Standardized micro-application-oriented operation and maintenance knowledge extraction method and system

Abstract

The application belongs to the technical field of service fault monitoring, and provides an operation and maintenance knowledge extraction method and system for standardized micro applications, which acquires service and operation and maintenance data of each micro application, pre-processes the data to obtain an initial sample data set after preprocessing, constructs a micro application hypergraph in units of micro applications based on the initial sample data set, establishes a domain knowledge model, constructs a base micro application operation and maintenance knowledge system with the domain knowledge model as a carrier, constructs a micro application knowledge graph in combination of the micro application hypergraph and the base micro application operation and maintenance knowledge system, constructs a probabilistic fuzzy cognitive graph, uses the probabilistic fuzzy cognitive graph to reason and extract operation and maintenance knowledge, and maps a fault path reasoned by the probabilistic fuzzy cognitive graph into an entity relationship in the micro application knowledge graph to obtain fault positioning and fault cause of micro application abnormal service, which solves the problems of slow discovery, inaccurate positioning and difficult tracing of key service operation faults of current standardized micro applications.

Description

Technical Field

[0001] This invention belongs to the field of smart grid technology, and in particular relates to a method and system for extracting operation and maintenance knowledge for standardized micro-applications. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] The new generation of data acquisition systems involves a large number and variety of servers, and has a complex topology, IT architecture, and data processing flow. Data storage employs a hybrid data storage architecture, making data querying and analysis relatively complex. While the new generation of acquisition systems ensures high performance to a certain extent, the components used are complex, and there are few related visualization and monitoring measures. This fails to effectively monitor the operation of various micro-application services, making it difficult to promptly identify the location and cause of abnormal service failures, thus affecting the operational efficiency of related services. To achieve effective monitoring of micro-applications, knowledge extraction is required from the business operation and maintenance data of each micro-application. However, most existing knowledge extraction methods do not consider the correlation and integrity of knowledge during extraction, resulting in the extraction of knowledge that cannot form a complete knowledge system or knowledge graph. At the same time, in some knowledge extraction tasks, expert participation is generally required to ensure the accuracy and reliability of knowledge extraction, which consumes certain human and time costs and makes it difficult to detect business failures in a timely and effective manner. This leads to problems such as slow detection, inaccurate location, and difficulty in tracing the source of critical business operation failures in micro-applications. Summary of the Invention

[0004] To address at least one of the technical problems mentioned above, this invention provides a method and system for extracting operation and maintenance knowledge for standardized micro-applications. This method and system accurately and efficiently extracts operation and maintenance-related knowledge for micro-applications and enables visualized fault monitoring and problem tracing based on operation and maintenance knowledge graphs, thereby effectively improving the operation and maintenance efficiency of micro-applications.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: The first aspect of the present invention provides a method for extracting operation and maintenance knowledge for standardized micro-applications, comprising the following steps: Acquire business and operational data from each micro-application, and preprocess the data to obtain a preprocessed initial sample dataset; Based on the initial sample dataset, a micro-application hypergraph is constructed, with each micro-application as a unit. Establish a domain knowledge model, use the domain knowledge model as a carrier to construct a base micro-application operation and maintenance knowledge system, and combine the micro-application hypergraph and the base micro-application operation and maintenance knowledge system to construct a micro-application knowledge graph; A probabilistic fuzzy cognitive graph is constructed, and the probabilistic fuzzy cognitive graph is used to reason and extract operation and maintenance knowledge. The fault paths reasoned by the probabilistic fuzzy cognitive graph are mapped to entity relationships in the micro-application knowledge graph to obtain the fault location and fault cause of abnormal micro-application services.

[0006] Furthermore, the business and operation and maintenance data of each micro-application specifically include base operation data, fault repair records, log record text, and power grid fault analysis report data; the preprocessing process includes data cleaning and data normalization processing.

