Method for determining development parameters of a natural gas hydrate reservoir and related apparatus

Abstract

The application provides a method for determining development parameters of a natural gas hydrate reservoir and related equipment. In the process of particle swarm intelligent optimization, a robot learning algorithm is used for data mining and training to obtain a training model. The potential particles are determined based on the training model. The global optimal objective function is determined based on the combination of the potential particles and the updated development parameters. In the case where the convergence condition is reached based on the development parameter combination corresponding to the global optimal objective function value, the development parameter combination corresponding to the global optimal objective function value is determined as the target development parameter combination. The calculation process of the intelligent optimization algorithm can be improved by the machine learning algorithm, and the optimization time of the development parameters of the natural gas hydrate reservoir is shortened.

Description

Technical Field

[0001] This application relates to the field of natural gas hydrate reservoir development, and in particular to a method and related equipment for determining development parameters of natural gas hydrate reservoirs. Background Technology

[0002] Natural gas hydrates are solid compounds formed from water and natural gas (mainly methane) molecules under low temperature and high pressure. Currently, natural gas hydrates, as a promising strategic alternative energy source, have become a major focus in contemporary earth science and the energy industry. Optimizing the development parameters of natural gas hydrate reservoirs is a crucial aspect of natural gas hydrate development schemes. How to efficiently and quickly optimize development parameters and determine a reasonable development plan is key to natural gas pilot production.

[0003] Currently, natural gas hydrate development schemes are mainly studied using physical experimental methods. However, numerical simulations are widely used in oil and gas field development engineering. Compared with the development of existing gas fields, the development mechanism of natural gas hydrate reservoirs is more complex. Due to the complexity of the development mechanism of natural gas hydrate reservoirs and the strong nonlinearity of numerical simulation models, a single simulation takes a long time. Using intelligent optimization algorithms for multi-factor optimization is a common method for designing development schemes, but its implementation requires repeated calls to the numerical simulator. Practice shows that when using particle swarm optimization algorithms to optimize multiple development parameters of natural gas hydrate reservoirs, the convergence speed is slow, and the calculation time can be as long as tens of days. How to accelerate the optimization process of natural gas hydrate reservoir development parameters is a current challenge. Summary of the Invention

[0004] To address the aforementioned issues, this application provides a method and related equipment for determining development parameters of natural gas hydrate reservoirs.

[0005] This application provides a method for determining development parameters of natural gas hydrate reservoirs, including:

[0006] A numerical simulation model of a natural gas hydrate reservoir is obtained, and the numerical simulation model is determined as the objective function model of the particle swarm optimization algorithm.

[0007] Determine the development parameter information of the particle swarm optimization algorithm, and determine the development parameter combination of the i-th particle based on the development parameter information;

[0008] Based on the objective function model, determine the objective function value corresponding to the combination of development parameters, and based on the objective function value and the historical objective function value, determine the current combination of development parameters for the i-th particle;

[0009] A training model is obtained by performing machine learning based on the current combination of development parameters and the corresponding objective function value, as well as the historical combination of development parameters and the corresponding objective function value; potential particles are determined based on the training model and the development parameter information.

[0010] The current development parameter combination of the i-th particle is updated based on the development parameter information to obtain the updated development parameter combination. The objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle are determined based on the objective function model. The global optimal objective function value is determined based on the objective function value corresponding to the updated development parameters and the objective function value corresponding to the potential particle.

[0011] If the convergence condition is met based on the development parameter combination corresponding to the global optimal objective function value, the development parameter combination corresponding to the global optimal objective function value is determined as the target development parameter combination for the natural gas hydrate reservoir, where 1≤i≤N and N is the number of particles.

[0012] In some embodiments, the development parameter information includes: the number of development parameters and the range corresponding to each development parameter, wherein the value of the j-th development parameter of the i-th particle is given by the following formula:

[0013]

[0014] Where j = 1, 2, ..., M, n For the number of iterations, Let M be the value of the j-th development parameter of the i-th particle, and P be the number of development parameters. j,low P is the lower limit value of the j-th development parameter. j,high The upper limit value of the j-th development parameter is given, and rand() is a random function in the range of 0 to 1.

[0015] In some embodiments, updating the development parameter combination of the i-th particle based on the development parameter information to obtain the updated development parameter combination includes:

[0016] The updated development parameter combination is obtained by updating the development parameter combination of the i-th particle based on the development parameter information and the preset update method, wherein the preset update method is given by the following formula:

[0017]

[0018] Where rand1() and rand2() are both random functions in the range of 0 to 1; ω is the inertia weight; c1 and c2 are both constants; This is the updated value of the j-th development parameter for the i-th particle; The updated value of the j-th development parameter for the i-th particle; Let j be the value of the development parameter in the current historical best development parameter combination for the i-th particle; This is the j-th development parameter value in the current globally optimal combination of development parameters.

[0019] In some embodiments, determining potential particles based on the trained model and the development parameter information includes:

[0020] The sampling points for each development parameter are determined based on the parameter range of each development parameter.

[0021] Based on the aforementioned development parameter sampling points, determine the combination of sampling development parameters;

[0022] The objective function value corresponding to the sampling development parameter combination is determined based on the sampling development parameter combination and the training model;

[0023] The objective function values ​​corresponding to the sampling development parameter combinations are sorted to obtain the sorting information of the objective function values ​​corresponding to the sampling development parameter combinations;

[0024] Potential particles are determined at least based on the sorting information.

[0025] In some embodiments, determining potential particles at least based on the sorting includes:

[0026] The top Z particles are determined based on the sorting information;

[0027] Determine the dimensionless distance between any two particles in the first Z particles;

[0028] Among the first Z particles, particles whose dimensionless distance between two particles is greater than a preset threshold are identified as potential particles. The dimensionless distance between the two particles is calculated using the following formula:

[0029]

[0030] Among them, L X,Y For dimensionless distance, P X,j For the j-th development parameter of the X-th particle, P Y,j Let be the j-th development parameter of the Y-th particle.

