A method and device for self-healing of a gigabit factory network based on SDN, equipment and storage medium

Abstract

This application discloses a self-healing method, apparatus, device, and storage medium for 10 Gigabit Ethernet power plant networks based on SDN, relating to the fields of software-defined networking and industrial internet technology. The method includes: using a pre-deployed edge SDN proxy to perform line-rate parsing of target detection packets of the target 10 Gigabit Ethernet power plant network to obtain target packet parameters, and reporting the target packet parameters to the SDN controller for fault judgment to determine the fault type and corresponding alarm level corresponding to the target packet parameters; extracting real-time feature vectors from the target packet parameters, and determining the similarity value between the real-time feature vectors and fault standard features in a pre-defined fault feature library based on a pre-defined similarity determination method; determining the fault type judgment result corresponding to the target packet parameters based on the similarity value, and calling resources from a pre-defined redundant resource pool to perform network self-healing operations on the target 10 Gigabit Ethernet power plant network. This enables accurate fault location and self-healing of 10 Gigabit Ethernet power plant networks.

Description

Technical Field

[0001] This invention relates to the fields of software-defined networking technology and industrial internet technology, and in particular to a self-healing method, apparatus, device and storage medium for a 10 Gigabit factory network based on SDN. Background Technology

[0002] Currently, 10 Gigabit factory networks face three major pain points that urgently need to be addressed: First, the self-healing capabilities of traditional industrial networks are severely inadequate. Most existing factory networks rely on hardware redundancy (such as dual fiber optic cables and dual switches) to ensure link reliability, but fault detection still primarily relies on manual inspections or simple alarm mechanisms, resulting in network fault recovery times ranging from seconds to minutes. For continuous production industrial environments, such long interruptions can lead to huge economic losses; for example, in automotive welding lines, the loss for every minute of network interruption can exceed ten thousand yuan. Second, redundancy mechanisms under 10 Gigabit bandwidth have poor adaptability. With the development of industrial automation, 10 Gigabit networks transmit a large amount of real-time control data that is extremely sensitive to latency, such as robot motion commands and precision equipment parameters, which typically require network latency below 50 microseconds. However, traditional static redundant path configurations not only easily waste valuable 10 Gigabit bandwidth resources, but more importantly, the latency generated during path switching often exceeds this stringent threshold, making it difficult to simultaneously meet the dual service requirements of high bandwidth and low latency. Finally, the complex industrial environment further exacerbates network instability. The factory contains numerous sources of strong electromagnetic interference, such as frequency converters and motors, as well as frequent movements of equipment like AGV robots, and mixed transmission of multiple services including control signals, video surveillance, and sensor data. In such a complex scenario, a single redundancy strategy or network architecture is insufficient to cover the reliability requirements of all business scenarios and struggles to cope with diverse physical and logical threats.

[0003] Therefore, how to achieve automatic fault detection and self-healing in 10 Gigabit networks is a problem that needs to be solved. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a self-healing method, apparatus, device, and storage medium for 10 Gigabit Ethernet power plant networks based on SDN, capable of accurately locating and self-healing faults in 10 Gigabit Ethernet power plant networks. The specific solution is as follows: In a first aspect, this application discloses a self-healing method for 10 Gigabit Ethernet networks based on SDN, including: The target detection packets of the target 10 Gigabit factory network are parsed at line speed using a pre-deployed edge SDN proxy to obtain target packet parameters, and the target packet parameters are reported to the SDN controller based on the target data reporting method; the target data reporting method is determined based on the data type of the target packet parameters. The SDN controller is used to perform fault judgment on the target packet parameters based on preset operating parameter thresholds, so as to determine the fault type and corresponding alarm level of the target packet parameters; Extract the real-time feature vector from the target message parameters, and determine the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on the preset similarity determination method; Based on the similarity value, the fault type determination result corresponding to the target message parameters is determined, and based on the fault type determination result, resources in the preset redundant resource pool are called to perform network self-healing operation on the target 10 Gigabit factory network; the preset redundant resource pool is a resource pool constructed based on multiple end-to-end transmission paths, network communication equipment and optical fiber physical links.

[0005] Optionally, the step of using a pre-deployed edge SDN proxy to perform line-rate parsing of target detection packets on the target 10 Gigabit factory network to obtain target packet parameters, and reporting the target packet parameters to the SDN controller based on the target data reporting method, includes: The target detection packets of the target 10 Gigabit factory network are obtained by using a pre-deployed edge SDN proxy. The target detection packets are BFD control messages and OAM protocol frames sent in parallel based on a preset period by the 10 Gigabit switch chip detection device to perform full-stack detection on the physical layer, link layer and network layer of the target 10 Gigabit factory network. Based on the edge SDN proxy, the line-speed parsing is performed to extract and verify the core parameters of the packet in order to obtain the target packet parameters; Based on the edge SDN proxy, normal data in the target packet parameters are uploaded to the SDN controller in batches using the OpenFlow protocol, and abnormal data in the target packet parameters and all data within the target detection period are immediately uploaded to the SDN controller.

[0006] Optionally, the step of using the SDN controller and based on preset operating parameter thresholds to perform fault judgment on the target packet parameters, in order to determine the fault type and corresponding alarm level corresponding to the target packet parameters, includes: The operating parameters of the target 10 Gigabit power plant network are collected within a preset time period, and the baseline value and dynamic fluctuation threshold of the operating parameters are determined to obtain the preset operating parameter threshold; the operating parameters are the operating data of the target 10 Gigabit power plant network under fault-free conditions; The SDN controller compares the target message parameters with the preset operating parameter thresholds to determine the fault type and corresponding alarm level corresponding to the target message parameters.

[0007] Optionally, before extracting the real-time feature vector from the target message parameters and determining the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on a preset similarity determination method, the method further includes: Collect network anomaly data in industrial scenarios and classify the network anomaly data into fault types to obtain each fault type; The preset fault feature library is constructed based on each of the fault types, and a standardized feature vector is constructed for each of the fault types. The standardized feature vector is normalized to obtain fault standard features, and corresponding weight coefficient values ​​are assigned to the fault standard features based on the business priority in the industrial scenario.

