Pipeline risk knowledge graph construction method and device and storage medium
By constructing a sub-risk knowledge map and building a pipeline risk knowledge map based on causal relationships, the problem that the existing technology cannot intuitively display the causal relationship of oil and gas pipelines is solved, and efficient identification and management of pipeline risks is achieved.
Patent Information
- Application Number
- CN202510161531.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-13
- Publication Date
- 2025-06-06
AI Technical Summary
The existing technology cannot intuitively demonstrate the causal driving relationship of oil and gas pipeline risks, and it is difficult to effectively identify and manage pipeline risks.
By obtaining multiple sets of risk information in the pipeline risk text, a sub-risk knowledge map is constructed, and a pipeline risk knowledge map is constructed based on the causal relationship, and the causal-driven relationship of risk is intuitively displayed.
It realizes the intuitive display of causal relationships of oil and gas pipeline risks, and improves the efficiency of risk identification and management.
Smart Images
Figure CN120106191A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of knowledge graph technology, and in particular to a pipeline risk knowledge graph construction method, device and storage medium. Background Art
[0002] Oil and gas pipelines involve major security interests, so it is crucial to ensure their safety and stability and to provide early warning of safety risks. Oil and gas pipelines face a variety of risks, and it is particularly important to build a pipeline risk knowledge map to effectively identify and manage these risks.
[0003] However, oil and gas pipelines have many failure modes and complex causes, and general technologies cannot intuitively display the causal driving relationship of oil and gas pipeline risks. Summary of the invention
[0004] The purpose of this application is to provide a pipeline risk knowledge graph construction method, device and storage medium, aiming to solve the problem of how to intuitively display the causal driving relationship of pipeline risks.
[0005] In order to achieve the above objectives, this application adopts the following technical solutions:
[0006] In a first aspect, the present application provides a method for constructing a pipeline risk knowledge graph, the method comprising: obtaining multiple groups of risk information in a pipeline risk text; a group of risk information includes multiple risk information elements with causal relationships, and the risk information elements include objects related to the pipeline and parameter information of the objects; for any target group risk information in the multiple groups of risk information, construct a sub-risk knowledge graph corresponding to the target group risk information to obtain a sub-risk knowledge graph corresponding to each group of risk information; multiple nodes in the sub-risk knowledge graph correspond one-to-one to multiple risk information elements in the target group risk information, and the connection relationship between the multiple nodes is determined according to the causal relationship between the multiple risk information elements; based on the sub-risk knowledge graphs corresponding to the multiple groups of risk information, construct a risk knowledge graph corresponding to the pipeline.
[0007] The pipeline risk knowledge graph construction method provided in the embodiment of the present application constructs a risk knowledge graph by establishing connection relationships between nodes based on the causal relationship between nodes, thereby being able to intuitively display the causal driving relationship of pipeline risks.
[0008] In a possible implementation, the risk information elements include one or more of the following: potential consequence elements; wherein the potential consequence is used to characterize risk events that exist in the pipeline; failure mode elements; wherein the failure mode is used to characterize risk events in which abnormalities in pipeline components affect the pipeline; potential cause elements; wherein the potential cause is used to characterize risk events in which operations performed on the pipeline affect the pipeline.
[0009] In a possible implementation, the parameter information of the failure mode element includes a risk index of the pipeline being in a failure mode; the risk index is determined based on the severity and probability of the failure mode when it occurs; the parameter information of the potential cause element includes a risk priority number of the potential cause; the risk priority number is determined based on the severity of the failure mode when it occurs, the probability of occurrence of the potential cause, and the detectability.
[0010] In a possible implementation, based on the sub-risk knowledge graphs corresponding to the multiple groups of risk information, a risk knowledge graph corresponding to the pipeline is constructed, including: determining the weight of each node in the risk knowledge graph in the sub-risk knowledge graph; based on the weight of each node in the risk knowledge graph, constructing the risk knowledge graph corresponding to the pipeline; wherein, the greater the weight of the node in the risk knowledge graph, the larger the area occupied by the node in the visualized risk knowledge graph.
[0011] In one possible implementation, the weight of each node in the risk knowledge graph is determined, including: determining the weight of each node in the risk knowledge graph according to parameter information of the risk information element corresponding to each node in the sub-risk knowledge graph.
[0012] In a possible implementation, multiple groups of risk information are obtained from a pipeline risk text, including: screening the pipeline risk text to obtain sentences including risk information elements; and annotating the sentences including risk information elements with semantic elements to obtain multiple groups of risk information.
[0013] In a second aspect, the present application provides a pipeline risk knowledge graph construction device, which includes: a communication unit and a processing unit; the communication unit is used to obtain multiple groups of risk information in the pipeline risk text; a group of risk information includes multiple risk information elements with causal relationships, and the risk information elements include objects related to the pipeline and parameter information of the objects; the processing unit is used to construct a sub-risk knowledge graph corresponding to any target group risk information in the multiple groups of risk information, so as to obtain a sub-risk knowledge graph corresponding to each group of risk information; multiple nodes in the sub-risk knowledge graph correspond one-to-one to multiple risk information elements in the target group risk information, and the connection relationship between the multiple nodes is determined according to the causal relationship between the multiple risk information elements; the processing unit is also used to construct a risk knowledge graph corresponding to the pipeline based on the sub-risk knowledge graphs corresponding to the multiple groups of risk information.