[0007] Furthermore, the construction of a micro-application hypergraph based on the initial sample dataset, with each micro-application as a unit, includes: The node set is defined based on the data in the initial sample dataset, the set of all entity objects is determined, and the hyperedge is constructed based on each data node and the set hyperedge rules to form a hyperedge set. The association matrix is ​​generated according to the relationship between vertices and edges to complete the construction of the hypergraph.

[0008] Furthermore, the process of building a domain knowledge model includes: Using a domain ontology as the knowledge carrier, this ontology forms a domain knowledge model by abstractly defining entity categories and their inter-domain relationships, represented as follows: ,in It is a set of concepts. It refers to the contextual relationship of concepts, also known as taxonomic knowledge; It is a set of attributes that describe the characteristics of a concept; It is a set of domain description rules; It is an instance set used to describe instance-property-value.

[0009] Furthermore, the construction of a micro-application operation and maintenance knowledge system based on the micro-application fault ontology includes: constructing an ontology concept model of the micro-application fault domain, which includes two primary concepts and three sub-concepts. The two primary concepts include: fault ontology and device ontology; the primary concepts are further divided into three sub-concepts, namely device, fault phenomenon and fault cause.

[0010] Furthermore, the probabilistic fuzzy cognitive map is as follows: , In the formula, The total number of all concepts. For the concept of results exist The state value at time t, For the concept of cause exist The state value at time t, For related concepts exist The state value at time t, In order to be in Time concept On the concept The strength of the causal relationship The influence factor of the previous time step on the state value of the next time step. For the concept Threshold function, Index representing the concept of cause, An index representing the concept of a result.

[0011] Furthermore, when mapping the fault paths inferred from the probabilistic fuzzy cognitive graph to entity relationships in the knowledge graph, the mapping is performed based on the state matrix and relation weight matrix of the probabilistic fuzzy cognitive graph.

[0012] A second aspect of the present invention provides an operation and maintenance knowledge extraction system for standardized micro-applications, comprising: The sample dataset construction module is used to acquire business and operation and maintenance data of each micro-application and preprocess the data to obtain the preprocessed initial sample dataset. The hypergraph building module is used to build a micro-application hypergraph based on the initial sample dataset, with each micro-application as a unit. The Operation and Maintenance Knowledge System Construction Module is used to establish a domain knowledge model. Using the domain knowledge model as a carrier, it constructs a base micro-application operation and maintenance knowledge system and combines the micro-application hypergraph and the base micro-application operation and maintenance knowledge system to construct a micro-application knowledge graph. The fault location module is used to construct a probabilistic fuzzy cognitive graph, use the probabilistic fuzzy cognitive graph to reason and extract operation and maintenance knowledge, and map the fault path reasoned from the probabilistic fuzzy cognitive graph into entity relationships in the micro-application knowledge graph to obtain the fault location and fault cause of abnormal micro-application services.

[0013] A third aspect of the present invention provides a computer-readable storage medium.

[0014] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the operation and maintenance knowledge extraction method for standardized micro-applications as described above.

[0015] A fourth aspect of the present invention provides a computer device.

[0016] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps in the operation and maintenance knowledge extraction method for standardized micro-applications as described above.

[0017] Compared with the prior art, the beneficial effects of the present invention are: This invention constructs a hypergraph based on micro-applications, completing the corresponding conversion of micro-application monitoring data, key nodes, and operation and maintenance knowledge extraction. It constructs a base micro-application operation and maintenance knowledge system through ontology modeling, abstracting and constraining the relationship between concept classification, concept attribute description, and knowledge data concepts. Based on fuzzy cognitive graphs, it performs reasoning and extraction of operation and maintenance knowledge. During reasoning analysis, it can intuitively analyze the location of fault occurrence and achieve root cause tracing, solving the problems of slow fault detection, inaccurate location, and difficulty in tracing the source of faults in current standardized micro-application critical business operations.

