Automatic operation and maintenance monitoring management method and system

Abstract

The invention relates to an automatic operation and maintenance monitoring management method and system, and belongs to the technical field of automatic operation and maintenance monitoring. The method comprises the steps that a system historical configuration data set and a system historical log data set are acquired, a fault data set and a system performance index data set are obtained through calculation of a classification model, and fault related parameters are obtained through calculation of a principal component analysis method; the method comprises the following steps: collecting real-time log data and real-time configuration parameters of a system, determining real-time performance data and real-time fault data, and constructing a performance data sequence and a configuration parameter sequence; a performance prediction result is obtained through calculation of the prediction model, a real-time fault related parameter sequence is obtained, and a fault prediction result is obtained through calculation of the diagnosis model; and obtaining a required performance index of the to-be-operated task, performing task scheduling according to the performance prediction result and the required performance index, and transmitting the performance prediction result, the real-time fault data and the fault prediction result to the operation and maintenance terminal by the intelligent terminal. And automatic operation and maintenance monitoring management of the system is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of automated operation and maintenance monitoring, and in particular relates to an automated operation and maintenance monitoring management method and system. Background Art

[0002] With the rapid development of information technology, various information systems are increasingly used in enterprises and organizations, and the stability and reliability of the system have become key factors in ensuring business continuity. Traditional operation and maintenance monitoring and management mainly rely on manual inspections and system alarms, which is not only inefficient, but also difficult to effectively prevent and predict failures before they occur. Therefore, automated operation and maintenance monitoring and management technology has emerged, aiming to achieve real-time monitoring of system status, fault warning and task scheduling through intelligent means, so as to improve operation and maintenance efficiency and service quality.

[0003] However, the existing automated operation and maintenance monitoring and management systems still have many deficiencies in data processing and fault prediction. On the one hand, when processing massive amounts of system historical configuration data and log data, the system often lacks effective data analysis and mining methods, making it difficult to accurately identify key parameters related to faults, thus affecting the accuracy and timeliness of fault prediction. On the other hand, in terms of real-time performance monitoring and fault prediction, most existing systems rely on simple threshold judgments or statistical methods, which often fail to provide accurate and reliable prediction results when faced with complex and changeable system faults.

[0004] In addition, most existing task scheduling strategies are based on static performance indicators and lack the prediction and evaluation of future performance changes, resulting in inflexible and inefficient task scheduling. At the same time, when receiving and processing operation and maintenance information, operation and maintenance personnel also face problems such as large amounts of information and low processing efficiency, making it difficult to quickly and accurately locate and resolve faults. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides an automated operation and maintenance monitoring management method and system. The purpose of the present invention can be achieved through the following technical solutions: Obtain a system history configuration data set and a system history log data set, calculate a fault data set and a system performance index data set based on the system history log data set through a classification model, and calculate fault-related parameters based on the system history configuration data set, the fault data set, and the system performance index data set through a principal component analysis method; Collect the real-time log data and real-time configuration parameters of the system through the log collection tool, calculate the real-time performance data and real-time fault data through the classification model according to the real-time log data, add timestamps according to the real-time performance data and the real-time configuration parameters to obtain time series performance data and time series configuration parameters, and store the time series performance data and the time series configuration parameters to obtain a performance data sequence and a configuration parameter sequence; According to the performance data sequence, a performance prediction result is obtained by calculating the prediction model, according to the fault-related parameters, the performance data sequence and the configuration parameter sequence are screened to obtain a real-time fault-related parameter sequence, and according to the real-time fault-related parameter sequence, a fault prediction result is obtained by calculating the diagnosis model; Obtain the required performance indicators of the tasks to be run, schedule the tasks according to the performance prediction results and the required performance indicators, obtain the operation and maintenance personnel information, and the intelligent terminal transmits the performance prediction results, the real-time fault data, and the fault prediction results to the operation and maintenance terminal according to the operation and maintenance personnel information.

[0006] Specifically, the specific calculation steps of the classification model are: According to the system history log data set, word segmentation data is obtained by using Jieba word segmentation calculation, and word vectors are obtained by using Skip-gram model calculation according to the word segmentation data; According to the word vector, a first feature vector and a second feature vector are calculated by a feature extraction model, the first feature vector is subjected to dimensionality reduction processing to obtain a reduced-dimensionality feature vector, and the reduced-dimensionality feature vector and the second feature vector are fused by a concat function to obtain a fused feature vector; The fused feature vector is refitted through a fully connected layer to obtain fitting data, a classification result is calculated through a softmax classifier based on the fitting data, and the system historical log data set is stored based on the classification to obtain the fault data set and the system performance indicator data set.

