An edge computing embedded application dependency system

Through the design of the edge computing embedded application dependency system, the mutual monitoring heartbeat signal and the Bayesian network model of the service processing module, application update module and edge monitoring module are used to monitor each other's heartbeat signal and the Bayesian network model fault prediction, which solves the difficulty of self-checking of embedded applications in the edge computing system. and automatic repair of problems, realize system self-test and automatic repair, reduce maintenance costs and improve service efficiency.

CN118740660BActive Publication Date: 2024-11-08广东知业科技有限公司
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202411231665.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-04
Publication Date
2024-11-08
Estimated Expiration
2044-09-04

AI Technical Summary

Technical Problem

During long-term operation, embedded applications are difficult to self-check and automatically repair, resulting in high after-sales maintenance costs and low service efficiency.

Method used

Design an edge computing embedded application dependency system, including a business processing module, an application update module and an edge monitoring module, self-test and automatic repair by monitoring each other's heartbeat signals, and use Bayesian network model to predict fault types and process them.

Benefits of technology

It realizes self-test and automatic repair of edge computing systems, reduces after-sales maintenance costs, improves service efficiency, and ensures the stability and reliability of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118740660B_ABST
    Figure CN118740660B_ABST
Patent Text Reader

Abstract

The present invention relates to an edge computing embedded application dependency system, in which a business processing module, an application update module and an edge monitoring module send heartbeat signals to each other in real time according to a set time step; when a receiving end module does not receive a heartbeat signal, a sending end module that does not send a heartbeat signal is located; the sending end module that does not send a heartbeat signal is restarted or restored; when a heartbeat signal is still not received after restarting or restoring, an alarm signal is sent to a cloud service end; the cloud service end obtains a heartbeat signal of a sequence of time before the fault; the heartbeat signal sequence data is input into an offline trained Bayesian network model, and a predicted fault type is output; according to the predicted fault type, the fault is processed according to previous processing experience. The present invention solves the problem that it is difficult for embedded applications to self-check and automatically repair when the edge computing system is running for a long time, resulting in increased after-sales maintenance costs and reduced service efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of edge computing and relates to an edge computing embedded application dependency system. Background Art

[0002] Embedded applications are an important part of edge computing systems and the basis for achieving intelligence and automation. For edge computing, embedded applications are connected to the underlying hardware devices, so it is necessary to consider their adaptability to different hardware platforms and operating systems, and to effectively integrate and manage different components and peripherals. The stability and reliability of embedded applications are very important for the operation and maintenance of equipment.

[0003] In actual applications, edge computing systems often need to run for a long time, and in special scenarios, they need to be frequently powered on, powered off, updated, and restarted, which may make it difficult for embedded applications to self-check and automatically repair. For example, when upgrading the system or software, problems such as a sudden power outage or loss of system files may cause the embedded application to fail to start normally, increase after-sales maintenance costs, and reduce service efficiency. Therefore, special attention needs to be paid to the stability and reliability of embedded applications and their robustness needs to be improved to reduce after-sales maintenance costs and improve service efficiency.

[0004] In order to ensure the stability and reliability of embedded applications, a series of countermeasures need to be adopted, including designing and developing high-quality embedded software and hardware, using system-level testing and continuous integration methods to ensure the quality and functionality of the software, and establishing a complete embedded application maintenance mechanism to ensure timely response and handling of problems in the event of failures or emergencies. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides an edge computing embedded application dependency system.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] The present application provides an edge computing embedded application dependency system, including a business processing module, an application update module and an edge monitoring module, wherein the business processing module, the application update module and the edge monitoring module are communicatively connected, wherein:

[0008] The business processing module is used to process the daily business of edge computing;

[0009] The application update module is used to monitor the operation of the business processing module, and when a new version is found, stop the business, update and start the business processing module;

[0010] The edge monitoring module is used to read the business logs of the business processing module and the application update module to monitor the operation of edge computing in real time;

[0011] The business processing module, the application update module and the edge monitoring module perform self-checking and automatic repair on the edge computing system by monitoring each other's heartbeat line, specifically including the following steps:

[0012] S1. According to the set time step, the business processing module, the application update module and the edge monitoring module send heartbeat signals to each other in real time, and the heartbeat signals are configured as the operating status of each module;

[0013] S2. When the receiving end module does not receive the heartbeat signal, locate the sending end module that does not send the heartbeat signal;

