Group obstacle identification method and device, and electronic device

Abstract

The application discloses a group obstacle identification method and device and electronic equipment. The method comprises the following steps: acquiring alarm information in a preset time period; determining a device address corresponding to the alarm information, and determining an uplink network device corresponding to the device address according to a device graph, wherein the device graph is used to represent the association relationship between each network device; determining the alarm state of the uplink network device according to an alarm graph, and determining a fault device from the uplink network device according to the alarm state, wherein the alarm graph is used to represent the association relationship between each alarm event; determining the association rule of the fault device in the alarm process through a fault identification model, and determining a group obstacle identification result according to the association rule. The application solves the technical problem that the group obstacle identification method in the related art is inefficient, especially when a large number of device alarms and user reports are faced, it is difficult to quickly and accurately locate the fault cause, and the timely discovery and processing of network faults are affected.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication, in particular to a group fault identification method and device and electronic equipment. BACKGROUND

[0002] With the widening of network coverage and the surge in the number of users, the complexity and operation and maintenance challenges of the network gradually increase. Especially the emergence of group faults (group obstacles) often involves multiple devices simultaneously alarming or a large number of users simultaneously reporting service interruption, which not only tests the robustness of the network, but also puts higher requirements on the operation and maintenance capabilities of the operator.

[0003] However, in the current network operation and maintenance field, the group obstacle identification method in the related art has significant deficiencies in efficiency, accuracy, response speed and resource utilization. For example, when dealing with a large number of alarms, the group obstacle identification method relying on manual analysis not only takes a long time, but also easily misses key information, and in the face of complex network structure and unknown fault modes, may lead to misjudgment or omission. At the same time, from receiving the alarm to formally confirming the group obstacle, and then locating the fault root cause, the process may take several hours or even longer, resulting in prolonged service recovery time and impaired user experience. In addition, ineffective fault location and repair attempts will waste valuable operation and maintenance resources and affect the efficiency of subsequent fault handling.

[0004] In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0005] The embodiments of the present application provide a group obstacle identification method, device and electronic equipment to at least solve the technical problem that the group obstacle identification method in the related art is inefficient, especially when facing a large number of device alarms and user reports, it is difficult to quickly and accurately locate the fault cause, affecting the timely discovery and handling of network faults.

[0006] According to an aspect of an embodiment of the present application, a group obstacle identification method is provided, comprising: acquiring alarm information in a preset time period, wherein the alarm information is used to prompt that a network device is in an abnormal state; determining a device address corresponding to the alarm information, and determining an uplink network device corresponding to the device address according to a device graph, wherein the device graph is used to represent the association relationship between each network device; determining an alarm state of the uplink network device according to an alarm graph, and determining a fault device from the uplink network device according to the alarm state, wherein the alarm graph is used to represent the association relationship between each alarm event; determining an association rule of the fault device in the alarm process through a fault identification model, and determining a group obstacle identification result according to the association rule, wherein the association rule is used to reflect the fault propagation mode and historical processing efficiency of the fault device.

[0007] Optionally, the alarm information in the preset time period is acquired through a preset calling mechanism from the operation and maintenance guarantee center, wherein the operation and maintenance guarantee center stores the device graph, the alarm graph and the operation and maintenance graph, and the operation and maintenance graph is used to provide repair guidance for the faulty device.

[0008] Optionally, the method further comprises: converting the group obstacle identification result into a preset format, and providing a preplan document link corresponding to the group obstacle identification result, wherein the preplan document link contains the fault handling measures determined according to the operation and maintenance graph.

[0009] Optionally, the device graph is determined by: acquiring link data of all network devices, and determining an initial device graph according to the link data, wherein the link data includes device information of the network devices and link information between the network devices; acquiring local station room data and power environment data of the network devices, wherein the local station room data is used to represent local station information and room information with explicit geographic coordinates and type attributes, and the power environment data is used to represent environment monitoring data in the local station and the room; and optimizing the initial device graph according to the local station room data and the power environment data to obtain the device graph.

[0010] Optionally, the device graph is constructed by taking a resource object corresponding to a network device as a node and a logical relationship between the resource objects as an edge, wherein the logical relationship at least includes a spatial position relationship, a physical connection relationship and a bearing relationship between the resource objects.

[0011] Optionally, the alarm graph is constructed by taking an alarm event corresponding to a network device as a node and a causal relationship between the alarm events as an edge.

[0012] Optionally, the fault identification model is trained by: acquiring historical alarm data and historical work order data; performing feature extraction on the historical alarm data and the historical work order data to obtain a first feature sequence corresponding to the historical alarm data and a second feature sequence corresponding to the historical work order data, wherein the first feature sequence includes a time sequence feature and a spatial distribution feature corresponding to the historical alarm data, and the second feature sequence is used to reflect a relationship between historical alarm events and historical fault handling measures; and training an initial model in a target application scenario according to the first feature sequence and the second feature sequence to obtain the fault identification model, wherein the target application scenario is a cluster offline fault scenario of an optical line terminal.

[0013] Optionally, the association rule of the faulty device in the alarm process includes: a frequent time period, a duration and a collinear mode with other network events of various fault types of the faulty device, wherein the fault types at least include a cluster offline fault of the optical line terminal.