[0031] In some embodiments, determining the globally optimal objective function value based on the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle includes:

[0032] Determine the first magnitude relationship between the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the historical optimal development parameters;

[0033] The current optimal objective function value is determined based on the first size relationship;

[0034] Determine a second magnitude relationship between the current optimal objective function value and the objective function value corresponding to the potential particle;

[0035] The largest objective function value is determined as the globally optimal objective function value.

[0036] In some embodiments, the method further includes:

[0037] The training model is trained based on the updated combination of development parameters and the corresponding objective function value.

[0038] This application provides a device for determining development parameters, including:

[0039] The acquisition module is used to acquire the numerical simulation model of the natural gas hydrate reservoir and determine the numerical simulation model as the objective function model of the particle swarm algorithm.

[0040] The first determining module is used to determine the development parameter information of the particle swarm algorithm, and to determine the development parameter combination of the i-th particle based on the development parameter information;

[0041] The second determining module is used to determine the objective function value corresponding to the development parameter combination based on the objective function model, and to determine the current development parameter combination of the i-th particle based on the objective function value and the historical objective function value.

[0042] The first training module is used to perform machine learning to obtain a training model based on the current combination of development parameters and the corresponding objective function value, and the historical combination of development parameters and the corresponding objective function value; and to determine potential particles based on the training model and the development parameter information.

[0043] The third determining module is used to update the development parameter combination of the i-th particle based on the current development parameter information to obtain the updated development parameter combination, and to determine the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle based on the objective function model, and to determine the global optimal objective function value based on the objective function value corresponding to the updated development parameters and the objective function value corresponding to the potential particle.

[0044] The fourth determining module is used to determine the development parameter combination corresponding to the global optimal objective function value as the target development parameter combination for natural gas hydrate reservoirs when the convergence condition is met, based on the development parameter combination corresponding to the global optimal objective function value. Here, i = 1, 2...N, and N is the number of particles.

[0045] This application provides an electronic device, including a memory and a processor. The memory stores a computer program, which, when executed by the processor, performs the method for determining the development parameters of any of the above-described natural gas hydrate reservoirs.

[0046] This application provides a storage medium storing a computer program that can be executed by one or more processors and can be used to implement the method for determining development parameters of natural gas hydrate reservoirs as described above.

[0047] This application provides a method and related equipment for determining development parameters of natural gas hydrate reservoirs. First, a numerical simulation model of the natural gas hydrate reservoir is constructed. This numerical simulation model is then used as the objective function model for a particle swarm optimization algorithm. During the intelligent optimization process of the particle swarm optimization, a robot learning algorithm is used for data mining and training to obtain a training model. Potential particles are determined based on the training model. Then, a globally optimal objective function is determined based on the potential particles and an updated combination of development parameters. When the convergence condition is met based on the development parameter combination corresponding to the globally optimal objective function value, the development parameter combination corresponding to the globally optimal objective function value is determined as the target development parameter combination. This method improves the computation process of the intelligent optimization algorithm through machine learning algorithms, shortening the optimization time required for the development parameters of natural gas hydrate reservoirs. Attached Figure Description

[0048] The present application will be described in more detail below based on embodiments and with reference to the accompanying drawings.

[0049] Figure 1 A schematic diagram illustrating the implementation process of a method for determining development parameters of a natural gas hydrate reservoir, provided in an embodiment of this application;

[0050] Figure 2 A schematic diagram illustrating the implementation process for determining potential particles provided in this application embodiment;

[0051] Figure 3 A schematic diagram illustrating another implementation process for determining potential particles provided in this application embodiment;

[0052] Figure 4 A schematic diagram illustrating the implementation process of another method for determining development parameters of a natural gas hydrate reservoir provided in this application embodiment;

[0053] Figure 5 This is a comparison of the performance of the algorithm of this invention and the traditional particle swarm optimization algorithm;

[0054] Figure 6 A schematic diagram of an apparatus for determining development parameters of a natural gas hydrate reservoir, provided in an embodiment of this application;

[0055] Figure 7 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application.

[0056] In the accompanying drawings, the same parts are referred to by the same reference numerals, and the drawings are not drawn to scale. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0059] If the application documents contain similar descriptions such as "first, second, third", the following explanation shall be added: In the following description, the terms "first, second, third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0061] To address the problems existing in related technologies, this application provides a method for determining the development parameters of natural gas hydrate reservoirs. This method is applied to electronic devices, such as computers and mobile terminals. The function implemented by the method for determining the development parameters of natural gas hydrate reservoirs provided in this application can be achieved by the processor of the electronic device calling program code, wherein the program code can be stored in a computer storage medium.

[0062] Example 1

[0063] This application provides a method for determining development parameters of natural gas hydrate reservoirs. Figure 1 A schematic diagram illustrating the implementation process of a method for determining development parameters of a natural gas hydrate reservoir provided in this application embodiment is shown below. Figure 1 As shown, it includes:

[0064] Step S101: Obtain the numerical simulation model of the natural gas hydrate reservoir, and determine the numerical simulation model as the objective function model of the particle swarm algorithm.

[0065] In this embodiment, the electronic device can connect to a server to obtain a numerical simulation model of the natural gas hydrate reservoir. In this embodiment, the numerical simulation model is used to simulate and calculate the development effects of different extraction parameters for the natural gas hydrate reservoir. The numerical simulation model is defined as the objective function model of a particle swarm optimization algorithm.

[0066] Step S102: Determine the development parameter information of the particle swarm algorithm, and determine the development parameter combination of the i-th particle based on the development parameter information.

[0067] In this embodiment, the development parameter information of a natural gas hydrate reservoir may include the number of development parameters for the natural gas hydrate reservoir and the range of development parameters for each natural gas hydrate reservoir. The number of development parameters can be represented by M, and M is usually greater than 3. In this embodiment, the range of each development parameter can be characterized by an upper limit and a lower limit, for example, P j,low P is the lower limit value of the j-th development parameter. j,high This is the upper limit value of the j-th development parameter.

