Construction method of electrical fault expert knowledge base and expert system

Abstract

The invention specifically discloses a construction method of an electrical fault expert knowledge base and an expert system. The method comprises the following steps: collecting fault maintenance records of electrical equipment of a production line and experience knowledge of maintenance experts; performing data preprocessing on the collected data and forming an electrical fault concept set, wherein the data preprocessing comprises data cleaning, format conversion and data classification; constructing a relation model of an electrical fault knowledge base based on the electrical fault concept set so as to carry out formatted standardization and digital storage on electrical fault knowledge; constructing a knowledge base table structure based on the relation model, wherein the knowledge base table structure comprises a rule table, a condition table, a variable table, a conclusion table and an interpretation table; and setting a primary key and a foreign key to establish an inter-table association relationship through the primary key and the foreign key so as to form an electrical fault expert knowledge base. The electrical fault detection efficiency can be improved, and standardization of maintenance experience management and knowledge training is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of expert knowledge bases, and more specifically, to a method for constructing an electrical fault expert knowledge base and an expert system. Background Art

[0002] Electrical failures occur from time to time in industrial sites, but due to their suddenness, concealment, randomness and other characteristics, electrical failures are sometimes difficult to quickly discover and solve. If electrical failures cannot be solved quickly and effectively, it may affect production and even cause serious safety accidents. On the other hand, in the daily maintenance of equipment, experienced operators can quickly and effectively solve problems with their rich practical experience, but this expert experience is often difficult to be effectively passed on. Therefore, how to quickly solve electrical failures and effectively pass on their solutions and experience is of great significance. Summary of the invention

[0003] The present invention provides a method for constructing an electrical fault expert knowledge base and an expert system, which solves the problem that electrical faults in existing industrial sites are difficult to solve quickly and accurately, and difficult to pass on experience, can improve the efficiency of electrical fault detection, and realize the standardization of maintenance experience management and knowledge training.

[0004] To achieve the above objectives, the present invention provides the following technical solutions:

[0005] A method for constructing an electrical fault expert knowledge base, comprising:

[0006] Collect the fault maintenance records of the electrical equipment on the production line and the experience and knowledge of maintenance experts;

[0007] Performing data preprocessing on the collected data and forming an electrical fault concept set, wherein the data preprocessing includes: data cleaning, format conversion and data classification;

[0008] Building a relational model of an electrical fault knowledge base based on the electrical fault concept set to format and standardize and digitally store electrical fault knowledge;

[0009] Constructing a knowledge base table structure based on the relational model, the knowledge base table structure comprising: a rule table, a condition table, a variable table, a conclusion table and an explanation table;

[0010] Set primary keys and foreign keys to establish relationships between tables through primary keys and foreign keys, thereby forming an electrical fault expert knowledge base.

[0011] Preferably, it also includes:

[0012] A fault tree is constructed according to the characteristics of each process in the production line and historical fault data, and a diagnosis model for each production process is established through the fault tree.

[0013] Preferably, it also includes:

[0014] Collect the fault phenomena of each production process and establish a fault phenomenon set for each process;

[0015] Assigning a corresponding membership value to each fault phenomenon in the fault phenomenon set, and establishing a membership table of the fault phenomena;

[0016] Fuzzy reasoning rules are set and an electrical fault diagnosis model based on fuzzy reasoning is constructed to associate the fault phenomenon with the fault cause.

[0017] Preferably, it also includes:

[0018] The association relationships within the knowledge base are visualized in combination with the graphic database to realize the interface of fault diagnosis.

[0019] Preferably, it also includes:

[0020] A collection layer is set up, and the collection layer is connected to the PLC, sensors, DCS system and control system signals on the production line through a programming interface to collect production data and production parameters in real time during the production process, and combine them with fault maintenance records and the experience and knowledge of maintenance experts to form fault knowledge data of the expert knowledge base.

