Method and device for identifying key scene in production process of deep coal-bed gas well
By obtaining the production data of deep coalbed methane wells and using key scene identification models, we automatically identify key scenarios in the production process, and solving the problem of relying on manual monitoring of the production status of deep coalbed methane wells, achieving efficient production management and collaborative production.
Patent Information
- Application Number
- CN202311845236.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-12-28
- Publication Date
- 2025-07-01
AI Technical Summary
During the production process of deep coalbed methane gas wells, changes in gas well production status rely on manual monitoring and judgment, resulting in low production management efficiency and difficulty in achieving efficient collaborative production.
By obtaining the production data of deep coalbed methane wells, key feature data are generated, and using pre-trained key scenario identification models, key scenarios in the production process are automatically identified, including production construction, production, gas production reduction, stable gas production, gas production increase and rapid attenuation.
It realizes automatic judgment of the production status of deep coalbed methane gas wells, improves production management efficiency, helps production personnel to adjust the production process in a timely manner, solves the problem of manual monitoring dependence, and supports the second-level adjustment and efficient management of coordinated production of multiple wells.
Smart Images
Figure CN120234710A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of deep coalbed methane development, and specifically relates to a method and device for identifying key scenarios in the production process of deep coalbed methane wells. Background Art
[0002] Currently, in some regions, the supply of oil and gas resources still falls short of demand, with a high degree of external dependence. With the gradual depletion of shallow coal and mineral resources, the utilization of deep coal resources has become an inevitable trend. During the development and production process of deep coalbed methane, due to the more complex deep environment and fluid gas quality, new requirements are put forward for the adaptability of surface gathering and transportation modes and production processes.
[0003] As a new energy growth point, deep coalbed methane has a promising development prospect. Under the background of high-quality development and the digital transformation of the energy field, problems such as large pressure differences in the surface gathering and transportation process of deep coalbed methane, energy waste in high-pressure wells, and development efficiency urgently need to be addressed through scientific and technological research to develop new methods and processes, thereby further improving the production capacity construction of deep coalbed methane, achieving cost reduction and efficiency increase, and efficient development. Summary of the Invention
[0004] In view of the problems in the prior art, the embodiments of this application provide a method and device for identifying key scenarios in the production process of deep coalbed methane wells, which can at least partially solve the problems existing in the prior art.
[0005] On the one hand, this application proposes a method for identifying key scenarios in the production process of deep coalbed methane wells, including:
[0006] Obtaining the actual production data during the production process of the first deep coalbed methane well;
[0007] Generating key feature data based on the actual production data;
[0008] Inputting the key feature data into a pre-trained key scenario recognition model to obtain the key scenarios identified by the key scenario recognition model.
[0009] In some embodiments, the actual production data includes: the bottom hole pressure, casing pressure, gas production volume, and water production volume of the first deep coalbed methane well.
[0010] In some embodiments, the generating key feature data based on the actual production data includes:
[0011] Generating the pressure gradient of the difference between the bottom hole pressure and the casing pressure of the first deep coalbed methane well with respect to the well depth based on the bottom hole pressure, casing pressure of the first deep coalbed methane well, and the well depth of the first deep coalbed methane well;
[0012] Generate the gas-phase superficial velocity and liquid-phase superficial velocity of the first deep coalbed methane well based on the gas production, water production, and wellbore diameter of the first deep coalbed methane well.
[0013] In some embodiments, the method further includes:
[0014] Obtain key scenario learning samples, where the key scenario learning samples are generated based on the actual production data of at least one second deep coalbed methane well in each historical period;
[0015] Train a preset machine learning model according to the key scenario learning samples to obtain a trained key scenario recognition model.
[0016] In some embodiments, the method further includes:
[0017] Obtain the actual production data of the second deep coalbed methane well in each historical period;
[0018] Classify the actual production data of each historical period by using an unsupervised classification algorithm to obtain the scenario labels corresponding to the actual production data of each historical period, where the scenario labels include well completion, production increase, gas production decrease, stable gas production, gas production increase, and rapid decay;
[0019] Generate learning samples of the key scenario recognition model according to the actual production data and the corresponding scenario labels of each historical period.