[0068] In this embodiment, the number of particles N can be selected according to actual needs, typically ranging from 10 to 50. In this embodiment, the values ​​of each development parameter for each particle can be determined based on the range of each development parameter in the development parameter information, thereby determining the development parameter combination for the i-th particle.

[0069] The value of the j-th development parameter of the i-th particle is taken from the following formula:

[0070]

[0071] Where j = 1, 2, ..., M, and n is the iteration number. Let M be the value of the j-th development parameter of the i-th particle, and P be the number of development parameters. j,low P is the lower limit value of the j-th development parameter. j,high The upper limit value of the j-th development parameter is given, and rand() is a random function in the range of 0 to 1.

[0072] In this embodiment of the application, the i-th particle can be any particle in the particle swarm optimization process, i.e., i = 1, 2, ... N.

[0073] In this embodiment of the application, the combination of development parameters for all N particles can be determined based on particle development parameter information.

[0074] Step S103: Determine the objective function value corresponding to the development parameter combination based on the objective function model, and determine the current development parameter combination of the i-th particle based on the objective function value and the historical objective function value.

[0075] In this embodiment, the combination of development parameters can be input into the objective function model to obtain the objective function value corresponding to the combination of development parameters. In this embodiment, the objective function value corresponding to the i-th particle... Where n is the number of iterations. In the embodiments of this application, the objective function values ​​of each particle can be compared. The size of , the maximum objective function value is . The corresponding combination of development parameters is the current globally optimal combination of development parameters. Current globally optimal combination of development parameters This represents the current combination of development parameters for the i-th particle.

[0076] In some embodiments, the N particles are independent of each other, in which case parallel computation can be used to determine the objective function value. Then, based on the objective function value of each particle, the current combination of development parameters for each particle is determined. Parallel computation can shorten the total computation time.

[0077] Step S104: Based on the current combination of development parameters and the corresponding objective function value, and the historical combination of development parameters and the corresponding objective function value, machine learning is performed to obtain a training model; potential particles are determined based on the training model and the development parameter information.

[0078] In this embodiment, the current development parameter combination and historical development parameter combinations can be used as inputs, and the objective function values ​​corresponding to the current development parameter combination and historical development parameter combinations can be used as outputs. An initial model using a preset algorithm is then trained to obtain a training model, which can be a regression algorithm. The training model is a correlation model between the development parameter combinations of the particles and the objective function values.

[0079] In some embodiments, a training model is obtained by performing machine learning based on the current combination of development parameters and the corresponding objective function value, and historical combinations of development parameters and the corresponding objective function values. When determining potential particles based on the training model and the development parameter information, the calculation can be performed in parallel with step S103. The electronic device can allocate certain computing resources to perform machine learning training while calculating the objective function values ​​of N particles to obtain the training model. During machine learning training, as the number of historical combinations of development parameters and the corresponding objective function values ​​increases with the number of iterations, the machine learning time will gradually increase. When the number of historical combinations of development parameters and the corresponding objective function values ​​exceeds a preset threshold, a selection method can be used to select a portion of historical combinations of development parameters and the corresponding objective function values ​​for machine learning training to reduce the machine learning time. In some embodiments, when the machine learning time exceeds the calculation time of step S103, a selection method can be used to reduce the machine learning time. In the above embodiments, only the M examples with the highest objective function values ​​can be selected for machine learning training.

[0080] In this embodiment, after the training model is determined, random or equidistant sampling can be performed based on the range of each development parameter in the development parameter information to obtain development parameter sampling points for each development parameter; based on the development parameter sampling points, a sampled development parameter combination is determined; based on the sampled development parameter combination and the training model, the objective function value corresponding to the sampled development parameter combination is determined; the objective function values ​​corresponding to the sampled development parameter combination are sorted to obtain sorting information of the objective function values ​​corresponding to the sampled development parameter combination; and at least the potential particles are determined based on the sorting information. For example, L potential particles are extracted.

[0081] Step S105: Update the current development parameter combination of the i-th particle based on the development parameter information to obtain the updated development parameter combination, and determine the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle based on the objective function model. Determine the global optimal objective function value based on the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle.

[0082] In this embodiment, the development parameters of the current development parameter combination can be updated according to the range of each development parameter in the development parameter information to obtain the updated development parameter combination. After determining the updated development parameter combination, the updated development parameter combination and the development parameter combination corresponding to the potential particle are input into the objective function model to calculate the objective function value corresponding to each updated development parameter combination and the objective function value corresponding to the potential particle.

[0083] In this embodiment of the application, a first size relationship can be determined between the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the historical optimal development parameter combination; the current optimal objective function value is determined based on the first size relationship.

[0084] In this embodiment, the objective function value corresponding to the newly developed parameter combination can be updated to match the objective function value corresponding to the historical best developed parameter combination, thereby determining the current optimal objective function value and the corresponding current optimal developed parameter combination. For example, the current objective function value of the i-th particle in the particle swarm optimization algorithm... (That is, the objective function value corresponding to the updated parameter combination) and the objective function value corresponding to the current historical best development parameter combination. like Greater than Then the objective function value corresponding to its current historical optimal combination of development parameters equal Its current historical best combination of development parameters Equal to the current development parameter combination P i n+1 (That is, the current optimal combination of development parameters), otherwise... equal equal

[0085] After determining the current optimal development objective function value, a second magnitude relationship is established between the current optimal development objective function value and the objective function value corresponding to the potential particle; the largest objective function value is determined as the globally optimal objective function value. For example, comparing the objective function values ​​corresponding to N particles in the particle swarm optimization algorithm and L potential particles selected by the machine learning method, the maximum value of the objective function value corresponding to the L potential particles is... The corresponding combination of development parameters is like Greater than the objective function value corresponding to the current globally optimal combination of development parameters. but equal Right now The current globally optimal combination of development parameters represents the globally optimal target parameter value. equal Right now corresponding This represents the optimal combination of development parameters corresponding to the globally optimal objective function value. Otherwise... equal equal

[0086] Step S106: If the convergence condition is met based on the development parameter combination corresponding to the global optimal objective function value, the development parameter combination corresponding to the global optimal objective function value is determined as the target development parameter combination for the natural gas hydrate reservoir, where i = 1, 2...N, and N is the number of particles.

