Transformer fault instance enhancement method based on multi-parameter and multi-index perception
Through the transformer failure example enhancement method based on multi-parameter and multi-index perception, the problems of inefficient and high cost of transformer fault diagnosis in the prior art are solved, the fault instance knowledge base is expanded, the performance of the large model is improved, and the foundation is laid for accurately inferring the cause of transformer failure.
Patent Information
- Application Number
- CN202510134765.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-02-07
- Publication Date
- 2025-05-30
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, transformer fault diagnosis relies on root cause analysis methods, however these methods are inefficient and costly, especially due to the small size of the existing grid knowledge base and the lack of sufficiently diverse transformer fault instance data, limiting the performance of the large model.
Provide a transformer failure instance enhancement method based on multi-parameter and multi-index perception. By combining multiple parameters based on transformer parameter documents, the cause of failure and failure phenomena are generated, and the original fault instance knowledge base is filtered and expanded by using the confidence of the big model and the confidence of the generation of fault instances.
The scale and diversity of the faulty instance knowledge base was successfully expanded, the performance of large models in fault attribution was improved, and the problems of small scale of the existing knowledge base and insufficient instance data were effectively solved, providing a foundation for accurately inferring the cause of transformer failure.
Smart Images
Figure CN120067686A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and particularly to a method for enhancing transformer fault instances based on multi-parameter and multi-index perception. Background Art
[0002] Transformers are crucial components in the power grid system, and their failures may trigger chain reactions, even leading to major accidents in severe cases. In recent years, with the continuous investment and development in the power grid field by the country, the number of devices connected to the power grid has been increasing, and the types and quantities of transformers have also risen accordingly. This trend has significantly increased the cost of transformer fault diagnosis and maintenance. At the same time, transformer faults have complex multiple characteristics, involving multiple factors such as the environment and equipment, making the fault diagnosis process often costly and inefficient. Therefore, developing efficient and intelligent transformer fault diagnosis methods is particularly urgent for the current power grid system. Current fault diagnosis usually relies on root cause analysis technology, that is, fault root cause analysis. This technology aims to trace the root cause by analyzing the abnormal phenomena generated by the fault.
[0003] In the prior art, root cause analysis methods combining machine learning, deep learning, or large language models have been widely applied. Machine learning-based methods require experts to manually construct a causal graph composed of fault causes and phenomena. When a fault occurs, the triggering cause of the fault is judged by the change in the node probability distribution in the graph. However, this method relies on manual cost for graph construction and has low efficiency. Deep learning-based methods judge the fault cause by constructing a neural network model, inputting the fault phenomenon and relevant domain rules, and outputting the fault cause by the model. However, this method requires a large amount of labeled data for supervised training, with low efficiency and high cost. Large language model-based methods rely on a high-quality fault instance knowledge base. When a fault phenomenon appears, relevant instance samples are specifically matched in the knowledge base using this fault phenomenon, and these samples are input to the large model together to assist in root cause reasoning. However, existing power grid knowledge bases are often small in scale and lack sufficient diverse transformer fault instance data for retrieval, restricting the performance of the large model.
[0004] Therefore, in the related art, there is an urgent need for a way to enhance transformer fault instances for root cause analysis. Summary of the Invention
[0005] Based on this, in view of the above technical problems, it is necessary to provide a method for enhancing transformer fault instances for root cause analysis, namely, a method for enhancing transformer fault instances based on multi-parameter and multi-index perception.
[0006] In the first aspect, this application provides a method for enhancing transformer fault instances based on multi-parameter and multi-index perception. The method includes:
[0007] Perform multi-parameter combination based on the transformer parameter document;
[0008] Set the offset situation based on the combination result, generate the fault cause and fault phenomenon, and determine the fault instance based on the corresponding fault cause and fault phenomenon;
[0009] Screen the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance, and expand the original fault instance knowledge base.
[0010] Optionally, in an embodiment of the present application, the multi-parameter combination includes:
[0011] Determine the multi-parameter combination according to the parameter high-frequency combination in the historical transformer fault cases and the number of combined parameters.
