An oil and gas pipeline failure detection method, system, storage medium and electronic device

Abstract

The present application relates to the technical field of oil and gas pipeline detection, and more particularly to an oil and gas pipeline failure detection method and system, a storage medium and an electronic device, the method comprising: storing failure multi-source knowledge and multiple fault trees of oil and gas field pipelines in a knowledge base, and storing collected oil and gas pipeline failure cases in the knowledge base; training a preset learning network model in combination with the failure multi-source knowledge, the fault trees and the oil and gas pipeline failure cases in the knowledge base to obtain a failure model; and inputting preset failure problem parameters into the failure model to predict failure causes and risks. The method can predict failure causes and risks, help eliminate hidden dangers in the safe operation of oil and gas pipelines, and ensure the safe operation of oil and gas pipelines.

Description

Technical Field

[0001] This invention relates to the field of oil and gas pipeline inspection technology, and in particular to an oil and gas pipeline failure detection method, system, storage medium, and electronic device. Background Technology

[0002] After years of digital and intelligent construction of oil and gas pipelines, a massive amount of operational data on oil and gas pipelines has been accumulated. Faced with the growing demand for knowledge retrieval and interaction from interdisciplinary fields, the key challenge in the knowledge base field is to comprehensively and accurately retrieve the necessary information and knowledge from this massive, heterogeneous data and intelligently recommend it to users. However, currently, data and knowledge related to pipeline design, manufacturing, operation, emergency response, management, and failure are not yet integrated, hindering decision-making regarding the causes of oil and gas pipeline failures and making it difficult to promptly identify potential safety hazards. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to address the shortcomings of the prior art by providing a method, system, storage medium and electronic device for detecting failures in oil and gas pipelines.

[0004] The technical solution of the oil and gas pipeline failure detection method of the present invention is as follows:

[0005] The knowledge of multiple sources of failure of oil and gas pipelines and multiple fault trees are stored in the knowledge base, and the collected failure cases of oil and gas pipelines are also stored in the knowledge base.

[0006] The failure model is obtained by training a preset learning network model by combining the multi-source failure knowledge, fault tree, and oil and gas pipeline failure cases in the knowledge base.

[0007] By inputting preset failure problem parameters into the failure model, the causes and risks of failure can be predicted.

[0008] The technical solution of the oil and gas pipeline failure detection system of the present invention is as follows:

[0009] It includes a storage module, a training module, and a detection module;

[0010] The storage module is used to: store multi-source knowledge of oil and gas pipeline failures and multiple fault trees into a knowledge base, and store collected oil and gas pipeline failure cases into the knowledge base;

[0011] The training module is used to: train a preset learning network model by combining the multi-source failure knowledge, fault tree and oil and gas pipeline failure cases in the knowledge base to obtain a failure model;

[0012] The detection module is used to: input preset failure problem parameters into the failure model to predict the causes and risks of failure.

[0013] The present invention provides a storage medium storing instructions, wherein when a computer reads the instructions, the computer executes any of the above-described methods for detecting oil and gas pipeline failures.

[0014] An electronic device according to the present invention includes a processor and the above-described storage medium, wherein the processor executes instructions in the storage medium.

[0015] The technical effects of this invention are as follows:

[0016] Being able to predict the causes and risks of failure helps to eliminate potential hazards to the safe operation of oil and gas pipelines and ensure their safe operation. Attached Figure Description

[0017] Other features, objects, and advantages of the invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0018] Figure 1 This is a schematic flowchart of a method for detecting failures in oil and gas pipelines according to an embodiment of the present invention.

[0019] Figure 2 This is a schematic diagram of a fault tree.

[0020] Figure 3 A schematic diagram of the structure of an oil and gas pipeline failure knowledge base system;

[0021] Figure 4 This is a schematic diagram illustrating the process of outputting failure causes and failure risks.

[0022] Figure 5 This is a schematic diagram of the training algorithm process;

[0023] Figure 6 This is a schematic diagram of the structure of an oil and gas pipeline failure detection system according to an embodiment of the present invention. Detailed Implementation

[0024] like Figure 1 As shown, an oil and gas pipeline failure detection method according to an embodiment of the present invention includes the following steps:

[0025] S1. Store the multi-source knowledge of oil and gas pipeline failures and multiple fault trees into the knowledge base, and store the collected oil and gas pipeline failure cases into the knowledge base.