[0087] In this embodiment of the application, after the global optimal objective function value is determined, steps S102 to S105 are executed. If the development parameter combination corresponding to the global optimal objective function value is not updated after a preset number of optimizations, it is determined that the convergence condition has been met. If the convergence condition has not been met, steps S102 to S105 are executed.

[0088] In this embodiment of the application, when the convergence condition is met, the combination of development parameters corresponding to the globally optimal objective function value is determined as the target development parameter combination.

[0089] This application provides a method for determining development parameters of natural gas hydrate reservoirs. First, a numerical simulation model of the natural gas hydrate reservoir is constructed. This numerical simulation model is then used as the objective function model for a particle swarm optimization algorithm. During the intelligent optimization process of the particle swarm optimization, a robot learning algorithm is used for data mining and training to obtain a training model. Potential particles are determined based on the training model. Then, a globally optimal objective function is determined based on the potential particles and an updated combination of development parameters. When the convergence condition is met based on the development parameter combination corresponding to the globally optimal objective function value, this combination of development parameters is determined as the target development parameter combination. This method improves the computation process of the intelligent optimization algorithm through machine learning algorithms, shortening the optimization time required for the development parameters of natural gas hydrate reservoirs.

[0090] Example 2

[0091] Based on the foregoing embodiments, this application further provides a method for determining development parameters of a natural gas hydrate reservoir, including:

[0092] Step S201: Obtain the numerical simulation model of the natural gas hydrate reservoir, and determine the numerical simulation model as the objective function model of the particle swarm optimization algorithm; determine the development parameter information of the particle swarm optimization algorithm, and determine the development parameter combination of the i-th particle based on the development parameter information.

[0093] In this embodiment, the objective function model is a numerical model of a natural gas hydrate reservoir. This model allows for the calculation of the objective function value corresponding to each particle during the optimization process. In this embodiment, the development parameter information may include the number of development parameters and the range of each parameter. The number of development parameters can be represented by M, which is typically greater than 3. In this embodiment, the range of each development parameter can be characterized by an upper limit and a lower limit, for example, P... j,low P is the lower limit value of the j-th development parameter. j,high This is the upper limit value of the j-th development parameter.

[0094] In this embodiment, the number of particles N can be selected according to actual needs, typically ranging from 10 to 50. In this embodiment, the values ​​of each development parameter for each particle can be determined based on the range of each development parameter in the development parameter information, thereby determining the development parameter combination for the i-th particle.

[0095] The value of the j-th development parameter of the i-th particle is taken from the following formula:

[0096]

[0097] Where j = 1, 2, ..., M, and n is the iteration number. Let M be the value of the j-th development parameter of the i-th particle, and P be the number of development parameters. j,low P is the lower limit value of the j-th development parameter. j,high The upper limit value of the j-th development parameter is given, and rand() is a random function in the range of 0 to 1.

[0098] In this embodiment of the application, the i-th particle can be any particle in the particle swarm optimization process, i.e., i = 1, 2, ... N.

[0099] Step S202: Determine the objective function value corresponding to the development parameter combination based on the objective function model, and determine the current development parameter combination of the i-th particle based on the objective function value and the historical objective function value.

[0100] In this embodiment, the combination of development parameters can be input into the objective function model to obtain the objective function value corresponding to the combination of development parameters. In this embodiment, the objective function value corresponding to the i-th particle... in, n Let P be the number of iterations. The development parameter combination P for the i-th particle. i n That is, its current historical optimal combination of development parameters. That is, its current historical best objective function value.

[0101] In this embodiment of the application, the target function values ​​of each particle can be compared. The size of , the maximum objective function value is . The corresponding combination of development parameters is the current globally optimal combination of development parameters. Current globally optimal combination of development parameters This represents the current combination of development parameters for the i-th particle.

[0102] Step S203: Based on the current combination of development parameters and the corresponding objective function value, and the historical combination of development parameters and the corresponding objective function value, machine learning is performed to obtain a training model; potential particles are determined based on the training model and the development parameter information.

[0103] Once the training model is determined, random or equidistant sampling can be performed based on the range of each development parameter in the development parameter information to obtain development parameter sampling points for each development parameter. Based on these sampling points, a combination of sampled development parameters is determined. The objective function value corresponding to the sampled development parameter combination is determined based on the sampled development parameter combination and the training model. The objective function values ​​corresponding to the sampled development parameter combinations are sorted to obtain sorting information. At least potential particles are determined based on this sorting information. For example, L potential particles are extracted.

[0104] The current and historical development parameter combinations can be used as inputs, and the objective function values ​​corresponding to the current and historical development parameter combinations can be used as outputs. A training model is obtained by training the model using a preset algorithm, which can be a regression algorithm. The training model is a correlation model between the development parameter combinations of the particles and the objective function values.

[0105] Step S204: Based on the development parameter information and the preset update method, update the development parameter combination of the i-th particle to obtain the updated development parameter combination.

[0106] In this embodiment of the application, the preset update method is described in the following formula:

[0107]

[0108] Where rand1() and rand2() are both random functions in the range of 0 to 1; ω is the inertia weight; c1 and c2 are both constants; This is the updated value of the j-th development parameter for the i-th particle; The updated value of the j-th development parameter for the i-th particle; Let j be the value of the development parameter in the current historical best development parameter combination for the i-th particle; This is the j-th development parameter value in the current globally optimal combination of development parameters.

