Microwave oven electrical fault diagnosis method and device, and storage medium

Abstract

The invention provides a microwave oven electrical fault diagnosis method and device and a storage medium, and the method comprises the steps: S1, obtaining the electrical fault data of a microwave oven, and carrying out the data preprocessing; s2, based on the preprocessed electrical fault data, entity identification, knowledge extraction and knowledge fusion processing are respectively carried out, and an electrical fault knowledge graph is established; s3, system fault retrieval is carried out based on the electrical fault knowledge graph, and a Bayesian network is constructed according to an electrical fault query result obtained through retrieval; and S4, determining the prior probability of occurrence of the fault cause and the conditional probability of each node in the Bayesian network, and performing reasoning by using the Bayesian network to obtain the fault cause causing the fault phenomenon. According to the method, the knowledge graph and the Bayesian network are combined, so that possible fault components can be quickly positioned, and the diagnosis accuracy is improved; and meanwhile, by continuously expanding and updating the knowledge graph and the Bayesian network, the diagnosis capability of the electrical fault can be continuously improved.

Description

Technical Field

[0001] The present invention relates to the technical field of household appliances, and in particular to a method, a device and a storage medium for diagnosing electrical faults of a microwave oven. Background Art

[0002] With the continuous development of science and technology and the continuous improvement of people's living standards, microwave ovens have become an indispensable kitchen appliance in modern families. Microwave ovens are bound to have various faults after working for a long time; the traditional fault diagnosis method mainly relies on the subjective description of users and manual troubleshooting and experience judgment of maintenance personnel, which is not only inefficient but also difficult to guarantee accuracy.

[0003] As the internal structure and functional components of microwave ovens become more complex, the difficulty of fault diagnosis continues to increase. Although some microwave ovens on the market are equipped with some simple fault self-detection functions, most of these functions are based on preset fault codes or simple sensor monitoring, and cannot achieve comprehensive intelligent fault diagnosis. Knowledge graphs are the cornerstone of the transition from perceptual intelligence to cognitive intelligence. They can make up for the lack of interpretability of traditional manual fault diagnosis results and solve the isolation of data. Using knowledge graphs, we can conduct a deeper analysis of microwave oven electrical faults, thereby better predicting and diagnosing electrical faults and promoting the intelligent upgrade of equipment repair and maintenance.

[0004] In view of this, the present invention is proposed. Summary of the invention

[0005] The problem solved by the invention is to provide a microwave oven fault diagnosis method which is low in cost and easy to implement to meet the market demand.

[0006] To solve the above problems, the present invention provides a method for diagnosing electrical faults of a microwave oven, comprising:

[0007] S1, obtaining microwave oven electrical fault data and performing data preprocessing;

[0008] S2. Based on the pre-processed electrical fault data, entity recognition, knowledge extraction and knowledge fusion processing are performed to establish an electrical fault knowledge graph;

[0009] S3. Perform system fault retrieval based on the electrical fault knowledge graph, and construct a Bayesian network based on the electrical fault query results obtained by the retrieval;

[0010] S4. Determine the prior probability of the fault cause and the conditional probability of each node in the Bayesian network, and use the Bayesian network to perform reasoning to obtain the fault cause that causes the fault phenomenon.

[0011] Preferably, step S2 comprises:

[0012] S21: Using ontology modeling technology, establish an ontology set of microwave ovens, including: an entity set of microwave oven structures, an entity set of various electrical fault phenomena manifested during electrical faults, an entity set of electrical fault causes, and an entity set of maintenance suggestions;

[0013] S22: Establish the ontology relationship of the microwave oven, and express the relationship between the concepts of the motor ontology, including: the cause of the electrical fault of the microwave oven, the sensor monitors the operating status of the microwave oven, the fault leads to the occurrence of other phenomena, the electrical fault phenomenon of the microwave oven occurs, and measures are taken to repair it;

[0014] S23: Based on the established knowledge ontology representation of microwave oven electrical fault diagnosis, information is extracted from the data and a microwave oven electrical fault triple relationship table is established, the relationships are marked, and the electrical fault knowledge is represented as a triple structure: entity-relationship-entity;

[0015] S24: The graph database Neo4j is used to store triple knowledge data. The nodes in the graph database represent entities, which correspond to the entity relationships in the triples. The directed edges represent the relationships between knowledge, which correspond to the relationships in the triples. The knowledge graph of microwave oven electrical faults is constructed.