[0014] S3, sending a restart instruction to the sending end module that has not sent the heartbeat signal, and the sending end module resends the heartbeat signal after restarting;

[0015] S4. When the heartbeat signal is still not received after restarting three times, a restore instruction is sent to the sending end module, that is, the sending end module is restored using the local backup;

[0016] S5. When the heartbeat signal is still not received after three restorations, an alarm signal is sent to the cloud server;

[0017] S6. After receiving the alarm signal, the cloud server obtains the heartbeat signal of the sending module in a certain period of time before the failure;

[0018] S7, inputting the heartbeat signal sequence data into the offline trained Bayesian network model, and outputting the predicted fault type;

[0019] S8. According to the predicted fault type, handle the fault according to previous handling experience;

[0020] Wherein, in step S7, the offline trained Bayesian network model includes the following training steps:

[0021] S71. Data collection: collecting heartbeat signal data and corresponding fault types when faults occurred in historical periods;

[0022] S72, data parameterization processing: dividing the heartbeat signal data into different operation levels according to the value size, and using the different operation levels as parameter values ​​of the heartbeat signal data;

[0023] S73, determining the model structure: using the heartbeat signal data as the parent node and the corresponding fault type as the child node; connecting the parent node and the child node to form a Bayesian network model structure;

[0024] S74, conditional probability table construction: construct a conditional probability table that correlates each operating level with the fault type based on the parameter values ​​obtained by the data parameterization processing.

[0025] Furthermore, in step S7, the output of the predicted fault type specifically outputs the occurrence probability of each fault type, and sorts each fault type from large to small according to the occurrence probability.

[0026] Furthermore, in step S8, the fault is processed according to the predicted fault type and previous processing experience, specifically, according to the order of the probability of occurrence of each fault type, the corresponding processing experience is selected to process the fault until the fault is resolved.

[0027] Furthermore, in step S72, the heartbeat signal data is divided into different operation levels according to the numerical values, specifically: the heartbeat signal data is divided into a plurality of clusters by using a cluster analysis method, and the operation levels are divided according to the numerical value ranges in the clusters.

[0028] Furthermore, the cluster analysis method is configured as a K-means clustering method.

[0029] Further, in step S74, the parameter values ​​obtained by the data parameterization processing are used to construct a conditional probability table of correlation between the operation level and the fault type, specifically: for each variable, given its parent node, the joint probability of the variable at the parent node is calculated, and the formula is:

[0030] ,

[0031] ,

[0032] In the formula, P ( U ) is represented as a set of nodes U = B 1 , B 2 ,……, B n The joint probability distribution of n is the number of nodes; Pa ( B i ) is a node in the Bayesian network B i The parent node set of P ( B i ) represents the target node B i Prior probability of; where Bi is the child node, that is, the fault type.

[0033] Furthermore, the cloud service end performs relevant remote operation and maintenance of edge computing through the edge monitoring module.

[0034] Furthermore, the edge monitoring module is also responsible for receiving various alarm signal processing operations from the cloud server and responding to the operation and maintenance results of the alarm signal from the cloud server.

[0035] Beneficial effects of the present invention:

[0036] According to the set time step, the business processing module, the application update module and the edge monitoring module send heartbeat signals to each other in real time; when the receiving end module does not receive the heartbeat signal, locate the sending end module that does not send the heartbeat signal; send a restart instruction to the sending end module that does not send the heartbeat signal; when the heartbeat signal is still not received after restarting three times, send a restore instruction to the sending end module; when the heartbeat signal is still not received after restoring three times, send an alarm signal to the cloud service end; after the cloud service end receives the alarm signal, obtain the heartbeat signal of the sending end module in a period of time before the failure; input the heartbeat signal sequence data into the Bayesian network model trained offline, and output the predicted fault type; according to the predicted fault type, the fault is processed according to previous processing experience. The present invention solves the problem that embedded applications are difficult to self-check and automatically repair when the edge computing system is running for a long time, resulting in increased after-sales maintenance costs and reduced service efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0038] Figure 1 It is a structural diagram of an edge computing embedded application dependency system in the present invention;

[0039] Figure 2 This is a flow chart of a business processing module, an application update module, and an edge monitoring module in one embodiment of the present invention performing self-inspection and automatic repair of an edge computing system by monitoring each other's heartbeat line. DETAILED DESCRIPTION

[0040] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation mode, structure, characteristics and effects of the present invention are described in detail below in combination with the accompanying drawings and preferred embodiments.

