Fault diagnosis method for communication data acquisition service, terminal equipment and storage medium

Abstract

The embodiment of the invention provides a fault diagnosis method for a communication data acquisition service, terminal equipment and a storage medium, and belongs to the technical field of communication. The method comprises the following steps: obtaining a first test model, and testing a target object by using the first test model to obtain test data of the target object; obtaining first abnormal data in the test data, and performing fault type diagnosis on the first abnormal data according to the initial fault diagnosis model to obtain a diagnosis result; under the condition that the fault type corresponding to the first abnormal data is determined to be an unknown fault type according to the diagnosis result, performing abnormal root cause analysis on the first abnormal data to obtain a root cause analysis result; and adjusting model parameters of the initial fault diagnosis model according to a root cause analysis result to obtain a target fault diagnosis model, and performing fault diagnosis on target abnormal data of the target object by using the target fault diagnosis model, thereby solving the problem that an unknown fault cannot be found to be in a passive state in the prior art, and improving the capability of identifying the unknown fault.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a fault diagnosis method, terminal equipment and storage medium for communication data acquisition services. Background Art

[0002] Based on the base station data in the communication field, the operation data of the collection service is obtained, and then the fault identification is performed based on the operation data of the collection service. In the prior art, the fault identification of the collection service operation data is usually carried out by experts analyzing the collected operation data, and then obtaining the fault cause of the data abnormality, so as to build a fault test case with specific conditions according to the known fault to perform the test and obtain the known fault result; but for the test scenario of unknown fault, the fault result is also unknown before it is actively encountered. Therefore, when performing fault analysis on a fault scenario that has not occurred, the fault identification effect cannot be achieved, which makes the abnormal test in product development passive. Summary of the invention

[0003] The main purpose of the embodiments of the present invention is to provide a fault diagnosis method, terminal and storage medium for communication data acquisition services, aiming to solve the problem in the prior art of testing scenarios for unknown faults, where the fault results are unknown before they are actively encountered, and the effect of fault identification cannot be achieved, making abnormal testing in product development in a passive state.

[0004] In a first aspect, an embodiment of the present invention provides a fault diagnosis method for a communication data collection service, comprising:

[0005] Obtaining a first test model, and using the first test model to test a target object to obtain test data of the target object;

[0006] Acquire first abnormal data in the test data, and perform fault type diagnosis on the first abnormal data according to the initial fault diagnosis model to obtain a diagnosis result;

[0007] Under the condition that the fault type corresponding to the first abnormal data is confirmed to be an unknown fault type according to the diagnosis result, performing abnormal root cause analysis on the first abnormal data to obtain a root cause analysis result;

[0008] The model parameters of the initial fault diagnosis model are adjusted according to the root cause analysis result to obtain the target fault diagnosis model, so as to use the target fault diagnosis model to perform fault diagnosis on the target abnormal data of the target object.

[0009] In a second aspect, an embodiment of the present invention further provides a terminal device, comprising a processor, a memory, a computer program stored in the memory and executable by the processor, and a data bus for realizing a connection and communication between the processor and the memory, wherein when the computer program is executed by the processor, the steps of a fault diagnosis method for a communication data acquisition service as provided in the specification of the present invention are implemented.

[0010] In a third aspect, an embodiment of the present invention further provides a storage medium for computer-readable storage, characterized in that the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any communication data acquisition service fault diagnosis method provided in the specification of the present invention.

[0011] The embodiment of the present invention provides a fault diagnosis method, terminal and storage medium for communication data acquisition service, the method comprising: obtaining a first test model to test a target object to obtain test data of the target object; analyzing the test data to obtain first abnormal data in the test data, and then using an initial fault diagnosis model to diagnose the fault type of the first abnormal data to obtain a diagnosis result, when it is confirmed according to the diagnosis result that the fault type corresponding to the first abnormal data is an unknown fault type, performing abnormal root cause analysis on the first abnormal data to obtain a root cause analysis result; and then adjusting the model parameters of the initial fault diagnosis model according to the root cause analysis result to obtain a target fault diagnosis model, so as to use the target fault diagnosis model to perform fault diagnosis on the target abnormal data of the target object. Thus, the problem that the test scenario of unknown faults in the prior art is solved, the fault result is unknown before actively encountering it, and the effect of fault identification cannot be achieved, which makes the abnormal test in product development in a passive state, and improves the ability of fault identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0013] Figure 1 A schematic diagram of a flow chart of a method for diagnosing a fault in a communication data collection service provided by an embodiment of the present invention;