[0014] In a possible implementation, the risk information elements include one or more of the following: potential consequence elements; wherein the potential consequence is used to characterize risk events that exist in the pipeline; failure mode elements; wherein the failure mode is used to characterize risk events in which abnormalities in pipeline components affect the pipeline; potential cause elements; wherein the potential cause is used to characterize risk events in which operations performed on the pipeline affect the pipeline.
[0015] In a possible implementation, the parameter information of the failure mode element includes a risk index of the pipeline being in a failure mode; the risk index is determined based on the severity and probability of the failure mode when it occurs; the parameter information of the potential cause element includes a risk priority number of the potential cause; the risk priority number is determined based on the severity of the failure mode when it occurs, the probability of occurrence of the potential cause, and the detectability.
[0016] In a possible implementation, the processing unit is further used to determine the weight of each node in the risk knowledge graph in the sub-risk knowledge graph; the processing unit is further used to construct a risk knowledge graph corresponding to the pipeline based on the weight of each node in the risk knowledge graph; wherein, the greater the weight of the node in the risk knowledge graph, the larger the area occupied by the node in the visualized risk knowledge graph.
[0017] In a possible implementation, the processing unit is further used to determine the weight of each node in the risk knowledge graph based on parameter information of the risk information element corresponding to each node in the sub-risk knowledge graph.
[0018] In a possible implementation, the processing unit is further used to screen the pipeline risk text to obtain sentences including risk information elements; the processing unit is further used to perform semantic element annotation on the sentences including risk information elements to obtain multiple groups of risk information.
[0019] In a third aspect, the present application provides a pipeline risk knowledge graph construction device, the device comprising: a processor and a communication interface; the communication interface and the processor are coupled, and the processor is used to run a computer program or instruction to implement the pipeline risk knowledge graph construction method described in the first aspect and any possible implementation method of the first aspect.
[0020] In a fourth aspect, the present application provides a computer-readable storage medium, which stores instructions. When the instructions are executed on a terminal, the terminal executes the pipeline risk knowledge graph construction method described in the first aspect and any possible implementation of the first aspect.
[0021] In a fifth aspect, the present application provides a computer program product comprising instructions. When the computer program product is run on a pipeline risk knowledge graph construction device, the pipeline risk knowledge graph construction device executes the pipeline risk knowledge graph construction method described in the first aspect and any possible implementation method of the first aspect.
[0022] In a sixth aspect, the present application provides a chip, the chip comprising a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run a computer program or instruction to implement the pipeline risk knowledge graph construction method as described in the first aspect and any possible implementation method of the first aspect.
[0023] Specifically, the chip provided in the present application also includes a memory for storing computer programs or instructions. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0025] Figure 1 A schematic diagram of the composition of a pipeline risk knowledge graph construction device provided in an embodiment of the present application;
[0026] Figure 2 A flowchart of a pipeline risk knowledge graph construction method provided in an embodiment of the present application;
[0027] Figure 3 A schematic diagram of a sub-risk knowledge graph provided in an embodiment of the present application;
[0028] Figure 4 A flowchart of a pipeline risk knowledge graph construction method provided in an embodiment of the present application;
[0029] Figure 5 A schematic diagram of a pipeline risk text provided in an embodiment of the present application;
[0030] Figure 6 A schematic diagram of using a large language model to identify and annotate pipeline risk text provided in an embodiment of the present application;
[0031] Figure 7 A flowchart of a pipeline risk knowledge graph construction method provided in an embodiment of the present application;
[0032] Figure 8 A schematic diagram of a risk knowledge graph provided in an embodiment of the present application;
[0033] Fig. 9 A schematic diagram of another risk knowledge graph provided in an embodiment of the present application;
[0034] Fig.10 A schematic diagram of the structure of a pipeline risk knowledge graph construction device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0035] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0036] In the description of this application, it should be understood that the terms "upper", "lower", "left", "right", "front", "back", "inside", "outside", etc. indicate directions or positional relationships based on the directions or relative positional relationships shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on this application. Unless otherwise specified, the above-mentioned directional description can be flexibly set in the process of actual application under the condition that the relative positional relationship shown in the accompanying drawings is met.
[0037] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.
[0038] In the present application, the terms "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, article or device including the element.
[0039] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.
[0040] In the description of this specification, specific features, structures, materials or characteristics may be combined in an appropriate manner in any one or more embodiments or examples.
[0041] Oil and gas pipelines involve major security interests, so it is crucial to ensure their safety and stability and to provide early warning of safety risks. Oil and gas pipelines face a variety of risks, such as physical damage, leakage, and aging. In order to effectively identify and manage these risks, it is particularly important to build a pipeline risk knowledge graph. As an important cornerstone of artificial intelligence, the knowledge graph can provide systematic knowledge support for risk management.
