An augmented reality-based motor operation and maintenance guidance method and system

Abstract

The application provides a motor operation and maintenance guidance method and system based on augmented reality, the motor fault knowledge graph is constructed, the motor fault knowledge graph is used for diagnosing the motor fault reason, when the motor fault reason is multiple, a three-dimensional simulation model and a similarity algorithm double-layer verification mode are used for accurately positioning the fault reason, according to the fault reason, an augmented reality technology is used for displaying an operation and maintenance model, and user operation actions are monitored in real time, real-time guidance and interaction are conducted on user operation and maintenance operation, the application can also form a maintenance three-dimensional animation guidance video by comparing the on-site motor operation state based on the motor fault knowledge base, and the maintenance three-dimensional animation guidance video is pushed to an AR device for guidance maintenance, the application improves the accuracy of motor fault diagnosis, and improves the flexibility and efficiency of the augmented reality model for fault maintenance guidance.

Description

Technical fields:

[0001] This invention relates to the field of augmented reality technology, and in particular to an augmented reality-based method and system for guiding the operation and maintenance of electric motors. Background technology:

[0002] Electric motors have a wide variety of components and complex structures. When an electric motor malfunctions, the symptoms can be varied, and the causes are diverse. Furthermore, even the simplest disassembly, assembly, and repair of an electric motor require specialized personnel. Ordinary workers often lack knowledge of the principles and structure of electric motors. Moreover, training an experienced electric motor repair technician requires several to over a decade of training and practical experience.

[0003] Therefore, in the face of various motor faults, how to improve the accuracy of fault diagnosis and enhance the repair capabilities of non-professional maintenance personnel in motors, thereby improving the company's ability to troubleshoot motor equipment and reducing the company's motor operation and maintenance costs, is an urgent problem to be solved. Summary of the Invention:

[0004] To address the current problem of low accuracy in diagnosing motor faults and the inability of non-professionals to effectively repair motors, this paper proposes an augmented reality-based method for guiding motor operation and maintenance. This method includes the following steps:

[0005] S1. When a motor malfunctions, obtain the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph; the motor malfunction knowledge graph is constructed based on historical motor malfunction data;

[0006] S2. Based on the node with the highest matching degree obtained in step S1, the fault cause nodes are associated layer by layer in the motor fault knowledge graph. When there are multiple fault cause nodes, a three-dimensional simulation model is used to verify the multiple fault cause nodes. When there is only one fault cause node after verification, step S4 is executed; when there are multiple fault cause nodes after verification, step S3 is executed.

[0007] S3. Use the similarity method to perform secondary verification on the multiple fault cause nodes after the first verification in step S2, obtain the fault cause nodes after secondary verification, and then execute step S4.

[0008] S4. Based on the fault cause nodes, use augmented reality technology to guide the operation and maintenance of motors.

[0009] The motor fault knowledge graph in step S1 is constructed based on historical motor fault data, and specifically includes the following steps:

[0010] S11. Obtain historical fault data of the electric motor, which includes: fault data of multiple components and correlation data of multiple component faults;

[0011] Each component's fault data is an array, which includes the following data: component name, fault type, fault description, and fault code;

[0012] The fault description includes: textual description data of component faults and video data of component faults;

[0013] Each component fault association data is an array, which includes the following data: faulty component name, cause component name, cause component fault code, faulty component fault code, and fault cause description;

[0014] S12. Using each component's fault data as a node in the motor fault knowledge graph, establish the association path between each component's fault data based on the component fault association data to form the motor fault knowledge graph.

[0015] The associated path is a directed path, with the path direction pointing from the node corresponding to the faulty component name and fault code in the component fault association data to the node corresponding to the cause component name and fault code.

[0016] Step S1, when a motor malfunctions, obtains the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph, specifically including the following steps:

[0017] S13. When a motor malfunctions, the user inputs motor malfunction description data;

[0018] The motor fault description data includes: textual description data of motor faults and video data of motor faults;

[0019] S14. Obtain multiple nodes in the fault knowledge graph whose component name is motor, and extract the component fault text description data and component fault video data corresponding to each of the multiple nodes;

[0020] S15. The text description data and video data of motor fault input by the user in step S13 are compared with the text description data and video data of component fault corresponding to each node in step S14. The node with the highest similarity is obtained and the node with the highest similarity is taken as the node with the highest matching degree in step S1.

[0021] In step S15, the text description data and video data of the motor fault input by the user in step S13 are compared with the text description data and video data of the component fault corresponding to each node in step S14 to calculate the similarity and obtain the node with the highest similarity. Specifically, this includes:

[0022] S151. The similarity calculation is performed between the user-input text description data and video data of motor faults and the text description data and video data of component faults corresponding to the nodes in step S14. Specifically:

[0023] r=δ1Z+δ2S

[0024] Where r is the similarity value, δ1 and δ2 are the first weight value and the second weight value, respectively, Z is the similarity value between the fault text description data, and S is the similarity value between the fault video data.

[0025] S152. Repeat step S151 to obtain multiple similarity values ​​between the user-inputted text description data of motor faults, video data of motor faults and the text description data of component faults and video data of component faults corresponding to each node in step S14.

[0026] S153. Sort the multiple similarity values ​​obtained in step S152, and feed back the component fault text description data and component fault video data corresponding to the top 3 nodes with the highest similarity values ​​to the user.

[0027] S154. The user selects one node from the three nodes as the node with the highest similarity obtained in step S15.

[0028] Step S2, based on the node with the highest matching degree obtained in step S1, progressively associates fault cause nodes in the motor fault knowledge graph. When multiple fault cause nodes are obtained, a three-dimensional simulation model is used to verify the multiple fault cause nodes. Specifically, this includes the following steps:

[0029] S21. Obtain the node pointed to by the node with the highest matching degree obtained in step S1, and take the pointed-to node as the fault cause node;

[0030] S22. When there is only one fault cause node, obtain the new fault cause node pointed to by the fault cause node;

[0031] S23. Repeat step S22 to obtain new fault cause nodes layer by layer until there are multiple new fault cause nodes, then execute step S25; when there is only one new fault cause node and the new fault cause node does not point to any node, take the new fault cause node as the fault cause node of the motor and execute step S4.