[0012] Optionally, in an embodiment of the present application, before setting the offset situation based on the combination result and generating the fault cause and fault phenomenon, it includes:
[0013] Retrieve in the original fault instance knowledge base for different combination results to obtain a generation reference template.
[0014] Optionally, in an embodiment of the present application, screening the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance includes:
[0015] Calculate the confidence based on the word list distribution of the fault phenomenon.
[0016] Optionally, in an embodiment of the present application, screening the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance further includes:
[0017] Calculate the credibility of the generated fault instance based on the semantic similarity between the generated fault instance and the original fault instance.
[0018] In a second aspect, the present application also provides a transformer fault instance enhancement device based on multi-parameter and multi-index perception. The device includes:
[0019] A parameter combination module for performing multi-parameter combination based on the transformer parameter document;
[0020] A fault instance generation module for setting the offset situation based on the combination result, generating the fault cause and fault phenomenon, and determining the fault instance based on the corresponding fault cause and fault phenomenon;
[0021] A multi-index perception instance data screening module for screening the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance, and expanding the original fault instance knowledge base.
[0022] In a third aspect, the present application also provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the steps of the methods described in the above various embodiments.
[0023] In a fourth aspect, the present application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, it implements the steps of the methods described in the above various embodiments.
[0024] For the above method for enhancing transformer fault instances based on multi-parameter and multi-index perception, first, multi-parameter combinations are performed based on the transformer parameter document; then, based on the combination results, offset situations are set to generate fault causes and fault phenomena, and fault instances are determined based on the corresponding fault causes and fault phenomena; finally, the fault causes and fault phenomena are screened based on the confidence of the large model and the credibility of the generated fault instances, and the original fault instance knowledge base is expanded. That is to say, first, the large model is guided by the parameter document to generate power fault instance samples, and then semantic similarity and the confidence of the large model generation are used to screen out effective instance samples, successfully expanding the scale and diversity of the fault instance knowledge base, thereby improving the performance of the large model based on the retrieval enhancement paradigm in fault attribution. It effectively solves the challenges of the small scale of the existing knowledge base and insufficient instance data, laying a foundation for accurately inferring the causes of transformer faults. Description of the Drawings
[0025] Figure 1 It is an application environment diagram of a method for enhancing transformer fault instances based on multi-parameter and multi-index perception in an embodiment;
[0026] Figure 2 It is a schematic flowchart of a method for enhancing transformer fault instances based on multi-parameter and multi-index perception in an embodiment;
[0027] Figure 3 It is a structural block diagram of a device for enhancing transformer fault instances based on multi-parameter and multi-index perception in an embodiment;
[0028] Figure 4 It is an internal structure diagram of a computer device in an embodiment. Detailed Embodiments
[0029] In order to make the objectives, technical solutions, and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.
[0030] An enhanced method for transformer fault instances based on multi-parameter and multi-index perception provided by an embodiment of the present application can be applied to, for example, Figure 1 the application environment shown. Among them, the terminal communicates with the server through the network. The data storage system can store the data that the server needs to process. The data storage system can be integrated on the server, or can be placed in the cloud or other network servers. Among them, the terminal can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices, and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The server can be implemented by an independent server or a server cluster composed of multiple servers.
[0031] In one embodiment, as shown in Figure 2 a method for enhancing transformer fault instances based on multi-parameter and multi-index perception is provided. Taking the server in Figure 1 as an example for illustration, the method includes the following steps:
[0032] S201: Perform multi-parameter combination based on the transformer parameter document.
