Network fault processing method and system, mobile robot and automatic manipulator

Abstract

The invention discloses a network fault processing method and system, a mobile robot and an automatic manipulator. The invention relates to the technical field of network fault processing, and discloses a network fault processing system, method and device and a storage medium, and the method comprises the steps: obtaining network equipment with an interface in a computer room, monitoring a port of each network equipment through edge monitoring equipment, and enabling the edge monitoring equipment to be installed on the port of the network equipment, and to be responsible for collecting the data of the port; when the edge monitoring device monitoring the network device gives an alarm, the mobile robot in the repair area is dispatched to repair the network device; when the edge monitoring equipment gives an alarm for the abnormal data of the port, executing a sectional alarm strategy, and repairing the port with the abnormal data; executing a fault risk prediction strategy on each port, and predicting a risk value of a fault of the port; and the robot deployment strategy is executed, and the movable robot is allocated to the port, so that potential faults can be responded prospectively, service interruption is effectively avoided, and the operation and maintenance efficiency and the customer satisfaction are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of network fault processing, and in particular to a network fault processing method, system, mobile robot and automated manipulator. Background Art

[0002] As enterprises and organizations rely more on digitalization, the complexity and scale of network infrastructure continue to expand, which leads to the demand for advanced operation and maintenance management. Modern network operation and maintenance not only includes traditional equipment monitoring and troubleshooting, but also involves automated configuration management, real-time performance analysis, network security monitoring and user experience optimization. At the same time, the popularity of cloud computing, the Internet of Things and mobile applications also requires operation and maintenance management to quickly adapt to the dynamically changing network environment. It has become a trend to use artificial intelligence and machine learning technologies for predictive maintenance and intelligent management to improve operation and maintenance efficiency and reduce operating costs.

[0003] In the existing technology, network fault handling usually relies on manual on-site intervention and the use of professional tools for diagnosis and repair. Manual troubleshooting is subject to time and space constraints, making it difficult to quickly locate and solve problems. Especially in large-scale network environments, fault response time is long, affecting service continuity.

[0004] This solution proposes a network fault handling method, system, mobile robot and automated manipulator, which have solved the problems existing in the prior art. Summary of the invention

[0005] The present invention provides a network fault processing system, method, device and storage medium, which promote the solution of the problems mentioned in the above background technology.

[0006] In a first aspect, the present application provides a network fault processing method, which adopts the following technical solution: A network fault processing method comprises: obtaining a network device with an interface in a computer room, and using an edge monitoring device to monitor the port of each network device, wherein: The edge monitoring device is installed on the port of the network device and is responsible for collecting data from the port, determining whether the data is abnormal, and issuing an alarm for abnormal data; By installing edge monitoring devices on the ports of network devices, port data can be collected in real time and judged whether the data is abnormal. Port anomalies can be discovered in time, potential network failures can be responded to quickly, and the stability and reliability of the network can be ensured.

[0007] Set the repair area radius; For any network device, the circle corresponding to the location of the network device as the center and the radius of the repair area as the radius is the repair area of ​​the network device; by setting a repair area for each network device, the center of the repair area is the location of the device, and the radius is set according to the actual needs of the repair area. It can accurately determine the coverage of the repair task, ensure that the repair robot can quickly arrive and perform the task when a fault occurs, and determine that the edge monitoring device can only dispatch the mobile robot in its repair area, avoiding the mobile robot in one repair area from being affected by the edge monitoring devices in other repair areas, thereby improving management efficiency.

[0008] By dynamically adjusting robot deployment according to the fault status and the size of the repair area, we can flexibly respond to faults of different scales, avoid waste or shortage of resources, and improve overall repair efficiency.

[0009] When the edge monitoring device that monitors the network equipment issues an alarm, a mobile robot in the repair area is dispatched to repair the network equipment; By associating edge monitoring devices with mobile robots in the repair area, when the edge monitoring device issues an alarm, the system dispatches mobile robots to perform repair tasks according to the repair area. This enables automated fault repair, reduces the need for manual intervention, improves response speed, and reduces network interruption time.