[0007] Specifically, the specific calculation steps of the feature extraction model include: Calculate a context feature vector according to the word vector through the LSTM model, concatenate the context feature vector and the word vector through the concat function to obtain a context concatenation feature vector, calculate a second context feature vector according to the context concatenation feature vector through the GRU, concatenate the context feature vector and the second context feature vector through the concat function to obtain a second concatenation vector, and calculate the first feature vector according to the second concatenation vector through the attention layer; A first local semantic feature vector and a second local semantic feature vector are calculated based on the word vector through two CNN models, a third concatenated vector is obtained by concatenating the first local semantic feature vector and the second local semantic feature vector through a concat function, and the second feature vector is calculated based on the third concatenated vector through an attention layer.

[0008] Specifically, the step of screening the performance data sequence and the configuration parameter sequence according to the fault-related parameters to obtain a real-time fault-related parameter sequence includes: The fault-related parameters include fault-related parameter name tags, and the performance data sequence and the configuration parameter sequence are screened according to the fault-related parameter name tags to obtain the real-time fault-related parameter sequence.

[0009] Specifically, the specific calculation steps of the prediction model are: Preset a time window, intercept the performance data sequence according to the time window to obtain a predicted performance sequence, fill missing values ​​according to the predicted performance sequence by using the mean method to obtain a filled sequence, and normalize the filled sequence to obtain a normalized sequence; According to the normalized sequence, dimension reduction is performed by using a t-distribution near embedding method to obtain dimension reduction data, and according to the dimension reduction data, clustering is performed by using density peak clustering to obtain a time series clustering feature sequence; The performance prediction result is obtained by calculating the time series clustering feature sequence through the LSTM model.

[0010] Specifically, the specific calculation steps of the diagnostic model are: According to the real-time fault-related parameter sequence, a preprocessing fault parameter sequence is obtained by calculating the preprocessing model; Performing forward scanning and reverse scanning through the BiGRU layer according to the preprocessed fault parameter sequence to obtain past context information and future context information, and performing modular addition operation on the past context information and the future context information to obtain a prediction feature vector; A state sequence is calculated through a Transformer layer according to the predicted feature vector, and a fault prediction result is calculated through a fully connected layer according to the state sequence.

[0011] Specifically, the specific calculation steps of the preprocessing model are: A normalized parameter sequence is obtained by normalizing the real-time fault-related parameter sequence, a time window is preset, the normalized parameter sequence is truncated according to the time window to obtain a shortened parameter sequence, and missing values ​​are filled by the mean method according to the shortened parameter sequence to obtain a preprocessed fault parameter sequence.

[0012] Specifically, the performing task scheduling according to the performance prediction result and the required performance indicator includes: Obtain the access time of the task to be run, preset a first weight value, multiply the first weight value and the performance prediction result to obtain a performance prediction weight value, multiply the first weight value and the required performance indicator to obtain a required performance weight value, calculate the absolute value of the difference based on the performance prediction weight value and the required performance weight value, preset a second weight value, calculate the sorting value based on the second weight value, the absolute value of the difference, and the access time of the task to be run, sort the tasks to be run according to the sorting value to obtain a task queue, and schedule tasks according to the task queue.

[0013] An automated operation and maintenance monitoring and management system, comprising: an analysis module, a collection module, a prediction module, and a scheduling module; The analysis module is used to obtain a system history configuration data set and a system history log data set, calculate a fault data set and a system performance index data set based on the system history log data set through a classification model, and calculate fault-related parameters based on the system history configuration data set, the fault data set, and the system performance index data set through a principal component analysis method; The collection module is used to collect real-time log data and real-time configuration parameters of the system through a log collection tool, calculate the real-time performance data and real-time fault data according to the real-time log data through the classification model, add timestamps according to the real-time performance data and the real-time configuration parameters to obtain time series performance data and time series configuration parameters, and store the time series performance data and the time series configuration parameters to obtain a performance data sequence and a configuration parameter sequence; The prediction module is used to obtain a performance prediction result by calculating the performance data sequence through a prediction model, to screen the performance data sequence and the configuration parameter sequence according to the fault-related parameters to obtain a real-time fault-related parameter sequence, and to obtain a fault prediction result by calculating the real-time fault-related parameter sequence through a diagnosis model; The scheduling module is used to obtain the required performance indicators of the tasks to be run, schedule the tasks according to the performance prediction results and the required performance indicators, and obtain the operation and maintenance personnel information. The intelligent terminal transmits the performance prediction results, the real-time fault data, and the fault prediction results to the operation and maintenance terminal according to the operation and maintenance personnel information.

