Fault identification method, electronic device, and computer-readable storage medium

Abstract

The present disclosure relates to the technical field of communications, and provides a fault identification method, an electronic device, and a computer-readable storage medium. The fault identification method comprises: acquiring network anomaly data, and upon determining, on the basis of the network anomaly data, that a fault has occurred, acquiring target network data; on the basis of the target network data, acquiring an inspection entry list, and acquiring topological relationships of network devices associated with the target network data; on the basis of the topological relationships and the inspection entry list, acquiring communication network management data, and, on the basis of the acquired communication network management data, determining an inspection result for each entry in the inspection entry list; and on the basis of the inspection results, generating a state mode vector corresponding to each network device, and, on the basis of the state mode vector, determining a fault identification result.

Description

Fault identification method, electronic device, and computer-readable storage medium

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to Chinese patent application No. 202410174339.7 filed on February 7, 2024, the contents of which are incorporated herein by reference in their entirety. Technical Field

[0003] The present disclosure relates to the field of communication technology, and in particular to a fault identification method, an electronic device, and a computer-readable storage medium. Background Art

[0004] With the rapid development of communication networks, customers have increasingly higher requirements and lower tolerance for network failures.

[0005] Fault demarcation in related communication networks often relies on experts manually analyzing data such as FCAPS (a common term in network management, referring to the five key aspects of comprehensive communication network management, including fault management, configuration management, accounting management, performance management, and security management) to draw conclusions, which takes a long time.

[0006] Communications networks contain numerous network support systems, each with distinct functions. When a fault occurs, operations and maintenance personnel often need to collect data from multiple systems to locate the problem. However, the topological relationships between the various network support systems are incomplete, making it difficult to establish a comprehensive fault topology.

[0007] In order to ensure the quality of communication networks and improve users' experience with communication networks, it is very important to improve the efficiency of network fault handling.

[0008] Public content

[0009] The first aspect of the present disclosure provides a fault identification method, comprising: obtaining network anomaly data, and when determining that a fault has occurred based on the network anomaly data, collecting target network data; obtaining a check item list based on the target network data, and obtaining a topological relationship of each network device associated with the target network data; collecting communication network management data based on the topological relationship and the check item list, and determining an inspection result of each item in the check item list based on the collected communication network management data, wherein the inspection result of one item is obtained by detecting the collected communication network management data using the item, and is used to indicate whether an anomaly has been detected; generating a state pattern vector corresponding to each network device based on the inspection result, and determining a fault identification result based on the state pattern vector, wherein the fault identification result is used to indicate the network device that has failed.

[0010] According to a second aspect of the present disclosure, there is provided an electronic device, comprising: at least one processor; a memory on which at least one computer program is stored, wherein when the at least one computer program is executed by the at least one processor, the at least one processor implements the method according to the first aspect; and at least one I / O interface connected between the processor and the memory and configured to implement information interaction between the processor and the memory.

[0011] A third aspect of the present disclosure provides a computer-readable storage medium having a computer program stored thereon, wherein the program is executed by a processor so that the processor implements the method according to the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] FIG1 is a schematic flow chart of a fault identification method provided in the present disclosure;

[0013] FIG2 is a schematic diagram of a training process of a neural network model provided in the present disclosure;

[0014] FIG3 is a schematic diagram of a specific process of fault demarcation provided in the present disclosure;

[0015] FIG4 is a schematic structural diagram of a fault identification device provided in the present disclosure;

[0016] FIG5 is a schematic structural diagram of a fault identification device provided in the present disclosure;

[0017] FIG6 is a schematic structural diagram of an electronic device provided in the present disclosure. DETAILED DESCRIPTION

[0018] The following describes the specific embodiments of the present disclosure in detail with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure and are not intended to limit the present disclosure.

[0019] As used in this disclosure, the term "and / or" includes any and all combinations of at least one of the associated listed items.

