An online fault diagnosis method based on interactive electronic manual

Abstract

The present application discloses an online fault diagnosis method based on an interactive electronic manual, which relates to the field of artificial intelligence technology, including: S1, determining the associated equipment corresponding to the fault information based on an uploaded standard fault information template; S2, decomposing the label fragments of the fault information according to the fault information template to obtain multiple levels of fault information, and outputting them as the first target area, wherein the first target area includes obtaining hierarchical paragraphs, paragraphs, illustrations, multimedia, tables, warnings, notes, and annotated data information fragments; S3, editing the obtained first target area to determine the connection relationship between different first target areas; the fault positioning effect can be improved, and the fault that occurs can be accurately identified.

Description

Technical Field

[0001] The invention relates to the field of artificial intelligence technology, and in particular to an online fault diagnosis method based on an interactive electronic manual. Background Art

[0002] IETM (Interactive Electronic Technical Manual), Chinese name Interactive Electronic Technical Manual. Simply put, the interactive electronic technical manual is to convert the content of the technical manual into digital form, recompile it, and consult it in an interactive way, and display the consulted content to maintenance technicians or system operators through computers and other devices.

[0003] For example, Chinese patent publication number CN112965460A discloses a fault diagnosis method, device and storage medium based on inquiry-based fault reporting and Bayesian theorem, the method comprising: constructing a fault diagnosis knowledge base, wherein the fault diagnosis knowledge base comprises a mapping relationship between fault phenomena and fault causes; obtaining the reported fault phenomenon based on the inquiry-based fault reporting process, automatically prompting other fault phenomena of the fault cause associated with the fault phenomenon according to the mapping relationship in the fault diagnosis knowledge base, obtaining the user's choice of the fault phenomenon for selective reporting; inputting the reported fault phenomenon into a pre-established fault diagnosis model based on Bayesian theorem, and obtaining the fault cause inferred by the fault diagnosis model based on the fault phenomenon.

[0004] The existing technology uses knowledge base to identify faults and processes fault information, but cannot interact in time when a fault occurs during the fault handling process, resulting in poor accuracy in locating the fault range, causing unnecessary repair attempts and misoperations of other devices when performing fault repairs. Summary of the invention

[0005] The embodiment of the present application solves the technical problem of low accuracy in locating the fault range and fault cause in the prior art by providing an online fault diagnosis method based on an interactive electronic manual, thereby improving the effect of fault locating.

[0006] The embodiment of the present application provides an online fault diagnosis method based on an interactive electronic manual, comprising: S1, determining an associated device corresponding to the fault information based on an uploaded standard fault information template;

[0007] S2, according to the fault information template, decompose the label fragments of the fault information to obtain multiple levels of fault information, and output them as the first target area, wherein the first target area includes obtaining hierarchical paragraphs, paragraphs, illustrations, multimedia, tables, warnings, notes, and annotated data information fragments;

[0008] S3, editing the acquired first target regions to determine the connection relationship between different first target regions;

[0009] S4, based on the connection relationship of the first target area, determining the change of the first target area when the fault occurs, and determining the second target area corresponding to the first target area;

[0010] S5, based on the acquired second target area, analyzing the faulty nodes in the second target area, and generating a corresponding fault information report.

[0011] The implementation of step S2 also includes:

[0012] S21, parsing the fault information based on the acquired fault information template, including determining the corresponding fault nodes in the fault information and the connection relationship between the fault nodes;

[0013] S22, based on the connection relationship of the faulty nodes, setting the fault information into multiple levels according to a preset mapping relationship;

[0014] S23, acquiring a first target area corresponding to the fault information according to the distribution, occurrence time, and frequency of the fault nodes.

[0015] The implementation of step S3 also includes:

[0016] S31, based on the acquired first target area, acquiring an impact range and a fault type corresponding to a faulty node in the first target area;

[0017] S32, determining a first correlation between different fault nodes according to the acquired fault type;

[0018] S33, determining a second correlation between different fault nodes according to the acquired impact range;

[0019] S34, determining connection relationships between different fault nodes based on the first correlation and the second correlation of the fault nodes.

