Fault maintenance method, device, equipment, storage medium and computer program product

Abstract

The application discloses a fault maintenance method and device, equipment, a storage medium and a computer program product, and relates to the technical field of computers. The fault maintenance method comprises the following steps: when a query question about the cause of a fault of a target device is received, performing intention recognition on the query question by using a target language large model to obtain an intention recognition result; calling a vocabulary resource library according to the intention recognition result, and querying a fault maintenance knowledge graph generated based on a target document large model according to a target vocabulary obtained by calling, to obtain a plurality of answer cases; performing network resource retrieval according to the answer cases to obtain a network retrieval result of fault classification; determining fault classification according to the answer cases and the network retrieval result, and generating a fault solution recommendation scheme according to the fault classification. In this way, customized and accurate fault elimination suggestions are provided, and the efficiency and accuracy of fault maintenance are improved by relying on a knowledge graph and a large model.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, in particular to a fault maintenance method, device, equipment, storage medium and computer program product. BACKGROUND

[0002] The fault maintenance of various high-end mechanical equipment, vehicles and technological products is a highly professional and complex task, which requires maintenance personnel to have solid theoretical knowledge and practical experience, and to be able to quickly and accurately locate the fault cause and take effective measures. The traditional maintenance mode relies on manual checking of a large number of technical materials and past cases, which is time-consuming and prone to errors.

[0003] The above content is only used to assist in understanding the technical solutions of the present application, and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0004] The main purpose of the present application is to provide a fault maintenance method, device, equipment, storage medium and computer program product, which aims to solve the technical problem of long time-consuming and easy-to-error in device fault troubleshooting.

[0005] To achieve the above purpose, the present application provides a fault maintenance method, which comprises:

[0006] When receiving a query question about the fault cause of a target device, performing intent recognition on the query question through a target language large model to obtain an intent recognition result;

[0007] According to the intent recognition result, calling a vocabulary resource library, and querying a fault maintenance knowledge graph generated based on a target document large model according to the target vocabulary obtained by calling, to obtain a plurality of answer cases;

[0008] According to the answer cases, performing network resource retrieval to obtain a network retrieval result of fault classification;

[0009] According to the answer cases and the network retrieval result, determining a fault classification, and generating a fault solution recommendation scheme according to the fault classification.

[0010] Optionally, before the step of calling the vocabulary resource library according to the intent recognition result, and querying the fault maintenance knowledge graph generated based on the target document large model according to the target vocabulary obtained by calling to obtain a plurality of answer cases, the method further comprises:

[0011] Dividing the content of a preset maintenance manual into a plurality of long text segments, and inputting the long text segments into a target document large model for fragmentation processing, to obtain maintenance manual content according to the processing result;

[0012] Generating a historical fault knowledge base according to historical fault data;

[0013] A fault maintenance knowledge graph is generated based on the historical fault knowledge base and the maintenance manual content.

[0014] Optionally, the step of dividing the content of the preset maintenance manual into a plurality of long text segments includes:

[0015] Obtaining a text file of a preset maintenance manual, and extracting the content of the target manual based on the text file, wherein the content extraction includes paragraph, picture, table and page number extraction;

[0016] Perform label extraction and summary extraction based on the target manual content to obtain the analysis manual content;

[0017] A plurality of long text segments are obtained based on the analysis of the manual content.

[0018] The step of obtaining a plurality of long text segments according to the content of the analysis manual comprises:

[0019] determining manual semantic information based on the analysis of the manual content;

[0020] The analyzed manual content is divided into a plurality of long text segments according to the manual semantic information.

[0021] Optionally, the step of generating a historical fault knowledge base based on historical fault data includes:

[0022] Converting the historical fault data into a question-answer pair format to obtain question-answer pair data;

[0023] Generate a historical fault file based on the question and answer data;

[0024] A historical fault knowledge base is generated according to the historical fault files.

[0025] Optionally, the step of generating a fault maintenance knowledge graph based on the historical fault knowledge base and the maintenance manual content includes:

[0026] Obtain maintenance keywords according to the maintenance manual content;

[0027] generating a keyword table according to the maintenance keywords;

[0028] The keyword table and the historical fault knowledge base are imported into the initial knowledge graph to obtain a fault maintenance knowledge graph.

[0029] Optionally, upon receiving a query question about a cause of a target device failure, performing intent recognition on the query question using a target language large model to obtain an intent recognition result includes:

[0030] The query question about the target device fault reason is input into a target language large model to obtain a user intention when the query question about the target device fault reason is received;

[0031] It is judged whether to call a vocabulary resource library according to the user intention, and an intention recognition result is obtained according to a judgment result.

[0032] Optionally, the step of inputting the query question about the target device fault reason into the target language large model to obtain the user intention when the query question about the target device fault reason is received comprises:

[0033] The query prompt word is determined according to the query question about the target device fault reason when the query question about the target device fault reason is received;

[0034] The subject name information in the query prompt word is extracted;

[0035] The subject name information and the query question are input into the target language large model to obtain the user intention.

[0036] Optionally, the step of calling the vocabulary resource library according to the intention recognition result, and querying a fault maintenance knowledge graph generated based on a target document large model according to a target vocabulary obtained by calling, to obtain a plurality of answer cases comprises:

[0037] The vocabulary resource library is called according to the intention recognition result, to obtain a plurality of target vocabularies corresponding to the query question;

[0038] The fault maintenance knowledge graph is queried according to the target vocabulary, to obtain a plurality of vocabulary-related questions;

[0039] The plurality of answer cases are obtained according to the vocabulary-related questions.

[0040] Optionally, the step of performing network resource retrieval according to the answer cases to obtain a network retrieval result of the fault classification comprises:

[0041] The answer cases are screened through a target agent node to obtain a screening result;

[0042] The fault elimination chapter number corresponding to a preset maintenance manual is determined according to the screening result;

[0043] The target network resource is retrieved according to the fault elimination chapter number to obtain the network retrieval result.

