Communication equipment monitoring method and device and medium

Abstract

The invention discloses a communication equipment monitoring method and device and a medium, and relates to the technical field of communication security, and the method comprises the steps: obtaining the real-time monitoring data of communication equipment, extracting the features of the real-time monitoring data, carrying out the clustering analysis, obtaining various types of classification index data, stacking the time sequence analysis and deep learning, and carrying out the clustering analysis. Constructing an abnormal operation monitoring model in combination with the obtained historical abnormal data of the plurality of periods; inputting the classification index data into an abnormal operation monitoring model, judging whether a classification index result is within a set threshold range or not, and switching an operation mainboard of the communication equipment to a standby mainboard when an identification result shows abnormality; setting a target function to minimize a false alarm rate, performing parameter adjustment on an abnormal operation monitoring model of an abnormal mainboard in combination with an abnormal identification result, inputting classification index data into the abnormal operation monitoring model of the adjusted mainboard, when an output result is normal, taking the adjusted abnormal mainboard as a standby mainboard, and if the output result is abnormal, taking the adjusted abnormal mainboard as the standby mainboard. And sending an abnormal identification result as an abnormal alarm.

Description

Technical Field

[0001] The present invention relates to the field of communication security, and in particular to a communication equipment monitoring method, device and medium. Background Art

[0002] With the rapid development of the Internet of Things and communication technology, communication equipment is a core component to maintain the normal operation of the communication system. Monitoring its operating status can timely discover and solve potential faults and ensure the stability and reliability of the communication system. The data capacity that communication equipment has to process is getting larger and larger, and the structure is becoming more and more complex. The use of communication equipment (such as switches, routers, smart boxes), environmental equipment (such as temperature, humidity, water immersion, smoke), power equipment (such as switching power supplies, uninterruptible power supplies UPS, battery packs, etc.), air conditioning equipment, etc. in the system is also increasing. The on-site use environment of these devices is complex and may be affected by various human and natural factors (such as lightning strikes, static electricity, power outages), causing equipment failures and shutdowns. These devices are relatively independent of each other. If a failure occurs between each device, it will not be possible to issue a warning to the outside world to indicate the occurrence of the failure.

[0003] The traditional communication equipment management system basically manages the equipment maintenance at each communication site. Each communication site is equipped with a certain number of operation and maintenance personnel to conduct inspections and equipment maintenance and repair. The amount of data processed by existing communication monitoring devices is very large, and it is very easy to have abnormal conditions caused by operating data, and then there are false alarms or missed alarms. Due to the low monitoring accuracy, it is difficult for administrators to accurately judge the operating status of the equipment and cannot timely discover abnormal phenomena in communication equipment. It may happen that the equipment is damaged and cannot work normally, and it is not discovered that the communication equipment needs to be repaired and replaced until the normal operation is affected. Early warning and early maintenance are not possible. Summary of the invention

[0004] In order to overcome the above-mentioned prior art communication monitoring devices that process a very large amount of data, it is very easy to have abnormal situations caused by operating data, and then there are false alarms or missed alarms. Due to the insufficient monitoring accuracy, it is difficult for administrators to accurately judge the operating status of the equipment and cannot promptly discover abnormal phenomena in the communication equipment. It may happen that the equipment is damaged and cannot work normally, and it is not discovered until the normal operation is affected that the communication equipment needs to be repaired and replaced, and it is impossible to provide early warning and early maintenance. The main purpose of the present invention is to provide a communication equipment monitoring method, device and medium.

[0005] To achieve the above object, the present invention adopts the following technical solution, a communication equipment monitoring method, comprising the following steps:

[0006] Acquire real-time monitoring data of communication equipment, extract features of real-time monitoring data and perform cluster analysis to obtain multiple classification index data, and store the multiple classification index data in a distributed manner;

[0007] Stacking time series analysis with deep learning, combined with historical abnormal data from multiple cycles, to build an abnormal operation monitoring model;

[0008] Input the classification indicator data into the abnormal operation monitoring model to determine whether the classification indicator result is within the set threshold range. When the identification result shows an abnormality, switch the operating mainboard of the communication equipment to the standby mainboard;

[0009] The objective function is set to minimize the false alarm rate. The parameters of the abnormal operation monitoring model of the abnormal mainboard are adjusted in combination with the abnormal recognition results to obtain the adjusted mainboard. The classification index data is input into the abnormal operation monitoring model of the adjusted mainboard. When the output result is normal, the adjusted abnormal mainboard is used as a spare mainboard. If it is abnormal, the abnormal recognition result is issued as an abnormal alarm.