[0014] Figure 2 for Figure 1 A flow chart of sub-steps of a method for diagnosing a fault in a communication data collection service;

[0015] Figure 3A schematic diagram of a scenario for implementing the fault diagnosis method of the communication data collection service provided in this embodiment;

[0016] Figure 4 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0019] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0020] The embodiment of the present invention provides a fault diagnosis method, terminal and storage medium for a communication data collection service. The fault diagnosis method for the communication data collection service can be applied to a terminal device, which can be an electronic terminal such as a tablet computer, a laptop computer, a desktop computer, a personal digital assistant and a wearable terminal.

[0021] Some embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0022] Please refer to Figure 1 , Figure 1 A schematic flow chart of a method for diagnosing a fault in a communication data collection service provided by an embodiment of the present invention.

[0023] like Figure 1 As shown, the fault diagnosis method of the communication data collection service includes steps S101 to S104.

[0024] Step S101: obtain a first test model, and use the first test model to test a target object to obtain test data of the target object.

[0025] Exemplarily, when testing a target object, a test case is created as a first test model, and the test case is injected through the ssh / sftp / ftp protocol to test the target object, thereby obtaining test data of the test case on the target object.

[0026] When creating a test case, you can create a test case with unknown test results, such as a test condition known but an unknown result; or a test case with a known test condition and a known test result.

[0027] Optionally, the method of injecting the test case into the target object for testing can be ssh / sftp / ftp protocol or http / https and other protocols that support information transmission. This application does not make specific restrictions and can be set according to needs.

[0028] Exemplarily, after the first test model is injected into the data collection service, the data collection service can be executed according to the steps of the first test model, or a scenario that meets the test requirements can be constructed according to the definition of the first test model, wherein the execution method can be an operating system script or an executable code in a certain programming language.

[0029] In one embodiment, a first test model is obtained, specifically, referring to Figure 2 , step S101 includes: sub-step S1011 to sub-step S1013.

[0030] Sub-step S1011, determining the test function points of the target object and the indicator parameters corresponding to the test function points.

[0031] Exemplarily, the creation, storage, injection, execution, and acquisition of the first test model are placed in the chaos test platform. When creating the first test model, the first test model is determined based on the test function points of the target object and the indicator parameters corresponding to the test function points.

[0032] For example, the target object is a data collection service, the test function point is the CPU load of the data collection service, and the indicator parameters are the number of CPU cores and the operating frequency of a single core.

[0033] Sub-step S1012, establishing a corresponding initial test model according to the test function points and the indicator parameters, and storing the initial test model in a corresponding data platform.

[0034] Exemplarily, a corresponding initial test model is established in the chaos test platform according to the test function points and indicator parameters, and then the initial test model is stored, so as to facilitate acquisition during subsequent testing of the target object.

[0035] For example, the existing chaos testing platform can establish an initial test model for the test function points of the target object. When the initial test model is established and saved, the initial test model is stored in a data storage platform that is connected to the chaos testing platform for communication. If the existing intelligent fault diagnosis platform that is connected to the data storage platform for communication needs to use the initial test model, it can be obtained from the data storage platform.

[0036] Optionally, the initial test model may be stored in a file or a database, and the storage format may be a custom binary or recognizable text encoding.

[0037] Sub-step S1013, obtaining a first test model corresponding to the test requirement from the data storage platform according to the test requirement.

[0038] Exemplarily, after the test requirements of the target object are determined, the test requirements are sent to the data storage platform for screening and matching, and when the matching results meet preset conditions, the screening results are used as the first test model corresponding to the test requirements.