[0016] Preferably, step S2 also includes S25, verifying the accuracy of the knowledge graph using a link prediction method.

[0017] Preferably, step S3 includes:

[0018] S31, after preprocessing the fault information, searching for fault nodes with the same fault phenomenon through the constructed microwave oven electrical fault knowledge graph, and searching for the structure and maintenance suggestions of the microwave ovens that have searched out these fault nodes;

[0019] S32, and establishing an electrical fault knowledge graph query subgraph according to the fault node, the structural entity of the microwave oven, the fault cause and the maintenance suggestion;

[0020] S33. Based on the fault knowledge graph query subgraph, the Bayesian network structure modeling is performed using the Bayesian network visualization method.

[0021] Preferably, step S4 includes: S41, selecting possible electrical fault causes based on the Bayesian network and constructing a clique tree; S42, initializing message transmission and continuously updating messages until convergence; S43, calculating the posterior probability of each fault cause using the converged clique tree.

[0022] Preferably, the message transmission formula in step S42 is: i→j (X j )=ΣP(X i |Xj )ΠM k→i (X i ), where M i→j (X j ) is the message sent by group i to group j, X i and X j are the node sets of cluster i and cluster j respectively, and Neighbors(i) is the set of neighboring clusters of cluster i.

[0023] Preferably, step S4 also includes: recording the probability values ​​of various fault causes of the fault information and the probabilities of various solutions under the fault causes into the microwave oven electrical fault knowledge graph.

[0024] The present invention also provides a device for implementing a method for diagnosing electrical faults in a microwave oven, comprising a processor and a memory; wherein the processor is configured to execute the method for diagnosing electrical faults in a microwave oven; and the memory is used to store executable instructions of the processor.

[0025] The present invention also provides a computer storable medium on which a computer program is stored. The computer program is executed by a processor to implement the microwave oven electrical fault diagnosis method.

[0026] Compared with the prior art, the microwave oven electrical fault diagnosis method and device described in the present invention have the following beneficial effects: 1) by real-time monitoring and analysis of the working status of each component of the microwave oven and semantic description for fault diagnosis, the fault point or fault range of the microwave oven can be accurately and quickly located, thereby improving the accuracy and efficiency of fault diagnosis; 2) the present invention can realize autonomous diagnosis of microwave oven electrical faults and give treatment suggestions, and professionals only need to perform targeted maintenance, thereby improving maintenance efficiency and reducing maintenance costs; 3) fault handling suggestions can be given through an intuitive user interaction interface, which can improve the user experience and enhance the user's trust and satisfaction with the product. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 The present invention is a flowchart of a method for diagnosing electrical faults in a microwave oven according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. The technical features of the embodiments of the present invention can be combined with each other without conflict.

[0029] The concept of knowledge graph was first proposed by Google. The project was initially intended to enhance the speed of search function and improve the retrieval ability of knowledge. Later, it was gradually extended to all knowledge bases similar to Google's knowledge graph architecture based on semantic network technology. Knowledge graph can convert data information into knowledge, and use natural language analysis technology from the knowledge level to identify, discover and infer computable models of complex relationships between things from data. Knowledge graph is used to integrate knowledge and data related to equipment failure to form a structured knowledge graph, so as to effectively diagnose microwave oven electrical failures more accurately and efficiently. To this end, the applicant proposes the following technical solutions:

[0030] Example 1

[0031] like Figure 1 As shown, a method for diagnosing electrical faults of a microwave oven comprises:

[0032] S1, obtaining microwave oven electrical fault data and performing data preprocessing;

[0033] S2. Based on the pre-processed electrical fault data, entity recognition, knowledge extraction and knowledge fusion processing are performed to establish an electrical fault knowledge graph;

[0034] S3. Perform system fault retrieval based on the electrical fault knowledge graph, and construct a Bayesian network based on the electrical fault query results obtained by the retrieval;

[0035] S4. Determine the prior probability of the fault cause and the conditional probability of each node in the Bayesian network, and use the Bayesian network to perform reasoning to obtain the fault cause that causes the fault phenomenon.

