CN118368652A - Method and device for generating network element operation scheme, electronic equipment and storage medium - Google Patents

Abstract

The invention provides a method and a device for generating a network element operation scheme, electronic equipment and a computer readable storage medium, and relates to the technical field of communication. The generation method comprises the following steps: acquiring network element full data and extracting network element capacity data from the network element full data, wherein the network element capacity data comprises: license capacity utilization, current user number, and split threshold; performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model to obtain capacity prediction data, wherein the capacity prediction data comprises: the maximum license capacity utilization rate and the maximum number of users in a preset time period of a target network element; and generating an optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model. The problems that the network element capacity maintenance in the related technology depends on manual work, errors and risks exist, and capacity alarming is delayed are at least solved. The method is suitable for network element capacity prediction and management scenes.

Description

Method and device for generating network element operation scheme, electronic equipment and storage medium

Technical Field

The present invention relates to the field of communications technologies, and in particular, to a method and apparatus for generating a network element operation scheme, an electronic device, and a computer readable storage medium.

Background

Currently, a 5G (the 5th Generation Mobile Communication Technology, fifth generation mobile communication technology) core network element is gradually clouded, the 5G network element adopts a service architecture to split the network element function into network services with fine granularity, and the equipment structure function is gradually complex due to the butt-jointed clouding NFV (Network Functions Virtualization, network function virtualization) platform lightweight deployment unit. The network is required to stably run, not only depends on the performance of the equipment, but also needs to find hidden trouble of equipment running through daily regular inspection, and timely performs relevant upgrading operation on network elements.

The operation mode and the corresponding problems of the network element of the existing 5G core network are as follows:

1. Relying on manual work: the core network maintainer logs in the equipment to acquire an execution result through manual execution instructions or scripts, manually checks and analyzes whether the equipment is abnormal, and then manually executes related operations of capacity expansion, cutting and upgrading. There are the following problems: the operation efficiency is low, a manual operation error exists, and if the operation error occurs, the network paralysis of the core network is possibly caused;

2. Relying on manufacturer network management: the network manager of some manufacturers has network element checking capability, and can periodically read the relevant indexes of the network element and manually analyze the indexes, so that various cutting and upgrading operations required by the network element are found. There are the following problems: because the measurement standard is formulated by a manufacturer, the real state of the equipment cannot be faithfully reflected, meanwhile, the operation scheme is manually written by maintenance personnel, manually executed and also has errors and risks;

3. In the conventional maintenance mode, when the capacity of the network element is exhausted, the network element is known only by receiving an alarm, and when the system really fails, a certain delay exists in the alarm.

Disclosure of Invention

The invention aims to solve the technical problems of the prior art and provides a method, a device, electronic equipment and a computer readable storage medium for generating a network element operation scheme, wherein the method can realize real-time monitoring and prediction of network element capacity, intelligent judgment of network element capacity prediction data and automatic generation of the network element operation scheme, realize timely pushing of capacity early warning information and the optimal operation scheme, ensure the stability of a 5G core network and reduce labor cost while improving quality and efficiency.

In a first aspect, the present invention provides a method for generating a network element operation scheme, including: acquiring network element full data and extracting network element capacity data from the network element full data, wherein the network element capacity data comprises: license capacity utilization, current user number, and split threshold; performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model to obtain capacity prediction data, wherein the capacity prediction data comprises: the maximum license capacity utilization rate and the maximum number of users in a preset time period of a target network element; and generating an optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model.

Optionally, after the extracting the network element capacity data from the network element full data and before the performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model, the generating method of the network element operation scheme further includes: performing data verification on network element capacity data, wherein the data verification comprises one or more of the following combinations: the method for generating the network element operation scheme further comprises the following steps of after the network element full data are acquired and before the capacity prediction of the target network element is carried out according to the network element capacity data and the capacity prediction model: and constructing a capacity prediction model and an optimal operation scheme recommendation model, wherein the capacity prediction model comprises an autoregressive AR model, a moving average MA model and an autoregressive moving average ARMA model.

Preferably, the performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model to obtain capacity prediction data specifically includes: determining a capacity prediction model based on the capacity data of the target network element, and estimating capacity prediction model parameters; performing capacity prediction on the target network element according to the capacity prediction model and the capacity data of the target network element to obtain capacity prediction data; calculating an error of the capacity prediction data based on the error indicator and the target network element capacity data actual value, wherein the error indicator comprises one or a combination of more of the following: average absolute error, root mean square error, average absolute percentage error; judging the magnitude of the error and a preset threshold value; outputting capacity prediction data in response to the error being less than a preset threshold; and re-estimating the capacity prediction model parameters and performing capacity prediction until the error is smaller than a preset threshold value in response to the error being greater than or equal to the preset threshold value.

Preferably, the generating the optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model specifically includes: judging whether the capacity of the target network element is too high according to the maximum license capacity utilization rate in the preset time period of the target network element; and responding to the excessive capacity of the target network element, and generating an optimal operation scheme according to the optimal operation scheme recommendation model and the maximum number of users in a preset time period of the target network element.

Preferably, the determining whether the capacity of the target network element is too high according to the maximum license capacity utilization rate within the preset time period of the target network element specifically includes: judging whether the maximum license capacity utilization rate in a preset time period of a target network element is larger than a first preset value; and determining that the capacity of the target network element is too high in response to the maximum license capacity utilization being greater than a first preset value.