[0039] For example, if the test requirement is a test case in which the target object A is missing data in the data collection service, after the test requirement is sent to the data storage platform, the test requirement is disassembled to obtain the target object A, the data collection service, and the missing data, and then screening and matching are performed based on the disassembled results to obtain the matching degree of the screened test cases. If the matching degree is greater than 90%, the corresponding test case is used as the first test model.

[0040] Step S102: acquiring first abnormal data in the test data, and performing fault type diagnosis on the first abnormal data according to the initial fault diagnosis model to obtain a diagnosis result.

[0041] Exemplarily, first abnormal data in the test data is obtained, and the first abnormal data is input into an initial fault diagnosis model for abnormal classification to implement fault type diagnosis, thereby determining a diagnosis result according to the abnormal classification result.

[0042] For example, the initial fault diagnosis model is to classify the communication anomalies and internal anomalies of the data acquisition service in the data acquisition service. After obtaining the first abnormal data in the test data, the first abnormal data can be input into the initial fault diagnosis model to classify the communication anomalies and internal anomalies of the acquisition service to obtain the diagnosis result corresponding to the first abnormal data.

[0043] In one embodiment, obtaining first abnormal data in test data includes: performing data conversion on the test data to obtain structured data, and preprocessing the structured data to obtain target data; performing abnormal information analysis on the target data, and then obtaining the first abnormal data from the target data.

[0044] Exemplarily, the test data is converted according to certain rules and formats to obtain structured data, for example, binary data is converted into structured data such as CSV and XML, and then the structured data is preprocessed by cleaning and normalization to obtain target data, and then the target data is analyzed for abnormal information to obtain the first abnormal data in the abnormal information.

[0045] In one embodiment, the abnormal scenarios of the initial fault diagnosis model include at least data missing anomalies, data delay anomalies, and data association anomalies caused by problems in the communication data acquisition service. The fault type of the first abnormal data is diagnosed according to the initial fault diagnosis model. Before obtaining the diagnosis result, it also includes: establishing a second test model, and using the second test model to test the target object to obtain the test result of the target object, and then using the test result to obtain the second abnormal data; comparing and analyzing the abnormal conditions corresponding to the second abnormal data with the data missing anomaly, data delay anomaly, and data association anomaly, to obtain the target abnormal scenario corresponding to the second abnormal data; determining the initial fault diagnosis model according to the second abnormal data and the target abnormal scenario.

[0046] Exemplarily, the abnormal scenarios of the initial fault diagnosis model are divided into data missing anomaly, data delay anomaly, and data association anomaly caused by problems in the communication data acquisition service, wherein the data missing anomaly refers to the situation that during the acquisition, decoding, output, and storage of the communication data acquisition service, the base station reports data continuously, but the data obtained in the communication data acquisition service is discontinuous, which is a data missing anomaly caused by problems in the communication data acquisition service; the data delay anomaly refers to the situation that the difference between the data acquisition time and the data generation time of the communication data acquisition service exceeds a threshold, which is a data delay anomaly caused by problems in the communication data acquisition service; the data association anomaly refers to the situation that any one or more of the data correlation rate, data statistical indicator accuracy, etc. in the collected data obtained in the communication data acquisition service is lower than the preset standard, which is a data association anomaly caused by problems in the communication data acquisition service.

[0047] Exemplarily, a second test model is established in the chaos test platform, and the second test model is injected into the target object to perform the test to obtain the test result, and the test result is analyzed to obtain the second abnormal data. The second abnormal data is compared and analyzed with the abnormal conditions corresponding to the data missing abnormality, data delay abnormality and data association abnormality to obtain the target abnormal scenario corresponding to the second abnormal data. The second abnormal data and the target abnormal scenario constitute a data pair, and training data is formed according to the data pair, so as to determine the initial fault diagnosis model according to the training data and the model target.

[0048] For example, the training data includes {second abnormal data A: data missing abnormality, second abnormal data B: data delay abnormality, ..., second abnormal data N: data association abnormality}, the model goal is that the accuracy on the model test data is greater than 90%, the loss value of the model training loss function is less than 0.8, and so on, and then training is performed based on the above training data. When the accuracy of the obtained training model on the model test data is greater than 90%, and the loss value of the model training loss function is less than 0.8, the obtained training model is determined as the initial fault diagnosis model.