[0109] Step S205: Based on the objective function model, determine the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle; and determine the globally optimal objective function value based on the objective function value corresponding to the updated development parameters and the objective function value corresponding to the potential particle.

[0110] Step S206: If the convergence condition is met based on the development parameter combination corresponding to the global optimal objective function value, the development parameter combination corresponding to the global optimal objective function value is determined as the target development parameter combination, where i = 1, 2...N, and N is the number of particles.

[0111] This application provides a method for determining development parameters of natural gas hydrate reservoirs. In the process of particle swarm optimization, a robot learning algorithm is used for data mining and training to obtain a training model. Potential particles are determined based on the training model. Then, a globally optimal objective function is determined based on the potential particles and an updated combination of development parameters. When the convergence condition is met based on the development parameter combination corresponding to the globally optimal objective function value, the development parameter combination corresponding to the globally optimal objective function value is determined as the target development parameter combination. This method can improve the calculation process of the intelligent optimization algorithm through machine learning algorithms, shortening the optimization time required for the development parameters of natural gas hydrate reservoirs.

[0112] Example 3

[0113] Based on the foregoing embodiments, this application further provides a method for determining potential particles. Figure 2 This application provides a schematic diagram of an implementation process for determining potential particles, as shown in the embodiments below. Figure 2 As shown, the method includes:

[0114] Step S301: Determine the sampling points for each development parameter based on the parameter range of each development parameter.

[0115] In this embodiment, sampling points for each development parameter can be obtained by random sampling based on the parameter range of each development parameter. In some embodiments, sampling points for each development parameter can be obtained by interval sampling based on the parameter range of each development parameter. For example, the number of sampling points for each development parameter is K. K is greater than or equal to 5.

[0116] Step S302: Based on the development parameter sampling points, determine the combination of sampling development parameters.

[0117] Continuing with the example above, we can combine the sampling points of the M development parameters to obtain K. M A combination of sampling development parameters.

[0118] Step S303: Determine the objective function value corresponding to the sampling development parameter combination based on the sampling development parameter combination and the training model.

[0119] In this embodiment of the application, K can be M Each combination of sampling development parameters is input into the training model to predict the objective function value of each sampling development parameter.

[0120] Step S304: Sort the objective function values ​​corresponding to the sampling development parameter combinations to obtain sorting information of the objective function values ​​corresponding to the sampling development parameter combinations.

[0121] In this embodiment of the application, K can be... M The objective function values ​​corresponding to each sampled development parameter combination are sorted to obtain K. M The sorting information of the objective function corresponding to each combination of sampling development parameters.

[0122] Step S305: Determine potential particles based at least on the sorting information.

[0123] In this embodiment, the top L sampled development parameter combinations with the highest objective function values ​​can be selected from the sorting, and each sampled development parameter combination is determined as a particle, thus obtaining a potential particle.

[0124] The method for determining development parameters of natural gas hydrate reservoirs provided in this application uses machine learning algorithms to perform data mining on the particle parameters and objective function values ​​in the particle swarm optimization process to obtain a training model. Potential particles are determined based on the training model. The globally optimal objective function value is calculated based on the combination of potential particles and updated development parameters of the particles. Then, the target development parameters are determined based on the globally optimal objective function value, thereby shortening the time required for optimizing the development parameters of natural gas hydrate reservoirs.

[0125] Example 4

[0126] Based on the foregoing embodiments, this application further provides a method for determining potential particles. Figure 3 A schematic diagram illustrating another method for determining potential particles provided in this application embodiment is shown below. Figure 3 As shown, the method includes:

[0127] Step S401: Determine the sampling points for each development parameter based on the parameter range of each development parameter.

[0128] In this embodiment, sampling points for each development parameter can be obtained by random sampling based on the parameter range of each development parameter. In some embodiments, sampling points for each development parameter can be obtained by interval sampling based on the parameter range of each development parameter. For example, the number of sampling points for each development parameter is K. K is greater than or equal to 5.

[0129] Step S402: Based on the development parameter sampling points, determine the combination of sampling development parameters.

[0130] Continuing with the example above, we can combine the sampling points of the M development parameters to obtain K. M A combination of sampling development parameters.

[0131] Step S403: Determine the objective function value corresponding to the sampling development parameter combination based on the sampling development parameter combination and the training model.

[0132] In this embodiment of the application, K can be M Each combination of sampling development parameters is input into the training model to predict the objective function value of each sampling development parameter.

[0133] Step S404: Sort the objective function values ​​corresponding to the sampling development parameter combinations to obtain sorting information of the objective function values ​​corresponding to the sampling development parameter combinations.

[0134] In this embodiment of the application, K can be... M The objective function values ​​corresponding to each sampled development parameter combination are sorted to obtain K. M The sorting information of the objective function corresponding to each combination of sampling development parameters.

[0135] Step S405: Determine the top Z particles based on the sorting information.

[0136] Step S406: Determine the dimensionless distance between two particles in the first Z particles.

[0137] In this embodiment of the application, the dimensionless distance between the two particles is calculated using the following formula:

[0138]

[0139] Among them, L X,Y For dimensionless distance, P X,j For the j-th development parameter of the X-th particle, P Y,j Let be the j-th development parameter of the Y-th particle.

[0140] Step S407: Identify potential particles as particles where the dimensionless distance between two particles in the first Z particles is greater than a preset threshold.

[0141] In this embodiment of the application, the preset threshold for dimensionless distance can be 2%.

[0142] The method for determining development parameters of natural gas hydrate reservoirs provided in this application uses a machine learning algorithm to mine the parameters of each particle and the objective function value in the particle swarm optimization process. The dimensionless distance between each potential particle should be greater than 2%, which avoids the potential particles selected by the machine learning method after data mining being too concentrated, thereby shortening the time required for optimizing the development parameters of natural gas hydrate reservoirs.