[0021] The present invention also provides an expert system of an electrical fault expert knowledge base, using the above construction method, comprising: a client layer, a business layer, a platform layer, a knowledge layer and a collection layer;

[0022] The acquisition layer is used to collect real-time data, historical data, fault repair data and expert experience knowledge of each process on the production line;

[0023] The client layer is used to divide the permissions into three categories: maintenance experts, maintenance teams, and training students according to different user responsibilities and tasks;

[0024] The business layer is used to set the business modules of the knowledge base, and the business modules include: status monitoring, fault reasoning, knowledge display, knowledge training and system management;

[0025] The knowledge layer provides a data foundation for the platform layer and the business layer;

[0026] The platform layer is used to set up an electrical fault diagnosis model, and through the connection with the knowledge layer, the expert knowledge in the knowledge base is called into the required electrical fault diagnosis, and the fault diagnosis results are displayed to the user through the business layer.

[0027] Preferably, the knowledge layer includes: PLC program data, expert knowledge data, collected real-time data and maintenance case data.

[0028] Preferably, the status monitoring includes: status detection of main equipment in each process, real-time data detection and abnormal alarm.

[0029] Preferably, the fault reasoning includes: maintenance knowledge, fault reasoning process and associative search;

[0030] Help users find fault results through associative search, and display the fault reasoning process and maintenance knowledge.

[0031] Preferably, the knowledge training includes: maintenance cases, online examinations and expert knowledge learning;

[0032] Users learn expert knowledge and maintenance cases and take tests through online exams.

[0033] The present invention provides a method and an expert system for constructing an electrical fault expert knowledge base, which collects the fault maintenance records of electrical equipment on a production line and the experience and knowledge of maintenance experts, constructs a relational model of the electrical fault knowledge base, performs grid standardization and digital storage, and then forms an electrical fault expert knowledge base by constructing a knowledge base table structure. The problem that electrical faults on existing industrial sites are difficult to solve quickly and accurately, and that experience is difficult to pass on, can be solved, the efficiency of electrical fault detection can be improved, and the standardization of maintenance experience management and knowledge training can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the specific embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below.

[0035] Figure 1 It is a schematic diagram of a method for constructing an electrical fault expert knowledge base provided by the present invention.

[0036] Figure 2 It is a schematic diagram of a vacuum rehumidification fault tree of a tobacco factory thread-making line provided by an embodiment of the present invention.

[0037] Figure 3 It is a structural schematic diagram of an electrical fault expert knowledge base provided by the present invention.

[0038] Figure 4 It is a functional architecture diagram of an expert system of an electrical fault expert knowledge base provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0039] In order to enable persons skilled in the art to better understand the solutions of the embodiments of the present invention, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings and implementation modes.

[0040] In view of the problem that electrical faults in existing industrial sites are difficult to solve quickly and accurately, and it is difficult to pass on experience, the present invention provides an electrical fault expert knowledge base and a construction method thereof, which solves the problem that electrical faults in existing industrial sites are difficult to solve quickly and accurately, and it is difficult to pass on experience, can improve the efficiency of electrical fault detection, and realize the standardization of maintenance experience management and knowledge training.

[0041] like Figure 1 As shown, a method for constructing an electrical fault expert knowledge base includes:

[0042] S1: Collect the fault maintenance records of electrical equipment on the production line and the experience and knowledge of maintenance experts;

[0043] S2: preprocessing the collected data and forming an electrical fault concept set, wherein the data preprocessing includes: data cleaning, format conversion and data classification;

[0044] S3: constructing a relational model of an electrical fault knowledge base based on the electrical fault concept set to format and standardize and digitally store electrical fault knowledge;

[0045] S4: constructing a knowledge base table structure based on the relational model, wherein the knowledge base table structure includes: a rule table, a condition table, a variable table, a conclusion table and an explanation table;

[0046] S5: Set primary keys and foreign keys to establish associations between tables through the primary keys and foreign keys, thereby forming an electrical fault expert knowledge base.

[0047] Specifically, internal data collection mainly involves communicating with professional maintenance technicians and then studying the PLC programs of the main controllers of different devices to obtain corresponding fault knowledge. External data collection involves obtaining relevant fault knowledge through some sensor hardware and detectors.

