Edge cloud cooperative dynamic equipment fault analysis method and device, electronic equipment and medium
By training in the cloud and deploying a lightweight fault diagnosis model on the edge, the problem of high communication consumption of real-time uploading of vibration data of rotating mechanical equipment is solved, and fault analysis at the edge is realized, reducing traffic volume and improving diagnostic accuracy.
Patent Information
- Application Number
- CN202410959932.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-17
- Publication Date
- 2024-11-08
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the prior art, the vibration data of rotating machinery equipment needs to be uploaded to the cloud in real time for analysis, resulting in large communication consumption and inability to effectively reduce communication volume.
Obtain vibration information and device information at the edge of the sensor in the cloud, train it through a heavyweight fault diagnosis model, generate a lightweight fault diagnosis model, and send it to the edge end, so that the edge end can perform fault analysis and reduce real-time vibration information Upload.
It effectively reduces the communication between the edge and the cloud, improves the accuracy of fault diagnosis results, and extends the battery life of the sensor.
Smart Images

Figure CN118913431A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of dynamic equipment technology, and in particular to an edge-cloud collaborative dynamic equipment fault analysis method, device, electronic device and medium. Background Art
[0002] In the related technology, there are a large number of rotating machinery (moving equipment) in process industry factories. When an abnormal situation occurs inside the rotating machinery, the vibration will increase. Strong vibration and impact will cause damage or abnormal failure of the equipment and parts. In order to timely and accurately understand the working conditions of rotating machinery and equipment, it is necessary to obtain the information contained in the mechanical vibration and analyze the vibration mechanism, so as to diagnose the operation of the equipment and find possible faults. Therefore, vibration sensors are needed to collect vibration data of rotating machinery, and judge the fault type, fault location, fault severity and expected life of the moving equipment by identifying vibration information such as waveform spectrum. The existing method is to send vibration data to the cloud for analysis, which results in high communication consumption.
[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention
[0004] The main purpose of the embodiments of the present application is to propose an edge-cloud collaborative dynamic equipment fault analysis method, device, electronic device and medium, which can effectively reduce the communication volume.
[0005] To achieve the above-mentioned purpose, an embodiment of the present application proposes an edge-cloud collaborative dynamic equipment fault analysis method, the method is applied to a cloud, and the cloud interacts with at least one sensor edge terminal; the method includes:
[0006] Acquire first vibration information of a moving device corresponding to each of the sensor edge ends, the first vibration information including a historical vibration waveform and a historical vibration spectrum; and acquire device information of a moving device corresponding to each of the sensor edge ends, the device information including a device model, device parameters, and device component type;
[0007] Performing fault relevance training on a heavyweight fault diagnosis model according to all the first vibration information, wherein the fault relevance is used to characterize the relevance between the first vibration information and the fault type;
[0008] Retraining the trained heavyweight fault diagnosis model according to the device information to generate a lightweight fault diagnosis model;
[0009] The lightweight fault diagnosis model is sent to the target sensor edge end corresponding to the device information, so that the target sensor edge end performs dynamic device fault analysis through the lightweight fault diagnosis model.
[0010] In some embodiments, sending the lightweight fault diagnosis model to the target sensor edge terminal corresponding to the device information includes:
[0011] Determine a target sensor edge end from a plurality of sensor edge ends according to the device information;
[0012] Determining a target lightweight fault diagnosis model from a plurality of lightweight fault diagnosis models according to the device information, wherein each of the device information is associated with one of the lightweight fault diagnosis models;
[0013] The target lightweight fault diagnosis model is sent to the target sensor edge.
[0014] In some embodiments, performing dynamic equipment fault analysis using the lightweight fault diagnosis model includes:
[0015] Acquiring real-time vibration information, wherein the real-time vibration information includes a real-time vibration waveform and a real-time vibration spectrum;
[0016] The real-time vibration information is input into the lightweight fault diagnosis model to perform dynamic equipment fault analysis to obtain a fault diagnosis result, which includes a fault location, a fault type, a fault severity or a fault life expectancy.