[0044] In addition, in order to achieve the above-mentioned purpose, the application further provides a fault maintenance device, which comprises:

[0045] An intention recognition module is configured to perform intention recognition on the query question about the target device fault reason through a target language large model to obtain an intention recognition result when the query question about the target device fault reason is received.

[0046] an answer recall module configured to invoke a vocabulary resource library according to the intention recognition result, query a fault maintenance knowledge graph generated based on a target document large model according to a target vocabulary obtained by the invocation, and obtain a plurality of answer recall results;

[0047] a network retrieval module configured to perform network resource retrieval according to the answer recall results and obtain network retrieval results of fault classification;

[0048] a scheme generation module configured to determine fault classification according to the answer recall results and the network retrieval results, and generate a fault solution recommendation scheme according to the fault classification.

[0049] Optionally, the answer recall module is further configured to divide content of a preset maintenance manual into a plurality of long text segments, input the long text segments into a target document large model for fragmentation processing, obtain maintenance manual content according to a processing result, generate a historical fault knowledge base according to historical fault data, and generate a fault maintenance knowledge graph according to the historical fault knowledge base and the maintenance manual content.

[0050] Optionally, the answer recall module is further configured to obtain a text file of a preset maintenance manual, extract target manual content according to the text file, and obtain analyzed manual content by performing label extraction and summary extraction on the target manual content; the content extraction includes paragraph, picture, table, and page number extraction; and the answer recall module is further configured to obtain a plurality of long text segments from the analyzed manual content, input the long text segments into a target document large model for fragmentation processing, and obtain maintenance manual content according to a processing result.

[0051] Optionally, the answer recall module is further configured to determine manual semantic information according to the analyzed manual content, divide the analyzed manual content into a plurality of long text segments according to the manual semantic information, input each long text segment into a target document large model for fragmentation processing, obtain a processing result, and obtain maintenance manual content according to the processing result.

[0052] Optionally, the answer recall module is further configured to convert the historical fault data into a question and answer pair form to obtain question and answer pair data, generate a historical fault file according to the question and answer pair data, and synthesize a historical fault knowledge base according to the historical fault file.

[0053] Optionally, the answer recall module is further configured to obtain maintenance key words according to the maintenance manual content, generate a key word table according to the maintenance key words, and import the key word table and the historical fault knowledge base into an initial knowledge graph to obtain a fault maintenance knowledge graph.

[0054] Optionally, the intention recognition module is further configured to, when receiving a query question about a fault cause of a target device, input the query question into a target language large model to obtain a user intention; determine whether to call a vocabulary resource library according to the user intention, and obtain an intention recognition result according to the determination result.

[0055] In addition, to achieve the above object, the application further provides a fault maintenance device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the fault maintenance method.

[0056] In addition, to achieve the above object, the application further provides a storage medium, which is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the fault maintenance method.

[0057] In addition, to achieve the above object, the application further provides a computer program product, which comprises a computer program, and the computer program is executed by a processor to implement the steps of the fault maintenance method.

[0058] The one or more technical solutions provided by the application have at least the following technical effects:

[0059] When receiving a query question about a fault cause of a target device, the scheme of the application performs intention recognition on the query question through a target language large model to obtain an intention recognition result; calls a vocabulary resource library according to the intention recognition result, and queries a fault maintenance knowledge graph generated based on a target document large model according to the target vocabulary obtained by calling, to obtain a plurality of answer cases; performs network resource retrieval according to the answer cases to obtain a network retrieval result of fault classification; determines a fault classification according to the answer cases and the network retrieval result, and generates a fault solution recommendation scheme according to the fault classification. In this way, intention recognition is performed according to the user query question, and then the answer is retrieved based on the pre-established knowledge graph, and at the same time, network retrieval is combined to perform fault classification and finally generate a solution scheme, which provides customized and accurate fault elimination suggestions, and at the same time, relies on the knowledge graph and the large model to improve the efficiency and accuracy of fault maintenance. BRIEF DESCRIPTION OF DRAWINGS

[0060] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application.

[0061] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0062] Figure 1 A flowchart of the first embodiment of the fault repair method of this application is provided;

[0063] Figure 2 This is a schematic diagram of model evaluation optimization in an embodiment of the fault repair method of this application;

[0064] Figure 3 A flowchart of the second embodiment of the fault repair method of this application is provided;

[0065] Figure 4 This is a brief flowchart of an embodiment of the fault repair method of this application;

[0066] Figure 5 This is a schematic diagram of the module structure of the fault repair device according to an embodiment of the present application;

[0067] Figure 6 Schematic diagram of the equipment structure of the hardware operating environment involved in the fault repair method in the embodiment of the present application.

[0068] The purpose, features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0069] It should be understood that the specific embodiments described herein are merely used to explain the technical solutions of the present application and are not intended to limit the present application.

[0070] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0071] The main solution of the embodiment of the present application is: when a query question about the cause of a target device failure is received, the query question is subjected to intent recognition through the target language large model to obtain an intent recognition result; the vocabulary resource library is called according to the intent recognition result, and the fault repair knowledge graph generated based on the target document large model is queried according to the target vocabulary obtained by the call to obtain multiple recall answers; network resource retrieval is performed according to the recall answers to obtain network retrieval results of fault classification; the fault classification is determined according to the recall answers and the network retrieval results, and a fault solution recommendation is generated according to the fault classification.

[0072] In this embodiment, for ease of description, the following description is made with the identification computer as the execution subject.

[0073] Since the fault maintenance of various high-end mechanical equipment, vehicles and technological products in the prior art is a highly professional and complex task, it not only requires maintenance personnel to have solid theoretical knowledge and practical experience, but also requires them to quickly and accurately locate the fault cause and take effective measures. The traditional maintenance mode relies on manual checking of a large number of technical materials and past cases, which is time-consuming and prone to errors.

