A dynamic updating method for the information model of digital fault diagnosis manual

Abstract

The present invention provides a method for dynamically updating a digital fault diagnosis manual information model, including: collecting basic information related to fault diagnosis, and processing the collected basic information into information elements; calling and combining multiple various information elements to form fault diagnosis information units; calling and combining fault diagnosis information units to organize fault diagnosis plans according to the manual outline, and forming a fault diagnosis manual after multiple fault diagnosis plans are released; automatically maintaining and dynamically updating the fault diagnosis manual according to the product's fault cases and diagnostic process. The present invention can improve the interactivity of the fault diagnosis manual, reduce the difficulty of operation, reduce the possibility of errors in the fault location and handling process, and reduce the event escalation and security impact caused by secondary faults; at the same time, through the dynamic update of the fault diagnosis manual, the time and labor costs consumed by the repeated production and update of the fault diagnosis manual are reduced, and the availability of the manual is improved.

Description

Technical Field

[0001] The invention relates to the technical field of electronic equipment maintenance, and in particular to a method for dynamically updating a digital fault diagnosis manual information model. Background Art

[0002] The fault diagnosis manual is an important part of the use information of electronic equipment. Its role is to guide users or professional maintenance personnel to identify, locate and eliminate faults safely, effectively and efficiently. It is generally delivered with the product and is widely used in various electronic equipment in aviation, aerospace, automobiles and ships. With the continuous advancement of modern science and technology, the application of new processes, new technologies and new materials has made the performance of electronic equipment develop towards complexity, automation and intelligence. Traditional fault diagnosis manuals can no longer meet the use requirements well. In this context, the development and use of fault diagnosis manuals organized in a digital way and with strong interactivity has become a trend around the world.

[0003] However, the digital fault diagnosis manuals for various electronic equipment currently have problems such as inconsistent information structure, irregular and imperfect content structure, and difficulty in updating. Modern high-end manufacturing equipment and key safety systems often have complex functions, structures, and technologies. Suppliers or users' understanding of product failures needs to be continuously deepened as product use and fault resolution cases continue to accumulate, requiring fault diagnosis manuals to have the ability to quickly update and iterate based on the collection of fault resolution cases. However, once the fault diagnosis manuals currently available on the market are formulated, the difficulty and workload of later updates are huge. They often rely on experts to manually sort out the fault conditions that are not correctly covered from a huge number of cases, and manually correct or add the content of the fault diagnosis manual. Therefore, the fault diagnosis manual is often updated infrequently, or even no longer updated after a period of use, which cuts off the connection between fault cases and fault diagnosis manuals, seriously affecting the correct troubleshooting and maintenance operation execution of after-sales tasks, and thus causing users to be dissatisfied with product services, product quality, and even corporate brands. Summary of the invention

[0004] The present invention aims to provide a method for dynamically updating the information model of a digital fault diagnosis manual, so as to solve the problems existing in the digital fault diagnosis manuals of various electronic equipments.

[0005] The present invention provides a method for dynamically updating a digital fault diagnosis manual information model, comprising the following steps:

[0006] S1, collect basic information related to fault diagnosis, and process the collected basic information into information elements;

[0007] S2, calling and combining multiple information elements to form a fault diagnosis information unit;

[0008] S3, calling and combining fault diagnosis information units, organizing fault diagnosis solutions according to the manual outline, and forming a fault diagnosis manual after multiple fault diagnosis solutions are released;

[0009] S4, automatically maintains and dynamically updates the fault diagnosis manual based on the product's fault cases and diagnosis process.

[0010] Further, step S1 includes:

[0011] Collect basic information related to fault diagnosis;

[0012] Classify and organize information elements based on basic information and fault diagnosis business requirements;

[0013] Standardize the classified and organized information elements according to unified rules for content and format;

[0014] The standardized information elements are encoded to form an information element library.

[0015] Furthermore, the basic information related to fault diagnosis includes product materials, fault information, fault diagnosis resource information, fault diagnosis safety information and other information; information elements include maintenance items, fault modes, schematics, physical pictures, tool information, material information, personnel information, general safety information, personnel safety information, equipment safety information and environmental safety information.

[0016] Further, step S2 includes:

[0017] Based on information elements, according to the characteristics and content of fault diagnosis work, information is combined and standardized to form fault diagnosis information units;

[0018] Fault diagnosis information units are classified based on fault codes and coded for storage management.

[0019] Furthermore, if the information elements cannot meet the requirements of the fault diagnosis information unit after being combined, relevant information is added to the fault diagnosis information unit.