[0014] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the automated operation and maintenance monitoring and management method as described above is implemented.

[0015] A storage medium containing computer executable instructions, which, when executed by a computer processor, are used to execute the automated operation and maintenance monitoring management method as described above.

[0016] The beneficial effects of the present invention are: (1) Improve the accuracy and timeliness of fault prediction: By obtaining the system's historical configuration data and log data sets, using the classification model to calculate the fault data set and system performance indicator data set, and combining the principal component analysis algorithm to determine the fault-related parameters, the system fault characteristics can be accurately extracted and identified. This not only improves the accuracy of fault prediction, but also provides strong data support for subsequent fault handling.

[0017] (2) Realize real-time performance monitoring and fault warning: By collecting real-time log data, constructing performance data sequences and real-time fault-related parameter sequences, and using prediction models and diagnostic models to calculate performance prediction results and fault prediction results. This real-time monitoring and warning mechanism can detect and warn potential risks in a timely manner before a fault occurs, saving valuable processing time for operation and maintenance personnel.

[0018] (3) Optimize task scheduling strategy: Task scheduling is performed based on the required performance indicators and performance prediction results of the tasks to be run, making task scheduling dynamic and intelligent. This scheduling strategy not only improves the efficiency and quality of task execution, but also ensures the rational use and allocation of system resources.

[0019] (4) Improve operation and maintenance efficiency and information processing capabilities: Performance prediction results, real-time fault data and fault prediction results are transmitted to the operation and maintenance terminal through the intelligent terminal, realizing the rapid transmission and sharing of operation and maintenance information. This not only improves the work efficiency of operation and maintenance personnel, but also reduces the difficulty and complexity of information processing, providing a strong guarantee for the smooth implementation of operation and maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0021] Figure 1 The present invention is a flowchart of an automated operation and maintenance monitoring management method. DETAILED DESCRIPTION

[0022] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0023] See also Figure 1 , an automated operation and maintenance monitoring management method, Obtain a system history configuration data set and a system history log data set, calculate a fault data set and a system performance index data set based on the system history log data set through a classification model, and calculate fault-related parameters based on the system history configuration data set, the fault data set, and the system performance index data set through a principal component analysis method; Collect the real-time log data and real-time configuration parameters of the system through the log collection tool, calculate the real-time performance data and real-time fault data through the classification model according to the real-time log data, add timestamps according to the real-time performance data and the real-time configuration parameters to obtain time series performance data and time series configuration parameters, and store the time series performance data and the time series configuration parameters to obtain a performance data sequence and a configuration parameter sequence; According to the performance data sequence, a performance prediction result is obtained by calculating the prediction model, according to the fault-related parameters, the performance data sequence and the configuration parameter sequence are screened to obtain a real-time fault-related parameter sequence, and according to the real-time fault-related parameter sequence, a fault prediction result is obtained by calculating the diagnosis model; Obtain the required performance indicators of the tasks to be run, schedule the tasks according to the performance prediction results and the required performance indicators, obtain the operation and maintenance personnel information, and the intelligent terminal transmits the performance prediction results, the real-time fault data, and the fault prediction results to the operation and maintenance terminal according to the operation and maintenance personnel information.

[0024] Specifically, the log collection tool includes intelligent agents or scripts deployed on each key node and key server; the system historical configuration data set includes historical network configuration, historical service settings, and historical user permissions; the system historical log data set includes but is not limited to the system performance indicator data set and the fault data set; the system performance indicator data set includes CPU usage, memory occupancy, disk space, network bandwidth, throughput, and response time; the fault data set includes but is not limited to data loss, system instability, and functional abnormality.