[0033] In the embodiment of the present application, transformers often have very strict parameter settings. The faults of transformers are often due to the deviation of one or more of its parameters from the original parameter setting standards, such as too low ambient temperature, overload of downstream equipment, etc. These parameter deviations constitute the causes of transformer faults and induce the occurrence of fault phenomena. The parameters of the transformer are often recorded in the parameter document D in the form of key-value pairs, where D = {(a 1 , c 1 ), (a 2 , c 2 ),..(a m , c m )}, (a m , c m ) represents the mth parameter a m and its corresponding numerical setting c m , for example, a m can represent the ambient temperature, while c m represents the temperature value specifically adapted by the transformer. The transformer fault instance knowledge base is also represented in the form of key-value pairs, that is, H = {(s 1 , r 1 ), (s 2 , r 2 ), …(s i , r i )}, (s i , r irespectively represent the transformer fault phenomena and their corresponding fault causes. In a real scenario, transformer faults are often not caused by the deviation of a single parameter, but by the simultaneous deviation of multiple parameters from their normal ranges or the result of their collaborative action. Therefore, based on the transformer parameter document, multiple parameters are introduced and randomly combined to be used for interactive modeling to simulate complex fault situations.
[0034] Specifically, in an embodiment of the present application, the multi-parameter combination includes:
[0035] Determine the multi-parameter combination according to the high-frequency combination of parameters and the number of combined parameters in historical transformer fault cases.
[0036] In an embodiment of the present application, for the parameters in document D, according to the high-frequency combination of parameters and the number of combined parameters (2-4) in historical transformer fault cases, several parameters are randomly selected, and at the same time, the parameters in the high-frequency combination are added as the input data of the fault instance to be generated. The m-th multi-parameter combination is denoted as d m ={a i , a i+1 …}, and the corresponding numerical setting combination is v m ={c i , c i+1 …}.
[0037] In an embodiment of the present application, before setting the offset situation based on the combination result and generating the fault cause and fault phenomenon, it includes:
[0038] For different combination results, retrieve in the original fault instance knowledge base to obtain a generation reference template.
[0039] In an embodiment of the present application, for different parameter combinations, retrieve respectively in the existing original fault instance knowledge base H, find historical fault cases similar to the current parameter combination situation, and analyze and integrate the fault causes in these cases as the reference template for generating new fault causes.
[0040] S203: Set the offset situation based on the combination result, generate the fault cause and fault phenomenon, and determine the fault instance based on the corresponding fault cause and fault phenomenon.
[0041] In the embodiment of the present application, by traversing the parameter combinations, the large model is called using a predefined Prompt. The large model generates targeted and credible fault causes according to the reference template and in combination with the specific offset situation of the current parameters. The specific process of the Prompt call is shown in Table 1.
[0042] Table 1 Prompt construction: Fault cause generation
[0043]
[0044] In the above way, for the parameter combination (d m , v m ), the fault cause r generated by the large model can be obtained m . Subsequently, combined with the examples in the knowledge base H, the large model generates the fault cause r through few-shot learning m that will cause the transformer fault phenomenon s m . The specific process of Prompt call is shown in Table 2
[0045] Table 2 Prompt construction: Fault phenomenon generation
[0046]
[0047] Among them, (s n , r n ) ∈ H. In the above way, for the fault cause r m , the fault phenomenon s induced by it can be obtained m . And finally, the fault instance pair (s m , r m ) generated by the large model based on the transformer parameter combination (d m , v m ) is obtained. All the enhanced instance data obtained in this process is defined as H ′ = {(s m1 , r m1 ), (s m2 , r m2 ),...}
[0048] S205: Screen the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance, and expand the original fault instance knowledge base
[0049] In the embodiment of the present application, based on the confidence of the large model and the credibility of the generated instance, the generated fault cause and fault phenomenon are dynamically screened, and finally a batch of high-quality fault instances are filtered out as the expansion of the original knowledge base H
[0050] In an embodiment of the present application, the screening of the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance includes
[0051] Calculating the confidence based on the vocabulary distribution of the fault phenomenon
[0052] In an embodiment of the present application, the hallucination problem of the large model always affects the quality of the data it generates. When the large model has hallucinations, its confidence is often low, and the quality of the generated data is also poor. First, with reference to the confidence score of the large model when generating instances, the generation results with confidence scores lower than the preset threshold θ are excluded, so as to ensure the quality of the generated instance data. Specifically, based on the vocabulary distribution when calling the large model to generate the fault phenomenon s m calculate its confidence p m when m , and define p m as the lowest probability value when generating each character of s m . In other words, if the probability of each character generated by the large model in s m is greater than θ, the large model is confident; otherwise, it is not confident. In addition, according to the overall distribution of the generated fault instance data, the confidence screening threshold θ is dynamically adjusted. For example, if most of the confidence scores of the generated instance data are high and the distribution is relatively concentrated, the threshold θ will be increased to further screen out the highest-quality instances; conversely, if the confidence score distribution is relatively dispersed and there are more instances with lower scores, the threshold θ will be decreased to avoid excessive screening resulting in too much data loss.