[0020] The implementation of step S4 also includes:

[0021] S41, determining a first target set associated with the faulty node and a second target set associated with the faulty node and the current node to be detected in the first target area;

[0022] S42, establishing a standard data set corresponding to each first target area as a third target set according to the historical records under normal operation;

[0023] S43, comparing and analyzing the third target set with the first target set and the second target set in sequence, to determine a first matching score between the third target set and the first target set and a second matching score between the third target set and the second target set;

[0024] S44, determining a corresponding fault node based on the first matching score and the second matching score, and filling the corresponding first target area with data, and outputting the filled area as the second target area;

[0025] S45, analyzing and processing the faulty nodes and fault information in the second target area until the second target area reaches a preset matching standard threshold.

[0026] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0027] By uploading a standard fault information template, the fault information is associated with the corresponding equipment, and then the fault information is formatted and decomposed to obtain multiple levels of fault information as the first target area. By editing the first target area and determining the connection relationship, the impact range when the fault occurs and the corresponding second target area can be more accurately located; in the second target area, the fault nodes are reorganized and analyzed to determine the impact range under different conditions, and the corresponding weights are set, and finally the fault information report is output.

[0028] The first target area is processed and optimized through data filling and candidate point evaluation, thereby improving the accuracy and reliability of fault diagnosis; by setting thresholds and similarity comparisons, information and data that are truly related to the fault can be screened out, eliminating the possibility of interference and misjudgment; the problems of low fault diagnosis efficiency, low accuracy and difficulty in dealing with complex system faults in the prior art are effectively solved, thereby achieving the technical effect of improving fault diagnosis efficiency, improving diagnostic accuracy and adapting to complex system fault handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic flow chart of an online fault diagnosis method based on an interactive electronic manual of the present invention;

[0030] Figure 2 It is a flow chart of step S2 of an online fault diagnosis method based on an interactive electronic manual of the present invention;

[0031] Figure 3 It is a flow chart of step S3 of an online fault diagnosis method based on an interactive electronic manual of the present invention;

[0032] Figure 4The figure is a flow chart of step S4 of an online fault diagnosis method based on an interactive electronic manual of the present invention. DETAILED DESCRIPTION

[0033] To facilitate the understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0034] It should be noted that the terms “vertical”, “horizontal”, “up”, “down”, “left”, “right” and similar expressions used in this document are only for illustrative purposes and do not represent the only implementation method.

[0035] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which the present invention belongs; the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more related listed items. Embodiment 1

[0036] like Figure 1 As shown, the present application provides an online fault diagnosis method based on an interactive electronic manual, comprising:

[0037] S1, based on the uploaded standard fault information template, determining the associated device corresponding to the fault information, wherein the fault information template is used to convert the format of the received fault information and associate the fault information with the corresponding device.

[0038] S2, according to the fault information template, decompose the label fragments of the fault information to obtain multiple levels of fault information, and output them as the first target area. The first target area includes obtaining hierarchical paragraphs, paragraphs, illustrations, multimedia, tables, warnings, notes, and annotated data information fragments.

[0039] S3, editing the acquired first target regions, and determining connection relationships between different first target regions.

[0040] S4, based on the connection relationship of the first target area, determining the change of the first target area when the fault occurs, and determining the second target area corresponding to the first target area.

[0041] S5, based on the acquired second target area, analyzing the faulty nodes in the second target area, and generating a corresponding fault information report.

[0042] By using the interactive electronic manual (IETM), a link to the IETM is added to the fault diagnosis knowledge base, so that when a certain fault phenomenon or fault cause is selected, the relevant electronic manual chapter can be automatically called up for user reference, and the troubleshooting guide, maintenance steps and diagrams in the IETM can be used to enhance the accuracy and efficiency of fault diagnosis.