[0036] As an example of the present invention, the sources of microwave oven electrical fault data obtained in step S1 include microwave oven maintenance records, user feedback, technical manuals, expert experience, etc.; data preprocessing includes removing duplicate, redundant and invalid data, such as removing useless characters, separating Chinese and English, and converting Traditional Chinese to ensure the accuracy and consistency of the data, and at the same time performing named entity recognition and labeling on key information such as fault phenomena, faulty components, and fault causes, wherein the named entity recognition includes component entity and fault cause entity recognition to facilitate subsequent processing.

[0037] As an example of the present invention, step S2 includes:

[0038] S21: Using ontology modeling technology, establish an ontology set of microwave ovens, including: an entity set of microwave oven structures, an entity set of various electrical fault phenomena manifested during electrical faults, an entity set of electrical fault causes, and an entity set of maintenance suggestions;

[0039] Use natural language processing technology to identify entities in text, such as electrical fault phenomenon entities: no heating (U), uneven heating (I), loud noise (N), display screen failure (D), door cannot be closed (C1), door cannot be opened (C2), magnetron failure (M), high-voltage transformer failure (H), door switch failure (S), control circuit board failure (B), fan failure (F), turntable motor failure (T), etc. The structural entities of microwave ovens include: magnetron (M), high-voltage transformer (H), door switch (S), control circuit board (B), fan (F), turntable motor (T), etc. The entities of maintenance suggestions include: replace magnetron (RM), replace high-voltage transformer (RH), adjust door switch (RS), repair control circuit board (RB), clean fan (RF), replace turntable motor (RT), etc.

[0040] Extract the relationship between entities from the text, such as the connection relationship between components, the correspondence between fault phenomena and fault causes; the relationship between component entities and fault phenomenon entities, indicating that a component failure may lead to a specific fault phenomenon, such as a magnetron failure may lead to no heating, which can be expressed as M→U; the relationship between maintenance operation entities and fault phenomenon entities, indicating that a maintenance operation can solve a specific fault phenomenon. For example, replacing a magnetron can solve the problem of no heating, which can be expressed as RM→U. Only when the extracted knowledge is organized into standardized, consistent format and accurate information can it be added to the knowledge graph.

[0041] S22: Establish the ontology relationship of the microwave oven, and express the relationship between the concepts of the motor ontology, including: the cause of the electrical fault of the microwave oven, the sensor monitors the operating status of the microwave oven, the fault leads to the occurrence of other phenomena, the electrical fault phenomenon of the microwave oven occurs, and measures are taken to repair it;

[0042] S23: Based on the established knowledge ontology representation of microwave oven electrical fault diagnosis, information is extracted from the data and a microwave oven electrical fault triple relationship table is established, the relationships are marked, and the electrical fault knowledge is represented as a triple structure: entity-relationship-entity;

[0043] S24: Neo4j, a graph database, is used to store triplet knowledge data. The nodes in the graph database represent entities, corresponding to the entity relationships in the triplet. Directed edges represent the relationships between knowledge, corresponding to the relationships in the triplet, and a knowledge graph of microwave oven electrical faults is constructed.

[0044] Take entities as nodes of the graph, relationships as edges between nodes, add attributes (such as component model, failure rate, etc.) and corresponding values ​​to each node, and use a graph database (such as Neo4j) to store the constructed knowledge graph. For example, one node represents "magnetron", another node represents "no heating", and there is an edge between them representing "leading to relationship".