[0049] In one embodiment, a target abnormal scenario corresponding to the second abnormal data is obtained by comparing and analyzing the abnormal conditions corresponding to the second abnormal data and the data missing abnormality, the data delay abnormality and the data association abnormality, including: when the second abnormal data satisfies the data discontinuity within a preset time period, the target abnormal scenario corresponding to the second abnormal data is the data missing abnormality; when the time difference between the output time and the generation time corresponding to the second abnormal data satisfies the preset time difference, the target abnormal scenario corresponding to the second abnormal data is the data delay abnormality; when the data association rate corresponding to the second abnormal data is lower than the preset association rate, the target abnormal scenario corresponding to the second abnormal data is the data association abnormality.

[0050] Exemplarily, when the second abnormal data is obtained, the second abnormal data is matched with the abnormal condition corresponding to the abnormal scenario type set in the abnormal scenario. If the data collected by the communication data collection service is reported continuously, the collected data collected by the communication data collection service is discontinuous within a preset time period (30 minutes), then it is determined that the target abnormal scenario corresponding to the second abnormal data obtained by the communication data collection service is a data missing abnormality.

[0051] Exemplarily, if the user has a standard time requirement for the data collection time, when the communication data collection service performs data collection, if the difference between the time of data output and the time when the communication data collection service generates data for data collection is greater than the standard time requirement, it is determined that the target abnormal scenario corresponding to the second abnormal data obtained by the communication data collection service is a data delay abnormality.

[0052] Exemplarily, if the communication data collection service is used to collect correlation data, such as CDT (Call Detail Trace) obtaining user call information in real time within a certain time period, but the call record correlation rate obtained by the communication data collection service in a certain time period is lower than the preset correlation rate, then the target abnormal scenario corresponding to the second abnormal data obtained by the communication data collection service is determined to be a data correlation abnormality.

[0053] Step S103: Under the condition that it is confirmed according to the diagnosis result that the fault type corresponding to the first abnormal data is an unknown fault type, an abnormal root cause analysis is performed on the first abnormal data to obtain a root cause analysis result.

[0054] Exemplarily, when the initial fault diagnosis model is trained, the reliability of subsequent model use is not guaranteed, and the category of unknown abnormalities is added to the abnormality category of the initial fault diagnosis model. That is, the abnormality category corresponding to the training data does not belong to any of the data missing abnormality, data delay abnormality, and data correlation abnormality caused by problems in the communication data acquisition service. Therefore, when the initial fault diagnosis model is used for fault diagnosis, the unknown fault type can be confirmed therefrom.

[0055] Exemplarily, when the initial fault diagnosis model performs fault diagnosis on the first abnormal data, the fault type obtained is an unknown fault type, then the first abnormal data is matched with the existing data missing anomalies, data delay anomalies, and data association anomalies, and the anomaly corresponding to the maximum matching result is used as the root cause analysis result.

[0056] For example, cluster analysis is performed on the first abnormal data and the target abnormal data corresponding to the data missing abnormality, data delay abnormality, and data association abnormality, respectively, and the category with the most type data corresponding to the data missing abnormality, data delay abnormality, and data association abnormality in the cluster where the first abnormal data is located is taken as the root cause analysis result of the first abnormal data.

[0057] In one embodiment, performing an abnormal root cause analysis on the first abnormal data to obtain a root cause analysis result includes: performing fault identification on the first abnormal data using a fault diagnosis identification model to obtain a fault identification result; and determining a root cause analysis result corresponding to the first abnormal data according to the fault identification result.

[0058] Exemplarily, a fault diagnosis and identification model is established based on a known fault type. When the first abnormal data is of an unknown fault type, the fault diagnosis and identification model is used to perform fault identification on the first abnormal data to obtain a fault identification result. If the fault identification result meets a preset condition, a root cause analysis result corresponding to the first abnormal data is obtained.

[0059] For example, if the fault diagnosis and identification model is a fault relationship extraction model, the first abnormal data is used to obtain the relationship between data in the first abnormal data using the fault relationship extraction model, and then the corresponding root cause analysis result corresponding to the first abnormal data is determined according to the relationship between data.