[0032] S24. If there are multiple fault cause nodes obtained in step S21, proceed to step S25;

[0033] S25. Obtain the component name, fault type, and fault code corresponding to each of the multiple new fault cause nodes obtained in step S23 or the multiple fault cause nodes obtained in step S24.

[0034] S26. Input the component name, fault type, and fault code corresponding to each node obtained in step S25 into the three-dimensional simulation model, run the three-dimensional simulation model, and obtain multiple simulation results of the three-dimensional simulation model;

[0035] Each time, only the component name, fault type, and fault code corresponding to a node are input to obtain the simulation results corresponding to that node.

[0036] The simulation results are fault video data and fault codes of the electric motor;

[0037] S27. Compare the motor fault code corresponding to the node with the highest matching degree obtained in step S1 with the fault code in the simulation run result obtained in step S26. If they are inconsistent, remove the node corresponding to the simulation run result. If they are consistent, the user compares whether the motor fault video data in the simulation run result is consistent with the motor fault video data obtained in step S13. If they are inconsistent, remove the node corresponding to the simulation run result.

[0038] S28. Repeat step S27 until all the multiple fault cause nodes obtained in step S25 have been processed, and the fault cause nodes after one verification are obtained.

[0039] Step S3 uses a similarity method to perform secondary verification on the multiple fault cause nodes that have undergone the first verification in step S2, and obtains the fault cause nodes after secondary verification. Specifically, it includes the following steps:

[0040] S31. Obtain the names of multiple components corresponding to the multiple fault cause nodes after one verification in step S2;

[0041] S32. Obtain the names of multiple components from multiple sample data in the fault experience database;

[0042] The fault experience database stores sample data of components of the motor that have failed.

[0043] The sample data of the faulty components of the motor includes: component name, usage time, and operating environment data;

[0044] The operating environment data includes: motor input voltage, ambient temperature, ambient humidity, and air dust data;

[0045] S33. Check if the component name obtained in step S31 exists in the sample data stored in the fault experience database. If it does not exist, remove the node corresponding to the component name from the multiple fault cause nodes after one verification in step S2.

[0046] S34. Repeat step S33 until all component names obtained in step S31 have been processed, and multiple fault cause nodes are retained after removal.

[0047] S35. Obtain the usage duration of the components in the component names corresponding to the multiple fault cause nodes retained after the removal process obtained in step S34, and obtain the current operating environment data of the motor.

[0048] The current operating environment data of the electric motor includes: motor input voltage, ambient temperature, ambient humidity, and air dust data;

[0049] S36. Based on the current operating environment data of the motor, the usage time of the components in the component names corresponding to the multiple fault cause nodes retained after removal processing in step S35, and the sample data in the experience database, perform secondary verification on the multiple fault cause nodes retained after removal processing obtained in step S34 to obtain the fault cause nodes after secondary verification.

[0050] Step S36 specifically includes the following steps:

[0051] S361. Obtain the component name of one of the multiple fault cause nodes retained after the removal process in step S35, and search for one or more sample data corresponding to the component name in the fault experience data. Calculate the similarity between the usage time of the component corresponding to the component name, the current environmental data of the motor, and the one or more sample data, to obtain one or more similarity values. The calculation method for the similarity values ​​is as follows:

[0052] R = ω1t + ω2v + ω3c + ω4h

[0053] Where R is the similarity value, ω1, ω2, ω3, and ω4 are the first weight value, the second weight value, the third weight value, and the fourth weight value, respectively, and t, v, c, and h are the usage time, input voltage, ambient temperature, and ambient humidity, respectively.

[0054] Among them, the weight values ​​of ω1, ω2, ω3, and ω4 corresponding to each component name are different;

[0055] The weight values ​​of ω1, ω2, ω3, and ω4 corresponding to each component name are pre-stored in the memory;

[0056] S362. If one or more similarity values ​​obtained in step S361 are all less than the preset threshold T, then discard the node obtained in step S361.

[0057] S363. Repeat steps S361 to S362 until all the fault cause nodes retained after the removal process in step S35 have been processed. The fault cause nodes retained after the removal process in steps S361 to S362 are used as the fault cause nodes after secondary verification.

[0058] Step S4, based on the fault cause node, uses augmented reality technology to guide the operation and maintenance of the electric motor, specifically including the following steps:

[0059] S41. When there is only one fault cause node, extract the disassembly augmented reality model, the assembly augmented reality model, and the maintenance augmented reality model corresponding to the fault code of the fault cause node, and combine them in the order of disassembly augmented reality model, maintenance augmented reality model, and assembly augmented reality model, and use augmented reality technology to provide users with motor operation and maintenance guidance.

[0060] S42. When there are multiple fault cause nodes, augmented reality technology is used to provide users with motor operation and maintenance guidance based on the hierarchical relationship of multiple fault cause nodes in the component BOM structure of the motor.

[0061] When there are multiple fault cause nodes, step S42 uses augmented reality technology to provide users with motor operation and maintenance guidance based on the hierarchical relationship of these nodes in the motor's component BOM structure. This includes the following steps:

[0062] S421. Obtain the hierarchy and inclusion relationship of the multiple fault cause nodes in the component BOM structure of the motor;

[0063] S422. Traverse the multiple fault cause nodes. When two or more fault cause nodes are in an inclusive relationship, form a dismantling path for the two or more fault cause nodes according to the hierarchy of each node, and finally form multiple dismantling paths.

[0064] S423. For each disassembly path, first obtain the disassembly augmented reality model of the fault cause node with the lowest level in the disassembly path, and guide the user to perform the disassembly operation;

[0065] S424. When the user confirms that the current fault cause node is faulty, the maintenance augmented reality model and the assembly augmented reality model corresponding to the fault code of the fault cause node are combined to provide the user with maintenance guidance and post-maintenance assembly, and the operation and maintenance guidance process ends.