[0048] For the knowledge base relational model, according to the association between the nodes of electrical faults and the logical management between the fault causes, the electrical fault information relational model is constructed to realize the formatting specification and digital storage of electrical fault knowledge. At present, there are two main types of fault knowledge data for the target silk thread, namely, the fault maintenance records recorded by the maintenance team and the expert experience knowledge based on the maintenance experts. Fault maintenance records belong to semi-structured data, while expert experience knowledge belongs to unstructured data, and most of them are some corpus information. Therefore, the fault knowledge data type is complex, including text, pictures, tables, videos, etc. In order to meet the high accuracy of fault knowledge and the effect of visual display, the electrical fault expert knowledge base of this system is designed based on the Neo4j database, which is a popular NOSQL graph database. For the construction of the knowledge base relational model, a combination of "top-down" and "bottom-up" is adopted.

[0049] First of all, determining the concept set is the first step in designing the electrical fault knowledge graph. After clarifying the design requirements, through in-depth research and analysis of the relevant knowledge in the field of electrical faults, the fault terms and knowledge that appear frequently are summarized. At the same time, the unstructured and semi-structured data such as electrical fault cases are summarized and sorted using a systematic approach.

[0050] Secondly, under the premise of ensuring the rationality of concept division, it is necessary to accurately describe the knowledge system and the relationship between knowledge points in the field of electrical faults, filter out knowledge with low relevance and high repetition rate, and make reasonable corrections to ambiguous and confusing terms, so as to form an accurate and complete set of core concepts of electrical faults. This core concept set mainly includes vacuum moisture recovery machine, loose moisture recovery machine, wire cutting machine, wire drying machine, feeder, temporary storage cabinet and other equipment in silk making equipment.

[0051] Finally, the focus of the study is the framework of the core concept set of electrical faults, and the main research scope is the common electrical faults of production equipment. Through such systematic analysis and organization, a clear-structured and complete electrical fault knowledge system can be established to provide strong support for fault diagnosis and processing.

[0052] After designing the knowledge base relationship model, the storage of electrical fault expert knowledge needs to be refined, so the knowledge base table needs to be designed. When designing the knowledge base table structure, consider using appropriate indexing and query optimization techniques to further improve the performance and efficiency of the system. By adding indexes to frequently queried fields, data retrieval can be accelerated and query time can be reduced. In addition, the relationship between knowledge base tables should be reasonably designed to avoid data redundancy and ensure data integrity and accuracy.

[0053] In addition to the design of the knowledge base table, it is also necessary to consider how the knowledge base tables are associated with each other. The tables are associated through primary keys and foreign keys. For the primary key, there are the rule number, condition number, and conclusion number in the rule table (as shown in Table 1), the condition table (as shown in Table 2), and the conclusion table (as shown in Table 3); compared with the primary key, the condition number, conclusion number, and explanation number in the rule table are set as foreign keys of the condition table, conclusion table, and explanation table (as shown in Table 4) respectively. In addition, consider introducing other constraints, such as unique constraints, default value constraints, etc., to ensure the consistency and integrity of the data. At the same time, regularly optimize the performance of the knowledge base and clean up the data, delete expired or useless data, reduce the burden on the knowledge base, and improve the overall operation efficiency of the system.

[0054] By designing the knowledge base table structure, making rational use of indexes and constraints, and regularly performing performance optimization and data cleaning, the system can be made more efficient and reliable in knowledge base reasoning and information retrieval, providing users with better services and support.

[0055]

[0056]

[0057]

[0058]

[0059] The method also includes: constructing a fault tree according to the characteristics of each process in the production line and historical fault data, and establishing a diagnosis model for each production process through the fault tree.

[0060] In practical applications, in view of the hidden and random characteristics of electrical faults, the fault tree method is used to obtain fault knowledge, fuzzy reasoning is used to diagnose electrical faults, and an electrical fault diagnosis model based on fuzzy reasoning is constructed to solve the problems of difficulty in acquiring fault knowledge and poor model consistency.