[0042] At present, the risk knowledge graph in general technology is associated according to the spatial position relationship of entities in 3-tuples, rather than according to the transmission relationship or causal relationship of risks or failures, which is inconsistent with the definition of risks or failures. Risk itself is a 2-tuple, including entities and failure modes.
[0043] In addition, the data used in some risk knowledge graphs are accident event data that have already occurred, which can prevent the recurrence of accidents, but cannot predict possible risks in advance. Risks satisfy Murphy's Law, especially the occurrence of low-probability events.
[0044] In view of this, an embodiment of the present application provides a method for constructing a pipeline risk knowledge graph, the method comprising: obtaining multiple groups of risk information in a pipeline risk text, constructing a sub-risk knowledge graph corresponding to each group of risk information in the multiple groups of risk information, and constructing a risk knowledge graph corresponding to the pipeline based on all sub-risk knowledge graphs. Among them, the multiple nodes in the sub-risk knowledge graph correspond one-to-one to the multiple risk information elements in the target group risk information, and the connection relationship between the multiple nodes is determined based on the causal relationship between the multiple risk information elements. In other words, the pipeline risk knowledge graph construction method provided in the embodiment of the present application is to establish a connection relationship between nodes based on the causal relationship between the nodes, and thus can intuitively display the causal driving relationship of pipeline risks.
[0045] For example, Figure 1 The following is a schematic diagram of the composition of a pipeline risk knowledge graph construction device 10 provided in an embodiment of the present application. Figure 1 As shown, the pipeline risk knowledge graph construction device 10 may include a processor 101 and a bus 102.
[0046] Furthermore, the pipeline risk knowledge graph construction device 10 may also include a communication interface 103 and a memory 104. The processor 101, the memory 104 and the communication interface 103 may be connected via a bus 102.
[0047] The processor 101 is a central processing unit (CPU), a general processor, a network processor (NP), a digital signal processor (DSP), a microprocessor, a microcontroller, a programmable logic device (PLD), or any combination thereof. The processor 101 may also be other devices with processing functions, such as circuits, devices, or software modules, without limitation.
[0048] The bus 102 is used to transmit information between the components included in the pipeline risk knowledge graph construction device 10.
[0049] The communication interface 103 is used to communicate with other devices or other communication networks. The other communication networks may be Ethernet, radio access network (RAN), wireless local area network (WLAN), etc. The communication interface 103 may be a module, a circuit, a communication interface or any device capable of realizing communication.
[0050] The memory 104 is used to store instructions, where the instructions may be computer programs.
[0051] The memory 104 may be a read-only memory (ROM) or other types of static storage devices that can store static information and / or instructions, or a random access memory (RAM) or other types of dynamic storage devices that can store information and / or instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), magnetic disk storage media or other magnetic storage devices, etc., without limitation.
[0052] It should be noted that the memory 104 can exist independently of the processor 101, or can be integrated with the processor 101. The memory 104 can be used to store instructions or program codes or some data, etc. The memory 104 can be located in the pipeline risk knowledge graph construction device 10, or can be located outside the pipeline risk knowledge graph construction device 10, without limitation. The processor 101 is used to execute the instructions stored in the memory 104 to implement the pipeline risk knowledge graph construction method provided in the following embodiments of the present application.
[0053] In an example, the processor 101 may include one or more CPUs, for example, CPU0 and CPU1 (not shown in the figure).
[0054] As an optional implementation, the pipeline risk knowledge graph construction device 10 includes multiple processors.
[0055] As an optional implementation, the pipeline risk knowledge graph construction device 10 also includes an output device and an input device. For example, the input device is a keyboard, a mouse, a microphone, a joystick, and the output device is a display screen, a speaker, and the like.
[0056] It should be noted that the pipeline risk knowledge graph construction device 10 can be a desktop computer, a portable computer, a network server, a mobile phone, a tablet computer, a wireless terminal, an embedded device, a chip system or a computer with Figure 1 In addition, Figure 1 The composition shown in the Figure 1 The limitations of each device in Figure 1 In addition to the parts shown, Figure 1 The various devices in the drawings may include more or fewer components than shown, or combine certain components, or arrange the components differently.
[0057] In the embodiment of the present application, the chip system may be composed of a chip, or may include a chip and other discrete devices.
[0058] In addition, the actions, terms, etc. involved in the various embodiments of the present application can refer to each other without limitation. The message name or parameter name in the message exchanged between the various devices in the embodiments of the present application is only an example, and other names can also be used in the specific implementation without limitation.
[0059] The following is a description of the pipeline risk knowledge graph construction method provided by the embodiment of the present application in conjunction with the accompanying drawings. Among them, the actions, terms, etc. involved in the various embodiments of the present application can refer to each other without limitation. The message name or parameter name in the message exchanged between the various devices in the embodiment of the present application is only an example, and other names can also be used in the specific implementation without limitation. The actions involved in the various embodiments of the present application are only an example, and other names can also be used in the specific implementation, such as: "included in" in the embodiment of the present application can also be replaced by "carried on" or "carried in", etc.