[0066] When the user confirms that the current fault cause node does not have a fault and that the current fault cause node has a subordinate fault cause node, the user directly obtains the disassembly augmented reality model of the subordinate fault cause node, guides the user to perform the disassembly operation, and repeats step S424.

[0067] If the user confirms that the current fault cause node does not have a fault and that the fault cause node does not have a subordinate fault cause node, obtain the assembled augmented reality model of the fault cause node and its superior fault cause node, and guide the user to assemble it.

[0068] S425. For the multiple disassembly paths obtained in step S422, repeat steps S423 to S424 until the faulty node is found, and provide the user with repair guidance and post-repair assembly, thus ending the operation and maintenance guidance process.

[0069] Furthermore, steps S41 and S42 employ augmented reality technology to provide users with motor operation and maintenance guidance, specifically including the following steps:

[0070] S4a. The user wears VR glasses, and an augmented reality model of motor operation and maintenance is displayed in the VR glasses. The augmented reality model of motor operation and maintenance is superimposed on the location of the component to be operated according to the operation and maintenance steps of the motor.

[0071] S4b. According to the operation and maintenance steps in the electric motor operation and maintenance augmented reality model, the user operates the electric motor components; the VR glasses monitor the user's hand movements in real time. When the similarity between the hand movements and the virtual hand movements in the electric motor operation and maintenance augmented reality model is lower than a preset threshold X, the virtual hand is displayed in red in the electric motor operation and maintenance augmented reality model displayed in the VR glasses.

[0072] The preset threshold X is 80%;

[0073] The augmented reality model for motor operation and maintenance displays textual descriptions of precautions for the virtual hand gesture, the fault location of the component being operated by the current hand gesture, and a textual description of the fault next to the virtual hand gesture.

[0074] S4c. When the virtual hand is red in the augmented reality model of motor maintenance displayed in the VR glasses, the user can adjust the hand gesture, the display progress of the augmented reality model of motor maintenance, or the display mode of the augmented reality model of motor maintenance.

[0075] The display progress of the adjusted electric motor operation and maintenance augmented reality model includes: rewind, pause, fast forward, and stop;

[0076] The display methods for adjusting the augmented reality model of motor operation and maintenance include: local zoom-in and local zoom-out.

[0077] An augmented reality-based electric motor operation and maintenance guidance system, characterized in that the system performs the following steps:

[0078] S1. When a motor malfunctions, obtain the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph; the motor malfunction knowledge graph is constructed based on historical motor malfunction data;

[0079] S2. Based on the node with the highest matching degree obtained in step S1, the fault cause nodes are associated layer by layer in the motor fault knowledge graph. When there are multiple fault cause nodes, a three-dimensional simulation model is used to verify the multiple fault cause nodes. When there is only one fault cause node after verification, step S4 is executed; when there are multiple fault cause nodes after verification, step S3 is executed.

[0080] S3. Use the similarity method to perform secondary verification on the multiple fault cause nodes after the first verification in step S2, obtain the fault cause nodes after secondary verification, and then execute step S4.

[0081] S4. Based on the fault cause nodes, use augmented reality technology to guide the operation and maintenance of motors.

[0082] The beneficial effects of this invention are as follows:

[0083] 1. This invention constructs a knowledge graph of motor faults using component fault data and component fault association data; it utilizes real data of historical motor faults, uses component fault data as graph nodes, and uses component fault association data as the pointing paths between nodes, which differs from the conventional triplet construction method, providing accurate data support for motor fault diagnosis; by constructing the fault knowledge graph, the accuracy of motor fault diagnosis can be effectively improved.

[0084] 2. In this invention, two types of data, text description and fault video, are used to locate the motor fault input by the user in the knowledge graph, which improves the accuracy of motor fault diagnosis;

[0085] 3. This invention uses knowledge graph-based fault location to diagnose the causes of motor faults, and employs a two-layer verification method using a three-dimensional simulation model and a similarity algorithm to accurately locate the causes of motor faults, thereby improving the accuracy of motor fault diagnosis.

[0086] 4. The similarity algorithm used in the secondary verification of this invention incorporates the usage time of motor components, input voltage, ambient temperature, and ambient humidity, that is, it incorporates environmental factors of the motor to accurately locate the possible causes of motor failure, thereby improving the overall accuracy of the system in fault diagnosis.

[0087] 5. In this invention, when there are multiple fault cause nodes, the guidance order of the augmented reality model is set by the optimal path method, and the investigation of other fault cause nodes is terminated in time when the fault is located, thereby improving the overall operating efficiency of the system.

[0088] 6. In this invention, the disassembly, repair, and assembly of components are divided into three augmented reality models, and the above three augmented reality models are dynamically combined according to the optimal path, which improves the flexibility of augmented reality models in guiding repair and improves the overall operating efficiency of the system.

[0089] 7. This invention uses augmented reality technology to display an augmented reality model of motor operation and maintenance in VR glasses to guide the user's hand movements during operation and maintenance. At the same time, the use of augmented reality technology can dynamically monitor the accuracy of the user's hand movements and support the interaction between the user and the augmented reality model of motor operation and maintenance, so that the user can quickly understand the fault point, standardize operation and maintenance actions, and enable users without maintenance experience to quickly repair the equipment, thereby improving the efficiency and accuracy of motor maintenance and reducing the requirements for motor maintenance personnel.