[0041] See also Figure 1-Figure 2 The present application provides an edge computing embedded application dependency system, including a business processing module, an application update module and an edge monitoring module, wherein the business processing module, the application update module and the edge monitoring module are communicatively connected, wherein:

[0042] The business processing module is used to process the daily business of edge computing;

[0043] In this embodiment, the business processing module is used to handle the daily business of the edge computing device, including analyzing and processing sensor data or other device status information, making intelligent decisions according to preset rules or specific needs, adjusting the working mode of the device or performing specific tasks to meet the needs of different business scenarios. In the edge computing system, the business processing module is the core part of the entire system. It can use the computing, storage, and communication resources of the edge computing nodes to achieve real-time data processing, intelligent decision-making, and flexible response to different business needs. At the same time, the business processing module can monitor and control the daily operation of the edge computing device based on various business models and algorithms, predict the probability of equipment failure, and thus optimize the operating efficiency and stability of the equipment.

[0044] The application update module is used to monitor the operation of the business processing module, and when a new version is found, stop the business, update and start the business processing module;

[0045] In this embodiment, the application update module is mainly used to monitor the operation of the service processing module, and when a new version is found, the service processing module is stopped, updated, and started. Specifically, the application update module can achieve the following functions:

[0046] 1) Real-time monitoring of the operation status of the business processing module, including various hardware indicators, network connection status, business processing load and other information.

[0047] 2) Based on known rules or algorithms, determine whether the business processing module needs to be updated. When a new version is found, the application update module can notify.

[0048] 3) Automated update process of business processing modules: Specifically, the application update module can control operations such as stopping services, downloading new versions, installing new versions, configuring parameters, and starting services, so that the system update process can be completed automatically.

[0049] 4) Monitor the progress and status of the update process to ensure the efficiency, stability and security of the update process.

[0050] With the support of the application update module, the edge computing system can realize the rapid and automatic update of the business processing module, so as to quickly adjust when the business scenario changes or the system is upgraded, and improve the reliability and adaptability of the system. At the same time, the application update module can also ensure the controllability and reliability of the update process, prevent unexpected situations, and improve the stability and security of the system.

[0051] The edge monitoring module is used to read the business logs of the business processing module and the application update module to monitor the operation of edge computing in real time;

[0052] In this embodiment, the edge monitoring module is mainly used to read the business logs of the business processing module and the application update module, and monitor the operation of edge computing in real time. Specifically, the edge monitoring module can achieve the following functions:

[0053] 1) Read the business logs of the business processing module and application update module, including operation records, error logs, performance indicators and other information.

[0054] 2) Analyze and process business logs to extract key information, such as business processing load, system response time, error rate, etc.

[0055] 3) Real-time monitoring of the operation of edge computing devices, including device status, network connection status, user behavior, etc.

[0056] 4) When abnormal situations are discovered, timely warnings and alarms are issued to handle and provide feedback on the problems.

[0057] With the support of the edge monitoring module, real-time monitoring and management of edge computing devices can be achieved to improve the stability, security and availability of the system. At the same time, by monitoring the performance and service quality indicators of the system, it can help relevant personnel to discover and solve problems in a timely manner, provide strong decision-making support, and thus improve business efficiency and user satisfaction.

[0058] The business processing module, application update module and edge monitoring module perform self-checking and automatic repair on the edge computing system by monitoring each other's heartbeat line;

[0059] It should be noted that heartbeat refers to the signal or data packet used to detect the operating status of a system or device in a computer network. These signals or data packets are periodically sent to other systems or devices to indicate that the sending system or device is still active. Heartbeat is usually used to monitor the availability and health status of devices or services in the network.

[0060] In actual applications, the heartbeat line can be a simple network communication protocol or a data packet in a specific format. The receiving system or device determines the status of the sending system or device by detecting the arrival of the heartbeat line. If the sending system or device stops sending the heartbeat line, the receiving system or device may think that the sender has failed or is unavailable.

[0061] Heartbeats are widely used in various systems and services, including network devices, servers, databases, etc. By monitoring and detecting heartbeats, administrators can promptly detect system or device failures or abnormalities and take appropriate measures to handle and repair them to ensure the availability and stability of the system or service.