[0010] Real-time data includes equipment operating status, performance parameters, and sensor data. The classification indicator results include data collection volume, collection time for a one-time collection of a set volume, and processing time per unit data volume.

[0011] The time series analysis model is stacked with the random forest model, and the historical abnormal data of multiple periods is used as input to build an abnormal operation monitoring model, including the following steps:

[0012] The time series analysis model selects ARIMA to fit the acquired historical abnormal data of multiple periods to obtain the residual between the predicted value and the actual value;

[0013] Extract features from the residual sequence, and use historical abnormal data to extract features before the abnormality occurs, and obtain the residual features and abnormal event features predicted by the model;

[0014] The residual features and abnormal event features predicted by the model are used as feature sets, and the random forest model is trained using the feature sets to build an abnormal operation monitoring model.

[0015] The method of extracting the features of the real-time data and performing cluster analysis to obtain a variety of classification index data includes the following steps:

[0016] Perform various features on the preprocessed real-time data to obtain various features of the real-time data, including time domain, frequency domain, time-frequency domain, statistical and dynamic features;

[0017] The multiple features extracted are reduced in dimension through principal component analysis to obtain multiple feature data after dimension reduction;

[0018] Hierarchical clustering is performed on the various feature data after dimensionality reduction to obtain various classification feature data as indicators.

[0019] The step of inputting the classification indicator data into the abnormal operation monitoring model and determining whether the classification indicator result meets the set threshold range includes the following steps:

[0020] Obtain pre-processed classification indicator data from the distributed storage system and input it into the abnormal operation monitoring model;

[0021] In the abnormal operation monitoring model, the threshold range of the classification index is pre-set and compared with the actual observed value;

[0022] Compare each classification indicator with the corresponding threshold to determine whether it exceeds the set range. If it exceeds the range, it will be marked as abnormal data;

[0023] For the marked abnormal data, anomaly identification is performed to determine the type and severity of the anomaly.

[0024] It also includes recording abnormal data, including the time when the abnormality occurred, the abnormality type, and the abnormal indicator value.

[0025] The objective function is set to minimize the false alarm rate, and the parameters of the abnormal motherboard are adjusted in combination with the abnormal identification result, including the following steps:

[0026] The objective function is set to minimize the false alarm rate, and the anomaly recognition result is used as input;

[0027] According to the abnormal recognition result and the function of setting the target, the target function value under the current parameter setting is obtained;

[0028] Using selected simulated annealing, searching for the parameter combination that makes the objective function optimal in the parameter space, and obtaining the iterative result;

[0029] According to the iterative results of the optimization algorithm, the parameter values ​​of the abnormal operation monitoring model of the abnormal mainboard are updated.

[0030] The simulated annealing comprises:

[0031] Randomly initialize the abnormal operation monitoring model parameter values ​​and operation strategies of the abnormal mainboard, and set the initial temperature and cooling rate;

[0032] Iterate optimization within a certain temperature range until the stop condition is reached, fine-tune or randomly change the current parameter values ​​and operating strategies, and generate new parameter values ​​and operating strategies;

[0033] Calculate the objective function value based on the new parameter values ​​and operation strategy, including indicators such as abnormality occurrence rate and equipment performance stability;

[0034] Determine whether to accept the new parameter values ​​and operation strategy; if the new objective function value is better, accept it directly; otherwise, accept it with a certain probability, which is related to the difference between the new and old objective function values ​​and the current temperature;

[0035] When the set stop condition is reached and the number of iterations reaches the upper limit, the iteration ends;

[0036] The final converged parameter values ​​and running strategy are returned as the adjusted abnormal mainboard settings.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows: the present application can continuously monitor the operating status of communication equipment through real-time monitoring data acquisition and feature extraction, ensuring that any potential problems can be responded to quickly. Cluster analysis further helps to identify key performance indicators in a large amount of data, so that the equipment status can be monitored more accurately. The abnormal operation monitoring model uses time series analysis and deep learning technology (random forest model stacking) to enhance the recognition ability of abnormal patterns and improve the accuracy and reliability of detection. When the monitoring model detects an abnormality, the operating motherboard of the device can be automatically switched to the standby motherboard, thereby reducing system downtime and maintaining service continuity and stability. Setting the objective function to minimize the false alarm rate and optimizing the parameters of the monitoring model through the simulated annealing algorithm can significantly reduce the situation of false alarms and missed alarms, thereby improving the overall efficiency of the system and the trust of users. Through the use of the simulated annealing algorithm, the monitoring parameters can be self-adjusted in a dynamic environment, so that the abnormal detection model can continuously adapt to new operating conditions and abnormal patterns. This adaptive adjustment strategy improves the robustness and flexibility of the model in a changing environment.