[0090] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above description and other objects, features and advantages of the present invention more obvious and understandable, preferred embodiments are provided and described in detail below. Attached Figure Description

[0091] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0092] Figure 1 This is a flowchart of an augmented reality-based method for guiding the operation and maintenance of electric motors. Detailed Implementation

[0093] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0094] In the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," "fixing," etc., should be interpreted broadly. For example, they can refer to a connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0095] Example 1

[0096] This paper proposes an augmented reality-based method for guiding the operation and maintenance of electric motors. The method includes the following steps:

[0097] S1. When a motor malfunctions, obtain the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph; the motor malfunction knowledge graph is constructed based on historical motor malfunction data;

[0098] The motor fault knowledge graph in step S1 is constructed based on historical motor fault data, and specifically includes the following steps:

[0099] S11. Obtain historical fault data of the electric motor, which includes: fault data of multiple components and correlation data of multiple component faults;

[0100] Each component's fault data is an array, which includes the following data: component name, fault type, fault description, and fault code;

[0101] Among them, the motor as a whole can be regarded as a component. For example, the component fault data of the motor is: motor, cannot start, the indicator light does not light up after the motor is powered on, and there is no response overall, 0012;

[0102] The fault description includes: textual description data of component faults and video data of component faults;

[0103] Each component fault association data is an array, which includes the following data: faulty component name, cause component name, cause component fault code, faulty component fault code, and fault cause description;

[0104] S12. Using each component's fault data as a node in the motor fault knowledge graph, establish the association path between each component's fault data based on the component fault association data to form the motor fault knowledge graph.

[0105] The associated path is a directed path, with the path direction pointing from the node corresponding to the faulty component name and fault code in the component fault association data to the node corresponding to the cause component name and fault code.

[0106] Step S1, when a motor malfunctions, obtains the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph, specifically including the following steps:

[0107] S13. When a motor malfunctions, the user inputs motor malfunction description data;

[0108] The motor fault description data includes: textual description data of motor faults and video data of motor faults;

[0109] S14. Obtain multiple nodes in the fault knowledge graph whose component name is motor, and extract the component fault text description data and component fault video data corresponding to each of the multiple nodes;

[0110] S15. The text description data and video data of motor fault input by the user in step S13 are compared with the text description data and video data of component fault corresponding to each node in step S14. The node with the highest similarity is obtained and the node with the highest similarity is taken as the node with the highest matching degree in step S1.

[0111] Step S15 specifically includes:

[0112] S151. The similarity calculation is performed between the user-input text description data and video data of motor faults and the text description data and video data of component faults corresponding to the nodes in step S14. Specifically:

[0113] r=δ1Z+δ2S

[0114] Where r is the similarity value, δ1 and δ2 are the first weight value and the second weight value, respectively, Z is the similarity value between the fault text description data, and S is the similarity value between the fault video data.

[0115] S152. Repeat step S151 to obtain multiple similarity values ​​between the user-inputted text description data of motor faults, video data of motor faults and the text description data of component faults and video data of component faults corresponding to each node in step S14.

[0116] S153. Sort the multiple similarity values ​​obtained in step S152, and feed back the component fault text description data and component fault video data corresponding to the top 3 nodes with the highest similarity values ​​to the user.

[0117] S154. The user selects one node from the three nodes as the node with the highest similarity obtained in step S15.

[0118] S2. Based on the node with the highest matching degree obtained in step S1, the fault cause nodes are associated layer by layer in the motor fault knowledge graph. When there are multiple fault cause nodes, a three-dimensional simulation model is used to verify the multiple fault cause nodes. When there is only one fault cause node after verification, step S4 is executed; when there are multiple fault cause nodes after verification, step S3 is executed.

[0119] Step S2, based on the node with the highest matching degree obtained in step S1, progressively associates fault cause nodes in the motor fault knowledge graph. When multiple fault cause nodes are obtained, a three-dimensional simulation model is used to verify the multiple fault cause nodes. Specifically, this includes the following steps:

[0120] S21. Obtain the node pointed to by the node with the highest matching degree obtained in step S1, and take the pointed-to node as the fault cause node;

[0121] S22. When there is only one fault cause node, obtain the new fault cause node pointed to by the fault cause node;

[0122] S23. Repeat step S22 to obtain new fault cause nodes layer by layer until there are multiple new fault cause nodes, then execute step S25; when there is only one new fault cause node and the new fault cause node does not point to any node, take the new fault cause node as the fault cause node of the motor and execute step S4.

[0123] S24. If there are multiple fault cause nodes obtained in step S21, proceed to step S25;

[0124] S25. Obtain the component name, fault type, and fault code corresponding to each of the multiple new fault cause nodes obtained in step S23 or the multiple fault cause nodes obtained in step S24.

[0125] S26. Input the component name, fault type, and fault code corresponding to each node obtained in step S25 into the three-dimensional simulation model, run the three-dimensional simulation model, and obtain multiple simulation results of the three-dimensional simulation model;

[0126] Each time, only the component name, fault type, and fault code corresponding to a node are input to obtain the simulation results corresponding to that node.

[0127] The simulation results are fault video data and fault codes of the electric motor;

[0128] S27. Compare the motor fault code corresponding to the node with the highest matching degree obtained in step S1 with the fault code in the simulation run result obtained in step S26. If they are inconsistent, remove the node corresponding to the simulation run result. If they are consistent, the user compares whether the motor fault video data in the simulation run result is consistent with the motor fault video data obtained in step S13. If they are inconsistent, remove the node corresponding to the simulation run result.

[0129] S28. Repeat step S27 until all the multiple fault cause nodes obtained in step S25 have been processed, and the fault cause nodes after one verification are obtained.

[0130] S3. Use the similarity method to perform secondary verification on the multiple fault cause nodes after the first verification in step S2, obtain the fault cause nodes after secondary verification, and then execute step S4.

[0131] Step S3 uses a similarity method to perform secondary verification on the multiple fault cause nodes that have undergone the first verification in step S2, and obtains the fault cause nodes after secondary verification. Specifically, it includes the following steps:

[0132] S31. Obtain the names of multiple components corresponding to the multiple fault cause nodes after one verification in step S2;

[0133] S32. Obtain the names of multiple components from multiple sample data in the fault experience database;

[0134] The fault experience database stores sample data of components of the motor that have failed.

[0135] The sample data of the faulty components of the motor includes: component name, usage time, and operating environment data;

[0136] The operating environment data includes: motor input voltage, ambient temperature, ambient humidity, and air dust data;

[0137] S33. Check if the component name obtained in step S31 exists in the sample data stored in the fault experience database. If it does not exist, remove the node corresponding to the component name from the multiple fault cause nodes after one verification in step S2.