[0008] Optionally, the step of extracting the real-time feature vector from the target message parameters and determining the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on a preset similarity determination method includes: Features are extracted from the target message parameters and normalized to obtain a real-time feature vector; The weighted Euclidean distance between each fault standard feature in the preset fault feature library and the real-time feature vector is calculated to obtain the similarity value between the real-time feature vector and each fault standard feature.

[0009] Optionally, determining the fault type determination result corresponding to the target message parameters based on the similarity value includes: The confidence level corresponding to the similarity value is determined based on the weighted Euclidean distance calculation method; If the confidence level is greater than the first preset confidence threshold, then the fault type determination result corresponding to the target message parameter is determined based on the fault standard features corresponding to the confidence level; If the confidence level is less than or equal to the first preset confidence threshold and greater than the second preset confidence threshold, then loopback test data, equipment operation logs, and electromagnetic spectrum data are acquired to update the target message parameters, and the process jumps to the step of extracting the real-time feature vector from the target message parameters. If the confidence level is less than or equal to the second preset confidence threshold, the target message parameter is determined to be an unknown fault type.

[0010] Optionally, the step of invoking resources from a preset redundant resource pool to perform network self-healing operations on the target 10 Gigabit factory network based on the fault type determination result includes: If the fault type determination result is a link interruption type fault, then the target backup path is called from the path resource pool in the preset redundant resource pool, so as to perform network self-healing operation on the target 10 Gigabit factory network based on the target backup path. If the fault type determination result is a device overload / traffic congestion type fault, then the low-priority services will be allocated to the backup link, and the traffic scheduling strategy will be adjusted to perform network self-healing operation on the target 10 Gigabit factory network. If the fault type determination result is an electromagnetic interference / link quality degradation fault, the modulation method of the 10 Gigabit port in the region is adjusted, and a network self-healing operation is performed on the target 10 Gigabit factory network based on the forward error correction anti-interference coding algorithm. If the fault type determination result is an unknown fault type, then all services of the faulty node / link will be switched to a preset backup path, and network self-healing operation will be performed on the target 10 Gigabit factory network.

[0011] Secondly, this application discloses a self-healing device for a 10 Gigabit power plant network based on SDN, comprising: The parameter reporting module is used to perform line-rate parsing of target detection packets of the target 10 Gigabit factory network using a pre-deployed edge SDN proxy to obtain target packet parameters, and to report the target packet parameters to the SDN controller based on the target data reporting method; the target data reporting method is determined based on the data type of the target packet parameters. The fault judgment module is used to use the SDN controller and based on preset operating parameter thresholds to judge the fault of the target packet parameters, so as to determine the fault type and corresponding alarm level of the target packet parameters; The similarity value determination module is used to extract the real-time feature vector from the target message parameters and determine the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on the preset similarity determination method. The network self-healing module is used to determine the fault type judgment result corresponding to the target message parameters based on the similarity value, and to call resources in the preset redundant resource pool to perform network self-healing operation on the target 10 Gigabit factory network based on the fault type judgment result; the preset redundant resource pool is a resource pool built based on multiple end-to-end transmission paths, network communication equipment and optical fiber physical links.

[0012] Thirdly, this application discloses an electronic device, including: Memory, used to store computer programs; A processor for executing the computer program to implement the aforementioned SDN-based 10 Gigabit factory network self-healing method.

[0013] Fourthly, this application discloses a computer-readable storage medium for storing a computer program, which, when executed by a processor, implements the aforementioned SDN-based 10 Gigabit Ethernet self-healing method.

[0014] As can be seen, in this application, a pre-deployed edge SDN proxy is used to perform line-rate parsing of target detection packets of the target 10 Gigabit factory network to obtain target packet parameters, and the target packet parameters are reported to the SDN controller based on a target data reporting method; the target data reporting method is determined based on the data type of the target packet parameters; the SDN controller is used to perform fault judgment on the target packet parameters based on preset operating parameter thresholds to determine the fault type and corresponding alarm level corresponding to the target packet parameters; real-time feature vectors are extracted from the target packet parameters, and the similarity value between the real-time feature vectors and the fault standard features in the preset fault feature library is determined based on a preset similarity determination method; the fault type determination result corresponding to the target packet parameters is determined based on the similarity value, and the network self-healing operation of the target 10 Gigabit factory network is performed by calling resources in the preset redundant resource pool based on the fault type determination result; the preset redundant resource pool is a resource pool constructed based on multiple end-to-end transmission paths, network communication equipment, and optical fiber physical links. In other words, based on the SDN architecture, line-speed parsing and intelligent reporting of packets are achieved by deploying agents at the edge. Combined with centralized decision-making by the controller and similarity comparison of the fault feature database, millisecond-level accurate fault detection and location can be realized. Furthermore, by dynamically invoking a multi-layered redundant resource pool covering end-to-end paths, devices, and links through software definition, fault isolation and self-healing can be quickly completed under 10 Gigabit bandwidth. This effectively ensures zero-interruption and low-latency transmission of industrial real-time control data, significantly improving the reliability, continuity, and production availability of the factory network. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0016] Figure 1 This application discloses a flowchart of a self-healing method for a 10 Gigabit factory network based on SDN. Figure 2 This is a schematic diagram of the structure of a 10 Gigabit factory network self-healing device based on SDN disclosed in this application; Figure 3 This is a structural diagram of an electronic device disclosed in this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] Traditional industrial networks rely on hardware redundancy, with fault recovery taking seconds to minutes, making it difficult to meet the stringent microsecond-level latency requirements of control data in 10 Gigabit environments. Furthermore, in the face of complex environments such as electromagnetic interference and equipment movement, a single redundancy strategy cannot guarantee millisecond-level failover and bandwidth efficiency, necessitating a highly reliable, low-latency intelligent redundancy mechanism. Therefore, this application will specifically introduce an SDN-based self-healing method for 10 Gigabit factory networks, which can solve the above problems.

