Oil and gas reservoir simulation method and device based on large language model, medium and equipment

Abstract

The invention discloses an oil and gas reservoir simulation method and device based on a large language model, a medium and equipment, and belongs to the technical field of oil and gas field development. The method comprises the following steps: S1, acquiring basic information of an oil and gas reservoir of a target block and an agent model of the target block, and storing the basic information and the agent model into a database; s2, an intelligent agent constructed based on a large language model is used for judging the purpose of a user according to the user input information, and related parameters are extracted from the user input information; s3, calculating the attribute information which needs to be calculated and is contained in the information input by the user through the intelligent agent; and S4, updating the basic information of the oil and gas reservoir by using the calculated attribute information, and performing oil and gas reservoir simulation by using the proxy model according to the related parameters and the updated basic information of the oil and gas reservoir to obtain an oil and gas reservoir simulation result. According to the method, the problem that a general large language model cannot solve the simulation problem in the professional field of oil and gas reservoirs is solved, rapid and accurate simulation of the oil and gas reservoirs is realized, and the difficulty of oil and gas reservoir simulation is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas field development, and in particular to an oil and gas reservoir simulation method, device, medium and equipment based on a large language model. Background Art

[0002] With the rapid development of deep learning technology and hardware computing power, the parameter scale and training data volume of deep neural network models are constantly expanding. In the field of natural language processing, researchers have found that by expanding the number of model parameters and increasing training data, the performance and generalization of language models can be enhanced, and even the phenomenon of "emergence" can occur, which greatly improves the performance on some tasks. These language models with expanded parameter volume and data volume are called large language models (LLM). In order to further promote this concept to other fields, such as computer vision, the concepts of foundation model and large model are proposed. The foundation model refers to a model that is trained on a large amount of data and can adapt to various downstream tasks (generally obtained by pre-training through self-supervised learning algorithms). A large model refers to a model with a large number of parameters, trained on massive data, and has excellent data and task generalization. Therefore, to a certain extent, the concepts of large model and foundation model are equivalent.

[0003] As an emerging trend in cross-disciplinary research, the research and application of big models in the oil and gas industry are gradually attracting the attention of researchers and the industry. Big model technology based on artificial intelligence can efficiently analyze massive data in the oil and gas field, support intelligent decision-making, optimize production processes, and improve the overall efficiency of the industry. In the oil and gas industry, big language models are being applied to intelligent assistant development, question-answering systems, data analysis and visualization, reservoir modeling and other fields with their efficient natural language understanding and generation capabilities.

[0004] Although the application of large model technology in the oil and gas industry has shown broad prospects, it is currently mostly focused on text processing issues. The work that really bothers developers in the oil industry should be concentrated on the dynamic simulation and development of oil and gas reservoirs in the process of oil and natural gas drilling and production. At present, research in this area still faces some technical and industry-specific challenges, making it difficult to promote the application of large models in the dynamic simulation and development of oil and gas reservoirs. Summary of the invention

[0005] In order to solve the defects of the prior art, the present invention provides an oil and gas reservoir simulation method, device, medium and equipment based on a large language model, which solves the problem that the general large language model cannot solve the simulation problem in the professional field of oil and gas reservoirs, realizes rapid and accurate simulation of oil and gas reservoirs, and greatly reduces the difficulty of oil and gas reservoir simulation.

[0006] The present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for simulating oil and gas reservoirs based on a large language model, the method comprising:

[0008] S1: Obtain the basic information of the oil and gas reservoirs in the target block and the proxy model of the target block, and store them in the database;

[0009] S2: using an agent built based on a large language model to determine the user's purpose based on the user's input information, and extracting relevant parameters from the user's input information;

[0010] S3: calculating the attribute information to be calculated contained in the user input information by the agent;

[0011] Wherein, the attribute information belongs to one or more of the basic information of the oil and gas reservoir;

[0012] S4: using the calculated attribute information to update the basic information of the oil and gas reservoir, and using the proxy model to perform oil and gas reservoir simulation according to the relevant parameters and the updated basic information of the oil and gas reservoir to obtain an oil and gas reservoir simulation result.