[0053] In an embodiment of the present application, the screening of the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instances further includes:
[0054] Calculating the credibility of the generated fault instance based on the semantic similarity between the generated fault instance and the original fault instance.
[0055] In an embodiment of the present application, the generated fault instance and the instances in the existing original fault instance knowledge base H are obtained through the embedding model to obtain their semantic embedding vectors, and the cosine similarity sim m between the vectors is calculated. If the similarity is less than the threshold γ (set to 0.5 in this example), it cannot be guaranteed that the generated transformer fault instance conforms to the real logic, and the generated instance will be temporarily discarded, and further joint decision-making with experts is carried out to determine whether to retain the instance. The specific calculation formula of the cosine similarity is as follows.
[0056] sim i =cos(E(s i +r i ,α),E(s m +r m ,α))
[0057] where E represents the pre-trained embedding model encoder, α represents the parameters of the encoder, cos represents the calculation of the cosine similarity of the vectors, s m and r mRepresents the generated fault phenomena and fault causes, s i and r i Represents the fault phenomena and fault causes in the original fault instance knowledge base, sim i Represents the similarity score between the i-th instance in H and the fault instance.
[0058] Finally, through the joint screening of multiple indicators, the enhanced fault instance H of the large model is obtained "" , and this process is represented by the following formula:
[0059] H″ = (s mi , r mi ) │ (s mi , r mi ) ∈ H′ & p mi > θ & sim mi > γ
[0060] Subsequently, the original knowledge base H and the knowledge base H″ enhanced by the large model are merged to obtain the new transformer fault instance knowledge base H * = H ∪ H″.
[0061] In the above method for enhancing transformer fault instances based on multi-parameter and multi-index perception, first, multi-parameter combinations are performed based on the transformer parameter document; then, the offset situation is set based on the combination result to generate fault causes and fault phenomena, and fault instances are determined based on the corresponding fault causes and fault phenomena; finally, the fault causes and fault phenomena are screened based on the confidence of the large model and the credibility of the generated fault instances, and the original fault instance knowledge base is expanded. That is to say, first, the large model is guided by the parameter document to generate power fault instance samples, and then semantic similarity and the confidence of the large model generation are used to screen out effective instance samples, successfully expanding the scale and diversity of the fault instance knowledge base, thereby improving the performance of the large model in fault attribution based on the retrieval-enhanced paradigm. It effectively solves the challenges of the small scale of the existing knowledge base and insufficient instance data, and lays a foundation for accurately inferring the causes of transformer faults.
[0062] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this document, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.
[0063] Based on the same inventive concept, an embodiment of the present application further provides a device for enhancing transformer fault instances based on multi-parameter and multi-index perception for implementing the method for enhancing transformer fault instances based on multi-parameter and multi-index perception described above. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the device for enhancing transformer fault instances based on multi-parameter and multi-index perception provided below can refer to the limitations on the method for enhancing transformer fault instances based on multi-parameter and multi-index perception in the above text, and will not be repeated here.
[0064] In one embodiment, as Figure 3 shown, a device 300 for enhancing transformer fault instances based on multi-parameter and multi-index perception is provided, including: a parameter combination module 301, a fault instance generation module 303, and a multi-index perception instance data screening module 305, where:
[0065] The parameter combination module 301 is used to perform multi-parameter combination based on the transformer parameter document.