[0020] On the other hand, the present application proposes a key scenario recognition device for the production process of a deep coalbed methane well, including:
[0021] A first acquisition module for acquiring the actual production data in the production process of the first deep coalbed methane well;
[0022] A generation module for generating key feature data according to the actual production data;
[0023] An identification module for inputting the key feature data into a pre-trained key scenario recognition model to obtain the key scenario recognized by the key scenario recognition model.
[0024] In some embodiments, the actual production data includes: the bottom hole pressure, casing pressure, gas production, and water production of the first deep coalbed methane well.
[0025] In some embodiments, the generation module is specifically configured to:
[0026] Generate the pressure gradient of the difference between the bottom hole pressure and the casing pressure of the first deep coalbed methane well with respect to the well depth according to the bottom hole pressure, casing pressure, and well depth of the first deep coalbed methane well;
[0027] Generate the gas-phase superficial velocity and liquid-phase superficial velocity of the first deep coalbed methane well according to the gas production, water production, and wellbore diameter of the first deep coalbed methane well.
[0028] In some embodiments, the device further includes:
[0029] A second acquisition module, configured to acquire key scenario learning samples, where the key scenario learning samples are generated according to the actual production data of at least one second deep coalbed methane well in each historical period;
[0030] A training module, configured to train a preset machine learning model according to the key scenario learning samples to obtain a trained key scenario recognition model.
[0031] In some embodiments, the device further includes:
[0032] A third acquisition module, configured to acquire the actual production data of the second deep coalbed methane well in each historical period;
[0033] A classification module, configured to classify the actual production data in each historical period by using an unsupervised classification algorithm to obtain a scenario label corresponding to the actual production data in each historical period, where the scenario labels include well completion, production increase, gas production decrease, stable gas production, gas production increase, and rapid decay;
[0034] A second generation module, configured to generate learning samples of the key scenario recognition model according to the actual production data in each historical period and the corresponding scenario labels.
[0035] An embodiment of the present application further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the key scenario recognition method for the deep coalbed methane well production process described in any one of the above embodiments are implemented.
[0036] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the key scenario recognition method for the deep coalbed methane well production process described in any one of the above embodiments are implemented.
[0037] The key scenario recognition method for the production process of deep coalbed methane wells provided by the embodiments of the present application generates key feature data in the production process of the first deep coalbed methane well according to the actual production data in the production process of the first deep coalbed methane well, and then uses the pre-trained key scenario recognition model to recognize the key scenarios in the production process of the first deep coalbed methane well according to the key feature data in the production process of the first deep coalbed methane well. In this way, it can support the automatic judgment of the production status of gas wells in the trial production and development of deep coalbed methane, thereby assisting production management and decision-making, solving the current situation that the change of the production status of gas wells in the production process of deep coalbed methane wells highly depends on manual monitoring and judgment, effectively improving the efficiency of production management of deep coalbed methane wells, helping production personnel timely master the well status, and adjusting the production process. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings. In the drawings:
[0039] Figure 1 is a schematic flowchart of the key scenario recognition method for the production process of deep coalbed methane wells provided by an embodiment of the present application.
[0040] Figure 2 is a partial schematic flowchart of the key scenario recognition method for the production process of deep coalbed methane wells provided by an embodiment of the present application.
[0041] Figure 3 is a partial schematic flowchart of the key scenario recognition method for the production process of deep coalbed methane wells provided by a specific embodiment of the present application.
[0042] Figure 4 is a partial schematic flowchart of the key scenario recognition method for the production process of deep coalbed methane wells provided by a specific embodiment of the present application.
[0043] Figure 5 is a partial schematic flowchart of the key scenario recognition method for the production process of deep coalbed methane wells provided by a specific embodiment of the present application.
[0044] Figure 6 is a schematic diagram of the key scenario recognition result for the production process of deep coalbed methane wells obtained by summarizing the recognition results of each stage of the production of deep coalbed methane wells provided by a specific embodiment of the present application.
[0045] Figure 7 is a schematic diagram of the key production data of a certain deep coalbed methane well extracted by using the method provided by the present application.
[0046] Figure 8 is generated according to Figure 7 the key production data of a certain deep coalbed methane well in
[0047] Figure 9 is generated according to Figure 7 the key production data of a certain deep coalbed methane well in
[0048] Figure 10 is a schematic diagram of the clustering results after clustering the converted deep coalbed methane characteristic data by using the K-means, EM, and FarthestFirst methods respectively.