[0014] According to another aspect of the embodiments of the present application, a group fault identification device is also provided, comprising: an acquisition module configured to acquire alarm information in a preset time period, wherein the alarm information is used to prompt that a network device is in an abnormal state; a first determination module configured to determine a device address corresponding to the alarm information, and determine an upstream network device corresponding to the device address according to a device graph, wherein the device graph is used to represent an association relationship between each network device; a second determination module configured to determine an alarm state of the upstream network device according to an alarm graph, and determine a fault device from the upstream network device according to the alarm state, wherein the alarm graph is used to represent an association relationship between each alarm event; and a third determination module configured to determine an association rule of the fault device in an alarm process by using a fault identification model, and determine a group fault identification result according to the association rule, wherein the association rule is used to reflect a fault propagation mode and a historical processing efficiency of the fault device.

[0015] According to still another aspect of the embodiments of the present application, an electronic device is also provided, comprising: a memory and a processor, wherein the memory is configured to store program instructions; and the processor is connected with the memory and is configured to execute the group fault identification method.

[0016] According to still another aspect of the embodiments of the present application, a non-volatile storage medium is also provided, comprising a stored computer program, wherein a device in which the non-volatile storage medium is located executes the group fault identification method by running the computer program.

[0017] According to still another aspect of the embodiments of the present application, a computer program product is also provided, comprising computer instructions, which, when executed by a processor, implement the group fault identification method.

[0018] In the embodiment of the present application, the alarm information in a preset time period is acquired, wherein the alarm information is used to prompt that the network device is in an abnormal state; the device address corresponding to the alarm information is determined, and the uplink network device corresponding to the device address is determined according to the device graph, wherein the device graph is used to represent the association relationship between the network devices; the alarm state of the uplink network device is determined according to the alarm graph, and the fault device is determined from the uplink network device according to the alarm state, wherein the alarm graph is used to represent the association relationship between the alarm events; the association rule of the fault device in the alarm process is determined through the fault identification model, and the group obstacle identification result is determined according to the association rule, wherein the association rule is used to reflect the fault propagation mode and the historical processing efficiency of the fault device, so as to achieve the purpose of quickly and accurately identifying the network group obstacle and its influence range, thereby realizing the technical effects of improving the network operation efficiency, shortening the fault recovery time and improving the service quality, and further solving the technical problems that the group obstacle identification method in the related art is inefficient, especially when a large number of device alarms and user reports are faced, it is difficult to quickly and accurately locate the fault cause, and the network fault is affected by the timely discovery and processing. BRIEF DESCRIPTION OF DRAWINGS

[0019] The drawings described herein are used to provide further understanding of the present application, and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application, and do not constitute an improper limitation on the present application. In the drawings:

[0020] Figure 1 Fig. 1 is a hardware structure diagram of a computer terminal for implementing a group obstacle identification method according to an embodiment of the present application;

[0021] Figure 2 Fig. 2 is a flowchart of a group obstacle identification method according to an embodiment of the present application;

[0022] Figure 3 Fig. 3 is an architecture schematic diagram of a group obstacle identification system according to an embodiment of the present application;

[0023] Figure 4 Fig. 4 is a structure diagram of a group obstacle identification device according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0025] It should be noted that the terms "first", "second", and the like in the description and in the claims of the present application and the above-mentioned accompanying drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but can include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] First, some of the nouns or terms that appear in the process of explaining the embodiments of the present application are applicable to the following explanations:

[0027] OLT (Optical Line Terminal): The core equipment in the fiber access network, responsible for converting the signal from the user side into a format suitable for fiber transmission, and then sending it to the network center or upper layer equipment. At the same time, the OLT can also convert the data signal from the center into a format that the user side can receive, realizing two-way data transmission. In the FTTH (Fiber To The Home) or FTTC (Fiber To The Curb) network, the OLT plays the role of a gateway, connecting the fiber distribution network and the local area network.

[0028] OTN (Optical Transport Network): A high-bandwidth, long-distance transmission network based on WDM (Wavelength Division Multiplexing) technology. OTN is mainly responsible for efficiently transmitting large-particle services such as high-speed Ethernet, SDH signals, etc. on fiber media, and provides strong network protection and recovery capabilities to ensure the stability of data transmission and service quality.

[0029] BAS (Broadband Access Server): A process used to manage and control broadband user access networks, the main functions include user authentication, authorization and charging (AAA service). The BAS device is usually located at the edge of the network, connected with the user terminal (such as a modem), responsible for processing user data packets, sending them to the core network or the Internet.

[0030] Function Calling: Function calling refers to the ability of a function to call another function to perform a specific task in software development. In the field of AI, especially in large language models, Function Calling allows the model to call external functions or services when generating responses, such as querying or updating databases, performing specific calculations, or obtaining real-time data, thereby enhancing the interactivity and practicality of the model.

[0031] Knowledge Graph: A data structure that represents entities and their relationships in a graphical manner, aiming to simulate the knowledge network in human cognition. In communication network operation and maintenance, the knowledge graph can include device graph, alarm graph, and operation and maintenance graph, etc., which are used to record the attributes, location, connection relationship of network devices, as well as historical fault data and processing experience, thereby assisting fault diagnosis and problem solving.