[0020] Further, step S3 includes:

[0021] Create a fault diagnosis scheme based on fault code, and let the fault code / name library be F = {f1,f2,...,f n}, where f i Represents a fault code in the library, which is consistent with the fault code f i Related elements have fault phenomenon fp i 、Location methodfm i and diagnostic resource collection Ft i ={ft1,ft2,...,fm}, Ft i ∈FT, FT is the diagnostic resource library; the fault phenomenon fp i Cause of failure Fr i There may be one or more, expressed mathematically as fp i ←{fr1,fr2,...,fr q}, where {fr1,fr2,...,fr q}∈FR, FR is the fault cause library, There are solutions fs o ;

[0022] The fault diagnosis plan is compiled into a fault diagnosis manual according to the diagnosis preparation, fault cause, fault location and solution measures:

[0023] Diagnostic preparation: Select fault code f i The corresponding diagnostic resource set Ft i ={ft1,ft2,...,f m}, Ft i ∈FT, and call the fault diagnosis information unit, associated with Ft i Relevant operating manuals or usage conditions are compiled into precautions and safe operating conditions before fault diagnosis;

[0024] Fault cause: Select fault code f i The underlying cause set Fr i ={fr1,fr2,...,fr q}, Fr i ∈FR, and displayed as a list;

[0025] Fault location: compilation and fault cause Fr i Corresponding fault location method fm i , guide the troubleshooting of fault causes;

[0026] Solution: If there is any fault phenomenon fp i Cause of failure Fr i , all need to provide solutions fs o .

[0027] Furthermore, in step S4, the dynamic update of the fault diagnosis manual includes two categories:

[0028] On the one hand, the content of the fault diagnosis manual is automatically optimized according to the newly added fault diagnosis cases;

[0029] On the other hand, it is to count the probability of failure and optimize the ranking of failure causes and failure solutions.

[0030] Furthermore, let the fault phenomenon of the newly added fault case be fcp j , the positioning method is fcm j , the diagnostic resource is fct j , the fault reason is fcr j , the solution is fcs j ;

[0031] First, the fault phenomenon fcp is identified based on the fault matching model j Can it correspond to a fault code in the fault diagnosis manual? i Fault phenomenon fp i , mathematically expressed as At the same time, fcr j ∈Fr i , fct j ∈Ft i , Explain the cause of the failure case in f i The fault cause set, and the fault case location method, solution and f i Consistent, the required diagnostic resources are in f i In the diagnostic resource collection of the fault diagnosis manual, the content does not need to be updated, only the fault case needs to be associated with the f i and the cause of the fault, and add 1 to the number of fault occurrences;

[0032] like fcr j ∈Fr i , but there is fcm j ≠fm i , fs o ≠fcs j In any case, it means that the fault case can be mapped to a fault code in the fault diagnosis manual, but the positioning method, solution and required diagnostic resources are different from those in the fault diagnosis manual. Then, the different parts are reprocessed and produced as information elements according to the method in step S3, and are associated with the corresponding fault code and fault cause, and the number of fault occurrences is increased by 1;

[0033] like If the fault cause of the fault case is not in the fault cause set of the fault code in the matched fault diagnosis manual, you need to add the fault cause fcr of the new fault case. j With fault code f i Associate and fcm j , fct j , fcs j , process and produce the information element according to the method in step S3, and compare it with the fault code f iAssociated, add 1 to the number of fault occurrences;

[0034] If fcp j ≠fp j , which means that the fault case matching fails, and a new fault code f needs to be created according to the encoding rules new , fp j , fcm j , fct j , fcr j , fcs j , process and produce the information element according to the method in step S3, and compare it with the fault code f new The number of fault occurrences is set to 1.

[0035] Furthermore, the content of the fault diagnosis manual is automatically optimized based on the newly added fault diagnosis cases, and all of them need to be passed to experts for verification and review before being published. When optimizing the sorting of fault causes and fault solutions, the fault diagnosis manual uses the number of fault occurrences as a parameter and weights them in combination with the maintenance difficulty coefficient to re-sort the fault code and fault cause lists, and automatically updates without review.

[0036] Furthermore, the fault matching model construction method includes:

[0037] The fault matching model uses a deep learning pre-training model as a base, and uses the fine-grained differences of fault feature words to enhance the matching degree. Specifically, the representative fault words in the fault description text are used as fine-grained features. By building a fault word library, the word vectors corresponding to the difference feature words are used as difference feature vectors, and the word vectors corresponding to the similar feature words are used as similar feature vectors. The differences between the difference feature words are amplified by the confidence level, and the matching degree between the similar feature words is enhanced.