[0061] In one embodiment, the silk production line in a tobacco factory has the characteristics of complex structure, numerous production equipment, and various fault forms. The fault tree analysis method is used to transform the complex production line fault diagnosis task into the fault diagnosis problem of each process according to the characteristics of each process in the silk production line. Through in-depth analysis of each process and its working principle of the silk production line, and after summarizing and analyzing the historical fault data and expert experience of the silk production line, a fault tree of the vacuum rehumidification section of the silk production line is established, such as Figure 2 shown.

[0062] The method also includes: collecting fault phenomena of each production process and establishing a fault phenomenon set for each process; assigning a corresponding membership value to each fault phenomenon in the fault phenomenon set and establishing a membership table of the fault phenomena; setting fuzzy reasoning rules and constructing an electrical fault diagnosis model based on fuzzy reasoning to associate the fault phenomenon with the fault cause.

[0063] Specifically, when building an electrical fault diagnosis expert system, the first thing to solve is to obtain the fuzzy knowledge of wire thread faults and write all electrical faults into the expert system as accurately as possible. The diagnostic model of each subsystem can be established separately through the fault tree analysis results.

[0064] (1) For each different production process, all possible fault phenomena are collected and extracted, and a possible fault phenomenon set X = {x 1 ,x 2,⋯,x n}.

[0065] (2) Since the occurrence of production equipment failure is a gradual process, there is usually a transition process from normal state to failure. During this process, the state of the production equipment is between normal and failure, neither completely "normal" nor completely "failure". Fuzzy semantics are usually used to describe the phenomenon in this process. Therefore, it is necessary to communicate with maintenance experts to obtain the membership values ​​of the fault phenomena corresponding to different semantic descriptions to better serve the fault reasoning process. The results are shown in Table 5.

[0066]

[0067] (3) Analyze the causes of the failures as much as possible in each production process and establish a set of failure causes for each process. .

[0068] (4) Establish fuzzy reasoning expert rules. The general form is: IF condition THEN conclusion, which means that if the condition is met, the corresponding conclusion is established; otherwise, the rule base will be searched to see if there is an unused rule that matches the known condition. If it exists, it will be executed. If not, the relevant known conditions need to be supplemented. The production rule adopts the standard structure "if-then". This structure makes it easy to combine the production rule with the credibility factor to form a CF model, thereby realizing the function of uncertainty reasoning.

[0069] (5) Establish an electrical fault diagnosis model based on fuzzy reasoning. Before diagnosing production equipment through fuzzy reasoning, a corresponding fuzzy diagnosis model must be created. The specific steps are as follows:

[0070] 1) Create a set of fault symptoms , that is, the set of all possible fault phenomena that may appear when a fault occurs. , where the element Represents the phenomena when various faults occur.

[0071] 2) Create a fault cause set , which is the set of all possible causes of failure:

[0072] , where the element Indicates various reasons that cause the failure.

[0073] 3) Create a fuzzy matrix , that is, the fuzzy matrix between fault phenomena and fault causes:

[0074] ;

[0075] in, , , , fuzzy matrix As an important characterization matrix, it is used to describe fault phenomena and cause of failure In this matrix, each element Both represent a quantitative fuzzy value, reflecting the fault phenomenon Cause of failure The magnitude of these values ​​indicates the depth of the association between the two, which is crucial to the accuracy of the diagnosis results. However, the initial setting of these fuzzy values ​​can only rely on methods such as expert knowledge and experience.

[0076] As time goes by, the environment and conditions of the object being diagnosed will change, such as changes in performance and parameters after long-term operation. This change will lead to continuous changes in the relationship between the cause of the fault and the fault phenomenon. In order to ensure that the diagnostic matrix can accurately reflect this relationship and adapt to changes, it is necessary to continuously adaptively adjust the diagnostic fuzzy matrix. Through a certain learning mechanism, the diagnostic matrix can continuously adapt to this change and improve the accuracy and reliability of the diagnosis. This adaptive adjustment is a key step in the diagnostic process, which can help the system better understand the complex relationship between the fault phenomenon and the cause of the fault, thereby improving the efficiency and accuracy of the diagnosis.