[0074] The present application provides a solution in this way, which realizes intent recognition according to the user's query question, then searches for the answer based on the pre-established knowledge graph, and combines network search to classify faults and generate the final solution, providing customized and accurate troubleshooting suggestions, while relying on the knowledge graph and large model to improve the efficiency and accuracy of fault maintenance.

[0075] It should be noted that the execution subject of the present embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device, a server, etc. that can realize the above functions. The following will take a fault maintenance device or a computer installed with corresponding programs as an example to describe the present embodiment and the following embodiments.

[0076] Based on this, the present embodiment provides a fault maintenance method, referring to Figure 1 , Figure 1 The flowchart of the first embodiment of the fault maintenance method of the present application.

[0077] In the present embodiment, the fault maintenance method comprises steps S10-S40:

[0078] Step S10, when receiving a query question about the fault cause of a target device, performing intent recognition on the query question through a target language large model to obtain an intent recognition result;

[0079] It should be noted that the query question input by the user is first received through the UI interface, then the user's intent is analyzed based on the query question, and then it is determined whether to call the vocabulary resource library based on the result of the intent analysis.

[0080] In a feasible implementation, in order to accurately analyze the user's intent, step S10 comprises: when receiving a query question about the fault cause of a target device, importing the query question into a target language large model to obtain a user intent; determining whether to call a vocabulary resource library according to the user intent, and obtaining an intent recognition result according to the determination result.

[0081] It should be understood that the system performs an intent recognition process for the question raised by the user, which is driven by a large language model, aiming to evaluate whether it is necessary to call a specialized vocabulary repository, which ensures a deep understanding of the user's query intent and lays the foundation for subsequent processing. Among them, the large language model is a target language large model, which is any type of pre-trained large model.

[0082] In a feasible implementation, in order to accurately obtain the user's intent, when receiving the query question of the fault reason of the target device, the step of introducing the query question into the target language large model to obtain the user's intent includes: when receiving the query question of the fault reason of the target device, determining a query prompt word according to the query question; extracting subject name information in the query prompt word; introducing the subject name information and the query question into the target language large model to obtain the user's intent.

[0083] In specific implementation, the user's question is extracted by the prompt word+large model, and the subject name such as fault code and part name in the extracted question is extracted. Subsequently, the knowledge graph retrieval and full-text retrieval of the data source are performed based on the extracted keywords.

[0084] Step S20, calling a vocabulary repository according to the intent recognition result, and querying the fault maintenance knowledge graph generated based on the target document large model according to the target vocabulary obtained by calling to obtain a plurality of call answer cases;

[0085] It should be noted that after obtaining the intent recognition result, the vocabulary repository is called according to the intent recognition result, and the vocabulary repository is also a pre-set and stored repository, which stores all the tendency words corresponding to the user's intent, such as maintenance, repair, inspection, etc.

[0086] It should be understood that after obtaining the target vocabulary, the target vocabulary is introduced into the fault maintenance knowledge graph, so that a plurality of call answer cases can be obtained.

[0087] In specific implementation, the vector similarity does not have the transitivity feature. For example, in the vector space, if historical question A is similar to historical question B, and historical question B is similar to historical question C, it does not mean that historical question A is similar to historical question C. In contrast, the knowledge graph stores a large number of explicit facts in the form of accurate, updatable and interpretable knowledge triples, i.e. (entity) -> [relation] -> (entity). In order to ensure that the relevant historical questions can be correctly retrieved, this paper proposes to increase the exploration of similar historical questions. In addition, the association of synonyms and near-synonyms of proper nouns is also realized to improve the accuracy and comprehensiveness of retrieval.

[0088] In an implementable embodiment, step S20 comprises: calling a vocabulary resource library according to the intention recognition result to obtain a plurality of target vocabularies corresponding to the query question; querying the fault maintenance knowledge graph according to the target vocabularies to obtain a plurality of vocabulary-related questions; and obtaining a plurality of recall answers according to the vocabulary-related questions.

[0089] It should be noted that, first, the vocabulary resource library is called according to the intention recognition result, a plurality of target vocabularies corresponding to the semantic of the query question can be queried, and then the target vocabularies are brought into the fault maintenance knowledge graph for querying. By using the constructed fault maintenance knowledge graph, the system can generate a plurality of related questions for information recall. This strategy not only enriches the query dimension, but also promotes the joint learning of vector models and large language models on enterprise-specific terminology, significantly improving the accuracy and relevance of recall.

[0090] Step S30, network resource retrieval is performed according to the recall answer to obtain a network retrieval result of fault classification;

[0091] It should be understood that after obtaining the recall answer, another dimension of query and retrieval based on the recall answer is needed, so as to enrich and perfect the query result.

[0092] In a specific implementation, network resource retrieval refers to querying and retrieving information in a pre-set network resource library based on the recall answer, thereby obtaining a network retrieval result.

[0093] In an implementable embodiment, in order to accurately perform network retrieval, step S30 comprises: screening the recall answer by a target agent node to obtain a screening result; determining a fault elimination chapter number corresponding to a pre-set maintenance manual according to the screening result; and retrieving a target network resource according to the fault elimination chapter number to obtain a network retrieval result.

[0094] It should be noted that after the recall stage is completed, an intelligent agent node is introduced for advanced screening. The agent is responsible for analyzing the information fragments recalled, and identifying whether the information fragments contain a number pointing to a fault elimination chapter. Once it is confirmed that there is a relevant chapter number, the system immediately activates a special tool to retrieve the corresponding network resource URL. This step ensures that the obtained information is directly related to specific fault resolution guidelines, enhancing the practicality.

[0095] It should be understood that since a single vector search is difficult to comprehensively retrieve complex content, some historical faults may not be recalled. Therefore, a plurality of retrieval methods, including vector retrieval, graph exploration, full-text retrieval, and question and answer pair retrieval, will be used to ensure comprehensive recall of all relevant information. By comprehensively using these methods, the coverage and accuracy of the retrieval system can be improved, ensuring that all relevant historical faults can be effectively identified and processed.

