Fault diagnosis method, electronic device, medium, and computer program product

Abstract

The embodiment of the application provides a fault diagnosis method, comprising: receiving a first fault description; obtaining a first sentence vector corresponding to the first fault description; obtaining a first knowledge graph; wherein the first knowledge graph comprises a first set of fault descriptions and diagnosis information associated with each fault description in the first set of fault descriptions; calculating a sentence vector corresponding to each fault description in the first knowledge graph; comparing the first sentence vector with the sentence vector corresponding to each fault description in the first knowledge graph one by one to obtain a second sentence vector with the highest similarity; and outputting diagnosis information associated with the fault description corresponding to the second sentence vector in the first knowledge graph.

Description

Technical Field

[0001] The embodiments of this application mainly relate to the field of power systems, and in particular to a fault diagnosis method, electronic equipment, medium, and computer program product. Background Technology

[0002] Due to the combined effects of various factors, electrical equipment is prone to malfunctions during operation, thus affecting its normal operation. Therefore, fault diagnosis is crucial for the proper functioning of electrical equipment. However, in complex fault / event situations, rapid fault diagnosis is often difficult. Traditional methods rely on manually consulting product manuals for fault diagnosis knowledge or on the past experience of maintenance personnel. Clearly, traditional methods are not only inefficient but also cannot guarantee accuracy. Summary of the Invention

[0003] This application provides a fault diagnosis method, electronic device, medium, and computer program product, which can accurately and quickly complete the fault diagnosis of the device.

[0004] In a first aspect, a fault diagnosis method is provided, comprising: receiving a first fault description; obtaining a first sentence vector corresponding to the first fault description; obtaining a first knowledge graph; wherein the first knowledge graph includes a first group of fault descriptions and diagnostic information associated with each fault description in the first group of fault descriptions; calculating a sentence vector corresponding to each fault description in the first knowledge graph; comparing the similarity of the first sentence vector with the sentence vector corresponding to each fault description in the first knowledge graph one by one to obtain a second sentence vector with the highest similarity; and outputting the diagnostic information associated with the fault description corresponding to the second sentence vector in the first knowledge graph.

[0005] In a second aspect, an electronic device is provided, comprising: at least one memory configured to store computer-readable code; and at least one processor configured to invoke the computer-readable code to perform the steps of the method provided in the first aspect.

[0006] Thirdly, a computer-readable medium is provided, on which computer-readable instructions are stored, which, when executed by a processor, cause the processor to perform the steps of the method provided in the first aspect.

[0007] Fourthly, a computer program product is provided, the computer program product being tangibly stored on a computer-readable medium and including computer-executable instructions that, when executed, cause at least one processor to perform the steps of the method provided in the first aspect. Attached Figure Description

[0008] The following figures are intended only to illustrate and explain the embodiments of this application, and do not limit the scope of the embodiments of this application. Wherein:

[0009] Figure 1 This is a flowchart of a fault diagnosis method according to an embodiment of this application;

[0010] Figure 2 This is a schematic diagram of an electronic device according to an embodiment of this application.

[0011] Explanation of reference numerals in the attached figures

[0012] 100: Fault Diagnosis Methods 101-106: Method Steps

[0013] 200: Electronic device; 201: Processor; 202: Communication interface

[0014] 203: Memory; 204: Communication bus; 205: Program Detailed Implementation

[0015] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed merely to enable those skilled in the art to better understand and implement the subject matter described herein, and are not intended to limit the scope, applicability, or examples set forth in the claims. The function and arrangement of the elements discussed may be changed without departing from the scope of the embodiments of this application. Various processes or components may be omitted, substituted, or added as needed in the various examples. For example, the described methods may be performed in a different order than described, and steps may be added, omitted, or combined. Furthermore, features described in some examples may be combined in other examples.

[0016] As used herein, the term "comprising" and its variations are open terms meaning "including but not limited to". The term "based on" means "at least partially based on". The terms "one embodiment" and "an embodiment" mean "at least one embodiment". The term "another embodiment" means "at least one other embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other definitions, whether explicit or implicit, may be included below. Unless explicitly indicated by the context, the definition of a term shall remain consistent throughout the specification.

[0017] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0018] Figure 1 This is a flowchart of a fault diagnosis method according to an embodiment of this application, such as... Figure 1 As shown, the fault diagnosis method 100 includes:

[0019] Step 101: Receive the first fault description.