[0077] When using fuzzy reasoning for fault diagnosis, whether it is a single fault phenomenon or a multi-fault phenomenon, the diagnostic results obtained may be cross-linked and overlapping, so the diagnostic results should be highly comprehensive, that is, fuzzy comprehensive decision. The decision model is:

[0078] ;

[0079] in, Represents the weight vector, element Indicated by The weight of the diagnosis result corresponding to the fault phenomenon. Represents the final diagnosis result, element Indicates corresponding The credibility of the fault, the one with the highest credibility is selected as the conclusion inferred by the reasoning mechanism.

[0080] The method also includes: visualizing the association relationship in the knowledge base in combination with the graphic database to realize the interface of fault diagnosis.

[0081] In actual applications, Java is selected as the backend development language, and the convenient configuration and development methods provided by the Spring Boot framework are used. Tomcat is responsible for running and managing WAR packages as an application server, while Nginx is used as a reverse proxy server to process front-end requests, thereby achieving efficient operation and management of the system. The knowledge base in the system uses two database technologies: a large relational database MS SQL SERVER 2016 and a graph database based on Neo4j. MS SQL SERVER2016 is a single-process, multi-threaded database system that can more effectively utilize hardware resources than multi-process, single-threaded databases. As an extension of the Windows system, SQL Server is different from other databases. It does not need to sacrifice performance when running on different operating systems, and can give full play to the performance advantages of the Windows system.

[0082] The method also includes: setting up a collection layer, which is connected to the PLC, sensors, DCS system and control system signals on the production line through a programming interface to collect production data and production parameters in real time during the production process, and combine them with fault maintenance records and the experience and knowledge of maintenance experts to form fault knowledge data of the expert knowledge base.

[0083] It can be seen that the present invention provides a method for constructing an electrical fault expert knowledge base, which collects the fault maintenance records of electrical equipment on the production line and the experience knowledge of maintenance experts, constructs a relational model of the electrical fault knowledge base, and performs grid standardization and digital storage, and then forms an electrical fault expert knowledge base by constructing a knowledge base table structure. This solves the problem that electrical faults in existing industrial sites are difficult to solve quickly and accurately, and that it is difficult to pass on experience, can improve the efficiency of electrical fault detection, and realize the standardization of maintenance experience management and knowledge training.

[0084] Accordingly, if Figure 3 As shown, the present invention also provides an electrical fault expert knowledge base, using the above-mentioned construction method, including: a client layer, a business layer, a platform layer, a knowledge layer and a collection layer.

[0085] The acquisition layer is used to collect real-time data, historical data, fault repair data and expert experience knowledge of each process on the production line.

[0086] The client layer is used to divide the authority into three categories: maintenance experts, maintenance teams, and training students according to different user responsibilities and tasks.

[0087] The business layer is used to set the business modules of the knowledge base, and the business modules include: status monitoring, fault reasoning, knowledge display, knowledge training and system management.

[0088] The knowledge layer provides a data foundation for the platform layer and the business layer.

[0089] The platform layer is used to set up an electrical fault diagnosis model, and through the connection with the knowledge layer, the expert knowledge in the knowledge base is called into the required electrical fault diagnosis, and the fault diagnosis results are displayed to the user through the business layer.

[0090] Furthermore, the knowledge layer includes: PLC program data, expert knowledge data, collected real-time data and maintenance case data.

[0091] Furthermore, the status monitoring includes: status detection of main equipment in each process, real-time data detection and abnormal alarm.

[0092] Furthermore, the fault reasoning includes: maintenance knowledge, fault reasoning process and associative search;

[0093] Help users find fault results through associative search, and display the fault reasoning process and maintenance knowledge.

[0094] Furthermore, the knowledge training includes: maintenance cases, online examinations and expert knowledge learning;

[0095] Users learn expert knowledge and maintenance cases and take tests through online exams.