[0017] In some embodiments, inputting the real-time vibration information into the lightweight fault diagnosis model to perform dynamic equipment fault analysis includes:
[0018] Obtaining a diagnosis time interval, wherein the diagnosis time interval is adjusted according to a historical fault diagnosis result or a change in the real-time vibration information;
[0019] According to the diagnosis time interval, the real-time vibration information is input into the lightweight fault diagnosis model to perform dynamic equipment fault analysis.
[0020] In some embodiments, the sensor edge terminal further performs the following steps:
[0021] Calculating the confidence level of the fault diagnosis result;
[0022] An upload status of the real-time vibration information is determined according to the confidence level, and the upload status is used to represent a status of the sensor edge end sending the real-time vibration information to the cloud.
[0023] In some embodiments, the method further comprises the following steps:
[0024] When receiving the real-time vibration information uploaded by the sensor edge end, forming a training set with the real-time vibration information and all the first vibration information;
[0025] Updating the model parameters of the heavyweight fault diagnosis model through the training set;
[0026] Retraining the heavyweight fault diagnosis model after updating the model parameters according to the device information to regenerate a lightweight fault diagnosis model;
[0027] The regenerated lightweight fault diagnosis model is sent to the target sensor edge end corresponding to the device information, so that the target sensor edge end updates the model parameters of the lightweight fault diagnosis model.
[0028] In some embodiments, the heavyweight neural network model includes a mechanism model and a digital AI model.
[0029] To achieve the above-mentioned purpose, another aspect of the embodiment of the present application proposes an edge-cloud collaborative dynamic equipment fault analysis device, the device is applied to a cloud, and the cloud interacts with at least one sensor edge terminal; the device includes:
[0030] The first module is used to obtain first vibration information of the moving device corresponding to each of the sensor edge ends, wherein the first vibration information includes a historical vibration waveform and a historical vibration spectrum; and obtain device information of the moving device corresponding to each of the sensor edge ends, wherein the device information includes a device model, device parameters, and device component type;
[0031] A second module is used to perform fault relevance training on a heavyweight fault diagnosis model according to all the first vibration information, wherein the fault relevance is used to characterize the relevance between the first vibration information and the fault type;
[0032] The third module is used to retrain the trained heavyweight fault diagnosis model according to the device information to generate a lightweight fault diagnosis model;
[0033] The fourth module is used to send the lightweight fault diagnosis model to the target sensor edge end corresponding to the device information, so that the target sensor edge end performs dynamic device fault analysis through the lightweight fault diagnosis model.
[0034] To achieve the above objective, another aspect of an embodiment of the present application provides an electronic device, the electronic device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the above method when executing the computer program.
[0035] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.
[0036] The embodiments of the present application include at least the following beneficial effects: The present application provides an edge-cloud collaborative dynamic equipment fault analysis method, apparatus, electronic device and medium. The scheme obtains the first vibration information of the dynamic equipment corresponding to each sensor edge end and the equipment information of the dynamic equipment corresponding to each sensor edge end in the cloud, and then trains the heavyweight fault diagnosis model for fault correlation according to all the first vibration information. Then, the trained heavyweight fault diagnosis model is re-trained according to the equipment information to generate a lightweight fault diagnosis model, and then the lightweight fault diagnosis model is sent to the target sensor edge end corresponding to the equipment information, so that the target sensor edge end performs dynamic equipment fault analysis through the lightweight fault diagnosis model, so that the edge end does not need to upload the vibration information to the cloud in real time, and the edge end can complete the dynamic equipment fault analysis, effectively reducing the communication volume between the edge end and the cloud. At the same time, fault analysis is performed through the lightweight fault diagnosis model, effectively improving the accuracy of the fault diagnosis results. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 It is a flow chart of the edge-cloud collaborative dynamic device fault analysis method provided in an embodiment of the present application;
[0038] Figure 2 It is a schematic diagram of an edge-cloud collaborative dynamic equipment fault analysis device provided in an embodiment of the present application;
[0039] Figure 3 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION
[0040] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application, they are only examples of devices and methods consistent with some aspects of the embodiments of the present application.
[0041] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another concept. For example, without departing from the scope of the embodiment of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".