[0143] Example 5

[0144] Based on the foregoing embodiments, after step S103 "perform machine learning based on the current combination of development parameters and the corresponding objective function value, and the historical combination of development parameters and the corresponding objective function value to obtain a training model", the method further includes:

[0145] Step S106: Train the training model based on the updated development parameter combination and the corresponding objective function value.

[0146] In this embodiment of the application, it is necessary to train all known objective function values ​​and the development parameter combinations corresponding to the particles with known objective function values, so as to continuously enhance the confidence of the correlation model between the trained development parameter combinations and the objective function during the iteration process.

[0147] Example 6

[0148] This application provides a method for determining development parameters of a natural gas hydrate reservoir. Figure 4 A schematic diagram illustrating the implementation process of another method for determining development parameters of a natural gas hydrate reservoir provided in this application embodiment is shown below. Figure 4 As shown, the above includes:

[0149] Step S601: Determine the objective function model and development parameter information.

[0150] In this embodiment of the application, the development parameter information includes: the number of development parameters and the range corresponding to each development parameter.

[0151] Step S602: Determine the number of particles in the particle swarm algorithm and initialize them randomly.

[0152] In this embodiment, the number of particles can be selected from 10 to 50. In this embodiment, the combination of development parameters for the i-th particle can be determined through random initialization.

[0153] Step S603: Calculate the objective function value of the particles in the particle swarm optimization algorithm.

[0154] In this embodiment, the combination of development parameters corresponding to the particles is input into the objective function model to calculate the objective function value.

[0155] Step S604: Obtain the historical optimal value and global optimal value of the particles in the particle swarm optimization algorithm.

[0156] In this embodiment, the objective function values ​​of the particle and historical particles can be compared to determine the historical optimum and the global optimum (i.e., the maximum objective function value).

[0157] Step S605: Update the particle development parameter combination in the particle swarm algorithm.

[0158] In this embodiment of the application, the update of the development parameter combination can be performed in the following manner, and the preset update method is shown in the following formula:

[0159]

[0160] Where rand1() and rand2() are both random functions in the range of 0 to 1; ω is the inertia weight; c1 and c2 are both constants; This is the updated value of the j-th development parameter for the i-th particle; The updated value of the j-th development parameter for the i-th particle; Let j be the value of the development parameter in the current historical best development parameter combination for the i-th particle; This is the j-th development parameter value in the current globally optimal combination of development parameters.

[0161] Step S606: Perform machine learning based on the particle data to obtain a trained model.

[0162] In this embodiment of the application, when using machine learning methods for data mining, it is necessary to train all known objective function values ​​and the development parameter combinations corresponding to the particles with known objective function values, so as to continuously enhance the confidence of the correlation model between the trained development parameter combinations and the objective function during the iteration process.

[0163] Training the model involves developing the computational relationship between parameter combinations and the objective function value.

[0164] Step S607: Obtain the potential particles for machine learning based on the trained model.

[0165] To avoid the potential particles selected by machine learning methods after data mining being too concentrated, the dimensionless distance between any two potential particles should be greater than 2%. The dimensionless distance between the m-th and n-th particles is calculated as follows:

[0166]

[0167] Where P m,j For the j-th development parameter of the m-th particle, P n,j This is the j-th development parameter of the n-th particle.

[0168] Step S608: Calculate the objective function values ​​of particles and potential particles in the particle swarm optimization algorithm.

[0169] In this embodiment, the development parameters of the particles are updated, and the target function value is calculated using the updated development parameter combination with the target function model. The target function value of the development parameter combination corresponding to the potential particle is calculated by training the model.

[0170] Step S609: Update the historical best value of the particle based on the objective function value of the particle in the particle swarm algorithm.

[0171] In this embodiment, the historical data of particles is updated in real time to the highest quality.

[0172] Step S610: Update the global optimum based on the objective function values ​​of the particles and potential particles in the particle swarm optimization algorithm.

[0173] In this embodiment, the global optimum can be updated by comparing the objective function value of a particle with the objective function value of a potential particle.

[0174] Step S611: Determine whether the convergence condition has been met based on the combination of development parameters corresponding to the global optimal value.

[0175] In this embodiment, if the convergence condition is met, step S612 is executed; if the convergence condition is not met, step S605 is executed. In this embodiment, the convergence condition can be that the combination of development parameters corresponding to the global optimal value has not been updated in 5 to 10 iterations, in which case convergence is considered achieved.

[0176] Step S612: Optimization ends, and the development parameter combination corresponding to the global optimal value is determined as the target development parameter combination.

[0177] Based on the above method for determining development parameters, the timing of pressure reduction to hot water drive, hot water injection temperature, hot water injection rate, and bottom hole flowing pressure of the production well after hot water injection are optimized, with the objective function being hot water injection efficiency. Figure 5 For comparison of the performance of the algorithm of this invention and the traditional particle swarm optimization algorithm, such as Figure 5 As shown in the figure, the method for determining development parameters provided in this embodiment of the application shows that the objective function no longer changes after 56 iterations, with a final energy efficiency of 6.19. In contrast, the conventional particle swarm optimization algorithm requires 118 iterations to converge, with a final energy efficiency of 5.96. The algorithm in this invention requires fewer iterations and has a higher final objective function value, demonstrating significantly better performance than the conventional particle swarm optimization algorithm.

[0178] This application provides a method for determining development parameters of natural gas hydrate reservoirs. The method uses machine learning algorithms to mine data from the particle parameters and objective function values ​​in the particle swarm optimization process, shortening the time required for optimizing development parameters. The dimensionless distance between any two potential particles should be greater than 2%, avoiding an over-concentration of potential particles selected after data mining by machine learning. When using machine learning for data mining, it is necessary to train all known objective function values ​​and the corresponding development parameter combinations for particles with known objective function values, thereby continuously enhancing the confidence of the correlation model between the trained development parameter combinations and the objective function during the iteration process.