[0043] In the consultation-style fault reporting process, not only can the associated fault causes be automatically prompted based on the fault phenomenon selected by the user, but interactive troubleshooting guidance can also be provided in combination with IETM, such as step-by-step inspection, testing or maintenance guides.

[0044] Through the interactive elements in IETM (such as animations, videos, 3D models, etc.), users are guided to perform more complex fault diagnosis and maintenance operations; the historical maintenance data and expert system in IETM are used, combined with the fault diagnosis model of Bayesian theorem, to perform more accurate fault prediction and diagnosis; the maintenance records in IETM are associated with the node paths in the fault diagnosis knowledge base to continuously update and optimize the accuracy of fault diagnosis.

[0045] By integrating IETM's detailed maintenance information and historical data, the accuracy and efficiency of fault diagnosis can be significantly improved, and interactive troubleshooting guidance and maintenance guides can be provided, enabling users to understand and solve fault problems more easily, improving user satisfaction, reducing unnecessary maintenance attempts and misoperations, and thus reducing maintenance costs and time. Embodiment 2

[0046] In order to match the obtained fault information with the data in the electronic interactive manual, the fault information can be decomposed into multiple problems according to the fault type and the specific implementation method, and each problem can be matched to a suitable first target area. At this time, the first target area is the main detection and modification area, and the overall fault is diagnosed accordingly based on the obtained first target area.

[0047] Specifically, Figure 2 As shown, in step S2, the division form of the first target area includes:

[0048] S21, parsing the fault information based on the acquired fault information template, including determining the corresponding fault nodes in the fault information and the connection relationship between the fault nodes;

[0049] S22, based on the connection relationship of the faulty nodes, setting the fault information into multiple levels according to a preset mapping relationship;

[0050] S23, acquiring a first target area corresponding to the fault information according to the distribution, occurrence time, and frequency of the fault nodes.

[0051] Preferably, when the distribution density of the faulty nodes reaches a preset value, the corresponding area is selected as the first target area for output.

[0052] Preferably, the implementation method for decomposing the label fragments of the fault information is to determine the device node corresponding to the fault information based on each acquired fault information, and select the corresponding fault node based on the fault type corresponding to the fault node, and construct a decision tree. Based on the constructed decision tree, the acquired fault information is divided into multiple levels of fault nodes to determine the distribution of the fault information.

[0053] By constructing a decision tree based on the acquired fault information's labels, types, and mapping relationships between fault nodes, the output first target area can fully reflect the clustering relationship between the fault nodes, making it easier to judge the relationship between different fault nodes in the future. Embodiment 3

[0054] After acquiring the corresponding first target area, the fault information in the first target area is processed to determine the visual matching effect of the fault information, so that the auxiliary staff can process the fault in the first target area in real time according to the obtained visual information, thereby reducing the error of the staff in fault identification.

[0055] Specifically, Figure 3 As shown, the implementation of step S3 includes:

[0056] S31, based on the acquired first target area, acquiring an impact range and a fault type corresponding to a faulty node in the first target area;

[0057] S32, determining a first correlation between different fault nodes according to the acquired fault type;

[0058] S33, determining a second correlation between different fault nodes according to the acquired impact range;

[0059] S34, determining connection relationships between different fault nodes based on the first correlation and the second correlation of the fault nodes.

[0060] The first correlation is used to refer to the connection between different fault nodes based on common or similar fault types. If two fault nodes have the same or similar fault types, they can be considered to be related in the first correlation.

[0061] The second correlation is used to refer to the connection between different fault nodes based on overlapping or interrelated influence ranges. If the influence ranges of two fault nodes overlap or are interdependent, they can be considered to be related in terms of the second correlation.