[0026] Among them, the multi-source knowledge of oil and gas field pipeline failure includes static knowledge and dynamic knowledge:

[0027] 1) Static knowledge includes basic information about oil and gas pipelines and station facilities, failure modes of oil and gas pipelines, causes of oil and gas pipeline failure, and failure mechanisms of oil and gas pipelines and station facilities. Specifically:

[0028] ①Basic information about oil and gas pipelines and station facilities includes: pipeline specifications, design pressure, material, medium type, welding process, welding specifications, applicable standards, equipment geometry, dimensions, service life, and operating conditions.

[0029] ② Failure modes of oil and gas pipelines include: explosion, fracture, deformation, corrosion, and external mechanical damage;

[0030] ③ The causes of oil and gas pipeline failure include: corrosion, external interference, welding and material defects, equipment and operation, natural disasters, etc.

[0031] ④ Failure mechanisms of oil and gas pipelines and station facilities include: ductile fracture, brittle fracture, fatigue fracture, creep fracture and corrosion failure mechanisms.

[0032] 2) Dynamic knowledge includes: fault tree rules and empirical knowledge, as well as learning outcomes, specifically:

[0033] ① Fault Tree Rule: To accurately reflect failure information, failure events in oil and gas pipelines are categorized into different levels, such as single failure mode, specific manifestations of failure modes, sets of specific causes of failure, and sets of failure response measures, to establish a fault tree. A fault tree is a hierarchical model, where each node corresponds to a set of specific knowledge. This set of knowledge reflects the principles, experiences, and other heuristic knowledge accumulated over a long period in the field of failure analysis, such as... Figure 2 As shown.

[0034] The failure tree is stored in the knowledge base in the form of production rules, including basic failure rules, theorems, mechanisms, and models. A rule typically consists of two parts: a precondition and a conclusion. The precondition is the necessary condition to be met during rule reasoning, and the conclusion is the result of the reasoning. The logical relationship is briefly described as IF <condition> THEN <conclusion>. For example, IF the fracture morphology of the circumferential weld is a flat surface, without a shear lip, and the fracture surface is relatively bright and gray. THEN the circumferential weld undergoes brittle fracture.

[0035] ②Experience knowledge: Experts' experience in failure analysis, diagnosis and prediction. Experts can dynamically add experience knowledge to the knowledge base;

[0036] ③ Learning results: Weights and thresholds after training a pre-defined learning network model, such as an artificial neural network.

[0037] The specific process of storing multi-source failure knowledge of oil and gas field pipelines into a knowledge base is as follows:

[0038] Multi-source knowledge storage can be stored in the form of data using a specific structure, taking into account storage costs and the efficiency of operations such as adding, querying, modifying, and deleting. Furthermore, based on the subsequent analysis and knowledge reasoning needs of this paper, it also considers the support for reasoning capabilities and efficiency, and adopts a graph database format to store the static and dynamic knowledge of oil and gas multi-source knowledge.

[0039] The specific process for collecting cases of oil and gas pipeline failures is as follows:

[0040] For failure cases of oil and gas pipelines, it is necessary to establish a failure case data collection table to record as many factors as possible that affect the failure, forming a failure case of oil and gas pipelines. This includes collecting and storing data such as the surrounding environment of the oil and gas pipeline, operating condition data, basic pipeline information, mechanical performance data, failure location, and defect detection information.

[0041] S2. Train the preset learning network model by combining the multi-source failure knowledge, fault tree and oil and gas pipeline failure cases in the knowledge base to obtain the failure model;

[0042] S3. Input the preset failure problem parameters into the failure model to predict the causes and risks of failure.