[0038] Let the fault phenomenon feature word vector of the fault case be f a , the fault phenomenon feature word vector in the fault diagnosis manual is f b , then the cosine similarity is:

[0039]

[0040] Let P represent the probability of successful matching, and C(a,b) be the feature confidence of each pair of word vectors:

[0041]

[0042] Among them, P∈[0,1], C(a,b)∈(0,1).

[0043] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0044] The present invention can improve the interactivity of the fault diagnosis manual, reduce the difficulty of operation, reduce the possibility of errors in the fault location and handling process, and reduce the event escalation and safety impact caused by secondary faults; at the same time, through the method of dynamically updating the diagnostic manual, the time and labor costs consumed in repeated production and updating of the manual are reduced, and the usability of the manual is improved, thereby guiding on-site maintenance personnel to complete fault identification, isolation and repair tasks more quickly and efficiently, shortening the troubleshooting time and improving the company's product image and service quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0046] Figure 1 It is a basic flow chart of the method for dynamically updating the information model of the digital fault diagnosis manual in an embodiment of the present invention.

[0047] Figure 2 A schematic diagram of a fault diagnosis manual in an embodiment of the present invention.

[0048] Figure 3 Schematic diagram of the automatic update process of the fault diagnosis manual in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0050] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] Example

[0052] like Figure 1 As shown, this embodiment proposes a method for dynamically updating a digital fault diagnosis manual information model, comprising the following steps:

[0053] S1, collect basic information related to fault diagnosis, and process the collected basic information into information elements; specifically including:

[0054] S11, collecting basic information related to fault diagnosis; in this embodiment, product materials, fault information, fault diagnosis resource information, fault diagnosis safety information and other information are used as basic information related to fault diagnosis; the details are as follows:

[0055] Product materials: product structure information, schematic diagram and description, wiring diagram or table, physical materials, description of the product's built-in self-test device, etc.;

[0056] Failure information: theoretical failure analysis information (such as FMECA, FTA, etc.), trial production process failure cases, use process failure cases and failure cases of similar products (generally speaking, failure cases will include failure phenomenon, cause and solution information);

[0057] Fault diagnosis resource information: including the collection of information on personnel, tools, and materials involved in troubleshooting operations;

[0058] Fault diagnosis safety information: including general safety information and personnel safety, equipment safety and environmental safety information related to troubleshooting operations;

[0059] Other information: such as abbreviations of terms, public material information, etc.

[0060] S12, classify and organize information elements based on basic information and fault diagnosis business requirements; information elements include maintenance items, fault modes, schematics, physical pictures, tool information, material information, personnel information, general safety information, personnel safety information, equipment safety information, environmental safety information, etc.;

[0061] S13, standardize the classified and organized information elements according to unified rules of content and format, and store them in media such as pictures, three-dimensional models, and rich media;

[0062] S14, encoding the standardized information elements to form an information element library.

[0063] S2, calling and combining multiple information elements to form a fault diagnosis information unit;

[0064] Specifically, based on information elements such as the fault phenomenon library FP, the positioning method library FM, the fault cause library FR, the solution library FS and the diagnostic resource FT, information is combined and standardized according to the characteristics and content of the fault diagnosis operation to form a fault diagnosis information unit, and classified based on the fault code, the fault cause can be standardized according to the combination of the fault object and the fault mode; information elements such as general safety information, personnel safety information, equipment safety information, and environmental safety information are called to combine and standardize the information to form an operation safety information unit; if the requirements of the fault diagnosis information unit cannot be met after the combination of information elements, relevant information can be added to the fault diagnosis information unit.

[0065] S3, call and combine fault diagnosis information units, organize fault diagnosis plans according to the manual outline, and form a fault diagnosis manual after multiple fault diagnosis plans are released. After verifying the completeness, correctness, applicability and operability of the plan, it can be put into the database and released.

[0066] like Figure 2 As shown in the figure, a fault diagnosis scheme is created with fault code as the subject, and the fault code / name library is F = {f1, f2, ..., f n}, where f i Represents a fault code in the library, which is consistent with the fault code f i Related elements have fault phenomenon fp i 、Location methodfm i and diagnostic resources Ft i ={ft1,ft2,...,f m}, Ft i ∈FT, FT is the diagnostic resource library; the fault phenomenon fp i Cause of failure Fr i There may be one or more, expressed mathematically as fp i ←{fr1,fr2,...,fr q}, where {fr1,fr2,...,fr q}∈FR, FR is the fault cause library, There are solutions fs o .