[0013] Further, the S1 includes:

[0014] S11: Modeling the production status of the oil and gas reservoirs contained in the target block to obtain basic information of the oil and gas reservoirs;

[0015] S12: storing basic information of oil and gas reservoirs in a database in a key-value format;

[0016] S13: constructing a corresponding proxy model for the target block and storing it in a database;

[0017] S14: Constructing the oil and gas reservoir simulation thinking chain as the reasoning process of the large language model.

[0018] Furthermore, the agent includes a user purpose judgment agent and an information extraction agent, and S2 includes:

[0019] S21: using the user purpose judgment agent to judge the user purpose according to the user input information, wherein the user purpose includes obtaining data and oil and gas reservoir simulation;

[0020] S22: If the user's purpose is to obtain data, then directly read the corresponding data from the database and end; if the user's purpose is to simulate oil and gas reservoirs, then execute the next step;

[0021] S23: Using the information extraction agent to extract relevant parameters contained in the user input information.

[0022] Furthermore, the agent also includes a problem solving agent and a calculation agent, and S3 includes:

[0023] S31: using the problem decomposition agent to decompose the user input information to obtain the premise and the final purpose;

[0024] S32: Determine whether the prerequisite is empty, if so, execute S4, otherwise call the computing agent to calculate the attribute information that needs to be calculated according to the input information.

[0025] Further, the S4 includes:

[0026] S41: Reading all basic information of the target block from the database;

[0027] S42: updating the basic information of the oil and gas reservoir according to the calculated attribute information;

[0028] S43: Based on the updated basic information of the oil and gas reservoir and the relevant parameters extracted from the user input information, the constructed oil and gas reservoir simulation thinking chain is called, and the updated basic information of the oil and gas reservoir is passed into each step of the decomposition function to obtain the oil and gas reservoir simulation result.

[0029] In a second aspect, the present invention provides an oil and gas reservoir simulation device based on a large language model, the device comprising:

[0030] The data acquisition module is used to obtain the basic information of the oil and gas reservoirs in the target block and the proxy model of the target block, and store them in the database;

[0031] A purpose judgment and parameter extraction module, used to judge the user's purpose according to the user input information using an intelligent agent built based on a large language model, and extract relevant parameters from the user input information;

[0032] An attribute calculation module, used to calculate the attribute information to be calculated contained in the user input information through the agent;

[0033] Wherein, the attribute information belongs to one or more of the basic information of the oil and gas reservoir;

[0034] The oil and gas reservoir simulation module is used to update the basic information of the oil and gas reservoir using the calculated attribute information, and to perform oil and gas reservoir simulation using the proxy model according to the relevant parameters and the updated basic information of the oil and gas reservoir to obtain the oil and gas reservoir simulation results.

[0035] Furthermore, the data acquisition module includes:

[0036] A data acquisition unit, used to model the production status of the oil and gas reservoirs contained in the target block to obtain basic information of the oil and gas reservoirs;

[0037] A storage unit is used to store basic information of oil and gas reservoirs into a database in a key-value format;

[0038] A model building unit, used to build a corresponding proxy model for the target block and store it in a database;

[0039] The thought chain construction unit is used to construct the oil and gas reservoir simulation thought chain as the reasoning process of the large language model.

[0040] Furthermore, the agent includes a user purpose judgment agent and an information extraction agent, and the purpose judgment and parameter extraction module includes:

[0041] A user purpose judgment unit, used to use the user purpose judgment agent to judge the user purpose according to the user input information, wherein the user purpose includes obtaining data and oil and gas reservoir simulation;

[0042] An execution judgment unit, for directly reading corresponding data from the database and ending the process if the user's purpose is to obtain data; and for executing a parameter extraction unit if the user's purpose is to simulate oil and gas reservoirs;

[0043] A parameter extraction unit is used to use the information extraction agent to extract relevant parameters contained in the user input information.

[0044] Furthermore, the agent also includes a problem solving agent and a calculation agent, and the attribute calculation module includes:

[0045] A problem splitting unit, used to use the problem splitting agent to split the user input information to obtain the premise and the final purpose;

[0046] The prerequisite judgment unit is used to judge whether the prerequisite is empty. If so, the oil and gas reservoir simulation module is executed; otherwise, the computing agent is called to calculate the attribute information that needs to be calculated according to the input information.