[0096] In specific implementation, when hybrid retrieval is used, a large number of historical fault records may be recalled, thereby providing too much irrelevant content for the large model, causing interference and increasing the consumption of computing power. In order to solve this problem, the bge-reranker reordering model is introduced in this paper to reorder the recalled historical faults. This model can effectively filter out the few historical faults with the highest similarity, thereby reducing knowledge interference, optimizing model performance and reducing computing power consumption.

[0097] Step S40, determining a fault classification according to the recall result and the network retrieval result, and generating a fault solution recommendation scheme according to the fault classification.

[0098] It should be understood that in combination with the original question and the answer extracted from the knowledge base, the model further judges the specific classification to which the fault belongs. This comprehensive analysis process not only improves the speed of fault identification, but also improves the accuracy of classification, providing more accurate fault solution recommendations for users. Through the integration of these strategies, the intelligent level and user experience of the fault diagnosis system are effectively improved.

[0099] It should be noted that the scheme of the embodiment has a user-friendly interface design, allowing users to accurately locate the required machine model and detailed DE (defect code) number through intuitive selection boxes. This mechanism ensures that the system can provide customized and accurate fault elimination suggestions for each specific machine model, avoiding generalization and hitting the core of the problem.

[0100] It should be understood that the system can also take full advantage of big data to dynamically aggregate and analyze user consultation logs, instantly extract frequently occurring hot issues, and display them in an intuitive manner. Users can easily obtain verified answers by clicking on these hot issues, significantly shortening the problem solving period and achieving a significant leap in efficiency.

[0101] In specific implementation, the system can also record the consultation history of each user, which not only facilitates subsequent personal inquiry and tracking, but also opens up new ways for team collaboration and knowledge inheritance. Users can conveniently share their consultation records with other maintenance engineers or management personnel, promoting efficient information flow and experience sharing between teams, and jointly promoting the continuous improvement of maintenance efficiency. In summary, the auxiliary system based on the large model brings unprecedented efficiency improvement to the maintenance field through intelligent, personalized and collaborative optimization.

[0102] It should be noted that, as Figure 2As shown, in order to optimize the model and the whole process, after step S40, the data set can also be tested for four rounds by using the RAGAS evaluation framework to comprehensively score from multiple dimensions, including Context Relevance, Context Recall, Faithfulness and Answer Relevance. The final result shows that the overall accuracy has been significantly improved from the original 60% to about 90%.

[0103] The embodiment provides a fault maintenance method. When a query question about a fault cause of a target device is received, an intention of the query question is recognized through a target language large model to obtain an intention recognition result. A vocabulary resource library is called according to the intention recognition result, and a fault maintenance knowledge graph generated based on a target document large model is queried according to a target vocabulary obtained by calling to obtain a plurality of answer cases. Network resource retrieval is performed according to the answer cases to obtain a network retrieval result of fault classification. The fault classification is determined according to the answer cases and the network retrieval result, and a fault solution recommendation scheme is generated according to the fault classification. In this way, the intention of the user query is recognized, and then the answer is retrieved based on the pre-established knowledge graph, and the network retrieval is combined to perform fault classification and finally generate a solution scheme, to provide customized and accurate fault elimination suggestions, and to improve the efficiency and accuracy of fault maintenance by relying on the knowledge graph and the large model.

[0104] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as the above-mentioned first embodiment can be referred to the above introduction, and the subsequent will not be repeated. On this basis, please refer to Figure 3 , before step S20, the fault maintenance method further includes steps S201-S203:

[0105] Step S201, dividing the content of the preset maintenance manual into a plurality of long text segments, and inputting the long text segments into a target document large model for fragmentation processing to obtain the maintenance manual content according to the processing result.

[0106] It should be noted that first, the text content of the preset maintenance manual is divided into a plurality of long text segments, then the long text segments are imported into the target document large model, the fragmentation processing is performed through the target document large model, and then the summary and information extraction are performed to obtain the maintenance manual content.

[0107] In an implementable embodiment, in order to extract long text segments, step S201 comprises: obtaining a text file of a preset maintenance manual, and performing content extraction on the text file to obtain target manual content, wherein the content extraction comprises paragraph, picture, table and page number extraction; performing label extraction and summary extraction on the target manual content to obtain analysis manual content; and obtaining a plurality of long text segments from the analysis manual content.

[0108] It should be understood that the system deeply analyzes the provided Boeing manual content through layout recognition technology. First, the content such as paragraphs, pictures, tables and page numbers in the PDF file is extracted in a semantic understanding manner.

[0109] In a specific implementation, after content extraction, label extraction and summary extraction are performed on the paragraph content to obtain analysis manual content, thereby improving the readability and retrieval efficiency of the document content. Finally, text division is performed according to the analysis manual content, and a plurality of long text segments can be obtained.

[0110] In an implementable embodiment, in order to perform fragmentation processing, the step of dividing the content of the preset maintenance manual into a plurality of long text segments comprises: determining manual semantic information according to the analysis manual content; and dividing the analysis manual content into a plurality of long text segments according to the manual semantic information.

[0111] It should be noted that dividing the analysis manual content into a plurality of long text segments ensures the semantic completeness of each segment. Then, the segments are input into a large model for intelligent fragmentation processing, thereby ensuring data integrity and consistency. Finally, according to the processing result, a document is formed to realize accurate information extraction and efficient document management, thereby improving the use efficiency of the maintenance manual. The target document large model herein is a 32K document large model, or other large models that can realize this function, and the present embodiment is not limited thereto.

[0112] Step S202: generating a historical fault knowledge base according to historical fault data;

[0113] It should be understood that the historical fault data is first extracted, which includes the historical fault, maintenance information record and data of the current equipment, and can also include the user's query record.