[0043] The process of obtaining samples:

[0044] Specific data on the influencing factors of any failure event are obtained from multi-source failure knowledge, fault trees, and oil and gas pipeline failure cases, and then normalized to obtain the normalized data of the influencing factors of the failure event. The corresponding output of the failure event is the cause and risk of failure. This process is repeated to construct a sample, and so on, until multiple samples are constructed. The specific training process is as follows: Based on multiple samples, a neural network containing a determined number of nodes in the input layer, hidden layer, and output layer is trained to obtain the failure model.

[0045] Optionally, the above technical solution also includes: making decisions based on predicted failure causes and risks, as follows:

[0046] When the output risk is low, normal monitoring measures are adopted; when the output risk is medium, planned repair and reinforcement measures are adopted; when the output risk is high, immediate repair and reinforcement measures or pipe replacement measures are adopted.

[0047] Optionally, in the above technical solution, in step S1, the collected oil and gas pipeline failure cases are stored in a knowledge base, including:

[0048] S10. According to the failure mode, classify the collected oil and gas pipeline failure cases, normalize the values ​​and parameters of the failure influencing factors in each oil and gas pipeline failure case in each category, and store the normalized data in the knowledge base. Specifically:

[0049] For failure cases of oil and gas pipelines, it is necessary to classify the failure cases according to failure modes, and then normalize the values ​​and parameters of the failure influencing factors in the failure cases so that the failure cases can be input in the form of data. The failure results of the failure cases are also classified into high risk, medium risk, and low risk. High risk is defined as an output value of 8-10, medium risk as an output value of 4-7, and low risk as an output value of 1-3. The failure cases are then stored in the form of data for artificial neural networks to learn from. Finally, the input and output data of the failure case library are stored in a database.

[0050] In S2, the specific training process is as follows:

[0051] The output results of failure cases can be divided into three categories: "high risk", "medium risk", and "low risk". The training formula for failure cases is: r = g(c0, c1, ..., c m-1 ), where c0, c1, ..., c m-1 represents the parameters corresponding to the failure cases. g represents the mapping relationship. r represents the output result.

[0052] Input quantities c0~c m-1 It involves various data types, including continuous, discrete, and ensemble quantities, with the output being discrete. Backpropagation (BP) neural networks possess the ability to classify arbitrarily complex patterns and have excellent multidimensional function mapping capabilities, enabling them to effectively achieve the required mappings for prediction through learning. The BP neural network algorithm is used to train failure cases. The weights and thresholds obtained after training a large number of failure cases using artificial neural networks are stored in a knowledge base.

[0053] Based on the above steps, a knowledge base system for oil and gas pipeline failures can also be constructed, such as... Figure 3 As shown, the specific functional modules are configured as follows:

[0054] 1) Human-computer interaction module:

[0055] The features include importing and exporting oil and gas pipeline failure data and knowledge, user management, user permission allocation, and the ability to input, add, delete, and save user information. It can easily and quickly input oil and gas pipeline failure cases and failure system knowledge into the knowledge base system.

[0056] 2) Database module:

[0057] It mainly manages the data of oil and gas pipeline failure cases, intermediate data of failure tree logical reasoning, and intermediate data of artificial neural network learning, training and operation. This includes adding, modifying, saving, filtering and deleting data, and importing data from other databases and calling database data.

[0058] 3) Knowledge Base Module:

[0059] The system primarily comprises static knowledge (basic information about the failure knowledge base system and failure system knowledge), dynamic knowledge (failure tree rules and the results of artificial neural network learning and training), a learning system (artificial neural network learning and training model), and a knowledge acquisition and management system (knowledge acquisition methods and management). The knowledge base module can acquire user-input knowledge, knowledge extracted through web crawling, knowledge learned from failure case studies, and knowledge imported from other knowledge bases. It represents this knowledge in knowledge graph and relational data formats, stores it in a graph database, and facilitates its use by application systems.

[0060] 4) Application Module:

[0061] The application module includes failure statistics and failure prediction sub-modules. Inputting the characteristics of a failure event provides the corresponding failure cause, mechanism, and countermeasures; inputting the values ​​of failure influencing factors provides the failure risk.

[0062] 5) Help module:

[0063] The help module is used to introduce the functions of the failure knowledge base system, provide system usage instructions, and offer solutions to some common problems.

[0064] The development tools for the oil and gas pipeline failure knowledge base system are shown in Table 1.

