Construction update and fault handling method of intelligent operation and maintenance-oriented fault knowledge base

Abstract

The application provides a fault knowledge base construction updating and fault processing method for intelligent operation and maintenance, relates to data processing technology, and comprises the following steps: obtaining a data acquisition request; the data acquisition request comprises a part data identifier of an electronic device; according to the data acquisition request, part data corresponding to the part identifier is acquired. The part data is subjected to fault analysis and processing, and a fault sample library of the electronic device is determined. Based on preset node definition information of a knowledge graph, the node type label of each part and the fault mode information of each part is determined, and a target fault library is generated according to the node type label, so that the electronic device is subjected to fault diagnosis according to the target fault library and diagnosis result information is obtained. The method of the application can comprehensively show the structure composition of intelligent equipment and the mutual relationship between fault modes, provides a basis for subsequent fault diagnosis, and solves the technical problem of low accuracy of the diagnosis result of the fault of the intelligent equipment.

Description

Technical Field

[0001] This application relates to data processing technology, and more particularly to a method for constructing, updating and handling fault knowledge bases for intelligent operation and maintenance. Background Technology

[0002] Currently, many complex tasks in real life often require the cooperation of intelligent equipment and humans or the intelligent equipment itself to complete. Therefore, it is necessary to ensure the reliability and lifespan of intelligent equipment based on a fault sample database.

[0003] In existing technologies, the construction of fault sample libraries mainly employs methods based on expert experience, data, and models. However, these methods have the following problems: taking mechanical structure objects in intelligent equipment as an example, their fault modes and mechanisms are diverse, making it difficult to comprehensively express and demonstrate them; in actual use, the probability of fault occurrence is low, and degradation characteristics are not obvious. Therefore, it is necessary to determine a fault sample library capable of diagnosing the diagnostic results of intelligent equipment to improve the accuracy of the diagnostic results.

[0004] Therefore, there is an urgent need for a method that can accurately determine a fault sample library suitable for diagnosing smart devices. Summary of the Invention

[0005] This application provides a method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, in order to solve the technical problem of low accuracy in fault diagnosis results of intelligent devices.

[0006] Firstly, this application provides a method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, including:

[0007] Obtain a data acquisition request; wherein the data acquisition request includes component data identifiers of the electronic device; and obtain component data corresponding to the component identifiers according to the data acquisition request;

[0008] The component data is subjected to fault analysis processing to determine the fault sample library of the electronic device; wherein, the fault sample library includes fault mode information corresponding to each component in the electronic device;

[0009] Based on the node definition information of the preset knowledge graph, the node type label of each component and the fault mode information of each component is determined, and a target fault library is generated according to the node type label, so as to realize the fault diagnosis of electronic equipment and obtain the diagnosis result information based on the target fault library.

[0010] Furthermore, the component data includes component list information, preset component rule information, design data for each component, and test data.

[0011] Furthermore, the step of performing fault analysis processing on the component data to determine the fault sample library of the electronic device includes:

[0012] The component list information and preset component rule information are subjected to fault analysis and processing to determine the composition relationship between multiple components in the electronic device and the first fault mode information corresponding to each of the multiple components.

[0013] Based on the compositional relationships between the multiple components, a component composition library in the fault sample library is generated; and based on the first fault mode information corresponding to each of the multiple components, a fault mode library in the fault sample library is generated.

[0014] The design data and test data of each component are subjected to fault analysis processing to determine the second fault mode information of each component;

[0015] Based on the second fault mode information of each component, a data feature library is generated in the fault sample library.

[0016] Furthermore, the step of performing fault analysis processing on the design data and test data of each component to determine the second fault mode information of each component includes:

[0017] Feature extraction processing is performed on the design data and test data of each component to determine the data characteristics of each component;

[0018] The data characteristics of each component are determined as the second fault mode information of each component.

[0019] Furthermore, the method also includes:

[0020] Based on the compositional relationships between components in the component composition library, the first fault mode information corresponding to each component in the fault mode library, and the second fault mode information corresponding to each component in the data feature library, a first mapping relationship between the components in the component composition library and the first fault mode information in the fault mode library, and / or a second mapping relationship between the components in the component composition library and the second fault mode information in the data feature library are determined.

[0021] Furthermore, the node type label for determining each component and its fault mode information based on the preset knowledge graph node definition information includes:

[0022] Based on the node definition information of the preset knowledge graph, the first mapping relationship, and the second mapping relationship, the first node type label of each component and the second node type label of the fault mode information of each component are determined.

[0023] Further, generating the target fault database based on the node type label includes:

[0024] Import the component and / or the failure mode information of the component corresponding to the first node type label, and the component and / or the failure mode information of the component corresponding to the second node type label into the component node in the preset knowledge graph;

[0025] The hierarchical relationship between component nodes is determined based on the first node type label and the second node type label;

[0026] Based on the hierarchical relationships between the component nodes, a target fault database is generated.

[0027] Secondly, this application provides a device for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, comprising:

[0028] An acquisition unit is configured to acquire a data acquisition request; wherein the data acquisition request includes a component data identifier of an electronic device; and acquire component data corresponding to the component identifier according to the data acquisition request;

[0029] An analysis unit is used to perform fault analysis processing on the component data to determine the fault sample library of the electronic device; wherein, the fault sample library includes fault mode information corresponding to each component in the electronic device;

[0030] The first determining unit is used to determine the node type label of each component and the fault mode information of each component based on the node definition information of the preset knowledge graph.