[0019] See Figure 1 As shown in the figure, this application discloses a self-healing method for 10 Gigabit Ethernet networks based on SDN, including: Step S11: Use a pre-deployed edge SDN agent to perform line-rate parsing on the target detection packets of the target 10 Gigabit factory network to obtain the target packet parameters, and report the target packet parameters to the SDN controller based on the target data reporting method; the target data reporting method is determined based on the data type of the target packet parameters.

[0020] In this embodiment, firstly, 10 Gigabit industrial switches (port speed ≥10Gbps) supporting the OpenFlow 1.5 protocol are deployed. The core layer adopts a dual-machine cluster architecture, forming a ring fiber optic backbone link between the core layer and the access layer. The access layer uses a star topology to connect production equipment (such as PLCs, robot controllers, and sensor gateways). Access switches in key production areas use dual uplinks to connect to two core switches respectively. A centralized SDN (Software Defined Network) controller (supporting cluster redundancy) is deployed, which connects to the factory's MES (Manufacturing Execution System) and SCADA (Supervisory Control and Data Acquisition) systems through the northbound interface to obtain real-time data on production business priorities, production cycle time, and equipment operating status. The southbound OpenFlow interface enables unified management, flow table distribution, and status acquisition of all 10 Gigabit switches and edge gateways in the network. Deploy lightweight SDN proxies on 10 Gigabit switches and edge gateways at the access layer to handle real-time collection, preprocessing, and rapid response of local detection data. This enables local pre-handling of emergency faults, reduces transmission latency caused by centralized controller decision-making, and adapts to the low latency requirements of 10 Gigabit industrial networks.

[0021] In this embodiment, the solution employs BFD (Bidirectional Forwarding) The system employs a dual-protocol collaborative detection mechanism combining Detection (bidirectional forwarding detection) and IEEE 802.3ah OAM (operation, maintenance, and management) protocols. This mechanism configures hardware-level detection sessions for 10 Gigabit industrial Ethernet, specifically as follows: Each 10 Gigabit physical link in the network is configured with an independent BFD asynchronous mode session, with each session bound to a specific link port. The core link BFD detection interval is configured at 3.3ms, while non-core links have a 10ms interval. All BFD sessions are offloaded to the 10 Gigabit switch ASIC chip for hardware-level forwarding, eliminating the need for CPU processing and preventing equipment load from affecting detection accuracy. The system also configures OAM protocol functionality for all 10 Gigabit ports, enabling three frame structures: Information OAMPDU, Event Notification OAMPDU, and Loopback Control OAMPDU. The OAM frame packet transmission period is fixed at 500μs, adapting to the millisecond-level degradation detection requirements of the 10 Gigabit physical layer. Finally, based on the factory-standard IEEE 1588v2PTP precise time protocol, time synchronization is achieved for all detection nodes, with a synchronization error ≤1μs. This provides a unified time reference for timestamp and delay calculation of detection messages, avoiding detection deviations caused by time asynchrony.

[0022] In this embodiment, the step of using a pre-deployed edge SDN proxy to perform line-rate parsing of target detection packets from the target 10 Gigabit Ethernet network to obtain target packet parameters, and then reporting the target packet parameters to the SDN controller based on the target data reporting method, includes: obtaining target detection packets from the target 10 Gigabit Ethernet network using a pre-deployed edge SDN proxy; the target detection packets are BFD control messages and OAM protocol frames sent in parallel at a preset period by a 10 Gigabit switch chip detection device to perform full-stack detection of the physical layer, link layer, and network layer of the target 10 Gigabit Ethernet network; extracting and verifying the core parameters of the packets based on the line-rate parsing by the edge SDN proxy to obtain the target packet parameters; and uploading normal data in the target packet parameters to the SDN controller in batches using the OpenFlow protocol based on the edge SDN proxy, and immediately uploading abnormal data in the target packet parameters and all data within the target detection period to the SDN controller. Specifically, after initialization, the 10 Gigabit switch ASIC chip-level detection module sends two types of detection packets in parallel at a preset period to achieve full-stack detection of the physical layer, link layer, and network layer. The messages are as follows: ① BFD control message: encapsulated in a 10 Gigabit Ethernet MAC frame, carrying the session number, transmission sequence number, transmission timestamp, and local port status information, and transmitted at a preset interval at line speed. The receiving end monitors the message reception status in real time and calculates link connectivity and round-trip delay; ② OAM protocol frame: information OAMPDU is sent every 500μs, carrying port optical transmit / receive power, link bit error rate, frame transmit / receive count, and link bandwidth utilization data; event notification OAMPDU is triggered immediately when bit error rate exceeds the limit or frame loss is detected; loopback control OAMPDU is sent on demand and used for link segment fault location. The receiving edge SDN proxy performs line-rate parsing of detection packets, extracts and verifies the core parameters of the packets, and completes the aggregation and preprocessing of detection data every 1ms, eliminating duplicate and erroneous packets. Data reporting adopts a "dual-track system": regular health data is uploaded to the SDN controller in batches every 10ms via Packet-in messages of the OpenFlow protocol; emergency abnormal data (such as continuous frame loss or session timeout) immediately triggers interrupt reporting, while the full data of the most recent 100 detection cycles is cached locally for subsequent fault backtracking analysis.

[0023] Step S12: Use the SDN controller and a preset operating parameter threshold to perform fault judgment on the target packet parameters to determine the fault type and corresponding alarm level of the target packet parameters.

[0024] In this embodiment, the step of using the SDN controller and based on a preset operating parameter threshold to determine the fault type and corresponding alarm level of the target packet parameters includes: collecting the operating parameters of the target 10 Gigabit Ethernet network within a preset time period, and determining the baseline value and dynamic fluctuation threshold of the operating parameters to obtain the preset operating parameter threshold; the operating parameters are the operating data of the target 10 Gigabit Ethernet network in a fault-free state; and using the SDN controller to compare the target packet parameters with the preset operating parameter threshold to determine the fault type and corresponding alarm level of the target packet parameters.