[0025] Specifically, the specific calculation steps of the classification model are: According to the system history log data set, word segmentation data is obtained by using Jieba word segmentation calculation, and word vectors are obtained by using Skip-gram model calculation according to the word segmentation data; According to the word vector, a first feature vector and a second feature vector are calculated by a feature extraction model, the first feature vector is subjected to dimensionality reduction processing to obtain a reduced-dimensionality feature vector, and the reduced-dimensionality feature vector and the second feature vector are fused by a concat function to obtain a fused feature vector; Refitting the fused feature vector through a fully connected layer to obtain fitting data, calculating the classification result through a softmax classifier based on the fitting data, and storing the system history log data set according to the classification combination to obtain the fault data set and the system performance indicator data set; The classification result calculation formula is: , in, P(y(i)=j|x(i);θ) is the classification result, i is the set of model parameters, j For categories, k To fit the data size, i、n is a variable parameter, y （i） is the classification label, x （i） To fit the data.

[0026] Specifically, the specific calculation steps of the feature extraction model include: Calculate a context feature vector according to the word vector through the LSTM model, concatenate the context feature vector and the word vector through the concat function to obtain a context concatenation feature vector, calculate a second context feature vector according to the context concatenation feature vector through the GRU, concatenate the context feature vector and the second context feature vector through the concat function to obtain a second concatenation vector, and calculate the first feature vector according to the second concatenation vector through the attention layer; A first local semantic feature vector and a second local semantic feature vector are calculated based on the word vector through two CNN models, a third concatenated vector is obtained by concatenating the first local semantic feature vector and the second local semantic feature vector through a concat function, and the second feature vector is calculated based on the third concatenated vector through an attention layer.

[0027] Specifically, the step of screening the performance data sequence and the configuration parameter sequence according to the fault-related parameters to obtain a real-time fault-related parameter sequence includes: The fault-related parameters include fault-related parameter name tags, and the performance data sequence and the configuration parameter sequence are screened according to the fault-related parameter name tags to obtain the real-time fault-related parameter sequence.

[0028] Specifically, the specific calculation steps of the prediction model are: Preset a time window, intercept the performance data sequence according to the time window to obtain a predicted performance sequence, fill missing values ​​according to the predicted performance sequence by using the mean method to obtain a filled sequence, and normalize the filled sequence to obtain a normalized sequence; According to the normalized sequence, dimension reduction is performed by using a t-distribution near embedding method to obtain dimension reduction data, and according to the dimension reduction data, clustering is performed by using density peak clustering to obtain a time series clustering feature sequence; The performance prediction result is obtained by calculating the time series clustering feature sequence through the LSTM model.

[0029] Specifically, the specific calculation steps of the diagnostic model are: According to the real-time fault-related parameter sequence, a preprocessing fault parameter sequence is obtained by calculating the preprocessing model; Performing forward scanning and reverse scanning through the BiGRU layer according to the preprocessed fault parameter sequence to obtain past context information and future context information, and performing modular addition operation on the past context information and the future context information to obtain a prediction feature vector; The prediction feature vector calculation expression is: , in, X is the predicted feature vector, H t1 is the past context information, H t2 for future contextual information; A state sequence is calculated through a Transformer layer according to the predicted feature vector, and a fault prediction result is calculated through a fully connected layer according to the state sequence; The fault prediction result calculation expression is: , in, Y is the fault prediction result, d(*) is the activation function of the fully connected layer, W f is the weight matrix, H is the state sequence, b f is the bias term.

[0030] Specifically, the specific calculation steps of the preprocessing model are: A normalized parameter sequence is obtained by normalizing the real-time fault-related parameter sequence, a time window is preset, the normalized parameter sequence is truncated according to the time window to obtain a shortened parameter sequence, and missing values ​​are filled by the mean method according to the shortened parameter sequence to obtain a preprocessed fault parameter sequence.

[0031] Specifically, the performing task scheduling according to the performance prediction result and the required performance indicator includes: Obtain the access time of the task to be run, preset a first weight value, multiply the first weight value and the performance prediction result to obtain a performance prediction weight value, multiply the first weight value and the required performance indicator to obtain a required performance weight value, calculate the absolute value of the difference based on the performance prediction weight value and the required performance weight value, preset a second weight value, calculate the sorting value based on the second weight value, the absolute value of the difference, and the access time of the task to be run, sort the tasks to be run according to the sorting value to obtain a task queue, and schedule tasks according to the task queue.