[0060] like Figure 2 As shown, the embodiment of the present application proposes a pipeline risk knowledge graph construction method, which includes:
[0061] S201. Obtain multiple groups of risk information from the pipeline risk text.
[0062] A set of risk information includes a plurality of risk information elements with causal relationships, and the risk information elements include objects related to the pipeline and parameter information of the objects.
[0063] The above risk information elements include one or more of the following: potential consequence elements, failure mode elements or potential cause elements. Among them, potential consequence is used to characterize risk events that exist in the pipeline. Failure mode is used to characterize risk events that affect the pipeline due to abnormal components of the pipeline. Potential cause is used to characterize risk events that affect the pipeline due to operations performed on the pipeline.
[0064] Optionally, the parameter information of the above failure mode element includes a risk index of the pipeline being in failure mode. The risk index is determined according to the severity and probability of occurrence of the failure mode. For example, risk index = failure mode severity × failure mode occurrence probability.
[0065] Optionally, the parameter information of the potential cause element includes a risk priority number of the potential cause. The risk priority number is determined according to the severity of the failure mode when it occurs, the probability of occurrence of the potential cause, and the detectability. For example, risk priority number = failure mode severity × potential cause probability × potential cause detectability.
[0066] It should be noted that the above-mentioned failure mode severity, failure mode occurrence probability, failure mode severity, potential cause occurrence probability and potential cause detectability can be obtained through a preset accident event structured database. Specifically, read in the preset accident event structured database, and perform field mapping on the preset accident event structured database to obtain the above-mentioned parameters. In addition, if there are missing parameters, they can be filled in with manual scoring. Among them, the scoring principle is: severity 1-10 points, the more serious the loss, the higher the score. Probability of occurrence 1-10 points, the more frequent the occurrence, the higher the score. Detectability 1-10 points, the more difficult it is to detect, the higher the score.
[0067] In a possible implementation, pipeline risk text is obtained, and the pipeline risk text is processed according to a preset risk semantic structure to obtain multiple groups of risk information.
[0068] Exemplarily, the above-mentioned preset risk semantic structure can be a two-level semantic structure, specifically: [potential consequences: (object, parameter status)]-->[failure mode: (object, parameter status, severity, probability of occurrence, risk index)]-->[potential causes: (object, parameter status, probability of occurrence, detectability, risk priority number)]. Among them, the first layer is a 3-tuple structure from effect to cause of (potential consequences, failure mode, potential cause). The second layer is the object and parameter information included in the three risk information elements of potential consequences, failure mode and potential cause. The parameter information of potential consequences includes parameter status, the parameter information of failure mode includes parameter status, severity, probability of occurrence and risk index (risk index, RI), and the parameter information of potential causes includes parameter status, probability of occurrence, detectability and risk priority number (risk priority number, RPN).
[0069] It should be noted that the above preset risk semantic structure is based on the risk semantic structure defined by failure mode and effects analysis (FMEA). FMEA is a systematic risk assessment tool that helps the team take preventive measures by identifying potential failure modes and their effects. The FMEA semantic framework structures failure modes, effects, causes and their interrelationships to facilitate knowledge visualization and management. This framework provides a basis for the construction of the pipeline risk knowledge graph.
[0070] Although FMEA also includes three parts: failure mode, failure consequence and potential cause, the definition is relatively broad and non-standard, and only the RPN indicator of the potential cause is given, and only the severity or loss size is given for the failure mode. In view of this, the preset risk semantic structure defined in this application adds the indicator of the probability of occurrence of the failure mode required for risk assessment.
[0071] For example, Table 1 gives an exemplary description of a set of risk information. As shown in Table 1, the set of risk information includes three risk information elements. Each risk information element includes object and parameter information.
[0072] Table 1
[0073]
[0074] It should be noted that the data in Table 1 are obtained by manually labeling after statistically analyzing the quality data of enterprises, accident data, and expert experience. Its dynamic changes can be updated according to the actual collected data. For example, weather disasters can be updated according to real-time weather data, thereby adjusting the risk rating of the entire oil and gas pipeline network.
[0075] Furthermore, the data in Table 1 can be stored in a risk database, and the risk database can be continuously updated. In addition, when storing in the risk database, information such as proofreader or storage time can also be added.
[0076] Exemplarily, Table 2 shows the original sentences, annotated sentences and causal verbs corresponding to the risk information in Table 1 above.
[0077] Table 2
[0078]
[0079] S202. For any target group risk information among the multiple groups of risk information, construct a sub-risk knowledge graph corresponding to the target group risk information to obtain a sub-risk knowledge graph corresponding to each group of risk information.