[0020] The terms used in the present disclosure are only used to describe specific embodiments and are not intended to limit the present disclosure.As used in the present disclosure, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0021] When the terms “comprising” and / or “made of…” are used in the present disclosure, it specifies the existence of certain features, integers, steps, operations, elements and / or components, but does not preclude the existence or addition of at least one other feature, integer, step, operation, element, component and / or group thereof.

[0022] Unless otherwise defined, all terms (including technical and scientific terms) used in this disclosure have the same meanings as those commonly understood by those skilled in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and this disclosure, and will not be interpreted as having an idealized or overly formal meaning unless expressly defined in this disclosure.

[0023] The present disclosure provides a fault identification method that can be applied to any electronic device. The fault identification method provided by the present disclosure can be applied to network fault demarcation in a core network, and can also be applied to network device fault demarcation in other scenarios.

[0024] FIG1 is a flow chart of a fault identification method provided by the present disclosure. The fault identification method includes the following steps: 101 to 104 .

[0025] Step 101: Acquire network abnormality data, and when it is determined that a fault has occurred based on the network abnormality data, collect target network data.

[0026] In some implementations, the network abnormality data includes at least one of the following: alarm data; abnormal performance data.

[0027] In some embodiments, the collecting target network data when a fault is determined to have occurred based on the network abnormality data includes: determining whether a network fault is identified based on pre-configured fault identification rules and monitored network abnormality data, determining the time when the network fault occurs when a network fault is identified, and collecting data before and after the time when the network fault occurs as target network data.

[0028] The data before and after the network failure occurs specifically includes network data collected within a first preset time period before the network failure occurs, and network data collected within a second preset time period after the network failure occurs.

[0029] The time of network failure occurrence can be determined based on the time when the network anomaly data is generated or collected. The network anomaly data is the network data used to determine the occurrence of the network failure. For example, the time when the network anomaly data is generated, or the time when the network data used to determine the occurrence of the network failure is collected, can be used as the time of network failure occurrence.

[0030] In some embodiments, before obtaining the network abnormality data, the fault identification method further includes: pre-configuring fault identification rules, where the fault identification rules are formulated based on the network abnormality data and the fault type, and after monitoring the network abnormality data, determining whether a fault has occurred based on the fault identification rules and the monitored network abnormality data.

[0031] Fault identification rules include but are not limited to at least one of the following: the alarm data in the collected network anomaly data meets the preset conditions, for example, an alarm code or an alarm code of a specified type is collected; the number of times the alarm data appears per unit time in the collected network anomaly data reaches a threshold, etc. The unit time is set according to the agreement, for example, it can be set to hours, months, half a year, etc.; the performance data in the collected network anomaly data exceeds the predefined threshold.

[0032] Step 102: Acquire a check item list according to the target network data, and acquire a topological relationship of each network device associated with the target network data.

[0033] In some embodiments, the check item list (also referred to as a checklist) includes at least one item, and each item includes a check item and a check rule corresponding to the check item.

[0034] The inspection items include categories of network data that need to be inspected.

[0035] The inspection rules include rules for determining whether network data corresponding to the inspection item is abnormal.

[0036] For example, the inspection items include at least one of categories such as alarm, performance, configuration, and log.

[0037] The inspection rules corresponding to the inspection items include but are not limited to the following: judging whether the network data corresponding to the inspection item is within the range specified by the fixed threshold (such as whether it is less than the fixed threshold, or whether it is greater than the fixed threshold, etc.), also known as the fixed threshold inspection rule; judging whether the difference between the network data corresponding to the inspection item and the historical reference value is within the set range, also known as the historical reference inspection rule; judging whether the difference between the network data corresponding to the inspection item and the average value is within the set range, also known as the moving average inspection rule; judging whether the correlation analysis results of the network data corresponding to the inspection item and other network data meet the set conditions, also known as the correlation analysis inspection rule.