[0179] Example 7

[0180] Based on the foregoing embodiments, this application provides a device for determining the development parameters of a natural gas hydrate reservoir. The various modules and units included in the device can be implemented by a processor in a computer device; of course, they can also be implemented by specific logic circuits. In the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA), etc.

[0181] This application provides an apparatus for determining the development parameters of a natural gas hydrate reservoir. Figure 6 A schematic diagram of a device for determining development parameters of a natural gas hydrate reservoir provided in this application embodiment is shown below. Figure 6 As shown, the apparatus 500 for determining the development parameters of a natural gas hydrate reservoir includes:

[0182] The acquisition module 501 is used to acquire the numerical simulation model of the natural gas hydrate reservoir and to determine the numerical simulation model as the objective function model of the particle swarm algorithm.

[0183] The first determining module 502 is used to determine the development parameter information of the particle swarm algorithm, and to determine the development parameter combination of the i-th particle based on the development parameter information.

[0184] The second determining module 503 is used to determine the target function value corresponding to the development parameter combination based on the target function model, and to determine the current development parameter combination of the i-th particle based on the target function value and the historical target function value.

[0185] The first training module 504 is used to perform machine learning to obtain a training model based on the current combination of development parameters and the corresponding objective function value, and the historical combination of development parameters and the corresponding objective function value; and to determine potential particles based on the training model and the development parameter information.

[0186] The third determining module 505 is used to update the development parameter combination of the i-th particle based on the current development parameter information to obtain an updated development parameter combination, and to determine the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle based on the objective function model, and to determine the global optimal objective function value based on the objective function value corresponding to the updated development parameters and the objective function value corresponding to the potential particle.

[0187] The fourth determining module 506 is used to determine the development parameter combination corresponding to the global optimal objective function value as the target development parameter combination for natural gas hydrate reservoirs when the convergence condition is met, based on the development parameter combination corresponding to the global optimal objective function value. Here, i = 1, 2...N, and N is the number of particles.

[0188] In some embodiments, the development parameter information includes: the number of development parameters and the range corresponding to each development parameter. The value of the j-th development parameter of the i-th particle is given by the following formula:

[0189]

[0190] Where j = 1, 2, ..., M, n For the number of iterations, Let M be the value of the j-th development parameter of the i-th particle, and P be the number of development parameters. j,low P is the lower limit value of the j-th development parameter. j,high The upper limit value of the j-th development parameter is given, and rand() is a random function in the range of 0 to 1.

[0191] In some embodiments, the third determining module 505 includes:

[0192] The first determining unit is used to update the development parameter combination of the i-th particle based on the development parameter information and a preset update method to obtain an updated development parameter combination, wherein the preset update method is given by the following formula:

[0193]

[0194] Where rand1() and rand2() are both random functions in the range of 0 to 1; ω is the inertia weight; c1 and c2 are both constants; This is the updated value of the j-th development parameter for the i-th particle; The updated value of the j-th development parameter for the i-th particle; Let j be the value of the development parameter in the current historical best development parameter combination for the i-th particle; This is the j-th development parameter value in the current globally optimal combination of development parameters.

[0195] In some embodiments, the first training module 504 includes:

[0196] The sampling unit is used to determine the sampling points of each development parameter based on the parameter range of each development parameter.

[0197] The second determining unit is used to determine the combination of sampling development parameters based on the sampling points of the development parameters;

[0198] The third determining unit is used to determine the objective function value corresponding to the sampling development parameter combination based on the sampling development parameter combination and the training model;

[0199] The sorting unit is used to sort the objective function values ​​corresponding to the sampling development parameter combinations to obtain the sorting information of the objective function values ​​corresponding to the sampling development parameter combinations;

[0200] The fourth determining unit is used to determine potential particles based at least on the sorting information.

[0201] In some embodiments, the fourth determining unit includes:

[0202] The first determining subunit is used to determine the first Z particles based on the sorting information;

[0203] The second determining subunit is used to determine the dimensionless distance between two particles in the first Z particles;

[0204] The third determining subunit is used to determine potential particles among the first Z particles whose dimensionless distance between two particles is greater than a preset threshold. The dimensionless distance between the two particles is calculated using the following formula:

[0205]

[0206] Among them, L X,Y For dimensionless distance, P X,j For the j-th development parameter of the X-th particle, P Y,j Let be the j-th development parameter of the Y-th particle.

[0207] In some embodiments, the third determining module 505 includes:

[0208] The fifth determining unit is used to determine the first magnitude relationship between the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the historical optimal development parameters;

[0209] The sixth determining unit is used to determine the current optimal objective function value based on the first size relationship;

[0210] The seventh determining unit is used to determine the second magnitude relationship between the current optimal objective function value and the objective function value corresponding to the potential particle;

[0211] The eighth determining unit is used to determine the largest objective function value as the globally optimal objective function value.

[0212] In some embodiments, the apparatus 500 for determining the development parameters of a natural gas hydrate reservoir further includes:

[0213] The second training module trains the training model based on the updated development parameter combination and the corresponding objective function value.

[0214] It should be noted that, in the embodiments of this application, if the above-mentioned method for determining development parameters is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium 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 methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0215] Accordingly, this application provides a storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, it implements the steps in the method for determining the development parameters of natural gas hydrate reservoirs provided in the above embodiments.

[0216] Example 8

[0217] This application provides an electronic device; Figure 7 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application, such as... Figure 7As shown, the electronic device 700 includes: a processor 701, at least one communication bus 702, a user interface 703, at least one external communication interface 704, and a memory 705. The communication bus 702 is configured to enable communication between these components. The user interface 703 may include a display screen, and the external communication interface 704 may include standard wired and wireless interfaces. The processor 701 is configured to execute a program for determining development parameters stored in the memory, to implement the steps in the method for determining development parameters of a natural gas hydrate reservoir provided in the above embodiment.

