parameters, development parameters, and engineering parameters of the gas field at its current development stage are obtained. These parameters are then combined to generate several parameter combinations, and a corresponding conceptual numerical model is constructed for each combination. In this embodiment, the geological parameters of the target gas reservoir are shown in Table 1, and the conceptual numerical model is as follows: Figure 2 As shown.

[0152] Table 1 Geological Parameters

[0153]

[0154] Annotate each conceptual numerical model separately Development and non-payment The developed simulation generates annotations. Development and recovery rate and uninjected Development and recovery rate; such as Figure 3 As shown, the injection Development and recovery rate (recovery rate in carbon dioxide flooding development) minus uninjected The development and recovery rate (the degree of depletion-type recovery) is used to generate the conceptual numerical model. Driven development and recovery rate range (continued) (The extent to which recovery rate was increased).

[0155] Calculate each dynamic parameter and The correlation coefficients of the recovery rate of oil recovery in drive development are shown in the figure below. Figure 4 As shown, in this embodiment, based on the correlation coefficient calculation results, the net-to-coarse ratio and current extraction level are removed. The updated dynamic parameters are then calculated and... The Pearson coefficient for the recovery rate of oil recovery in flooding is shown below. The updated Pearson coefficient calculation results for each dynamic parameter are as follows: Figure 5 As shown. Dynamic parameters with Pearson coefficients greater than a preset Pearson coefficient are selected to influence the target gas reservoir. In this embodiment, the key factors for oil recovery are calculated based on the Pearson coefficient, excluding current formation pressure, current daily gas production, current formation temperature, and top depth. The remaining key factors include remaining reserves, reservoir thickness, permeability, porosity, well spacing, and water cut. A single-factor analysis method is used to generate the value ranges for each key factor. In this embodiment, the value ranges are shown in Table 2.

[0156] Table 2 Value Ranges for Key Factors

[0157]

[0158] Within the range of values ​​for the aforementioned key factors, several combinations of key parameters are generated, and numerical simulation models corresponding to each combination of key parameters are constructed; the numerical simulation models are then calculated. The extent to which oil recovery is improved; using the aforementioned key factors as inputs to the input layer. The increase in recovery rate is used as the output of the output layer, and a machine learning model is used for training.

[0159] In this embodiment, the training result of the artificial neural network is as follows: Figure 6 As shown, the training results of the gradient boosting decision tree are as follows: Figure 7 As shown, the training results of the extreme gradient boosting decision tree are as follows: Figure 8 As shown, the training results of the lightweight gradient booster are as follows: Figure 9 As shown.

[0160] The evaluation matrix of each machine learning model is calculated. In this embodiment, the calculation results of the evaluation matrix are shown in Table 3.

[0161] Table 3 Evaluation Matrix of Machine Learning Models

[0162]

[0163] Based on the evaluation matrices of each machine learning model, a machine learning model is selected. In this embodiment, the Lightweight Gradient Boosting Machine algorithm is preferred, and the target gas reservoir is constructed based on the Lightweight Gradient Boosting Machine. A predictive model for the potential to enhance oil recovery.

[0164] Input the parameters of the target area into the target gas reservoir. In this embodiment, the input parameters for the enhanced oil recovery potential prediction model include an effective thickness of 11.37 m and a permeability of 3200 × 10⁻⁶ m. -3 μm 2 The conditions at the time of the intervention were: water cut 71.43%, formation temperature 41.18℃, formation pressure 1.86MPa, recovery rate 14.90%, and remaining reserves 244.24×10⁻⁶. 8 m 3 Well spacing 120m, target area prediction adopted After development, the recovery rate can be increased by up to 25%, and the economic recovery rate of the target gas reservoir is 21.63%, indicating that the target gas reservoir is suitable for development. Drive development.

[0165] like Figure 10 As shown, based on the aforementioned key factors, a basic numerical simulation model of the actual development well group of the target gas reservoir is constructed, and production dynamic history is fitted; the production dynamic history fitting results are as follows. Figure 11 As shown.

[0166] The natural gas extraction method in the basic numerical simulation model was adjusted to injection. ;injection Monitoring During the process, the maximum formation pressure in the basic numerical simulation model is analyzed to determine if the maximum formation pressure equals a pressure threshold. If so, the injection process is stopped. Otherwise, injection is urgently needed. In this embodiment, the injection is stopped. At that time, the net present value of the economy was 90.6558 million yuan; continuing... Simulation of the geological sealing process yields the final result. The geological reserves are 644.71 × 10⁻⁶. 4 t.

[0167] A Bayesian adaptive direct search algorithm is used to generate the transition timing and corresponding simulation engineering parameters, and iterative optimization is performed. The optimization process is as follows: Figure 12 As shown. In this embodiment, the optimal rotation was obtained after 200 simulations. The displacement was carried out at a formation pressure of 18.35 MPa; the injection rate of the injection well was 190.5 m / s. 3 ·d -1 111.4m 3 ·d -1 193.7m 3 ·d -1 189.3m 3 ·d -1 169.5m 3 ·d -1 67.2m 3 ·d -1 169.2m 3 ·d -1 200m 3 ·d -1 221.6m 3 ·d -1 200m 3 ·d -1 221.4m 3 ·d -1 215.8m 3 ·d -1 The production rate of the production well was 143.4 t·d. -1 52.3t·d -1 113.8t·d -1 173.8t·d -1 119.1t·d -1 173.8t·d -1 The net present value of this plan is 114,366,800 yuan, and ultimately... The geological reserves are 812.18 × 10⁻⁶. 4 t.