[0042] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.
[0043] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.
[0044] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.
[0045] Dynamic equipment: refers to rotating equipment driven by a driving machine (i.e. equipment that consumes energy), such as pumps, compressors, fans, etc. Its energy can be electric power, pneumatic power, steam power, etc.
[0046] Process factory: refers to the production process from the initial processing of raw materials to the output of finished products, in which materials are always transferred through interconnected equipment, pipelines or other channels.
[0047] Mechanism model: A model established through mathematical modeling methods based on the understanding of the internal mechanisms and principles of a physical system. This model can reflect the essential laws and causal relationships of a physical system and is usually used to predict and control system behavior.
[0048] Data-driven model: A model built based on data analysis and machine learning techniques. This model uses a large amount of data for training and optimization, and can mine hidden laws and patterns in the data to model and predict physical systems.
[0049] Artificial Intelligence (AI) is a field of science and technology that studies how to enable computers to simulate and perform human intelligent tasks. It is committed to developing intelligent systems that can perceive, understand, learn, reason, make decisions, and interact with humans.
[0050] The embodiments of the present application provide an edge-cloud collaborative dynamic equipment fault analysis method, device, electronic device and medium. The embodiments deploy a lightweight fault diagnosis model on the edge and a heavyweight neural network model on the cloud. The lightweight fault diagnosis model corresponding to the sensor edge is obtained through integrated training of the heavyweight fault diagnosis model on the cloud, and the lightweight fault diagnosis model is sent to the corresponding sensor edge. Thus, the sensor edge can perform fault analysis on the dynamic equipment through the lightweight fault diagnosis model, without uploading vibration information to the cloud in real time, thereby effectively reducing communication volume.
[0051] The edge-cloud collaborative dynamic device fault analysis method provided in the embodiment of the present application relates to the field of dynamic device technology. The edge-cloud collaborative dynamic device fault analysis method provided in the embodiment of the present application can be applied to a server, and can also be software running in a server. In some embodiments, the server side can be configured as an independent physical server, or it can be configured as a server cluster or distributed system composed of multiple physical servers. It can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in the blockchain network; the software can be an application that implements the edge-cloud collaborative dynamic device fault analysis method, etc., but is not limited to the above forms.
[0052] Figure 1 This is an optional flow chart of the edge-cloud collaborative dynamic device fault analysis method provided in the embodiment of the present application. Taking the method applied to the cloud as an example, the cloud of the embodiment of the present application can interact with at least one sensor edge. During the interaction process, Figure 1 The method may include but is not limited to steps S110 to S140:
[0053] Step S110, obtaining first vibration information of a moving device corresponding to each of the sensor edge ends; and obtaining device information of a moving device corresponding to each of the sensor edge ends; wherein the first vibration information includes a historical vibration waveform and a historical vibration spectrum, and the device information includes a device model, device parameters, and device component type;
[0054] Step S120: performing fault relevance training on a heavyweight fault diagnosis model according to all first vibration information, wherein the fault relevance is used to characterize the relevance between the first vibration information and the fault type;
[0055] Step S130: retrain the trained heavyweight fault diagnosis model according to the device information to generate a lightweight fault diagnosis model;
[0056] Step S140: sending the lightweight fault diagnosis model to the target sensor edge end corresponding to the device information, so that the target sensor edge end performs dynamic device fault analysis through the lightweight fault diagnosis model.
[0057] In an embodiment of the present application, a vibration waveform or a vibration spectrum may reflect the intensity or envelope peak of the vibration. Equipment parameters may include, but are not limited to, power, shaft height, rated speed, rated voltage, etc. Equipment component types may include, but are not limited to, motors, pumps, fans, reducers, screw machines, compressors, etc. A unique moving device can be determined by the device model, device parameters, and device component type. Therefore, during the training process, the cloud can generate a lightweight fault diagnosis model corresponding to the moving device based on the device information, and send the lightweight fault diagnosis model to the sensor edge corresponding to the moving device, so that the sensor edge can perform fault analysis and judgment on the moving device through the lightweight fault diagnosis model.