[0018] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0019] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0020] Figure 1 This is a flowchart of the abnormal business tracing method for standardized micro-applications provided in the embodiments of the present invention; Figure 2 This is the micro-application fault monitoring knowledge graph ontology design scheme provided in the embodiments of the present invention; Figure 3 This is the relational weight matrix and corresponding PFCM provided in the embodiments of the present invention. Detailed Implementation

[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0022] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0024] To achieve effective monitoring of micro-applications, knowledge extraction from the operational data of each micro-application is necessary. Many research institutions and scholars have studied knowledge extraction methods. However, due to the difficulty in guaranteeing the F1 score during feature extraction in the construction of knowledge graphs for power grid infrastructure projects, a feature extraction method combining the BERT model for power grid infrastructure engineering knowledge graphs has been proposed. This method is based on RDF theory and utilizes triples to construct a knowledge graph model for power grid infrastructure projects, effectively improving the accuracy of target feature extraction. Addressing the over-reliance on expert experience in traditional distribution network problem diagnosis, a distribution network problem association analysis model based on grey relational analysis and the Apriori algorithm has been proposed. Using association rules for data feature extraction effectively overcomes the difficulties of relying on human experience, improving the efficiency and reliability of distribution network maintenance. Finally, addressing the problems of low utilization of unstructured text data, difficulties in data fusion, and limited data application methods in the maintenance of transmission and transformation equipment, a knowledge graph-based intelligent question-and-answer system for operation and maintenance knowledge has been proposed. Entity-relationship-attribute triples are extracted from multi-source data and combined with the Neo4j graph database to construct a knowledge graph for the operation and maintenance of power transmission and transformation equipment. This provides a knowledge network for power transmission and transformation equipment operation and maintenance systems that is oriented towards multiple business scenarios, and provides reliable methods and means for equipment status detection and data application decision-making.

[0025] However, most existing knowledge extraction methods do not consider the relevance and integrity of knowledge during extraction, resulting in the extraction of knowledge that cannot form a complete knowledge system or knowledge graph. At the same time, in some knowledge extraction tasks, in order to ensure the accuracy and reliability of knowledge extraction, experts are generally required, which consumes certain human and time costs. This makes it difficult to discover business faults in a timely and effective manner, resulting in slow detection, inaccurate location, and difficulty in tracing the source of faults in the critical business operations of micro-applications.

[0026] This invention addresses the problems of slow fault detection, inaccurate location, and difficulty in tracing the root causes of critical operational faults in current standardized micro-applications. It proposes a method for extracting operational and maintenance (O&M) knowledge from standardized micro-applications, enabling the monitoring and root cause tracing of abnormal system operations. First, based on a new-generation data acquisition system, a hypergraph is constructed at the micro-application level, completing the corresponding conversion of micro-application monitoring data, key nodes, and O&M knowledge extraction. Then, an ontology modeling method is used to construct the O&M knowledge system for the base micro-applications, abstracting and constraining the relationships between concept classification, concept attribute descriptions, and knowledge data concepts. Finally, based on a fuzzy cognitive graph, O&M knowledge is inferred and extracted. During inference analysis, the location of faults can be intuitively analyzed, and root cause tracing can be achieved. Experiments show that this method can more accurately and efficiently extract O&M-related knowledge for micro-application operations, while simultaneously realizing visualized fault monitoring and root cause tracing based on O&M knowledge graphs, effectively improving the O&M efficiency of micro-applications.

[0027] Specifically, it includes the following aspects: First, in response to the problems of strong reliance on manual labor, difficulty in handling complex and diverse relationships, and lack of intuitiveness in the traditional knowledge extraction process, HyperGraph is used for knowledge extraction from sample data. HyperGraph can visualize knowledge extraction and can more intuitively show the knowledge extraction process through images, reducing the difficulty of knowledge extraction-related operations, making the knowledge extraction process easier to understand, and facilitating subsequent analysis and modeling. Second, based on the definitions and properties of information systems and hypergraphs, the correspondence between the components of information systems and hypergraphs is analyzed, and a method for mutual transformation between information systems and hypergraphs is proposed. Knowledge extraction is then performed on this basis.