[0020] The first fault description can be entered by the user or obtained by reading alarm information from the system or equipment. For example, in a circuit breaker fault diagnosis scenario, the first set of fault descriptions could be natural language descriptions such as: the cabinet door cannot be opened, the circuit breaker cannot be closed, the circuit breaker temperature rises too high, or the circuit breaker cannot be closed after pressing the close button.

[0021] Step 102: Obtain the first sentence vector corresponding to the first fault description.

[0022] Preferably, the first fault description is segmented into multiple sub-words. Each sub-word is then converted into a corresponding word vector using a first word embedding model, resulting in multiple word vectors. Based on these multiple word vectors, the first sentence vector is obtained.

[0023] Alternatively, the first sentence vector can be obtained by averaging multiple word vectors.

[0024] Optionally, the first sentence vector can be obtained by calculating a weighted average of multiple word vectors. The corresponding weights can be pre-set based on the relevance of the words.

[0025] Alternatively, the first fault description can be directly converted into the first sentence vector using a pre-trained language representation model, such as a Bidirectional Encoder Representations from Transformers (BERT) model.

[0026] In one embodiment, before converting each word among multiple sub-words into corresponding word vectors using the first word embedding model, information related to the domain of the first knowledge graph can be input into a designated word embedding model for incremental training to enhance the designated word embedding model, thus obtaining the first word embedding model. By enhancing the designated word embedding model, its output related word vectors can be made more accurate.

[0027] In one embodiment, before converting each word in a plurality of sub-words into a corresponding word vector using a first word embedding model, the similarity of each word in the specified word embedding model is compared with that of each word in the first knowledge graph. Words in the specified word embedding model whose similarity exceeds a preset threshold are extracted to obtain the first word embedding model. This first embedding model is more lightweight and occupies less computing and storage resources.

[0028] In one embodiment, before converting each word among multiple sub-words into corresponding word vectors using the first word embedding model, information related to the domain of the first knowledge graph can be input into the designated word embedding model for incremental training to enhance the designated word embedding model. Then, each word in the enhanced designated word embedding model is compared with each word in the first knowledge graph for similarity. Words in the designated word embedding model whose similarity exceeds a preset threshold are extracted, thus obtaining the first word embedding model.

[0029] Optionally, information related to the domain to which the first knowledge graph belongs, such as papers, patents, or textbooks.

[0030] Step 103: Obtain the first knowledge graph. This first knowledge graph includes a first set of fault descriptions and diagnostic information associated with each fault description in the first set of fault descriptions.

[0031] Diagnostic information includes: cause and / or solution.

[0032] Optionally, a first fault cause table is input into a relational database to obtain a first file. The first fault cause table contains the fault cause corresponding to each fault description in the first group of fault descriptions. The first file is input into a mapping relation generation tool to obtain a first mapping file. The first mapping file is instantiated using a first ontology to obtain a second mapping file. The first ontology includes classes corresponding to fault descriptions and classes corresponding to fault causes. Specifically, based on the classes corresponding to fault descriptions and fault causes in the first ontology, the fault descriptions and fault causes in the first mapping file are instantiated separately in the first ontology to obtain the second mapping file. The second mapping file and the first file are input into a data format conversion tool to obtain a first knowledge graph.

[0033] Optionally, a first fault solution table is input into a relational database to obtain a first file. The first fault cause table contains fault solutions corresponding to each fault description in the first group of fault descriptions. The first file is input into a mapping relationship generation tool to obtain a first mapping file. The first mapping file is instantiated using a first ontology to obtain a second mapping file. The first ontology includes classes corresponding to fault descriptions and classes corresponding to fault solutions. The second mapping file and the first file are input into a data format conversion tool to obtain a first knowledge graph.

[0034] Storing various types of data in the form of knowledge graphs not only offers excellent scalability and strong expressive power, but also solves the problem of multi-source and heterogeneous data being unable to be accessed uniformly. Furthermore, it improves knowledge retrieval efficiency and facilitates semantic retrieval.

[0035] In one embodiment, the first set of fault descriptions in the first knowledge graph can be obtained based on historical fault events stored in the fault information collection system. The historical fault events are then input into a natural language processing model, such as the ChatGPT model, to obtain the diagnostic information corresponding to the fault descriptions.

[0036] Step 104: Calculate the sentence vector corresponding to each fault description in the first knowledge graph.

[0037] Step 105: Compare the similarity of the first sentence vector with the sentence vector corresponding to each fault description in the first knowledge graph to obtain the second sentence vector with the highest similarity.