[0025] First, during the network initialization phase, the SDN controller continuously collects full-link operating parameters under 72 hours of fault-free operation, calculates the baseline values ​​and dynamic fluctuation thresholds for each parameter, and defines the core parameter thresholds as follows: ① Link round-trip delay baseline value T0, alarm threshold T_max=T0×1.2, i.e., a sudden increase in delay exceeding 20% ​​triggers an alarm; ② Link bit error rate critical threshold BER_th=1e-12, exceeding this threshold is considered link quality degradation; ③ OAM frame reception rate baseline value R0=100%, minimum threshold R_th=99.99%, a 100% frame loss rate for 3 consecutive periods is considered link interruption; ④ Optical power reception baseline value P0, the normal operating range is [P0-3dBm, P0+3dBm], a sudden drop exceeding 5dBm is considered fiber optic link failure. Then, based on the comparison of collected data and thresholds, alarms are divided into three levels to achieve initial fault screening: ① Level 1 Emergency Alarm: Complete loss of OAM frames for 3 consecutive cycles, timeout of BFD session for 3 consecutive packet transmission cycles, and sudden drop in optical power exceeding 5dBm, directly identified as emergency faults of link interruption / equipment failure, immediately triggering subsequent fault matching procedures; ② Level 2 Important Alarm: Link latency exceeding T_max, bit error rate exceeding BER_th, and frame loss rate exceeding 0.01%, identified as faults of link quality degradation, triggering deep detection and fault diagnosis. Matching; ③ Level 3 early warning alarm: When the link bandwidth utilization exceeds 85% and the device CPU / memory utilization exceeds 90%, it is judged as a congestion / overload fault, triggering traffic monitoring and load balancing prediction; 3) Abnormal data noise filtering: The edge SDN proxy adopts a 5-cycle sliding window filtering mechanism to remove single burst noise data (such as single packet loss, instantaneous latency jitter). Only when the parameters continuously exceed the threshold within 2 consecutive detection cycles is it judged as a valid anomaly and reported to the SDN controller, keeping the false alarm rate within 0.1%. In actual operation, 1) for Level 1 emergency alarms, the loopback OAMPDU hop-by-hop detection is immediately triggered, sending loopback control frames from both ends of the faulty link to the other end, locking the faulty optical cable segment and faulty port segment by segment, with positioning accuracy reaching the single link / single port level; 2) for Level 2 important alarms, the SDN controller retrieves the historical detection data of the faulty link for the most recent 100 cycles, analyzes the parameter degradation trend, and distinguishes between gradual degradation (such as fiber aging) and sudden degradation (such as electromagnetic interference); 3) for Level 3 early warning alarms, service traffic data within the link is collected, distinguishing between high-priority control services and low-priority non-control services, analyzing the source of bandwidth congestion, and providing data support for subsequent load balancing.

[0026] In this way, through the hardware-level dual-protocol collaborative detection mechanism, the fault identification time is shortened to less than 20μs, and the fault location accuracy is ≥99%. Compared with the traditional industrial network SNMP polling detection mechanism (detection cycle ≥30s), the detection efficiency is improved by more than 1.5 million times, which is fully adapted to the low latency and high reliability detection requirements of 10 Gigabit industrial control networks.

[0027] Step S13: Extract the real-time feature vector from the target message parameters, and determine the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on the preset similarity determination method.

[0028] In this embodiment, before extracting the real-time feature vector from the target message parameters and determining the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on a preset similarity determination method, the method further includes: collecting network anomaly data in industrial scenarios and classifying the network anomaly data into fault types; constructing the preset fault feature library based on each fault type and constructing standardized feature vectors for each fault type; normalizing the standardized feature vectors to obtain fault standard features, and assigning corresponding weight coefficient values ​​to the fault standard features based on the service priority in the industrial scenario. That is, a fault feature library for 10 Gigabit factory networks is pre-constructed, covering more than 99% of network anomalies in industrial scenarios, divided into four core fault types. Each fault type corresponds to a unique fault ID, standardized feature vector, and preset hierarchical self-healing strategy. The feature library supports dynamic iterative updates. The core fault classifications are shown in Table 1 below. Table 1 Core Fault Classification Table

[0029] Construction of standardized fault feature vector: Define a 6-dimensional standardized feature vector for each type of fault, complete 0-1 normalization processing, eliminate the influence of dimensions, and the feature vector formula is: F=[f1,f2,f3,f4,f5,f6]; The dimensions are defined as follows: ①f1: Link connectivity characteristic, 0 = complete interruption, 1 = complete normal operation, with the intermediate value representing the degree of link degradation; ②f2: Physical layer optical power characteristic, the deviation of normalized optical power from the baseline value; ③f3: Link bit error rate characteristic, the ratio of normalized real-time bit error rate to the critical threshold; ④f4: Delay jitter characteristic, the deviation of normalized real-time delay from the baseline value; ⑤f5: Equipment load characteristic, the normalized equipment CPU / memory / port bandwidth utilization; ⑥f6: Environmental interference characteristic, the frequency matching degree between normalized electromagnetic spectrum data and typical interference sources. Feature weight configuration: A weight coefficient W=[w1,w2,w3,w4,w5,w6] is set for each feature dimension, with a total weight of 1, dynamically adjusted based on service priority and the degree of fault impact; for example, for high-priority control links, the connectivity characteristic weight w1=0.4, and the optical power characteristic weight w2=0.2, prioritizing the accuracy of link connectivity determination.

[0030] In this embodiment, the step of extracting the real-time feature vector from the target message parameters and determining the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on a preset similarity determination method includes: extracting features from the target message parameters and performing normalization processing to obtain the real-time feature vector; calculating the weighted Euclidean distance between each fault standard feature in the preset fault feature library and the real-time feature vector to obtain the similarity value between the real-time feature vector and each of the fault standard features.