[0067] The fault diagnosis plan is compiled into a fault diagnosis manual according to the diagnosis preparation, fault cause, fault location and solution measures:

[0068] Diagnostic preparation: Select fault code f i The corresponding diagnostic resource set Ft i ={ft1,ft2,...,f m}, Ft i∈FT, and call the fault diagnosis information unit, associated with Ft i Relevant operating manuals or usage conditions are compiled into precautions and safe operating conditions before fault diagnosis;

[0069] Fault cause: Select fault code f i The underlying cause set Fr i ={fr1,fr2,...,fr q}, Fr i ∈FR and display it in a list. The list of fault causes can be arranged according to the probability of fault occurrence or in a preferred order such as easy first then difficult, external first then internal, etc., to predetermine the best troubleshooting process.

[0070] Fault location: compilation and fault cause Fr i Corresponding fault location method fm i , guide the troubleshooting of the fault. The fault location method is in fm i According to the operation sequence, it can be divided into multiple diagnostic steps (j is the order of operations), Before execution, prompt the security information of the current step as needed and link to the required resource collection Diagnostic steps When executed, describe the action and expected results and responses of the operation. For example: If phenomenon fmp1 occurs, jump to step If phenomenon fmp2 occurs, jump to step If phenomenon fmp3 occurs, positioning is completed and the cause of the fault is determined to be fr i Each The decision branches should be clear and should not form an infinite loop. Possible error messages and recovery procedures should be explained. In addition, The technical information required for troubleshooting (schematic diagrams, flow charts, wiring diagrams, location diagrams, etc.) should be linked and provided.

[0071] Solution. If there is any malfunction fp i Cause of failure Fr i ,Right now All need to provide solutions fs o For the selected solution, you can link to the maintenance manual and repair / replacement work instructions to guide the implementation of the solution; link to the functional inspection file after the fault is repaired to guide the confirmation of fault elimination.

[0072] S4, automatically maintains and dynamically updates the fault diagnosis manual based on the product's fault cases and diagnosis process.

[0073] like Figure 3 As shown in the figure, the dynamic update of the troubleshooting manual includes two categories:

[0074] On the one hand, the content of the fault diagnosis manual is automatically optimized according to the newly added fault diagnosis cases;

[0075] On the other hand, it is to count the probability of failure and optimize the ranking of failure causes and failure solutions.

[0076] Let the fault phenomenon of the newly added fault case be fcp j , the positioning method is fcm j , the diagnostic resource is fct j , the fault reason is fcr j , the solution is fcs j .

[0077] First, the fault phenomenon fcp is identified based on the fault matching model j Can it correspond to a fault code in the fault diagnosis manual? i Fault phenomenon fp i , mathematically expressed as At the same time, fcr j ∈Fr i , fct j ∈Ft i , Explain the cause of the failure case in f i The fault cause set, and the fault case location method, solution and f i Consistent, the required diagnostic resources are in f i In the diagnostic resource collection of the fault diagnosis manual, the content does not need to be updated, only the fault case needs to be associated with the f i and its cause, and add 1 to the number of fault occurrences;

[0078] like fcr j ∈Fr i , but there is fcm j ≠fm i , fs o ≠fcs j In any case (“≠” means mismatch), it means that the fault case can be matched to a fault code in the fault diagnosis manual, but the location method, solution and required diagnostic resources are different from the manual information model. Then the different part is reprocessed and produced as an information element according to the method in step S3, and is associated with the corresponding fault code and fault cause, and the number of fault occurrences is increased by 1;

[0079] like If the fault cause of the fault case is not in the fault cause set of the fault code in the matched fault diagnosis manual, you need to add the fault cause fcr of the new fault case. j With fault code f i Associate and fcm j , fct j , fcs j , process and produce the information element according to the method in step S3, and compare it with the fault code f i Associated, add 1 to the number of fault occurrences;

[0080] If fcp j ≠fp j , which means that the fault case matching fails, and a new fault code f needs to be created according to certain coding rules new , fp j , fcm j , fct j , fcr j , fcs j , process and produce the information element according to the method in step S3, and compare it with the fault code f new The number of fault occurrences is set to 1.