[0049] Figure 11 is a schematic diagram of the clustering results after clustering the converted deep coalbed methane characteristic data by using the EM method.
[0050] Figure 12 is a schematic diagram of the classification results of the key scenarios in the production process of a certain deep coalbed methane well and the gas-liquid apparent velocity distribution.
[0051] Figure 13 is a schematic diagram of the structure of the key scenario recognition device for the production process of a deep coalbed methane well provided by an embodiment of the present application.
[0052] Figure 14 is a schematic diagram of the physical structure of the electronic device provided by an embodiment of the present application. Detailed implementation manners
[0053] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer and more understandable, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings. Herein, the illustrative embodiments of the present application and their descriptions are used to explain the present application, but not to limit the present application. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be arbitrarily sorted with each other.
[0054] Regarding the "first", "second",... used herein, they do not particularly refer to the order or sequence, nor are they used to limit the present application. They are only used to distinguish the elements or operations described with the same technical terms.
[0055] Regarding the "including", "comprising", "having", "containing", etc. used herein, they are all open-ended terms, that is, they mean including but not limited to.
[0056] Regarding the "and / or" used herein, it includes any or all of the sorted items.
[0057] To better understand this application, the research background of this application will be described in detail below.
[0058] The execution subject of the key scenario recognition method for the production process of deep coalbed methane wells provided by the embodiments of this application includes, but is not limited to, a computer.
[0059] Figure 1 It is a schematic flowchart of the key scenario recognition method for the production process of deep coalbed methane wells provided by an embodiment of this application. As Figure 1 shown, the key scenario recognition method for the production process of deep coalbed methane wells provided by the embodiments of this application includes:
[0060] S101. Obtain the actual production data during the production process of the first deep coalbed methane well;
[0061] S102. Generate key feature data according to the actual production data;
[0062] S103. Input the key feature data into a pre-trained key scenario recognition model to obtain the key scenarios recognized by the key scenario recognition model.
[0063] The key scenario recognition method for the production process of deep coalbed methane wells provided by the embodiments of this application generates the key feature data during the production process of the first deep coalbed methane well according to the actual production data during the production process of the first deep coalbed methane well, and then uses the pre-trained key scenario recognition model to recognize the key scenarios during the production process of the first deep coalbed methane well according to the key feature data during the production process of the first deep coalbed methane well. In this way, it can support the automatic judgment of the production status of gas wells in the trial production and development of deep coalbed methane, thereby assisting production management and decision-making, solving the current situation that the change of the production status of gas wells in the production process of deep coalbed methane wells highly depends on manual monitoring and judgment, effectively improving the efficiency of production management of deep coalbed methane wells, helping production personnel timely master the well status, and adjusting the production process.
[0064] In some embodiments, the actual production data includes: the bottom hole pressure, casing pressure, gas production volume, and water production volume of the first deep coalbed methane well.
[0065] As Figure 2 shown, in some embodiments, the generating key feature data according to the actual production data includes:
[0066] S1021. Generate the pressure gradient of the difference between the bottom hole pressure and the casing pressure of the first deep coalbed methane well with the well depth according to the bottom hole pressure, casing pressure of the first deep coalbed methane well, and the well depth of the first deep coalbed methane well;
[0067] S1022. Generate the gas-phase superficial velocity and liquid-phase superficial velocity of the first deep coalbed methane well based on the gas production volume, water production volume, and wellbore diameter of the first deep coalbed methane well.
[0068] Specifically, the calculation formulas for the above key feature data are as follows:
[0069]
[0070] In the formula: dP l is the pressure gradient of the difference between the bottom-hole pressure and the casing pressure of the deep coalbed methane well with respect to the well depth; P f is the bottom-hole pressure of the deep coalbed methane well; P c is the casing pressure of the deep coalbed methane well; H w is the depth of the deep coalbed methane well; v sg is the gas-phase superficial velocity of the deep coalbed methane well; v sl is the liquid-phase superficial velocity of the deep coalbed methane well; Q g is the gas production volume of the deep coalbed methane well; Q l is the liquid production volume of the deep coalbed methane well; d w is the wellbore diameter of the deep coalbed methane well.