[0045] Step S2 also includes: S25, map optimization and verification;

[0046] Remove noise data, merge duplicate nodes and edges, and invite experts to review and verify the graph to ensure its accuracy and completeness. At the same time, you can use link prediction methods to verify the accuracy of the knowledge graph and remove known correct knowledge relationships. If the removed relationships can be predicted through multiple link predictions, then the constructed knowledge graph is relatively reliable.

[0047] Step S3 includes:

[0048] S31, after preprocessing the fault information, searching for fault nodes with the same fault phenomenon through the constructed microwave oven electrical fault knowledge graph, and searching for the structure and maintenance suggestions of the microwave ovens that have searched out these fault nodes;

[0049] S32, and establishing an electrical fault knowledge graph query subgraph according to the fault node, the structural entity of the microwave oven, the fault cause and the maintenance suggestion;

[0050] S33. Based on the fault knowledge graph query subgraph, the Bayesian network structure modeling is performed using the Bayesian network visualization method.

[0051] After the Bayesian network is built, it can be inferred by analyzing the joint probability distribution of variable nodes, the correlation between variable nodes, and the strength of correlation, that is, the probability of other variables being inferred by the observed values ​​of some variables. However, to use the Bayesian network for reasoning, the network structure of the Bayesian network must first be determined, and the knowledge graph is used to provide the required network structure for the Bayesian network. The Bayesian network is then used for reasoning to complete the microwave oven electrical fault diagnosis, which can solve the problem of requiring field experts to build Bayesian networks by searching for a large amount of information and based on experience.

[0052] Step S4 includes:

[0053] S41, selecting possible electrical fault causes according to the Bayesian network and constructing a cluster tree;

[0054] For example, for the failure of a microwave oven to heat (U) and make loud noise (N), the Bayesian network found that the possible causes are: magnetron failure (M) and fan failure (F); on this basis, the maximum cluster is determined, such as {U,M,I}, {N,F}, {D,B}, {C1,S}, {C2,S}, etc., and then the cluster tree structure is constructed: the nodes of the cluster tree are these maximum clusters, and the separation is determined based on the shared nodes. For example, the shared node of {U,M,I} and {N,F} is "U", so the separation is {U}.

[0055] S42, initialize message transmission and continuously update messages until convergence;

[0056] The initial probability distribution of a clique can be determined by multiplying the prior probabilities of its nodes. For example, for the clique {U,M,I}, the initial probability distribution is P(M)P(U|M)P(I|M). The initial probability distribution of a partition can be determined similarly; choose a starting clique and send messages to its neighboring cliques.

[0057] The message passing formula is: Among them, M i→j (X j ) is the message sent by group i to group j, X i and X j are the node sets of cluster i and cluster j respectively, and Neighbors(i) is the set of neighboring clusters of cluster i. For example, if cluster {U,M,I} sends a message to partition {U}, it can be calculated as: . Message delivery is considered to have converged when all messages no longer change.

[0058] S43. Calculate the posterior probability of each fault cause using the converged clique tree.

[0059] For a fault cause node, such as magnetron fault “M”, its posterior probability can be obtained by dividing the probability distribution of the cluster containing “M” by the probability distribution of the cluster not containing “M”.

[0060] Assuming that the known fault phenomena are no heating "U" and loud noise "N", the posterior probability of magnetron failure is calculated as:

[0061] Fault reasoning is the core link of fault diagnosis using Bayesian networks. In the fault reasoning process, the Bayesian reasoning algorithm is used to calculate the posterior probability of each fault cause based on the input observation data (i.e., fault phenomenon), so as to determine the most likely fault cause.

[0062] Step S4 also includes: recording the probability values ​​of various fault causes of the fault information and the probabilities of various solutions under the fault causes into the microwave oven electrical fault knowledge graph.