[0080] Among them, multiple nodes in the sub-risk knowledge graph correspond one-to-one to multiple risk information elements in the target group risk information, and the connection relationship between the multiple nodes is determined according to the causal relationship between the multiple risk information elements.
[0081] In a possible implementation, a group of risk information is selected from multiple groups of risk information as target group risk information, multiple risk information elements in the target group risk information are visualized as multiple nodes, and the multiple nodes are connected according to the causal relationship between the risk information elements.
[0082] For example, Table 1 is used as the target group risk information. Figure 3 is the sub-risk knowledge graph corresponding to the target group risk information. Figure 3As shown in the figure, the potential consequence, failure mode and potential cause each correspond to a circular node, the node size is equal, and the text in the node is the object name and parameter status corresponding to the node. Among them, the text in the potential cause node is "Manufacturing and Application-Stress Concentration", the text in the failure mode node is "Weld, Bend-Corrosion Perforation", and the text in the potential consequence node is "Pipeline-Operation Years". The potential cause and failure mode are connected by an arrow, and the arrow points from the potential cause to the failure mode. The failure mode and potential consequence are connected by an arrow, and the arrow points from the failure mode to the potential consequence. Among them, the direction of the arrow indicates from cause to effect.
[0083] Optionally, a causal relationship between all risk information elements may be constructed based on multiple sets of risk information.
[0084] For example, Table 3 provides an exemplary description of the causal relationship between multiple risk information elements. As shown in Table 3, it includes two columns, cause and result, which can be the causal relationship between the failure mode and the potential result, or the causal relationship between the potential cause and the failure mode.
[0085] Table 3
[0086]
[0087] It should be noted that failure modes, potential results, and potential causes are all tuples of (object, parameter state). For example, "Manufacturing and Construction-Stress Concentration" in Table 3 indicates that the object of the potential cause is "Manufacturing and Construction" and the parameter state is "Stress Concentration".
[0088] It is understandable that the construction of the knowledge graph in the general technology is constructed by mining the 2-point 1-line relationship of the text, that is, the 3-tuple. The present application is constructed by mining the 3-point 2-line relationship of the text, that is, the 5-tuple. Among them, the 5-tuple includes potential consequences, failure modes, potential causes, the relationship between potential consequences and failure modes, and the relationship between failure modes and potential causes. In other words, the pipeline risk knowledge graph construction method provided in the embodiment of the present application can more accurately express the relationship between risks.
[0089] S203. Based on the sub-risk knowledge graphs corresponding to the multiple groups of risk information, a risk knowledge graph corresponding to the pipeline is constructed.
[0090] In one possible implementation, the weight of each node in the sub-risk knowledge graph in the risk knowledge graph is determined, and the area occupied by each node in the visualized risk knowledge graph is determined based on the weight of each node in the risk knowledge graph. The same nodes are merged to construct the risk knowledge graph corresponding to the pipeline. For details, please refer to the following Figure 7 The described embodiments are not described in detail here.
[0091] In the pipeline risk knowledge graph construction method provided in the present application, multiple groups of risk information in the pipeline risk text are obtained, a sub-risk knowledge graph corresponding to each group of risk information in the multiple groups of risk information is constructed, and a risk knowledge graph corresponding to the pipeline is constructed based on all the sub-risk knowledge graphs. Among them, the multiple nodes in the sub-risk knowledge graph correspond one-to-one to the multiple risk information elements in the target group risk information, and the connection relationship between the multiple nodes is determined according to the causal relationship between the multiple risk information elements. In other words, the pipeline risk knowledge graph construction method provided in the embodiment of the present application is to establish a connection relationship between nodes based on the causal relationship between the nodes, and thus can intuitively display the causal driving relationship of pipeline risks.
[0092] In one embodiment, Figure 4 As shown, the above S201 can be specifically determined through the following S401 to S402.
[0093] S401. Filter the pipeline risk text to obtain sentences including risk information elements.
[0094] In a possible implementation, keyword recognition is performed on the pipeline risk text according to a risk factor dictionary to obtain a sentence including risk information elements.
[0095] For example, the above pipeline risk text may be risk knowledge related to the pipeline, for example, relevant literature, such as research literature, production reports, etc.
[0096] For example, Figure 5 A schematic diagram of a pipeline risk text provided in an embodiment of the present application. Figure 5 As shown, this is a relatively long accident analysis report with a chapter structure, titled "Oil and Gas Pipeline Cracking Failure Analysis Case".
[0097] For example, Table 4 gives an exemplary description of a risk factor dictionary. As shown in Table 4, the risk factor dictionary includes a key and a value. The key represents an object name or a parameter state or a causal verb, and the value is a corresponding risk information element.
[0098] Table 4
[0099]
[0100]
[0101] It should be noted that the same keyword may belong to multiple semantic fields. For example, the “pipeline” in Table 4 can be the object name of potential impact, failure mode, and potential cause.
[0102] For example, the risk factor dictionary provided in Table 4 is used from Figure 5 The pipeline risk text is filtered and the sentences including risk information elements are obtained as follows:
[0103] The first sentence: "An oil and gas pipeline cracking accident occurred at a gas production plant, seriously affecting the work progress."