[0114] In an implementable embodiment, in order to generate a more accurate and convenient-to-query knowledge base, step S202 comprises: converting the historical fault data into a question and answer pair form to obtain question and answer pair data; generating a historical fault file according to the question and answer pair data; and synthesizing a historical fault knowledge base according to the historical fault file.

[0115] In a specific implementation, the system can automatically process historical fault measure data through built-in tools. First, data is arranged: historical fault data is uniformly arranged and converted into question and answer pairs, specifically, historical fault data is converted into pairs of questions and answer combinations, and each question and answer combination is a question and answer pair. Then, files are generated: the arranged data generates new historical fault files for subsequent management and query. Finally, the knowledge base is entered: the generated historical fault files are entered into the historical fault knowledge base as a data source for historical fault measures, providing a reference for future fault diagnosis.

[0116] Step S203, generating a fault maintenance knowledge graph according to the historical fault knowledge base and the maintenance manual content.

[0117] It should be noted that after obtaining the historical fault knowledge base and the maintenance manual content, the fault maintenance knowledge graph is constructed, specifically, first, the keywords are extracted, then the initial knowledge graph is imported, and the fault maintenance knowledge graph for the target device is formed.

[0118] In a feasible implementation, in order to accurately construct the knowledge graph, step S203 includes: obtaining maintenance keywords according to the maintenance manual content; generating a keyword table according to the maintenance keywords; importing the keyword table and the historical fault knowledge base into the initial knowledge graph to obtain the fault maintenance knowledge graph.

[0119] It should be understood that the maintenance keywords are entered into the knowledge graph through the table import method, which enhances the function of the query tool. First, the keywords are entered, a plurality of maintenance keywords can be obtained according to the maintenance manual content, then the maintenance keyword table is imported into the knowledge graph, which is convenient for the system to call when querying. Then, the historical fault data is entered, the historical fault problems are entered into the knowledge graph, and the association between the historical problems is established.

[0120] In a specific implementation, the fault maintenance knowledge graph also includes an intelligent query tool: the knowledge graph serves as a backup for the intelligent query tool, and supports the intelligent agent to call relevant data in the process of fault diagnosis and problem solving.

[0121] It should be noted that the maintenance of the aircraft is taken as an example, as shown in Figure 4 The flowchart shows that answer retrieval can be achieved with the help of the knowledge graph and the historical fault data knowledge base.

[0122] The embodiment divides the content of the preset maintenance manual into a plurality of long text segments, inputs the long text segments into a target document large model for fragmentation processing, obtains the maintenance manual content according to a processing result, generates a historical fault knowledge base according to historical fault data, and generates a fault maintenance knowledge graph according to the historical fault knowledge base and the maintenance manual content. In this way, the construction of the knowledge graph is realized in combination with the historical data and the maintenance manual, so that the subsequent problem query can quickly query the closest and accurate answer.

[0123] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the fault maintenance method of the present application. More forms of simple transformation based on this technical concept are within the protection scope of the present application.

[0124] The present application also provides a fault maintenance device, please refer to Figure 5 , the fault maintenance device comprises:

[0125] The intent recognition module 10 is configured to perform intent recognition on the query question of the fault cause of the target device through a target language large model when the query question is received, and obtain an intent recognition result.

[0126] The answer recall module 20 is configured to call a vocabulary resource library according to the intent recognition result, query the fault maintenance knowledge graph generated based on the target document large model according to the target vocabulary obtained by calling, and obtain a plurality of recall answers.

[0127] The network retrieval module 30 is configured to perform network resource retrieval according to the recall answers, and obtain a network retrieval result of the fault classification.

[0128] The scheme generation module 40 is configured to determine the fault classification according to the recall answers and the network retrieval result, and generate a fault solution recommendation scheme according to the fault classification.

[0129] In an embodiment, the answer recall module 20 is further configured to divide the content of the preset maintenance manual into a plurality of long text segments, input the long text segments into a target document large model for fragmentation processing, obtain the maintenance manual content according to a processing result, generate a historical fault knowledge base according to historical fault data, and generate a fault maintenance knowledge graph according to the historical fault knowledge base and the maintenance manual content.

[0130] In an embodiment, the answer recall module 20 is further configured to: obtain a text file of the preset maintenance manual, and perform content extraction on the text file to obtain target manual content, the content extraction including paragraph extraction, picture extraction, table extraction, and page number extraction; perform tag extraction and summary extraction on the target manual content to obtain analysis manual content; and divide the analysis manual content into a plurality of long text segments according to the analysis manual content.

[0131] In an embodiment, the answer recall module 20 is further configured to: determine manual semantic information according to the analysis manual content; and divide the analysis manual content into a plurality of long text segments according to the manual semantic information.

[0132] In an embodiment, the answer recall module 20 is further configured to: convert the historical fault data into a question and answer pair form to obtain question and answer pair data; generate a historical fault file according to the question and answer pair data; and synthesize a historical fault knowledge base according to the historical fault file.

[0133] In an embodiment, the answer recall module 20 is further configured to: obtain a maintenance keyword according to the maintenance manual content; generate a keyword table according to the maintenance keyword; and import the keyword table and the historical fault knowledge base into an initial knowledge graph to obtain a fault maintenance knowledge graph.

[0134] In an embodiment, the intent recognition module 10 is further configured to: when receiving a query question about a fault cause of a target device, import the query question into a target language large model to obtain a user intent; determine whether to call a vocabulary resource library according to the user intent, and obtain an intent recognition result according to a determination result.

[0135] In an embodiment, the intent recognition module 10 is further configured to: when receiving a query question about a fault cause of a target device, determine a query prompt word according to the query question; extract subject name information in the query prompt word; and import the subject name information and the query question into a target language large model to obtain a user intent.