Preferably, the generating the optimal operation scheme of the target network element according to the optimal operation scheme recommendation model and the maximum user number in the preset time period of the target network element in response to the target network element capacity being too high specifically includes: judging whether the maximum number of users in a preset time period of a target network element is smaller than a splitting threshold of the target network element or not; generating a target network topological graph, a capacity expansion threshold and a capacity expansion scheme based on an optimal operation scheme recommendation model to serve as an optimal operation scheme of the target network element in response to the fact that the maximum number of users in a preset time period of the target network element is smaller than a splitting threshold of the target network element; and generating a target network topological graph and a newly-built network element scheme as optimal operation schemes of the target network element based on the optimal operation scheme recommendation model in response to the maximum number of users in a preset time period of the target network element being greater than or equal to a splitting threshold of the target network element.

Optionally, after the generating the optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model, the generating method of the network element operation scheme further includes: generating an optimal operation scheme report based on an optimal operation scheme and a preset report structure frame; performing page presentation on network element capacity data, capacity prediction data and an optimal operation scheme report of a target network element, and storing the page presentation in a database; and responding to the excessive capacity of the target network element, and carrying out push early warning on the capacity prediction data and the optimal operation scheme.

In a second aspect, the present invention further provides a device for generating a network element operation scheme, which is characterized by including: the system comprises an acquisition module, a prediction module and a generation module, wherein the acquisition module is used for acquiring network element full data and extracting network element capacity data from the network element full data, and the network element capacity data comprises: the system comprises a license capacity utilization rate, a current user number and a splitting threshold, a prediction module, an acquisition module and a target network element, wherein the prediction module is connected with the acquisition module and is used for carrying out capacity prediction on the target network element according to network element capacity data and a capacity prediction model to obtain capacity prediction data, and the capacity prediction data comprises: the generation module is connected with the prediction module and is used for generating an optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model.

In a third aspect, the present invention also provides an electronic device comprising a memory and a processor, the memory storing a computer program, the processor being arranged to run the computer program to implement the method of generating a network element operating scheme provided in the first aspect above.

In a fourth aspect, the present invention also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method for generating a network element operation scheme provided in the first aspect.

The invention provides a method, a device, electronic equipment and a computer readable storage medium for generating a network element operation scheme, which are used for carrying out capacity prediction on a network element based on collected network element capacity data and a capacity prediction model, carrying out classification judgment on the capacity prediction data, and identifying a capacity abnormal result to output an optimal operation scheme of the network element based on the capacity prediction data and an optimal operation scheme recommendation model. Therefore, the invention can realize the real-time monitoring and prediction of the network element capacity, the intelligent judgment of the network element capacity prediction data and the automatic generation of the network element operation scheme, realize the timely pushing of the capacity early warning information and the optimal operation scheme, ensure the stability of the 5G core network and reduce the labor cost while improving the quality and enhancing the efficiency.

Drawings

Fig. 1 is a flowchart of a method for generating a network element operation scheme according to embodiment 1 of the present invention;

fig. 2 is a flowchart of a capacity prediction method based on network element capacity data and a capacity prediction model according to embodiment 1 of the present invention;

Fig. 3 is a flowchart of a network element operation scheme generation method based on a capacity prediction model and an optimal operation scheme recommendation model according to embodiment 1 of the present invention;

FIG. 4 is a flow chart of a method for generating a report of an optimal operation scheme according to embodiment 1 of the present invention;

fig. 5 is a flowchart of a method for generating a network element operation scheme according to embodiment 2 of the present invention;

fig. 6 is a schematic structural diagram of a generating device of a network element operation scheme in embodiment 3 of the present invention.

Detailed Description

In order to make the technical scheme of the present invention better understood by those skilled in the art, the following detailed description of the embodiments of the present invention will be given with reference to the accompanying drawings.

It is to be understood that the specific embodiments and figures described herein are merely illustrative of the invention, and are not limiting of the invention.

It is to be understood that the various embodiments of the invention and the features of the embodiments may be combined with each other without conflict.

It is to be understood that only the portions relevant to the present invention are shown in the drawings for convenience of description, and the portions irrelevant to the present invention are not shown in the drawings.

It should be understood that each unit and module in the embodiments of the present invention may correspond to only one physical structure, may be formed by a plurality of physical structures, or may be integrated into one physical structure.

It will be appreciated that, without conflict, the functions and steps noted in the flowcharts and block diagrams of the present invention may occur out of the order noted in the figures.

It is to be understood that the flowcharts and block diagrams of the present invention illustrate the architecture, functionality, and operation of possible implementations of systems, apparatuses, devices, methods according to various embodiments of the present invention. Where each block in the flowchart or block diagrams may represent a unit, module, segment, code, or the like, which comprises executable instructions for implementing the specified functions. Moreover, each block or combination of blocks in the block diagrams and flowchart illustrations can be implemented by hardware-based systems that perform the specified functions, or by combinations of hardware and computer instructions.

It should be understood that the units and modules related in the embodiments of the present invention may be implemented by software, or may be implemented by hardware, for example, the units and modules may be located in a processor.