[0047] Furthermore, the oil and gas reservoir simulation module includes:

[0048] A data reading unit, used for reading all basic information of the target block from the database;

[0049] A data updating unit, used for updating the basic information of the oil and gas reservoir according to the calculated attribute information;

[0050] The oil and gas reservoir simulation unit is used to call the constructed oil and gas reservoir simulation thinking chain based on the updated basic information of the oil and gas reservoir and the relevant parameters extracted from the user input information, pass the updated basic information of the oil and gas reservoir into each step of the decomposition function, and obtain the oil and gas reservoir simulation results.

[0051] In a third aspect, the present invention provides a computer-readable storage medium for oil and gas reservoir simulation based on a large language model, comprising a memory for storing processor-executable instructions, which, when executed by the processor, implement the steps of the oil and gas reservoir simulation method based on a large language model described in the first aspect.

[0052] In a fourth aspect, the present invention provides a device for simulating oil and gas reservoirs based on a large language model, characterized in that it comprises at least one processor and a memory storing computer executable instructions, and when the processor executes the instructions, it implements the steps of the oil and gas reservoir simulation method based on a large language model described in the first aspect.

[0053] The present invention has the following beneficial effects:

[0054] The present invention discloses a rapid simulation scheme for oil and gas reservoirs based on a large language model. Based on the large language model, combined with a thinking chain prompt from less to more and a professional oil and gas reservoir proxy model, while ensuring the simulation accuracy of the oil and gas reservoir, the production dynamics of the oil and gas reservoir can be quickly simulated by text and voice. The present invention is an application of a combination of a large language model and a special oil and gas reservoir simulation software in the field of oil and gas development, which can solve the simulation problem in the professional field of oil and gas reservoirs that cannot be solved by a general large language model, realizes rapid and accurate simulation of oil and gas reservoirs, and can also reduce the high requirements of professional oil and gas reservoir simulation on personnel, greatly reduce the difficulty of oil and gas reservoir simulation, give full play to the huge innovative application of the large language model in the vertical field of oil and gas, and has important significance for the intelligent construction of oil and gas reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flow chart of the oil and gas reservoir simulation method based on the large language model of the present invention;

[0056] Figure 2 An example diagram of the results of oil and gas reservoir simulation

[0057] Figure 3 It is a schematic diagram of the oil and gas reservoir simulation device based on the large language model of the present invention. DETAILED DESCRIPTION

[0058] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings and specific embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present invention.

[0059] The embodiment of the present invention provides an oil and gas reservoir simulation method based on a large language model. This embodiment is based on a specific oil and gas reservoir and uses a combination of a large language model and an agent model to simulate its results. In order to better illustrate the working principle of the present invention, this embodiment is based on a question, such as "the production and pressure changes in each month in the next year when the well control in the xx oil reservoir is expanded by 2 times" to illustrate the specific embodiment of the present invention.

[0060] like Figure 1 As shown, the method includes:

[0061] S1: Obtain the basic information of the oil and gas reservoirs in the target block and the proxy model of the target block, and store them in the database.

[0062] This step collects basic information about the oil and gas reservoirs and stores it in a database (such as an ES database) based on the production status of the target block and the oil and gas reservoirs it contains. This step is usually done before the user uses it and is completed automatically. In one example, this step includes the following specific contents:

[0063] S11: Modeling is performed based on the production status of the oil and gas reservoirs contained in the target block to obtain basic information of the oil and gas reservoirs.

[0064] These basic information include basic geology, well control, perforation, phase permeability curve and other related information.

[0065] S12: The basic information of the oil and gas reservoir is stored in the database in the form of key-value.

[0066] S13: Construct a corresponding proxy model for the target block and store it in the database.

[0067] S14: Constructing the oil and gas reservoir simulation thinking chain as the reasoning process of the large language model.

[0068] S2: Use an intelligent agent built based on a large language model to determine the user's purpose based on the user's input information and extract relevant parameters from the user's input information.