[0062] The specific steps include:

[0063] S1. According to the set time step, the business processing module, the application update module and the edge monitoring module send heartbeat signals to each other in real time, and the heartbeat signals are configured as the operating status of each module;

[0064] In this embodiment, the heartbeat signal is configured as the running state, which means that each module will periodically send a signal to confirm its own running state. This signal is a simple notification to inform other modules that they are still running and have not encountered any problems. By configuring the heartbeat signal as the running state, the system can monitor the status of each module in real time to ensure that they are working properly. The implementation of this configuration is usually completed during the system design phase, and the developer will set a time interval for each module to send the heartbeat signal during this interval. When other modules receive this signal, they know that the sending module is still running and the system is normal.

[0065] S2. When the receiving end module does not receive the heartbeat signal, locate the sending end module that does not send the heartbeat signal;

[0066] It should be noted that in this process, the communication link must be checked first to ensure that it is normal. It may be due to a network failure or a communication device failure that the heartbeat signal fails to reach the receiving module. If the communication link is normal, the sending module must be checked next. It may be due to a failure or stoppage of the sending module itself that causes the heartbeat signal to fail to be sent.

[0067] S3, sending a restart instruction to the sending end module that has not sent the heartbeat signal, and the sending end module resends the heartbeat signal after restarting;

[0068] In this embodiment, restarting the sending end module that did not send the heartbeat signal is applicable to the situation where most edge computing systems fail. After the restart, the sending end module can restart and send a heartbeat signal to restore the normal operation of the system. Of course, before performing the restart operation, it is necessary to try to check whether there are other faults that may cause problems, and obtain detailed information about the failure to send the heartbeat signal as much as possible. This helps to avoid unnecessary system interruptions and reduce the impact on the system.

[0069] It is also important to note that before sending a restart command, it is recommended to send a stop signal to the sending module to ensure that it stops running and releases resources safely. Because in some cases, sending a restart command may cause data loss and system crash, so you need to consider it carefully before doing so.

[0070] At the same time, if the heartbeat signal is still not sent after the sending end module is restarted, the restart will continue until the heartbeat signal is still not sent after three restarts.

[0071] S4. When the heartbeat signal is still not received after restarting three times, a restore instruction is sent to the sending end module, that is, the sending end module is restored using the local backup;

[0072] In this embodiment, if the heartbeat signal cannot be received after restarting three times, local restoration of the sending end module can be considered. Local restoration refers to restoring the system using backup data to roll back to the previous state. Before performing local restoration, it is necessary to ensure that the data of the sending end module has been backed up, including configuration files, databases, logs, etc. In addition, a backup device that can maintain the backup data and ensure the integrity of the backup data must be arranged. When performing local restoration, it is necessary to stop the current sending end module and restore the backup data. Then restart the sending end module and monitor whether it sends a heartbeat signal.

[0073] It should be noted that local restoration may introduce new problems or lose updated data. Therefore, before performing a local restoration, you need to carefully evaluate the problems and risks and ensure the integrity and accuracy of the backup data. In addition, after performing the restoration operation, you need to further test and monitor the system to ensure that it is fully restored to normal operation.

[0074] In general, the interdependent business processing modules, application update modules, and edge monitoring modules in edge computing embedded applications can solve most fault problems by sending heartbeat signals to each other and restarting or restoring in emergencies, realizing self-checking and self-repair of the system without the need for on-site maintenance, thus solving the problems of high after-sales maintenance costs and low service efficiency. When encountering a fault problem that cannot be solved by the restart and restore program, an alarm signal needs to be sent to solve the corresponding fault problem.

[0075] S5. When the heartbeat signal is still not received after three restorations, an alarm signal is sent to the cloud server;

[0076] S6. After receiving the alarm signal, the cloud server obtains the heartbeat signal of the sending module in a certain period of time before the failure;

[0077] S7, inputting the heartbeat signal sequence data into the offline trained Bayesian network model, and outputting the predicted fault type;