[0031] The generation unit is used to generate a target fault database based on the node type label, so as to realize fault diagnosis of electronic devices and obtain diagnostic result information based on the target fault database.

[0032] Furthermore, the component data includes component list information, preset component rule information, design data for each component, and test data.

[0033] Furthermore, the analysis unit includes:

[0034] The first analysis module is used to perform fault analysis processing on the component list information and the preset component rule information to determine the composition relationship between multiple components in the electronic device and the first fault mode information corresponding to each of the multiple components.

[0035] The first determining module is used to determine the component composition library in the fault sample library based on the composition relationship between the plurality of components; and to determine the fault mode library in the fault sample library based on the first fault mode information corresponding to each of the plurality of components.

[0036] The second analysis module is used to perform fault analysis processing on the design data and test data of each component to determine the second fault mode information of each component.

[0037] The second determining module is used to determine the data feature library in the fault sample library based on the second fault mode information of each component.

[0038] Furthermore, the second analysis module includes:

[0039] The extraction submodule is used to perform feature extraction processing on the design data and test data of each component to determine the data features of each component;

[0040] The determination submodule is used to determine the data characteristics of each component as the second fault mode information of each component.

[0041] Furthermore, the device also includes:

[0042] The second determining unit is used to determine, based on the composition relationship between the components in the component composition library, the first fault mode information corresponding to each component in the fault mode library, and the second fault mode information corresponding to each component in the data feature library, a first mapping relationship between the components in the component composition library and the first fault mode information in the fault mode library, and / or a second mapping relationship between the components in the component composition library and the second fault mode information in the data feature library.

[0043] Furthermore, the first determining unit is specifically used for:

[0044] Based on the node definition information of the preset knowledge graph, the first mapping relationship, and the second mapping relationship, the first node type label of each component and the second node type label of the fault mode information of each component are determined.

[0045] Further, the generation unit includes:

[0046] The import module is used to import the component and / or the failure mode information of the component corresponding to the first node type label and the component and / or the failure mode information of the component corresponding to the second node type label into the component node in the preset knowledge graph.

[0047] The third determining module is used to determine the subordinate relationship between each component node based on the first node type label and the second node type label;

[0048] The generation module is used to generate a target fault library based on the hierarchical relationships between the component nodes.

[0049] Thirdly, this application provides a server, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the method described in the first aspect.

[0050] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the method described in the first aspect.

[0051] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0052] This application provides a method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance. The method involves obtaining a data acquisition request, which includes component data identifiers for electronic devices. Based on the data acquisition request, component data corresponding to the component identifiers is acquired. Fault analysis processing is performed on the component data to determine a fault sample library for the electronic device. This fault sample library includes fault mode information corresponding to each component in the electronic device. Based on preset knowledge graph node definition information, node type labels are determined for each component and its fault mode information. A target fault library is generated based on these node type labels to enable fault diagnosis of the electronic device and obtain diagnostic results. In this solution, a data acquisition request is first obtained, and based on the request, component data corresponding to the component identifiers is acquired. Then, fault analysis processing is performed on the component data to determine the fault sample library for the electronic device. Because the pre-defined node definition information of the knowledge graph defines the node type labels to which each node belongs, the server can determine the node type label of each component and the node type label of each component's fault mode information based on the pre-defined node definition information. Then, based on the node type labels of each component and the node type labels of each component's fault mode information, a target fault library is generated, which includes the correspondence between each node type label. Furthermore, the server can perform fault diagnosis on the equipment and obtain diagnostic results based on the target fault library generated on the server. Therefore, it is possible to determine the relationship between the occurrence of faults and the structural hierarchy of intelligent equipment during the initial design stage of intelligent equipment, generating a target fault library including the correspondence between each node type label. The generation method is simple and efficient, comprehensively displaying the structural composition of intelligent equipment and the interrelationships between fault modes, providing a basis for subsequent fault diagnosis, and solving the technical problem of low accuracy in fault diagnosis results of intelligent equipment. Attached Figure Description

[0053] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0054] Figure 1 A flowchart illustrating a method for constructing, updating, and handling faults in an intelligent operation and maintenance system, as provided in an embodiment of this application.

[0055] Figure 2 A flowchart illustrating another method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, provided in an embodiment of this application.

[0056] Figure 3A flowchart illustrating another method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, provided in an embodiment of this application.

[0057] Figure 4 A flowchart illustrating a knowledge graph in a target fault database, provided as an embodiment of this application;

[0058] Figure 5 A flowchart illustrating the relationship between a target fault database and a knowledge graph, provided as an embodiment of this application;

[0059] Figure 6 This application provides a schematic diagram of an information collection process.

[0060] Figure 7 A flowchart illustrating another method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, provided in an embodiment of this application;

[0061] Figure 8 A schematic diagram of a device for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, provided in an embodiment of this application;

[0062] Figure 9 A schematic diagram of another device for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, provided in an embodiment of this application;

[0063] Figure 10 This is a schematic diagram of the structure of a server provided in an embodiment of this application.