[0060] In one embodiment, a fault diagnosis and recognition model is used to perform fault identification on the first abnormal data to obtain a fault identification result, including: classifying the first abnormal data according to the fault diagnosis and recognition model to obtain a fault classification result corresponding to the first abnormal data; and determining the fault identification result corresponding to the first abnormal data according to the fault classification result.

[0061] Exemplarily, the first abnormal data is fault-classified using a fault diagnosis and identification model to obtain a corresponding fault classification result, which indicates that the fault type corresponding to the first abnormal data is a known fault type, and then the fault identification result corresponding to the first abnormal data is determined according to the fault classification result.

[0062] In one embodiment, fault classification is performed on the first abnormal data according to a fault diagnosis and recognition model to obtain a fault classification result corresponding to the first abnormal data, including: performing a similarity analysis on the first abnormal data according to the fault diagnosis and recognition model to obtain a similarity result; fault classification is performed on the first abnormal data according to the similarity result, and then a fault classification result corresponding to the first abnormal data is obtained.

[0063] Exemplarily, target abnormal data corresponding to a known fault type is obtained, the target abnormal data is represented by a vector according to a fault diagnosis and recognition model, a first target vector corresponding to the target abnormal data is obtained, the first abnormal data is also represented by a vector through the fault diagnosis and recognition model, a second target vector corresponding to the first abnormal data is obtained, and similarity is calculated between the first target vector and the second target vector to obtain a similarity result, thereby taking the known fault type corresponding to the maximum similarity result as the fault classification result corresponding to the first abnormal data.

[0064] For example, the target abnormal data corresponding to the data missing anomaly are data1, data2, data3, and the target abnormal data corresponding to the data association anomaly are data5, data6, and data7. Then, data1, data2, data3, data5, data6, and data7 are respectively represented by vectors to obtain the first target vector: v1, v2, v3, v5, v6, and v7. Then, the second target vector corresponding to the first abnormal data is v8. Then, the similarity between v1, v2, v3, v5, v6, and v7 and v8 is calculated to obtain sim1, sim2, sim3, sim5, sim6, and sim7. The data missing anomaly The similarity results in are sim1, sim2, and sim3, and the similarity results in the data-associated anomaly are sim5, sim6, and sim7. The maximum value max(sim1, sim2, sim3) of the similarity result in the data-missing anomaly is taken as the probability that the first abnormal data is the data-missing anomaly; the maximum value max(sim5, sim6, sim7) of the similarity result in the data-associated anomaly is taken as the probability that the first abnormal data is the data-associated anomaly, and then the category corresponding to the maximum value of max(max(sim1, sim2, sim3), max(sim5, sim6, sim7)) is obtained, which is the fault classification result corresponding to the first abnormal data.

[0065] Optionally, in one embodiment, if the root cause analysis result corresponding to the first abnormal data cannot be obtained according to the fault diagnosis and identification model, the first test model can be adjusted to reduce or increase indicator parameters, etc., to adjust the influencing factors of the test function point, so that the first test model is more concise and easier to diagnose the root cause, and then the root cause analysis result corresponding to the first abnormal data can be obtained according to the elimination method. When the newly obtained first test model re-triggers the same test process and obtains a test result that is the same as the root cause analysis result, the process of determining the root cause is terminated, and the root cause analysis result corresponding to the first abnormal data is used as a new known fault result.

[0066] Optionally, after obtaining the fault diagnosis identification model, the fault diagnosis identification model is stored in an intelligent fault diagnosis platform, wherein the intelligent fault diagnosis platform is used to obtain monitoring data of the collection service in real time. If the first abnormal data determines the root cause analysis result, the first test model is re-triggered to reproduce and confirm the fault, forming a self-feedback process. When the root cause analysis result is the same as the known fault result, the process ends.

[0067] Optionally, under the condition that it is confirmed according to the diagnosis result that the fault type corresponding to the first abnormal data is a known fault type, the fault diagnosis process of the communication data acquisition service ends.