[0038] The optimized parameter settings can be applied to further update and upgrade of the monitoring system, ensuring long-term stable operation and less maintenance requirements. The detailed abnormal data recorded by the system, such as the time, type and indicator value of the abnormality, provides rich data support for subsequent analysis and decision-making. This data can be used for trend analysis, performance evaluation and risk management, helping managers make more information-based and scientific decisions.

[0039] In summary, this technical solution improves the stability, reliability and operational efficiency of communication equipment through highly integrated monitoring, diagnosis and recovery functions, while reducing the time cost of manual intervention and troubleshooting, enhancing the real-time and intelligence of the system, and enhancing the ability to respond to complex environments and emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the structural flow of the present invention;

[0041] Figure 2 It is a schematic diagram of the process of data classification of the present invention;

[0042] Figure 3 It is a schematic diagram of the detection process of the anomaly detection model of the present invention;

[0043] Figure 4 It is a schematic diagram of the process of adjusting an abnormal mainboard according to the present invention. DETAILED DESCRIPTION

[0044] The present invention will be further described below in conjunction with the accompanying drawings and implementation modes.

[0045] Example:

[0046] See also Figure 1 A communication equipment monitoring method is to monitor relevant parameters of the communication equipment and monitor the relevant parameters to ensure the normal operation of the communication equipment, and to promptly handle the abnormal situation of the communication equipment when the communication equipment is abnormal, specifically including:

[0047] Real-time data collection and preprocessing of communication equipment, generation of classification indicator results, identification and optimization training of abnormal operation monitoring models, and handling of abnormal situations:

[0048] Among them, using sensors and monitoring devices to collect data from communication equipment in real time, real-time collection and preprocessing of equipment operating status, performance parameters and sensor data are key steps to ensure data quality and availability. Streaming processing technologies such as Apache Kafka, Apache Flink, etc. can be used to implement operating status, performance parameters such as temperature, voltage, frequency, etc., sensor data, etc., and preprocess the collected data, including data cleaning, denoising, normalization and other operations to ensure data quality and accuracy.

[0049] Extract features from the preprocessed data, such as the operating frequency of the equipment, temperature change trend, voltage fluctuation, etc., and use cluster analysis such as K-means clustering and DBSCAN to discover the potential patterns and cluster structures of the feature data, and obtain multiple classification indicator results, such as equipment status such as normal, warning, abnormal, operation performance score, etc., for subsequent abnormality detection and operation optimization. Store the extracted classification indicator data in distributed storage systems such as Hadoop HDFS, Apache Cassandra, etc. for fast access and processing.

[0050] By stacking time series analysis with deep learning and combining historical abnormal data from multiple cycles, an abnormal operation monitoring model is constructed. The classification indicator results are input into the abnormal operation monitoring model for identification and judgment. The model determines whether the classification indicator results are abnormal based on the preset threshold range. When the abnormal operation monitoring model identifies an abnormal situation, it immediately switches to the backup mainboard to ensure the continuous operation and reliability of the equipment and the stable operation of the communication system.

[0051] The abnormal data is used as a training set to optimize the training of the abnormal operation monitoring model to improve its ability and accuracy in identifying abnormal situations.

[0052] The objective function is set to minimize the false alarm rate. Combined with the abnormal identification results, the parameters and operation strategies of the abnormal motherboard are adjusted to try to solve the abnormal situation and optimize the operation performance and stability of the motherboard.

[0053] After adjustment, test again. If it is still abnormal, an error alarm will be sent to the remote control center to notify the administrator for further processing and scheduling.

[0054] During the entire process, the collection and processing of real-time data can promptly detect abnormal conditions of communication equipment. The generation of classification indicator results and the identification and optimization training of monitoring models can effectively monitor and respond to abnormal conditions and ensure the stable operation of the communication system.