[0136] In an embodiment, the answer recall module 20 is further configured to: call a vocabulary resource library according to the intent recognition result to obtain a plurality of target vocabularies corresponding to the query question; query the fault maintenance knowledge graph according to the target vocabularies to obtain a plurality of vocabulary-related questions; and obtain a plurality of recall answers according to the vocabulary-related questions.

[0137] In an embodiment, the network retrieval module 30 is further configured to filter the query question through the target agent node to obtain a filtering result; determine a troubleshooting chapter number corresponding to a preset maintenance manual according to the filtering result; and retrieve a target network resource according to the troubleshooting chapter number to obtain a network retrieval result.

[0138] When the query question about the cause of the failure of the target device is received, the embodiment performs intent recognition on the query question through the target language large model to obtain an intent recognition result; calls a vocabulary resource library according to the intent recognition result, and queries a fault maintenance knowledge graph generated based on a target document large model according to the target vocabulary obtained by the calling to obtain a plurality of query answers; performs network resource retrieval according to the query answers to obtain a network retrieval result of fault classification; determines fault classification according to the query answers and the network retrieval result, and generates a fault solution recommendation scheme according to the fault classification. In this way, the intent recognition is performed according to the query question of the user, and then the answer is retrieved based on the pre-established knowledge graph, and the fault classification and the generation of the final solution are combined with the network retrieval to provide customized and accurate fault elimination suggestions, and the efficiency and accuracy of the fault maintenance are improved by relying on the knowledge graph and the large model.

[0139] The fault maintenance device provided in the present application adopts the fault maintenance method in the above embodiments, and can solve the technical problems of long time-consuming and easy to make mistakes in device fault troubleshooting. Compared with the prior art, the fault maintenance device provided in the present application has the same beneficial effects as the fault maintenance method provided in the above embodiments, and other technical features in the fault maintenance device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0140] The present application provides a fault maintenance device, which comprises at least one processor and a memory in communication connection with the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the fault maintenance method in the above embodiment one.

[0141] The following will be described with reference to the drawings Figure 6, which shows a schematic diagram of the structure of a fault repair device suitable for implementing the embodiments of the present application. The fault repair device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 6 The fault repair device shown is only an example and should not limit the functions and scope of use of the embodiments of the present application.

[0142] like Figure 6 As shown, the fault repair device may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 1002 or programs loaded from a storage device 1003 into a random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the fault repair device. Processing device 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems may be connected to I / O interface 1006: input devices 1007 including, for example, a touchscreen, touchpad, keyboard, mouse, image sensor, microphone, accelerometer, gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage device 1003 including, for example, a magnetic tape, hard disk, etc.; and communication device 1009. The communication device 1009 can allow the fault repair device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a fault repair device with various systems, it should be understood that it is not required to implement or have all of the systems shown. More or fewer systems can be implemented or have alternatively.

[0143] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program codes for executing the method shown in the flowchart. In such embodiments, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiments disclosed in the present application are executed.

[0144] The fault maintenance device provided by the present application adopts the fault maintenance method in the above-mentioned embodiments, and can solve the technical problems of long time-consuming and easy to make mistakes in troubleshooting of equipment faults. Compared with the prior art, the beneficial effects of the fault maintenance device provided by the present application are the same as those of the fault maintenance method provided by the above-mentioned embodiments, and other technical features in the fault maintenance device are the same as those disclosed in the previous embodiment method, which will not be repeated here.

[0145] It should be understood that parts of the present application can be realized by hardware, software, firmware or a combination thereof. In the description of the above-mentioned embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0146] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed in the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0147] The present application provides a computer readable storage medium having stored thereon computer readable program instructions (i.e. computer program) for executing the fault maintenance method in the above-mentioned embodiments.

[0148] The computer readable storage medium provided in the application may be, for example, a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electric connection with one or more conductive wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted by any suitable medium, including but not limited to an electric wire, an optical cable, an RF (Radio Frequency), and the like, or any suitable combination of the above.

[0149] The above computer readable storage medium can be included in the fault maintenance device, or can exist separately without being assembled into the fault maintenance device.

[0150] The above computer readable storage medium carries one or more programs, which, when executed by the fault maintenance device, cause the fault maintenance device to: when receiving a query question about the cause of the fault of the target device, perform intent recognition on the query question by a target language large model to obtain an intent recognition result; call a vocabulary resource library according to the intent recognition result, and query a fault maintenance knowledge graph generated based on a target document large model according to the target vocabulary obtained by calling, to obtain a plurality of answer cases; perform network resource retrieval according to the answer cases to obtain a network retrieval result of fault classification; determine a fault classification according to the answer cases and the network retrieval result, and generate a fault solution recommendation scheme according to the fault classification.

[0151] Computer program code for carrying out operations of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0152] The flow diagrams and the block diagrams in the drawings are illustrations of architectures, functionalities, and operations of possible implementations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may

[0153] The modules involved in the embodiments of the present application can be implemented in the form of software or in the form of hardware. In some cases, the name of the module does not constitute a limitation on the module itself.

[0154] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e., a computer program) for executing the above-mentioned fault maintenance method, and can solve the technical problems that the device fault troubleshooting is time-consuming and prone to errors. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the fault maintenance method provided by the above-mentioned embodiments, which will not be described here.

[0155] The application further provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of the fault maintenance method as described above.

[0156] The computer program product provided by the application can solve the technical problem that device troubleshooting is time-consuming and prone to errors. Compared with the prior art, the beneficial effects of the computer program product provided by the application are the same as those of the fault maintenance method provided by the above-mentioned embodiments, and are not described here.

[0157] The above only describes some embodiments of the application, and does not limit the patent scope of the application. Any equivalent structural transformation based on the technical concept of the application, or direct / indirect application in other related technical fields is included in the patent protection scope of the application.