[0104] Second sentence: "In summary, the main reason for the cracking of this section of the pipeline is the special geographical location of this section of the pipeline, and the sudden increase in pipeline pressure after the well drilling operation caused the local bulging of the pipe body, which further led to the bursting of the pipeline."
[0105] Optionally, a similarity algorithm may be used to screen the pipeline risk text to find sentences similar to the preset example sentences as sentences including risk information elements. The preset example sentences may be sentences that have been determined to include risk information elements, such as the original sentences in Table 2 above.
[0106] S402: Annotate the sentences including risk information elements with semantic elements to obtain multiple groups of risk information.
[0107] In a possible implementation, the risk factor dictionary provided in Table 4 is used to perform semantic element annotation on sentences including risk information elements to obtain multiple groups of risk information.
[0108] Exemplarily, the risk factor dictionary provided in Table 4 is used to perform semantic element annotation on the sentence including the risk information element obtained in S401 above, and the annotated result is:
[0109] First sentence: "An accident occurred in a gas production plant / [cause and effect verb] oil and gas pipeline body / [failure mode-object] cracking / [failure mode-parameter state], which seriously affected / [cause and effect verb] the work progress / [potential impact-parameter state]."
[0110] Second sentence: "In summary, the main reason for the cracking of this section of the pipeline is the special geographical location of this section of the pipeline, and the well drilling operation / [potential cause-object] caused / [causal verb] a sudden increase in pipeline pressure / [potential cause-parameter state], causing / [causal verb] local expansion of the pipe body / [causal verb], which further led to / [causal verb] pipeline bursting."
[0111] It should be noted that the second sentence above contains multiple causal verbs, involving a response sequence. For such sentences, only one typical risk information element is selected, rather than segmenting the risk information element into each word. For example, this sentence only selects "well opening operation / [potential cause-object] and sudden increase in pressure / [potential cause-parameter status]" as the object and parameter status of the potential cause, which is more essential and representative. In this way, risk information element mining is performed at the sentence granularity, avoiding the unfair results caused by the length of the sentence due to the too fine segmentation granularity. In addition, since the same keyword may belong to multiple semantic fields, disambiguation processing is required when annotating semantic elements.
[0112] Further, the above-mentioned annotated results are sorted to obtain a set of risk information. By processing multiple pipeline risk texts, multiple sets of risk information can be obtained, and the multiple sets of risk information are stored in the risk database.
[0113] Optionally, a large language model may be used to screen and annotate pipeline risk texts. Specifically, a large language model (eg, Wenxinyiyan) is used to identify and annotate pipeline risk texts, and the large language model outputs candidate sentences and annotated sentences corresponding to the candidate sentences.
[0114] For example, Figure 6 A schematic diagram of using a large language model to identify and annotate pipeline risk text is provided in an embodiment of the present application. Figure 6 As shown, Figure 6 The a in it is the content of the input large language model. Figure 6 The b in it is the output result of the large language model.
[0115] Optional, above Figure 6 The candidate example sentences and the annotated example sentences used in a may be the original sentences and the annotated sentences in the aforementioned Table 2. The original sentences correspond to the candidate example sentences, and the annotated example sentences correspond to the annotated sentences.
[0116] Furthermore, after using the risk factor dictionary to identify and annotate the pipeline risk text, a large language model can be used for further screening and annotation, which can make up for the omission of candidate sentences due to the incomplete risk factor dictionary.
[0117] Furthermore, the structured data identified during the annotation process can be stored in the accident event structured database, and then stored in the risk database through field mapping.
[0118] In one embodiment, Figure 7 As shown, the above S203 can be specifically determined through the following S701 to S702.
[0119] S701. Determine the weight of each node in the sub-risk knowledge graph in the risk knowledge graph.
[0120] In one possible implementation, the weight of each node in the risk knowledge graph is determined based on the parameter information of the risk information element corresponding to each node in the sub-risk knowledge graph.
[0121] In one example, the weight of each node in the risk knowledge graph is determined according to the risk index included in the risk information element corresponding to each node in the sub-risk knowledge graph. The risk index is the RI in Table 1.
[0122] In another example, the weight of each node in the risk knowledge graph is determined according to the risk priority number included in the risk information element corresponding to each node in the sub-risk knowledge graph. The risk priority number is the RPN in Table 1.
[0123] It should be noted that since the parameter information of the potential consequence factor does not include RI or RPN, when visualizing the node corresponding to the potential consequence factor, the weight of the node in the risk knowledge graph uses the default value.
[0124] S702: Based on the weight of each node in the risk knowledge graph, construct a risk knowledge graph corresponding to the pipeline.
[0125] Among them, the greater the weight of the node in the risk knowledge graph, the larger the area occupied by the node in the visualized risk knowledge graph.