[0055] The basic principle of a communication equipment monitoring device is that the collection and processing of real-time data can timely discover abnormal conditions of communication equipment, and the generation of classification indicator results and the identification and optimization training of monitoring models can effectively monitor and respond to abnormal conditions and ensure the stable operation of the communication system.

[0056] Example 2

[0057] A communication monitoring method, in a specific practical application, uses sensors to collect real-time data of equipment, including temperature, humidity, voltage, etc. The collected data is preprocessed, including data cleaning, denoising and normalization operations, to ensure the quality and availability of the data. Features, such as average temperature, maximum voltage, etc., are extracted from the preprocessed data. Cluster analysis technology, such as K-means clustering, is used to classify the device status into two categories: normal and abnormal. The extracted classification indicator data is stored in a distributed storage system, such as Hadoop HDFS. The classification indicator data is input into an abnormal operation monitoring model to detect whether there is an abnormality. If the output result is abnormal, the communication device will switch to a backup mainboard to ensure the continuous operation of the communication device. At the same time, the abnormal data is used as a training set to optimize the training and update the parameters of the abnormal operation monitoring model to improve the accuracy and robustness of the model. Based on a multi-objective optimization algorithm, the parameters and operation strategy of the abnormal mainboard are adjusted in combination with the abnormal recognition result to optimize the performance and stability of the device. If the output result of the adjusted abnormal mainboard is within the normal range, the adjusted mainboard is used as a backup mainboard.

[0058] If the output result of the abnormal motherboard after adjustment is still abnormal, the monitoring device will immediately issue an error alarm and notify the remote control center so that timely measures can be taken to deal with the problem. In this embodiment, the steps of real-time data collection, preprocessing, abnormality monitoring, and automatic switching are integrated together to build a complete fault detection and automatic switching system to ensure the reliable operation of the communication equipment under monitoring and the timely and effective handling of faults.

[0059] More specifically, the classification indicator results include the amount of data collected, the time it takes to collect a set amount of data at one time, and the processing time per unit of data volume.

[0060] Extracting the features of the preprocessed data and performing cluster analysis to obtain a variety of classification index data includes the following steps:

[0061] Perform various features on the preprocessed real-time data to obtain various features of the real-time data, including time domain, frequency domain, time-frequency domain, statistical and dynamic features;

[0062] The multiple features extracted are reduced in dimension through principal component analysis to obtain multiple feature data after dimension reduction;

[0063] Hierarchical clustering is performed on the various feature data after dimensionality reduction to obtain various classification feature data as indicators.

[0064] The classification indicator data is input into the abnormal operation monitoring model to determine whether the classification indicator result meets the set threshold range, including the following steps:

[0065] Obtain pre-processed classification indicator data from the distributed storage system and input it into the abnormal operation monitoring model;

[0066] Preset the threshold range of classification indicators in the abnormal operation monitoring model to compare with the actual observed values;

[0067] Compare each classification indicator with the corresponding threshold to determine whether it exceeds the set range. If it exceeds the range, it will be marked as abnormal data;

[0068] For the marked abnormal data, anomaly identification is performed to determine the type and severity of the anomaly.

[0069] Record abnormal data, including the time when the abnormality occurred, the abnormality type, and the abnormal indicator value.

[0070] Based on the multi-objective optimization algorithm, the parameters of the abnormal motherboard and the operation strategy adjustment are adjusted in combination with the abnormal identification results, including the following steps:

[0071] Taking the abnormality identification results as input, the multi-objective optimization algorithm selects simulated annealing, and defines reducing the occurrence rate of mainboard abnormalities as the optimization goal;

[0072] According to the abnormal recognition result and the function of the set target, the target function value under the current parameter setting is calculated;

[0073] Using selected simulated annealing, searching for the parameter combination that makes the objective function optimal in the parameter space, and obtaining the iterative result;

[0074] According to the iterative results of the optimization algorithm, the parameter values ​​and operation strategies of the abnormal motherboard are updated.

[0075] The simulated annealing includes:

[0076] Randomly initialize the parameter values ​​and operation strategies of the abnormal motherboard, and set the initial temperature and cooling rate;

[0077] Iterate optimization within a certain temperature range until the stop condition is reached, fine-tune or randomly change the current parameter values ​​and operating strategies, and generate new parameter values ​​and operating strategies;

[0078] Calculate the objective function value based on the new parameter values ​​and operation strategy, including indicators such as abnormality occurrence rate and equipment performance stability;

[0079] Determine whether to accept the new parameter values ​​and operation strategy; if the new objective function value is better, accept it directly; otherwise, accept it with a certain probability, which is related to the difference between the new and old objective function values ​​and the current temperature;

[0080] When the set stop condition is reached and the number of iterations reaches the upper limit, the iteration ends;

[0081] The final converged parameter values ​​and running strategy are returned as the settings of the adjusted anomaly mainboard.