[0138] S34. Repeat step S33 until all component names obtained in step S31 have been processed, and multiple fault cause nodes are retained after removal.

[0139] S35. Obtain the usage duration of the components in the component names corresponding to the multiple fault cause nodes retained after the removal process obtained in step S34, and obtain the current operating environment data of the motor.

[0140] The current operating environment data of the electric motor includes: motor input voltage, ambient temperature, ambient humidity, and air dust data;

[0141] S36. Based on the current operating environment data of the motor, the usage time of the components in the component names corresponding to the multiple fault cause nodes retained after removal processing in step S35, and the sample data in the experience database, perform secondary verification on the multiple fault cause nodes retained after removal processing obtained in step S34 to obtain the fault cause nodes after secondary verification.

[0142] Step S36 specifically includes the following steps:

[0143] S361. Obtain the component name of one of the multiple fault cause nodes retained after the removal process in step S35, and search for one or more sample data corresponding to the component name in the fault experience data. Calculate the similarity between the usage time of the component corresponding to the component name, the current environmental data of the motor, and the one or more sample data, to obtain one or more similarity values. The calculation method for the similarity values ​​is as follows:

[0144] R = ω1t + ω2v + ω3c + ω4h

[0145] Where R is the similarity value, ω1, ω2, ω3, and ω4 are the first weight value, the second weight value, the third weight value, and the fourth weight value, respectively, and t, v, c, and h are the usage time, input voltage, ambient temperature, and ambient humidity, respectively.

[0146] Among them, the weight values ​​of ω1, ω2, ω3, and ω4 corresponding to each component name are different;

[0147] For example, the weight values ​​corresponding to component A are: ω 1A ω 2A ω 3A ω4A The weight value corresponding to component B is: ω 1B ω 2B ω 3B ω 4B However, because the environmental factors affecting each component are different, the two sets of weight values ​​are different, that is, ω. 1A ω 2A ω 3A ω 4A With ω 1B ω 2B ω 3B ω 4B Set different values;

[0148] The weight values ​​of ω1, ω2, ω3, and ω4 corresponding to each component name are pre-stored in the memory;

[0149] S362. If one or more similarity values ​​obtained in step S361 are all less than the preset threshold T, then discard the node obtained in step S361.

[0150] The threshold T can be set flexibly according to actual conditions;

[0151] S363. Repeat steps S361 to S362 until all the fault cause nodes retained after the removal process in step S35 have been processed. The fault cause nodes retained after the removal process in steps S361 to S362 are used as the fault cause nodes after secondary verification.

[0152] S4. Based on the fault cause nodes, use augmented reality technology to guide the operation and maintenance of motors.

[0153] Step S4, based on the fault cause node, uses augmented reality technology to guide the operation and maintenance of the electric motor, specifically including the following steps:

[0154] S41. When there is only one fault cause node, extract the disassembly augmented reality model, the assembly augmented reality model, and the maintenance augmented reality model corresponding to the fault code of the fault cause node, and combine them in the order of disassembly augmented reality model, maintenance augmented reality model, and assembly augmented reality model, and use augmented reality technology to provide users with motor operation and maintenance guidance.

[0155] S42. When there are multiple fault cause nodes, augmented reality technology is used to provide users with motor operation and maintenance guidance based on the hierarchical relationship of multiple fault cause nodes in the component BOM structure of the motor.

[0156] Step S42 specifically includes the following steps:

[0157] S421. Obtain the hierarchy and inclusion relationship of the multiple fault cause nodes in the component BOM structure of the motor;

[0158] S422. Traverse the multiple fault cause nodes. When two or more fault cause nodes are in an inclusive relationship, form a dismantling path for the two or more fault cause nodes according to the hierarchy of each node, and finally form multiple dismantling paths.

[0159] S423. For each disassembly path, first obtain the disassembly augmented reality model of the fault cause node with the lowest level in the disassembly path, and guide the user to perform the disassembly operation;

[0160] The motor has a minimum level of 0, and according to the disassembly sequence of the motor, the lower the level of the component, the higher the level.

[0161] S424. When the user confirms that the current fault cause node is faulty, the maintenance augmented reality model and the assembly augmented reality model corresponding to the fault code of the fault cause node are combined to provide the user with maintenance guidance and post-maintenance assembly, and the operation and maintenance guidance process ends.

[0162] When the user confirms that the current fault cause node does not have a fault and that the current fault cause node has a subordinate fault cause node, the user directly obtains the disassembly augmented reality model of the subordinate fault cause node, guides the user to perform the disassembly operation, and repeats step S424.

[0163] The lower-level fault cause node refers to the node in the disassembly path that is at a higher level than the current fault cause node and has the smallest level difference.

[0164] If the user confirms that the current fault cause node does not have a fault and that the fault cause node does not have a subordinate fault cause node, obtain the assembled augmented reality model of the fault cause node and its superior fault cause node, and guide the user to assemble it.

[0165] S425. For the multiple disassembly paths obtained in step S422, repeat steps S423 to S424 until the faulty node is found, and provide the user with repair guidance and post-repair assembly, thus ending the operation and maintenance guidance process.

[0166] Furthermore, steps S41 and S42 employ augmented reality technology to provide users with motor operation and maintenance guidance, specifically including the following steps:

[0167] S4a. The user wears VR glasses, and an augmented reality model of motor operation and maintenance is displayed in the VR glasses. The augmented reality model of motor operation and maintenance is superimposed on the location of the component to be operated according to the operation and maintenance steps of the motor.

[0168] S4b. According to the operation and maintenance steps in the electric motor operation and maintenance augmented reality model, the user operates the electric motor components; the VR glasses monitor the user's hand movements in real time. When the similarity between the hand movements and the virtual hand movements in the electric motor operation and maintenance augmented reality model is lower than a preset threshold X, the virtual hand is displayed in red in the electric motor operation and maintenance augmented reality model displayed in the VR glasses.