[0031] Real-time anomaly feature extraction: The SDN controller extracts the corresponding real-time feature vector F from the valid anomaly data reported in step 3. real =[f 1real ,f 2real ,f 3real ,f 4real ,f 5real ,f 6real ], complete 0-1 normalization processing to maintain the same dimension as the standard fault feature vector in the feature library; 2) Weighted Euclidean distance similarity calculation: for each standard fault feature vector in the feature library Calculate the weighted Euclidean distance D between the real-time feature vector and the standard feature vector. i The formula for quantifying the similarity between the two is as follows: ; Where j is the feature dimension index, w j D represents the weight coefficient for the corresponding dimension. i The smaller the value, the higher the similarity and matching degree between the real-time abnormal data and the standard fault of that type.

[0032] Step S14: Determine the fault type determination result corresponding to the target message parameters based on the similarity value, and call the resources in the preset redundant resource pool to perform network self-healing operation on the target 10 Gigabit factory network based on the fault type determination result; the preset redundant resource pool is a resource pool constructed based on multiple end-to-end transmission paths, network communication equipment and optical fiber physical links.

[0033] In this embodiment, determining the fault type judgment result corresponding to the target message parameter based on the similarity value includes: determining the confidence level corresponding to the similarity value based on a weighted Euclidean distance calculation method; if the confidence level is greater than a first preset confidence threshold, determining the fault type judgment result corresponding to the target message parameter based on the fault standard features corresponding to the confidence level; if the confidence level is less than or equal to the first preset confidence threshold and greater than a second preset confidence threshold, acquiring loopback test data, equipment operation logs, and electromagnetic spectrum data to update the target message parameter, and jumping to the step of extracting the real-time feature vector from the target message parameter; if the confidence level is less than or equal to the second preset confidence threshold, determining the target message parameter as an unknown fault type. Fault matching confidence calculation: calculating matching confidence based on weighted Euclidean distance. i The similarity is converted into a quantifiable criterion, and the calculation formula is as follows: ; The confidence level ranges from 0 to 100%, with higher values ​​indicating higher accuracy in fault type determination. It's important to note the fault type classification rules: ① When the highest confidence level is reached... max When ≥90%, it is directly identified as the corresponding standard fault type, and the preset self-healing strategy for that fault is immediately triggered; ② When 70%≤Conf max When the accuracy is less than 90%, initiate a second-stage deep detection, supplement loopback test data, equipment operation logs, and electromagnetic spectrum data, update the real-time feature vector, recalculate the matching degree, and complete the accurate fault determination; ③ When Conf max When the failure rate is less than 70%, it is determined to be an unknown fault type, and the fallback redundancy strategy is immediately triggered (full and seamless switching of the primary and backup paths). At the same time, the abnormal feature vector is stored in the sample library to be learned for subsequent feature library iteration and optimization.

[0034] In this embodiment, the step of calling resources from a preset redundant resource pool to perform network self-healing operations on the target 10 Gigabit Ethernet network based on the fault type determination result includes: if the fault type determination result is a link interruption fault, then calling a target backup path from the path resource pool in the preset redundant resource pool to perform network self-healing operations on the target 10 Gigabit Ethernet network based on the target backup path; if the fault type determination result is a device overload / traffic congestion fault, then allocating low-priority services to backup links and adjusting traffic scheduling strategies to perform network self-healing operations on the target 10 Gigabit Ethernet network; if the fault type determination result is an electromagnetic interference / link quality degradation fault, then adjusting the modulation method of the regional 10 Gigabit ports and performing network self-healing operations on the target 10 Gigabit Ethernet network based on a forward error correction anti-interference coding algorithm; if the fault type determination result is an unknown fault type, then switching all services of the faulty node / link to the preset backup path to perform network self-healing operations on the target 10 Gigabit Ethernet network.

[0035] First, it should be noted that this application establishes a multi-layered redundancy resource pool. This includes: constructing a path redundancy resource pool, which involves pre-calculating 3-5 end-to-end primary and backup paths based on the entire network topology using the SDN controller. Path calculation dimensions include shortest path, lowest transmission latency, lowest device load, and least electromagnetic interference coverage area. Each path completes end-to-end bandwidth reservation and flow table pre-configuration, and is stored in the path resource pool; the primary and backup paths achieve complete physical link separation to avoid simultaneous failure of primary and backup paths due to failures in the same cable or trench; and configuring a device redundancy hot standby mechanism, namely: configuring a dual-machine hot standby mechanism for the core switch and SDN controller, with the SDN controller cluster using the Raft consensus algorithm. The system ensures seamless synchronization of current status and fault takeover; the core switch implements virtual gateway redundancy through VRRP (Virtual Router Redundancy Protocol), with a device fault switching time of ≤50ms; key access layer devices are configured with dual power supplies and dual main control modules, providing hardware-level fault redundancy coverage; link redundancy resources are reserved, i.e., dual-fiber physical links are deployed in key production areas (such as semiconductor clean rooms and vehicle welding lines), supporting mixed networking of single-mode / multi-mode fiber types, with ≥50% bandwidth reserved as redundancy resources for each link; the optical power monitoring module collects the optical transmit and receive power of the links in real time, enabling continuous assessment of link health and providing data support for fault prediction.

[0036] Specifically, if the matching result is a link interruption fault: the SDN controller immediately calls the optimal backup path from the path resource pool and quickly switches the switch flow table action bucket through the OpenFlow 1.5 protocol's GroupTable mechanism, without needing to reissue the full flow table, achieving an end-to-end path switching time of ≤8ms. Simultaneously, it notifies edge devices to adjust transmission parameters, ensuring zero packet loss and uninterrupted operation for high-priority services. If the matching result is a device overload / traffic congestion fault: a dynamic load balancing mechanism is triggered. Based on service priority rules, low-priority services (such as video surveillance and operation and maintenance management data) are temporarily migrated to backup links, reserving ≥80% dedicated bandwidth for high-priority control services. At the same time, traffic scheduling strategies are adjusted to achieve link load balancing and prevent further congestion. If the matching result indicates an electromagnetic interference / link quality degradation fault: Dynamically adjust the modulation scheme of the 10 Gigabit ports in the affected area (e.g., switch from PAM4 (Phase Amplitude Modulation) to DMT (Discrete Multi Tone) modulation), initiate the Forward Error Correction (FEC) anti-interference coding algorithm, and simultaneously add a backup path to bypass the interference area for the link. Continuously monitor link quality, and gradually switch back services after the link returns to normal. If the matching result indicates an unknown fault type: Trigger a full redundancy strategy, seamlessly switch all services of the faulty node / link to the preset backup path to ensure uninterrupted production services, and simultaneously initiate fault backtracking analysis to complete the feature database iteration.