[0028] Third, in response to the difficulty in tracing the root cause of business anomalies in existing micro-applications, probabilistic fuzzy cognitive graphs can not only optimize the root cause ranking based on fuzzy rules, but also utilize multi-source data and domain knowledge to build a problem tracing model for complex operation and maintenance scenarios, thereby improving the efficiency and accuracy of fault tracing for provincial-level micro-applications and enhancing the scientific nature and reliability of operation and maintenance decisions.

[0029] Example 1 like Figure 1 As shown, this embodiment provides a method for tracing abnormal business processes in standardized micro-applications, including the following steps: Step 1: Obtain the business and operation and maintenance data of each micro-application, and preprocess the data to obtain the initial sample dataset; In this embodiment, based on the new generation of electricity information collection system, business and operation and maintenance data of each micro-application are acquired, and the data is preprocessed to obtain an initial sample dataset; In this embodiment, the business and operation and maintenance data of each micro-application specifically include base operation data, fault repair records, log records and other text data, as well as power grid fault analysis reports and other data; the preprocessing process includes data cleaning and data normalization processing.

[0030] Step 2: Based on the initial sample dataset, construct a micro-application hypergraph with micro-applications as the unit; In this embodiment, the node set is defined based on the data of the initial sample dataset, the set of all entity objects is determined, and the hyperedge is constructed based on each data node and combined with the set hyperedge rules to form a hyperedge set. The association matrix is ​​generated according to the relationship between vertices and edges to complete the hypergraph construction. Specifically, define a hypergraph ,in For a set of points, For a set of superedges, It is called a hyperedge and satisfies .when At that time, the super-border It is called an empty hyperedge.

[0031] Step 3: Establish a domain knowledge model. Using the domain knowledge model as a carrier, construct a base micro-application operation and maintenance knowledge system. Combine the micro-application hypergraph and the base micro-application operation and maintenance knowledge system to construct a micro-application knowledge graph. Specifically, the steps include the following: Step 301: Establish a domain knowledge model; Defining knowledge space: memorization The domain is a finite, non-empty set. For the domain The set of equivalence relations on the knowledge space Representing the set of equivalence relations All possible relations are related to the domain of discourse The classification; ontology modeling, as a form of knowledge modeling, can not only satisfy the construction of knowledge graph schema layers, but also unify data representation. Through ontology modeling methods, a micro-application operation and maintenance knowledge system can be constructed, abstracting and constraining the relationships between concept classification, concept attribute descriptions, and knowledge data concepts.

[0032] Knowledge graphs use domain ontology as the knowledge carrier. This ontology forms a domain knowledge model by abstractly defining entity categories and their inter-domain relationships, which can be represented as follows: ,in It is a set of concepts, such as the concepts of importance and event classes; It refers to the contextual relationship of concepts, also known as taxonomic knowledge; It is a set of attributes that describe the characteristics of a concept; It is a set of domain description rules; It is an instance set used to describe instance-property-value.

[0033] The domain ontology forms a domain knowledge model O by abstractly defining entity categories and their inter-domain relationships. The knowledge graph uses the domain ontology as the knowledge carrier, that is, O as the carrier, to construct a micro-application business fault monitoring knowledge graph.

[0034] Step 302: Construct a base micro-application operation and maintenance knowledge system using a micro-application fault ontology; In a given fault monitoring domain, a well-defined domain ontology clarifies the structure of the knowledge system, reducing redundancy and errors when constructing the knowledge graph. This embodiment proposes a micro-application fault monitoring domain ontology design scheme; for the specific architecture, please refer to [link to specific architecture]. Figure 2 .

[0035] This embodiment defines the micro-application fault ontology from industry standards and categorizes the knowledge under the guidance of experts in the field of micro-application fault monitoring. The ontology concept modeling of two main concepts and three sub-concepts in the micro-application fault domain is achieved using the ontology building tool Protégé.

[0036] The two primary concepts include: fault entity and equipment entity; each primary concept is further divided into three sub-concepts: equipment, fault phenomenon, and fault cause. Correlation analysis of these three types of entities can determine two semantic relationships: (1) Occurrence: [Equipment - Occurrence - Cause of Failure].