[0071] As Figure 3 shown, in some embodiments, the method further includes:
[0072] S104. Obtain key scenario learning samples, where the key scenario learning samples are generated based on the actual production data of at least one second deep coalbed methane well in each historical period;
[0073] S105. Train a preset machine learning model according to the key scenario learning samples to obtain a trained key scenario recognition model.
[0074] Specifically, the second deep coalbed methane well is different from the first deep coalbed methane well. The key scenario recognition model trained using the key scenario learning samples generated from the actual production data of the second deep coalbed methane well in each historical period can be used to identify the key scenarios in the production process of a new deep coalbed methane well, such as identifying the key scenarios in the production process of the first deep coalbed methane well. The preset machine learning model can be a supervised learning algorithm model.
[0075] As Figure 4 shown, in some embodiments, the method further includes:
[0076] S106. Obtain the actual production data of the second deep coalbed methane well in each historical period;
[0077] S107. Classify the actual production data of each historical period using an unsupervised classification algorithm to obtain the scenario labels corresponding to the actual production data of each historical period. The scenario labels include well completion, production increase, gas production decrease, stable gas production, gas production increase, and rapid decline.
[0078] S108. Generate learning samples for the key scenario recognition model based on the actual production data and corresponding scenario labels of each historical period.
[0079] Specifically, classify the samples using an unsupervised learning classification algorithm. For the selection and evaluation of the classification algorithm, six types of scenario identifiers are used for evaluation, including well completion, production increase, gas production decrease, stable gas production, gas production increase, and rapid decline. Use the above classification results as learning samples and perform learning using a supervised learning algorithm to obtain the recognition model for key scenarios of deep coalbed methane.
[0080] To better understand the present application, the following provides a detailed introduction to the method for recognizing key scenarios in the production process of deep coalbed methane wells provided by the present application through a specific embodiment.
[0081] The purpose of this embodiment is to propose a method for recognizing key scenarios in the production process of deep coalbed methane wells. Through the intelligent extraction, conversion, and processing of key production data of deep coalbed methane wells, intelligent recognition of various key scenarios in the production process of deep coalbed methane wells is realized to support the automatic judgment of the production status of gas wells in the trial production and development of deep coalbed methane, thereby assisting production management and decision-making. Solve the current situation that the change of the production status of gas wells in the production process of deep coalbed methane wells highly depends on manual monitoring and judgment, effectively improve the efficiency of production management of deep coalbed methane wells, help production personnel timely master the well status, and adjust the production process.
[0082] As Figure 5 described, a method for recognizing key scenarios in the production process of deep coalbed methane wells provided by this embodiment includes:
[0083] S1. Extract key production data of deep coalbed methane wells by extracting specific parameters and features from various production parameters of deep coalbed methane wells.
[0084] S2. Convert the key production data of deep coalbed methane wells to obtain learning samples for key scenarios of deep coalbed methane through the conversion of the extracted key data of deep coalbed methane.
[0085] S3. Build an intelligent recognition model for key scenarios of deep coalbed methane, and establish an intelligent recognition model for key scenarios of deep coalbed methane using the learning samples for key scenarios of deep coalbed methane wells.
[0086] S4. Obtain the actual production data. Through filtering and processing the actual production data, use the generated intelligent recognition model for key scenarios of deep coalbed methane to calculate, so as to obtain the discrimination results of key scenarios in the production of deep coalbed methane wells. Summarize the recognition results of each stage in the production of deep coalbed methane wells, and the key scenario recognition and classification results in the production process of deep coalbed methane wells as shown in Figure 6 can be obtained.
[0087] In the above method for identifying key scenarios in the production process of deep coalbed methane wells, for the parameter extraction in the production process of deep coalbed methane wells, the key characteristic parameters of deep coalbed methane wells extracted include: bottom-hole pressure of the gas well, casing pressure, gas production rate, and water production rate. Among them, the bottom-hole pressure of the gas well, casing pressure, gas production rate, and water production rate include the full-life cycle data of deep coalbed methane test wells and the real-time collected data of production wells.