[0032] Alarm Graph: A type of knowledge graph specific to network operation and maintenance, which records alarm information in the network, including the type, level, time, location, description of the alarm, and their causal relationships. The construction and update of the alarm graph help to quickly locate the root cause of the fault and reduce the group obstacle response time.

[0033] Device Graph: A knowledge graph that shows the internal logical relationships between various resource objects in the network, including spatial location, physical connection, logical connection, and bearing relationship, etc. The construction of the device graph is crucial for understanding the network structure and resource distribution, and helps with fault analysis and resource scheduling.

[0034] Operation and Maintenance Graph: Established through manufacturer manuals and fixed rules, it records the operation and maintenance strategies, operation manuals, and emergency plans of various devices, which are used to guide the decision-making and actions of operation and maintenance personnel when handling faults. The operation and maintenance graph can recommend repair schemes according to fault types, improving operation and maintenance efficiency.

[0035] To solve the problem of poor group obstacle identification efficiency in related technologies, the embodiment of the present application provides a group obstacle identification method, which can run in Figure 1 The computer terminal is illustrated as follows.

[0036] The group obstacle identification method embodiment provided by the present application can be executed in a mobile terminal, a computer terminal, or a similar computing device. Figure 1 A hardware structure block diagram of a computer terminal for implementing the group obstacle identification method is shown. As Figure 1As shown, the computer terminal 10 may include one or more processors (shown as 102a, 102b, ..., 102n in the figure) (the processor may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.), a memory 104 for storing data, and a transmission module 106 for communication functions connected via wired and / or wireless networks. In addition, it may also include: a display, a keyboard, a cursor control device, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the I / O interface), a network interface, and a BUS bus. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, computer terminal 10 may also include... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0037] It should be noted that the aforementioned one or more processors and / or other data processing circuits are generally referred to herein as "data processing circuits". These data processing circuits may be implemented wholly or partially as software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuits may be a single, independent processing module, or may be wholly or partially integrated into any other element in the computer terminal 10. As involved in the embodiments of this application, the data processing circuits serve as processor control (e.g., selection of a variable resistor termination path connected to an interface).

[0038] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the group obstacle identification method in this embodiment. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory 104, thereby realizing the aforementioned group obstacle identification method. The memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor, and these remote memories can be connected to the computer terminal 10 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0039] The transmission module 106 is configured to receive or send data via a network. The network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission module 106 includes a network interface controller (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In one example, the transmission module 106 can be a radio frequency (RF) module that is configured to communicate with the Internet wirelessly.

[0040] The display can be a liquid crystal display (LCD) that is touch screen type, for example, which can enable a user to interact with a user interface of the computer terminal 10.

[0041] It should be noted that, in some optional embodiments, the above Figure 1 The computer terminal shown can include hardware elements (including circuitry), software elements (including computer code stored on a computer readable medium), or a combination of both hardware and software elements. It should be noted that, Figure 1 is merely one example of a particular implementation and is intended to illustrate the types of components that can be present in the computer terminal described above.

[0042] In the operating environment described above, an embodiment of the present application provides a group obstacle identification method. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0043] Figure 2 is a flowchart of a group obstacle identification method according to an embodiment of the present application, as shown, the method includes the following steps: Figure 2

[0044] In step S202, alarm information in a preset time period is obtained, wherein the alarm information is used to prompt that a network device is in an abnormal state.

[0045] In step S202, the alarm information contains real-time feedback that a network device is in an abnormal state, such as OLT off-network alarm, PON port receiving no light, dynamic system alarm (such as battery hidden danger, power failure, high temperature, water immersion, smoke sensing, etc.), wave division system failure, OLT uplink unilateral alarm, OLT uplink device alarm (SW, BAS), and device link state abnormality, etc. OLT associated alarm, which is the starting point of group obstacle identification.

[0046] ​Each alarm message includes real-time alarm event information (such as alarm ID, alarm type, alarm level, alarm time, alarm location, and alarm description) and real-time status information (device operating status, device performance indicators, and link status), which are used for real-time analysis and alarm generation.

[0047] Step S204: Determine the device address corresponding to the alarm information, and determine the uplink network device corresponding to the device address based on the device map, wherein the device map is used to represent the association relationship between various network devices.

[0048] In step S204 above, by analyzing the obtained alarm information, device addresses, such as the IP address of the OLT device, can be identified. Subsequently, uplink tracing is performed based on the device graph to identify all uplink network devices associated with that device address. The device graph is a comprehensive knowledge graph that records in detail the physical and logical connections between devices in the network, including but not limited to connections between city / county / district, central office, data center, OLT, SW, BAS, OTN, and other devices, as well as their configuration information and geographical layout. Through the device graph, the uplink structure of the alarming device can be quickly located, thereby pinpointing potential fault locations.

[0049] Step S206: Determine the alarm status of the uplink network devices based on the alarm map, and identify the faulty devices from the uplink network devices based on the alarm status. The alarm map is used to represent the correlation between various alarm events.

[0050] In step S206 above, by further analyzing the real-time alarm status of the uplink network devices and using the alarm graph, faulty devices experiencing faults can be identified. The alarm graph is another type of knowledge graph; it not only records each alarm event but also displays the causal relationships and time-series characteristics between these events. By analyzing the alarm graph, the model can understand how faults propagate between different devices, helping to accurately determine which uplink devices are the direct source of the fault, laying the foundation for subsequent fault root cause localization.