[0069] This step is used to build a related intelligent agent based on the large language model to determine the user's input intention and extract the relevant parameters passed in by the user. The related intelligent agents used include the user purpose judgment agent and the information extraction agent. For example, the user input information is "the production and pressure changes in each month in the next year when the well control in the xx oil reservoir is expanded by 2 times", and the user input information can be text input or language input. The specific implementation process of this step includes:

[0070] S21: Using the user purpose judgment agent to judge the user purpose according to the user input information, the user purpose includes obtaining data and oil and gas reservoir simulation.

[0071] The user purpose judgment intelligence is built on the basis of the llama3.1-70b model and is used to judge the user's purpose. The results returned by the user purpose judgment intelligence are only "get data (get_data)" or "oil and gas reservoir simulation (simulation)". The user input information in this case is "the production and pressure changes in each month in the next year when the well control in the xx oil reservoir is expanded by 2 times". Therefore, the result finally returned by the user purpose judgment intelligence is: {"user purpose (question_object)": "oil and gas reservoir simulation (simulation)"}.

[0072] S22: If the user's purpose is to obtain data, the corresponding data is directly read from the database and the process ends; if the user's purpose is to simulate oil and gas reservoirs, the next step is executed.

[0073] This step determines the result returned by the agent according to the user's purpose and performs different operations. If "get_data" is returned, it means that oil and gas reservoir simulation is not required, and the corresponding data is directly read from the database to get the final result and return it. If "simulation" is returned, it means that oil and gas reservoir numerical simulation is required, and enter step S23. Since oil and gas reservoir simulation is required in this case, it will enter step S323.

[0074] S33: Use the information extraction agent to extract relevant parameters contained in the user input information.

[0075] The information extraction agent mainly constructs corresponding prompt statements through prompt engineering, requiring the extraction of attribute fields, well names, reservoir names, and the time required for simulation from the user input information. In this specific case, the relevant parameters returned by the information extraction agent are: {" name":"xx reservoir"," start_time":"January 1, 2025"," end_time":"January 1, 2026","property":["Oil production","pressure"]," well_name":["None"];

[0076] S3: Calculate the attribute information that needs to be calculated contained in the user input information through the intelligent agent.

[0077] The intelligent agents used in this step include problem decomposition agents and calculation agents, which mainly use prompt word engineering to determine whether the user's question involves the calculation and calculation method of some attribute information, where the attribute information belongs to one or more of the basic information of the aforementioned oil and gas reservoirs.

[0078] As an example, this step includes

[0079] S31: Use the problem decomposition agent to decompose the user input information into the premise and the final goal.

[0080] The premise contains the process that needs to be inferred. If there is no premise, it is empty or None. The final objective indicates the attribute information that the user ultimately wants to obtain. In this case, the premise returned is {'premise': "The control conditions of the well are expanded by 2 times"}, and the final objective returned is {"objective": "The production and pressure changes in each month in the next year"}.

[0081] S32: Determine whether the prerequisite is empty, if so, execute S4, otherwise call the computing agent to calculate the attribute information that needs to be calculated according to the input information.

[0082] The computational agent determines the attribute information and calculation method that the user needs to calculate through the prompt word engineering judgment, and then constructs the python code to calculate it through the code. This case contains prerequisites, so the computational agent needs to be called for reasoning calculations. Specifically, when implementing, first extract the attribute information and calculation method that need to be calculated. In this problem, the attribute information we need to calculate is "the control conditions of the well", and the calculation method is "expand by 2 times"; then extract the control conditions of all wells from the database; finally, input "the control conditions of the well" and "expand by 2 times" to the computational agent at the same time to obtain the calculated control conditions of the well.

[0083] S4: using the calculated attribute information to update the basic information of the oil and gas reservoir, and using the proxy model to perform oil and gas reservoir simulation according to the relevant parameters and the updated basic information of the oil and gas reservoir to obtain the oil and gas reservoir simulation results.

[0084] This step is used to update parameters and call the proxy model or numerical simulator to quickly obtain the final result. An example includes:

[0085] S41: Read all basic information of the target block from the database.

[0086] In this example, it is necessary to extract information such as the gas saturation field, pressure field, and well control conditions on January 1, 2025 from the database.

[0087] S42: Update the basic information of the oil and gas reservoir according to the calculated attribute information.