[0037] Finally, after the self-healing strategy is executed, the SDN controller monitors the link and service recovery status in real time through the BFD+OAM dual protocol. If the service does not return to normal within 10ms, it immediately triggers the suboptimal backup path switch and generates a fault alarm to notify the operation and maintenance personnel. If the service returns to normal, the fault and self-healing process data are stored in the historical database for subsequent strategy optimization.

[0038] In addition, after fault recovery, the SDN controller performs a full-process retrospective analysis of the fault's occurrence time, fault type, matching process, and self-healing execution effect, extracting core fault features and updating the fault feature library and weight coefficients. For unknown fault types, manual annotation is performed and the fault feature library is added to improve the accuracy of subsequent fault matching. Edge nodes continuously collect data on electromagnetic interference, production business traffic, and link operation status in the factory environment. The cloud platform uses historical fault data to train reinforcement learning models through federated learning, optimizing path weight calculation rules, fault feature matching algorithms, and self-healing strategy libraries. For example, the number of backup paths is increased in high-frequency interference areas, and the bandwidth reservation ratio is dynamically adjusted during peak production periods, so that the system's anti-interference capability and self-healing capability continue to improve over time. During the factory's non-production hours every morning, the SDN controller automatically executes redundancy strategy drills, simulating various fault scenarios such as fiber optic cable breakage, equipment downtime, and electromagnetic interference, verifying the effectiveness of the entire process of fault detection, feature matching, and self-healing switching, and generating drill reports and optimization suggestions. Full network redundancy failover tests are conducted quarterly to ensure that network availability remains stable at over 99.999%.

[0039] As can be seen, in this embodiment, a pre-deployed edge SDN proxy is used to perform line-rate parsing of target detection packets of the target 10 Gigabit factory network to obtain target packet parameters, and the target packet parameters are reported to the SDN controller based on a target data reporting method; the target data reporting method is determined based on the data type of the target packet parameters; the SDN controller is used to perform fault judgment on the target packet parameters based on preset operating parameter thresholds to determine the fault type and corresponding alarm level corresponding to the target packet parameters; real-time feature vectors are extracted from the target packet parameters, and the similarity value between the real-time feature vectors and the fault standard features in the preset fault feature library is determined based on a preset similarity determination method; the fault type determination result corresponding to the target packet parameters is determined based on the similarity value, and the resources in the preset redundant resource pool are called to perform network self-healing operation on the target 10 Gigabit factory network based on the fault type determination result; the preset redundant resource pool is a resource pool built based on multiple end-to-end transmission paths, network communication devices, and optical fiber physical links. In other words, based on the SDN architecture, line-speed parsing and intelligent reporting of packets are achieved by deploying agents at the edge. Combined with centralized decision-making by the controller and similarity comparison of the fault feature database, millisecond-level accurate fault detection and location can be realized. Furthermore, by dynamically invoking a multi-layered redundant resource pool covering end-to-end paths, devices, and links through software definition, fault isolation and self-healing can be quickly completed under 10 Gigabit bandwidth. This effectively ensures zero-interruption and low-latency transmission of industrial real-time control data, significantly improving the reliability, continuity, and production availability of the factory network.

[0040] refer to Figure 2The present application also discloses a self-healing device for a 10 Gigabit power plant network based on SDN, comprising: The parameter reporting module 11 is used to perform line-rate parsing of target detection packets of the target 10 Gigabit factory network using a pre-deployed edge SDN proxy to obtain target packet parameters, and to report the target packet parameters to the SDN controller based on the target data reporting method; the target data reporting method is determined based on the data type of the target packet parameters. The fault judgment module 12 is used to use the SDN controller and based on preset operating parameter thresholds to judge the fault of the target packet parameters, so as to determine the fault type and corresponding alarm level of the target packet parameters; The similarity value determination module 13 is used to extract the real-time feature vector from the target message parameters and determine the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on the preset similarity determination method. The network self-healing module 14 is used to determine the fault type judgment result corresponding to the target message parameters based on the similarity value, and to call resources in the preset redundant resource pool to perform network self-healing operation on the target 10 Gigabit factory network based on the fault type judgment result; the preset redundant resource pool is a resource pool constructed based on multiple end-to-end transmission paths, network communication equipment and optical fiber physical links.

[0041] As can be seen, in this embodiment, based on the SDN architecture, line-speed parsing and intelligent reporting of packets are achieved by deploying agents at the edge. Combined with centralized decision-making by the controller and similarity comparison of the fault feature database, millisecond-level accurate fault detection and location can be achieved. Furthermore, by dynamically invoking a multi-layered redundant resource pool covering end-to-end paths, devices, and links through software definition, fault isolation and self-healing can be quickly completed under 10 Gigabit bandwidth. This effectively ensures zero-interruption and low-latency transmission of industrial real-time control data, significantly improving the reliability, continuity, and production availability of the factory network.

[0042] In some specific embodiments, the parameter reporting module 11 may specifically include: The message acquisition unit is used to acquire the target detection message of the target 10 Gigabit factory network using a pre-deployed edge SDN proxy; the target detection message is a BFD control message and an OAM protocol frame sent in parallel based on a preset period by the 10 Gigabit switch chip detection device to perform full-stack detection on the physical layer, link layer and network layer of the target 10 Gigabit factory network. The parameter extraction unit is used to extract and verify the core parameters of the packet based on the edge SDN proxy to obtain the target packet parameters. The packet uploading unit is used to upload normal data in the target packet parameters to the SDN controller in batches based on the OpenFlow protocol using the edge SDN proxy, and to immediately upload abnormal data in the target packet parameters and all data within the target detection period to the SDN controller.