[0037] (2) Cause: [Cause of failure - Cause - Fault phenomenon].

[0038] The device implementation includes multiple attributes, such as: name, model, code, location, upper port, and lower port. These six attributes and the device entity can form six attribute triples.

[0039] The fault phenomenon entity contains multiple attributes: name, category (single fault, multiple faults), and status (attention, abnormal, severe).

[0040] The entity representing the cause of the fault includes the following attributes: name, category (A, B, C, D), and handling method.

[0041] Step 303: Construct a micro-application knowledge graph by combining the micro-application hypergraph and the base micro-application operation and maintenance knowledge system; The base standardization micro application is a quadruple ,in This represents the domain of discourse for each micro-application; Represents a non-empty finite set of attributes; , Representative attribute The range of values; The representative information function can be expressed as: , Each attribute can be assigned a value, that is, a value that can be assigned to it. ,have .

[0042] For base-standardized micro-applications There must exist a corresponding hypergraph. And it meets the following conditions: , .

[0043] Information systems He Chaotu The following correspondences are satisfied: .

[0044] For attribute value range is not Information systems There also exists a hypergraph. Correspondingly, Table 1 shows the correspondence between hypergraphs, knowledge spaces, and information systems.

[0045] Table 1. Correspondence between Hypergraph, Knowledge Space, and Information System

[0046] Table 1 shows that there is a one-to-one correspondence between the terms information system, knowledge space, and hypergraph, that is, for every known information system... Each of them has a unique corresponding hypergraph. .

[0047] Based on the logical architecture of standardized micro-applications, corresponding hypergraphs are generated, which can then leverage the visualization advantages of hypergraphs to monitor anomalies in key business metrics of micro-applications.

[0048] The algorithm for constructing a hypergraph using an information system is as follows:

[0049] The first step of the hypergraph algorithm is to initialize the set of points that the information system uses to construct the hypergraph. With the set of superedges Steps 2-4 of the algorithm construct the point set of the hypergraph based on the object set of the information system. Each object needs to be assigned a value, and a total of [number] operations are required. This operation; steps 5-9 of the algorithm involve performing each loop based on the attributes of the information system, determining the appropriate attribute value for each object in each loop, for a total of [number] operations. This is the second operation.

[0050] Step 4: Construct a probabilistic fuzzy cognitive graph. Use the fuzzy cognitive graph to reason and extract operational knowledge, and map the fault paths reasoned from the probabilistic fuzzy cognitive graph into entity relationships in the knowledge graph to obtain the fault location and fault cause of abnormal micro-application services, thereby improving the quality and efficiency of operation and maintenance.

[0051] Specifically, the steps include the following: Step 401: Construct a probabilistic fuzzy cognitive map; In this embodiment, causal relationships are introduced into the probabilistic fuzzy cognitive graph to obtain a causal network model; Probabilistic Fuzzy Cognitive Maps (PFCMs) generally solve reasoning problems involving incomplete expert knowledge and ambiguity in causal relationships between concepts. PFCMs introduce causal relationships into Fuzzy Cognitive Maps (FCMs), facilitating better causal network modeling and reasoning. Their specific form is as follows: , In the formula, The total number of all concepts. For the concept of results exist The state value at time t, For the concept of cause exist The state value at time t, For related concepts exist The state value at time t, In order to be in Time concept On the concept The strength of the causal relationship The influence factor of the previous time step on the state value of the next time step. For the concept Threshold function.