[0082] Example 3

[0083] A specific implementation case is to monitor the server health status of a communication base station. This implementation case collects and analyzes real-time monitoring data of the server to identify and prevent potential failures. The following is a detailed description of the entire process, including the specific steps of data collection, model building, anomaly detection and parameter optimization:

[0084] Collects server CPU usage, memory usage, temperature readings, and network traffic data. Collects data every 5 minutes.

[0085] The collected data is analyzed in the time domain, frequency domain, and time-frequency domain to extract statistical and dynamic features. The features are reduced in dimension using principal component analysis (PCA). Hierarchical clustering is performed on the reduced-dimensional features to generate a variety of classification indicator data.

[0086] The ARIMA model is used to fit the time series of historical abnormal data.

[0087] Calculate the residual between the predicted value and the actual value, and extract features from the residual sequence. Combine historical abnormal data to extract features before the abnormal event occurs. Use the output of the time series model and the abnormal features as feature sets, and use these feature sets to train the random forest model.

[0088] The output of the model is set to be abnormal and non-abnormal. The classification indicator data is obtained from the distributed storage and input into the abnormal operation monitoring model. The model is judged according to the preset threshold range, and the data outside the range is marked as abnormal. If the data is marked as abnormal, the server will automatically switch to the backup motherboard.

[0089] The abnormal motherboard is automatically trimmed, the objective function is set to minimize the false alarm rate, and the parameters of the abnormal motherboard are adjusted in combination with the abnormal recognition results, that is, the initial temperature and cooling rate are set, and iterative optimization is performed until the stop condition is reached. According to the iterative results obtained by simulated annealing, the monitoring model parameters of the abnormal motherboard are updated. The classification index data is input into the abnormal operation monitoring model of the adjusted motherboard. When the output result is normal, the adjusted abnormal motherboard is used as a spare motherboard. If it is abnormal, the abnormal recognition result is sent to the remote control center as an abnormal alarm, and the staff is required to further adjust or replace the abnormal motherboard.

[0090] The results of the implementation case application show that in the early stage of implementation, the false alarm rate of the model was 5%, which was reduced to 1% after optimization. The false alarm rate calculation formula is:

[0091] The health monitoring system of communication equipment can identify potential problems in real time, respond to abnormal events in a timely manner, and continuously improve the stability and reliability of the system through intelligent optimization.

[0092] It should be noted that, in the present invention, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0093] The above embodiments are merely examples of the present invention and do not limit the protection scope of the present invention. All designs that are the same or similar to the present invention fall within the protection scope of the present invention.

Claims

1. A communication equipment monitoring method, characterized in that: The following steps are involved: Acquire real-time monitoring data of communication equipment, extract features of real-time monitoring data and perform cluster analysis to obtain multiple classification index data, and store the multiple classification index data in a distributed manner; Stacking time series analysis with deep learning, combined with historical abnormal data from multiple cycles, to build an abnormal operation monitoring model; Input the classification indicator data into the abnormal operation monitoring model to determine whether the classification indicator result is within the set threshold range. When the identification result shows an abnormality, switch the operating mainboard of the communication equipment to the standby mainboard; The objective function is set to minimize the false alarm rate. The parameters of the abnormal operation monitoring model of the abnormal mainboard are adjusted in combination with the abnormal recognition results to obtain the adjusted mainboard. The classification index data is input into the abnormal operation monitoring model of the adjusted mainboard. When the output result is normal, the adjusted abnormal mainboard is used as a spare mainboard. If it is abnormal, the abnormal recognition result is issued as an abnormal alarm.

2. The communication device monitoring method according to claim 1, characterized in that: The real-time data includes equipment operating status, performance parameters, and sensor data. The classification index results include data collection volume, collection time for a one-time collection of a set volume, and processing time per unit data volume.