[0064] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0065] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure.

[0066] Currently, with the development of science and technology, the application of intelligent equipment has become widespread. Many complex tasks in real life often require intelligent equipment to work in conjunction with humans or to be completed independently. In such situations, the requirements for the reliability and lifespan of intelligent equipment are becoming increasingly stringent. This is especially true for intelligent equipment operating in extreme environments such as deep space and the nuclear industry, where operating conditions are harsh, maintenance and support are difficult, and functional structures are complex. Intelligent equipment is highly susceptible to functional or performance degradation, and may even malfunction with serious consequences. Considering that intelligent equipment possesses intelligent functions such as perception, decision-making, and execution, when performance degradation, exceeding tolerances, or even malfunctions occur, its own intelligent maintenance system can predict and handle these situations. However, constructing a fault sample database remains a significant challenge.

[0067] In one example, the construction of the fault sample library mainly employs three approaches: expert experience-based, data-driven, and model-driven. The expert experience-based approach constructs the fault sample library using an intelligent expert system designed based on long-term practical experience and a large amount of fault information. The data-driven approach processes the acquired fault data through information processing and feature extraction methods to obtain fault sample information. The model-driven approach compares the measurable information of the intelligent equipment with consistent information expressed by a model, analyzes and processes the residuals, and thus obtains fault sample information. However, the above methods have the following problems: taking mechanical structure objects in intelligent equipment as an example, their fault modes and mechanisms are diverse, making it difficult to comprehensively express and demonstrate them; the probability of fault occurrence is low in actual use, and degradation characteristics are not obvious; typically, the development cycle is long, the cost of conducting tests is high, and it is difficult to obtain test data; the data acquired by sensors during the testing phase simultaneously contains degradation information, fault information, noise, etc., making it difficult to extract useful information; and the extracted information is difficult to use for fault location.

[0068] As the foundation for intelligent equipment analysis, the fault sample library can provide a technical basis and basic samples for subsequent research on fault diagnosis, experimental verification, etc. How to generate and construct the fault sample library is currently a research hotspot and a research challenge.

[0069] This application provides a method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, aiming to solve the above-mentioned technical problems in the prior art.

[0070] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0071] Figure 1This application provides a flowchart illustrating a method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, as well as a fault handling method, as shown in the embodiments below. Figure 1 As shown, the method includes:

[0072] Step 101: Obtain a data acquisition request; wherein, the data acquisition request includes the component data identifier of the electronic device; and according to the data acquisition request, obtain the component data corresponding to the component identifier.

[0073] For example, the execution entity in this embodiment can be a server. First, a data acquisition request needs to be obtained. When collecting information for the electronic device to be analyzed, the server can respond to the user's request operation and obtain a data acquisition request, or receive a data acquisition request transmitted by other devices. The data acquisition request includes component data identifiers of the electronic device, wherein the electronic device includes intelligent equipment, etc. The server can obtain the component data corresponding to the component identifier based on the component identifier in the data acquisition request. For example, the component data includes knowledge information such as component list information and preset component rule information, as well as design data and test data of each component. The component list information includes multiple components that make up the electronic device, and the multiple components include components of the same type but different parameters as each component. The preset component rule information includes preset expert experience, which is component rule information summarized from experiments. The design data includes parameters of the components that make up the electronic device, and the test data includes parameter change data of the components during the experiment.

[0074] Step 102: Perform fault analysis on the component data to determine the fault sample library of the electronic equipment; wherein, the fault sample library includes fault mode information corresponding to each component in the electronic equipment.

[0075] For example, the server performs fault analysis on the component list information, preset component rule information, design data for each component, and test data to determine the compositional relationships between multiple components in the electronic device, the first fault mode information corresponding to each of the multiple components, and the second fault mode information for each component. Based on the compositional relationships between the multiple components, a component composition library is generated in the fault sample library; based on the first fault mode information corresponding to each of the multiple components, a fault mode library is generated in the fault sample library; and based on the second fault mode information for each component, a data feature library is generated in the fault sample library.

[0076] For example, Figure 2 A flowchart illustrating another method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance provided in this application embodiment is shown below. Figure 2As shown, this includes the correspondence between the structural and functional levels of the collected information. The structural level includes: electronic device definition layer, subsystem definition layer, device definition layer, component definition layer, and part definition layer. The functional level includes: part function definition layer, component function definition layer, device function definition layer, subsystem function definition layer, and electronic device function definition layer. Specifically, the functional level includes: electronic device + electronic device code, multiple subsystems + subsystem codes, devices constituting a subsystem + device codes, components constituting a device + component codes, and parts constituting a component + part codes. The functional level includes: part code + part function, component code + component function, device code + device function, subsystem code + subsystem function, and electronic device code + electronic device function. Pre-defined part rule information, part list information, and other knowledge information are typically stored as character-based information, primarily for fault mode analysis. Figure 2 The compositional relationship between the bottom layer (components) and the top layer (electronic devices) shown in the diagram is analyzed. This analysis includes examining the component composition relationships from the bottom layer to the top layer and troubleshooting, resulting in a component composition library PT = {PT1, PT2, ..., PT...}. m} and the fault mode library FM = {FM1, FM2, ..., FM n The database contains information on m components and n primary failure modes. The primary failure mode information in the failure mode library should include types such as fracture, crack, pitting, plastic deformation, torsional deformation, indentation, aging, spalling, abnormal wear, loosening, improper stroke, improper clearance, interference, functional failure, performance degradation, and performance exceeding limits.