[0062] The first correlation and the second correlation obtained also include:

[0063] Analyze the fault type of each fault node, compare the similarities existing under the same fault node when the fault occurs, and output the data whose similarity value is greater than the preset threshold, and take the data whose similarity of the fault node is greater than the preset threshold as the first correlation of the fault node;

[0064] When analyzing the influence range of the faulty node, the influence ranges corresponding to the faulty nodes are compared, the overlapping parts between the faulty nodes are determined, and the data whose overlapping parts are larger than the preset range are output, and the data whose overlapping parts of the faulty nodes are larger than the preset range are taken as the second correlation of the faulty node.

[0065] The overlapping part described here refers to the hardware components, software applications, system processes, etc. that can be affected by the current faulty node, and there is an upper and lower connection relationship between the devices affected by the current faulty node. At this time, obtaining the corresponding data can assist in troubleshooting the faulty device.

[0066] In the fault diagnosis process, the fault type of each fault node is first analyzed in detail. By comparing the similarities that exist under the same fault node when the fault occurs, the common characteristics between the faults can be identified. When the similarity value is greater than the preset threshold, these data will be extracted and output as an important basis for measuring the first correlation between fault nodes.

[0067] Next, the impact range of the faulty nodes is further analyzed. By comparing the impact ranges corresponding to different faulty nodes, the overlapping parts between them can be determined. When the overlapping parts exceed the preset range, this indicates that there is a significant correlation in the impact ranges of these faulty nodes, so it is used as the second correlation between the faulty nodes.

[0068] Specifically, after obtaining the first correlation and the second correlation, step S34 further includes:

[0069] When analyzing the fault node, in addition to obtaining the similarity of the fault type corresponding to the fault node, obtain the fault description corresponding to the fault node where the fault occurs. The fault description includes the symptoms, manifestations, and environmental conditions of the fault. Compare the text information similar to the current fault description and use the text similarity algorithm to determine the similarity score between the fault nodes.

[0070] Based on the obtained similarity scores, the weights between the current fault nodes are determined;

[0071] Based on the acquired weights between the faulty nodes, the connection relationship between the faulty nodes is determined, and the local influence of each faulty node is set to determine the final correlation of each faulty node under the current local influence as the connection relationship of the faulty nodes output.

[0072] When establishing a connection relationship for a fault node, different fault nodes are connected based on the first correlation and the second correlation of the fault node as edges and the weights between the fault nodes as points. For the connected fault nodes, the local influence of the fault node is determined based on the connection relationship between the fault node and other fault nodes. Based on the local influence and weights between the fault nodes, the superimposed mean of the fault nodes is output as the connection relationship of the first target area.

[0073] The local impact of the faulty nodes indicates the mutual impact between the faulty nodes, to determine whether each faulty node can be caused to fail by other faulty nodes or be significantly affected.

[0074] By associating the first correlation and the second correlation with the fault description, when evaluating the scope of the fault, the impact range of the fault and the correlation between faults under different impacts can be accurately obtained, making the identification of the fault scope faster and facilitating the processing and positioning of the fault. Embodiment 4

[0075] After obtaining the connection relationship between the corresponding fault nodes in the first target area, the fault information in each area is processed to determine the corresponding content of the fault information after processing and the changes in the corresponding nodes after receiving the fault information, as well as the actual impact between different nodes when the fault occurs.

[0076] After the first target area is determined, the fault information inside each first target area is processed, and the matching effect between the attributes and elements in the first target area and the data under normal operation is determined.

[0077] Specifically, Figure 4 As shown, the implementation of step S4 also includes:

[0078] S41, determining a first target set associated with the faulty node and a second target set associated with the faulty node and the current node to be detected in the first target area.

[0079] The first target set obtained is used to represent the corresponding connection relationship between similar fault nodes and similar fault nodes under the corresponding connection relationship in the first target area. The second target set is used to select the data set between the nodes that currently need to be detected, and is used to determine the data related to the actual fault nodes.

[0080] S42: Establish a standard data set corresponding to each first target area according to the historical records under normal operation as a third target set.