[0038] Step 106: Output the diagnostic information associated with the fault description corresponding to the second sentence vector in the first knowledge graph.

[0039] Optionally, the output can be in the form of a subgraph.

[0040] This application proposes an automated fault diagnosis scheme. It obtains the sentence vector corresponding to the received first fault description, compares this sentence vector with the sentence vectors corresponding to each fault description in a first knowledge graph for similarity, and determines the associated diagnostic information in the first knowledge graph based on the fault description corresponding to the sentence vector with the highest similarity. Finally, the diagnostic information is output. This application can quickly and accurately output the corresponding diagnostic information based solely on the fault description. Furthermore, using a knowledge graph as the data carrier gives this application good scalability. If the diagnostic information includes multiple diagnostic information items associated with the fault description in the knowledge graph, such as fault causes and fault solutions, then when a user inputs a fault description text, this method can provide a wealth of relevant diagnostic information, thereby significantly saving various costs.

[0041] Figure 2 This is a schematic diagram of an electronic device according to one embodiment of this application. The specific embodiments of this application do not limit the specific implementation of the electronic device. Figure 2 As shown, the electronic device 200 may include: a processor 201, a communications interface 202, a memory 203, and a communication bus 204. Wherein:

[0042] The processor 201, communication interface 202, and memory 203 communicate with each other through the communication bus 204.

[0043] Communication interface 202 is used to communicate with other electronic devices or servers.

[0044] The processor 201 is used to execute program 205, which can specifically execute the relevant steps in any of the aforementioned method embodiments.

[0045] Specifically, program 205 may include program code that includes computer operation instructions.

[0046] The processor 201 may be a CPU, an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The smart device may include one or more processors of the same type, such as one or more CPUs; or it may include processors of different types, such as one or more CPUs and one or more ASICs.

[0047] Memory 203 is used to store program 205. Memory 203 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0048] Specifically, program 205 can be used to cause processor 201 to execute any of the methods in the foregoing embodiments.

[0049] The specific implementation of each step in program 205 can be found in the corresponding steps and units described in the aforementioned data interaction model update method embodiment, and will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the devices and modules described above can be referred to the corresponding process descriptions in the aforementioned method embodiment, and will not be repeated here.

[0050] This application also provides a computer-readable storage medium storing instructions for causing a machine to perform any of the methods described in the various method embodiments herein. Specifically, a system or apparatus equipped with a storage medium storing software program code that implements the functions of any of the embodiments described above, and enabling the computer (or CPU or MPU) of the system or apparatus to read and execute the program code stored in the storage medium.

[0051] In this case, the program code read from the storage medium can itself implement the function of any of the above embodiments, and therefore the program code and the storage medium storing the program code constitute part of this application.

[0052] Examples of storage media used to provide program code include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer via a communication network.

[0053] This application also provides a computer program product, including computer instructions that instruct a computing device to perform any corresponding operation in the above-described plurality of method embodiments.

[0054] It should be noted that, depending on the implementation needs, the various components / steps described in the embodiments of this application can be broken down into more components / steps, or two or more components / steps or parts of the operation of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of this application.

[0055] The methods described in the embodiments of this application can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code downloaded over a network that is originally stored in a remote recording medium or a non-transitory machine-readable medium and will be stored in a local recording medium. Thus, the methods described herein can be processed by software stored on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an ASIC or FPGA). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., RAM, ROM, flash memory, etc.) capable of storing or receiving software or computer code that, when accessed and executed by the computer, processor, or hardware, implements the methods described herein. Furthermore, when a general-purpose computer accesses code used to implement the methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for executing the methods shown herein.

[0056] It should be noted that not all steps and modules in the above processes and system structure diagrams are necessary; some steps or modules can be omitted as needed. The execution order of each step is not fixed and can be adjusted as required. The system structure described in the above embodiments can be a physical structure or a logical structure. That is, some modules may be implemented by the same physical entity, or some modules may be implemented by multiple physical entities, or they may be jointly implemented by certain components in multiple independent devices.

[0057] In the above embodiments, the hardware modules can be implemented mechanically or electrically. For example, a hardware module may include permanent dedicated circuitry or logic (such as a dedicated processor, FPGA, or ASIC) to perform the corresponding operations. The hardware module may also include programmable logic or circuitry (such as a general-purpose processor or other programmable processor), which can be temporarily configured by software to perform the corresponding operations. The specific implementation method (mechanical, dedicated permanent circuitry, or temporarily configured circuitry) can be determined based on cost and time considerations.