Example 1:

as shown in fig. 1, this embodiment provides a method for generating a network element operation scheme. The method for generating the network element operation scheme comprises the following steps:

Step S101, acquiring network element full data and extracting network element capacity data from the network element full data, wherein the network element capacity data comprises: license capacity utilization, current number of users, and split threshold.

In this embodiment, a network element capacity monitoring platform is built to perform unified management on network element capacity, the network element capacity monitoring platform adopts an SSL (Secure Socket Layer ) manner, opens a north command interface of a network element network manager, simulates login of the network element network manager and the network element, and realizes full acquisition of network element data by using various manners such as socket protocol, API (Application Programming Interface ), telnet, SSH (Secure Shell protocol), etc., and sets a monitoring period to be acquired once per hour (which can be adjusted as required). The network element capacity monitoring platform comprises a real-time monitoring module and a capacity extraction module, wherein the real-time monitoring module is used for periodically acquiring network element total data, the network element total data comprises various indexes, and the capacity data is only a part of the indexes; the capacity extraction module is used for extracting network element capacity data, namely license capacity utilization rate X, current user number Y and splitting threshold Z (namely maximum loadable user number) index data from the network element total data monitored by the real-time monitoring module, wherein X=Y/Z.

Step S102, carrying out capacity prediction on a target network element according to the network element capacity data and a capacity prediction model to obtain capacity prediction data, wherein the capacity prediction data comprises: the target network element presets the maximum license capacity utilization rate and the maximum number of users in a time period.

Optionally, after extracting the network element capacity data from the network element full data, and in step S102: before the capacity prediction of the target network element is performed according to the network element capacity data and the capacity prediction model, the method further comprises the following steps: performing data verification on network element capacity data, wherein the data verification comprises one or more of the following combinations: and (5) checking stability and white noise.

In this embodiment, after obtaining the network element capacity data, performing stability test to determine whether the network element capacity data has a random trend, if the network element capacity data has a random trend, a phenomenon of "pseudo regression" will be generated, and then the random trend and seasonal change in the network element capacity data need to be eliminated through differential operation, so that the network element capacity data is more stable, and in this embodiment, the stability test can be performed by observing a timing diagram. After the stability test is performed, whether the time sequence of the network element capacity data is a white noise sequence is tested, namely, the white noise test is performed on the network element capacity data, if the time sequence is the white noise sequence, the useful information in the time sequence is completely extracted, and the rest is random interference and cannot be predicted and used. White noise testing generally employs an LB statistic (Ljung-Box statistical, a statistic used to test whether or not there is auto-correlation in time series data) test method. According to the method, the characteristics and the structure of the network element capacity data are evaluated through the stability test and the white noise test, guidance is provided for subsequent modeling and prediction, the property of the data can be better understood, a proper modeling method is selected, and the accuracy and the reliability of prediction are improved.

Optionally, after acquiring the network element full data, and in step S102: before the capacity prediction of the target network element is performed according to the network element capacity data and the capacity prediction model, the method further comprises the following steps: and constructing a capacity prediction model and an optimal operation scheme recommendation model, wherein the capacity prediction model comprises an autoregressive AR model, a moving average MA model and an autoregressive moving average ARMA model.

In this embodiment, the network element capacity data may be classified into historical network element capacity data (including sample data and test data) for constructing a training capacity prediction model and target network element capacity data for performing prediction. The capacity prediction model is built based on historical network element capacity data in two ways, wherein the first way is to use a machine learning algorithm (such as a decision tree, a support vector machine and a neural network) to perform model training and evaluation on the historical network element capacity data so as to build the capacity prediction model.

The second mode is as follows: AR (Autoregressive ), MA (Moving Average) and ARMA (Autoregressive Moving Average ) models are constructed as capacity prediction models based on historical network element capacity data. The process of building AR, MA and ARMA models typically involves the steps of: 1) Identifying characteristics of the time series: firstly, the time series data is required to be visualized and analyzed to know the characteristics of trend, seasonality, periodicity and the like; 2) Determining the order: determining the order of the AR and MA models, i.e., the P and Q values in AR (P) and MA (Q), can be done by observing a plot of the autocorrelation function (ACF, autocorrelation Function) and the partial autocorrelation function (PACF, partial autocorrelation function); 3) Fitting a model: fitting an AR, MA or ARMA model to the time series data according to the determined order; 4) Model diagnosis: diagnosing the fitted model, and checking whether the residual sequence has autocorrelation or other features violating model assumptions; 5) Model selection: the optimal AR, MA or ARMA model is selected based on the diagnostic results, typically using information criteria such as AIC (Akaike information criterion, red pool information amount criteria) or BIC (Bayesian Information Criterion, bayesian information criteria) for model selection, resulting in a capacity prediction model, including the optimal AR, MA or ARMA model.