[0038] In some embodiments, obtaining a checklist of check items based on the target network data includes: determining a fault type corresponding to the target network data; and obtaining a checklist of check items corresponding to the determined fault type based on a preconfigured mapping relationship between the fault type and the checklist of check items. In this embodiment, by preliminarily identifying the fault type and obtaining a checklist of check items corresponding to the fault type, the obtained checklist of check items is adapted to the preliminarily determined fault type, ensuring the adaptability of the checklist of check items and providing a basis for efficient fault identification.

[0039] In some implementations, target network data is collected by accessing multiple systems, fault identification is performed based on the collected target network data, the fault type is preliminarily determined, and a corresponding checklist is obtained based on the fault type.

[0040] In some embodiments, obtaining the topological relationship of each network device associated with the target network data includes: determining at least one first network device to which the target network data belongs; obtaining at least one second network device associated with the first network device through a network management system, and obtaining the connection relationship between each first network device and each second network device through the network management system; and determining the topological relationship of each network device associated with the target network data based on the connection relationship between each first network device and each second network device.

[0041] The topological relationship of each network device associated with the fault is established through the target network data, thereby comprehensively considering the upper and lower associated devices of the fault, improving the coverage of fault demarcation, and providing support for rapid and accurate fault identification.

[0042] In some embodiments, based on the device information of network abnormality data such as alarms and performance, information is collected from the network management system to establish the topological relationship corresponding to the faulty equipment, including the topological relationship from the network element to the virtual machine, the topological relationship from the virtual machine to the host, the topological relationship from the host to the switch, the topological relationship from the switch to the router, etc.

[0043] The network management system includes the network element management system and the network management system.

[0044] Step 103: Collect communication network management data based on the topological relationship and the inspection item list, and determine the inspection results of each item in the inspection item list based on the collected communication network management data. The inspection result of one item is obtained by using the item to detect the collected communication network management data, and is used to indicate whether an abnormality is detected.

[0045] In some embodiments, the collecting of communication network management data based on the topological relationship and the check item list includes: collecting communication network management data from a network management system based on the check item list and the connection relationship between each of the network devices in the topological relationship; the collected communication network management data is the target data required for at least one item in the check item list; and the collected communication network management data belongs to at least one network device in the topological relationship, or belongs to a device that has a connection relationship with at least one network device in the topological relationship.

[0046] The communication network management system includes various management function data, including fault management data, configuration management data, financial management data, performance management data and security management data, namely FCAPS data.

[0047] In some embodiments, based on the inspection item list and the connection relationship between each network device in the topological relationship, communication network management data is collected from each management function data of the network management system, including: obtaining the inspection items included in each item in the inspection item list; based on the inspection items and the connection relationship between each network device in the topological relationship, communication network management data is collected from each management function data of the network management system.

[0048] There are two situations, depending on the source of the data required for the inspection item.

[0049] Case 1: When the data required for the inspection item comes from a network device, the communication network management data matching the inspection item is directly collected from the management function data of the network management system.

[0050] Case 2: When the data required for the inspection item comes from the network device and the upstream and downstream devices of the network device, the upstream and downstream devices of the network device are found according to the connection relationship between the network devices in the topology relationship. Based on the information of the upstream and downstream devices of the network device, the communication network management data matching the inspection item is collected from the various management function data of the network management system.

[0051] For example, when the inspection items include configuration item A of network device 1 and configuration item B of the upper-layer device of network device 1, the upper-layer device of network device 1 is further searched for as network device 2 based on the connection relationship between network devices. For this inspection item, the data of configuration item A of network device 1 and the data of configuration item B of network device 2 are collected from the management function data of the network management system.

[0052] In some embodiments, each communication network management data of the network management system includes FCAPS data, including fault management (Fault Management), configuration management (Configuration Management), accounting management (Accounting Management), performance management (Performance Management) and security management (Security Management) data.