[0058] The present invention has been shown and described in detail above with reference to the accompanying drawings and preferred embodiments. However, the present invention is not limited to these disclosed embodiments. Based on the above multiple embodiments, those skilled in the art will know that more embodiments of the present invention can be obtained by combining the code review methods in the different embodiments above. These embodiments are also within the protection scope of the present invention.

Claims

1. A fault diagnosis method, comprising: - Receive (101) a first fault description; the first fault description is input by the user or obtained by reading alarm information from the system or device; - Obtain the first sentence vector corresponding to the first fault description (102); - Obtain (103) a first knowledge graph; wherein the first knowledge graph includes a first set of fault descriptions and diagnostic information associated with each fault description in the first set of fault descriptions; - Calculate the sentence vector corresponding to each fault description in the first knowledge graph (104); - Compare the similarity of the first sentence vector with the sentence vector corresponding to each fault description in the first knowledge graph (105) to obtain the second sentence vector with the highest similarity. - Output (106) the diagnostic information associated with the fault description corresponding to the second sentence vector in the first knowledge graph; Wherein, obtaining the first sentence vector corresponding to the first fault description (102) includes: directly converting the first fault description into a first sentence vector through a pre-trained language representation model; or includes: - Segment the first fault description into multiple smaller words; - Each word in the plurality of words is converted into a corresponding word vector by the first word embedding model, resulting in multiple word vectors; - Based on the multiple word vectors, the first sentence vector is obtained.

2. The method according to claim 1, wherein, The diagnostic information includes: cause and / or solution.

3. The method according to claim 2, wherein, The acquisition of the (103) first knowledge graph includes: - Input the first fault cause table into the relational database to obtain the first file; wherein, the first fault cause table contains the fault cause corresponding to each fault description in the first group of fault descriptions; - Input the first file into the mapping relationship generation tool to obtain the first mapping file; - The first mapping file is instantiated using the first ontology to obtain the second mapping file; wherein, the first ontology includes a class corresponding to the fault description and a class corresponding to the fault cause; - Input the second mapping file and the first file into a data format conversion tool to obtain the first knowledge graph.

4. The method according to claim 3, wherein, The acquisition of the (103) first knowledge graph includes: - Input the first fault solution table into the relational database to obtain the first file; wherein, the first fault cause table contains the fault solution corresponding to each fault description in the first group of fault descriptions; - Input the first file into the mapping relationship generation tool to obtain the first mapping file; - The first mapping file is instantiated using the first ontology to obtain the second mapping file; wherein, the first ontology includes a class corresponding to the fault description and a class corresponding to the fault solution; - Input the second mapping file and the first file into a data format conversion tool to obtain the first knowledge graph.

5. The method according to claim 1, wherein, The process of obtaining the first sentence vector based on the multiple word vectors includes: - Calculate the average of the multiple word vectors; - Use the calculated average value as the first vector.

6. The method according to claim 1, wherein, Before converting each of the plurality of sub-words into a corresponding word vector using the first word embedding model, the method further includes: - Input information related to the domain of the first knowledge graph into a specified word embedding model for incremental training to enhance the specified word embedding model, thereby obtaining the first word embedding model; or - Compare the similarity between each word embedded in the model and each word in the first knowledge graph; - Extract the words corresponding to the specified word embedding model whose similarity exceeds a preset threshold to obtain the first word embedding model.

7. The method according to claim 1, wherein, Before converting each of the plurality of sub-words into a corresponding word vector using the first word embedding model, the method further includes: - Input information related to the domain to which the first knowledge graph belongs into the specified word embedding model for incremental training to enhance the specified word embedding model; - Compare the similarity between each word in the enhanced specified word embedded in the model and each word in the first knowledge graph; - Extract the words corresponding to the specified word embedding model whose similarity exceeds a preset threshold to obtain the first word embedding model.

8. An electronic device (200), comprising: The processor (201), communication interface (202), memory (203), and communication bus (204) communicate with each other through the communication bus (204). The memory (203) is used to store at least one executable instruction that causes the processor to perform the operation corresponding to the fault diagnosis method as described in any one of claims 1-7.

9. A computer storage medium having a computer program stored thereon, which, when executed by a processor, implements the fault diagnosis method as described in any one of claims 1-7.

10. A computer program product tangibly stored on a computer-readable medium and comprising computer-executable instructions that, when executed, cause at least one processor to perform the fault diagnosis method according to any one of claims 1-7.

Images