[0068] Step S104: adjusting the model parameters of the initial fault diagnosis model according to the root cause analysis result to obtain a target fault diagnosis model, so as to perform fault diagnosis on the target abnormal data of the target object using the target fault diagnosis model.

[0069] Exemplarily, after obtaining the root cause analysis result corresponding to the unknown fault type, the first abnormal data and the root cause analysis result are used as new training data, and then the model parameters of the initial fault diagnosis model are adjusted to obtain a target fault diagnosis model that can identify the first abnormal data, so that after receiving the new abnormal data, the target abnormal data of the target object is diagnosed according to the target fault diagnosis model.

[0070] For example, after taking the first abnormal data and the root cause analysis result as newly added training data, parameter adjustment can be performed on the basis of the initial fault diagnosis model to obtain a target fault diagnosis model that can identify the first abnormal data.

[0071] Alternatively, the original training data corresponding to the initial fault diagnosis model and the newly added training data corresponding to the first abnormal data are combined to form the latest training data, and the target fault diagnosis model is obtained by performing model training using the latest training data.

[0072] Please refer to Figure 3 , Figure 3 A schematic diagram of a scenario for implementing the fault diagnosis method of the communication data collection service provided in this embodiment, such as Figure 3 As shown, first, a first test model is established in the chaos test platform. After the first test model is established, the execution of the model is triggered by the chaos test platform, thereby starting the chaos test. During the test, the communication data acquisition service is monitored, and the monitoring data is collected in real time. The monitoring data is sent from the monitoring platform to the intelligent fault diagnosis platform for analysis by real-time reporting or real-time acquisition, and the collected data is judged to be abnormal. If an abnormality occurs, fault diagnosis is required. If no abnormality occurs, the process ends. If fault diagnosis is required, it is determined whether it is a known fault. If it is a known fault, the process ends; if it is an unknown fault type, the root cause analysis of the fault is performed, and it is determined whether the root cause can be diagnosed. If the root cause can be diagnosed, the test case will be re-triggered to reproduce and confirm the fault, forming a self-feedback process; if the root cause cannot be diagnosed, the first test model will be adjusted to perform chaos testing until the root cause of the fault is determined. After the root cause analysis result is determined, the abnormal data is used as new training data for training to obtain the target fault diagnosis model, so that the target fault diagnosis model performs fault diagnosis on the target abnormal data of the target object.

[0073] See also Figure 4 , Figure 4 A schematic block diagram of the structure of a terminal device provided in an embodiment of the present invention.

[0074] like Figure 4 As shown, the terminal 300 includes a processor 301 and a memory 302, and the processor 301 and the memory 302 are connected via a bus 303, such as an I2C (Inter-integrated Circuit) bus.

[0075] Specifically, the processor 301 is used to provide computing and control capabilities to support the operation of the entire terminal. The processor 301 can be a central processing unit (CPU), and the processor 301 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0076] Specifically, the memory 302 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0077] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a partial structure related to the embodiment of the present invention, and does not constitute a limitation on the terminal to which the embodiment of the present invention is applied. The specific server may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0078] The processor is used to run a computer program stored in the memory, and implement any one of the communication data acquisition service fault diagnosis methods provided by the embodiments of the present invention when executing the computer program.

[0079] In one embodiment, the processor 301 is used to run a computer program stored in a memory, and implement the following steps when executing the computer program:

[0080] Obtaining a first test model, and using the first test model to test a target object to obtain test data of the target object;

[0081] Acquire first abnormal data in the test data, and perform fault type diagnosis on the first abnormal data according to an initial fault diagnosis model to obtain a diagnosis result;

[0082] Under the condition that the fault type corresponding to the first abnormal data is confirmed to be an unknown fault type according to the diagnosis result, performing abnormal root cause analysis on the first abnormal data to obtain a root cause analysis result;

[0083] The model parameters of the initial fault diagnosis model are adjusted according to the root cause analysis result to obtain a target fault diagnosis model, so as to perform fault diagnosis on the target abnormal data of the target object using the target fault diagnosis model.