[0081] It should be noted that the content of the fault diagnosis manual is automatically optimized based on the newly added fault diagnosis cases, and it needs to be passed to experts for verification and review before being released; when optimizing the sorting of fault causes and fault solutions, the fault diagnosis manual uses the number of fault occurrences as a parameter and weights it in combination with the maintenance difficulty coefficient to re-sort the fault code and fault cause list, and automatically updates without review.

[0082] Furthermore, the fault matching model construction method includes:

[0083] The fault matching model can use deep learning pre-training models such as BiLSTM and BERT as the base, and use the fine-grained differences of fault feature words to enhance the matching degree. Specifically, the representative fault words in the fault description text are used as fine-grained features. For example, in the case phenomenon "The RF excitation terminal occasionally outputs a weak signal, resulting in intermittent speech" and the manual fault phenomenon "The RF excitation terminal receives insufficient gain", the matching focus is on the difference between "output signal" and "receive gain"; on the contrary, in the case phenomenon "The RF excitation terminal occasionally outputs a weak signal, resulting in intermittent speech" and the manual fault phenomenon "The RF excitation terminal transmits insufficient power", the matching focus is on the similarity between "output signal" and "transmit power". Therefore, this embodiment constructs a fault word library, and uses the word vector corresponding to the difference feature word as the difference feature vector, and uses the word vector corresponding to the similar feature word as the similar feature vector. The difference between the difference feature words is amplified by the confidence, and the matching degree between the similar feature words is enhanced, which effectively improves the model's recognition ability of fine-grained difference features of faults.

[0084] Let the fault phenomenon feature word vector of the fault case be f a , the fault phenomenon feature word vector in the fault diagnosis manual is f b , then the cosine similarity is:

[0085]

[0086] Let P represent the probability value (similarity) of successful matching, C(a,b) be the feature confidence of each pair of word vectors, P∈[0,1], C(a,b)∈(0,1).

[0087]

[0088] By scaling the fault matching value through the above formula, the matching model is not limited to text similarity, thus improving the success rate and accuracy of fault matching.

[0089] 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. A method for dynamically updating a digital fault diagnosis manual information model, characterized in that: The following steps are involved: S1, collect basic information related to fault diagnosis, and process the collected basic information into information elements; S2, calling and combining multiple information elements to form a fault diagnosis information unit; S3, calling and combining fault diagnosis information units, organizing fault diagnosis solutions according to the manual outline, and forming a fault diagnosis manual after multiple fault diagnosis solutions are released; S4, automatically maintains and dynamically updates the fault diagnosis manual based on the product's fault cases and diagnosis process.

2. The method for dynamically updating the information model of a digital fault diagnosis manual according to claim 1, characterized in that: Step S1 includes: Collect basic information related to fault diagnosis; Classify and organize information elements based on basic information and fault diagnosis business requirements; Standardize the classified and organized information elements according to unified rules for content and format; The standardized information elements are encoded to form an information element library.

3. The method for dynamically updating the information model of a digital fault diagnosis manual according to claim 2, characterized in that: The basic information related to fault diagnosis includes product materials, fault information, fault diagnosis resource information, fault diagnosis safety information and other information; information elements include maintenance items, fault modes, schematics, physical pictures, tool information, material information, personnel information, general safety information, personnel safety information, equipment safety information and environmental safety information.

4. The method for dynamically updating the digital fault diagnosis manual information model according to claim 1, characterized in that: Step S2 includes: Based on information elements, according to the characteristics and content of fault diagnosis work, information is combined and standardized to form fault diagnosis information units; Fault diagnosis information units are classified based on fault codes and coded for storage management.

5. The method for dynamically updating the information model of a digital fault diagnosis manual according to claim 4, characterized in that: If the combination of information elements cannot meet the requirements of the fault diagnosis information unit, relevant information is added to the fault diagnosis information unit.