[0126] For example, taking the example of determining the weight of a node in a risk knowledge graph according to RI, Figure 8 A schematic diagram of a risk knowledge graph provided in an embodiment of the present application. Figure 8 As shown, the size of the circle corresponding to the node is determined according to the RI corresponding to the node. Among them, the largest node is the weld, bend-corrosion perforation node, indicating that the greatest risk comes from corrosion perforation of welds, elbows, etc. The pipeline-operation life node is both a cause and a result, indicating that the operation life of the pipeline is of great significance in pipeline risk management.
[0127] Understandably, Figure 8 It can visually display the risk level of pipelines in the oil and gas pipeline network, thereby improving the level of risk assessment and reducing the risk losses of the oil and gas pipeline network.
[0128] For example, taking the determination of the weight of a node in a risk knowledge graph according to RPN as an example, Fig. 9 A schematic diagram of a risk knowledge graph provided in an embodiment of the present application. Fig. 9As shown, the size of the circle corresponding to the node is determined according to the RPN corresponding to the node. Among them, the largest node is the iron-containing bacteria-bacterial corrosion node, indicating that the corrosion effect of iron-containing bacteria is the largest potential cause. Therefore, controlling the growth of iron-containing bacteria inside the oil and gas pipeline is an urgent measure to be taken in advance.
[0129] Understandably, Fig. 9 It can visually display the priority of risk prevention measures for pipelines in the oil and gas pipeline network, thereby improving the level of risk management and reducing risk losses in the oil and gas pipeline network.
[0130] In summary, this application aims at the situation where there are many failure modes and complex failure causes in complex pipeline networks, constructs a risk semantic structure for risk assessment and risk prevention based on FMEA, constructs a risk knowledge graph by mining text and data, and displays failure risk points and means of risk prevention through the risk knowledge graph.
[0131] It is understandable that the above-mentioned pipeline risk knowledge graph construction method can be implemented by a pipeline risk knowledge graph construction device. In order to realize the above-mentioned functions, the pipeline risk knowledge graph construction device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the modules and algorithm steps of each example described in the embodiments disclosed herein, the embodiments disclosed in this application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the embodiments disclosed in this application.
[0132] The embodiments disclosed in the present application can divide the functional modules of the pipeline risk knowledge graph construction device generated according to the above method example. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one processing module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules. It should be noted that the division of modules in the embodiments disclosed in the present application is schematic and is only a logical functional division. There may be other division methods in actual implementation.
[0133] Fig.10 A schematic diagram of a pipeline risk knowledge graph construction device provided in an embodiment of the present application. Fig.10 As shown, the pipeline risk knowledge graph construction device 100 can be used to perform Figure 2 , Figure 4 as well as Figure 7The pipeline risk knowledge graph construction method shown in the figure. The pipeline risk knowledge graph construction device 100 includes: a communication unit 1001 and a processing unit 1002.
[0134] The communication unit 1001 is used to obtain multiple groups of risk information in the pipeline risk text; one group of risk information includes multiple risk information elements with causal relationships, and the risk information elements include objects related to the pipeline and parameter information of the objects; the processing unit 1002 is used to construct a sub-risk knowledge graph corresponding to any target group risk information in the multiple groups of risk information, so as to obtain a sub-risk knowledge graph corresponding to each group of risk information; multiple nodes in the sub-risk knowledge graph correspond one-to-one to multiple risk information elements in the target group risk information, and the connection relationship between the multiple nodes is determined according to the causal relationship between the multiple risk information elements; the processing unit 1002 is also used to construct a risk knowledge graph corresponding to the pipeline based on the sub-risk knowledge graphs corresponding to the multiple groups of risk information.
[0135] In a possible implementation, the risk information elements include one or more of the following: potential consequence elements; wherein the potential consequence is used to characterize risk events that exist in the pipeline; failure mode elements; wherein the failure mode is used to characterize risk events in which abnormalities in pipeline components affect the pipeline; potential cause elements; wherein the potential cause is used to characterize risk events in which operations performed on the pipeline affect the pipeline.
[0136] In a possible implementation, the parameter information of the failure mode element includes a risk index of the pipeline being in a failure mode; the risk index is determined based on the severity and probability of the failure mode when it occurs; the parameter information of the potential cause element includes a risk priority number of the potential cause; the risk priority number is determined based on the severity of the failure mode when it occurs, the probability of occurrence of the potential cause, and the detectability.
[0137] In one possible implementation, processing unit 1002 is also used to determine the weight of each node in the sub-risk knowledge graph in the risk knowledge graph; processing unit 1002 is also used to construct a risk knowledge graph corresponding to the pipeline based on the weight of each node in the risk knowledge graph; wherein, the greater the weight of the node in the risk knowledge graph, the larger the area occupied by the node in the visualized risk knowledge graph.
[0138] In a possible implementation, the processing unit 1002 is further used to determine the weight of each node in the risk knowledge graph according to the parameter information of the risk information element corresponding to each node in the sub-risk knowledge graph.