[0077] For design data, test data, and other data information of similar components, floating-point information is typically used for storage. The main tasks involve data feature extraction and processing. Specifically, the server extracts data features from the design data and test data of each component, determines the data features of each component as the second fault mode information, and generates a data feature library DC based on the second fault mode information. The data feature library DC = {DC1, DC2, ..., DC...} p} contains a total of p data features.

[0078] It should be noted that this embodiment uses the Fourier method to extract features from fault data. Wavelet analysis, artificial neural networks, deep learning, and other methods can also be used. There is no limitation on these methods, as the purpose is to achieve feature extraction from fault data.

[0079] Step 103: Based on the node definition information of the preset knowledge graph, determine the node type label of each component and the fault mode information of each component, and generate a target fault library according to the node type label, so as to realize the fault diagnosis of electronic equipment and obtain the diagnosis result information based on the target fault library.

[0080] For example, the preset knowledge graph node definition information defines the node type label to which each node belongs. Therefore, based on the preset knowledge graph node definition information, the server can determine the node type label of each component and the node type label of each component's fault mode information. Based on the node type labels of each component and the node type labels of each component's fault mode information, a target fault database is generated. Then, based on the target fault database generated on the server, the server can perform fault diagnosis on the equipment and obtain diagnostic results. Therefore, by using a knowledge graph, the relationships between components, faults, and data are comprehensively expressed in graph form, and it is easy to implement search functions through software.

[0081] It should be noted that this embodiment uses Neo4j software to construct the fault sample library. Other database software such as MySQL and ArangoDB can also be used, with the same goal of constructing a fault sample library. Depending on the application, target fault libraries can be generated for different analysis objects, providing a basis for subsequent fault diagnosis and other analytical work.

[0082] In this embodiment, a data acquisition request is obtained; wherein the data acquisition request includes component data identifiers of the electronic device; and according to the data acquisition request, component data corresponding to the component identifiers is acquired. Fault analysis processing is performed on the component data to determine a fault sample library for the electronic device; wherein the fault sample library includes fault mode information corresponding to each component in the electronic device. Based on the node definition information of a preset knowledge graph, node type labels for each component and the fault mode information of each component are determined, and a target fault library is generated according to the node type labels; so as to realize fault diagnosis of the electronic device based on the target fault library and obtain diagnostic result information. In this scheme, a data acquisition request is first obtained, and according to the data acquisition request, component data corresponding to the component identifiers is acquired. Then, fault analysis processing is performed on the component data to determine the fault sample library for the electronic device. Because the pre-defined node definition information of the knowledge graph defines the node type labels to which each node belongs, the server can determine the node type label of each component and the node type label of each component's fault mode information based on the pre-defined node definition information. Then, based on the node type labels of each component and the node type labels of each component's fault mode information, a target fault library is generated, which includes the correspondence between each node type label. Furthermore, the server can perform fault diagnosis on the equipment and obtain diagnostic results based on the target fault library generated on the server. Therefore, it is possible to determine the relationship between the occurrence of faults and the structural hierarchy of intelligent equipment during the initial design stage of intelligent equipment, generating a target fault library including the correspondence between each node type label. The generation method is simple and efficient, realizing the organization, analysis, and summarization of knowledge and data information. It can comprehensively display the structural composition of intelligent equipment and the interrelationships between fault modes, providing a basis for subsequent fault diagnosis and solving the technical problem of low accuracy in fault diagnosis results of intelligent equipment.

[0083] Figure 3 A flowchart illustrating another method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance provided in this application embodiment is shown below. Figure 3 As shown, the method includes:

[0084] Step 201: Obtain a data acquisition request; wherein, the data acquisition request includes the component data identifier of the electronic device; and according to the data acquisition request, obtain the component data corresponding to the component identifier.

[0085] In one example, the component data includes component list information, preset component rule information, design data for each component, and test data.

[0086] For example, this step can be referred to Figure 1 Step 101 in the text will not be repeated here.

[0087] Step 202: Perform fault analysis on the component list information and preset component rule information to determine the composition relationship between multiple components in the electronic device and the first fault mode information corresponding to each of the multiple components.

[0088] For example, such as Figure 2 As shown, the processing flow for knowledge information such as parts list information and preset parts rule information mainly includes the following steps:

[0089] (1) Define the corresponding conventions for components and functions; whereby the hierarchy includes structural hierarchy and functional hierarchy.

[0090] (2) Establish a coding system for components and failure modes; the coding system includes electronic equipment coding, subsystem coding, equipment coding, component coding and component coding, etc.

[0091] (3) Determine the task description and functional process analysis;

[0092] (4) Determine the fault mode information;

[0093] (5) Perform fault analysis and handling;

[0094] (6) Output the correspondence between structural analysis level, functional analysis level and fault mode;

[0095] Furthermore, based on the structural analysis level, functional analysis level, and fault mode correspondence, the compositional relationship between multiple components in an electronic device, as well as the first fault mode information corresponding to each of the multiple components, can be determined.