[0088] This step is used to update some parameters related to reservoir simulation. In this case, the control conditions of the well are doubled to replace the original control conditions of the well.

[0089] S43: Based on the updated basic information of the oil and gas reservoir and the relevant parameters extracted from the user input information, the constructed oil and gas reservoir simulation thinking chain is called, and the updated basic information of the oil and gas reservoir is passed into each step of the decomposition function to obtain the oil and gas reservoir simulation result.

[0090] In this specific example, the proxy model of the xx reservoir is directly selected, and then the model is loaded. The gas saturation field, pressure field and control conditions of the updated well on January 1, 2025 are used as input, and the interface of the proxy model is called to realize the prediction of dynamic field and production. And according to the user's purpose "oil production and pressure" changes, the simulated oil production and pressure information are automatically returned. The final reservoir simulation results are as follows Figure 2 shown.

[0091] The present invention discloses a rapid simulation scheme for oil and gas reservoirs based on a large language model. Based on the large language model, combined with a thinking chain prompt from less to more and a professional oil and gas reservoir proxy model, while ensuring the simulation accuracy of the oil and gas reservoir, the production dynamics of the oil and gas reservoir can be quickly simulated by text and voice. The present invention is an application of a combination of a large language model and a special oil and gas reservoir simulation software in the field of oil and gas development, which can solve the simulation problem in the professional field of oil and gas reservoirs that cannot be solved by a general large language model, realizes rapid and accurate simulation of oil and gas reservoirs, and can also reduce the high requirements of professional oil and gas reservoir simulation on personnel, greatly reduce the difficulty of oil and gas reservoir simulation, give full play to the huge innovative application of the large language model in the vertical field of oil and gas, and has important significance for the intelligent construction of oil and gas reservoirs.

[0092] The embodiment of the present invention also provides an oil and gas reservoir simulation device based on a large language model, such as Figure 3 As shown, the device comprises:

[0093] The data acquisition module 1 is used to acquire the basic information of the oil and gas reservoir of the target block and the proxy model of the target block, and store them in the database.

[0094] The purpose judgment and parameter extraction module 2 is used to use an intelligent agent built based on a large language model to judge the user's purpose according to the user's input information, and extract relevant parameters from the user's input information.

[0095] The attribute calculation module 3 is used to calculate the attribute information required to be calculated contained in the user input information through the intelligent agent.

[0096] Among them, the attribute information belongs to one or more basic information of the oil and gas reservoir.

[0097] The oil and gas reservoir simulation module 4 is used to update the basic information of the oil and gas reservoir using the calculated attribute information, and to perform oil and gas reservoir simulation using the proxy model according to the relevant parameters and the updated basic information of the oil and gas reservoir to obtain the oil and gas reservoir simulation results.

[0098] As an example, the data acquisition module includes:

[0099] The data acquisition unit is used to build a model based on the production status of the oil and gas reservoirs contained in the target block to obtain basic information of the oil and gas reservoirs.

[0100] The storage unit is used to store the basic information of oil and gas reservoirs into the database in the form of key-value.

[0101] The model building unit is used to build a corresponding proxy model for the target block and store it in the database.

[0102] The thought chain construction unit is used to construct the oil and gas reservoir simulation thought chain as the reasoning process of the large language model.

[0103] In the present invention, the aforementioned agent may include a user purpose judgment agent and an information extraction agent. Based on this, the purpose judgment and parameter extraction module of the present invention includes:

[0104] The user purpose judgment unit is used to use the user purpose judgment intelligent agent to judge the user purpose according to the user input information, and the user purpose includes obtaining data and oil and gas reservoir simulation.

[0105] The execution judgment unit is used to directly read the corresponding data from the database and end if the user's purpose is to obtain data; if the user's purpose is to simulate the oil and gas reservoir, the parameter extraction unit is executed.

[0106] The parameter extraction unit is used to extract relevant parameters contained in the user input information using an information extraction agent.

[0107] Furthermore, the aforementioned agent may also include a problem solving agent and a computing agent, and the attribute computing module at this time includes:

[0108] The problem splitting unit is used to use the problem splitting agent to split the user input information into the premise and the final purpose.