[0081] The acquired third target set is used to express the data of normal operation in the first target area, and these normal data are compared with the first target set and the second target set to determine the corresponding matching degree when an abnormality occurs, so as to assist in locating and analyzing the fault.

[0082] S43, comparing and analyzing the third target set with the first target set and the second target set in sequence, and determining a first matching score between the third target set and the first target set and a second matching score between the third target set and the second target set.

[0083] The obtained first matching score and the second matching score are calculated by using cosine similarity to determine the corresponding matching degree when a fault occurs.

[0084] S44, based on the first matching score and the second matching score, determine the corresponding fault node, fill the corresponding first target area with data, and output the filled area as the second target area.

[0085] The first matching score and the second matching score are compared to determine which set is more mismatched with the standard set; data is filled for nodes corresponding to the set with a higher mismatch, and nearest neighbor filling is used for data of corresponding fault nodes to obtain a sufficiently sufficient second target area.

[0086] Specifically, the implementation method of data filling includes: determining a fault node whose first matching score and second matching score are both lower than a preset threshold, and determining the corresponding fault information in the fault node as the first target information; obtaining the most recent data of the first target information as the second target information; when the first matching score and the second matching score of the second target information are both greater than the preset threshold, and the first matching score of the second target information is greater than the second matching score, assigning the value of the current second target information to the first target information; when the first matching score and the second matching score of the second target information are both greater than the preset threshold, and the first matching score of the second target information is less than the second matching score, assigning the value of the previous second target information to the first target information, and outputting the first target area in which the first target information is modified as the second target area.

[0087] After obtaining the first matching score and the second matching score, the processing method corresponding to the first target area also includes:

[0088] Based on the first matching score and the second matching score obtained, a candidate point corresponding to the fault information is selected, and different weights of the scores are set based on the fault type and fault impact corresponding to the candidate point to obtain a score corresponding to the fault of the candidate point. When the score of the candidate point is lower than a threshold, the corresponding candidate point is deleted and the candidate point is not matched;

[0089] When a candidate point is deleted, the deleted candidate point is evaluated based on the candidate points that have not been deleted subsequently, and the score of the deleted candidate point is re-determined. When the score of the deleted candidate point is higher than the threshold, the deleted candidate point is retained again; the similarity between each candidate point and the deleted candidate point that is retained again is calculated, and the fault information corresponding to the candidate point with the greatest similarity is filled with data.

[0090] At this time, the data with too low matching scores is removed so that the data with higher current scores can be processed.

[0091] S45, analyzing and processing the faulty nodes and fault information in the second target area until the second target area reaches a preset matching standard threshold.

[0092] This is an iterative optimization process, the purpose of which is to gradually bring the data in the first target area closer to the normal state by continuously analyzing and processing the faulty nodes; analyzing the filled second target area to check whether it meets the preset matching criteria (such as the matching score is higher than a certain threshold); if not, repeat step S44 to adjust the data filling strategy or identify new faulty nodes for processing; continue to optimize until the matching criteria are met or the preset number of iterations is reached.

[0093] By continuously iterating the data in the first target area, the data that has changed during the iteration process can be known, so that the location and node where the fault occurs can be located more quickly, making the fault diagnosis more efficient. Embodiment 5

[0094] After obtaining the second target area, the obtained second target area is judged again, and the corresponding fault nodes are shuffled and analyzed to determine whether the current fault node analysis results can meet the fault threshold under different conditions, and the final fault occurrence point and fault cause are output to assist in handling the fault.

[0095] Specifically, the implementation of step S5 includes:

[0096] Based on the acquired second target area, the fault nodes in the second target area are reorganized and analyzed to determine the influence range of the fault nodes under different conditions, weights are set for the fault nodes based on the influence range of the fault nodes, and according to the fault weights of the fault nodes, a fault information report corresponding to the second target area is output.