[0078] In this embodiment, when restarting and restoring cannot solve the fault of the edge computing module, it means that the edge computing system may have an external fault, and the operating state before the fault can be used to determine what kind of fault occurred, that is, to determine the fault type. However, since the causal relationship between the fault operation state and the fault type is unknown, and the different fault types corresponding to the operating state data have a certain probability of occurrence, it is difficult to judge the fault type by the module operation state. Therefore, this embodiment uses a Bayesian network model with a quantitative causal relationship to monitor the heartbeat signal to determine the fault type. Specifically, after the cloud service receives the alarm signal, the heartbeat signal of the sending end module in a period of time before the fault is obtained, and these signals include the operating state of the sending end module, and then these heartbeat signal sequence data are input into the Bayesian network model trained offline. In the model training stage, the known fault type and the corresponding heartbeat signal data can be input into the model for training to establish a mapping relationship between the heartbeat signal and the fault type. After training, the model can be used to predict the possible fault type based on the real-time heartbeat signal data, and further processing and alarming can be performed. The Bayesian network model is a probabilistic graph model used to describe the dependency relationship between random variables. It is based on Bayes' theorem and uses graph structure to represent the conditional independence between variables, so that inference, prediction and decision-making can be performed. In a Bayesian network, nodes represent random variables and directed edges represent dependencies between variables. Each node represents a random variable, while the edges represent causal relationships or conditional dependencies between variables. The Bayesian network represents the probability distribution of each node through a conditional probability table (CPT), where the probability distribution of each node takes into account the influence of its parent node. The advantages of the Bayesian network model include:

[0079] 1) Interpretability: Since the network structure is clear, the dependencies between variables can be clearly expressed, so it has good interpretability.

[0080] 2) Uncertainty handling: Ability to effectively handle uncertainty and make inferences and predictions based on observed data.

[0081] 3) Data efficiency: It can efficiently use limited data for learning and inference, especially performing well in the face of high-dimensional data and few samples.

[0082] 4) Scalability: The network structure can be flexibly expanded as needed to suit different problems and scenarios.

[0083] S8. According to the predicted fault type, handle the fault according to previous handling experience;

[0084] Wherein, in step S7, the offline trained Bayesian network model includes the following training steps:

[0085] S71. Data collection: collecting heartbeat signal data and corresponding fault types when faults occurred in historical periods;

[0086] S72, data parameterization processing: dividing the heartbeat signal data into different operation levels according to the value size, and using the different operation levels as parameter values ​​of the heartbeat signal data;

[0087] S73, determining the model structure: using the heartbeat signal data as the parent node and the corresponding fault type as the child node; connecting the parent node and the child node to form a Bayesian network model structure;

[0088] S74, conditional probability table construction: construct a conditional probability table that correlates each operating level with the fault type based on the parameter values ​​obtained by the data parameterization processing.

[0089] Furthermore, in step S7, the output of the predicted fault type specifically outputs the occurrence probability of each fault type, and sorts each fault type from large to small according to the occurrence probability.

[0090] In this embodiment, for different heartbeat signals (operating status data), the corresponding various fault types have a certain probability of occurrence. Through the quantification of the Bayesian network model, the probability of occurrence of each fault type corresponding to the heartbeat signal can be predicted, and maintenance can be performed based on the probability of occurrence.

[0091] Furthermore, in step S8, the fault is processed according to the predicted fault type and previous processing experience, specifically, according to the order of the probability of occurrence of each fault type, the corresponding processing experience is selected to process the fault until the fault is resolved.

[0092] In this embodiment, according to the order of the probability of occurrence of each fault type, the corresponding processing experience is selected to handle the fault, and the fault type with a high probability of occurrence can be processed first, which is expected to accurately and quickly solve the fault problem.

[0093] Furthermore, in step S72, the heartbeat signal data is divided into different operation levels according to the numerical values, specifically: the heartbeat signal data is divided into a plurality of clusters by using a cluster analysis method, and the operation levels are divided according to the numerical value ranges in the clusters.

[0094] Furthermore, the cluster analysis method is configured as a K-means clustering method.

[0095] Further, in step S74, the parameter values ​​obtained by the data parameterization processing are used to construct a conditional probability table of correlation between the operation level and the fault type, specifically: for each variable, given its parent node, the joint probability of the variable at the parent node is calculated, and the formula is:

[0096] ,

[0097] ,

[0098] In the formula, P ( U ) is represented as a set of nodes U = B 1 , B 2 ,……, B n The joint probability distribution of n is the number of nodes; Pa ( B i ) is a node in the Bayesian network B i The parent node set of P ( B i ) represents the target node B i Prior probability of; where Bi is the child node, that is, the fault type.

[0099] Furthermore, the cloud service end performs relevant remote operation and maintenance of edge computing through the edge monitoring module.