[0084] In one embodiment, the processor 301 executes, in the process of obtaining the first test model:

[0085] Determine the test function points of the target object and the indicator parameters corresponding to the test function points;

[0086] Establishing a corresponding initial test model according to the test function points and the indicator parameters, and storing the initial test model in a corresponding data platform;

[0087] According to the test requirement, a first test model corresponding to the test requirement is obtained from the data storage platform.

[0088] In one embodiment, the processor 301 executes, during the process of acquiring the first abnormal data in the test data:

[0089] Convert the test data into structured data, and preprocess the structured data into target data;

[0090] An abnormal information analysis is performed on the target data, and then first abnormal data is obtained from the target data.

[0091] In one embodiment, the abnormal scenarios of the initial fault diagnosis model include at least data missing abnormality, data delay abnormality, and data association abnormality caused by problems in the communication data acquisition service. The processor 301 performs fault type diagnosis on the first abnormal data according to the initial fault diagnosis model, and before obtaining the diagnosis result, the processor 301 further executes:

[0092] Establishing a second test model, and using the second test model to test the target object to obtain a test result of the target object, and then using the test result to obtain second abnormal data;

[0093] Comparing and analyzing the abnormal conditions corresponding to the second abnormal data and the data missing abnormality, the data delay abnormality, and the data association abnormality, to obtain a target abnormal scenario corresponding to the second abnormal data;

[0094] The initial fault diagnosis model is determined according to the second abnormal data and the target abnormal scenario.

[0095] In one embodiment, the processor 301 performs comparative analysis according to the abnormal conditions corresponding to the second abnormal data and the data missing abnormality, the data delay abnormality, and the data association abnormality to obtain a target abnormal scenario corresponding to the second abnormal data:

[0096] When the second abnormal data satisfies the data discontinuity within the preset time period, the target abnormal scenario corresponding to the second abnormal data is data missing abnormality;

[0097] When the time difference between the output time and the generation time corresponding to the second abnormal data satisfies the preset time difference, the target abnormal scenario corresponding to the second abnormal data is a data delay abnormality;

[0098] When the data association rate corresponding to the second abnormal data is lower than a preset association rate, the target abnormal scenario corresponding to the second abnormal data is a data association abnormality.

[0099] In one embodiment, the processor 301 performs the following steps when performing an abnormality root cause analysis on the first abnormal data to obtain a root cause analysis result:

[0100] Using a fault diagnosis and identification model to perform fault identification on the first abnormal data to obtain a fault identification result;

[0101] A root cause analysis result corresponding to the first abnormal data is determined according to the fault identification result.

[0102] In one embodiment, the processor 301 performs the following steps when performing fault identification on the first abnormal data using the fault diagnosis identification model to obtain a fault identification result:

[0103] Performing fault classification on the first abnormal data according to the fault diagnosis and identification model to obtain a fault classification result corresponding to the first abnormal data;

[0104] A fault identification result corresponding to the first abnormal data is determined according to the fault classification result.

[0105] In one embodiment, when the processor 301 classifies the first abnormal data according to the fault diagnosis and identification model to obtain the fault classification result corresponding to the first abnormal data, the processor 301 executes:

[0106] Performing similarity analysis on the first abnormal data according to the fault diagnosis and recognition model to obtain a similarity result;

[0107] Fault classification is performed on the first abnormal data according to the similarity result, so as to obtain a fault classification result corresponding to the first abnormal data.

[0108] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the terminal described above can refer to the corresponding process in the aforementioned communication data acquisition service fault diagnosis method embodiment, which will not be repeated here.

[0109] An embodiment of the present invention further provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of any communication data acquisition service fault diagnosis method provided in the specification of the embodiment of the present invention.

[0110] The storage medium may be an internal storage unit of the terminal described in the foregoing embodiment, such as a hard disk or memory of the terminal. The storage medium may also be an external storage terminal of the terminal, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal.

[0111] 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 disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, 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 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 include, but are 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 tapes, 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 of ordinary skill in the art that communication media typically contain 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.