[0158] The application discloses A1. A fault maintenance method, the method comprises:

[0159] Upon receiving a query question about the cause of the fault of the target device, performing intent recognition on the query question by a target language large model to obtain an intent recognition result;

[0160] According to the intent recognition result, calling a vocabulary resource library, and according to the target vocabulary obtained by calling, querying a fault maintenance knowledge graph generated based on a target document large model to obtain a plurality of answer cases;

[0161] According to the answer cases, performing network resource retrieval to obtain a network retrieval result of fault classification;

[0162] According to the answer cases and the network retrieval result, determining a fault classification, and generating a fault solution recommendation scheme according to the fault classification.

[0163] A2. The method of A1, before the step of calling a vocabulary resource library according to the intent recognition result, and querying a fault maintenance knowledge graph generated based on a target document large model according to the target vocabulary obtained by calling to obtain a plurality of answer cases, further comprising:

[0164] Dividing the content of a preset maintenance manual into a plurality of long text segments, and inputting the long text segments into a target document large model for fragmentation processing, and obtaining maintenance manual content according to the processing result;

[0165] Generating a historical fault knowledge base according to historical fault data;

[0166] Generating a fault maintenance knowledge graph according to the historical fault knowledge base and the maintenance manual content.

[0167] A3. The method of A2, wherein the step of dividing the content of the preset maintenance manual into a plurality of long text segments comprises:

[0168] obtaining a text file of a preset maintenance manual, and performing content extraction on the text file to obtain target manual content, wherein the content extraction comprises paragraph extraction, picture extraction, table extraction, and page number extraction;

[0169] performing tag extraction and summary extraction on the target manual content to obtain analysis manual content;

[0170] dividing the analysis manual content into a plurality of long text segments.

[0171] A4. The method of A3, wherein the step of dividing the analysis manual content into a plurality of long text segments comprises:

[0172] determining manual semantic information from the analysis manual content;

[0173] dividing the analysis manual content into a plurality of long text segments according to the manual semantic information.

[0174] A5. The method of A2, wherein the step of generating a historical fault knowledge base from historical fault data comprises:

[0175] converting the historical fault data into a question and answer pair form to obtain question and answer pair data;

[0176] generating a historical fault file from the question and answer pair data;

[0177] synthesizing a historical fault knowledge base from the historical fault file.

[0178] A6. The method of A2, wherein the step of generating a fault maintenance knowledge graph from the historical fault knowledge base and the maintenance manual content comprises:

[0179] obtaining maintenance keywords from the maintenance manual content;

[0180] generating a keyword table from the maintenance keywords;

[0181] importing the keyword table and the historical fault knowledge base into an initial knowledge graph to obtain a fault maintenance knowledge graph.

[0182] A7. The method of A1, wherein the step of, upon receiving a query question about the cause of a fault of a target device, performing intent recognition on the query question by a target language large model to obtain an intent recognition result comprises:

[0183] upon receiving a query question about the cause of a fault of a target device, importing the query question into a target language large model to obtain a user intent;

[0184] determining whether to call a vocabulary resource library according to the user intention, and obtaining an intention recognition result according to a result of the determination.

[0185] A8. The method of A7, wherein the step of, upon receiving a query question about a cause of a fault of a target device, inputting the query question into a target language large model to obtain a user intention, comprises:

[0186] upon receiving a query question about a cause of a fault of a target device, determining a query prompt word according to the query question;

[0187] extracting subject name information in the query prompt word;

[0188] inputting the subject name information and the query question into a target language large model to obtain a user intention.

[0189] A9. The method of A1, wherein the step of, according to the intention recognition result, calling a vocabulary resource library, and querying a fault maintenance knowledge graph generated based on a target document large model according to a target vocabulary obtained by the calling, to obtain a plurality of call answering cases, comprises:

[0190] according to the intention recognition result, calling a vocabulary resource library to obtain a plurality of target vocabularies corresponding to the query question;

[0191] querying the fault maintenance knowledge graph according to the target vocabulary to obtain a plurality of vocabulary related questions;

[0192] obtaining a plurality of call answering cases according to the vocabulary related questions.

[0193] A10. The method of A1, wherein the step of, according to the call answering cases, performing network resource retrieval to obtain a network retrieval result of a fault classification, comprises:

[0194] performing screening on the call answering cases through a target agent node to obtain a screening result;

[0195] determining a fault elimination chapter number corresponding to a preset maintenance manual according to the screening result;

[0196] retrieving a target network resource according to the fault elimination chapter number to obtain a network retrieval result.

[0197] The application also discloses B11. A fault maintenance device, comprising:

[0198] an intention recognition module, configured to, upon receiving a query question about a cause of a fault of a target device, perform intention recognition on the query question through a target language large model to obtain an intention recognition result;

[0199] an answer recall module configured to invoke a vocabulary resource library according to the intention recognition result, query a fault maintenance knowledge graph generated based on a target document large model according to a target vocabulary obtained by the invocation, and obtain a plurality of answer recall results;

[0200] a network retrieval module configured to perform network resource retrieval according to the answer recall results, and obtain network retrieval results of fault classification;

[0201] a scheme generation module configured to determine fault classification according to the answer recall results and the network retrieval results, and generate a fault solution recommendation scheme according to the fault classification.

[0202] B12. The apparatus of B11, wherein the answer recall module is further configured to divide content of a preset maintenance manual into a plurality of long text segments, input the long text segments into a target document large model for fragmentation processing, obtain maintenance manual content according to a processing result, generate a historical fault knowledge base according to historical fault data, and generate a fault maintenance knowledge graph according to the historical fault knowledge base and the maintenance manual content.

[0203] B13. The apparatus of B12, wherein the answer recall module is further configured to obtain a text file of a preset maintenance manual, perform content extraction on the text file to obtain target manual content, the content extraction including paragraph, picture, table, and page number extraction, perform label extraction and summary extraction on the target manual content to obtain analyzed manual content, obtain a plurality of long text segments according to the analyzed manual content, input the long text segments into a target document large model for fragmentation processing, and obtain maintenance manual content according to a processing result.