[0096] Specifically, at the customer layer: according to different user responsibilities and tasks, permissions are divided into three categories: maintenance experts, maintenance teams, and training students. All system functional permissions are opened for maintenance experts; fault reasoning management and maintenance management permissions are opened for maintenance teams; and knowledge learning and examination permissions are opened for training personnel.

[0097] Business layer: data interface and background services for the five functional modules of status monitoring, fault reasoning, knowledge display, knowledge training and system management. It is also the interface between users and the system. Users interact directly with the system through the business layer and use specific functions in the functional modules according to the permission distribution. In addition, the business layer and the knowledge layer cannot directly exchange data, and only extract information from the knowledge base through fixed interfaces and drivers. When changing or updating the knowledge base system, the redesign of the business layer is avoided through data access, and only the data interface of the data access method needs to be changed.

[0098] Platform layer: Contains fuzzy knowledge acquisition and electrical fault diagnosis models. Through the connection with the knowledge layer, the expert knowledge in the knowledge base is called to the required electrical fault diagnosis link, and the fault diagnosis results are better presented to users through the business layer.

[0099] Knowledge layer: The knowledge layer provides a data foundation for fault reasoning and the business layer. The knowledge layer contains four types of data: 1) PLC program. By checking the PLC program in the main controller of each device on the silk-making line, we can observe which step in the PLC program cannot be executed when an electrical fault occurs. 2) Expert knowledge. By collecting the accumulated expert experience of silk-making workshop maintenance experts, we can help the fault diagnosis reasoning process and locate the cause of the fault more quickly. 3) Real-time data. The data reflecting the real-time status of the silk-making line obtained through the collection system by measuring instruments such as temperature and humidity sensors is usually used for real-time status monitoring of silk-making equipment. 4) Maintenance cases. Collect typical maintenance history on the silk-making line and record it in a book to help quickly solve the problem when similar faults occur.

[0100] In one embodiment, according to the production characteristics of the target tobacco factory and the needs of the enterprise, an electrical fault expert system is established, which is divided into five main functional modules: status monitoring, fault reasoning, knowledge display, knowledge training and system management. The functional architecture of the electrical fault expert system is as follows: Figure 4 shown.

[0101] Status monitoring: mainly includes overview of main equipment in each process, real-time data detection and abnormal alarm. According to the actual situation of the target silk thread, the silk thread is divided into 10 production links, and each production link is displayed in the expert system with a 2D model. In the overview screen of each production link, the working status parameters of the production equipment can be seen in real time, and an alarm will be issued when an abnormality occurs.

[0102] Fault reasoning: In the fault reasoning function, associative search can be performed to help users find the desired results more quickly; the fault reasoning process can be viewed, and the PLC program of the main controller that causes the fault is presented to the user, helping users quickly identify the ultimate source of the fault and provide relevant maintenance knowledge.

[0103] Knowledge display: The fault expert knowledge base is displayed to the user. In this functional module, the user can see the fault tree model used for fault reasoning and share the fault expert knowledge.

[0104] Knowledge training: Users can learn fault expert knowledge in the system and check their learning results by taking exams, changing the traditional training model of senior employees helping the new employees, which can help companies reduce expenses on personnel training.

[0105] System management: mainly implements the basic settings of the system including system configuration and user management. The administrator creates accounts for users and assigns different permissions according to different roles such as maintenance experts, maintenance teams and training personnel, making maintenance management more hierarchical and multifunctional.

[0106] It can be seen that the present invention provides an expert system of an electrical fault expert knowledge base, which constructs five levels, namely, the customer layer, the business layer, the platform layer, the knowledge layer and the collection layer, to form an electrical fault expert system. It solves the problem that electrical faults in existing industrial sites are difficult to solve quickly and accurately, and it is difficult to pass on experience, which can improve the efficiency of electrical fault detection and realize the standardization of maintenance experience management and knowledge training.