[0043] In some specific embodiments, the fault determination module 12 may specifically include: The parameter collection unit is used to collect the operating parameters of the target 10 Gigabit factory network within a preset time period, and determine the baseline value and dynamic fluctuation threshold of the operating parameters to obtain the preset operating parameter threshold; the operating parameters are the operating data of the target 10 Gigabit factory network under fault-free conditions; The fault type determination unit is used to compare the target message parameters with the preset operating parameter thresholds using the SDN controller to determine the fault type and corresponding alarm level corresponding to the target message parameters.

[0044] In some specific embodiments, the SDN-based 10 Gigabit factory network self-healing device may further include: The fault classification module is used to collect network anomaly data in industrial scenarios and classify the network anomaly data into fault types. The feature construction module is used to construct the preset fault feature library based on each of the fault types, and to construct a standardized feature vector for each of the fault types. The weight coefficient determination module is used to normalize the standardized feature vector to obtain fault standard features, and assign corresponding weight coefficient values ​​to the fault standard features based on the business priority in the industrial scenario.

[0045] In some specific embodiments, the similarity value determination module 13 may specifically include: The feature normalization unit is used to extract features from the target message parameters and perform normalization processing to obtain a real-time feature vector. The similarity calculation unit is used to calculate the weighted Euclidean distance between each fault standard feature in the preset fault feature library and the real-time feature vector, so as to obtain the similarity value between the real-time feature vector and each fault standard feature.

[0046] In some specific embodiments, the network self-healing module 14 may specifically include: A confidence determination unit is used to determine the confidence level corresponding to the similarity value based on a weighted Euclidean distance calculation method. The determination result unit is used to determine the fault type determination result corresponding to the target message parameter based on the fault standard features corresponding to the confidence level if the confidence level is greater than the first preset confidence level threshold. The step jump unit is used to obtain loopback test data, equipment operation logs, and electromagnetic spectrum data if the confidence level is less than or equal to the first preset confidence threshold and greater than the second preset confidence threshold, so as to update the target message parameters and jump to the step of extracting the real-time feature vector in the target message parameters. The parameter determination unit is used to determine the target message parameter as an unknown fault type if the confidence level is less than or equal to the second preset confidence threshold.

[0047] In some specific embodiments, the network self-healing module 14 may specifically include: The first network self-healing unit is used to call the target backup path from the path resource pool in the preset redundant resource pool if the fault type determination result is a link interruption type fault, so as to perform network self-healing operation on the target 10 Gigabit factory network based on the target backup path. The second network self-healing unit is used to allocate low-priority services to backup links and adjust the traffic scheduling strategy to perform network self-healing operations on the target 10 Gigabit factory network if the fault type determination result is a device overload / traffic congestion type fault. The third network self-healing unit is used to adjust the modulation mode of the regional 10 Gigabit port and perform network self-healing operation on the target 10 Gigabit factory network based on the forward error correction anti-interference coding algorithm if the fault type determination result is an electromagnetic interference / link quality degradation fault. The fourth network self-healing unit is used to switch all services of the faulty node / link to a preset backup path and perform network self-healing operation on the target 10 Gigabit factory network if the fault type determination result is an unknown fault type.

[0048] Furthermore, embodiments of this application also disclose an electronic device, Figure 3 This is a structural diagram of an electronic device 20 according to an exemplary embodiment. The content of the diagram should not be construed as limiting the scope of this application.

[0049] Figure 3This is a schematic diagram of the structure of an electronic device 20 provided in an embodiment of this application. Specifically, the electronic device 20 may include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input / output interface 25, and a communication bus 26. The memory 22 stores a computer program, which is loaded and executed by the processor 21 to implement the relevant steps in the SDN-based 10 Gigabit Ethernet self-healing method disclosed in any of the foregoing embodiments. Alternatively, the electronic device 20 in this embodiment may specifically be a computer.

[0050] In this embodiment, the power supply 23 is used to provide operating voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows can be any communication protocol applicable to the technical solution of this application, and is not specifically limited here; the input / output interface 25 is used to acquire external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs, and is not specifically limited here.

[0051] In addition, the memory 22, as a carrier for resource storage, can be a read-only memory, random access memory, disk or optical disk, etc. The resources stored thereon can include operating system 221, computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0052] The operating system 221 is used to manage and control the various hardware devices on the electronic device 20 and the computer program 222, which may be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program capable of performing the SDN-based 10 Gigabit factory network self-healing method executed by the electronic device 20 as disclosed in any of the foregoing embodiments, the computer program 222 may further include computer programs capable of performing other specific tasks.

[0053] Furthermore, this application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, it implements the aforementioned SDN-based 10 Gigabit Ethernet self-healing method. Specific steps of this method can be found in the corresponding content disclosed in the foregoing embodiments, and will not be repeated here.

[0054] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section.

[0055] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0056] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0057] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0058] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A self-healing method for 10 Gigabit factory networks based on SDN, characterized in that, include: The target detection packets of the target 10 Gigabit factory network are parsed at line speed using a pre-deployed edge SDN proxy to obtain target packet parameters, and the target packet parameters are reported to the SDN controller based on the target data reporting method; the target data reporting method is determined based on the data type of the target packet parameters. The SDN controller is used to perform fault judgment on the target packet parameters based on preset operating parameter thresholds, so as to determine the fault type and corresponding alarm level of the target packet parameters; Extract the real-time feature vector from the target message parameters, and determine the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on the preset similarity determination method; Based on the similarity value, the fault type determination result corresponding to the target message parameter is determined, and based on the fault type determination result, resources in the preset redundant resource pool are called to perform network self-healing operation on the target 10 Gigabit factory network. The preset redundant resource pool is a resource pool constructed based on multiple end-to-end transmission paths, network communication devices, and optical fiber physical links.