[0051] Step S208: Determine the correlation pattern of faulty equipment in the alarm process through the fault identification model, and determine the group fault identification result based on the correlation pattern. The correlation pattern is used to reflect the fault propagation mode and historical processing efficiency of the faulty equipment.

[0052] In step S208, the pre-trained fault identification model can be used to analyze the correlation rules of the faulty device in the alarm process. These correlation rules integrate the fault propagation mode and the historical processing efficiency, which can help the model understand how a specific fault spreads in the device and network, and how a similar fault was solved in the past. Based on these correlation rules, the model can comprehensively determine whether a group fault has occurred, the impact range of the group fault, and the potential root cause, forming a group fault identification result.

[0053] Through steps S202 to S208, the purpose of quickly and accurately identifying network group faults and their impact range is achieved, thereby achieving the technical effects of improving network operation efficiency, shortening fault recovery time, and improving service quality, thereby solving the technical problems of low efficiency of group fault identification methods in related technologies, especially when facing a large number of device alarms and user reports, it is difficult to quickly and accurately locate the fault cause, affecting the timely discovery and processing of network faults. The following will be described in detail.

[0054] In step S202, the alarm information in the preset time period is obtained, including: obtaining the alarm information in the preset time period from the operation and maintenance guarantee center through a preset calling mechanism, wherein the operation and maintenance guarantee center stores a device graph, an alarm graph, and an operation and maintenance graph, and the operation and maintenance graph is used to provide repair guidance for a faulty device.

[0055] In the embodiments of the present application, in order to monitor the network health state in real time, the Function Calling mechanism (i.e. the above-mentioned preset calling mechanism) can be used to give the large model the ability to dynamically query the current state of the network device. For example, when receiving a warning signal such as multiple OLT uplink abnormalities, the model immediately starts Function Calling and sends a query request to the network management system to collect real-time alarm information of key devices including but not limited to OLT, switch, wavelength division multiplexing system. This process is not limited to top-level devices, but also penetrates down to lower-level network components to ensure the comprehensiveness of the evaluation.

[0056] Figure 3 is an architecture schematic diagram of a group fault identification system according to an embodiment of the present application. As shown in Figure 3As shown, the group fault identification system includes three core components: a network operation platform 30 (control platform), a network management system 32 (resource system), and a smart operation and maintenance guarantee center 34 (referred to as operation and maintenance center), and a resource relational database 36 supporting the entire system. Among them, the operation and maintenance center 34 is a newly added processing center in the group fault identification system. The operation and maintenance center includes: an alarm graph construction and update module 302, a device graph construction and update module 304, an operation and maintenance graph construction and update module 306, a fault identification model 308, a root cause positioning module 310, a repair suggestion (intelligent) recommendation model 312, and a repair suggestion pushing module 314.

[0057] Specifically, first, a device alarm is generated on the network operation platform 30, and real-time alarm information is provided to the operation and maintenance center 34 by the network management system 32. The resource relational database 36 provides detailed device resource information, which is the cornerstone of device graph construction. The operation and maintenance center, as the intelligent brain of the system, has an alarm graph construction and update module 302 and a device graph construction and update module 304 below it, respectively responsible for constructing and maintaining dynamically updated graphs according to real-time alarm information and device resource information; the fault identification model 308 analyzes the alarm information in depth according to the constructed alarm graph and device graph, combined with the training ability of the large model, identifies group fault events and assesses potential risks; the root cause positioning module 310 further analyzes the fault root cause to provide accurate guidance for fault handling; the operation and maintenance graph construction module 308 integrates operation manuals and rules to establish a knowledge framework for fault repair; the repair suggestion pushing module 314 and the repair suggestion recommendation model 312 generate and push customized repair suggestions based on fault types, root cause positioning, and operation and maintenance graphs, ensuring the efficiency and intelligence of fault handling. Through close data interaction and logical analysis, the entire system builds a comprehensive and intelligent operation and maintenance guarantee system covering fault prediction, identification, positioning, and repair guidance, significantly improving the efficiency and accuracy of network operation.

[0058] In the above step S204, the device graph is determined by: obtaining link data of all network devices, and determining an initial device graph according to the link data, wherein the link data includes device information of the network devices and link information between the network devices; obtaining local station room data and power environment data of the network devices, wherein the local station room data is used to represent local station information and room information with clear geographic coordinates and type attributes, and the power environment data is used to represent environment monitoring data in the local station and the room; and optimizing the initial device graph according to the local station room data and the power environment data to obtain the device graph.

[0059] Optionally, the device graph is constructed with resource objects corresponding to network devices as nodes and logical relationships between the resource objects as edges, wherein the logical relationships at least include spatial position relationships, physical connection relationships and bearing relationships between the resource objects.