[0088] In the above method for identifying key scenarios in the production process of deep coalbed methane wells, according to the fluid flow law and generation characteristics in the deep coalbed methane well, transform the extracted bottom-hole pressure of the gas well, casing pressure, gas production rate, and water production rate. Among them, referring to the above formula (1), convert the bottom-hole pressure and casing pressure of each well into the pressure gradient of the difference between the bottom-hole pressure and casing pressure of the gas well with the well depth using the well depth of the deep coalbed methane well, and convert the gas production rate and water production rate into the apparent gas velocity and apparent liquid velocity of the deep coalbed methane well using the wellbore diameter of the deep coalbed methane well, so as to obtain the learning samples of key scenarios of deep coalbed methane wells.
[0089] In the above method for identifying key scenarios in the production process of deep coalbed methane wells, construct an intelligent recognition model for key scenarios of deep coalbed methane, and classify the samples through an unsupervised learning classification algorithm. Among them, the selection and evaluation of the classification algorithm are evaluated using six types of scenario identifiers, including well completion, production increase, gas production reduction, stable gas production, gas production increase, and rapid decay. Mark the scenario that can obtain the most significant classification with the corresponding algorithm (that is to say, first manually label some samples, classify the samples with labels using the unsupervised learning classification algorithm, adjust the parameters of the unsupervised learning classification algorithm according to the classification results until the classification results of the unsupervised learning classification algorithm meet the expectations, and then use the unsupervised learning classification algorithm to classify all the samples to obtain the labels of each sample), and at the same time, manually classify the samples that cannot be clearly classified significantly, and use the above classification results as learning samples and learn using a supervised learning algorithm, so as to obtain the intelligent recognition model for key scenarios of deep coalbed methane.
[0090] In the above method for identifying key scenarios in the production process of deep coalbed methane wells, obtain production data. The filtering of key parameters in the actual production data includes two processes: outlier filtering and Kalman filtering. Through filtering, obtain the corresponding characteristic parameters of the actual generation, and use the intelligent recognition model for key scenarios of deep coalbed methane to calculate to obtain the key scenarios of deep coalbed methane wells.
[0091] The key scenario recognition method for the production process of deep coalbed methane wells proposed by the present invention can change the traditional practice of relying on manual monitoring and judgment for production conditions during the trial production and formal production of deep coalbed methane, and theoretically can achieve second-level adjustment suggestions and continuous optimization for the regulation of multi-well collaborative production devices, realize the efficient management of integrated production of multi-well fields, solve the problem of large differences in well conditions between wells caused by significant differences in production pressure, gas production, and water production of deep coalbed methane wells at different production stages, and the difficulty of multi-well collaboration, and provide key technical support for the collaborative, stable, continuous, and efficient production of deep coalbed methane wells.
[0092] For example, first, key production data of deep coalbed methane wells need to be extracted. As Figure 7 shown, four characteristics, namely the bottom hole pressure, casing pressure, gas production, and water production of deep coalbed methane wells, are extracted from various production parameters of a certain deep coalbed methane well.
[0093] In the above key scenario recognition method for the production process of deep coalbed methane wells, the obtained key production data of deep coalbed methane wells are converted using formula (1), where Figure 8 is the corresponding distribution relationship between the gas-phase apparent velocity and the water-phase apparent velocity of a certain deep coalbed methane well obtained by conversion, and Figure 9 is the change situation of the gradient of the difference between the bottom hole pressure and the casing pressure of a certain deep coalbed methane well obtained by conversion.
[0094] In the above key scenario recognition method for the production process of deep coalbed methane wells, the converted characteristic data of deep coalbed methane are optimized and automatically classified using an unsupervised classification algorithm. As Figure 10 shown, the clustering results obtained by using the K-means, EM, and FarthestFirst methods respectively are presented. Among them, the optimized classification method is EM, and the optimized number of classification categories is 7, as Figure 11 shown.
[0095] In the above key scenario recognition method for the production process of deep coalbed methane wells, a key scenario learning sample is constructed using the optimized key scenario classification method and classification quantity of deep coalbed methane, and a key scenario intelligent recognition model for deep coalbed methane is established. Through learning, the intelligent recognition model can be used to process and predict new condition data obtained during the production process of deep coalbed methane wells. As Figure 12 shown, it is the key scenario classification result and the gas-liquid apparent velocity distribution diagram of a certain deep coalbed methane well production process. It can be seen that the intelligent classification method has well realized the recognition of key production scenario conditions, and the coincidence degree between the intelligent prediction classification scenario and the manually defined classification reaches 95.78%.