[0052] Step 402: Use probabilistic fuzzy cognitive graphs to reason and extract operational knowledge, and map the fault paths reasoned from the probabilistic fuzzy cognitive graphs into entity relationships in the knowledge graph to obtain the fault location and fault cause of abnormal micro-application services. Specifically, based on the state matrix and relation weight matrix of the probabilistic fuzzy cognitive graph, the fault path inferred from the probabilistic fuzzy cognitive graph is mapped to entity relations in the knowledge graph, the state output of each node is calculated, and the fault of the business is traced based on the state information of each node. Suppose there is an operations and maintenance knowledge point that includes If there are 10 information points, then its probability fuzzy cognitive map should have 1000 information points. There are 10 nodes, so at this point... state matrix and Relationship weight matrix It can be defined as: , , In the state matrix, yes The state values ​​of each information point. Its state transition function is: , Then the state output of each node is obtained as follows: , in, The knowledge function representing all forward information points, This is the status output function for the information point.

[0053] Information points new status It can be represented as: , in, Indicates from node To the node Relationship weight value, Represents a node The state value.

[0054] Based on the above reasoning process, the causal influence strength is calculated as the confidence weight of entity relationships in the knowledge graph, and the nodes are... To the node Instantaneous causal effect intensity Defined as: , Based on this, Aggregation is performed at key time steps, mapping them to relationships in a knowledge graph. confidence level : , in, This represents the nodes during the fault propagation process. For nodes The greatest impact; This represents a mapping function that maps the calculated maximum influence to the range required for the knowledge graph relation weights.

[0055] Each state change of a knowledge point is considered as a reasoning step. The reasoning process of a knowledge point can be completed by performing matrix multiplication on the knowledge points and combining it with the state output function. This process can intuitively analyze the impact of cause nodes on result nodes and realize fault tracing of key business of provincial micro-applications.

[0056] This invention uses a new-generation electricity consumption information collection system of a State Grid province as the experimental background platform. The experimental data includes base operation data, fault maintenance records, log records, and power grid fault analysis reports from January to December 2024, collected by the base micro-application on the provincial side. The sample data is divided into a training sample set and a test sample set in a 7:3 ratio.

[0057] Let's take a micro-application A as an example to perform knowledge reasoning for PFCM. Assume that the operational knowledge points of this micro-application include... Establish a state matrix based on information points. And relational weight matrix for: , , Based on the established knowledge reasoning model of probabilistic fuzzy cognitive graphs, the corresponding structure diagram is obtained as follows: Figure 3 : Assumption , , , , , The thresholds are 0.3, 0.3, 0.2, 0.2, 0.4, and 0.1, respectively. The required relevant knowledge can be obtained after calculation through the knowledge reasoning model.

[0058] To verify the effectiveness of the proposed method in tracing the source of abnormal business issues in micro-applications, the method of this invention is compared with existing knowledge extraction methods. The comparison methods include: (1) Knowledge extraction from cloud computing platform: Based on cloud computing platform, a knowledge extraction framework is constructed using Hadoop combined with MapReduce programming method; (2) Based on semantic relations: Construct a knowledge extraction system based on semantic relations; (3) Conceptual level; (4) The method of the present invention.

[0059] The evaluation metrics chosen are Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE). Smaller metric values ​​indicate smaller errors, meaning higher model prediction accuracy. The definitions are as follows: , , in, This is the root cause prediction value for abnormal business failures in micro-applications. This represents the true root cause of abnormal business failures in micro-applications. This indicates the number of test samples.

[0060] The final experimental comparison results are shown in Table 2: Table 2. Experimental Comparison Results

[0061] As can be seen from Table 2, compared with the other three algorithms, the method of the present invention has higher recognition accuracy. It can be effectively applied to the fault root cause identification scenario of abnormal micro-application services in the cold start state during the construction of the provincial side base, improve the management and control capabilities of provincial side micro-application services, provide a data foundation for the orderly operation of provincial side micro-application services, improve the operation and maintenance efficiency of provincial side micro-application services, and thus improve the level of refined management of provincial side micro-application services.