[0058] In the embodiment of the present application, the fault types include motor fault, centrifugal pump fault, equipment foundation fault, rotor fault, coupling fault, rolling bearing fault, sliding bearing fault and gear fault, etc. The vibration waveform and vibration spectrum corresponding to each fault type have different characteristics. Therefore, the cloud can train a heavyweight fault diagnosis model based on the historical vibration information of all devices, so that the heavyweight fault diagnosis model has the correlation between vibration information and fault type, and then the generated lightweight fault diagnosis model can analyze the corresponding fault of the dynamic equipment.
[0059] In the embodiment of the present application, the cloud performs iterative classification training on the heavyweight neural network model based on the acquired vibration information, and when the model parameters meet the preset requirements, a lightweight fault diagnosis model corresponding to each sensor edge cloud is generated. Specifically, the heavyweight neural network model of this embodiment includes a mechanism model and a digital AI model, and these two models are integrated into an integrated model, which has both the ability to summarize and generalize, and a certain deductive reasoning ability, so that it can effectively complete the vibration waveform, spectrum and fault correlation training, reasoning and other processes of the dynamic equipment, and generate lightweight fault diagnosis models suitable for embedded deployment of different devices according to different equipment models, parameters and equipment types.
[0060] It is understandable that after the sensors in the edge cloud are installed and configured, the corresponding lightweight fault diagnosis model is downloaded from the cloud, so that when the equipment fault is judged, the edge cloud can perform fault reasoning through the lightweight fault diagnosis model, thereby eliminating the need to upload the vibration information to the cloud for analysis in real time, effectively reducing the communication volume. Specifically, when the cloud sends the lightweight fault diagnosis model to the target sensor edge corresponding to the device information, the target sensor edge can be determined from multiple sensor edge according to the device information; at the same time, the target lightweight fault diagnosis model is determined from multiple lightweight fault diagnosis models according to the device information, and then the target lightweight fault diagnosis model is sent to the target sensor edge, thereby improving the accuracy of the model sending result. It is understandable that each device information in this embodiment is associated with one of the lightweight fault diagnosis models, and through this association, the unique correspondence between the sensor edge and the lightweight fault diagnosis model can be achieved.
[0061] In this embodiment, after obtaining the corresponding lightweight fault diagnosis model, the sensor edge cloud inputs the real-time vibration information into the lightweight fault diagnosis model, and then the lightweight fault diagnosis model can analyze the vibration intensity of the real-time vibration information, the equipment temperature, or the change of the waveform spectrum envelope to generate the corresponding fault diagnosis result. Specifically, the fault diagnosis result of this embodiment includes the fault location, fault type, fault severity, or expected life of the fault.
[0062] In the embodiment of the present application, since the dynamic equipment will not always have faults, the sensor edge does not need to analyze the faults of the dynamic equipment in real time. This embodiment adjusts the diagnosis time interval based on the historical fault diagnosis results or the changes in real-time vibration information, and controls the dynamic equipment fault analysis process of the lightweight fault diagnosis model through the adjusted diagnosis time interval, thereby reducing the workload and working power consumption of the sensor edge while improving the real-time performance of data collection and analysis.
[0063] In an embodiment of the present application, when the sensor edge performs fault diagnosis analysis, it can also determine the upload status of the real-time vibration information by calculating the confidence of the fault diagnosis result. The upload status is used to characterize the state of the sensor edge sending the real-time vibration information to the cloud. For example, when it is determined that the equipment failure of the sensor edge is more serious according to the confidence of the fault diagnosis result, the real-time vibration information at this time can be uploaded to the cloud, so that the cloud updates the training data according to the latest vibration information, and retrains the heavyweight fault diagnosis model with the updated training data to update the model parameters of the heavyweight fault diagnosis model. After completing the retraining of the heavyweight fault diagnosis model, this embodiment will also regenerate the corresponding lightweight fault diagnosis model, and return the latest lightweight fault diagnosis model to the sensor edge, so that the sensor edge updates the model parameters of the lightweight fault diagnosis model to the latest model parameters, thereby improving the fault analysis accuracy of the sensor edge model.