[0169] The preset threshold X is 80%;

[0170] S4c. When the virtual hand is red in the augmented reality model of motor maintenance displayed in the VR glasses, the user can adjust the hand gesture, the display progress of the augmented reality model of motor maintenance, or the display mode of the augmented reality model of motor maintenance.

[0171] The display progress of the adjusted electric motor operation and maintenance augmented reality model includes: rewind, pause, fast forward, and stop;

[0172] The display methods for adjusting the augmented reality model of motor operation and maintenance include: local zoom-in and local zoom-out.

[0173] Example 2

[0174] This invention also proposes an augmented reality-based electric motor operation and maintenance guidance system, characterized in that the system performs the following steps:

[0175] S1. When a motor malfunctions, obtain the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph; the motor malfunction knowledge graph is constructed based on historical motor malfunction data;

[0176] S2. Based on the node with the highest matching degree obtained in step S1, the fault cause nodes are associated layer by layer in the motor fault knowledge graph. When there are multiple fault cause nodes, a three-dimensional simulation model is used to verify the multiple fault cause nodes. When there is only one fault cause node after verification, step S4 is executed; when there are multiple fault cause nodes after verification, step S3 is executed.

[0177] S3. Use the similarity method to perform secondary verification on the multiple fault cause nodes after the first verification in step S2, obtain the fault cause nodes after secondary verification, and then execute step S4.

[0178] S4. Based on the fault cause nodes, use augmented reality technology to guide the operation and maintenance of motors.

[0179] Optionally, based on the above implementation method, a 3D animation guidance video for maintenance can be generated by comparing the on-site motor operating status with a motor fault knowledge base and then pushed to an AR device for guidance and maintenance.

[0180] The beneficial effects of this invention are as follows:

[0181] 1. This invention constructs a knowledge graph of motor faults using component fault data and component fault association data; it utilizes real data of historical motor faults, uses component fault data as graph nodes, and uses component fault association data as the pointing paths between nodes, which differs from the conventional triplet construction method, providing accurate data support for motor fault diagnosis; by constructing the fault knowledge graph, the accuracy of motor fault diagnosis can be effectively improved.

[0182] 2. In this invention, two types of data, text description and fault video, are used to locate the motor fault input by the user in the knowledge graph, which improves the accuracy of motor fault diagnosis;

[0183] 3. This invention uses knowledge graph-based fault location to diagnose the causes of motor faults, and employs a two-layer verification method using a three-dimensional simulation model and a similarity algorithm to accurately locate the causes of motor faults, thereby improving the accuracy of motor fault diagnosis.

[0184] 4. The similarity algorithm used in the secondary verification of this invention incorporates the usage time of motor components, input voltage, ambient temperature, and ambient humidity, that is, it incorporates environmental factors of the motor to accurately locate the possible causes of motor failure, thereby improving the overall accuracy of the system in fault diagnosis.

[0185] 5. In this invention, when there are multiple fault cause nodes, the guidance order of the augmented reality model is set by the optimal path method, and the investigation of other fault cause nodes is terminated in time when the fault is located, thereby improving the overall operating efficiency of the system.

[0186] 6. In this invention, the disassembly, repair, and assembly of components are divided into three augmented reality models, and the above three augmented reality models are dynamically combined according to the optimal path, which improves the flexibility of augmented reality models in guiding repair and improves the overall operating efficiency of the system.

[0187] 7. This invention uses augmented reality technology to display an augmented reality model of motor operation and maintenance in VR glasses to guide the user's hand movements during operation and maintenance. At the same time, the use of augmented reality technology can dynamically monitor the accuracy of the user's hand movements and support the interaction between the user and the augmented reality model of motor operation and maintenance, so that the user can quickly understand the fault point, standardize operation and maintenance actions, and enable users without maintenance experience to quickly repair the equipment, thereby improving the efficiency and accuracy of motor maintenance and reducing the requirements for motor maintenance personnel.

[0188] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for guiding the operation and maintenance of electric motors based on augmented reality, characterized in that, Includes the following steps: S1. When a motor malfunctions, obtain the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph; the motor malfunction knowledge graph is constructed based on historical motor malfunction data; S2. Based on the node with the highest matching degree obtained in step S1, the fault cause nodes are associated layer by layer in the motor fault knowledge graph. When there are multiple fault cause nodes, a three-dimensional simulation model is used to verify the multiple fault cause nodes. When there is only one fault cause node after verification, step S4 is executed; when there are multiple fault cause nodes after verification, step S3 is executed. S3. Use the similarity method to perform secondary verification on the multiple fault cause nodes after the first verification in step S2, obtain the fault cause nodes after secondary verification, and then execute step S4. S4. Based on the fault cause nodes, augmented reality technology is used to guide the operation and maintenance of motors; Step S4 specifically includes the following steps: S41. When there is only one fault cause node, extract the disassembly augmented reality model, the assembly augmented reality model, and the maintenance augmented reality model corresponding to the fault code of the fault cause node, and combine them in the order of disassembly augmented reality model, maintenance augmented reality model, and assembly augmented reality model, and use augmented reality technology to provide users with motor operation and maintenance guidance. S42. When there are multiple fault cause nodes, augmented reality technology is used to provide users with motor operation and maintenance guidance based on the hierarchical relationship of multiple fault cause nodes in the component BOM structure of the motor. Step S42 specifically includes the following steps: S421. Obtain the hierarchy and inclusion relationship of the multiple fault cause nodes in the component BOM structure of the motor; S422. Traverse the multiple fault cause nodes. When two or more fault cause nodes are in an inclusive relationship, form a dismantling path for the two or more fault cause nodes according to the hierarchy of each node, and finally form multiple dismantling paths. S423. For each disassembly path, first obtain the disassembly augmented reality model of the fault cause node with the lowest level in the disassembly path, and guide the user to perform the disassembly operation; The motor has a minimum level of 0, and according to the disassembly sequence of the motor, the lower the level of the component, the higher the level. S424. When the user confirms that the current fault cause node is faulty, the maintenance augmented reality model and the assembly augmented reality model corresponding to the fault code of the fault cause node are combined to provide the user with maintenance guidance and post-maintenance assembly, and the operation and maintenance guidance process ends. When the user confirms that the current fault cause node does not have a fault and that the current fault cause node has a subordinate fault cause node, the user directly obtains the disassembly augmented reality model of the subordinate fault cause node, guides the user to perform the disassembly operation, and repeats step S424. The lower-level fault cause node refers to the node in the disassembly path that is at a higher level than the current fault cause node and has the smallest level difference. If the user confirms that the current fault cause node does not have a fault and that the fault cause node does not have a subordinate fault cause node, obtain the assembled augmented reality model of the fault cause node and its superior fault cause node, and guide the user to assemble it. S425. For the multiple disassembly paths obtained in step S422, repeat steps S423 to S424 until the faulty node is found, and provide the user with repair guidance and post-repair assembly, thus ending the operation and maintenance guidance process.