[0065] Table 1:

[0066]

[0067]

[0068] After constructing the oil and gas pipeline failure knowledge base system, failure decision-making questions are input into the human-computer interaction module, including questions on failure analysis and failure prediction. For failure analysis questions, the main input in the human-computer interaction module is the failure mode of the failure event that has occurred, or the specific manifestation of the failure mode. For failure influencing factors that have not yet occurred, the normalized data involves various data types, including continuous, discrete, and ensemble quantities.

[0069] The input parameters from the human-computer interaction module are fed into the inference engine of the oil and gas pipeline failure knowledge base system. Different algorithm strategies are assigned based on the problem type (failure analysis and failure prediction). Failure analysis problems correspond to logical reasoning methods, while failure prediction problems correspond to artificial neural network methods. The logical reasoning in the inference engine calls the failure fault tree rules stored in the knowledge base, and intermediate data during the reasoning process is stored in the database. The artificial neural network method in the inference engine calls the weights and thresholds of pre-trained failure cases (already stored in the knowledge base). Finally, for failure analysis problems, the failure cause is output; for failure prediction problems, the specific numerical value of the failure risk is output. The specific analysis process is as follows: Figure 4 As shown.

[0070] Based on the given failure causes and failure risk results, the failure decision scheme is shown in Table 2 below.

[0071] Table 2:

[0072]

[0073]

[0074] The invention will now be described through another embodiment:

[0075] S101. Collected 54 failure analysis reports, 32 pipeline company cut performance test reports, and 27,713 excavation test results from various pipeline companies, of which 2,905 were for repairing circumferential welds.

[0076] S102. Regarding the impact of factors affecting circumferential weld failure, the evaluation index system directly affects the accuracy of failure judgment. Based on the analysis of actual pipeline failure causes, the factors causing failure can be divided into three categories: pipe material type and performance-related indicators, defect-related indicators, and load-related indicators. Table 3 shows the various factors that can be obtained in practice and influence failure.

[0077] Table 3:

[0078]

[0079]

[0080] The failure case input is completed by selecting corresponding parameters from the data source or supplementing them according to the indicator system established in Table 4. The output definition is shown in Table 4. Then, the failure case input and output data are stored in the database.

[0081] Table 4:

[0082] Risk level Output range High risk 8-10 points Medium risk 4-7 points Low risk 1-3 points

[0083] S103, Construction of the Ring Welding Neural Network:

[0084] A backpropagation (BP) neural network was used, consisting of an input layer, hidden layers, and an output layer. Each layer is composed of neurons. Based on sample characteristics, a network structure with 19 input layer nodes, 25 hidden layer nodes, and 3 output layer nodes was selected for model training. Figure 5 The algorithm shown ultimately determines the learning sample set.

[0085] S104, Training:

[0086] First, a database was established containing information on 44 high-risk circumferential welds, 1823 medium-risk circumferential welds, and 2950 low-risk circumferential welds. The neural network structure was set to three layers: 19 nodes in the input layer, 3 nodes in the output layer, and 30 nodes in the hidden layer. The initial training sample size K for the high, medium, and low risk categories was defined as 10, and the learning step size was 0.001. The neural network was then trained. The trained neural network, i.e., the failure model, was used to conduct identification experiments on all weld data.

[0087] S105. Input preset failure problem parameters:

[0088] Table 5 shows 10 case studies of circumferential weld failure risks at a valve chamber of a pipeline company. Then, a knowledge base system for oil and gas pipeline failures was used to predict the failure risks of these circumferential welds.

[0089] Table 5:

[0090]

[0091] The specific parameters are input into the failure model, that is, the preset failure problem parameters are input into the failure model. The specific preset failure problem parameters are shown in Table 6.

[0092] Table 6:

[0093]

[0094]

[0095] S106, Failure Risk Output:

[0096] Failure risk prediction for circumferential welds was conducted using an oil and gas pipeline failure knowledge base system. The results of the failure risk prediction are shown in Table 7.