[0109] The prerequisite judgment unit is used to judge whether the prerequisite is empty. If so, the oil and gas reservoir simulation module is executed. Otherwise, the computing agent is called to calculate the attribute information that needs to be calculated according to the input information.

[0110] As a specific implementation method, the aforementioned oil and gas reservoir simulation module includes:

[0111] The data reading unit is used to read all basic information of the target block from the database.

[0112] The data updating unit is used to update the basic information of the oil and gas reservoir according to the calculated attribute information.

[0113] The oil and gas reservoir simulation unit is used to call the constructed oil and gas reservoir simulation thinking chain based on the updated basic information of the oil and gas reservoir and the relevant parameters extracted from the user input information, pass the updated basic information of the oil and gas reservoir into each step of the decomposition function, and obtain the oil and gas reservoir simulation results.

[0114] The device provided in the above embodiment, its implementation principle and the technical effect produced correspond to the embodiment of the above method one by one. For the sake of brief description, the part not mentioned in the embodiment of the device can refer to the corresponding content in the embodiment of the above method. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the modules and units described in the device can refer to the corresponding process in the embodiment of the above method, and will not be repeated here.

[0115] The oil and gas reservoir simulation method based on a large language model described in the above embodiment provided by the present invention can implement business logic through a computer program and record it on a storage medium, and the storage medium can be read and executed by a computer to achieve the effect of the scheme described in the method embodiment of this specification. Therefore, the embodiment of the present invention also provides a computer-readable storage medium for oil and gas reservoir simulation based on a large language model, including a memory for storing processor-executable instructions, and when the instructions are executed by the processor, the steps of the oil and gas reservoir simulation method based on a large language model including the above embodiment are implemented.

[0116] The storage medium may include a physical device for storing information, which is usually a medium that digitizes the information and then stores it in an electrical, magnetic or optical manner. The storage medium may include: a device that stores information in an electrical energy manner, such as various memories, such as RAM, ROM, etc.; a device that stores information in a magnetic energy manner, such as a hard disk, a floppy disk, a magnetic tape, a magnetic core memory, a magnetic bubble memory, a USB flash drive; a device that stores information in an optical manner, such as a CD or a DVD. Of course, there are other readable storage media, such as quantum memory, graphene memory, etc.

[0117] The storage medium described above may also include other implementation methods according to the description of the method embodiment. The implementation principle and technical effects produced by this embodiment are the same as those of the aforementioned method embodiment. For details, please refer to the description of the relevant method embodiment, and no further description will be given here.

[0118] The embodiment of the present invention also provides a device for oil and gas reservoir simulation based on a large language model, and the device may be a separate computer, or may include an actual operating device using one or more of the methods or one or more of the embodiments of this specification. The device for oil and gas reservoir simulation based on a large language model may include at least one processor and a memory storing computer executable instructions, and when the processor executes the instructions, any one or more steps of the oil and gas reservoir simulation method based on a large language model are implemented.

[0119] The device described above may also include other implementation methods according to the description of the method embodiment. The implementation principle and technical effects produced by this embodiment are the same as those of the aforementioned method embodiment. For details, please refer to the description of the relevant method embodiment, and no further description will be given here.

[0120] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. They should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for simulating oil and gas reservoirs based on a large language model, characterized in that: The method comprises: S1: Obtain the basic information of the oil and gas reservoirs in the target block and the proxy model of the target block, and store them in the database; S2: using an agent built based on a large language model to determine the user's purpose based on the user's input information, and extracting relevant parameters from the user's input information; S3: calculating the attribute information to be calculated contained in the user input information by the agent; Wherein, the attribute information belongs to one or more of the basic information of the oil and gas reservoir; S4: using the calculated attribute information to update the basic information of the oil and gas reservoir, and using the proxy model to perform oil and gas reservoir simulation according to the relevant parameters and the updated basic information of the oil and gas reservoir to obtain an oil and gas reservoir simulation result.

2. The oil and gas reservoir simulation method based on a large language model according to claim 1, characterized in that: The S1 includes: S11: Modeling the production status of the oil and gas reservoirs contained in the target block to obtain basic information of the oil and gas reservoirs; S12: storing basic information of oil and gas reservoirs in a database in a key-value format; S13: constructing a corresponding proxy model for the target block and storing it in a database; S14: Constructing the oil and gas reservoir simulation thinking chain as the reasoning process of the large language model.