[0063] This operation can make the microwave oven electrical fault diagnosis method more accurate and reliable, and improve the effect of electrical fault diagnosis in the future. In addition, the diagnosis results are presented to the user in a clear and understandable manner, such as through a display screen, report or voice prompt; according to the diagnosis results, corresponding maintenance suggestions are provided. For example, if it is judged that the magnetron is faulty, the user can be advised to contact a professional maintenance person to perform a maintenance operation to replace the magnetron; in addition, asking the user whether he is satisfied with the diagnosis results and maintenance suggestions, and whether he has other questions or suggestions, can also help improve the diagnosis method and improve the quality of service.

[0064] Example 2

[0065] A method for diagnosing electrical faults of a microwave oven, comprising:

[0066] S1 obtains microwave oven electrical fault data and performs data preprocessing;

[0067] The sources for obtaining microwave oven electrical fault data include microwave oven maintenance records, user feedback, technical manuals, expert experience, etc., among which maintenance data include fault causes, maintenance methods, replaced parts, etc. The technical manual can be used to understand the structure, working principle, common faults and solutions of the microwave oven; data preprocessing includes removing duplicate, redundant and invalid data, such as removing useless characters, separating Chinese and English, and converting traditional Chinese to ensure the accuracy and consistency of the data, and at the same time performing named entity recognition and labeling of key information such as fault phenomena, faulty parts, and fault causes. The named entity recognition includes component entity and fault cause entity recognition to facilitate subsequent processing.

[0068] S2. Based on the pre-processed electrical fault data, entity recognition, knowledge extraction and knowledge fusion processing are performed respectively to establish an electrical fault knowledge graph, including:

[0069] S21: Using ontology modeling technology, establish an ontology set of microwave ovens, including: an entity set of microwave oven structures, an entity set of various electrical fault phenomena manifested during electrical faults, an entity set of electrical fault causes, and an entity set of maintenance suggestions;

[0070] S22: Establish the ontology relationship of the microwave oven, and express the relationship between the concepts of the motor ontology, including: the cause of the electrical fault of the microwave oven, the sensor monitors the operating status of the microwave oven, the fault leads to the occurrence of other phenomena, the electrical fault phenomenon of the microwave oven occurs, and measures are taken to repair it;

[0071] S23: Based on the established knowledge ontology representation of microwave oven electrical fault diagnosis, information is extracted from the data and a microwave oven electrical fault triple relationship table is established, the relationships are marked, and the electrical fault knowledge is represented as a triple structure: entity-relationship-entity;

[0072] S24: Neo4j, a graph database, is used to store triplet knowledge data. The nodes in the graph database represent entities, corresponding to the entity relationships in the triplet. Directed edges represent the relationships between knowledge, corresponding to the relationships in the triplet, and a knowledge graph of microwave oven electrical faults is constructed.

[0073] S25, map optimization and verification;

[0074] Remove noise data, merge duplicate nodes and edges, and invite experts to review and verify the graph to ensure its accuracy and completeness. At the same time, you can use link prediction methods to verify the accuracy of the knowledge graph and remove known correct knowledge relationships. If the removed relationships can be predicted through multiple link predictions, then the constructed knowledge graph is relatively reliable.

[0075] S3. Perform system fault retrieval based on the electrical fault knowledge graph, and construct a Bayesian network based on the electrical fault query results obtained by the retrieval;

[0076] For example, if a microwave oven fails to heat (U) or makes a lot of noise (N), the Bayesian network can be used to find that a magnetron failure (M) may cause the microwave oven to fail to heat and a fan failure (F) may cause a lot of noise. This provides a preliminary direction for electrical fault diagnosis.

[0077] S31, after preprocessing the fault information, searching for fault nodes with the same fault phenomenon through the constructed microwave oven electrical fault knowledge graph, and searching for the structure and maintenance suggestions of the microwave ovens that have searched out these fault nodes;

[0078] S32, and establishing an electrical fault knowledge graph query subgraph according to the fault node, the structural entity of the microwave oven, the fault cause and the maintenance suggestion;

[0079] S33. Based on the fault knowledge graph query subgraph, the Bayesian network structure modeling is performed using the Bayesian network visualization method. The nodes of the Bayesian network include no heating (U), loud noise (N), magnetron failure (M), fan failure (F), etc.