[0096] Figure 13 is the structural schematic diagram of the key scenario recognition device for the production process of deep coalbed methane wells provided by an embodiment of the present application. AsFigure 13 As shown in Figure 13 , the key scenario recognition device for the production process of deep coalbed methane wells provided by the embodiments of the present application includes:
[0097] A first acquisition module 21, configured to acquire actual production data during the production process of the first deep coalbed methane well;
[0098] A generation module 22, configured to generate key feature data according to the actual production data;
[0099] An identification module 23, configured to input the key feature data into a pre-trained key scenario recognition model to obtain the key scenarios recognized by the key scenario recognition model.
[0100] The key scenario recognition device for the production process of deep coalbed methane wells provided by the embodiments of the present application generates key feature data during the production process of the first deep coalbed methane well according to the actual production data during the production process of the first deep coalbed methane well, and then uses the pre-trained key scenario recognition model to recognize the key scenarios during the production process of the first deep coalbed methane well according to the key feature data during the production process of the first deep coalbed methane well. In this way, it can support the automatic judgment of the production status of gas wells during the trial production and development of deep coalbed methane, thereby assisting production management and decision-making, solving the current situation that the change of the production status of gas wells during the production process of deep coalbed methane wells highly depends on manual monitoring and judgment, effectively improving the efficiency of production management of deep coalbed methane wells, and helping production personnel timely master the well status and adjust the production process.
[0101] In some embodiments, the actual production data includes: the bottom-hole pressure, casing pressure, gas production volume, and water production volume of the first deep coalbed methane well.
[0102] In some embodiments, the generation module is specifically configured to:
[0103] Generate a pressure gradient of the difference between the bottom-hole pressure and the casing pressure of the first deep coalbed methane well with respect to the well depth according to the bottom-hole pressure, casing pressure of the first deep coalbed methane well, and the well depth of the first deep coalbed methane well;
[0104] Generate the apparent gas velocity and apparent liquid velocity of the first deep coalbed methane well according to the gas production volume, water production volume of the first deep coalbed methane well, and the wellbore diameter of the first deep coalbed methane well.
[0105] In some embodiments, the device further includes:
[0106] A second acquisition module, configured to acquire key scenario learning samples, where the key scenario learning samples are generated according to the actual production data of at least one second deep coalbed methane well in each historical period;
[0107] A training module for training a preset machine learning model according to the key scenario learning samples to obtain a trained key scenario recognition model.
[0108] In some embodiments, the apparatus further includes:
[0109] A third acquisition module for acquiring the actual production data of the second deep coalbed methane well in each historical period;
[0110] A classification module for classifying the actual production data in each historical period by using an unsupervised classification algorithm to obtain a scenario label corresponding to the actual production data in each historical period, where the scenario label includes well completion, production increase, gas production decrease, stable gas production, gas production increase, and rapid decay;
[0111] A second generation module for generating learning samples of the key scenario recognition model according to the actual production data in each historical period and the corresponding scenario label.
[0112] The embodiments of the apparatus provided in the embodiments of the present application can specifically be used to execute the processing procedures of the above method embodiments, and their functions will not be elaborated here. Reference can be made to the detailed descriptions of the above method embodiments.
[0113] Figure 14 It is a schematic physical structure diagram of an electronic device provided in an embodiment of the present application. As Figure 14 shown, the electronic device may include: a processor 301, a communication interface 302, a memory 303, and a communication bus 304. Among them, the processor 301, the communication interface 302, and the memory 303 communicate with each other through the communication bus 304. The processor 301 can call the logical instructions in the memory 303 to execute the method described in any of the above embodiments.
[0114] In addition, when the logical instructions in the above-mentioned memory 303 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.
[0115] This embodiment discloses a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided in the above-mentioned method embodiments.
[0116] This embodiment provides a computer-readable storage medium. The computer-readable storage medium stores a computer program. The computer program causes the computer to execute the methods provided in the above-mentioned method embodiments.
[0117] Those skilled in the art should understand that the embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, this application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) that contain computer-usable program codes.