[0139] In a possible implementation, the processing unit 1002 is further used to screen the pipeline risk text to obtain sentences including risk information elements; the processing unit 1002 is further used to perform semantic element annotation on the sentences including risk information elements to obtain multiple groups of risk information.
[0140] Through the description of the above implementation methods, technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.
[0141] The present disclosure also provides a computer-readable storage medium on which instructions are stored. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the pipeline risk knowledge graph construction method provided by the above-mentioned embodiment of the present disclosure.
[0142] The embodiments of the present disclosure also provide a computer program product including instructions, which, when executed on an electronic device, enables the electronic device to execute the pipeline risk knowledge graph construction method provided in the embodiments of the present disclosure.
[0143] Among them, the computer readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a register, a hard disk, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above, or any other form of computer readable storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (ASIC). In the embodiments of the present application, a computer-readable storage medium may be any tangible medium that contains or stores a program, which may be used by or in conjunction with an instruction execution system, apparatus, or device.
[0144] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.
Claims
1. A pipeline risk knowledge graph construction method, characterized in that: The method comprises: Acquire multiple groups of risk information in the pipeline risk text; a group of risk information includes multiple risk information elements with causal relationships, and the risk information elements include objects related to the pipeline and parameter information of the objects; For any target group risk information in the multiple groups of risk information, a sub-risk knowledge graph corresponding to the target group risk information is constructed to obtain a sub-risk knowledge graph corresponding to each group of risk information; the multiple nodes in the sub-risk knowledge graph correspond one-to-one to the multiple risk information elements in the target group risk information, and the connection relationship between the multiple nodes is determined according to the causal relationship between the multiple risk information elements; Based on the sub-risk knowledge graphs corresponding to the multiple groups of risk information, a risk knowledge graph corresponding to the pipeline is constructed.
2. The method according to claim 1, characterized in that The risk information elements include one or more of the following: Potential consequence elements; wherein the potential consequence is used to characterize the risk events existing in the pipeline; Failure mode elements; wherein the failure mode is used to characterize the risk event of the abnormality of the pipeline components affecting the pipeline; Potential cause elements; wherein the potential cause is used to characterize a risk event that an operation performed on the pipeline affects the pipeline.
3. The method according to claim 2, characterized in that The parameter information of the failure mode element includes a risk index of the pipeline being in a failure mode; the risk index is determined according to the severity and probability of occurrence of the failure mode; The parameter information of the potential cause factor includes the risk priority number of the potential cause; the risk priority number is determined according to the severity of the failure mode when it occurs, the probability of occurrence of the potential cause, and the detectability.
4. The method according to any one of claims 1 to 3, characterized in that The constructing the risk knowledge graph corresponding to the pipeline based on the sub-risk knowledge graphs corresponding to the multiple groups of risk information includes: Determining the weight of each node in the sub-risk knowledge graph in the risk knowledge graph; Based on the weight of each node in the risk knowledge graph, a risk knowledge graph corresponding to the pipeline is constructed; wherein, the greater the weight of the node in the risk knowledge graph, the larger the area occupied by the node in the visualized risk knowledge graph.
5. The method according to claim 4, characterized in that The determining the weight of each node in the sub-risk knowledge graph in the risk knowledge graph includes: According to the parameter information of the risk information element corresponding to each node in the sub-risk knowledge graph, the weight of each node in the risk knowledge graph is determined.
6. The method according to claim 1, characterized in that The obtaining of multiple sets of risk information in the pipeline risk text includes: Screening the pipeline risk text to obtain sentences including the risk information elements; Semantic element annotation is performed on the sentences including the risk information elements to obtain the multiple groups of risk information.
7. A pipeline risk knowledge graph construction device, characterized in that: The device comprises: a communication unit and a processing unit; The communication unit is used to obtain multiple groups of risk information in the pipeline risk text; a group of risk information includes multiple risk information elements with causal relationships, and the risk information elements include objects related to the pipeline and parameter information of the objects; The processing unit is used to construct a sub-risk knowledge graph corresponding to any target group risk information in the multiple groups of risk information, so as to obtain a sub-risk knowledge graph corresponding to each group of risk information; the multiple nodes in the sub-risk knowledge graph correspond to the multiple risk information elements in the target group risk information one-to-one, and the connection relationship between the multiple nodes is determined according to the causal relationship between the multiple risk information elements; The processing unit is further used to construct a risk knowledge graph corresponding to the pipeline based on the sub-risk knowledge graphs corresponding to the multiple groups of risk information.
8. A pipeline risk knowledge graph construction device, characterized in that: include: Memory and processor; The memory is coupled to the processor; The memory is used to store instructions executable by the processor; When the processor executes the instruction, the pipeline risk knowledge graph construction method according to any one of claims 1-6 is performed.
9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed on a computer, the computer executes the pipeline risk knowledge graph construction method according to any one of claims 1 to 6.
10. A computer program product, characterized in that The computer program product includes computer program instructions, and when the computer program instructions are executed by a processor, the pipeline risk knowledge graph construction method according to any one of claims 1 to 6 is implemented.