The network element total data comprises configuration files and performance files of the network element besides the network element capacity data. The configuration files and performance files of the network element typically include configuration information and performance indicators for the network element itself, as well as message passing and link state information with other network elements. And pushing the configuration file and the performance file of the network element to the backup server by opening the service interface of the network element network management. And constructing and training an optimal operation scheme recommendation model based on configuration files and performance files of the network elements in the network element total data, wherein the optimal operation scheme model comprises network topology self-learning and scheme output. Assume that there is a message and link relationship between the following network elements: a link exists between A and B, A sends a message to B, B and C exist links, B sends a message to C, C and D exist links, and C sends a message to D; according to the above information, a Python program can be written to automatically parse to obtain a network topology map, for example:

Running the above code will result in the following network topology:

A->{'B'}

B->{'C'}

C->{'D'}

D->set()。

specifically, step S102: performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model to obtain capacity prediction data, including steps S1021-S1026:

S1021, determining a capacity prediction model based on the capacity data of the target network element, and estimating parameters of the capacity prediction model.

And S1022, carrying out capacity prediction on the target network element according to the capacity prediction model and the capacity data of the target network element to obtain capacity prediction data.

S1023, calculating the error of the capacity prediction data based on the error index and the actual value of the capacity data of the target network element, wherein the error index comprises one or more of the following combinations: mean absolute error, root mean square error, mean absolute percentage error.

S1024, judging the magnitude of the error and the preset threshold.

And S1025, outputting capacity prediction data in response to the error being smaller than a preset threshold.

And S1026, re-estimating the capacity prediction model parameters and performing capacity prediction until the error is smaller than a preset threshold value in response to the error being larger than or equal to the preset threshold value.

In this embodiment, as shown in fig. 2, the capacity data of the target network element is the observation value sequence in fig. 2, the actual value of the capacity data of the target network element is the actual value in fig. 2, the capacity prediction data is the prediction result in fig. 2, and the capacity prediction is performed based on the capacity data of the network element and the capacity prediction model, which specifically includes: ① And after acquiring the capacity data of the target network element, carrying out stability test, and if the stability test of the capacity data of the target network element is not passed, enabling the capacity data of the target network element to be more stable through differential operation. ② And after the stability test is carried out, carrying out white noise test on the capacity data of the target network element. ③ By observing the autocorrelation diagrams and the partial autocorrelation diagrams of the target network element capacity data to determine P, Q parameters, it is identified which of AR, MA and ARMA the capacity prediction model used to predict the target network element capacity data belongs to. ④ Other parameters of the capacity prediction model are estimated, and the embodiment can adopt maximum likelihood estimation and a conditional least square method to estimate the model parameters. ⑤ And carrying out capacity prediction on the capacity data of the target network element based on the capacity prediction model to obtain capacity prediction data. ⑥ Three statistic indexes for measuring the prediction precision of the model are adopted: and carrying out error analysis on the capacity prediction data and the actual value of the capacity data of the target network element by using the average absolute error, the root mean square error and the average absolute percentage error, and reflecting the prediction precision of the capacity prediction model from different sides. ⑦ And if the error is smaller than the preset threshold, outputting capacity prediction data, and if the error is larger than or equal to the preset threshold, re-estimating the capacity prediction model parameters and performing capacity prediction until the error is smaller than the preset threshold.

And step S103, generating an optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model.

Specifically, step S103: generating an optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model, wherein the optimal operation scheme comprises steps S1031-S1032:

Step S1031, judging whether the capacity of the target network element is too high according to the maximum license capacity utilization rate in the preset time period of the target network element.

Specifically, step S1031: judging whether the capacity of the target network element is too high according to the maximum license capacity utilization rate in the preset time period of the target network element, comprising: judging whether the maximum license capacity utilization rate in a preset time period of a target network element is larger than a first preset value; and determining that the capacity of the target network element is too high in response to the maximum license capacity utilization being greater than a first preset value.

Step S1032, generating an optimal operation scheme according to the optimal operation scheme recommendation model and the maximum user number in the preset time period of the target network element in response to the target network element capacity being too high.

Specifically, step S1032: responding to the overhigh capacity of the target network element, generating an optimal operation scheme according to an optimal operation scheme recommendation model and the maximum number of users in a preset time period of the target network element, wherein the method comprises the following steps: judging whether the maximum number of users in a preset time period of a target network element is smaller than a splitting threshold of the target network element or not; generating a target network topological graph, a capacity expansion threshold and a capacity expansion scheme based on an optimal operation scheme recommendation model to serve as an optimal operation scheme of the target network element in response to the fact that the maximum number of users in a preset time period of the target network element is smaller than a splitting threshold of the target network element; and generating a target network topological graph and a newly-built network element scheme as optimal operation schemes of the target network element based on the optimal operation scheme recommendation model in response to the maximum number of users in a preset time period of the target network element being greater than or equal to a splitting threshold of the target network element.