[0062] Example 2 This embodiment provides an operation and maintenance knowledge extraction system for standardized micro-applications, including: The sample dataset construction module is used to acquire business and operation and maintenance data of each micro-application and preprocess the data to obtain the preprocessed initial sample dataset. The hypergraph building module is used to build a micro-application hypergraph based on the initial sample dataset, with each micro-application as a unit. The Operation and Maintenance Knowledge System Construction Module is used to establish a domain knowledge model. Using the domain knowledge model as a carrier, it constructs a base micro-application operation and maintenance knowledge system and combines the micro-application hypergraph and the base micro-application operation and maintenance knowledge system to construct a micro-application knowledge graph. The fault location module is used to construct a probabilistic fuzzy cognitive graph, use the probabilistic fuzzy cognitive graph to reason and extract operation and maintenance knowledge, and map the fault path reasoned from the probabilistic fuzzy cognitive graph into entity relationships in the micro-application knowledge graph to obtain the fault location and fault cause of abnormal micro-application services.

[0063] The hypergraph construction module, which involves constructing a micro-application hypergraph based on the initial sample dataset, includes: The node set is defined based on the data in the initial sample dataset, the set of all entity objects is determined, and the hyperedge is constructed based on each data node and the set hyperedge rules to form a hyperedge set. The association matrix is ​​generated according to the relationship between vertices and edges to complete the construction of the hypergraph.

[0064] In the module for building the operations and maintenance knowledge system, the process of establishing the domain knowledge model includes: Using a domain ontology as the knowledge carrier, this ontology forms a domain knowledge model by abstractly defining entity categories and their inter-domain relationships, represented as follows: ,in It is a set of concepts. It refers to the contextual relationship of concepts, also known as taxonomic knowledge; It is a set of attributes that describe the characteristics of a concept; It is a set of domain description rules; It is an instance set used to describe instance-property-value.

[0065] In the operation and maintenance knowledge system construction module, the micro-application operation and maintenance knowledge system based on the micro-application fault ontology is constructed, which includes: constructing an ontology concept model of the micro-application fault domain, which includes two primary concepts and three sub-concepts. The two primary concepts include: fault ontology and device ontology; the primary concepts are further divided into three sub-concepts, namely device, fault phenomenon and fault cause.

[0066] In the fault location module, the probabilistic fuzzy cognitive map is: , In the formula, The total number of all concepts. For the concept of results exist The state value at time t, For the concept of cause exist The state value at time t, For related concepts exist The state value at time t, In order to be in Time concept On the concept The strength of the causal relationship The influence factor of the previous time step on the state value of the next time step. For the concept Threshold function, Index representing the concept of cause, An index representing the concept of a result.

[0067] When mapping the fault paths inferred from the probabilistic fuzzy cognitive graph to entity relationships in the knowledge graph, the mapping is performed based on the state matrix and relation weight matrix of the probabilistic fuzzy cognitive graph.

[0068] It should be noted that the specific implementation of the operation and maintenance knowledge extraction system for standardized micro-applications in this embodiment of the invention is similar to the specific implementation of the operation and maintenance knowledge extraction method for standardized micro-applications in this embodiment of the invention. For details, please refer to the description in the method section. To reduce redundancy, it will not be repeated here.

[0069] Example 3 This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in the operation and maintenance knowledge extraction method for standardized micro-applications described above.

[0070] Example 4 This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the operation and maintenance knowledge extraction method for standardized micro-applications described above.

[0071] Example 5 This embodiment provides a program product, which is a computer program product including a computer program. When the computer program is executed by a processor, it implements the steps in the operation and maintenance knowledge extraction method for standardized micro-applications described above.

[0072] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0073] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0074] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0075] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0076] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0077] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for extracting operation and maintenance knowledge for standardized micro-applications, characterized in that, Includes the following steps: Acquire business and operational data from each micro-application, and preprocess the data to obtain a preprocessed initial sample dataset; Based on the initial sample dataset, a micro-application hypergraph is constructed, with each micro-application as a unit. Establish a domain knowledge model, use the domain knowledge model as a carrier to construct a base micro-application operation and maintenance knowledge system, and combine the micro-application hypergraph and the base micro-application operation and maintenance knowledge system to construct a micro-application knowledge graph; A probabilistic fuzzy cognitive graph is constructed, and the probabilistic fuzzy cognitive graph is used to reason and extract operation and maintenance knowledge. The fault paths reasoned by the probabilistic fuzzy cognitive graph are mapped to entity relationships in the micro-application knowledge graph to obtain the fault location and fault cause of abnormal micro-application services.