[0064] In summary, the method provided in the embodiments of the present application has the following beneficial effects:
[0065] First, at the edge of the sensor, the waveform and spectrum can be used to diagnose and analyze equipment faults, forming a judgment on the equipment fault location, fault type, fault severity, and expected life span.
[0066] Second, waveforms and spectra do not need to be frequently uploaded to the cloud for diagnostic analysis, which greatly reduces the amount of communication data and communication power consumption, and enhances the battery life of edge sensors.
[0067] Third, the time interval for data collection and diagnosis at the edge of the sensor can be further shortened, and the real-time performance of data collection and diagnostic analysis can be improved.
[0068] Fourth, the edge sensor only downloads the lightweight fault diagnosis model corresponding to the monitored equipment. The model is lightweight, occupies little resources, and has a short diagnosis time.
[0069] Fifth, the lightweight fault diagnosis model at the sensor edge can be dynamically updated and iterated with the cloud, continuously improving the accuracy of the sensor edge's fault diagnosis of dynamic equipment.
[0070] See also Figure 2 The embodiment of the present application also provides an edge-cloud collaborative dynamic equipment fault analysis device, which is applied to the cloud, and the cloud interacts with at least one sensor edge terminal; wherein the device includes:
[0071] The first module 210 is used to obtain first vibration information of the moving device corresponding to each sensor edge end, the first vibration information includes a historical vibration waveform and a historical vibration spectrum; and obtain device information of the moving device corresponding to each sensor edge end, the device information includes a device model, device parameters and device component type;
[0072] The second module 220 is used to perform fault relevance training on the heavyweight fault diagnosis model according to all the first vibration information, where the fault relevance is used to characterize the relevance between the first vibration information and the fault type;
[0073] The third module 230 is used to retrain the trained heavyweight fault diagnosis model according to the device information to generate a lightweight fault diagnosis model;
[0074] The fourth module 240 is used to send the lightweight fault diagnosis model to the target sensor edge end corresponding to the device information, so that the target sensor edge end performs dynamic device fault analysis through the lightweight fault diagnosis model.
[0075] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0076] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above-mentioned edge-cloud collaborative dynamic device fault analysis method when executing the computer program. The electronic device can be any smart terminal including a tablet computer, a car computer, etc.
[0077] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0078] See also Figure 3 , Figure 3 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:
[0079] The processor 310 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.
[0080] The memory 320 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 320 can store an operating system and other applications. When the technical solution provided in the embodiment of this specification is implemented by software or firmware, the relevant program code is stored in the memory 320, and the processor 310 calls and executes the edge-cloud collaborative dynamic device fault analysis method of the embodiment of this application;
[0081] Input / output interface 330, used to implement information input and output;
[0082] The communication interface 340 is used to realize the communication interaction between the device and other devices. The communication can be realized through a wired manner (such as USB, network cable, etc.) or a wireless manner (such as mobile network, WI FI, Bluetooth, etc.);
[0083] bus 350 , which transmits information between the various components of the device (e.g., processor 310 , memory 320 , input / output interface 330 , and communication interface 340 );
[0084] The processor 310 , the memory 320 , the input / output interface 330 and the communication interface 340 are connected to each other in communication within the device via the bus 350 .
[0085] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned edge-cloud collaborative dynamic device fault analysis method.
[0086] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.
[0087] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.
[0088] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.
[0089] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.
[0090] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0091] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units 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 is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0092] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0093] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.
Claims
1. A method for analyzing dynamic equipment failures in edge-cloud collaboration, characterized in that: The method is applied to a cloud, and the cloud interacts with at least one sensor edge terminal; the method comprises: Acquire first vibration information of a moving device corresponding to each of the sensor edge ends, the first vibration information including a historical vibration waveform and a historical vibration spectrum; and acquire device information of a moving device corresponding to each of the sensor edge ends, the device information including a device model, device parameters, and device component type; Performing fault relevance training on a heavyweight fault diagnosis model according to all the first vibration information, wherein the fault relevance is used to characterize the relevance between the first vibration information and the fault type; Retraining the trained heavyweight fault diagnosis model according to the device information to generate a lightweight fault diagnosis model; The lightweight fault diagnosis model is sent to the target sensor edge end corresponding to the device information, so that the target sensor edge end performs dynamic device fault analysis through the lightweight fault diagnosis model.