[0080] S4. Determine the prior probability of the fault cause and the conditional probability of each node in the Bayesian network, and use the Bayesian network to infer the fault cause that causes the fault phenomenon, including:

[0081] S41, selecting possible electrical fault causes according to the Bayesian network and constructing a cluster tree;

[0082] Determine the prior probability, assuming that the prior probability of magnetron failure is P(M) = 0.1, and the prior probability of fan failure is P(F) = 0.08. The conditional probability of no heating when the magnetron fails is P(U|M) = 0.8, and the conditional probability of loud noise when the fan fails is P(N|F) = 0.7.

[0083] S42, initialize message transmission and continuously update messages until convergence;

[0084] On this basis, the largest cluster is determined, such as {U, M, I}, {N, F} is the largest cluster, separated into {U} and {N}. For the cluster {U, M, I}, the initial probability distribution is P(M)P(U|M)P(I|M)=0.1×0.8×0.2 (assuming that the prior probability of uneven heating is P(I)=0.2). For the cluster {N, F}, the initial probability distribution is P(F)P(N|F)=0.08×0.7.

[0085] To pass messages, group {U, M, I} sends a message to partition {U}:

[0086]

[0087] Assuming that the message is m1 after calculation, group {N,F} sends a message to partition {N}:

[0088]

[0089] Assuming that the message is m2, partition {U} sends the message to group {U, M, I}: M {U}→{U,M,I} (M,U,I)=m2. Since partition {U} has only two adjacent groups, the message sent by another group is forwarded. Partition {N} sends a message to group {N,F}: M {N}→{N,F} (F,N)=m1, repeat the above process until the message converges.

[0090] S43. Calculate the posterior probability of each fault cause using the converged clique tree.

[0091] The posterior probability of magnetron failure is calculated as:

[0092] Calculation shows that P(M|U,N)=0.4.

[0093] The posterior probability of fan failure is calculated as:

[0094] Calculation shows that P(F|U,N)=0.3.

[0095] Combining the fault knowledge graph and the cluster tree propagation algorithm, it is determined that the microwave oven may have a magnetron fault and a fan fault. Since the posterior probability of a magnetron fault is relatively high, it is possible to prioritize checking whether the magnetron is faulty. If a magnetron fault is confirmed, the magnetron can be replaced for maintenance.

[0096] S44, recording the probability values ​​of the fault information corresponding to various fault causes and the probabilities of various solutions under the fault causes into the microwave oven electrical fault knowledge graph.

[0097] The microwave oven electrical fault diagnosis method described in the present invention realizes efficient and accurate diagnosis of microwave oven faults by integrating the advantages of the knowledge graph and the Bayesian network. The present invention can locate faulty components more quickly, improve diagnostic accuracy, adapt to complex fault conditions, and continuously optimize diagnostic capabilities as new fault data and experience are accumulated.

[0098] The present invention also provides a device for implementing a method for diagnosing electrical faults in a microwave oven, comprising a processor and a memory; wherein the processor is configured to execute the method for diagnosing electrical faults in a microwave oven; and the memory is used to store executable instructions of the processor.

[0099] The present invention also provides a computer storable medium, on which a computer program is stored, and the computer program is executed by a processor to implement the microwave oven electrical fault diagnosis method. The computer readable storage medium can be a readable storage medium or a readable signal medium, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk or an optical disk. It should be understood in the embodiments provided in this application that the disclosed device and method can be implemented in other ways. The device embodiment described above is only schematic, and the division of the unit is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms. The air conditioner and the computer storage medium have the same beneficial effects as the variable frequency air conditioner control method, which will not be repeated here.

[0100] Although the present invention is disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the scope defined by the claims.