[0053] In some embodiments, determining the inspection results of each item in the inspection item list based on the collected communication network management data includes: performing the following processing on each of the network devices included in the topological relationship: for each entry in the inspection item list, obtaining the target data required for the entry from the communication network management data corresponding to the network device, processing the target data using the inspection rules contained in the entry, and generating the inspection result corresponding to the entry.

[0054] In some implementations, the checklist includes entries at multiple levels, such as the network element level, virtual machine level, host level, switch level, and storage pool level.

[0055] In some embodiments, before determining the inspection results of each item in the inspection item list based on the collected communication network management data, the fault detection method also includes: cleaning the collected communication network management data, including cleaning duplicate alarms, cleaning duplicate logs, eliminating invalid data, and eliminating discontinuous time series data, etc.

[0056] Step 104 : Generate a state pattern vector corresponding to each of the network devices based on the inspection result, and determine a fault identification result based on the state pattern vector, where the fault identification result is used to indicate the network device where the fault occurs.

[0057] In some embodiments, determining the fault identification result based on the state pattern vector includes: inputting the state pattern vector into a pre-trained neural network model to obtain an output fault identification result.

[0058] In some embodiments, generating a state mode vector corresponding to each of the network devices based on the inspection results includes: for any of the network devices, obtaining the inspection results of each item in the inspection item list corresponding to the network device as a target inspection result set, determining the state value corresponding to each target inspection result in the target inspection result set, and generating a state mode vector corresponding to the network device based on the state value corresponding to each target inspection result; the inspection result of any of the items is used to indicate whether the corresponding item is faulty.

[0059] In some embodiments, a corresponding inspection result is obtained according to the inspection rule corresponding to each item in the inspection item list, and a state pattern vector is constructed according to the inspection result, which is expressed as follows:

[0060] P i,j Indicates the status value corresponding to the inspection result of inspection item j of device i, P = [P i,1 ,...P i,m ] represents the complete state pattern vector of device i.

[0061] In some embodiments, the process of obtaining the neural network model includes: obtaining a training sample set, the training sample set including historical normal data and historical fault data of the network device, the historical normal data being the communication network management data collected when the network device is in normal operation, and the historical fault data being the communication network management data collected before and after the network device fault occurs; using the training sample set to train the initial neural network model until the initial neural network model can converge to obtain the final neural network model.

[0062] In some embodiments, the neural network model is trained by a self-learning algorithm based on historical data.

[0063] Figure 2 shows a schematic diagram of the training process of the neural network model, which mainly includes the following steps:

[0064] Collect data on network devices in historical normal states, as well as FCAPS data on network devices near the time of various historical failures. Network devices include virtual machines, hosts, switches, etc.

[0065] Clean the collected FCAPS data, including cleaning duplicate alarms and logs, and eliminating invalid data.

[0066] A portion of the cleaned data samples is randomly selected as training samples, and the rest are used as test samples.

[0067] According to the inspection rules corresponding to each item in the inspection item list, the inspection results corresponding to each item are obtained after calculation based on the training samples. The state pattern vector is constructed based on the inspection results, which is expressed as:

[0068] P i,j The state value of entry j of device i, P = [P i,1 ,...P i,m ] represents the complete state pattern vector of device i. After normalizing the data, it is easier to extract features. These state pattern vectors will then be used as the representation vectors of the device for learning.

[0069] The neural network model is trained using a deep learning method called self-learning. More specifically, a stacked denoising autoencoder can be used for self-learning. This stacked denoising autoencoder is a deep neural network model composed of multiple autoencoders. It requires layer-by-layer training and fine-tuning to achieve the desired neural network model.

[0070] Taking a simple three-layer autoencoder as an example, a single layer consists of an input layer, a hidden layer, and an output layer. The encoder layer runs from the input layer to the hidden layer, while the decoder layer runs from the hidden layer to the output layer. Typically, the number of neurons in the hidden layer is much lower than that in the input layer, allowing fewer features to be used to represent the input data, thus achieving dimensionality reduction. The unsupervised training of each autoencoder layer is as follows: For the input layer, sample data P is first encoded, Y = f(W1P + b1).