[0096] Step 203: Determine the component composition library in the fault sample library based on the composition relationship between multiple components; and determine the fault mode library in the fault sample library based on the first fault mode information corresponding to each of the multiple components.

[0097] For example, this step can be referred to Figure 1 Step 102 in the text will not be repeated here.

[0098] Step 204: Perform fault analysis and processing on the design data and test data of each component to determine the second fault mode information of each component.

[0099] In one example, feature extraction processing is performed on the design data and test data of each component to determine the data characteristics of each component; the data characteristics of each component are determined as the second failure mode information of each component.

[0100] For example, based on design data and test data for each component, information such as component health and lifespan prediction can be calculated. To determine component health, the fault analysis and processing flow based on design data and test data for each component mainly includes the following steps:

[0101] ①Signal preprocessing

[0102] Since the extracted signal contains noise, it is first denoised, for example, by using the Fourier transform method; then the denoised signal is normalized.

[0103] ②Distance determined

[0104] Various vector distance or similarity calculations are performed on the collected actual signals and the preset health signals. The health status is diagnosed by comparing the distance or similarity between the vector to be evaluated and the standard health vector. Commonly used calculation methods include Euclidean distance, Mahalanobis distance, and cosine similarity.

[0105] ③ Health Calculation

[0106] The health status of components is directly calculated based on the distance calculation results. Health status is generally represented by a range of 0 to 100, where 0 indicates a faulty state and 100 indicates an excellent health state.

[0107] To obtain life prediction data for components, the fault analysis and processing flow based on design data and test data for each component mainly includes the following steps:

[0108] ① Determine the fault mode information and its corresponding mechanism

[0109] Based on the failure characteristics of the current equipment, conduct fault analysis, identify the main failure modes and corresponding mechanisms of the research object, and determine the failure mechanism according to the environmental factors and load conditions of the current equipment.

[0110] ② Select physical and mathematical models to describe the mechanism.

[0111] Based on the current environmental factors and load conditions of the equipment, a suitable physical model is determined, and the physical model is then mathematically processed.

[0112] ③ Threshold determination

[0113] Based on the fault characteristics, and with reference to the current equipment design standards and the system's established fault standards, the threshold for life calculation is determined.

[0114] ④ Lifespan prediction

[0115] Based on the determined analysis model and threshold, a corresponding life reliability model is established to complete life prediction and obtain life prediction data for components.

[0116] Therefore, the final output format of the health and life expectancy prediction data is shown in Table 1 below:

[0117] Table 1

[0118]

[0119] Step 205: Determine the data feature library in the fault sample library based on the second fault mode information of each component.

[0120] For example, this step can be referred to Figure 1 Step 102 in the text will not be repeated here.

[0121] Step 206: Based on the composition relationship between components in the component composition library, the first fault mode information corresponding to each component in the fault mode library, and the second fault mode information corresponding to each component in the data feature library, determine the first mapping relationship between the components in the component composition library and the first fault mode information in the fault mode library, and / or the second mapping relationship between the components in the component composition library and the second fault mode information in the data feature library.

[0122] For example, based on the composition relationship between components in the component composition library, the first fault mode information corresponding to each component in the fault mode library, and the second fault mode information corresponding to each component in the data feature library, the server can determine a first mapping relationship between the components in the component composition library and the first fault mode information in the fault mode library, and / or a second mapping relationship between the components in the component composition library and the second fault mode information in the data feature library.

[0123] Therefore, based on the first and second mapping relationships mentioned above, a fault framework mapping relationship for the current equipment can be constructed, forming a fault sample framework for the current equipment. The fault sample framework includes components corresponding to at least one first fault mode information, or components corresponding to at least one second fault mode information, or components corresponding to at least one first fault mode information and at least one second fault mode information. Taking a fault sample framework where components correspond to at least one first fault mode information as an example, the fault sample framework... Represented as:

[0124] in, It contains 2 n -1 type of first fault mode information.

[0125] Step 207: Based on the node definition information, first mapping relationship and second mapping relationship of the preset knowledge graph, determine the first node type label of each component and the second node type label of the fault mode information of each component.

[0126] For example, the preset knowledge graph node definition information defines the node type label to which the node belongs. The node is a triple <head entity, relation, tail entity>, where the head entity and tail entity are the data pointed to by the two ends of the first mapping relation and the two ends of the second mapping relation, respectively. The head entity is the starting node, the tail entity is the ending node pointed to by the head entity, and the relation is the semantic relationship between the head entity and the tail entity. The server determines the first node type label of each component and the second node type label of the fault mode information of each component according to the knowledge graph node definition information, the first mapping relation, and the second mapping relation, thereby obtaining the fault triple <head entity, relation, tail entity>.

[0127] For example, the essence of a knowledge graph is a structured semantic knowledge base and a graph-based knowledge representation method used to symbolically describe concepts and relationships in the real world. Typically, a knowledge graph is a semantic graph composed of multiple nodes and edges. Entities represent fault mode information or components; fault mode information includes data features from a data feature library or fault descriptions from a fault mode library. Edges represent semantic relationships between two nodes. After the server performs fault analysis on the current equipment, it forms corresponding component nodes, fault description nodes, and data feature nodes, which serve as input information for the knowledge graph.