[0060] In the embodiments of the present application, the device graph not only reflects the direct physical and logical connections between network devices, but also comprehensively considers the environmental factors and the position relationships of the devices in the global network, and is a multi-dimensional and multi-level knowledge graph. The device graph is constructed in the following manner:

[0061] Firstly, link data of all network devices is collected, which includes not only detailed information of the devices (such as device ID, device type, device model, device location, etc.), but also physical and logical connection information between the devices (such as link ID, start device ID, end device ID, port number, transmission rate and transmission medium, etc.). By integrating these link data, the system can construct the basic connection structure of the devices in the network and form an initial device graph, laying a foundation for subsequent fault location and root cause analysis.

[0062] Secondly, in addition to the link information of the devices, detailed data of stations and machine rooms, as well as power environment data in these places, need to be collected. The station and machine room data provides the geographical position and type attributes of the devices, such as station ID, machine room ID, geographical position (latitude and longitude) of the station and machine room, environmental monitoring data (such as temperature, humidity, power state, etc.); the power environment data covers key information such as temperature, humidity, power supply state, etc., such as power supply state (such as the state of main power and backup power), air conditioning system state (such as the running state and parameters of air conditioning equipment) and environmental alarm data (such as alarm information of the power environment system, such as power failure, abnormal temperature and humidity, etc.). These data can help the system understand the environmental conditions of the devices and how these conditions affect the normal operation of the devices.

[0063] Finally, the station and machine room data and the power environment data are integrated on the basis of the initial device graph to form the final device graph. This operation not only enhances the geographical information of the device graph, making the geographical redundancy and potential regional risks of the network layout intuitively displayed, but also supplements the analysis dimension of the influence of environmental factors on the devices. The optimized device graph can more comprehensively reflect the spatial position relationships, physical connection relationships and bearing relationships between the network devices, providing more abundant information for fault analysis.

[0064] It should be noted that in the device graph construction, the resource objects corresponding to the network devices are used as nodes in the graph, where the resource objects cover various entities in the network, such as OLT, SW, BAS, OTN devices, and their environment, such as city, county, station, and machine room. The edges between nodes represent the logical relationship between these resource objects, including but not limited to:

[0065] Spatial position relationship: such as the geographical position association of city-county-station-machine room-device.

[0066] Physical connection relationship: such as the direct connection of OLT port and SW port.

[0067] Bearing relationship: such as the logical association of OLT device bearing on OTN circuit through link.

[0068] This device graph combined with geographical information can not only help analysts quickly identify and locate network faults, but also predict potential regional risks, providing a more comprehensive and in-depth perspective for network operation and maintenance.

[0069] In addition, the device graph also records key configuration information such as double-uplink routing configuration of each link, including but not limited to primary and backup routing path information (such as path ID, starting device ID, ending device ID, path node list, and path edge list) and routing attribute information (such as routing transmission rate, path delay, and path error rate). These configuration details, such as the existence of single or non-redundant link paths, are crucial for identifying potential network vulnerabilities. By analyzing the configuration data in the device graph, the system can intelligently determine the dependency relationship between devices, especially when facing complex faults such as OLT group obstacles, it can quickly locate possible single-point faults or configuration defects, such as the absence or improper of double-uplink routing, providing a forward-looking hazard identification and risk assessment means for network operation and maintenance, greatly enhancing the stability and fault prevention capability of the network.

[0070] In the above step S206, the alarm graph is constructed with the alarm events corresponding to the network devices as nodes and the causal relationship between the alarm events as edges.

[0071] In the embodiments of the present application, in the construction of the alarm graph, each alarm event occurring in the network is adopted as a node in the alarm graph, each node contains but is not limited to alarm ID, alarm state, alarm time, location, type and description and the like, and these information collectively constitute a comprehensive description of the alarm event; and the edges between the nodes represent the potential causal relationship between the alarm events, and through the analysis of the causal relationship, the system can identify the subsequent alarms that a alarm event can trigger or the potential root cause that multiple alarm events jointly point to. The construction of the causal relationship is not only based on the physical or logical connection between the devices, but also considers the propagation law of the fault and the actual experience of network operation, which provides a powerful analysis tool for the identification of group obstacle phenomenon, root cause analysis and fault prediction.

[0072] In the step S208, the fault identification model is obtained by the following manner: obtaining historical alarm data and historical work order data; performing feature extraction on the historical alarm data and the historical work order data to obtain a first feature sequence corresponding to the historical alarm data and a second feature sequence corresponding to the historical work order data, wherein the first feature sequence includes time sequence features and spatial distribution features corresponding to the historical alarm data, and the second feature sequence is used to reflect the relationship between the historical alarm events and the historical fault processing measures; and training an initial model under a target application scenario according to the first feature sequence and the second feature sequence to obtain the fault identification model, wherein the target application scenario is a cluster offline fault scenario of an optical line terminal.

[0073] In the embodiments of the present application, the training and application of the fault identification model is the key link of the group obstacle identification method, and is particularly for the target application scenario of the cluster offline fault of the optical line terminal (OLT). The specific process can be as follows:

[0074] Firstly, historical alarm data and historical work order data are obtained from a resource relational database, wherein the historical work order data includes historical work order and merged work order records. Specifically, the historical alarm data contains various alarm events occurring in the past network operation, including but not limited to historical alarm ID, type, level, time, location and description and the like. The historical work order data records the detailed description of the fault event, including but not limited to historical work order ID, fault description, processing measure, processing time, processing result and associated alarm ID, device ID and path ID and the like.