[0032] An automated operation and maintenance monitoring and management system, comprising: an analysis module, a collection module, a prediction module, and a scheduling module; The analysis module is used to obtain a system history configuration data set and a system history log data set, calculate a fault data set and a system performance index data set based on the system history log data set through a classification model, and calculate fault-related parameters based on the system history configuration data set, the fault data set, and the system performance index data set through a principal component analysis method; The collection module is used to collect real-time log data and real-time configuration parameters of the system through a log collection tool, calculate the real-time performance data and real-time fault data according to the real-time log data through the classification model, add timestamps according to the real-time performance data and the real-time configuration parameters to obtain time series performance data and time series configuration parameters, and store the time series performance data and the time series configuration parameters to obtain a performance data sequence and a configuration parameter sequence; The prediction module is used to obtain a performance prediction result by calculating the performance data sequence through a prediction model, to screen the performance data sequence and the configuration parameter sequence according to the fault-related parameters to obtain a real-time fault-related parameter sequence, and to obtain a fault prediction result by calculating the real-time fault-related parameter sequence through a diagnosis model; The scheduling module is used to obtain the required performance indicators of the tasks to be run, schedule the tasks according to the performance prediction results and the required performance indicators, and obtain the operation and maintenance personnel information. The intelligent terminal transmits the performance prediction results, the real-time fault data, and the fault prediction results to the operation and maintenance terminal according to the operation and maintenance personnel information.

[0033] An electronic device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the automated operation and maintenance monitoring and management method as described above is implemented.

[0034] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in combination with an instruction execution system, device or device.

[0035] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0036] The program code included on the computer readable medium can be transmitted with any appropriate medium, including but not limited to wireless, electric wire, optical cable, RF, etc., or any suitable combination of the above. The computer program code for performing the operation of the present invention can be written in one or more programming languages ​​or their combinations, and the programming language includes object-oriented programming languages-such as Java, Smalltalk, C++, and also includes conventional procedural programming languages-such as "C" language or similar programming languages. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0037] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. An automated operation and maintenance monitoring management method, characterized in that: include: Obtain a system history configuration data set and a system history log data set, calculate a fault data set and a system performance index data set based on the system history log data set through a classification model, and calculate fault-related parameters based on the system history configuration data set, the fault data set, and the system performance index data set through a principal component analysis method; Collect the real-time log data and real-time configuration parameters of the system through the log collection tool, calculate the real-time performance data and real-time fault data through the classification model according to the real-time log data, add timestamps according to the real-time performance data and the real-time configuration parameters to obtain time series performance data and time series configuration parameters, and store the time series performance data and the time series configuration parameters to obtain a performance data sequence and a configuration parameter sequence; According to the performance data sequence, a performance prediction result is obtained by calculating the prediction model, according to the fault-related parameters, the performance data sequence and the configuration parameter sequence are screened to obtain a real-time fault-related parameter sequence, and according to the real-time fault-related parameter sequence, a fault prediction result is obtained by calculating the diagnosis model; Obtain the required performance indicators of the tasks to be run, schedule the tasks according to the performance prediction results and the required performance indicators, obtain the operation and maintenance personnel information, and the intelligent terminal transmits the performance prediction results, the real-time fault data, and the fault prediction results to the post-operation and maintenance terminal according to the operation and maintenance personnel information.

2. The method according to claim 1, characterized in that The specific calculation steps of the classification model are: According to the system history log data set, word segmentation data is obtained by using Jieba word segmentation calculation, and word vectors are obtained by using Skip-gram model calculation according to the word segmentation data; According to the word vector, a first feature vector and a second feature vector are calculated by a feature extraction model, the first feature vector is subjected to dimensionality reduction processing to obtain a reduced-dimensionality feature vector, and the reduced-dimensionality feature vector and the second feature vector are fused by a concat function to obtain a fused feature vector; The fused feature vector is refitted through a fully connected layer to obtain fitting data, a classification result is calculated through a softmax classifier based on the fitting data, and the system historical log data set is stored based on the classification to obtain the fault data set and the system performance indicator data set.

3. The method according to claim 2, characterized in that The specific calculation steps of the feature extraction model include: Calculate a context feature vector according to the word vector through the LSTM model, concatenate the context feature vector and the word vector through the concat function to obtain a context concatenation feature vector, calculate a second context feature vector according to the context concatenation feature vector through the GRU, concatenate the context feature vector and the second context feature vector through the concat function to obtain a second concatenation vector, and calculate the first feature vector according to the second concatenation vector through the attention layer; A first local semantic feature vector and a second local semantic feature vector are calculated based on the word vector through two CNN models, a third concatenated vector is obtained by concatenating the first local semantic feature vector and the second local semantic feature vector through a concat function, and the second feature vector is calculated based on the third concatenated vector through an attention layer.