[0010] The mobile robot is equipped with an automated manipulator for connecting to the port with abnormal data and replacing the port to establish a connection with the upper network device for data transmission; By equipping the mobile robot with an automated manipulator, connecting the port with abnormal data, and replacing the port to establish a connection with the upper-layer device, the stability of the network connection is ensured, and the data transmission link is repaired in time to avoid data loss and network instability caused by port failure.

[0011] Use a centralized monitoring device to connect all edge monitoring devices. When the edge monitoring device issues an alarm for abnormal data on a port, a segmented alarm strategy is executed to repair the port with abnormal data. Connect all edge monitoring devices through centralized monitoring equipment, execute segmented alarm strategies and schedule repair tasks. Through unified monitoring platform for data aggregation and alarm processing, high-priority fault tasks can be processed first to ensure that network problems are solved efficiently.

[0012] For any network device, obtain all ports of the network device, execute the fault risk prediction strategy for each port, and predict the risk value of the port failure; Set the risk threshold to determine whether the port is a high-risk port; When the risk value of a port is greater than or equal to the risk threshold, the port is identified as a high-risk port; When the risk value of a port is less than the risk threshold, the port is identified as a low-risk port; By using a neural network model to predict the risk value of port failure and identifying high-risk ports based on risk thresholds, potential faulty ports can be identified in advance and preventive maintenance can be performed to reduce the probability of sudden failures and the risk of network interruption.

[0013] Get all mobile robots, execute robot deployment strategies, and assign mobile robots to ports.

[0014] According to the risk value and prediction results of the port, mobile robots are assigned to high-risk ports and deployment strategies are formulated to ensure that high-risk ports can be repaired first, improve the efficiency of fault response and repair accuracy, and avoid small problems from turning into big faults.

[0015] Preferably, when the edge monitoring device issues an alarm for abnormal data of a port, a segmented alarm strategy is executed to repair the port with abnormal data, including: The mobile robot that is repairing the port is recorded as the first type of robot; The movable robots waiting for the repair port are recorded as the second type of robots; When the edge monitoring device issues an alarm for abnormal data of the port, the network device where the port is located is obtained, the repair area corresponding to the network device is obtained, and all the second-type robots in the repair area are obtained; Determine whether the number of the second type of robots is 0; When the number of the second type of robots is 0, the edge monitoring device is controlled to continuously alarm; When the number of second-category robots is not 0, any second-category robot is dispatched to repair the port with abnormal data and stop the edge monitoring device from alarming.

[0016] Preferably, when the edge monitoring device issues an alarm for abnormal data of a port, a segmented alarm strategy is executed to repair the port with abnormal data, and the method further includes: The moment when the edge monitoring device starts to alarm is recorded as the first moment; Set the alarm duration threshold; The time after the first time and the interval of the alarm duration is recorded as the second time; At the second moment, determining whether the edge monitoring device stops alarming; If the edge monitoring device alarms at the second moment, the alarm information is uploaded to the centralized monitoring device, and the alarm information includes the port location and the alarm duration of the edge monitoring device; The centralized monitoring device obtains all uploaded alarm information, and sets priorities for the ports of edge monitoring devices according to the alarm duration in the alarm information from large to small. The longer the alarm duration, the higher the priority. Obtain each port in order from high to low priority, and record the obtained port as the execution port; Obtain the positions of all second-type robots, and calculate the straight-line distance from the position of each second-type robot to the position of the execution port; Obtain the second type of robot with the shortest straight-line distance, which is recorded as the execution robot; The dispatch execution robot repairs the execution port and stops the edge monitoring device from alarming.