[0075] Secondly, feature extraction is performed on the historical alarm data to obtain a first feature sequence corresponding thereto, and the first feature sequence not only contains time sequence features (such as fault frequent period, alarm duration and the like), but also covers spatial distribution features (such as the distribution mode of the fault in the geographical space).

[0076] Meanwhile, the historical work order data is analyzed to extract a second feature sequence corresponding thereto, which mainly reflects the correlation between the fault handling measures and the alarm events, including the priority of fault handling, the efficiency and effectiveness of the handling measures, etc.

[0077] Finally, the initial model is trained under a target application scenario (in this application, the offline fault of an optical line terminal (OLT) cluster is taken as a specific training scenario) according to the extracted first and second feature sequences. The model training process can be implemented by a sequence-to-sequence (SFT) model, aiming to enable the model to learn and master the time sequence pattern of the fault events, the spatial distribution law, and the relationship between the fault handling measures and the fault types. In the process of model training, the parameters are constantly adjusted, the feature weights are optimized, and new learning strategies are introduced to improve the recognition accuracy and processing efficiency of the model for the cluster offline fault, and finally a fault recognition model is obtained.

[0078] Optionally, the correlation of the faulty device in the alarm process includes the frequent time period, duration, and co-linear mode with other network events of various fault types of the faulty device, wherein the fault types at least include the cluster offline fault of the OLT.

[0079] In the embodiments of this application, after the model training is completed, the correlation of the faulty device in the alarm process can be determined by the model. These correlation laws include but are not limited to the frequent time period of various fault types of the faulty device in time, the average duration of the fault, and the co-linear mode between the fault and other network events. For example, when the model identifies that multiple OLT devices simultaneously appear offline fault in a similar time period, and the uplink of these OLT devices or the region where they are located has a common alarm mode, the model will determine that it is an OLT cluster offline fault, forming a group fault recognition result.

[0080] Based on the determined correlation, the model can further analyze the root cause of the fault to obtain a root cause positioning result, including but not limited to identifying the initial fault point that triggers the group fault, and how the fault propagates in the network. For example, if multiple OLT devices are offline at the same time, the model will analyze the uplink state of these OLT devices, the dynamic system state of the computer room where they are located, and the state of the associated SW and BAS devices, to locate the fault root cause, such as whether there is a dynamic system fault, an upper device fault, or an optical cable interruption. After determining the group fault recognition result and the fault root cause, the analysis results can be integrated to provide decision support for the operation and maintenance personnel.

[0081] After the above step S208, the above method can further include: converting the group fault recognition result into a preset format, and providing a preplan document link corresponding to the group fault recognition result, wherein the preplan document link contains the fault handling measures determined according to the operation and maintenance graph.

[0082] In the embodiments of the present application, after completing the group obstacle identification and root cause positioning, the analysis results can be converted into a preset JSON format, which contains all important identification information such as root alarm ID, root alarm type, judgment basis, group obstacle judgment, hidden danger identification, and other key fields, which can ensure that all necessary fault details are clearly recorded and presented. Subsequently, according to the fault identification results, automatically associate the operation and maintenance atlas to find the processing measures matching the fault type and root cause. Among them, the operation and maintenance atlas contains detailed device operation manual, fixed rules and historical fault cases, which provide the basis for generating repair suggestions.

[0083] Specifically, the system generates a (emergency) plan document link that directly points to a document containing specific fault handling measures, which helps operators quickly review and execute. For example, based on the current fault description and device status, the model can recommend the most appropriate handling steps and preventive measures, which significantly improves the relevance and efficiency of fault handling.

[0084] With the handling of each group obstacle event, the system collects feedback information, including the actual effect of fault handling and the operation experience of the operation and maintenance personnel, which will be used for continuous training and optimization of the model. Through continuous learning, the model can more accurately predict fault patterns and provide more intelligent fault handling suggestions, forming a closed-loop process from fault identification to plan generation to feedback optimization.

[0085] In addition to outputting standardized group obstacle identification results, the system also provides multi-dimensional fault analysis, including the frequency of the fault, the duration, the impact range, and the correlation with other network events. These information not only help operators understand the whole picture of the fault, but also guide them to make more reasonable repair decisions to ensure the stability of the network.

[0086] In the embodiments of the present application, an interface calling function is also provided to realize timely identification and analysis of OLT group obstacles. Specifically, a series of intelligent operations can be automatically triggered at a specified time point (for example, every 5 minutes, the alarm data within the previous and next 5 minutes is analyzed), including database query, uplink device analysis, alarm root cause positioning, etc., and finally returning easy-to-understand fault diagnosis information and coping strategy links.

[0087] For example, select the alarm information within 10 minutes before and after 2025-02-24 12:50:00 after entering the OLT group obstacle capability scenario to identify the group obstacle. Based on the provided calling time period, after the system performs the automated analysis process, the expected returned JSON format data contains but is not limited to the following fields, and the output parameters (API response example) are shown in Table 1:

[0088]

[0089]

[0090] In the embodiments of the present application, the construction of the knowledge graph and the real-time data processing technology of Function Calling are ingeniously combined to provide a new group obstacle identification method for the field of communication network operation and maintenance. Specifically, by deeply mining the time sequence features, spatial distribution rules and fault handling measures in historical alarm data and work order records, detailed device graphs and alarm graphs are constructed. In the model training and fault identification process, the Function Calling mechanism allows the large model to call the network device state in real time, realizes the dynamic association of alarm information and resource information, and effectively improves the accuracy and response speed of group obstacle identification. In addition, the system also has intelligent decision-making and plan recommendation capabilities, and can quickly generate repair suggestions according to the fault identification results, further optimizing the fault handling process. That is, through its unique data processing and analysis strategy, the present application significantly enhances the intelligence and efficiency of the operation and maintenance system, providing a revolutionary tool for the prevention and solution of network faults.