[0100] Furthermore, the edge monitoring module is also responsible for receiving various alarm signal processing operations from the cloud server and responding to the operation and maintenance results of the alarm signal from the cloud server.

[0101] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. An edge computing embedded application dependent system, characterized in that: It includes a business processing module, an application update module and an edge monitoring module, which are communicatively connected to each other, wherein: The business processing module is used to process the daily business of edge computing; The application update module is used to monitor the operation of the business processing module, and when a new version is found, stop the business, update and start the business processing module; The edge monitoring module is used to read the business logs of the business processing module and the application update module to monitor the operation of edge computing in real time; The business processing module, the application update module and the edge monitoring module perform self-checking and automatic repair on the edge computing system by monitoring each other's heartbeat line, specifically including the following steps: S1. According to the set time step, the business processing module, the application update module and the edge monitoring module send heartbeat signals to each other in real time, and the heartbeat signals are configured as the operating status of each module; S2. When the receiving end module does not receive the heartbeat signal, locate the sending end module that does not send the heartbeat signal; S3, sending a restart instruction to the sending end module that has not sent the heartbeat signal, and the sending end module resends the heartbeat signal after restarting; S4. When the heartbeat signal is still not received after restarting three times, a restore instruction is sent to the sending end module, that is, the sending end module is restored using the local backup; S5. When the heartbeat signal is still not received after three restorations, an alarm signal is sent to the cloud server; S6. After receiving the alarm signal, the cloud server obtains the heartbeat signal of the sending module in a certain period of time before the failure; S7, inputting the heartbeat signal sequence data into the offline trained Bayesian network model, and outputting the predicted fault type; S8. According to the predicted fault type, handle the fault according to previous handling experience; Wherein, in step S7, the offline trained Bayesian network model includes the following training steps: S71. Data collection: collecting heartbeat signal data and corresponding fault types when faults occurred in historical periods; S72, data parameterization processing: dividing the heartbeat signal data into different operation levels according to the value size, and using the different operation levels as parameter values ​​of the heartbeat signal data; S73, determining the model structure: using the heartbeat signal data as the parent node and the corresponding fault type as the child node; connecting the parent node and the child node to form a Bayesian network model structure; S74, conditional probability table construction: construct a conditional probability table that correlates each operating level with the fault type based on the parameter values ​​obtained by the data parameterization processing.

2. The edge computing embedded application dependency system according to claim 1, characterized in that: In step S7, the output of the predicted fault type specifically outputs the occurrence probability of each fault type, and sorts the fault types from large to small according to the occurrence probability.

3. The edge computing embedded application dependency system according to claim 2, characterized in that: In step S8, the fault is processed according to the predicted fault type and previous processing experience, specifically, according to the order of the probability of occurrence of each fault type, the corresponding processing experience is selected to process the fault until the fault is resolved.

4. The edge computing embedded application dependency system according to claim 1, characterized in that: In step S72, the heartbeat signal data is divided into different operation levels according to the numerical values, specifically: the heartbeat signal data is divided into several clusters by using a cluster analysis method, and the operation levels are divided according to the numerical value ranges in the clusters.

5. The edge computing embedded application dependency system according to claim 4, characterized in that: The cluster analysis method is configured as a K-means clustering method.

6. The edge computing embedded application dependency system according to claim 1, characterized in that: In step S74, the parameter values ​​obtained by the data parameterization processing are used to construct a conditional probability table of the correlation between the operation level and the fault type. Specifically, for each variable, given its parent node, its joint probability at the parent node is calculated. The formula is: , , In the formula, P ( U ) is represented as a set of nodes U = B 1 , B 2 ,……, B n The joint probability distribution of n is the number of nodes; P a ( B i ) is a node in the Bayesian network B i The parent node set of P ( B i ) represents the target node B i The prior probability of Where Bi is a child node, that is, the fault type.

7. The edge computing embedded application dependency system according to claim 1, characterized in that: The cloud server performs relevant remote operation and maintenance of edge computing through the edge monitoring module.

8. The edge computing embedded application dependency system according to claim 1, characterized in that: The edge monitoring module is also responsible for receiving various alarm signal processing operations from the cloud server and responding to the cloud server's operation and maintenance results of the alarm signal.

Citation Information

Patent Citations

  • Method for searching register point location based on PLC and edge calculation

    CN115878186A

  • Distributed object recognition system and method, and edge computing apparatus

    WO2021203862A1