[0071] W1 is the weight, b1 is the bias, and Y is the value obtained after a layer of autoencoder encodes the input sample, that is, the output of the hidden layer.

[0072] Next, reconstruct the encoded value into data.

[0073] W1, W2, b1, b2 are the network parameters of the autoencoder.

[0074] Noise is introduced into the input samples during training to increase the robustness of the encoding.

[0075] The training goal of the autoencoder is to make the reconstructed data as similar as possible to the input sample data. In order to extract the main features and ensure the training goal, the objective function can be expressed as:

[0076] m represents the number of input sample data.

[0077] The back-propagation algorithm is used to adjust the weights and configuration of the neural network model. When the neural network model converges, it means that the layer has been trained. After fixing the network parameters, the next layer is trained.

[0078] During the training process of the stacked denoising autoencoder, after the first layer of autoencoder is trained using the above process, the parameters and results of the hidden layer in the first layer of autoencoder are fixed, and the output layer of the first layer of autoencoder and the corresponding weights and biases are removed; the results of the hidden layer are used as the input of the second layer of autoencoder, and after the second layer of autoencoder is trained using the above process, the parameters and results of the hidden layer in the second layer of autoencoder are fixed, and the output layer of the second layer of autoencoder and the corresponding weights and biases are removed; and so on, until the parameters and results of the hidden layer in the last layer of autoencoder are obtained, and after the output layer and the corresponding weights and biases of the last layer of autoencoder are removed, a classifier is added after the hidden layer of the last layer of autoencoder to obtain the architecture of the entire neural network model, and then the network parameters of the entire neural network model are fine-tuned in combination with the labeled sample data so that the classifier outputs a vector for indicating the category, the labeled sample data includes the state pattern vector of the network device and the state of the pre-labeled network device, and the vector output by the classifier is used to indicate the state of the network device.

[0079] After the neural network model training is completed, the BP algorithm is used to perform overall reverse tuning to fine-tune the parameters of the neural network model. That is, the entire network is trained through the back propagation algorithm to fine-tune the parameters.

[0080] The test samples are input into the trained neural network model, and the neural network model is used for fault diagnosis.

[0081] In one example, FIG3 shows a specific process example of fault demarcation, which mainly includes steps 301 to 307 .

[0082] Step 301: training a neural network model.

[0083] Step 302: Collect data, identify faults, and obtain a checklist of items.

[0084] Step 303: Establish a topological relationship related to the faulty device.

[0085] Step 304: Collect FCAPS data based on the topological relationship and the checklist.

[0086] Step 305 : Calculate based on the FCAPS data and the checklist to obtain a state mode vector of the device.

[0087] Step 306: The state pattern vector of the device is transferred to the neural network model.

[0088] Step 307: Obtain the diagnosis result output by the neural network model.

[0089] In the present disclosure, when a fault is determined to have occurred through network anomaly data, target network data is collected, and by obtaining a check item list associated with the target network data and the topological relationship of each network device, communication network management data is collected based on the topological relationship and the check item list. In this way, communication network management data can be intelligently collected based on the preliminarily obtained target network data, without the need for manual intervention. Operation and maintenance personnel do not need to manually collect data from multiple systems to locate the fault, which reduces manual operations and improves the collection efficiency of communication network management data, thereby providing a possibility for improving fault identification efficiency. In addition, the inspection results of each item in the check item list are determined based on the collected communication network management data, and then a state pattern vector corresponding to each of the network devices is generated based on the inspection results, thereby obtaining the status of each network device related to the network fault and obtaining an output fault identification result, so that the fault can be quickly identified based on the status of the network device related to the fault, and fault demarcation can be quickly achieved, which reduces the time for fault demarcation, improves the accuracy of demarcation, reduces the cost of fault demarcation, improves the fault identification efficiency, and thus improves the fault handling efficiency.