[0017] By implementing a segmented alarm strategy, the repair robot is scheduled according to the alarm duration and priority order to ensure timely response and fault handling. The alarm delay is effectively reduced to avoid the continued expansion of faults due to unhandled alarms, ensuring that network equipment resumes normal operation as soon as possible. The robot closest to the faulty port is scheduled for repair based on the straight-line distance between the robot position and the execution port. The task execution efficiency is optimized, the repair time is reduced, the flexibility and response speed of robot scheduling are improved, and the repair task is completed quickly.

[0018] Preferably, the acquiring of all ports of the network device, executing a fault risk prediction strategy for each port, and predicting a risk value of a port failure includes: Get any port of any network device; extract the characteristic data of the port, expressed as ,in, Indicates the characteristics of the port; Execute the following formula to calculate the risk value y of the port; ,in, is the activation function of the neural network, W is the weight matrix, and b is the bias term.

[0019] Preferably, the acquiring of all ports of the network device, executing a fault risk prediction strategy for each port, and predicting a risk value of a port failure includes: Get the number of all ports, recorded as m; Calculate the binary loss function value of the neural network: Get the current status of each port , where if a data anomaly occurs on the port, =1, otherwise, =0; calculate ,in, is the risk value of port i; Set the loss function threshold; Compare the binary loss function value with the loss function threshold. If the binary loss function value ≧ the loss function threshold, execute the parameter update strategy, which is: calculate , the result is used as the new weight matrix; calculate , the result is used as the new bias term, where is the learning rate, are the parameters of the loss function for the weight matrix and bias term; Calculate the risk value of each port again; If the binary loss function value is less than the loss function threshold, stop executing the fault risk prediction strategy to obtain the number of all ports, denoted as m; Calculate the binary loss function value of the neural network: Get the current status of each port , where if a data anomaly occurs on the port, =1, otherwise, =0; calculate ,in, is the risk value of port i; Set the loss function threshold; Compare the binary loss function value with the loss function threshold. If the binary loss function value ≧ the loss function threshold, execute the parameter update strategy, which is: calculate , the result is used as the new weight matrix; calculate , the result is used as the new bias term, where is the learning rate, are the parameters of the loss function for the weight matrix and bias term; Calculate the risk value of each port again; If the binary loss function value is less than the loss function threshold, the fault risk prediction strategy is stopped.

[0020] Preferably, the acquiring of all movable robots, executing the robot deployment strategy, and allocating the movable robots to the ports includes: Get all the movable robots and number them from 1 to h; Set deployment values ​​for each port , the deployment value indicates whether to deploy the mobile robot j to port i. If the mobile robot j is deployed to port i, then =1, otherwise, =0, where 1≦j≦h, 1≦i≦m; calculate ,in, represents the straight-line distance from the mobile robot to port i; And stipulate ≧1, where 1≦i≦m, indicating that at least one mobile robot must be deployed at each high-risk port; Regulation ≦1, where 1≦j≦h, indicating that each mobile robot is deployed to at most one port.

[0021] The neural network model predicts the risk value of port failure by learning and calculating the port's historical data. Accurately predict the port's failure risk and take corresponding measures based on the risk prediction to reduce the possibility of failure.

[0022] In a second aspect, the present application provides a network fault handling system, which adopts the following technical solution: a network fault handling system, comprising: Edge monitoring devices are installed at the ports of network devices to monitor port data in real time, detect anomalies and issue alarms; Centrally monitor equipment, aggregate alarm information from edge devices, schedule repair tasks based on priority, and implement segmented alarm strategies to ensure timely repairs; The fault risk prediction module uses a neural network to predict the port fault risk, determines the high-risk ports based on the predicted values, and dispatches robots for repair; The robot deployment and scheduling module assigns a mobile robot to each port to ensure that high-risk ports are repaired first.

[0023] On the third aspect, the present application provides a mobile robot, which adopts the following technical solution: a device, including: each mobile robot is only equipped with one automated manipulator, connected to the edge monitoring device and the centralized monitoring device, and when a port with abnormal data appears, it locates, moves, operates and repairs the port with abnormal data.