3. The oil and gas reservoir simulation method based on a large language model according to claim 2, characterized in that: The agent includes a user purpose judgment agent and an information extraction agent, and S2 includes: S21: using the user purpose judgment agent to judge the user purpose according to the user input information, wherein the user purpose includes obtaining data and oil and gas reservoir simulation; S22: If the user's purpose is to obtain data, then directly read the corresponding data from the database and end; if the user's purpose is to simulate oil and gas reservoirs, then execute the next step; S23: Using the information extraction agent to extract relevant parameters contained in the user input information.

4. The oil and gas reservoir simulation method based on a large language model according to claim 3, characterized in that: The agent also includes a problem-solving agent and a computing agent, and S3 includes: S31: using the problem decomposition agent to decompose the user input information to obtain the premise and the final purpose; S32: Determine whether the prerequisite is empty, if so, execute S4, otherwise call the computing agent to calculate the attribute information that needs to be calculated according to the input information.

5. The oil and gas reservoir simulation method based on a large language model according to claim 4, characterized in that: The S4 includes: S41: Reading all basic information of the target block from the database; S42: updating the basic information of the oil and gas reservoir according to the calculated attribute information; S43: Based on the updated basic information of the oil and gas reservoir and the relevant parameters extracted from the user input information, the constructed oil and gas reservoir simulation thinking chain is called, and the updated basic information of the oil and gas reservoir is passed into each step of the decomposition function to obtain the oil and gas reservoir simulation result.

6. An oil and gas reservoir simulation device based on a large language model, characterized in that: The device comprises: The data acquisition module is used to obtain the basic information of the oil and gas reservoirs in the target block and the proxy model of the target block, and store them in the database; A purpose judgment and parameter extraction module, used to judge the user's purpose according to the user input information using an intelligent agent built based on a large language model, and extract relevant parameters from the user input information; An attribute calculation module, used to calculate the attribute information to be calculated contained in the user input information through the agent; Wherein, the attribute information belongs to one or more of the basic information of the oil and gas reservoir; The oil and gas reservoir simulation module is used to update the basic information of the oil and gas reservoir using the calculated attribute information, and to perform oil and gas reservoir simulation using the proxy model according to the relevant parameters and the updated basic information of the oil and gas reservoir to obtain oil and gas reservoir simulation results.

7. The oil and gas reservoir simulation device based on a large language model according to claim 6, characterized in that: The data acquisition module comprises: A data acquisition unit, used to model the production status of the oil and gas reservoirs contained in the target block to obtain basic information of the oil and gas reservoirs; A storage unit is used to store basic information of oil and gas reservoirs into a database in a key-value format; A model building unit, used to build a corresponding proxy model for the target block and store it in a database; The thought chain construction unit is used to construct the oil and gas reservoir simulation thought chain as the reasoning process of the large language model.

8. The oil and gas reservoir simulation device based on a large language model according to claim 7, characterized in that: The agent includes a user purpose judgment agent and an information extraction agent, and the purpose judgment and parameter extraction module includes: A user purpose judgment unit, used to use the user purpose judgment agent to judge the user purpose according to the user input information, wherein the user purpose includes obtaining data and oil and gas reservoir simulation; An execution judgment unit, for directly reading corresponding data from the database and ending the process if the user's purpose is to obtain data; and for executing a parameter extraction unit if the user's purpose is to simulate oil and gas reservoirs; A parameter extraction unit is used to use the information extraction agent to extract relevant parameters contained in the user input information.

9. A computer-readable storage medium for oil and gas reservoir simulation based on a large language model, characterized in that: It comprises a memory for storing processor executable instructions, and when the instructions are executed by the processor, the steps of the oil and gas reservoir simulation method based on the large language model according to any one of claims 1-5 are implemented.

10. A device for oil and gas reservoir simulation based on a large language model, characterized in that: The method comprises at least one processor and a memory storing computer executable instructions, wherein when the processor executes the instructions, the steps of the oil and gas reservoir simulation method based on a large language model described in any one of claims 1 to 5 are implemented.

Images