[0168] This approach first predicts the potential for resource extraction through the measures, effectively avoiding the problem of underperformance due to blind implementation of measures; secondly, it... The two sub-processes of gas extraction and geological storage are integrated into one process, ensuring that the optimization results can simultaneously guarantee... The economic efficiency of gas-driven development and The safety and effectiveness of geological sealing can provide on-site [information / reliability]. It provides strong support for rapid decision-making regarding gas extraction and geological storage.

[0169] The above are merely embodiments of the present invention. Commonly known structures and characteristics are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, under the guidance of this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for improving gas reservoir recovery and its storage, characterized by: Includes the following steps: S100: Obtain dynamic parameters of the developed target gas reservoir in its current state. These dynamic parameters include geological parameters, development parameters, and engineering parameters. Analyze the factors affecting the target gas reservoir. Key factors for oil recovery; The geological parameters include one or more of the following: rock porosity, permeability, oil / gas / water saturation, wettability, reservoir depth, oil / gas / water interface depth, effective thickness, and degree of heterogeneity; the development parameters include one or more of the following: formation temperature, formation pressure, recovery rate, daily gas production, water cut, and remaining reserves. The engineering parameters include one or more of the following: well location coordinates, perforation location, gas production rate, and bottom hole flowing pressure for each individual well. S100 includes: The geological parameters, development parameters, and engineering parameters of each region are combined to generate several parameter combinations, and corresponding conceptual numerical models are constructed for each parameter combination. Using the aforementioned conceptual numerical model, annotations are performed respectively. Development and non-payment The developed simulation generates annotations. Development and recovery rate and uninjected Develop and improve the recovery rate; inject Development recovery rate minus uninjected Develop the recovery rate and generate the conceptual numerical model. The extent to which the recovery rate is driven by development; Calculate each dynamic parameter and The correlation coefficient of the recovery rate of the oil recovery is calculated using the following formula: In the formula, The correlation coefficient is given by X1 and X2, where X1 and X2 are variables, DX represents variance, Cov represents covariance, and E represents expectation. Calculate each dynamic parameter and The Pearson coefficient for the recovery rate of oil recovery is calculated using the following formula: In the formula, Indicates the standard deviation of each variable; Dynamic parameters whose corresponding correlation coefficients are less than a preset correlation coefficient and whose corresponding Pearson coefficients are greater than a preset Pearson coefficient are selected as those affecting the target gas reservoir. Key factors for oil recovery; Based on the single-factor analysis method, the value range of each key factor is generated; S200, construct the target gas reservoir based on the aforementioned key factors. A potential prediction model for enhanced oil recovery; using the potential prediction model, target gas reservoirs are generated. The extent to which recovery rate was increased; S300, Analyze the target gas reservoir Does the increase in recovery rate exceed the economic recovery rate? If so, then adopt... If not, then the current development method will be used for development. S400, based on the aforementioned key factors, construct a basic numerical simulation model of the actual development well group of the target gas reservoir; through the basic numerical simulation model, simulate injection... Natural gas extraction and geological storage generate economic net present value and Geological reserves; S500 uses a Bayesian adaptive direct search algorithm to generate the conversion timing and corresponding simulation engineering parameters, and calculates the objective function based on the simulation engineering parameters; S600 analyzes the objective function corresponding to each simulated engineering parameter and generates the optimal engineering parameters.

2. As described in claim 1 A method for improving gas reservoir recovery and its storage, characterized by: S200 includes: Within the range of values ​​for each key factor, several combinations of key parameters are generated, and numerical simulation models corresponding to each combination of key parameters are constructed. Calculate the numerical simulation models The extent to which oil recovery is improved; using the aforementioned key factors as inputs to the input layer. The increase in oil recovery rate is used as the output of the output layer. Machine learning models are trained and evaluation matrices are calculated for each machine learning model. The machine learning models include one or more of artificial neural networks, gradient boosting decision trees, extreme gradient boosting decision trees, and lightweight gradient boosting machines. The evaluation matrices include mean absolute error, mean relative error, root mean square error, and coefficient of determination. The formula for calculating the evaluation matrix is ​​as follows: In the formula, MAE is the mean absolute error, MRE is the mean relative error, RMSE is the root mean square error, and R0 is the mean square error. 2 As the coefficient of determination, For calculation using a numerical simulation model The extent to which recovery rate was increased Predicted by machine learning models The extent to which recovery rate was increased All calculated The average increase in oil recovery rate n The number of samples; Based on the evaluation matrices of each machine learning model, a machine learning model is selected, and the target gas reservoir is constructed based on the selected machine learning model. A predictive model for the potential to enhance oil recovery.