3. The communication device monitoring method according to claim 1, characterized in that: The method of stacking the time series analysis model and the random forest model and combining the acquired historical abnormal data of multiple periods as input to construct an abnormal operation monitoring model includes the following steps: The time series analysis model selects ARIMA to fit the acquired historical abnormal data of multiple periods to obtain the residual between the predicted value and the actual value; Extract features from the residual sequence, and use historical abnormal data to extract features before the abnormality occurs, and obtain the residual features and abnormal event features predicted by the model; The residual features and abnormal event features predicted by the model are used as feature sets, and the random forest model is trained using the feature sets to build an abnormal operation monitoring model.

4. The communication device monitoring method according to claim 1, characterized in that: The method of extracting the features of the real-time data and performing cluster analysis to obtain a variety of classification index data includes the following steps: Perform various features on the preprocessed real-time data to obtain various features of the real-time data, including time domain, frequency domain, time-frequency domain, statistical and dynamic features; The multiple features extracted are reduced in dimension through principal component analysis to obtain multiple feature data after dimension reduction; Hierarchical clustering is performed on the various feature data after dimensionality reduction to obtain various classification feature data as indicators.

5. The communication equipment monitoring method according to claim 1, characterized in that: The step of inputting the classification indicator data into the abnormal operation monitoring model and determining whether the classification indicator result meets the set threshold range includes the following steps: Obtain pre-processed classification indicator data from the distributed storage system and input it into the abnormal operation monitoring model; In the abnormal operation monitoring model, the threshold range of the classification index is pre-set and compared with the actual observed value; Compare each classification indicator with the corresponding threshold to determine whether it exceeds the set range. If it exceeds the range, it will be marked as abnormal data; For the marked abnormal data, anomaly identification is performed to determine the type and severity of the anomaly.

6. The communication equipment monitoring method according to claim 1, characterized in that: It also includes recording abnormal data, including the time when the abnormality occurred, the abnormality type, and the abnormal indicator value.

7. The communication equipment monitoring method according to claim 1, characterized in that: The objective function is set to minimize the false alarm rate, and the parameters of the abnormal motherboard are adjusted in combination with the abnormal identification result, including the following steps: The objective function is set to minimize the false alarm rate, and the anomaly recognition result is used as input; According to the abnormal recognition result and the function of setting the target, the target function value under the current parameter setting is obtained; Using selected simulated annealing, searching for the parameter combination that makes the objective function optimal in the parameter space, and obtaining the iterative result; According to the iterative results of the optimization algorithm, the parameter values ​​of the abnormal operation monitoring model of the abnormal mainboard are updated.

8. A communication equipment monitoring method according to claim 7, characterized in that: The simulated annealing comprises: Randomly initialize the abnormal operation monitoring model parameter values ​​and operation strategies of the abnormal mainboard, and set the initial temperature and cooling rate; Iterate optimization within a certain temperature range until the stop condition is reached, fine-tune or randomly change the current parameter values ​​and operating strategies, and generate new parameter values ​​and operating strategies; Calculate the objective function value based on the new parameter values ​​and operation strategy, including indicators such as abnormality occurrence rate and equipment performance stability; Determine whether to accept the new parameter values ​​and operation strategy; if the new objective function value is better, accept it directly; otherwise, accept it with a certain probability, which is related to the difference between the new and old objective function values ​​and the current temperature; When the set stop condition is reached and the number of iterations reaches the upper limit, the iteration ends; The final converged parameter values ​​and running strategy are returned as the adjusted abnormal mainboard settings.

9. A communication equipment monitoring system, characterized in that: include: The data processing module is used to obtain the real-time monitoring data of the communication equipment, extract the characteristics of the real-time monitoring data and perform cluster analysis, obtain multiple classification index data, and store the multiple classification index data in a distributed manner; The anomaly detection model building module is used to stack time series analysis with deep learning, and combine the historical anomaly data of multiple periods to build an abnormal operation monitoring model; An exception handling module is used to input the classification index data into the abnormal operation monitoring model, determine whether the classification index result is within the set threshold range, and when the identification result shows an abnormality, switch the operating mainboard of the communication equipment to the standby mainboard; The monitoring abnormality warning module is used to set the objective function to minimize the false alarm rate, adjust the parameters of the abnormal operation monitoring model of the abnormal motherboard in combination with the abnormal identification results, obtain the adjusted motherboard, input the classification indicator data into the abnormal operation monitoring model of the adjusted motherboard, and when the output result is normal, the adjusted abnormal motherboard will be used as a spare motherboard. If it is abnormal, the abnormal identification result will be used and an error alarm will be issued.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images