2. The augmented reality-based electric motor operation and maintenance guidance method according to claim 1, characterized in that, The motor fault knowledge graph in step S1 is constructed based on historical motor fault data, and specifically includes the following steps: S11. Obtain historical fault data of the electric motor, which includes: fault data of multiple components and correlation data of multiple component faults; Each component's fault data is an array, which includes the following data: component name, fault type, fault description, and fault code; The fault description includes: textual description data of component faults and video data of component faults; Each component fault association data is an array, which includes the following data: faulty component name, cause component name, cause component fault code, faulty component fault code, and fault cause description; S12. Using each component's fault data as a node in the motor fault knowledge graph, establish the association path between each component's fault data based on the component fault association data to form the motor fault knowledge graph. The associated path is a directed path, with the path direction pointing from the node corresponding to the faulty component name and fault code in the component fault association data to the node corresponding to the cause component name and fault code.

3. The augmented reality-based electric motor operation and maintenance guidance method according to claim 2, characterized in that, Step S1, when a motor malfunctions, obtains the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph, specifically including the following steps: S13. When a motor malfunctions, the user inputs motor malfunction description data; The motor fault description data includes: textual description data of motor faults and video data of motor faults; S14. Obtain multiple nodes in the fault knowledge graph whose component name is motor, and extract the component fault text description data and component fault video data corresponding to each of the multiple nodes; S15. The text description data and video data of motor fault input by the user in step S13 are compared with the text description data and video data of component fault corresponding to each node in step S14. The node with the highest similarity is obtained and the node with the highest similarity is taken as the node with the highest matching degree in step S1.

4. The augmented reality-based electric motor operation and maintenance guidance method according to claim 3, characterized in that, Step S2, based on the node with the highest matching degree obtained in step S1, progressively associates fault cause nodes in the motor fault knowledge graph. When multiple fault cause nodes are obtained, a three-dimensional simulation model is used to verify the multiple fault cause nodes. Specifically, this includes the following steps: S21. Obtain the node pointed to by the node with the highest matching degree obtained in step S1, and take the pointed-to node as the fault cause node; S22. When there is only one fault cause node, obtain the new fault cause node pointed to by the fault cause node; S23. Repeat step S22 to obtain new fault cause nodes layer by layer until there are multiple new fault cause nodes, then execute step S25; when there is only one new fault cause node and the new fault cause node does not point to any node, take the new fault cause node as the fault cause node of the motor and execute step S4. S24. If there are multiple fault cause nodes obtained in step S21, proceed to step S25; S25. Obtain the component name, fault type, and fault code corresponding to each of the multiple new fault cause nodes obtained in step S23 or the multiple fault cause nodes obtained in step S24. S26. Input the component name, fault type, and fault code corresponding to each node obtained in step S25 into the three-dimensional simulation model, run the three-dimensional simulation model, and obtain multiple simulation results of the three-dimensional simulation model; Each time, only the component name, fault type, and fault code corresponding to a node are input to obtain the simulation results corresponding to that node. The simulation results are fault video data and fault codes of the electric motor; S27. Compare the motor fault code corresponding to the node with the highest matching degree obtained in step S1 with the fault code in the simulation run result obtained in step S26. If they are inconsistent, remove the node corresponding to the simulation run result. If they are consistent, the user compares whether the motor fault video data in the simulation run result is consistent with the motor fault video data obtained in step S13. If they are inconsistent, remove the node corresponding to the simulation run result. S28. Repeat step S27 until all the multiple fault cause nodes obtained in step S25 have been processed, and the fault cause nodes after one verification are obtained.

5. The augmented reality-based motor operation and maintenance guidance method according to claim 4, characterized in that, Step S3 uses a similarity method to perform secondary verification on the multiple fault cause nodes that have undergone the first verification in step S2, and obtains the fault cause nodes after secondary verification. Specifically, it includes the following steps: S31. Obtain the names of multiple components corresponding to the multiple fault cause nodes after one verification in step S2; S32. Obtain the names of multiple components from multiple sample data in the fault experience database; The fault experience database stores sample data of components of the motor that have failed. The sample data of the faulty components of the motor includes: component name, usage time, and operating environment data; The operating environment data includes: motor input voltage, ambient temperature, ambient humidity, and air dust data; S33. Check if the component name obtained in step S31 exists in the sample data stored in the fault experience database. If it does not exist, remove the node corresponding to the component name from the multiple fault cause nodes after one verification in step S2. S34. Repeat step S33 until all component names obtained in step S31 have been processed, and multiple fault cause nodes are retained after removal. S35. Obtain the usage duration of the components in the component names corresponding to the multiple fault cause nodes retained after the removal process obtained in step S34, and obtain the current operating environment data of the motor. The current operating environment data of the electric motor includes: motor input voltage, ambient temperature, ambient humidity, and air dust data; S36. Based on the current operating environment data of the motor, the usage time of the components in the component names corresponding to the multiple fault cause nodes retained after removal processing in step S35, and the sample data in the experience database, perform secondary verification on the multiple fault cause nodes retained after removal processing obtained in step S34 to obtain the fault cause nodes after secondary verification.