[0024] Through the collaboration between edge monitoring equipment and mobile robots, fault information is fed back in a timely manner, and the robots perform repairs. This ensures that edge devices can monitor and respond to network faults in a timely manner, and the automation of the repair process reduces human intervention, thereby improving network reliability and efficiency.

[0025] In the fourth aspect, the present application provides an automated manipulator that adopts the following technical solution: a computer-readable storage medium, comprising: the automated manipulator adopts a modular design and pre-installs a variety of adapters according to the interface types of different network devices; when receiving an instruction for a specific type of fault, the robot autonomously adjusts the shape of the manipulator's end effector to match the port of the device.

[0026] The manipulator adopts modular design and pre-installs a variety of adapters, which can automatically adjust the end effector shape according to different device interface types. This improves the versatility and adaptability of the manipulator, reduces the extra cost and time waste caused by changes in device types, and achieves a wider range of adaptation and flexible response.

[0027] The present invention has the following beneficial effects: 1. This network fault handling method uses a mobile robot with an automated manipulator to quickly locate the fault point and connect to the upper-layer equipment wirelessly or wired, which effectively solves the fault, greatly reduces the time for troubleshooting and repairing, and improves the stability and service availability of the network system. In particular, combined with the prediction function of the AI ​​algorithm and the design of a multi-purpose adapter, it can proactively respond to potential faults, effectively avoid service interruptions, and improve operation and maintenance efficiency and customer satisfaction.

[0028] 2. This network fault handling method can detect data anomalies of the port in a timely manner by real-time monitoring of the ports of network devices, ensuring that the device always maintains the best working condition. The edge monitoring device provides fine-grained real-time monitoring for each port, which enables network faults to be identified at the early stage of occurrence and quickly alarmed. This early detection mechanism can significantly reduce the occurrence rate of network faults, reduce the impact on business, and improve the availability and stability of equipment. At the same time, this also provides accurate data support for subsequent repair work and improves the overall operation and maintenance efficiency.

[0029] 3. This network fault handling method sets a repair area for each network device, with the center of the repair area being the device location and the radius adjusted as needed. The repair task can be spatialized to ensure that the repair robot can respond quickly within the predetermined range. This regional management not only optimizes the allocation of robot resources, but also reduces the waste of time and resources during the repair process. Through this precise spatial division, the robot can complete the task more effectively, reduce manual intervention, and improve the response speed of fault repair. This precise positioning method helps to improve the efficiency and reliability of the overall repair process and ensure that the repair work is more timely and effective.

[0030] 4. This network fault handling method can achieve seamless fault response through instant alarm and repair robot dispatch mechanism at the moment of fault discovery. Automated robots can quickly arrive at the site and perform precise repairs, avoiding possible delays and operational errors during manual repairs. Automated repairs greatly improve the speed of fault handling and shorten equipment downtime. This not only improves the recovery capability of network equipment, but also reduces the risk of network interruption, ensuring higher network availability, thereby improving overall operational efficiency.

[0031] 5. This network fault handling method uses an automated manipulator to ensure reliable connection between the abnormal data port and the upper-layer device, avoiding possible errors in manual operation. The robot can perform tasks autonomously and intelligently adjust the actuator to meet the needs of different interfaces. This automated repair not only speeds up fault handling, but also eliminates the risk of human intervention and improves the accuracy of repair. With the intelligent design of the manipulator, network repair becomes more efficient and reliable, ensuring the stable transmission of data streams, reducing the time for fault recovery, and improving the overall reliability of the network.

[0032] 6. This network fault handling method calculates the fault risk value of the port and identifies potential problems by learning historical data through a neural network. It can predict the possibility of failure of each port by learning a large amount of historical data, so that maintenance personnel can identify high-risk ports in advance and take preventive measures. Compared with traditional experience-based judgment, neural networks can find potential problems in a wider range and give more accurate risk assessments. By predicting risks in advance, network interruptions caused by sudden equipment failures are reduced, ensuring the stable operation of the network. Through intelligent fault prediction, the overall reliability of network equipment is improved, the operation and maintenance costs are reduced, and the efficiency of fault handling is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 Schematic diagram of the method of the present invention.