2. The method for extracting operation and maintenance knowledge for standardized micro-applications as described in claim 1, characterized in that, The business and operation and maintenance data of each micro-application specifically include base operation data, fault repair records, log record text, and power grid fault analysis report data; the preprocessing process includes data cleaning and data normalization.

3. The method for extracting operation and maintenance knowledge for standardized micro-applications as described in claim 1, characterized in that, The construction of a micro-application hypergraph based on the initial sample dataset, with each micro-application as a unit, includes: The node set is defined based on the data in the initial sample dataset, the set of all entity objects is determined, and the hyperedge is constructed based on each data node and the set hyperedge rules to form a hyperedge set. The association matrix is ​​generated according to the relationship between vertices and edges to complete the construction of the hypergraph.

4. The method for extracting operation and maintenance knowledge for standardized micro-applications as described in claim 1, characterized in that, The process of building a domain knowledge model includes: Using a domain ontology as the knowledge carrier, this ontology forms a domain knowledge model by abstractly defining entity categories and their inter-domain relationships, represented as follows: ,in It is a set of concepts. It refers to the contextual relationship of concepts, also known as taxonomic knowledge; It is a set of attributes that describe the characteristics of a concept; It is a set of domain description rules; It is an instance set used to describe instance-property-value.

5. The method for extracting operation and maintenance knowledge for standardized micro-applications as described in claim 1, characterized in that, The method of constructing a micro-application operation and maintenance knowledge system based on a micro-application fault ontology includes: constructing an ontology concept model of the micro-application fault domain, which includes two primary concepts and three sub-concepts. The two primary concepts are: fault ontology and device ontology. The primary concepts are further divided into three sub-concepts, namely, device, fault phenomenon and fault cause.

6. The method for extracting operation and maintenance knowledge for standardized micro-applications as described in claim 1, characterized in that, The probabilistic fuzzy cognitive map is as follows: ， In the formula, The total number of all concepts. For the concept of results exist The state value at time t, For the concept of cause exist The state value at time t, For related concepts exist The state value at time t, In order to be in Time concept On the concept The strength of the causal relationship The influence factor of the previous time step on the state value of the next time step. For the concept Threshold function, Index representing the concept of cause, An index representing the concept of a result.

7. The method for extracting operation and maintenance knowledge for standardized micro-applications as described in claim 1, characterized in that, When mapping the fault paths inferred from the probabilistic fuzzy cognitive graph to entity relationships in the knowledge graph, the mapping is performed based on the state matrix and relation weight matrix of the probabilistic fuzzy cognitive graph.

8. A knowledge extraction system for operation and maintenance of standardized micro-applications, characterized in that, include: The sample dataset construction module is used to acquire business and operation and maintenance data of each micro-application and preprocess the data to obtain the preprocessed initial sample dataset. The hypergraph building module is used to build a micro-application hypergraph based on the initial sample dataset, with each micro-application as a unit. The Operation and Maintenance Knowledge System Construction Module is used to establish a domain knowledge model. Using the domain knowledge model as a carrier, it constructs a base micro-application operation and maintenance knowledge system and combines the micro-application hypergraph and the base micro-application operation and maintenance knowledge system to construct a micro-application knowledge graph. The fault location module is used to construct a probabilistic fuzzy cognitive graph, use the probabilistic fuzzy cognitive graph to reason and extract operation and maintenance knowledge, and map the fault path reasoned from the probabilistic fuzzy cognitive graph into entity relationships in the micro-application knowledge graph to obtain the fault location and fault cause of abnormal micro-application services.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the operation and maintenance knowledge extraction method for standardized micro-applications as described in any one of claims 1-7.

10. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the operation and maintenance knowledge extraction method for standardized micro-applications as described in any one of claims 1-7.

Images