[0097] Specifically, the corresponding fault nodes in the acquired second target area are re-analyzed to determine the range of influence that each fault node can have under different combinations, so as to facilitate further restrictions on the scope and impact of the fault. At the same time, when restricting the fault nodes, corresponding weights are set for the fault nodes, and the impact of the fault nodes is calculated based on the weights. The detailed information and weight report of the fault nodes are output to obtain a fault information report corresponding to the second target area.

[0098] According to the impact range of each faulty node under different conditions, an initial weight is assigned to it, and nodes with a wider impact range should be given a higher weight.

[0099] The initial weights are adjusted considering the failure frequency and potential consequences of the faulty nodes, and the weights of nodes with higher failure frequency or more serious potential consequences should be increased.

[0100] All weights are normalized so that their sum equals a predetermined value (e.g., 100%).

[0101] Output the fault information report corresponding to the second target area, design a report template, including tables, charts, and text descriptions. The fault information report contains information such as the identification, location, impact range, weight, and potential consequences of each fault node; use visualization tools to enhance the readability and intuitiveness of the report, highlight the weight and impact range of different fault nodes through color coding, size changes, etc., review and verify the report to ensure the accuracy and completeness of the data, and then publish the report to relevant personnel or teams so that appropriate actions can be taken based on the information in the report.

[0102] The reorganization analysis of the faulty node includes obtaining the faulty node in the second target area, determining the minimum fault set and the maximum fault set in which the faulty node fails, determining the path overlap rate of the minimum fault set and the maximum fault set, and when the path overlap rate is greater than a threshold, outputting the minimum fault set corresponding to the faulty node.

[0103] Among them, the minimum fault set represents the smallest set of all possible paths when a fault event occurs, and the maximum fault set represents the largest set of possible paths when a fault occurs. The path overlap rate is determined based on the proportion of common elements in the minimum fault set and the maximum fault set. Statistical methods can be used to analyze the distribution of the overlap rate of fault sets, and based on this distribution, a suitable threshold can be set to adjust the minimum fault set selected for processing.