[0107] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any ordinary technician in the industry can smoothly implement the present invention as shown in the drawings and described above. However, any equivalent changes, modifications and evolutions made by technicians familiar with the profession without departing from the scope of the technical solution of the present invention using the technical content disclosed above are all equivalent embodiments of the present invention. At the same time, any equivalent changes, modifications and evolutions made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the technical solution of the present invention.

Claims

1. A method for constructing an electrical fault expert knowledge base, characterized in that: include: Collect the fault maintenance records of the electrical equipment on the production line and the experience and knowledge of maintenance experts; Preprocessing the collected data and forming an electrical fault concept set, wherein the data preprocessing includes: data cleaning, format conversion and data classification; Building a relational model of an electrical fault knowledge base based on the electrical fault concept set to format and standardize and digitally store electrical fault knowledge; Constructing a knowledge base table structure based on the relational model, the knowledge base table structure comprising: a rule table, a condition table, a variable table, a conclusion table and an explanation table; Set primary keys and foreign keys to establish relationships between tables through primary keys and foreign keys, thereby forming an electrical fault expert knowledge base.

2. The method for constructing an electrical fault expert knowledge base according to claim 1, characterized in that: Also includes: A fault tree is constructed according to the characteristics of each process in the production line and historical fault data, and a diagnosis model for each production process is established through the fault tree.

3. The method for constructing an electrical fault expert knowledge base according to claim 2, characterized in that: Also includes: Collect the fault phenomena of each production process and establish a fault phenomenon set for each process; Assigning a corresponding membership value to each fault phenomenon in the fault phenomenon set, and establishing a membership table of the fault phenomena; Fuzzy reasoning rules are set and an electrical fault diagnosis model based on fuzzy reasoning is constructed to associate the fault phenomenon with the fault cause.

4. The method for constructing an electrical fault expert knowledge base according to claim 3, characterized in that: Also includes: The association relationships within the knowledge base are visualized in combination with the graphic database to realize the interface of fault diagnosis.

5. The method for constructing an electrical fault expert knowledge base according to claim 4, characterized in that: Also includes: A collection layer is set up, and the collection layer is connected to the PLC, sensors, DCS system and control system signals on the production line through a programming interface to collect production data and production parameters in real time during the production process, and combine them with fault maintenance records and the experience and knowledge of maintenance experts to form fault knowledge data of the expert knowledge base.

6. An expert system of an electrical fault expert knowledge base, using the construction method according to any one of claims 1 to 5, characterized in that: include: Customer layer, business layer, platform layer, knowledge layer and collection layer; The acquisition layer is used to collect real-time data, historical data, fault repair data and expert experience knowledge of each process on the production line; The client layer is used to divide the permissions into three categories: maintenance experts, maintenance teams, and training students according to different user responsibilities and tasks; The business layer is used to set the business modules of the knowledge base, and the business modules include: status monitoring, fault reasoning, knowledge display, knowledge training and system management; The knowledge layer provides a data foundation for the platform layer and the business layer; The platform layer is used to set up an electrical fault diagnosis model, and through the connection with the knowledge layer, the expert knowledge in the knowledge base is called into the required electrical fault diagnosis, and the fault diagnosis results are displayed to the user through the business layer.

7. The expert system of the electrical fault expert knowledge base according to claim 6, characterized in that: The knowledge layer includes: PLC program data, expert knowledge data, collected real-time data and maintenance case data.

8. The expert system of the electrical fault expert knowledge base according to claim 7, characterized in that: The status monitoring includes: status detection of main equipment in each process, real-time data detection and abnormal alarm.

9. The expert system of the electrical fault expert knowledge base according to claim 8, characterized in that: The fault reasoning includes: maintenance knowledge, fault reasoning process and associative search; Help users find fault results through associative search, and display the fault reasoning process and maintenance knowledge.

10. The expert system of electrical fault expert knowledge base according to claim 9, characterized in that: The knowledge training includes: maintenance cases, online examinations and expert knowledge learning; Users learn expert knowledge and maintenance cases and take tests through online exams.

Images