[0128] Step 208: Import the component and / or component failure mode information corresponding to the first node type label and the component and / or component failure mode information corresponding to the second node type label into the component node in the preset knowledge graph.

[0129] For example, the server imports the component and / or component failure mode information corresponding to the first node type label and the component and / or component failure mode information corresponding to the second node type label into the component node in the preset knowledge graph.

[0130] For example, the failure mode information of a component includes the first failure mode information in the failure mode library and the second failure mode information in the data feature library. The first failure mode information is the failure description information of the component, and the second failure mode information is the data feature information of the component. The server imports the component nodes containing component information, the failure description nodes containing failure description information, and the data feature nodes containing data feature information as node type labels into the component nodes in the knowledge graph.

[0131] Step 209: Determine the dependency relationship between each component node based on the first node type label and the second node type label.

[0132] For example, the server connects component nodes in the knowledge graph hierarchically according to the first node type label and the second node type label, and determines the subordinate relationship between the component nodes.

[0133] Step 210: Generate the target fault database based on the hierarchical relationship between each component node.

[0134] For example, the server can determine the backbone structure of the knowledge graph based on the hierarchical relationships between the component nodes. Furthermore, for faulty triples, the server projects the entity relation to its corresponding relation space, constructs a representation model, removes redundant nodes and relations, and forms a complete knowledge graph. Redundant nodes and relations include bidirectional relationships, etc., which are not restricted; for bidirectional relationships, one direction of the pointer can be randomly deleted.

[0135] For example, Figure 4 This application provides a flowchart illustrating a knowledge graph in a target fault database, as shown in the embodiments below. Figure 4 As shown, it includes: electronic devices, controllers, motion sensors, perception sensors, etc., where bold circles represent structural entities, light-colored circles represent functional entities, dashed circles represent fault entities, and horizontal lines between circles define the relationship between the two circles.

[0136] Figure 5 This application provides a flowchart illustrating the relationship between a target fault database and a knowledge graph, as shown in the embodiments of the present application. Figure 5 As shown, this includes the directional relationship between the target fault database and the knowledge graph. Specifically, the target fault database includes: serial number, component fault mode information, system fault mode, monitoring signals (features), health status, and predicted lifespan. The knowledge graph includes: component name (i.e., entity 1, entity 2, entity 3, etc.), health status (i.e., entity 1, entity 2, entity 3, etc.), fault mode information (i.e., entity 1, entity 2, entity 3, etc.), and predicted lifespan (i.e., entity 1, entity 2, entity 3, etc.). When performing fault diagnosis based on the target sample database, the knowledge graph pointed to by the target sample database is invoked, and the diagnostic result information of the electronic device is determined according to the pointing relationship in the invoked knowledge graph.

[0137] In this embodiment, a data acquisition request is obtained; wherein the data acquisition request includes component data identifiers of the electronic device; and according to the data acquisition request, component data corresponding to the component identifiers is acquired. Fault analysis processing is performed on the component list information and preset component rule information to determine the compositional relationships between multiple components in the electronic device, and the first fault mode information corresponding to each of the multiple components. Based on the compositional relationships between the multiple components, a component composition library in the fault sample library is determined; and based on the first fault mode information corresponding to each of the multiple components, a fault mode library in the fault sample library is determined. Fault analysis processing is performed on the design data and test data of each component to determine the second fault mode information of each component. Based on the second fault mode information of each component, a data feature library in the fault sample library is determined. Based on the compositional relationships between components in the component composition library, the first fault mode information corresponding to each component in the fault mode library, and the second fault mode information corresponding to each component in the data feature library, a first mapping relationship between the components in the component composition library and the first fault mode information in the fault mode library, and / or a second mapping relationship between the components in the component composition library and the second fault mode information in the data feature library, is determined. Based on the node definition information, first mapping relationship, and second mapping relationship of the pre-defined knowledge graph, a first node type label and a second node type label for the fault mode information of each component are determined. The component and / or its fault mode information corresponding to the first node type label and the component and / or its fault mode information corresponding to the second node type label are imported into the component nodes of the pre-defined knowledge graph. The dependency relationships between component nodes are determined according to the first and second node type labels. A target fault library is generated based on the dependency relationships between component nodes. Therefore, the relationship between the occurrence of faults and the structural hierarchy of intelligent equipment can be determined in the initial design stage of intelligent equipment, generating a target fault library including the correspondence between node type labels. The generation method is simple and efficient, comprehensively displaying the structural composition of intelligent equipment and the interrelationships between fault modes, providing a basis for subsequent fault diagnosis, and solving the technical problem of low accuracy in fault diagnosis results for intelligent equipment. Furthermore, during the generation process, qualitative and quantitative analysis are combined to construct a target fault database, which provides fault mode information, fault impact, and characteristic quantities. In addition, indicators such as the health of electronic devices are expressed as attributes of components, providing a data and analysis basis for subsequent fault diagnosis and life prediction.

[0138] For example, Figure 6 This application provides a schematic diagram of an information collection process, such as... Figure 6As shown, it includes: electronic device knowledge graph, electronic device fault description, electronic device structure, and electronic device function. It can be seen that, based on the function corresponding to any component in the electronic device structure, the fault description corresponding to any component can be determined, thereby generating the target fault library of the electronic device.