[0104] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An online fault diagnosis method based on an interactive electronic manual, characterized in that: include: S1, based on the uploaded standard fault information template, determine the associated device corresponding to the fault information; S2, according to the fault information template, decompose the label fragments of the fault information to obtain multiple levels of fault information, and output them as the first target area, wherein the first target area includes obtaining hierarchical paragraphs, paragraphs, illustrations, multimedia, tables, warnings, notes, and annotated data information fragments; S3, editing the acquired first target regions to determine the connection relationship between different first target regions; S4, based on the connection relationship of the first target area, determining the change of the first target area when the fault occurs, and determining the second target area corresponding to the first target area; S5, based on the acquired second target area, analyzing the faulty nodes in the second target area and generating a corresponding fault information report; The implementation of step S2 also includes: S21, parsing the fault information based on the acquired fault information template, including determining the corresponding fault nodes in the fault information and the connection relationship between the fault nodes; S22, based on the connection relationship of the faulty nodes, setting the fault information into multiple levels according to a preset mapping relationship; S23, acquiring a first target area corresponding to the fault information according to the distribution, occurrence time, and frequency of the fault nodes; The implementation of step S3 also includes: S31, based on the acquired first target area, acquiring an impact range and a fault type corresponding to a faulty node in the first target area; S32, determining a first correlation between different fault nodes according to the acquired fault type; S33, determining a second correlation between different fault nodes according to the acquired impact range; S34, determining connection relationships between different fault nodes based on the first correlation and the second correlation of the fault nodes; After obtaining the first correlation and the second correlation, step S34 further includes: Obtain the fault description corresponding to the fault node where the fault occurred. The fault description includes the symptoms, manifestations, and environmental conditions of the fault. Compare the text information similar to the current fault description to determine the similarity score between the fault nodes. Based on the obtained similarity scores, the weights between the current fault nodes are determined; Based on the acquired weights between the fault nodes, the connection relationship between the fault nodes is determined, and the local influence of each fault node is set to determine the final correlation of each fault node under the current local influence as the output connection relationship of the fault nodes; When establishing a connection relationship for a fault node, different fault nodes are connected according to the first correlation and the second correlation of the fault node as edges and the weights between the fault nodes as points. For the connected fault nodes, the local influence of the fault node is determined according to the connection relationship between the fault node and other fault nodes. Based on the local influence and weights between the fault nodes, the superimposed mean of the fault nodes is output as the connection relationship of the first target area. The implementation of step S4 also includes: S41, determining a first target set associated with the faulty node and a second target set associated with the faulty node and the current node to be detected in the first target area; S42, establishing a standard data set corresponding to each first target area as a third target set according to the historical records under normal operation; S43, comparing and analyzing the third target set with the first target set and the second target set in sequence, to determine a first matching score between the third target set and the first target set and a second matching score between the third target set and the second target set; S44, determining a corresponding fault node based on the first matching score and the second matching score, and filling the corresponding first target area with data, and outputting the filled area as the second target area; S45, analyzing and processing the fault nodes and fault information in the second target area until the second target area reaches a preset matching standard threshold; Step S5 further includes, based on the acquired second target area, performing a reorganization analysis on the faulty nodes in the second target area, determining the influence range of the faulty nodes under different conditions, setting weights for the faulty nodes based on the influence ranges of the faulty nodes, and outputting a fault information report corresponding to the second target area according to the fault weights of the faulty nodes; The implementation method of data filling in step S44 includes: determining a fault node whose first matching score and second matching score are both lower than a preset threshold, and determining the corresponding fault information in the fault node as the first target information; obtaining the most recent data of the first target information as the second target information; when the first matching score and the second matching score of the second target information are both greater than the preset threshold, and the first matching score of the second target information is greater than the second matching score, assigning the value of the current second target information to the first target information; when the first matching score and the second matching score of the second target information are both greater than the preset threshold, and the first matching score of the second target information is less than the second matching score, assigning the value of the previous second target information to the first target information, and outputting the first target area in which the first target information is modified as the second target area; After obtaining the first matching score and the second matching score, the processing method corresponding to the first target area also includes: Based on the first matching score and the second matching score obtained, a candidate point corresponding to the fault information is selected, and different weights of the scores are set based on the fault type and fault impact corresponding to the candidate point to obtain a score corresponding to the fault of the candidate point. When the score of the candidate point is lower than a threshold, the corresponding candidate point is deleted and the candidate point is not matched; When a candidate point is deleted, the deleted candidate point is evaluated based on the candidate points that have not been deleted subsequently, and the score of the deleted candidate point is re-determined. When the score of the deleted candidate point is higher than the threshold, the deleted candidate point is retained again; the similarity between each candidate point and the deleted candidate point that is retained again is calculated, and the fault information corresponding to the candidate point with the greatest similarity is filled with data.

2. The online fault diagnosis method based on an interactive electronic manual as claimed in claim 1, characterized in that: The first correlation and the second correlation obtained also include: Analyze the fault type of each fault node, compare the similarities existing under the same fault node when the fault occurs, and output the data whose similarity value is greater than the preset threshold, and take the data whose similarity of the fault node is greater than the preset threshold as the first correlation of the fault node; When analyzing the influence range of the faulty node, the influence ranges corresponding to the faulty nodes are compared, the overlapping parts between the faulty nodes are determined, and the data whose overlapping parts are larger than the preset range are output, and the data whose overlapping parts of the faulty nodes are larger than the preset range are taken as the second correlation of the faulty node.

3. The online fault diagnosis method based on interactive electronic manual as claimed in claim 1, characterized in that: The reorganization analysis of the faulty node includes obtaining the faulty node in the second target area, determining the minimum fault set and the maximum fault set in which the faulty node fails, determining the path overlap rate of the minimum fault set and the maximum fault set, and when the path overlap rate is greater than a threshold, outputting the minimum fault set corresponding to the faulty node.

Images