Claims

1. A method for diagnosing electrical faults in a microwave oven, characterized in that: include: S1, obtaining microwave oven electrical fault data and performing data preprocessing; S2. Based on the pre-processed electrical fault data, entity recognition, knowledge extraction and knowledge fusion processing are performed to establish an electrical fault knowledge graph; S3. Perform system fault retrieval based on the electrical fault knowledge graph, and construct a Bayesian network based on the electrical fault query results obtained by the retrieval; S4. Determine the prior probability of the fault cause and the conditional probability of each node in the Bayesian network, and use the Bayesian network to perform reasoning to obtain the fault cause that causes the fault phenomenon.

2. The method for diagnosing electrical faults of microwave ovens according to claim 1, characterized in that: Step S2 includes: S21: Using ontology modeling technology, establish an ontology set of microwave ovens, including: an entity set of microwave oven structures, an entity set of various electrical fault phenomena manifested during electrical faults, an entity set of electrical fault causes, and an entity set of maintenance suggestions; S22: Establish the ontology relationship of the microwave oven, and express the relationship between the concepts of the motor ontology, including: the cause of the electrical fault of the microwave oven, the sensor monitors the operating status of the microwave oven, the fault leads to the occurrence of other phenomena, the electrical fault phenomenon of the microwave oven occurs, and measures are taken to repair it; S23: Based on the established knowledge ontology representation of microwave oven electrical fault diagnosis, information is extracted from the data and a microwave oven electrical fault triple relationship table is established, the relationships are marked, and the electrical fault knowledge is represented as a triple structure: entity-relationship-entity; S24: The graph database Neo4j is used to store triple knowledge data. The nodes in the graph database represent entities, which correspond to the entity relationships in the triples. The directed edges represent the relationships between knowledge, which correspond to the relationships in the triples. The knowledge graph of microwave oven electrical faults is constructed.

3. The microwave oven electrical fault diagnosis method according to claim 2, characterized in that: Step S2 also includes S25, verifying the accuracy of the knowledge graph using a link prediction method.

4. The method for diagnosing electrical faults of microwave ovens according to claim 1, characterized in that: Step S3 includes: S31, after preprocessing the fault information, searching for fault nodes with the same fault phenomenon through the constructed microwave oven electrical fault knowledge graph, and searching for the structure and maintenance suggestions of the microwave ovens that have searched out these fault nodes; S32, and establishing an electrical fault knowledge graph query subgraph according to the fault node, the structural entity of the microwave oven, the fault cause and the maintenance suggestion; S33. Based on the fault knowledge graph query subgraph, the Bayesian network structure modeling is performed using the Bayesian network visualization method.

5. The method for diagnosing electrical faults of microwave ovens according to claim 1, characterized in that: Step S4 includes: S41, selecting possible electrical fault causes according to the Bayesian network and constructing a cluster tree; S42, initialize message transmission and continuously update messages until convergence; S43. Calculate the posterior probability of each fault cause using the converged clique tree.

6. The method for diagnosing electrical faults of a microwave oven according to claim 5, characterized in that: The message transmission formula in step S42 is: i→j (X j )=∑P(X i |X j )ΠM k→i (X i ), where M i→j (X j ) is the message sent by group i to group j, X i and X j are the node sets of cluster i and cluster j respectively, and Neighbors(i) is the set of neighboring clusters of cluster i.

7. The method for diagnosing electrical faults of a microwave oven according to claim 5, characterized in that: Step S4 also includes: recording the probability values ​​of various fault causes of the fault information and the probabilities of various solutions under the fault causes into the microwave oven electrical fault knowledge graph.

8. A device for implementing a method for diagnosing electrical faults in a microwave oven, comprising a processor and a memory; wherein: The processor is configured to execute the microwave oven electrical fault diagnosis method according to any one of claims 1 to 7; and the memory is used to store executable instructions of the processor.

9. A computer storable medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the microwave oven electrical fault diagnosis method according to any one of claims 1 to 7.

Images