In this embodiment, as shown in fig. 3, the preset time period is exemplified by 24 hours in the future, the first preset value is exemplified by 80%, and in the initial stage of planning of network element establishment, the splitting threshold, that is, the maximum bearing capacity of the artificially given network element, exceeds the maximum bearing capacity for splitting, and the network element operation scheme generation based on the capacity prediction model and the optimal operation scheme recommendation model specifically includes: 1) And (3) inputting the capacity prediction data output by the capacity prediction model in the step (S102) into an optimal operation scheme recommendation model for analysis so as to judge whether the maximum license capacity utilization rate in the preset time period of the target network element is larger than a first preset value and judge whether the maximum user number in the preset time period of the target network element is smaller than the splitting threshold of the target network element. 2) If the maximum license capacity utilization rate in the target network element preset time period is smaller than or equal to a first preset value, the capacity is normal; if the maximum license capacity utilization rate in the preset time period of the target network element is larger than the first preset value and the maximum user number in the preset time period of the target network element is smaller than the splitting threshold of the target network element, judging that: the capacity is too high, the number of users does not reach the splitting threshold, capacity expansion is needed, the capacity expansion threshold and the capacity expansion scheme are output, and the capacity expansion scheme comprises operation steps, operation instructions and operation influences; if the maximum license capacity utilization rate in the preset time period of the target network element is larger than the first preset value and the maximum user number in the preset time period of the target network element is larger than the splitting threshold of the target network element, judging that: the capacity is too high, the number of users reaches a splitting threshold, a new network element is required to be built for load balancing, a new network element scheme is output, and the new network element scheme comprises operation steps, operation instructions and operation influences. 3) The optimal operation scheme model analyzes big data aiming at configuration files and performance files of network elements, analyzes layer-by-layer nesting rules among the network elements, namely message and link states among the network elements, and performs network topology self-learning according to a Python program and the configuration files of the network elements and the message and link among the network elements in the performance files to obtain a current network topology graph. 4) According to the output scheme (i.e. the capacity expansion threshold and the capacity expansion scheme, or the newly-built network element scheme), the state information of the network element to be adjusted in the current network topology is modified to obtain the target network topology, for example, in the capacity expansion scheme, the current network element has only one A, and then becomes A1 and A2 in the target network topology after splitting.

It should be noted that, in the following step, according to the user requirement, when the optimal operation scheme recommendation model is built, the judgment condition and the output scheme of the optimal operation scheme recommendation model can be freely added.

In the present embodiment, in step S1032: in response to the target network element capacity being too high, generating an optimal operation scheme according to the optimal operation scheme recommendation model and the maximum number of users in a preset time period of the target network element, and then further comprising:

Step S1033, generating an optimal operation scheme report based on the optimal operation scheme and a preset report structure framework.

In this embodiment, as shown in fig. 4, generating the report of the optimal operation scheme specifically includes: 1) And (3) formulating a report structure framework, and generating a preset report structure framework containing report fixed content and report forms, so as to leave blank content. 2) And outputting the optimal operation scheme by the optimal operation scheme recommendation model to carry out program analysis, and backfilling blank contents, wherein the blank contents comprise operation steps, operation instructions, operation influences and network topology changes. 3) An optimal operating scheme report is generated. The report of the optimal operation scheme is generated based on the optimal operation scheme and the preset report structure framework, so that the quality and effect of the report can be improved, a user is helped to better understand and accept the optimal operation scheme, decision making and execution are supported, and the sustainable development and success of organization are promoted.

Step S1034, the network element capacity data, the capacity prediction data and the optimal operation scheme report of the target network element are subjected to page presentation and stored in a database.

And step S1035, in response to the excessively high capacity of the target network element, pushing and early warning is carried out on the capacity prediction data and the optimal operation scheme.

In this embodiment, the network element capacity data, the capacity prediction data and the optimal operation scheme of the target network element are synchronized to the front end of the page in real time, wherein the optimal operation scheme is presented in the form of an optimal operation scheme report. And pushing the capacity prediction data and the optimal operation scheme to equipment maintenance personnel through short messages or nails when the capacity of the target network element is too high, so that the equipment maintenance personnel can remind in real time and correct in time. According to the method and the device, the network element capacity data, the capacity prediction data and the optimal operation scheme report of the target network element are presented on the page, so that a user can know the change and trend of the data in time, the data is easier to understand and analyze, and the user experience is improved. The data is stored in the database in a lasting way through the network element capacity data, the capacity prediction data and the optimal operation scheme report of the target network element, so that the data is ensured to be stored for a long time and available at any time. By pushing and early warning the capacity prediction data and the optimal operation scheme when the capacity of the target network element is too high, a manager can know the data condition in time and take action in time, and decision making and problem solving are supported.

According to the network element operation scheme generation method, capacity prediction is carried out on the network elements based on the collected network element capacity data and the capacity prediction model, classification judgment is carried out on the capacity prediction data based on the optimal operation scheme recommendation model, the capacity abnormal result is identified to output the optimal operation scheme of the network elements, real-time monitoring and prediction of the network element capacity, intelligent judgment of the network element capacity prediction data and automatic generation of the network element operation scheme are achieved, timely pushing of capacity early warning information and the optimal operation scheme is achieved, quality improvement and efficiency improvement are achieved, meanwhile stability of the 5G core network is guaranteed, and labor cost is reduced. In addition, the characteristics and the structure of the network element capacity data are evaluated through the stability test and the white noise test, guidance is provided for subsequent modeling and prediction, the property of the data can be better understood, a proper modeling method is selected, and the accuracy and the reliability of prediction are improved. Generating the report of the optimal operation scheme based on the optimal operation scheme and the preset report structure framework can improve the quality and effect of the report, help users to better understand and accept the optimal operation scheme, support decision making and execution, and promote the sustainable development and success of organizations. By page presentation of network element capacity data, capacity prediction data and an optimal operation scheme report of the target network element, a user can know data change and trend in time, the data is easier to understand and analyze, and user experience is improved. The data is stored in the database in a lasting way through the network element capacity data, the capacity prediction data and the optimal operation scheme report of the target network element, so that the data is ensured to be stored for a long time and available at any time. By pushing and early warning the capacity prediction data and the optimal operation scheme when the capacity of the target network element is too high, a manager can know the data condition in time and take action in time, and decision making and problem solving are supported.