6. The augmented reality-based electric motor operation and maintenance guidance method according to claim 5, characterized in that, Step S36 specifically includes the following steps: S361. Obtain the component name of one of the multiple fault cause nodes retained after the removal process in step S35, and search for one or more sample data corresponding to the component name in the fault experience data. Calculate the similarity between the usage time of the component corresponding to the component name, the current environmental data of the motor, and the one or more sample data, to obtain one or more similarity values. The calculation method for the similarity values ​​is as follows: Where R is the similarity value. , , , These are the first weight value, the second weight value, the third weight value, and the fourth weight value, respectively. , , , These are usage time, input voltage, ambient temperature, and ambient humidity, respectively. Among them, the names of each component correspond to , , , The weight values ​​are all different; The corresponding name of each component , , , The weight values ​​are pre-stored in memory; S362. If one or more similarity values ​​obtained in step S361 are all less than the preset threshold T, then discard the node obtained in step S361. S363. Repeat steps S361 to S362 until all the fault cause nodes retained after the removal process in step S35 have been processed. The fault cause nodes retained after the removal process in steps S361 to S362 are used as the fault cause nodes after secondary verification.

7. The augmented reality-based electric motor operation and maintenance guidance method according to claim 6, characterized in that, Steps S41 and S42 employ augmented reality technology to provide users with motor operation and maintenance guidance, specifically including the following steps: S4a. The user wears VR glasses, and an augmented reality model of motor operation and maintenance is displayed in the VR glasses. The augmented reality model of motor operation and maintenance is superimposed on the location of the component to be operated according to the operation and maintenance steps of the motor. S4b. According to the operation and maintenance steps in the electric motor operation and maintenance augmented reality model, the user operates the electric motor components; the VR glasses monitor the user's hand movements in real time. When the similarity between the hand movements and the virtual hand movements in the electric motor operation and maintenance augmented reality model is lower than a preset threshold X, the virtual hand is displayed in red in the electric motor operation and maintenance augmented reality model displayed in the VR glasses. The preset threshold X is 80%; The augmented reality model for motor operation and maintenance displays textual descriptions of precautions for the virtual hand gesture, the fault location of the component being operated by the current hand gesture, and a textual description of the fault next to the virtual hand gesture. S4c. When the virtual hand is red in the augmented reality model of motor maintenance displayed in the VR glasses, the user can adjust the hand gesture, the display progress of the augmented reality model of motor maintenance, or the display mode of the augmented reality model of motor maintenance. The display progress of the adjusted electric motor operation and maintenance augmented reality model includes: rewind, pause, fast forward, and stop; The display methods for adjusting the augmented reality model of motor operation and maintenance include: local zoom-in and local zoom-out.

8. An augmented reality-based electric motor operation and maintenance guidance system, characterized in that, The system performs the following steps: S1. When a motor malfunctions, obtain the node with the highest matching degree of the motor malfunction in the motor malfunction knowledge graph; the motor malfunction knowledge graph is constructed based on historical motor malfunction data; S2. Based on the node with the highest matching degree obtained in step S1, the fault cause nodes are associated layer by layer in the motor fault knowledge graph. When there are multiple fault cause nodes, a three-dimensional simulation model is used to verify the multiple fault cause nodes. When there is only one fault cause node after verification, step S4 is executed; when there are multiple fault cause nodes after verification, step S3 is executed. S3. Use the similarity method to perform secondary verification on the multiple fault cause nodes after the first verification in step S2, obtain the fault cause nodes after secondary verification, and then execute step S4. S4. Based on the fault cause nodes, augmented reality technology is used to guide the operation and maintenance of motors; Step S4 specifically includes the following steps: S41. When there is only one fault cause node, extract the disassembly augmented reality model, the assembly augmented reality model, and the maintenance augmented reality model corresponding to the fault code of the fault cause node, and combine them in the order of disassembly augmented reality model, maintenance augmented reality model, and assembly augmented reality model, and use augmented reality technology to provide users with motor operation and maintenance guidance. S42. When there are multiple fault cause nodes, augmented reality technology is used to provide users with motor operation and maintenance guidance based on the hierarchical relationship of multiple fault cause nodes in the component BOM structure of the motor. Step S42 specifically includes the following steps: S421. Obtain the hierarchy and inclusion relationship of the multiple fault cause nodes in the component BOM structure of the motor; S422. Traverse the multiple fault cause nodes. When two or more fault cause nodes are in an inclusive relationship, form a dismantling path for the two or more fault cause nodes according to the hierarchy of each node, and finally form multiple dismantling paths. S423. For each disassembly path, first obtain the disassembly augmented reality model of the fault cause node with the lowest level in the disassembly path, and guide the user to perform the disassembly operation; The motor has a minimum level of 0, and according to the disassembly sequence of the motor, the lower the level of the component, the higher the level. S424. When the user confirms that the current fault cause node is faulty, the maintenance augmented reality model and the assembly augmented reality model corresponding to the fault code of the fault cause node are combined to provide the user with maintenance guidance and post-maintenance assembly, and the operation and maintenance guidance process ends. When the user confirms that the current fault cause node does not have a fault and that the current fault cause node has a subordinate fault cause node, the user directly obtains the disassembly augmented reality model of the subordinate fault cause node, guides the user to perform the disassembly operation, and repeats step S424. The lower-level fault cause node refers to the node in the disassembly path that is at a higher level than the current fault cause node and has the smallest level difference. If the user confirms that the current fault cause node does not have a fault and that the fault cause node does not have a subordinate fault cause node, obtain the assembled augmented reality model of the fault cause node and its superior fault cause node, and guide the user to assemble it. S425. For the multiple disassembly paths obtained in step S422, repeat steps S423 to S424 until the faulty node is found, and provide the user with repair guidance and post-repair assembly, thus ending the operation and maintenance guidance process.

Images