2. The self-healing method for 10 Gigabit factory networks based on SDN according to claim 1, characterized in that, The step of using a pre-deployed edge SDN proxy to perform line-rate parsing of target detection packets on the target 10 Gigabit factory network to obtain target packet parameters, and then reporting the target packet parameters to the SDN controller based on the target data reporting method, includes: The target detection packets of the target 10 Gigabit factory network are obtained by using a pre-deployed edge SDN proxy. The target detection packets are BFD control messages and OAM protocol frames sent in parallel based on a preset period by the 10 Gigabit switch chip detection device to perform full-stack detection on the physical layer, link layer and network layer of the target 10 Gigabit factory network. Based on the edge SDN proxy, the line-speed parsing is performed to extract and verify the core parameters of the packet in order to obtain the target packet parameters; Based on the edge SDN proxy, normal data in the target packet parameters are uploaded to the SDN controller in batches using the OpenFlow protocol, and abnormal data in the target packet parameters and all data within the target detection period are immediately uploaded to the SDN controller.

3. The self-healing method for 10 Gigabit factory networks based on SDN according to claim 1, characterized in that, The step of using the SDN controller and based on preset operating parameter thresholds to perform fault judgment on the target packet parameters, in order to determine the fault type and corresponding alarm level corresponding to the target packet parameters, includes: The operating parameters of the target 10 Gigabit power plant network are collected within a preset time period, and the baseline value and dynamic fluctuation threshold of the operating parameters are determined to obtain the preset operating parameter threshold; the operating parameters are the operating data of the target 10 Gigabit power plant network under fault-free conditions; The SDN controller compares the target message parameters with the preset operating parameter thresholds to determine the fault type and corresponding alarm level corresponding to the target message parameters.

4. The self-healing method for 10 Gigabit factory networks based on SDN according to claim 1, characterized in that, Before extracting the real-time feature vector from the target message parameters and determining the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on the preset similarity determination method, the method further includes: Collect network anomaly data in industrial scenarios and classify the network anomaly data into fault types to obtain each fault type; The preset fault feature library is constructed based on each of the fault types, and a standardized feature vector is constructed for each of the fault types. The standardized feature vector is normalized to obtain fault standard features, and corresponding weight coefficient values ​​are assigned to the fault standard features based on the business priority in the industrial scenario.

5. The self-healing method for 10 Gigabit factory networks based on SDN according to claim 1, characterized in that, The step of extracting the real-time feature vector from the target message parameters and determining the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on a preset similarity determination method includes: Features are extracted from the target message parameters and normalized to obtain a real-time feature vector; The weighted Euclidean distance between each fault standard feature in the preset fault feature library and the real-time feature vector is calculated to obtain the similarity value between the real-time feature vector and each fault standard feature.

6. The self-healing method for 10 Gigabit factory networks based on SDN according to claim 1, characterized in that, The step of determining the fault type judgment result corresponding to the target message parameters based on the similarity value includes: The confidence level corresponding to the similarity value is determined based on the weighted Euclidean distance calculation method; If the confidence level is greater than the first preset confidence threshold, then the fault type determination result corresponding to the target message parameter is determined based on the fault standard features corresponding to the confidence level; If the confidence level is less than or equal to the first preset confidence threshold and greater than the second preset confidence threshold, then loopback test data, equipment operation logs, and electromagnetic spectrum data are acquired to update the target message parameters, and the process jumps to the step of extracting the real-time feature vector from the target message parameters. If the confidence level is less than or equal to the second preset confidence threshold, the target message parameter is determined to be an unknown fault type.

7. The self-healing method for 10 Gigabit factory networks based on SDN according to claim 1, characterized in that, The step of invoking resources from a preset redundant resource pool to perform network self-healing operations on the target 10 Gigabit factory network based on the fault type determination result includes: If the fault type determination result is a link interruption type fault, then the target backup path is called from the path resource pool in the preset redundant resource pool, so as to perform network self-healing operation on the target 10 Gigabit factory network based on the target backup path. If the fault type determination result is a device overload / traffic congestion type fault, then the low-priority services will be allocated to the backup link, and the traffic scheduling strategy will be adjusted to perform network self-healing operation on the target 10 Gigabit factory network. If the fault type determination result is an electromagnetic interference / link quality degradation fault, the modulation method of the 10 Gigabit port in the region is adjusted, and a network self-healing operation is performed on the target 10 Gigabit factory network based on the forward error correction anti-interference coding algorithm. If the fault type determination result is an unknown fault type, then all services of the faulty node / link will be switched to a preset backup path, and network self-healing operation will be performed on the target 10 Gigabit factory network.

8. A self-healing device for a 10 Gigabit factory network based on SDN, characterized in that, include: The parameter reporting module is used to perform line-rate parsing of target detection packets of the target 10 Gigabit factory network using a pre-deployed edge SDN proxy to obtain target packet parameters, and to report the target packet parameters to the SDN controller based on the target data reporting method; the target data reporting method is determined based on the data type of the target packet parameters. The fault judgment module is used to use the SDN controller and based on preset operating parameter thresholds to judge the fault of the target packet parameters, so as to determine the fault type and corresponding alarm level of the target packet parameters; The similarity value determination module is used to extract the real-time feature vector from the target message parameters and determine the similarity value between the real-time feature vector and the fault standard features in the preset fault feature library based on the preset similarity determination method. The network self-healing module is used to determine the fault type judgment result corresponding to the target message parameters based on the similarity value, and to call resources in the preset redundant resource pool to perform network self-healing operation on the target 10 Gigabit factory network based on the fault type judgment result. The preset redundant resource pool is a resource pool constructed based on multiple end-to-end transmission paths, network communication devices, and optical fiber physical links.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing the computer program to implement the SDN-based 10 Gigabit factory network self-healing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, Used to store a computer program, which, when executed by a processor, implements the SDN-based 10 Gigabit factory network self-healing method as described in any one of claims 1 to 7.

Images