2. The method according to claim 1, characterized in that The step of sending the lightweight fault diagnosis model to the target sensor edge terminal corresponding to the device information includes: Determine a target sensor edge end from a plurality of sensor edge ends according to the device information; Determining a target lightweight fault diagnosis model from a plurality of lightweight fault diagnosis models according to the device information, wherein each of the device information is associated with one of the lightweight fault diagnosis models; The target lightweight fault diagnosis model is sent to the target sensor edge.
3. The method according to claim 1, characterized in that The performing of dynamic equipment fault analysis by using the lightweight fault diagnosis model includes: Acquiring real-time vibration information, wherein the real-time vibration information includes a real-time vibration waveform and a real-time vibration spectrum; The real-time vibration information is input into the lightweight fault diagnosis model to perform dynamic equipment fault analysis to obtain a fault diagnosis result, which includes a fault location, a fault type, a fault severity or a fault life expectancy.
4. The method according to claim 3, characterized in that The step of inputting the real-time vibration information into the lightweight fault diagnosis model to perform dynamic equipment fault analysis includes: Obtaining a diagnosis time interval, wherein the diagnosis time interval is adjusted according to a historical fault diagnosis result or a change in the real-time vibration information; According to the diagnosis time interval, the real-time vibration information is input into the lightweight fault diagnosis model to perform dynamic equipment fault analysis.
5. The method according to claim 3, characterized in that: The sensor edge terminal also performs the following steps: Calculating the confidence level of the fault diagnosis result; An upload status of the real-time vibration information is determined according to the confidence level, and the upload status is used to represent a status of the sensor edge end sending the real-time vibration information to the cloud.
6. The method according to claim 5, characterized in that The method further comprises the following steps: When receiving the real-time vibration information uploaded by the sensor edge end, forming a training set with the real-time vibration information and all the first vibration information; Updating model parameters of the heavyweight fault diagnosis model through the training set; Retraining the heavyweight fault diagnosis model after updating the model parameters according to the device information to regenerate a lightweight fault diagnosis model; The regenerated lightweight fault diagnosis model is sent to the target sensor edge end corresponding to the device information, so that the target sensor edge end updates the model parameters of the lightweight fault diagnosis model.
7. The method according to claim 1, characterized in that The heavyweight neural network model includes a mechanism model and a digital AI model.
8. An edge-cloud collaborative dynamic equipment fault analysis device, characterized in that: The device is applied to a cloud, and the cloud interacts with at least one sensor edge terminal; the device includes: The first module is used to obtain first vibration information of the moving device corresponding to each of the sensor edge ends, wherein the first vibration information includes a historical vibration waveform and a historical vibration spectrum; and obtain device information of the moving device corresponding to each of the sensor edge ends, wherein the device information includes a device model, device parameters, and device component type; A second module is used to perform fault relevance training on a heavyweight fault diagnosis model according to all the first vibration information, wherein the fault relevance is used to characterize the relevance between the first vibration information and the fault type; The third module is used to retrain the trained heavyweight fault diagnosis model according to the device information to generate a lightweight fault diagnosis model; The fourth module is used to send the lightweight fault diagnosis model to the target sensor edge end corresponding to the device information, so that the target sensor edge end performs dynamic device fault analysis through the lightweight fault diagnosis model.
9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
High-end equipment fault intelligent diagnosis system and method based on side cloud cooperation
CN112286751A
Federal learning-based equipment fault prediction method and device under cloud edge collaborative architecture
CN116520814A
Bearing fault diagnosis model training method based on local time-frequency feature transfer learning
CN116541713A
Industrial robot bearing fault edge detection method and system
CN116878881A
Edge cloud cooperative multistage fault diagnosis method and system based on stack type auto-encoder
CN117932307A
Cited By
Method and device for carrying out partial discharge detection through edge cloud cooperation
CN120847566A