[0139] For example, Figure 7 A flowchart illustrating another method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance provided in this application embodiment is shown below. Figure 7 As shown, the process includes: collecting data and knowledge information for the analyzed electronic equipment; processing the collected knowledge and data information separately; completing the fault framework mapping relationship of the electronic equipment to form a fault sample framework of the electronic equipment; screening fault triples <head entity, relation, tail entity> according to the definition of knowledge graph; using each component node containing component information, fault mode information, and data feature information as a node type label and importing it into the knowledge graph; connecting the component nodes in the knowledge graph hierarchically according to the node type labels, and determining the main structure of the knowledge graph according to the subordinate relationships between component nodes; projecting entity relations to the relation space in which they reside for triples to construct a representation model; deleting redundant nodes and relations to form a complete knowledge graph.

[0140] Figure 8 This application provides a schematic diagram of the structure of a device for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, as shown in the embodiments of this application. Figure 8 As shown, the device includes:

[0141] The acquisition unit 31 is used to acquire a data acquisition request; wherein the data acquisition request includes the component data identifier of the electronic device; and acquire the component data corresponding to the component identifier according to the data acquisition request.

[0142] The analysis unit 32 is used to perform fault analysis processing on component data and determine the fault sample library of electronic equipment; wherein, the fault sample library includes fault mode information corresponding to each component in the electronic equipment.

[0143] The first determining unit 33 is used to determine the node type label of each component and the fault mode information of each component based on the node definition information of the preset knowledge graph.

[0144] The generation unit 34 is used to generate a target fault library based on the node type label, so as to realize the fault diagnosis of electronic equipment and obtain the diagnosis result information based on the target fault library.

[0145] The apparatus in this embodiment can execute the technical solutions in the above method. Its specific implementation process and technical principles are the same, and will not be repeated here.

[0146] Figure 9 This is a schematic diagram of another device for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, provided in an embodiment of this application. Figure 8 Based on the illustrated embodiments, as Figure 9 As shown, the component data includes component list information, preset component rule information, design data for each component, and test data.

[0147] In one example, analysis unit 32 includes:

[0148] The first analysis module 321 is used to perform fault analysis processing on the component list information and the preset component rule information to determine the composition relationship between multiple components in the electronic device and the first fault mode information corresponding to each of the multiple components.

[0149] The first determining module 322 is used to determine the component composition library in the fault sample library based on the composition relationship between multiple components; and to determine the fault mode library in the fault sample library based on the first fault mode information corresponding to each of the multiple components.

[0150] The second analysis module 323 is used to perform fault analysis and processing on the design data and test data of each component to determine the second fault mode information of each component.

[0151] The second determining module 324 is used to determine the data feature library in the fault sample library based on the second fault mode information of each component.

[0152] In one example, the second analysis module 323 includes:

[0153] The extraction submodule 3231 is used to perform feature extraction processing on the design data and test data of each component to determine the data characteristics of each component.

[0154] The determination submodule 3232 is used to determine the data characteristics of each component as the second fault mode information of each component.

[0155] In one example, the device also includes:

[0156] The second determining unit 41 is used to determine, based on the composition relationship between components in the component composition library, the first fault mode information corresponding to each component in the fault mode library, and the second fault mode information corresponding to each component in the data feature library, a first mapping relationship between components in the component composition library and the first fault mode information in the fault mode library, and / or a second mapping relationship between components in the component composition library and the second fault mode information in the data feature library.

[0157] In one example, the first determining unit 33 is specifically used for:

[0158] Based on the node definition information, first mapping relationship and second mapping relationship of the preset knowledge graph, the first node type label of each component and the second node type label of the fault mode information of each component are determined.

[0159] In one example, generating unit 34 includes:

[0160] Import module 341 is used to import the component and / or component failure mode information corresponding to the first node type label and the component and / or component failure mode information corresponding to the second node type label into the component node in the preset knowledge graph.

[0161] The third determining module 342 is used to determine the subordinate relationship between component nodes based on the first node type label and the second node type label.

[0162] The generation module 343 is used to generate a target fault library based on the hierarchical relationship between the component nodes.

[0163] The apparatus in this embodiment can execute the technical solutions in the above method. Its specific implementation process and technical principles are the same, and will not be repeated here.

[0164] Figure 10 This application provides a schematic diagram of the structure of a server, as shown in the embodiment of the present application. Figure 10 As shown, the server includes: a memory 51 and a processor 52.

[0165] The memory 51 stores a computer program that can run on the processor 52.

[0166] Processor 52 is configured to perform the methods provided in the embodiments described above.

[0167] The server also includes a receiver 53 and a transmitter 54. The receiver 53 is used to receive instructions and data sent by external devices, and the transmitter 54 is used to send instructions and data to external devices.

[0168] This application also provides a non-transitory computer-readable storage medium, which, when the instructions in the storage medium are executed by the processor of a server, enables the server to perform the methods provided in the above embodiments.

[0169] This application also provides a computer program product, which includes: a computer program stored in a readable storage medium, at least one processor of the server can read the computer program from the readable storage medium, and the at least one processor executes the computer program to cause the server to perform the solution provided in any of the above embodiments.