[0034] Figure 2 Schematic diagram of the system of the present invention.

[0035] Figure 3 The figure is a schematic diagram of the process of calculating the port risk value according to the present invention.

[0036] Figure 4 Schematic diagram of segmented alarm strategy. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0038] Embodiment 1, refer to Figure 1 , a network fault processing method, comprising: Obtain network devices with interfaces in the computer room and use edge monitoring devices to monitor the ports of each network device, where: The edge monitoring device is installed on the port of the network device and is responsible for collecting data from the port, determining whether the data is abnormal, and issuing an alarm for abnormal data; Set the repair area radius; For any network device, a circle with the location of the network device as the center and the radius of the repair area as the radius is the repair area of ​​the network device; When the edge monitoring device that monitors the network equipment issues an alarm, a mobile robot in the repair area is dispatched to repair the network equipment; The mobile robot is equipped with an automated manipulator for connecting to the port with abnormal data and replacing the port to establish a connection with the upper network device for data transmission; Use a centralized monitoring device to connect all edge monitoring devices. When the edge monitoring device issues an alarm for abnormal data on a port, a segmented alarm strategy is executed to repair the port with abnormal data. For any network device, obtain all ports of the network device, execute the fault risk prediction strategy for each port, and predict the risk value of the port failure; Set the risk threshold to determine whether the port is a high-risk port; When the risk value of a port is greater than or equal to the risk threshold, the port is identified as a high-risk port; When the risk value of a port is less than the risk threshold, the port is identified as a low-risk port; Get all mobile robots, execute robot deployment strategies, and assign mobile robots to ports.

[0039] When the edge monitoring device issues an alarm for abnormal data of the port, a segmented alarm strategy is executed to repair the port with abnormal data, including: The mobile robot that is repairing the port is recorded as the first type of robot; The movable robots waiting for the repair port are recorded as the second type of robots; When the edge monitoring device issues an alarm for abnormal data of the port, the network device where the port is located is obtained, the repair area corresponding to the network device is obtained, and all the second-type robots in the repair area are obtained; Determine whether the number of the second type of robots is 0; When the number of the second type of robots is 0, the edge monitoring device is controlled to continuously alarm; When the number of second-category robots is not 0, any second-category robot is dispatched to repair the port with abnormal data and stop the edge monitoring device from alarming.

[0040] When the edge monitoring device issues an alarm for abnormal data of the port, a segmented alarm strategy is executed to repair the port with abnormal data, and the method further includes: The moment when the edge monitoring device starts to alarm is recorded as the first moment; Set the alarm duration threshold; The time after the first time and the interval of the alarm duration is recorded as the second time; At the second moment, determining whether the edge monitoring device stops alarming; If the edge monitoring device alarms at the second moment, the alarm information is uploaded to the centralized monitoring device, and the alarm information includes the port location and the alarm duration of the edge monitoring device; The centralized monitoring device obtains all uploaded alarm information, and sets priorities for the ports of edge monitoring devices according to the alarm duration in the alarm information from large to small. The longer the alarm duration, the higher the priority. Obtain each port in order from high to low priority, and record the obtained port as the execution port; Obtain the positions of all second-type robots, and calculate the straight-line distance from the position of each second-type robot to the position of the execution port; Obtain the second type of robot with the shortest straight-line distance, which is recorded as the execution robot; The dispatch execution robot repairs the execution port and stops the edge monitoring device from alarming.

[0041] In this embodiment, refer to Figure 4 , Schematic diagram of segmented alarm strategy.