6. The method for dynamically updating the information model of a digital fault diagnosis manual according to claim 1, characterized in that: Step S3 includes: Create a fault diagnosis scheme based on fault code, and let the fault code / name library be F = {f1,f2,...,f n }, where f i Represents a fault code in the library, which is consistent with the fault code f i Related elements have fault phenomenon fp i 、Location methodfm i and diagnostic resource collection Ft i ={ft1,ft2,...,f m }, Ft i ∈FT, FT is the diagnostic resource library; the fault phenomenon fp i Cause of failure Fr i There may be one or more, expressed mathematically as fp i ←{fr1,fr2,...,fr q }, where {fr1,fr2,...,fr q }∈FR, FR is the fault cause library, There are solutions fs o ; The fault diagnosis plan is compiled into a fault diagnosis manual according to the diagnosis preparation, fault cause, fault location and solution measures: Diagnostic preparation: Select fault code f i The corresponding diagnostic resource set Ft i ={ft1,ft2,...,f m }, Ft i ∈FT, and call the fault diagnosis information unit, associated with Ft i Relevant operating manuals or usage conditions are compiled into precautions and safe operating conditions before fault diagnosis; Fault cause: Select fault code f i The underlying cause set Fr i ={fr1,fr2,...,fr q }, Fr i ∈FR, and displayed as a list; Fault location: compilation and fault cause Fr i Corresponding fault location method fm i , guide the troubleshooting of fault causes; Solution: If there is any fault phenomenon fp i Cause of failure Fr i , all need to provide solutions fs o .

7. The method for dynamically updating the information model of a digital fault diagnosis manual according to claim 6, characterized in that: In step S4, the dynamic update of the fault diagnosis manual includes two types: On the one hand, the content of the fault diagnosis manual is automatically optimized according to the newly added fault diagnosis cases; On the other hand, it is to count the probability of failure and optimize the ranking of failure causes and failure solutions.

8. The method for dynamically updating the information model of a digital fault diagnosis manual according to claim 7, characterized in that: Let the fault phenomenon of the newly added fault case be fcp j , the positioning method is fcm j , the diagnostic resource is fct j , the fault reason is fcr j , the solution is fcs j ; First, the fault phenomenon fcp is identified based on the fault matching model j Can it correspond to a fault code in the fault diagnosis manual? i Fault phenomenon fp i , mathematically expressed as At the same time, fcr j ∈Fr i , fct j ∈Ft i , Explain the cause of the failure case in f i The fault cause set, and the fault case location method, solution and f i Consistent, the required diagnostic resources are in f i In the diagnostic resource collection of the fault diagnosis manual, the content does not need to be updated, only the fault case needs to be associated with the f i and the cause of the fault, and add 1 to the number of fault occurrences; like fcr j ∈Fr i , but there is fcm j ≠fm i , fs o ≠fcs j In any case, it means that the fault case can be mapped to a fault code in the fault diagnosis manual, but the positioning method, solution and required diagnostic resources are different from those in the fault diagnosis manual. Then, the different parts are reprocessed and produced as information elements according to the method in step S3, and are associated with the corresponding fault code and fault cause, and the number of fault occurrences is increased by 1; like If the fault cause of the fault case is not in the fault cause set of the fault code in the matched fault diagnosis manual, you need to add the fault cause fcr of the new fault case. j With fault code f i Associate and fcm j , fct j , fcs j , process and produce the information element according to the method in step S3, and compare it with the fault code f i Associated, add 1 to the number of fault occurrences; If fcp j ≠fp j , which means that the fault case matching fails, and a new fault code f needs to be created according to the encoding rules new , fp j , fcm j , fct j , fcr j , fcs j , process and produce the information element according to the method in step S3, and compare it with the fault code f new The number of fault occurrences is set to 1.

9. The method for dynamically updating the information model of a digital fault diagnosis manual according to claim 7 or 8, characterized in that: The contents of the fault diagnosis manual are automatically optimized based on newly added fault diagnosis cases. All contents need to be passed to experts for verification and review before being published. When optimizing the sorting of fault causes and fault solutions, the fault diagnosis manual uses the number of fault occurrences as a parameter and weights them in combination with the maintenance difficulty coefficient to re-sort the fault code and fault cause lists, and automatically updates without review.

10. The method for dynamically updating the information model of a digital fault diagnosis manual according to claim 8, characterized in that: The fault matching model construction method comprises: The fault matching model uses a deep learning pre-training model as a base, and uses the fine-grained differences of fault feature words to enhance the matching degree. Specifically, the representative fault words in the fault description text are used as fine-grained features. By building a fault word library, the word vectors corresponding to the difference feature words are used as difference feature vectors, and the word vectors corresponding to the similar feature words are used as similar feature vectors. The differences between the difference feature words are amplified by the confidence level, and the matching degree between the similar feature words is enhanced. Let the fault phenomenon feature word vector of the fault case be f a , the fault phenomenon feature word vector in the fault diagnosis manual is f b , then the cosine similarity is: Let P represent the probability of successful matching, and C(a,b) be the feature confidence of each pair of word vectors: Among them, P∈[0,1], C(a,b)∈(0,1).

Images