[0090] The fault identification method provided by this disclosure can proactively detect faults through monitoring, analyze data from multiple data sources, and localize the fault point. The collected data is strongly correlated with the fault, avoiding the collection of large amounts of irrelevant data. Compared with related technologies, the data is more comprehensive and the network topology is clearer.

[0091] The steps of the various methods above are divided only for clarity of description. During implementation, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this disclosure. Adding insignificant modifications or introducing insignificant designs to the algorithm or process without changing the core design of the algorithm and process are all within the scope of protection of this disclosure.

[0092] The present disclosure provides a fault identification device. The specific implementation of the device can be found in the relevant description of the above method implementation, which will not be repeated here. FIG4 shows a schematic structural diagram of the device, which mainly includes: a first acquisition module 401, a second acquisition module 402, a collection module 403 and an identification module 404.

[0093] The first acquisition module 401 is configured to acquire network abnormality data, and collect target network data when a fault is determined to have occurred based on the network abnormality data.

[0094] The second acquisition module 402 is configured to acquire a check item list according to the target network data, and acquire a topological relationship of each network device associated with the target network data.

[0095] The collection module 403 is configured to collect communication network management data based on the topological relationship and the inspection item list, and determine the inspection results of each item in the inspection item list based on the collected communication network management data. The inspection result of one item is obtained by using the item to detect the collected communication network management data, and is used to indicate whether an abnormality is detected.

[0096] The identification module 404 is configured to generate a state pattern vector corresponding to each of the network devices based on the inspection result, and determine a fault identification result based on the state pattern vector, where the fault identification result is used to indicate the network device that has a fault.

[0097] In one example, FIG5 shows a schematic structural diagram of a faulty equipment device, which includes: a model training module, a fault identification module, a data acquisition module, a data analysis module, and a fault delimitation module.

[0098] The model training module is configured to train a neural network model for fault identification.

[0099] The fault identification module includes a first acquisition module 401 configured to acquire target network data and initially identify the fault type;

[0100] The data acquisition module includes a second acquisition module 402 and an acquisition module 403, which are configured to obtain a check item list based on the initially identified fault type, and obtain a topological relationship between each network device associated with the network fault data, collect communication network management data based on the topological relationship and the check item list, and determine the inspection result of each item in the check item list based on the collected communication network management data.

[0101] The identification module 404 includes a data analysis module and a fault delimitation module; the data analysis module is configured to generate a state pattern vector corresponding to each of the network devices based on the inspection results; the fault delimitation module is configured to input the state pattern vector corresponding to the network device into a pre-trained neural network model to obtain an output fault identification result.

[0102] The functions or modules included in the device provided by the present disclosure can be used to execute the method described in the method implementation method. Its specific implementation and technical effects can be referred to the description of the method implementation method above. For the sake of brevity, they will not be repeated here.

[0103] It should be noted that all modules involved in this embodiment are logical modules. In actual applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. In addition, to highlight the innovations of this disclosure, this embodiment does not include units that are not closely related to solving the technical problems proposed by this disclosure. However, this does not mean that other units do not exist in this embodiment.

[0104] 6 , an embodiment of the present disclosure provides an electronic device, comprising: at least one processor 601; a memory 602 on which at least one computer program is stored, and when the at least one computer program is executed by the at least one processor, the at least one processor implements the above method; and at least one I / O interface 603, connected between the processor 601 and the memory 602, and configured to implement information interaction between the processor 601 and the memory 602.

[0105] The processor 601 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 602 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically such as SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read-write interface) 603 is connected between the processor 601 and the memory 602, and can realize information exchange between the processor 601 and the memory 602, including but not limited to a data bus (Bus), etc.

[0106] In some implementations, the processor 601 , the memory 602 , and the I / O interface 603 are connected to each other via a bus, and further connected to other components of the computing device.