[0097] Table 7:

[0098] Case Number Weld number Output Failure risk 1# ZMQ-JH06-QAC112-1+001-W 2 Low risk 2# ZMQ-JH06-QAC112-1+002-M 3 Low risk 3# ZMQ-JH06-QAC112-1+003-L 5 Medium risk 4# ZMQ-JH06-QAC112-2+001-L 1 Low risk 5# ZMQ-JH06-QAC112-2+002-WL 6 Medium risk 6# ZMQ-JH06-QAC112-2+003-W 2 Low risk 7# ZMQ-JH06-QAC112-2+004-M 2 Low risk 8# ZMQ-JH06-QAC112-2+005-M 1 Low risk 9# ZMQ-JH06-QAC112-2+006-M 2 Low risk 10# ZMQ-JH06-QAC112-2+007-W 5 Medium risk

[0099] S107. Make a failure decision:

[0100] Failure risk prediction for circumferential welds was conducted using an oil and gas pipeline failure knowledge base system. The results of the failure risk prediction are shown in Table 8.

[0101] Table 8:

[0102]

[0103]

[0104] This invention primarily leverages multi-source oil and gas pipeline knowledge and case data, integrating oil and gas pipeline failure fault tree rules and artificial neural network training results into a knowledge base system. By designing the knowledge base system structure, logical reasoning and artificial neural networks are combined within the system's inference engine. Basic information about oil and gas pipelines, failure knowledge, rule knowledge, expert experience, and artificial neural network training results are input into the knowledge base, while failure case sample data is stored in the database. Finally, through the operation of the expert knowledge base system, failure problem parameters are input into the system. The inference engine processes the failure problem type and then calls upon the database and knowledge base to provide failure risks and causes.

[0105] This invention integrates knowledge from multiple knowledge bases, continuously discovering knowledge from data and dynamically updating the professional domain knowledge base to guide professional and technical personnel and managers in making scientific and efficient decisions. The knowledge base can also track the evolution, development, and cutting-edge aspects of knowledge, effectively solving knowledge-based reasoning and decision-making problems in related fields, such as intelligent search, intelligent question answering, and personalized recommendations. This helps eliminate potential safety hazards in oil and gas pipeline operations and ensures their safe operation.

[0106] In the above embodiments, although the steps are numbered S1, S2, etc., they are only specific embodiments given in this application. Those skilled in the art can adjust the execution order of S1, S2, etc. according to the actual situation, which is also within the protection scope of this invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0107] like Figure 6 As shown, an oil and gas pipeline failure detection system 200 according to an embodiment of the present invention includes a storage module 210, a training module 220 and a detection module 230;

[0108] The storage module 210 is used to: store multi-source knowledge of oil and gas pipeline failures and multiple fault trees into the knowledge base, and store collected oil and gas pipeline failure cases into the knowledge base;

[0109] Training module 220 is used to: train a preset learning network model by combining multi-source failure knowledge, fault tree and oil and gas pipeline failure cases in the knowledge base to obtain a failure model;

[0110] The detection module 230 is used to: input preset failure problem parameters into the failure model to predict the cause and risk of failure.

[0111] Optionally, in the above technical solution, the process by which the storage module 210 stores the collected oil and gas pipeline failure cases into the knowledge base includes:

[0112] According to the failure mode, the collected oil and gas pipeline failure cases are classified, and the values ​​and parameters of the failure influencing factors in each oil and gas pipeline failure case in each category are normalized. The normalized data is then stored in the knowledge base.

[0113] Optionally, in the above technical solution, the multi-source knowledge of oil and gas field pipeline failure includes static knowledge and dynamic knowledge;

[0114] Static knowledge includes basic information about oil and gas pipelines and station facilities, failure modes of oil and gas pipelines, causes of oil and gas pipeline failure, and failure mechanisms of oil and gas pipelines and station facilities.

[0115] Dynamic knowledge includes fault tree rules and empirical knowledge.

[0116] Optionally, in the above technical solution, the preset learning network model is a neural network.

[0117] The parameters and steps for each unit module to achieve their respective functions in the oil and gas pipeline failure detection system 200 of the present invention described above can be referred to the parameters and steps in the embodiments of the oil and gas pipeline failure detection method described above, and will not be repeated here.