3. As described in claim 1 A method for improving gas reservoir recovery and its storage, characterized by: The S300 also includes: The economic extraction margin is calculated using the following formula: In the formula, For the first k Year The injection cost For the first k Year The amount of injection, N t To evaluate time, N p Geological reserves, For the first k Annual natural gas sales price, To increase the economic growth rate.

4. As described in claim 1 A method for improving gas reservoir recovery and its storage, characterized by: The S400 includes: Based on key factors, a basic numerical simulation model of the actual development well group of the target gas reservoir is constructed, and production dynamic history is fitted. The natural gas extraction method in the basic numerical simulation model was adjusted to injection. ;injection Monitoring During the process, the maximum formation pressure in the basic numerical simulation model is analyzed to determine if the maximum formation pressure equals a pressure threshold. If so, the injection process is stopped. ; Computational injection The economic net present value (NPV) during the process is calculated using the following formula: In the formula, NPV For economic net present value, , The first i Koujing in the k Annual gas and water production The selling price of oil. The cost of treating the produced water, For gas injection costs, N t To evaluate time, N p For the number of production wells, N i For the number of injection wells, b The discount rate is... For the kth year The amount injected; Stop injection Subsequently, the geological storage process is simulated using the aforementioned basic numerical simulation model to obtain the tectonic storage capacity, dissolution storage capacity, residual phase storage capacity, and mineralization storage capacity, and the final storage capacity of the target gas reservoir is calculated. Geological reserves; The The formula for calculating geological reserves is as follows: In the formula, N s For the final target gas reservoir Geological reserves N c To construct the storage quantity, N g This refers to the amount of residual phase stored. N a This refers to the amount dissolved and sealed. N m This refers to the amount of mineralized sequestration.

5. The method according to claim 4 A method for improving gas reservoir recovery and its storage, characterized by: The S500 includes: The Bayesian adaptive direct search algorithm is used to generate the conversion timing and corresponding simulation engineering parameters. The Bayesian adaptive direct search algorithm continuously updates the iteration points by alternating between searching and polling. The search process includes searching for regions where the Gaussian value is higher than the first preset value and the Gaussian value is lower than the preset value, and iteratively selecting new evaluation points. The polling process includes evaluating the deployment in any direction. The transition timing includes when the target gas reservoir is converted to injection using existing development methods. The timing of gas reservoir development is quantified by the average formation pressure during the development of the target gas reservoir. Timing of driver development transition; The simulation engineering parameters include one or more of the following: injection rate, injection pressure, total gas injection volume, and gas production rate; Calculate the implementation based on the simulated engineering parameters. The objective function for maximizing gas recovery and storage capacity is defined by the following formula: In the formula: F ( x Let ) be the objective function. The weight of the net present value. for The weight of geological reserves.

6. The method according to claim 1 A method for improving gas reservoir recovery and its storage, characterized by: In S600, the cumulative number of operations in S500 is obtained, and it is determined whether the cumulative number of operations is greater than the preset maximum number of updates. If not, S500 is returned to be executed. If so, the objective function corresponding to each simulation engineering parameter is analyzed, and the optimal engineering parameters are generated.

7. The system for enhancing gas reservoir recovery and its storage uses any one of claims 1-6. A method for improving gas reservoir recovery and its storage, characterized by: Includes a key factor analysis module, a potential prediction model construction module, The module includes a potential extraction analysis module, a basic numerical simulation model construction module, an objective function calculation module, and an objective function analysis module. The key factor analysis module is used to obtain dynamic parameters of the developed target gas reservoir in its current state. These dynamic parameters include geological parameters, development parameters, and engineering parameters, and analyze the factors affecting the target gas reservoir. Key factors for oil recovery rate; The potential prediction model construction module is used to construct the target gas reservoir based on the key factors. A potential prediction model for enhanced oil recovery; using the potential prediction model, target gas reservoirs are generated. The extent to which recovery rate was increased; The The extraction potential analysis module is used to analyze the target gas reservoir. Does the increase in recovery rate exceed the economic recovery rate? If so, then adopt... If not, then the current development method will be used for development. The basic numerical simulation model construction module is used to construct a basic numerical simulation model of the actual development well group of the target gas reservoir based on the key factors; and to simulate injection using the basic numerical simulation model. Natural gas extraction and geological storage generate economic net present value and Geological reserves; The objective function calculation module is used to generate the transformation timing and corresponding simulation engineering parameters using the Bayesian adaptive direct search algorithm, and to calculate the objective function based on the simulation engineering parameters. The objective function analysis module is used to analyze the objective function corresponding to each simulated engineering parameter and generate the optimal engineering parameters.

8. Equipment for enhancing gas reservoir recovery and storage, characterized in that: Includes a processor and a memory configured to store computer-executable instructions, which, when executed, cause the processor to perform any one of claims 1-6. Methods to improve gas reservoir recovery and storage.

9. A storage medium for enhancing gas reservoir recovery and its sequestration, used to store computer-executable instructions, characterized in that: the computer-executable instructions, when executed, implement any one of claims 1-6. Methods to improve gas reservoir recovery and storage.

Images