[0107] The present disclosure also provides a computer-readable storage medium having a computer program stored thereon. The program is executed by a processor, so that the processor implements the method provided by the present disclosure. To avoid repeated description, the specific steps of the method are not repeated here.

[0108] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods applied for above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor (such as a central processing unit, a digital signal processor, or a microprocessor), or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As is well known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable, and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0109] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0110] Those skilled in the art will understand that although some embodiments described herein include certain features included in other embodiments but not other features, the combination of features from different embodiments is meant to be within the scope of the present embodiment and to form different embodiments.

[0111] It is understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present disclosure, and the present disclosure is not limited thereto. Those skilled in the art may make various modifications and improvements without departing from the spirit and substance of the present disclosure, and such modifications and improvements are also considered to be within the scope of protection of the present disclosure.

Claims

1. A fault identification method, comprising: Acquiring network abnormality data, and in the case where a fault is determined to have occurred based on the network abnormality data, collecting target network data; Acquire a check item list according to the target network data, and acquire a topological relationship of each network device associated with the target network data; collecting communication network management data according to the topological relationship and the check item list, and determining a check result for each item in the check item list based on the collected communication network management data, wherein the check result for one item is obtained by testing the collected communication network management data using the item and is used to indicate whether an abnormality is detected; A state pattern vector corresponding to each of the network devices is generated based on the inspection result, and a fault identification result is determined based on the state pattern vector, wherein the fault identification result is used to indicate the network device where the fault occurs.

2. The method according to claim 1, wherein The network abnormality data includes at least one of the following: alarm data; abnormal performance data.

3. The method according to claim 1, wherein The obtaining of a checklist of items according to the target network data includes: Determining a fault type corresponding to the target network data; According to the pre-configured mapping relationship between the fault type and the check item list, the check item list corresponding to the determined fault type is obtained.

4. The method according to claim 1, wherein The acquiring of the topological relationship of each network device associated with the target network data includes: Determining at least one first network device to which the target network data belongs; Acquire at least one second network device associated with the first network device through a network management system, and acquire a connection relationship between each of the first network devices and each of the second network devices through the network management system; A topological relationship between each network device associated with the target network data is determined based on a connection relationship between each first network device and each second network device.

5. The method according to claim 4, wherein The collecting of communication network management data according to the topological relationship and the check item list includes: Acquire communication network management data from the network management system based on the inspection item list and the connection relationship between the network devices in the topological relationship; Among them, the communication network management data collected is the target data required for at least one item in the inspection item list; and the communication network management data collected belongs to at least one network device in the topological relationship, or belongs to a device that has a connection relationship with at least one network device in the topological relationship.

6. The method according to claim 1, wherein Determining the inspection result of each item in the inspection item list based on the collected communication network management data includes: Perform the following processing on each of the network devices included in the topology relationship: For each item in the inspection item list, target data required by the item is obtained from the communication network management data corresponding to the network device, and the target data is processed using the inspection rule contained in the item to generate an inspection result corresponding to the item.

7. The method according to any one of claims 1 to 6, wherein Generating a state mode vector corresponding to each of the network devices based on the inspection result includes: For any of the network devices, obtain the inspection results of each item in the inspection item list corresponding to the network device as a target inspection result set, determine the status value corresponding to each target inspection result in the target inspection result set, and generate a status mode vector corresponding to the network device based on the status value corresponding to each target inspection result.

8. The method according to claim 7, wherein: Determining a fault identification result based on the state pattern vector includes: The state pattern vector is input into a pre-trained neural network model to obtain an output fault identification result.

9. An electronic device comprising: at least one processor; a memory having at least one computer program stored thereon, wherein when the at least one computer program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 8; At least one I / O interface is connected between the processor and the memory and is configured to implement information interaction between the processor and the memory.

10. A computer-readable storage medium having a computer program stored thereon, wherein the program is executed by a processor, causing the processor to implement the method according to any one of claims 1 to 8.

Images