[0204] B14. The apparatus of B13, wherein the answer recall module is further configured to determine manual semantic information according to the analyzed manual content, divide the analyzed manual content into a plurality of long text segments according to the manual semantic information, input each long text segment into a target document large model for fragmentation processing to obtain a processing result, and obtain maintenance manual content according to the processing result.

[0205] B15. The apparatus of B12, wherein the answer recall module is further configured to convert the historical fault data into a question and answer pair form to obtain question and answer pair data, generate a historical fault file according to the question and answer pair data, and synthesize a historical fault knowledge base according to the historical fault file.

[0206] B16. The apparatus of B12, wherein the answer recall module is further configured to obtain maintenance key words according to the maintenance manual content, generate a key word table according to the maintenance key words, and import the key word table and the historical fault knowledge base into an initial knowledge graph to obtain a fault maintenance knowledge graph.

[0207] B17. The apparatus of B11, wherein the intent recognition module is further configured to, when a query question about a cause of a failure of the target device is received, input the query question into a target language large model to obtain a user intent, determine whether to call a vocabulary resource library according to the user intent, and obtain an intent recognition result according to a result of the determination.

[0208] The application further discloses C18. A fault maintenance device, the device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, the computer program being configured to implement the steps of the fault maintenance method.

[0209] The application further discloses D19. A storage medium, the storage medium being a computer-readable storage medium, and the storage medium storing a computer program, the computer program being executable by a processor to implement the steps of the fault maintenance method.

[0210] The application further discloses E20. A computer program product, the computer program product comprising a computer program, the computer program being executable by a processor to implement the steps of the fault maintenance method.

Claims

1. A method of failure repair, characterized by, The method comprises: Upon receiving a query question about the cause of failure of a target device, performing intent recognition on the query question by a target language large model to obtain an intent recognition result; According to the intent recognition result, calling a vocabulary resource library, and querying a failure maintenance knowledge graph generated based on a target document large model according to the target vocabulary obtained by calling to obtain a plurality of answer cases; According to the answer cases, performing network resource retrieval to obtain a network retrieval result of failure classification; According to the answer cases and the network retrieval result, determining a failure classification, and generating a failure solution recommendation scheme according to the failure classification; The step of performing network resource retrieval according to the answer cases to obtain a network retrieval result of failure classification comprises: Filtering the answer cases by a target proxy node to obtain a filtering result; According to the filtering result, determining a fault elimination chapter number corresponding to a preset maintenance manual; According to the fault elimination chapter number, retrieving target network resources to obtain a network retrieval result.

2. The method of claim 1, wherein, Before the step of calling a vocabulary resource library according to the intent recognition result, and querying a failure maintenance knowledge graph generated based on a target document large model according to the target vocabulary obtained by calling to obtain a plurality of answer cases, the method further comprises: Dividing the content of a preset maintenance manual into a plurality of long text segments, and inputting the long text segments into a target document large model for fragmentation processing to obtain maintenance manual content according to the processing result; Generating a historical failure knowledge base according to historical failure data; Generating a failure maintenance knowledge graph according to the historical failure knowledge base and the maintenance manual content.

3. The method of claim 2, wherein, The step of dividing the content of a preset maintenance manual into a plurality of long text segments comprises: Obtaining a text file of a preset maintenance manual, and extracting content from the text file to obtain target manual content, wherein the content extraction includes paragraph, picture, table, and page number extraction; According to the target manual content, performing label extraction and abstract extraction to obtain analyzed manual content; According to the analyzed manual content, obtaining a plurality of long text segments.

4. The method of claim 3, wherein, The step of obtaining a plurality of long text segments according to the analyzed manual content comprises: According to the analyzed manual content, determining manual semantic information; According to the manual semantic information, dividing the analyzed manual content into a plurality of long text segments.

5. The method of claim 2, wherein, The step of generating a historical failure knowledge base according to historical failure data comprises: Converting the historical failure data into a question and answer pair form to obtain question and answer pair data; Generating a historical failure file according to the question and answer pair data; Generating a historical failure knowledge base according to the historical failure file.

6. The method of claim 2, wherein, The step of generating a failure maintenance knowledge graph according to the historical failure knowledge base and the maintenance manual content comprises: According to the maintenance manual content, obtaining maintenance keywords; According to the maintenance keywords, generating a keyword table; Importing the keyword table and the historical failure knowledge base into an initial knowledge graph to obtain a failure maintenance knowledge graph.

7. A fault repair apparatus, characterized by, The device comprises: An intent recognition module configured to, upon receiving a query question about the cause of failure of a target device, perform intent recognition on the query question by a target language large model to obtain an intent recognition result; The answer recall module is configured to invoke a vocabulary resource library according to the intention recognition result, query a fault maintenance knowledge graph generated based on a target document large model according to a target vocabulary obtained by the invocation, and obtain a plurality of recall answers; The network retrieval module is configured to perform network resource retrieval according to the recall answers, and obtain a network retrieval result of fault classification; The network retrieval module is further configured to filter the recall answers through a target proxy node to obtain a filtering result, determine a fault elimination chapter number corresponding to a preset maintenance manual according to the filtering result, and retrieve target network resources according to the fault elimination chapter number to obtain the network retrieval result; The scheme generation module is configured to determine fault classification according to the recall answers and the network retrieval result, and generate a fault solution recommendation scheme according to the fault classification.

8. A fault repair apparatus characterized by comprising: The device comprises a memory, a processor, and a computer program stored on the memory and executable on the processor, and the computer program is configured to implement the steps of the fault maintenance method according to any one of claims 1 to 6.

9. A storage medium, characterized by The storage medium is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the fault maintenance method according to any one of claims 1 to 6.

10. A computer program product, characterised in that, The computer program product comprises a computer program, and the computer program is executed by the processor to implement the steps of the fault maintenance method according to any one of claims 1 to 6.

Images