Example 2:

As shown in fig. 5, this embodiment provides a method for generating a network element operation scheme. The method for generating the network element operation scheme comprises the following steps:

step S201, a network element capacity monitoring platform is built to monitor the network element full data in real time and extract the network element capacity data.

In this embodiment, the network element capacity data is extracted, i.e. the capacity extraction in fig. 5. The network element capacity monitoring platform is built to uniformly manage the network element capacity, the network element capacity monitoring platform adopts an SSL (Secure Socket Layer ) mode, a north command interface of the network element network manager is opened, the network element management and the network element are simulated to log in, the total acquisition of network element data is realized by utilizing various modes such as socket protocol, API (Application Programming Interface ), telnet, SSH (Secure Shell), and the like, and the monitoring period is set to be acquired once per hour (can be adjusted according to the requirement). The network element capacity monitoring platform comprises a real-time monitoring module and a capacity extraction module, wherein the real-time monitoring module is used for periodically acquiring network element total data, the network element total data comprises various indexes, and the capacity data is only a part of the indexes; the capacity extraction module is used for extracting network element capacity data, namely license capacity utilization rate X, current user number Y and splitting threshold Z (namely maximum loadable user number) index data from the network element total data monitored by the real-time monitoring module, wherein X=Y/Z.

Step S202, building a database platform.

Step S203, an evaluation model is established, and the network element capacity data is evaluated based on the evaluation model to obtain capacity prediction data and an optimal operation scheme report, wherein the evaluation model comprises a capacity prediction model and an optimal operation scheme recommendation model.

In this embodiment, the capacity prediction model is the prediction model in fig. 5, and the optimal operation scheme recommendation model is the optimal decision recommendation model in fig. 5. And after acquiring the capacity data of the target network element, carrying out stability test, and if the stability test of the capacity data of the target network element is not passed, enabling the capacity data of the target network element to be more stable through differential operation. ② And after the stability test is carried out, carrying out white noise test on the capacity data of the target network element. ③ By observing the autocorrelation diagrams and the partial autocorrelation diagrams of the target network element capacity data to determine P, Q parameters, it is identified which of AR, MA and ARMA the capacity prediction model used to predict the target network element capacity data belongs to. ④ Other parameters of the capacity prediction model are estimated, and the embodiment can adopt maximum likelihood estimation and a conditional least square method to estimate the model parameters. ⑤ And carrying out capacity prediction on the capacity data of the target network element based on the capacity prediction model to obtain capacity prediction data. ⑥ Three statistic indexes for measuring the prediction precision of the model are adopted: and carrying out error analysis on the capacity prediction data and the actual value of the capacity data of the target network element by using the average absolute error, the root mean square error and the average absolute percentage error, and reflecting the prediction precision of the capacity prediction model from different sides. ⑦ And if the error is smaller than the preset threshold, outputting capacity prediction data, and if the error is larger than or equal to the preset threshold, re-estimating the capacity prediction model parameters and performing capacity prediction until the error is smaller than the preset threshold. According to the method, the characteristics and the structure of the network element capacity data are evaluated through the stability test and the white noise test, guidance is provided for subsequent modeling and prediction, the property of the data can be better understood, a proper modeling method is selected, and the accuracy and the reliability of prediction are improved.

And inputting the capacity prediction data output by the capacity prediction model into an optimal operation scheme recommendation model for analysis so as to judge whether the maximum license capacity utilization rate in a preset time period of the target network element is larger than a first preset value and judge whether the maximum user number in the preset time period of the target network element is smaller than the splitting threshold of the target network element. If the maximum license capacity utilization rate in the target network element preset time period is smaller than or equal to a first preset value, the capacity is normal; if the maximum license capacity utilization rate in the preset time period of the target network element is larger than the first preset value and the maximum user number in the preset time period of the target network element is smaller than the splitting threshold of the target network element, judging that: the capacity is too high, the number of users does not reach the splitting threshold, capacity expansion is needed, the capacity expansion threshold and the capacity expansion scheme are output, and the capacity expansion scheme comprises operation steps, operation instructions and operation influences; if the maximum license capacity utilization rate in the preset time period of the target network element is larger than the first preset value and the maximum user number in the preset time period of the target network element is larger than the splitting threshold of the target network element, judging that: the capacity is too high, the number of users reaches a splitting threshold, a new network element is required to be built for load balancing, a new network element scheme is output, and the new network element scheme comprises operation steps, operation instructions and operation influences. The optimal operation scheme model analyzes big data aiming at configuration files and performance files of network elements, analyzes layer-by-layer nesting rules among the network elements, namely message and link states among the network elements, and performs network topology self-learning according to a Python program and the configuration files of the network elements and the message and link among the network elements in the performance files to obtain a current network topology graph. According to the output scheme (i.e. the capacity expansion threshold and the capacity expansion scheme, or the newly-built network element scheme), the state information of the network element to be adjusted in the current network topology is modified to obtain the target network topology, for example, in the capacity expansion scheme, the current network element has only one A, and then becomes A1 and A2 in the target network topology after splitting.