[0042] By integrating and classifying the alarm information of multiple edge devices through centralized monitoring equipment, the priority of faults can be determined more accurately and repair tasks can be reasonably scheduled. This segmented alarm strategy can ensure that the most urgent faults are repaired first, avoiding waste of resources and conflicts in repair tasks. At the same time, the system can dynamically adjust the repair strategy and arrange repair tasks according to the duration and severity of the alarm, ensuring that high-priority faults are not ignored, improving the repair efficiency of the overall network, and reducing the impact of system failures.

[0043] The method of obtaining all ports of the network device, executing a fault risk prediction strategy for each port, and predicting a risk value of a port failure includes: Get any port of any network device; Extract the characteristic data of the port, expressed as ,in, Indicates the characteristics of the port; Execute the following formula to calculate the risk value y of the port; ,in, is the activation function of the neural network, W is the weight matrix, and b is the bias term.

[0044] The method of obtaining all ports of the network device, executing a fault risk prediction strategy for each port, and predicting a risk value of a port failure includes: Get the number of all ports, recorded as m; Calculate the binary loss function value of the neural network: Get the current status of each port , where if a data anomaly occurs on the port, =1, otherwise, =0; calculate ,in, is the risk value of port i; Set the loss function threshold; Compare the binary loss function value with the loss function threshold. If the binary loss function value ≧ the loss function threshold, execute the parameter update strategy, which is: calculate , the result is used as the new weight matrix; calculate , the result is used as the new bias term, where is the learning rate, are the parameters of the loss function for the weight matrix and bias term; Calculate the risk value of each port again; If the binary loss function value is less than the loss function threshold, the fault risk prediction strategy is stopped.

[0045] In this embodiment, refer to Figure 3 ,Flowchart for calculating port risk value.

[0046] Through the intelligent risk prediction model, potential high-risk ports can be identified in advance to avoid large-scale network downtime caused by failures. With the predictive ability of neural networks, the system can accurately predict the possibility of failure of each port based on historical data, thereby achieving more accurate preventive maintenance. This method effectively reduces the frequency of unexpected failures and reduces the impact of emergencies on business. By identifying and repairing high-risk ports in advance, the long-term stability of network equipment can be effectively improved, the self-healing ability of the system can be enhanced, and the overall operation and maintenance efficiency can be improved.

[0047] The obtaining of all movable robots, executing the robot deployment strategy, and allocating the movable robots to the ports include: Get all the movable robots and number them from 1 to h; Set deployment values ​​for each port , the deployment value indicates whether to deploy the mobile robot j to port i. If the mobile robot j is deployed to port i, then =1, otherwise, =0, where 1≦j≦h, 1≦i≦m; calculate ,in, represents the straight-line distance from the mobile robot to port i; And stipulate ≧1, where 1≦i≦m, indicating that at least one mobile robot must be deployed at each high-risk port; Regulation ≦1, where 1≦j≦h, indicating that each mobile robot is deployed to at most one port.

[0048] By allocating mobile robots to high-risk ports based on the risk prediction value of the port, a reasonable robot deployment strategy is implemented. A reasonable robot deployment strategy can ensure that high-risk ports are given priority in repair resources, and prevent low-risk ports from occupying limited robot resources. Through the high-risk priority strategy, repair tasks can be concentrated in the most needed places to ensure that the network environment can be restored to normal as soon as possible. By optimizing the allocation of robot resources, the efficiency of resource utilization is improved, robot task conflicts are reduced, and unnecessary resource waste is avoided. In this way, the overall repair efficiency and system stability are greatly enhanced, which can effectively respond to various fault conditions in the network environment and ensure a smoother operation and maintenance process.

[0049] Example 2, refer to Figure 2 , a network fault handling system, comprising: Edge monitoring devices are installed at the ports of network devices to monitor port data in real time, detect anomalies and issue alarms; Centrally monitor equipment, aggregate alarm information from edge devices, schedule repair tasks based on priority, and implement segmented alarm strategies to ensure timely repairs; The fault risk prediction module uses a neural network to predict the port fault risk, determines the high-risk ports based on the predicted values, and dispatches robots for repair; The robot deployment and scheduling module assigns a mobile robot to each port to ensure that high-risk ports are repaired first.