[0218] The descriptions of the display device and storage medium embodiments above are similar to those of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the computer device and storage medium embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0219] It should be noted that the descriptions of the storage medium and device embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the storage medium and device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0220] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0221] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0222] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0223] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0224] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0225] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0226] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a controller to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0227] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for determining development parameters of a natural gas hydrate reservoir, characterized in that, include: A numerical simulation model of a natural gas hydrate reservoir is obtained, and the numerical simulation model is determined as the objective function model of the particle swarm optimization algorithm. Determine the development parameter information for the particle swarm optimization algorithm, and determine the first [unclear] based on the development parameter information. The combination of development parameters for each particle; Based on the objective function model, the objective function value corresponding to the development parameter combination is determined. Based on the objective function value and the historical objective function value, the first... The current development parameter combination for the nth particle; the current development parameter combination is the nth particle. The current historical optimal combination of development parameters for each particle; A training model is obtained by performing machine learning based on the current combination of development parameters and the corresponding objective function value, as well as the historical combination of development parameters and the corresponding objective function value; potential particles are determined based on the training model and the development parameter information. Update the first based on the development parameter information. The current development parameter combination of each particle is used to obtain an updated development parameter combination. Based on the objective function model, the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle are determined. Based on the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle, the global optimal objective function value is determined. If the convergence condition is met based on the development parameter combination corresponding to the globally optimal objective function value, then the development parameter combination corresponding to the globally optimal objective function value is determined as the target development parameter combination for the natural gas hydrate reservoir, wherein... , The number of particles.

2. The method according to claim 1, characterized in that, The development parameter information includes: the number of development parameters and the range corresponding to each development parameter, wherein the first... The first particle The values ​​for each development parameter are shown in the following formula: ； in, , For the number of iterations, For the first The first particle Each development parameter can take a value. To increase the number of development parameters, For the first The lower limit value of each development parameter, For the first The upper limit of the development parameters, and rand() is a random function in the range of 0 to 1.

3. The method according to claim 2, characterized in that, The update based on the development parameter information The development parameter combination for each particle is used to obtain an updated development parameter combination, including: Update the first [item] based on the aforementioned development parameter information and preset update method. The development parameter combination of each particle is used to obtain the updated development parameter combination, wherein the preset update method is given by the following formula: Both rand1() and rand2() are random functions in the range of 0 to 1; For inertial weights, ; and All are constants; For the first The first particle Updated values ​​for each development parameter; For the first The first particle The updated values ​​of the development parameters; For the first The current historical optimal development parameter combination of the nth particle One development parameter value; The first in the current globally optimal combination of development parameters Each development parameter value.

4. The method according to claim 2, characterized in that, Determining potential particles based on the training model and the development parameter information includes: The sampling points for each development parameter are determined based on the parameter range of each development parameter. Based on the aforementioned development parameter sampling points, determine the combination of sampling development parameters; The objective function value corresponding to the sampling development parameter combination is determined based on the sampling development parameter combination and the training model; The objective function values ​​corresponding to the sampling development parameter combinations are sorted to obtain the sorting information of the objective function values ​​corresponding to the sampling development parameter combinations; Potential particles are determined at least based on the sorting information.

5. The method according to claim 4, characterized in that, The determination of potential particles based at least on the sorting includes: The top Z particles are determined based on the sorting information; Determine the dimensionless distance between any two particles in the first Z particles; Among the first Z particles, particles whose dimensionless distance between two particles is greater than a preset threshold are identified as potential particles. The dimensionless distance between the two particles is calculated using the following formula: in, The distance is dimensionless. For the Xth particle One development parameter, For the Y-th particle One development parameter.

6. The method according to claim 2, characterized in that, The process of determining the globally optimal objective function value based on the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle includes: Determine the primary relationship between the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the historical optimal development parameters; The current optimal objective function value is determined based on the first size relationship; Determine a second magnitude relationship between the current optimal objective function value and the objective function value corresponding to the potential particle; The largest objective function value is determined as the globally optimal objective function value.

7. The method according to claim 1, characterized in that, The method further includes: The training model is trained based on the updated combination of development parameters and the corresponding objective function value.

8. An apparatus for determining development parameters of a natural gas hydrate reservoir, characterized in that, include: The acquisition module is used to acquire the numerical simulation model of the natural gas hydrate reservoir and determine the numerical simulation model as the objective function model of the particle swarm algorithm. The first determining module is used to determine the development parameter information of the particle swarm optimization algorithm, and to determine the second module based on the development parameter information. The combination of development parameters for each particle; The second determining module is used to determine the objective function value corresponding to the development parameter combination based on the objective function model, and to determine the first objective function value based on the objective function value and historical objective function values. The current development parameter combination for the nth particle; the current development parameter combination is the nth particle. The current historical optimal combination of development parameters for each particle; The first training module is used to perform machine learning to obtain a training model based on the current combination of development parameters and the corresponding objective function value, and the historical combination of development parameters and the corresponding objective function value; and to determine potential particles based on the training model and the development parameter information. The third determining module is used to update the first [module] based on the development parameter information. The current development parameter combination of each particle is used to obtain an updated development parameter combination. Based on the objective function model, the objective function value corresponding to the updated development parameter combination and the objective function value corresponding to the potential particle are determined. The global optimal objective function value is determined based on the objective function value corresponding to the updated development parameters and the objective function value corresponding to the potential particle. The fourth determining module is used to determine the development parameter combination corresponding to the globally optimal objective function value as the target development parameter combination for the natural gas hydrate reservoir, provided that the convergence condition has been met. , The number of particles.

9. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, performs the method for determining the development parameters of a natural gas hydrate reservoir as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The computer program stored in the storage medium can be executed by one or more processors and can be used to implement the method for determining the development parameters of the natural gas hydrate reservoir as described in any one of claims 1 to 7.

Images