[0112] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0113] The serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for diagnosing a fault in a communication data collection service, characterized in that: The method comprises: Obtaining a first test model, and using the first test model to test a target object to obtain test data of the target object; Acquire first abnormal data in the test data, and perform fault type diagnosis on the first abnormal data according to an initial fault diagnosis model to obtain a diagnosis result; Under the condition that the fault type corresponding to the first abnormal data is confirmed to be an unknown fault type according to the diagnosis result, performing abnormal root cause analysis on the first abnormal data to obtain a root cause analysis result; The model parameters of the initial fault diagnosis model are adjusted according to the root cause analysis result to obtain a target fault diagnosis model, so as to perform fault diagnosis on the target abnormal data of the target object using the target fault diagnosis model.

2. The method according to claim 1, characterized in that: The obtaining of the first test model comprises: Determine the test function points of the target object and the indicator parameters corresponding to the test function points; Establishing a corresponding initial test model according to the test function points and the indicator parameters, and storing the initial test model in a corresponding data platform; According to the test requirement, a first test model corresponding to the test requirement is obtained from the data storage platform.

3. The method according to claim 1, characterized in that: The obtaining of first abnormal data in the test data includes: Performing data conversion on the test data to obtain structured data, and preprocessing the structured data to obtain target data; An abnormal information analysis is performed on the target data, and then first abnormal data is obtained from the target data.

4. The method according to claim 1, characterized in that: The abnormal scenarios of the initial fault diagnosis model at least include data missing abnormalities, data delay abnormalities, and data association abnormalities caused by problems in the communication data acquisition service. Before the fault type diagnosis of the first abnormal data is performed according to the initial fault diagnosis model and the diagnosis result is obtained, the method further includes: Establishing a second test model, and using the second test model to test the target object to obtain a test result of the target object, and then using the test result to obtain second abnormal data; Comparing and analyzing the abnormal conditions corresponding to the second abnormal data and the data missing abnormality, the data delay abnormality, and the data association abnormality, to obtain a target abnormal scenario corresponding to the second abnormal data; The initial fault diagnosis model is determined according to the second abnormal data and the target abnormal scenario.

5. The method according to claim 4, characterized in that: The comparing and analyzing the abnormal conditions corresponding to the second abnormal data and the data missing abnormality, the data delay abnormality, and the data association abnormality to obtain a target abnormal scenario corresponding to the second abnormal data includes: When the second abnormal data satisfies the data discontinuity within the preset time period, the target abnormal scenario corresponding to the second abnormal data is data missing abnormality; When the time difference between the output time and the generation time corresponding to the second abnormal data satisfies the preset time difference, the target abnormal scenario corresponding to the second abnormal data is a data delay abnormality; When the data association rate corresponding to the second abnormal data is lower than a preset association rate, the target abnormal scenario corresponding to the second abnormal data is a data association abnormality.

6. The method according to claim 1, characterized in that: The performing abnormal root cause analysis on the first abnormal data to obtain a root cause analysis result includes: Using a fault diagnosis and identification model to perform fault identification on the first abnormal data to obtain a fault identification result; A root cause analysis result corresponding to the first abnormal data is determined according to the fault identification result.

7. The method according to claim 6, characterized in that: The using the fault diagnosis and identification model to perform fault identification on the first abnormal data to obtain a fault identification result includes: Performing fault classification on the first abnormal data according to the fault diagnosis and identification model to obtain a fault classification result corresponding to the first abnormal data; A fault identification result corresponding to the first abnormal data is determined according to the fault classification result.

8. The method according to claim 7, characterized in that: The performing fault classification on the first abnormal data according to the fault diagnosis and identification model to obtain a fault classification result corresponding to the first abnormal data includes: Performing similarity analysis on the first abnormal data according to the fault diagnosis and recognition model to obtain a similarity result; Fault classification is performed on the first abnormal data according to the similarity result, so as to obtain a fault classification result corresponding to the first abnormal data.

9. A terminal device, characterized in that: The terminal device includes a processor and a memory; The memory is used to store computer programs; The processor is used to execute the computer program and implement the fault diagnosis method for the communication data collection service according to any one of claims 1 to 8 when executing the computer program.

10. A computer-readable storage medium, characterized in that: When the computer-readable storage medium is executed by one or more processors, the one or more processors execute the steps of the fault diagnosis method for the communication data collection service according to any one of claims 1 to 8.

Images