[0091] According to the embodiments of the present application, a group obstacle identification device is provided. It should be noted that the group obstacle identification device of the embodiments of the present application can be used to execute the group obstacle identification method provided by the embodiments of the present application. The group obstacle identification device provided by the embodiments of the present application is introduced below.

[0092] Figure 4 is a structural diagram of a group obstacle identification device according to the embodiments of the present application. As shown in Figure 4 , the device includes:

[0093] The acquisition module 40 is configured to acquire alarm information in a preset time period, wherein the alarm information is used to prompt that a network device is in an abnormal state.

[0094] The first determination module 42 is configured to determine a device address corresponding to the alarm information, and determine an uplink network device corresponding to the device address according to a device graph, wherein the device graph is used to represent the association relationship between each network device.

[0095] The second determination module 44 is configured to determine an alarm state of the uplink network device according to an alarm graph, and determine a fault device from the uplink network device according to the alarm state, wherein the alarm graph is used to represent the association relationship between each alarm event.

[0096] The third determining module 46 is configured to determine an association rule of the fault device in the alarm process by using the fault identification model, and determine a group fault identification result according to the association rule, wherein the association rule is used to reflect a fault propagation mode and a historical processing efficiency of the fault device.

[0097] Through the acquisition module, the first determining module, the second determining module and the third determining module in the group fault identification device, the purpose of quickly and accurately identifying the network group fault and the influence range is achieved, so that the technical effects of improving the network operation efficiency, shortening the fault recovery time and improving the service quality are realized, and the technical problems that the group fault identification method in the related art is inefficient, especially when a large number of device alarms and user reports are faced, it is difficult to quickly and accurately locate the fault cause, and the network fault is difficult to be discovered and processed in time are solved.

[0098] In the group fault identification device provided in the embodiments of the present application, the acquisition module is further configured to acquire alarm information in a preset time period from an operation and maintenance guarantee center by using a preset calling mechanism, wherein the operation and maintenance guarantee center stores a device graph, an alarm graph and an operation and maintenance graph, and the operation and maintenance graph is used to provide repair guidance for the fault device.

[0099] In the group fault identification device provided in the embodiments of the present application, the processing module 48 is further configured to convert the group fault identification result into a preset format, and provide a preplan document link corresponding to the group fault identification result, wherein the preplan document link contains fault processing measures determined according to the operation and maintenance graph.

[0100] In the group fault identification device provided in the embodiments of the present application, the processing module is further configured to acquire link data of all network devices, and determine an initial device graph according to the link data, wherein the link data includes device information of the network devices and link information between the network devices; acquire local station and machine room data of the network devices, and power environment data, wherein the local station and machine room data are used to represent local station information and machine room information with clear geographic coordinates and type attributes, and the power environment data is used to represent environment monitoring data in the local station and the machine room; and optimize the initial device graph according to the local station and machine room data and the power environment data to obtain the device graph.

[0101] In the group obstacle identification device provided in the embodiments of the present application, the processing module is further configured to acquire historical alarm data and historical work order data; perform feature extraction on the historical alarm data and the historical work order data to obtain a first feature sequence corresponding to the historical alarm data and a second feature sequence corresponding to the historical work order data, wherein the first feature sequence includes a time sequence feature and a spatial distribution feature corresponding to the historical alarm data, and the second feature sequence is used to reflect a relationship between historical alarm events and historical fault handling measures; and train an initial model under a target application scenario to obtain a fault identification model according to the first feature sequence and the second feature sequence, wherein the target application scenario is a cluster offline fault scenario of an optical line terminal.

[0102] The embodiments of the present application further provide an electronic device, including a memory and a processor, wherein the memory is configured to store program instructions; and the processor is connected with the memory and is configured to execute the group obstacle identification method.

[0103] It should be noted that the electronic device is configured to execute the group obstacle identification method as shown in the above method, and the related explanations in the group obstacle identification method are also applicable to the electronic device, which will not be repeated here. Figure 2

[0104] The embodiments of the present application further provide a non-volatile storage medium, including a stored computer program, wherein a device in which the non-volatile storage medium is located executes the group obstacle identification method by running the computer program.

[0105] It should be noted that the non-volatile storage medium is configured to execute the group obstacle identification method as shown in the above method, and the related explanations in the group obstacle identification method are also applicable to the non-volatile storage medium, which will not be repeated here. Figure 2

[0106] The embodiments of the present application further provide a computer program product, including computer instructions, which are executed by a processor to implement the group obstacle identification method.