[0170] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0171] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, characterized in that, include: Obtain a data acquisition request; wherein the data acquisition request includes component data identifiers of electronic devices; and according to the data acquisition request, acquire component data corresponding to the component data identifiers, wherein the component data includes component list information, preset component rule information, design data of each component, and test data; The component data is subjected to fault analysis processing to determine the fault sample library of the electronic device; wherein, the fault sample library includes fault mode information corresponding to each component in the electronic device, the fault sample library also includes a component composition library and a fault mode library obtained by analyzing the component list information and the preset component rule information, and a data feature library obtained by feature extraction from the design data and test data of each component; Based on the composition relationship between components in the component composition library, the first fault mode information corresponding to each component in the fault mode library, and the second fault mode information corresponding to each component in the data feature library, a first mapping relationship between the components in the component composition library and the first fault mode information in the fault mode library, and / or a second mapping relationship between the components in the component composition library and the second fault mode information in the data feature library is determined. Based on the node definition information of the preset knowledge graph, the first mapping relationship and the second mapping relationship, the first node type label of each component and the second node type label of the fault mode information of each component are determined. Based on the first node type label and the second node type label, a target fault database is generated to realize the fault diagnosis of electronic equipment and obtain the diagnosis result information according to the target fault database.

2. The method according to claim 1, characterized in that, The step of performing fault analysis processing on the component data to determine the fault sample library of the electronic device includes: The component list information and preset component rule information are subjected to fault analysis and processing to determine the composition relationship between multiple components in the electronic device and the first fault mode information corresponding to each of the multiple components. Based on the compositional relationships between the multiple components, a component composition library is generated in the fault sample library; and based on the first fault mode information corresponding to each of the multiple components... Information is used to generate a fault mode library in the fault sample library; The design data and test data of each component are subjected to fault analysis processing to determine the second fault mode information of each component; Based on the second fault mode information of each component, a data feature library is generated in the fault sample library.

3. The method according to claim 2, characterized in that, The process of performing fault analysis on the design data and test data of each component to determine the second fault mode information of each component includes: Feature extraction processing is performed on the design data and test data of each component to determine the data characteristics of each component; The data characteristics of each component are determined as the second fault mode information of each component.

4. The method according to claim 1, characterized in that, The step of generating a target fault database based on the first node type label and the second node type label includes: Import the component and / or the fault mode information of the component corresponding to the first node type label, and the component and / or the fault mode information of the component corresponding to the second node type label into the component node in the preset knowledge graph; The hierarchical relationship between component nodes is determined based on the first node type label and the second node type label; Based on the hierarchical relationships between the component nodes, a target fault database is generated.

5. A device for constructing, updating, and handling fault knowledge bases for intelligent operation and maintenance, characterized in that, include: An acquisition unit is used to acquire a data acquisition request; wherein the data acquisition request includes component data identifiers of electronic devices; and according to the data acquisition request, acquires component data corresponding to the component identifiers, wherein the component data includes component list information, preset component rule information, design data of each component, and test data; The analysis unit is used to perform fault analysis processing on the component data to determine a fault sample library for the electronic device. The fault sample library includes fault mode information corresponding to each component in the electronic device. It also includes a component composition library and a fault mode library obtained by analyzing the component list information and the preset component rule information, as well as feature analysis based on the design data and test data of each component. The extracted data feature library; The first determining unit is configured to determine, based on the composition relationship between components in the component composition library, the first fault mode information corresponding to each component in the fault mode library, and the second fault mode information corresponding to each component in the data feature library, a first mapping relationship between components in the component composition library and the first fault mode information in the fault mode library, and / or a second mapping relationship between components in the component composition library and the second fault mode information in the data feature library. Based on the node definition information of the preset knowledge graph, the first mapping relationship and the second mapping relationship, the first node type label of each component and the second node type label of the fault mode information of each component are determined. The generation unit is used to generate a target fault database based on the first node type label and the second node type label, so as to realize fault diagnosis of electronic devices and obtain diagnostic result information based on the target fault database.

6. The apparatus according to claim 5, characterized in that, The analysis unit includes: The first analysis module is used to perform fault analysis processing on the component list information and the preset component rule information to determine the composition relationship between multiple components in the electronic device and the first fault mode information corresponding to each of the multiple components. The first determining module is used to determine the component composition library in the fault sample library based on the composition relationship between the plurality of components; and to determine the fault mode library in the fault sample library based on the first fault mode information corresponding to each of the plurality of components. The second analysis module is used to perform fault analysis processing on the design data and test data of each component to determine the second fault mode information of each component. The second determining module is used to determine the data feature library in the fault sample library based on the second fault mode information of each component.

7. The apparatus according to claim 6, characterized in that, The second analysis module includes: The extraction submodule is used to perform feature extraction processing on the design data and test data of each component to determine the data features of each component; The determination submodule is used to determine the data characteristics of each component as the second fault mode information of each component.

8. The apparatus according to claim 5, characterized in that, The generation unit includes: The import module is used to import the component and / or the failure mode information of the component corresponding to the first node type label and the component and / or the failure mode information of the component corresponding to the second node type label into the component node in the preset knowledge graph. The third determining module is used to determine the subordinate relationship between each component node based on the first node type label and the second node type label; The generation module is used to generate a target fault library based on the hierarchical relationships between the component nodes.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-4.

10. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1-4.

Images