And (3) formulating a report structure framework, and generating a preset report structure framework containing report fixed content and report forms, so as to leave blank content. And outputting the optimal operation scheme by the optimal operation scheme recommendation model to carry out program analysis, and backfilling blank contents, wherein the blank contents comprise operation steps, operation instructions, operation influences and network topology changes. An optimal operating scheme report is generated. The report of the optimal operation scheme is generated based on the optimal operation scheme and the preset report structure framework, so that the quality and effect of the report can be improved, a user is helped to better understand and accept the optimal operation scheme, decision making and execution are supported, and the sustainable development and success of organization are promoted.

And step S204, carrying out interface presentation and early warning pushing on the network element capacity data, the capacity prediction data and the optimal operation scheme report.

In this embodiment, the network element capacity data, the capacity prediction data and the optimal operation scheme of the target network element are synchronized to the front end of the page in real time, wherein the optimal operation scheme is presented in the form of an optimal operation scheme report. And pushing the capacity prediction data and the optimal operation scheme to equipment maintenance personnel through short messages or nails when the capacity of the target network element is too high, so that the equipment maintenance personnel can remind in real time and correct in time. According to the method and the device, the network element capacity data, the capacity prediction data and the optimal operation scheme report of the target network element are presented on the page, so that a user can know the change and trend of the data in time, the data is easier to understand and analyze, and the user experience is improved. The data is stored in the database in a lasting way through the network element capacity data, the capacity prediction data and the optimal operation scheme report of the target network element, so that the data is ensured to be stored for a long time and available at any time. By pushing and early warning the capacity prediction data and the optimal operation scheme when the capacity of the target network element is too high, a manager can know the data condition in time and take action in time, and decision making and problem solving are supported.

According to the network element operation scheme generation method, capacity prediction is carried out on the network elements based on the collected network element capacity data and the capacity prediction model, classification judgment is carried out on the capacity prediction data based on the optimal operation scheme recommendation model, the capacity abnormal result is identified to output the optimal operation scheme of the network elements, real-time monitoring and prediction of the network element capacity, intelligent judgment of the network element capacity prediction data and automatic generation of the network element operation scheme are achieved, timely pushing of capacity early warning information and the optimal operation scheme is achieved, quality improvement and efficiency improvement are achieved, meanwhile stability of the 5G core network is guaranteed, and labor cost is reduced. In addition, the characteristics and the structure of the network element capacity data are evaluated through the stability test and the white noise test, guidance is provided for subsequent modeling and prediction, the property of the data can be better understood, a proper modeling method is selected, and the accuracy and the reliability of prediction are improved. Generating the report of the optimal operation scheme based on the optimal operation scheme and the preset report structure framework can improve the quality and effect of the report, help users to better understand and accept the optimal operation scheme, support decision making and execution, and promote the sustainable development and success of organizations. By page presentation of network element capacity data, capacity prediction data and an optimal operation scheme report of the target network element, a user can know data change and trend in time, the data is easier to understand and analyze, and user experience is improved. The data is stored in the database in a lasting way through the network element capacity data, the capacity prediction data and the optimal operation scheme report of the target network element, so that the data is ensured to be stored for a long time and available at any time. By pushing and early warning the capacity prediction data and the optimal operation scheme when the capacity of the target network element is too high, a manager can know the data condition in time and take action in time, and decision making and problem solving are supported.

Example 3:

as shown in fig. 6, this embodiment provides a generating device of a network element operation scheme, including: the acquiring module 31, the predicting module 32, and the generating module 33, the acquiring module 31 is configured to acquire network element full data, and extract network element capacity data from the network element full data, where the network element capacity data includes: the prediction module 32 is connected with the acquisition module 31, and is configured to perform capacity prediction on the target network element according to the network element capacity data and the capacity prediction model, so as to obtain capacity prediction data, where the capacity prediction data includes: the generation module 33 is connected with the prediction module 32 and is used for generating an optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model.

Optionally, the generating device of the network element operation scheme further includes: a checking module 30 and a constructing module 34, the checking module 30 is configured to perform data checking on network element capacity data, where the data checking includes one or more of the following combinations: and the construction module is used for constructing a capacity prediction model and an optimal operation scheme recommendation model, wherein the capacity prediction model comprises an autoregressive AR model, a moving average MA model and an autoregressive moving average ARMA model.

Specifically, the prediction module 32 includes: the device comprises a determining unit 321, a predicting unit 322, a calculating unit 323, a first judging unit 324, a first responding unit 325 and a second responding unit 326, wherein the determining unit 321 is used for determining a capacity predicting model based on target network element capacity data and estimating capacity predicting model parameters, the predicting unit 322 is used for carrying out capacity prediction on the target network element according to the capacity predicting model and the target network element capacity data to obtain capacity predicting data, and the calculating unit 323 is used for calculating errors of the capacity predicting data based on error indexes and actual values of the target network element capacity data, wherein the error indexes comprise one or more of the following combinations: the capacity prediction method comprises the steps of average absolute error, root mean square error and average absolute percentage error, a first judging unit 324 used for judging the magnitude of the error and a preset threshold value, a first responding unit 325 used for responding to the error being smaller than the preset threshold value and outputting capacity prediction data, and a second responding unit 326 used for responding to the error being larger than or equal to the preset threshold value and re-estimating capacity prediction model parameters and carrying out capacity prediction until the error is smaller than the preset threshold value.