4. The method according to claim 1, characterized in that: The step of screening the performance data sequence and the configuration parameter sequence according to the fault-related parameters to obtain a real-time fault-related parameter sequence comprises: The fault-related parameters include fault-related parameter name tags, and the performance data sequence and the configuration parameter sequence are screened according to the fault-related parameter name tags to obtain the real-time fault-related parameter sequence.

5. The method according to claim 1, characterized in that The specific calculation steps of the prediction model are: Preset a time window, intercept the performance data sequence according to the time window to obtain a predicted performance sequence, fill missing values ​​according to the predicted performance sequence by using the mean method to obtain a filled sequence, and normalize the filled sequence to obtain a normalized sequence; According to the normalized sequence, dimension reduction is performed by using a t-distribution near embedding method to obtain dimension reduction data, and according to the dimension reduction data, clustering is performed by using density peak clustering to obtain a time series clustering feature sequence; The performance prediction result is obtained by calculating the time series clustering feature sequence through the LSTM model.

6. The method according to claim 1, characterized in that The specific calculation steps of the diagnostic model are: According to the real-time fault-related parameter sequence, a preprocessing fault parameter sequence is obtained by calculating the preprocessing model; Performing forward scanning and reverse scanning through the BiGRU layer according to the preprocessed fault parameter sequence to obtain past context information and future context information, and performing modular addition operation on the past context information and the future context information to obtain a prediction feature vector; A state sequence is calculated through a Transformer layer according to the predicted feature vector, and a fault prediction result is calculated through a fully connected layer according to the state sequence.

7. The method according to claim 6, characterized in that The specific calculation steps of the preprocessing model are: A normalized parameter sequence is obtained by normalizing the real-time fault-related parameter sequence, a time window is preset, the normalized parameter sequence is truncated according to the time window to obtain a shortened parameter sequence, and missing values ​​are filled by the mean method according to the shortened parameter sequence to obtain a preprocessed fault parameter sequence.

8. The method according to claim 1, characterized in that The performing of task scheduling according to the performance prediction result and the required performance index comprises: Obtain the access time of the task to be run, preset a first weight value, multiply the first weight value and the performance prediction result to obtain a performance prediction weight value, multiply the first weight value and the required performance indicator to obtain a required performance weight value, calculate the absolute value of the difference based on the performance prediction weight value and the required performance weight value, preset a second weight value, calculate the sorting value based on the second weight value, the absolute value of the difference, and the access time of the task to be run, sort the tasks to be run according to the sorting value to obtain a task queue, and schedule tasks according to the task queue.

9. An automated operation and maintenance monitoring and management system, characterized in that: include: Analysis module, collection module, prediction module, and scheduling module; The analysis module is used to obtain a system history configuration data set and a system history log data set, calculate a fault data set and a system performance index data set based on the system history log data set through a classification model, and calculate fault-related parameters based on the system history configuration data set, the fault data set, and the system performance index data set through a principal component analysis method; The collection module is used to collect real-time log data and real-time configuration parameters of the system through a log collection tool, calculate the real-time performance data and real-time fault data according to the real-time log data through the classification model, add timestamps according to the real-time performance data and the real-time configuration parameters to obtain time series performance data and time series configuration parameters, and store the time series performance data and the time series configuration parameters to obtain a performance data sequence and a configuration parameter sequence; The prediction module is used to obtain a performance prediction result by calculating the performance data sequence through a prediction model, to screen the performance data sequence and the configuration parameter sequence according to the fault-related parameters to obtain a real-time fault-related parameter sequence, and to obtain a fault prediction result by calculating the real-time fault-related parameter sequence through a diagnosis model; The scheduling module is used to obtain the required performance indicators of the tasks to be run, schedule the tasks according to the performance prediction results and the required performance indicators, and obtain the operation and maintenance personnel information. The intelligent terminal transmits the performance prediction results, the real-time fault data, and the fault prediction results to the operation and maintenance terminal according to the operation and maintenance personnel information.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the automated operation and maintenance monitoring management method as described in any one of claims 1 to 8 is implemented; A storage medium containing computer executable instructions, characterized in that: The computer executable instructions, when executed by a computer processor, are used to execute the automated operation and maintenance monitoring management method as described in any one of claims 1-8.

Images