[0118] This application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of this application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.
[0119] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one process Figure 1 one process or more processes and / or blocks Figure 1 or multiple blocks.
[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, such that a series of operation steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process Figure 1 one process or more processes and / or blocks Figure 1 or multiple blocks.
[0121] In the description of this specification, the descriptions with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.
[0122] The above specific embodiments have further elaborated on the purpose, technical solutions, and beneficial effects of the present application. It should be understood that the above are only specific embodiments of the present application and are not used to limit the protection scope of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
Claims
1. A method for identifying key scenarios in the production process of deep coalbed methane wells, characterized in that, Including: Obtain the actual production data during the production process of the first deep coalbed methane well; Generate key feature data according to the actual production data; Input the key feature data into a pre-trained key scenario recognition model to obtain the key scenarios recognized by the key scenario recognition model.
2. The method according to claim 1, wherein The actual production data includes: the bottom hole pressure, casing pressure, gas production volume, and water production volume of the first deep coalbed methane well.
3. The method according to claim 2, wherein The generating the key feature data according to the actual production data includes: Generate the pressure gradient of the difference between the bottom hole pressure and the casing pressure of the first deep coalbed methane well with the well depth according to the bottom hole pressure, casing pressure of the first deep coalbed methane well, and the well depth of the first deep coalbed methane well; Generate the apparent gas velocity and apparent liquid velocity of the first deep coalbed methane well according to the gas production volume, water production volume of the first deep coalbed methane well, and the wellbore diameter of the first deep coalbed methane well.
4. The method according to any one of claims 1 to 3, characterized in that The method further includes: Obtain key scenario learning samples, where the key scenario learning samples are generated according to the actual production data of at least one second deep coalbed methane well in each historical period; Train a preset machine learning model according to the key scenario learning samples to obtain a trained key scenario recognition model.
5. The method according to claim 4, wherein The method further includes: Obtain the actual production data of the second deep coalbed methane well in each historical period; Use an unsupervised classification algorithm to classify the actual production data of each historical period to obtain the scenario labels corresponding to the actual production data of each historical period, and the scenario labels include well completion, production increase, gas production decrease, stable gas production, gas production increase, and rapid decay; Generate learning samples of the key scenario recognition model according to the actual production data and the corresponding scenario labels of each historical period.
6. A key scenario recognition device for the production process of deep coalbed methane wells, characterized in that, Including: A first acquisition module for obtaining the actual production data during the production process of the first deep coalbed methane well; A generation module for generating key feature data according to the actual production data; An identification module for inputting the key feature data into a pre-trained key scenario recognition model to obtain the key scenarios recognized by the key scenario recognition model.
7. The device according to claim 6, characterized in that, The actual production data includes: the bottom hole pressure, casing pressure, gas production volume, and water production volume of the first deep coalbed methane well.
8. The device according to claim 7, characterized in that, The generation module is specifically used for: Generate the pressure gradient of the difference between the bottom hole pressure and the casing pressure of the first deep coalbed methane well with the well depth according to the bottom hole pressure, casing pressure of the first deep coalbed methane well, and the well depth of the first deep coalbed methane well; Generate the apparent gas velocity and apparent liquid velocity of the first deep coalbed methane well according to the gas production volume, water production volume of the first deep coalbed methane well, and the wellbore diameter of the first deep coalbed methane well.
9. The device according to any one of claims 6 to 8, characterized in that, The device further includes: A second acquisition module for obtaining key scenario learning samples, where the key scenario learning samples are generated according to the actual production data of at least one second deep coalbed methane well in each historical period; A training module for training a preset machine learning model according to the key scenario learning samples to obtain a trained key scenario recognition model.
10. The device according to claim 9, wherein The device further includes: A third acquisition module for obtaining the actual production data of the second deep coalbed methane well in each historical period; A classification module, configured to classify the actual production data of each historical period by using an unsupervised classification algorithm, so as to obtain a scenario label corresponding to the actual production data of each historical period, where the scenario label includes production startup, production increase, gas production reduction, stable gas production, gas production increase, and rapid decay; A second generation module, configured to generate learning samples of a key scenario recognition model according to the actual production data of each historical period and the corresponding scenario label.
11. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.
12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 5 are implemented.