[0066] The fault instance generation module 303 is used to set the offset situation based on the combination result, generate the fault cause and the fault phenomenon, and determine the fault instance based on the corresponding fault cause and fault phenomenon.
[0067] The multi-index perception instance data screening module 305 is used to screen the fault cause and the fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance, and expand the original fault instance knowledge base.
[0068] In an embodiment of the present application, the parameter combination module is further used for:
[0069] Determine the multi-parameter combination according to the high-frequency combination of parameters and the number of combined parameters in the historical transformer fault cases.
[0070] In one embodiment of the present application, the fault instance generation module is further configured to:
[0071] For different combination results, retrieve in the original fault instance knowledge base to obtain a generation reference template.
[0072] In one embodiment of the present application, the instance data screening module for multi-index perception is further configured to:
[0073] Calculate the confidence based on the word list distribution of the fault phenomenon.
[0074] In one embodiment of the present application, the instance data screening module for multi-index perception is further configured to:
[0075] Calculate the credibility of the generated fault instance based on the semantic similarity between the generated fault instance and the original fault instance.
[0076] Each module in the above-mentioned transformer fault instance enhancement device based on multi-parameter and multi-index perception can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above-mentioned modules.
[0077] In one embodiment, a computer device is provided. The computer device can be a terminal, and its internal structure diagram can be as Figure 4 shown. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner. The wireless manner can be achieved through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for enhancing transformer fault instances based on multi-parameters and multi-index perception. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, a touchpad, or a mouse, etc.
[0078] Those skilled in the art can understand, Figure 4The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.
[0079] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.
[0080] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0081] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the steps in the above method embodiments are implemented.
[0082] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties.
[0083] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include Read-Only Memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.
[0084] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.
[0085] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.
Claims
1. A transformer fault instance enhancement method based on multi-parameter and multi-index perception, characterized in that: The method comprises: Combine multiple parameters based on transformer parameter documents; Based on the combination result, an offset condition is set, a fault cause and a fault phenomenon are generated, and a fault instance is determined based on the corresponding fault cause and fault phenomenon; Based on the confidence of the large model and the credibility of the generated fault instance, the fault cause and fault phenomenon are screened to expand the original fault instance knowledge base.
2. A transformer fault instance enhancement method based on multi-parameter and multi-index perception according to claim 1, characterized in that: The multi-parameter combination includes: According to the high-frequency combinations of parameters in historical transformer failure cases and the number of combined parameters, a multi-parameter combination is determined.
3. A transformer fault instance enhancement method based on multi-parameter and multi-index perception according to claim 1, characterized in that: The step of setting the offset condition based on the combined result and generating the fault cause and fault phenomenon includes: According to different combination results, the original fault instance knowledge base is searched to obtain the generated reference template.
4. The transformer fault instance enhancement method based on multi-parameter and multi-index perception according to claim 1 is characterized in that: The screening of the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance includes: The confidence level is calculated based on the vocabulary distribution of the fault phenomenon.
5. The transformer fault instance enhancement method based on multi-parameter and multi-index perception according to claim 1 is characterized in that: The screening of the fault cause and fault phenomenon based on the confidence of the large model and the credibility of the generated fault instance also includes: The credibility of the generated fault instance is calculated based on the semantic similarity between the generated fault instance and the original fault instance.
6. A transformer fault instance enhancement device based on multi-parameter and multi-index perception, characterized in that: The device comprises: Parameter combination module, used to combine multiple parameters based on transformer parameter documents; A fault instance generation module, used to set an offset condition based on the combination result, generate a fault cause and a fault phenomenon, and determine a fault instance based on the corresponding fault cause and fault phenomenon; The multi-indicator-aware instance data screening module is used to screen the fault causes and fault phenomena based on the confidence of the large model and the credibility of the generated fault instance, and expand the original fault instance knowledge base.
7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.