[0118] The present invention provides a storage medium storing instructions, wherein when a computer reads the instructions, the computer executes any of the above-described methods for detecting oil and gas pipeline failures.

[0119] An electronic device according to the present invention includes a processor and the aforementioned storage medium, wherein the processor executes instructions in the storage medium, and the electronic device may be a computer, a mobile phone, or the like.

[0120] Those skilled in the art will know that this invention can be implemented as a system, method, or computer program product.

[0121] Therefore, this disclosure can be implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, the invention can also be implemented as a computer program product in one or more computer-readable media, the computer-readable medium containing computer-readable program code.

[0122] Any combination of one or more computer-readable media may be used. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0123] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A method of detecting failure of an oil and gas pipeline, characterized by, The method comprises the following steps: storing failure multi-source knowledge and multiple fault trees of oil and gas field pipelines in a knowledge base, and storing collected oil and gas pipeline failure cases in the knowledge base; training a preset learning network model in combination with the failure multi-source knowledge, the fault trees and the oil and gas pipeline failure cases in the knowledge base to obtain a failure model; inputting preset failure problem parameters into the failure model to predict failure causes and risks; storing the collected oil and gas pipeline failure cases in the knowledge base, which comprises the following steps: classifying the collected oil and gas pipeline failure cases according to failure modes, normalizing the values and parameters of failure influencing factors in each oil and gas pipeline failure case in each class, and storing the normalized data in the knowledge base; the failure multi-source knowledge of the oil and gas field pipelines comprises static knowledge and dynamic knowledge; the static knowledge comprises basic information of oil and gas pipelines and in-station facilities, failure modes of oil and gas pipelines, failure causes of oil and gas pipelines and failure mechanisms of oil and gas pipelines and in-station facilities; the dynamic knowledge comprises fault tree rules and experience knowledge; the fault trees are stored in the knowledge base in the form of production rules, the production rules are composed of premise conditions and conclusions, the premise conditions are necessary conditions to be met during reasoning, the conclusions are results after reasoning, and the logical relationship is briefly described as IF<condition> THEN<conclusion>.

2. A method of detecting failure of an oil and gas pipeline according to claim 1, characterized in that, the preset learning network model is a neural network.

3. An oil and gas pipeline failure detection system characterized by, The method comprises a storage module, a training module and a detection module; the storage module is configured to store failure multi-source knowledge and multiple fault trees of oil and gas field pipelines in a knowledge base, and store collected oil and gas pipeline failure cases in the knowledge base; the training module is configured to train a preset learning network model in combination with the failure multi-source knowledge, the fault trees and the oil and gas pipeline failure cases in the knowledge base to obtain a failure model; the detection module is configured to input preset failure problem parameters into the failure model to predict failure causes and risks; the storage module stores the collected oil and gas pipeline failure cases in the knowledge base, which comprises the following steps: classifying the collected oil and gas pipeline failure cases according to failure modes, normalizing the values and parameters of failure influencing factors in each oil and gas pipeline failure case in each class, and storing the normalized data in the knowledge base; the failure multi-source knowledge of the oil and gas field pipelines comprises static knowledge and dynamic knowledge; the static knowledge comprises basic information of oil and gas pipelines and in-station facilities, failure modes of oil and gas pipelines, failure causes of oil and gas pipelines and failure mechanisms of oil and gas pipelines and in-station facilities; the dynamic knowledge comprises fault tree rules and experience knowledge; the fault trees are stored in the knowledge base in the form of production rules, the production rules are composed of premise conditions and conclusions, the premise conditions are necessary conditions to be met during reasoning, the conclusions are results after reasoning, and the logical relationship is briefly described as IF<condition> THEN<conclusion>.

4. A system for detecting failure in an oil and gas pipeline according to claim 3, wherein the preset learning network model is a neural network.

5. A storage medium, characterized by The storage medium stores instructions, and when a computer reads the instructions, the computer executes the oil and gas pipeline failure detection method according to any one of claims 1 to 2.

6. An electronic device, comprising: including a processor and the storage medium of claim 5, the processor executing the instructions in the storage medium.

Images