[0050] Embodiment three, a mobile robot, characterized in that it includes: each mobile robot is configured with only one automated manipulator, connected to edge monitoring equipment and centralized monitoring equipment, and when a port with abnormal data appears, the port with abnormal data is located, moved, operated and repaired.

[0051] Embodiment 4, an automated manipulator, characterized in that the automated manipulator adopts a modular design and is pre-installed with a variety of adapters according to the interface types of different network devices; when receiving an instruction for a specific type of fault, the robot autonomously adjusts the shape of the manipulator's end effector to match the port of the device.

[0052] The modular design of the automated manipulator automatically adjusts the end effector shape according to the different device interface types. The modular design of the automated manipulator greatly enhances its adaptability and can be flexibly adjusted according to the needs of different network device interfaces. This design not only reduces maintenance costs, but also improves the versatility of the robot between different devices, ensuring that it can cope with various types of faults. The robot can quickly adapt and complete the repair task after receiving the repair instruction, avoiding the operational errors that may occur in traditional manual repairs. The modular design makes the manipulator more flexible and can automatically adapt to various interfaces, improving the efficiency and accuracy of equipment repair, thereby shortening the fault recovery time and reducing the need for manual intervention.

[0053] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0054] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A network fault handling method, characterized in that: include: Obtain network devices with interfaces in the computer room and use edge monitoring devices to monitor the ports of each network device; The edge monitoring device is installed on the port of the network device and is responsible for collecting data from the port, determining whether the data is abnormal, and issuing an alarm for abnormal data; Set the repair area radius; For any network device, a circle with the location of the network device as the center and the radius of the repair area as the radius is the repair area of ​​the network device; When the edge monitoring device that monitors the network equipment issues an alarm, a mobile robot in the repair area is dispatched to repair the network equipment; The mobile robot is equipped with an automated manipulator for connecting to the port with abnormal data and replacing the port to establish a connection with the upper network device for data transmission; Use a centralized monitoring device to connect all edge monitoring devices. When the edge monitoring device issues an alarm for abnormal data on a port, a segmented alarm strategy is executed to repair the port with abnormal data. For any network device, obtain all ports of the network device, execute the fault risk prediction strategy for each port, and predict the risk value of the port failure; Set the risk threshold to determine whether the port is a high-risk port; When the risk value of a port is greater than or equal to the risk threshold, the port is identified as a high-risk port; When the risk value of a port is less than the risk threshold, the port is identified as a low-risk port; Get all mobile robots, execute robot deployment strategies, and assign mobile robots to ports.

2. A network fault handling method according to claim 1, characterized in that: When the edge monitoring device issues an alarm for abnormal data of the port, a segmented alarm strategy is executed to repair the port with abnormal data, including: The mobile robot that is repairing the port is recorded as the first type of robot; The movable robots waiting for the repair port are recorded as the second type of robots; When the edge monitoring device issues an alarm for abnormal data of the port, the network device where the port is located is obtained, the repair area corresponding to the network device is obtained, and all the second-type robots in the repair area are obtained; Determine whether the number of the second type of robots is 0; When the number of the second type of robots is 0, the edge monitoring device is controlled to continuously alarm; When the number of second-category robots is not 0, any second-category robot is dispatched to repair the port with abnormal data and stop the edge monitoring device from alarming.