[0107] It should be noted that the computer program product is configured to execute the group obstacle identification method as shown in the above method, and the related explanations in the group obstacle identification method are also applicable to the computer program product, which will not be repeated here. Figure 2

[0108] The serial numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0109] In the above embodiments of the present application, the description of each embodiment has its own emphasis, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0110] ​​​In several embodiments provided in the present application, it should be understood that the disclosed technology can be implemented in other manners. The above described apparatus embodiments are merely exemplary, for example, the division of units can be different, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, access layers, or a component.

[0111] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place or can be distributed on a plurality of places. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments.

[0112] In addition, each functional unit in the embodiments of the present application can be integrated in a processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0113] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such an understanding, the technical solutions of the present application essentially or the part that makes a contribution to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in the embodiments of the present application. The foregoing storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various other media that can store program codes.

[0114] The above are only the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, without departing from the principles of the present application, some improvements and refinements can be made, and these improvements and refinements should be regarded as the protection scope of the present application.

Claims

1. A method for identifying group obstacles, characterized in that, include: Obtain alarm information within a preset time period, wherein the alarm information is used to indicate that the network device is in an abnormal state; The device address corresponding to the alarm information is determined, and the uplink network device corresponding to the device address is determined according to the device map, wherein the device map is used to represent the association relationship between various network devices; The alarm status of the uplink network device is determined based on the alarm map, and the faulty device is identified from the uplink network device based on the alarm status. The alarm map is used to represent the correlation between various alarm events. The correlation pattern of the faulty device in the alarm process is determined by the fault identification model, and the group fault identification result is determined based on the correlation pattern. The correlation pattern is used to reflect the fault propagation mode and historical processing efficiency of the faulty device.

2. The method according to claim 1, characterized in that, Retrieve alarm information within a preset time period, including: Alarm information within the preset time period is obtained from the operation and maintenance support center through a preset invocation mechanism. The operation and maintenance support center stores the device map, the alarm map, and the operation and maintenance map. The operation and maintenance map is used to provide repair guidance for the faulty equipment.

3. The method according to claim 2, characterized in that, The method further includes: The group fault identification results are converted into a preset format, and a link to a contingency plan document corresponding to the group fault identification results is provided. The contingency plan document link contains fault handling measures determined based on the operation and maintenance map.

4. The method according to claim 1, characterized in that, The equipment map is determined in the following way: Obtain link data for all network devices and determine an initial device map based on the link data, wherein the link data includes device information of the network devices and link information between the network devices; Acquire the local station data and power environment data of the network device, wherein the local station data is used to represent local station information and data center information with clear geographical coordinates and type attributes, and the power environment data is used to represent environmental monitoring data within the local station and data center; The initial equipment map is optimized based on the data from the central office equipment room and the power environment data to obtain the equipment map.

5. The method according to claim 4, characterized in that, The device map is constructed using the resource objects corresponding to the network devices as nodes and the logical relationships between the resource objects as edges. The logical relationships include at least the spatial location relationships, physical connection relationships, and carrying relationships between the resource objects.

6. The method according to claim 1, characterized in that, The alarm graph is constructed using the alarm events corresponding to the network devices as nodes and the causal relationships between the alarm events as edges.

7. The method according to claim 1, characterized in that, The fault identification model is trained in the following way: Obtain historical alarm data and historical work order data; Feature extraction is performed on the historical alarm data and the historical work order data to obtain a first feature sequence corresponding to the historical alarm data and a second feature sequence corresponding to the historical work order data. The first feature sequence includes time series features and spatial distribution features corresponding to the historical alarm data, and the second feature sequence is used to reflect the relationship between historical alarm events and historical handling measures. Based on the first feature sequence and the second feature sequence, the initial model is trained in the target application scenario to obtain the fault identification model, wherein the target application scenario is the cluster offline fault scenario of optical line terminal.

8. The method according to claim 1, characterized in that, The correlation patterns of the faulty devices during the alarm process include: the frequency period, duration, and co-location pattern of various fault types of the faulty devices with other network events, wherein the fault types include at least: cluster offline faults of optical line terminals.

9. A group obstacle identification device, characterized in that, include: The acquisition module is used to acquire alarm information within a preset time period, wherein the alarm information is used to indicate that the network device is in an abnormal state; The first determining module is used to determine the device address corresponding to the alarm information and to determine the uplink network device corresponding to the device address based on the device map, wherein the device map is used to represent the association relationship between various network devices; The second determining module is used to determine the alarm status of the uplink network device based on the alarm map, and to determine the faulty device from the uplink network device based on the alarm status, wherein the alarm map is used to represent the correlation between various alarm events; The third determining module is used to determine the correlation pattern of the faulty device in the alarm process through the fault identification model, and to determine the group fault identification result based on the correlation pattern, wherein the correlation pattern is used to reflect the fault propagation mode and historical processing efficiency of the faulty device.

10. An electronic device, characterized in that, include: A memory and a processor, wherein the memory is used to store program instructions; The processor, connected to the memory, is used to execute the obstacle identification method according to any one of claims 1 to 8.

11. A non-volatile storage medium, characterized in that, The non-volatile storage medium includes a stored computer program, wherein the device containing the non-volatile storage medium executes the group obstacle identification method according to any one of claims 1 to 8 by running the computer program.

12. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they implement the obstacle identification method according to any one of claims 1 to 8.

Images