3. A network fault handling method according to claim 1, characterized in that: When the edge monitoring device issues an alarm for abnormal data of the port, a segmented alarm strategy is executed to repair the port with abnormal data, and the method further includes: The moment when the edge monitoring device starts to alarm is recorded as the first moment; Set the alarm duration threshold; The time after the first time and the interval of the alarm duration is recorded as the second time; At the second moment, determining whether the edge monitoring device stops alarming; If the edge monitoring device alarms at the second moment, the alarm information is uploaded to the centralized monitoring device, and the alarm information includes the port location and the alarm duration of the edge monitoring device; The centralized monitoring device obtains all uploaded alarm information, and sets priorities for the ports of edge monitoring devices according to the alarm duration in the alarm information from large to small. The longer the alarm duration, the higher the priority. Obtain each port in order from high to low priority, and record the obtained port as the execution port; Obtain the positions of all second-type robots, and calculate the straight-line distance from the position of each second-type robot to the position of the execution port; Obtain the second type of robot with the shortest straight-line distance, which is recorded as the execution robot; The dispatch execution robot repairs the execution port and stops the edge monitoring device from alarming.

4. A network fault handling method according to claim 1, characterized in that: The method of obtaining all ports of the network device, executing a fault risk prediction strategy for each port, and predicting a risk value of a port failure includes: Get any port of any network device; Extract the characteristic data of the port, expressed as ,in, Indicates the characteristics of the port; Execute the following formula to calculate the risk value y of the port; ,in, is the activation function of the neural network, W is the weight matrix, and b is the bias term.

5. A network fault handling method according to claim 4, characterized in that: The method of obtaining all ports of the network device, executing a fault risk prediction strategy for each port, and predicting a risk value of a port failure includes: Get the number of all ports, recorded as m; Calculate the binary loss function value of the neural network: Get the current status of each port , where if a data anomaly occurs on the port, =1, otherwise, =0; calculate ,in, is the risk value of port i; Set the loss function threshold; Compare the binary loss function value with the loss function threshold. If the binary loss function value ≧ the loss function threshold, execute the parameter update strategy, which is: calculate , the result is used as the new weight matrix; calculate , the result is used as the new bias term, where is the learning rate, are the parameters of the loss function for the weight matrix and bias term; Calculate the risk value of each port again; If the binary loss function value is less than the loss function threshold, the fault risk prediction strategy is stopped.

6. A network fault handling method according to claim 1, characterized in that: The obtaining of all movable robots, executing the robot deployment strategy, and allocating the movable robots to the ports include: Get all the movable robots and number them from 1 to h; Number all ports from 1 to m; Set deployment values ​​for each port , the deployment value indicates whether to deploy the mobile robot j to port i. If the mobile robot j is deployed to port i, then =1, otherwise, =0, where 1≦j≦h, 1≦i≦m; calculate ,in, represents the straight-line distance from the mobile robot to port i; And stipulate ≧1, where 1≦i≦m, indicating that at least one mobile robot must be deployed at each high-risk port; Regulation ≦1, where 1≦j≦h, indicating that each mobile robot is deployed to at most one port.

7. A network fault handling system, characterized in that: include: Edge monitoring devices are installed at the ports of network devices to monitor port data in real time, detect anomalies and issue alarms; Centrally monitor equipment, aggregate alarm information from edge devices, schedule repair tasks based on priority, and implement segmented alarm strategies to ensure timely repairs; The fault risk prediction module uses a neural network to predict the port fault risk, determines the high-risk ports based on the predicted values, and dispatches robots for repair; The robot deployment and scheduling module assigns a mobile robot to each port to ensure that high-risk ports are repaired first.

8. A mobile robot, characterized in that: include: Each mobile robot is configured with only one automated manipulator, which is connected to the edge monitoring device and the centralized monitoring device. When a port with abnormal data appears, the port with abnormal data is located, moved, operated and repaired, thereby implementing the network fault handling method described in any one of claims 1 to 6.

9. An automated manipulator, characterized in that The automated manipulator adopts a modular design and is pre-installed with a variety of adapters according to the interface types of different network devices. When receiving an instruction for a specific type of fault, the robot autonomously adjusts the shape of the manipulator's end effector to match the port of the device, thereby implementing the network fault handling method described in any one of claims 1 to 6.

Images