Specifically, the generation module 33 includes: a second judging unit 331 and a third responding unit 332, where the second judging unit 331 is configured to judge whether the capacity of the target network element is too high according to the maximum license capacity utilization rate in the preset time period of the target network element, and the third responding unit 332 is configured to generate an optimal operation scheme according to the optimal operation scheme recommendation model and the maximum number of users in the preset time period of the target network element in response to the target network element capacity being too high.

Specifically, the second judgment unit 331 includes: the system comprises a first judging subunit and a first responding subunit, wherein the first judging subunit is used for judging whether the maximum license capacity utilization rate in a preset time period of a target network element is larger than a first preset value, and the second responding subunit is used for responding to the fact that the maximum license capacity utilization rate is larger than the first preset value and determining that the capacity of the target network element is too high.

Specifically, the third response unit 332 includes: the system comprises a first judging subunit, a first response subunit and a second response subunit, wherein the first judging subunit is used for judging whether the maximum user number in a preset time period of a target network element is smaller than the splitting threshold of the target network element, the second response subunit is used for generating a target network topological graph, a capacity expansion threshold and a capacity expansion scheme based on an optimal operation scheme recommendation model to serve as an optimal operation scheme of the target network element, and the third response subunit is used for generating the target network topological graph and a newly built network element scheme based on the optimal operation scheme recommendation model to serve as the optimal operation scheme of the target network element in response to the fact that the maximum user number in the preset time period of the target network element is larger than or equal to the splitting threshold of the target network element.

Optionally, the generating device of the network element operation scheme further includes: the system comprises a report generating module 35, a presenting module 36 and an early warning module 37, wherein the second generating module 35 is used for generating an optimal operation scheme report based on an optimal operation scheme, a preset report structure frame and configuration files and performance files of network elements, the presenting module 36 is used for carrying out page presentation on network element capacity data, capacity prediction data and the optimal operation scheme report of a target network element and storing the page presentation into a database, and the early warning module 37 is used for carrying out push early warning on the capacity prediction data and the optimal operation scheme in response to the fact that the capacity of the target network element is too high.

It can be understood that the generating device of the network element operation scheme provided above is configured to execute the method corresponding to the embodiment 1 provided above, so that the beneficial effects achieved by the generating device can refer to the method of the embodiment 1 and the beneficial effects of the corresponding scheme in the following detailed description, which are not repeated herein.

Example 4:

The present embodiment also provides an electronic device, including a memory and a processor, where the memory stores a computer program, and the processor is configured to run the computer program to implement the method for generating the network element operation scheme in embodiment 1.

Example 5:

the present embodiment also provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method for generating the network element operation scheme in embodiment 1 above.

It is to be understood that the above embodiments are merely illustrative of the application of the principles of the present invention, but not in limitation thereof. Various modifications and improvements may be made by those skilled in the art without departing from the spirit and substance of the invention, and are also considered to be within the scope of the invention.

Claims (10)

1. A method for generating a network element operating scheme, comprising:

Acquiring network element full data and extracting network element capacity data from the network element full data, wherein the network element capacity data comprises: license capacity utilization, current user number, and split threshold;

Performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model to obtain capacity prediction data, wherein the capacity prediction data comprises: the maximum license capacity utilization rate and the maximum number of users in a preset time period of a target network element;

And generating an optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model.

2. The method according to claim 1, wherein after the extracting the network element capacity data from the network element full data and before the performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model, further comprises:

Performing data verification on network element capacity data, wherein the data verification comprises one or more of the following combinations: the stability test, the white noise test,

After the acquiring the network element full data and before the performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model, the method further comprises:

and constructing a capacity prediction model and an optimal operation scheme recommendation model, wherein the capacity prediction model comprises an autoregressive AR model, a moving average MA model and an autoregressive moving average ARMA model.

3. The method for generating a network element operation scheme according to claim 1, wherein the performing capacity prediction on the target network element according to the network element capacity data and the capacity prediction model to obtain capacity prediction data specifically includes:

Determining a capacity prediction model based on the capacity data of the target network element, and estimating capacity prediction model parameters;

Performing capacity prediction on the target network element according to the capacity prediction model and the capacity data of the target network element to obtain capacity prediction data;

calculating an error of the capacity prediction data based on the error indicator and the target network element capacity data actual value, wherein the error indicator comprises one or a combination of more of the following: average absolute error, root mean square error, average absolute percentage error;

Judging the magnitude of the error and a preset threshold value;

Outputting capacity prediction data in response to the error being less than a preset threshold;

And re-estimating the capacity prediction model parameters and performing capacity prediction until the error is smaller than a preset threshold value in response to the error being greater than or equal to the preset threshold value.

4. The method for generating a network element operation scheme according to claim 1, wherein the generating the optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model specifically comprises:

judging whether the capacity of the target network element is too high according to the maximum license capacity utilization rate in the preset time period of the target network element;

And responding to the excessive capacity of the target network element, and generating an optimal operation scheme according to the optimal operation scheme recommendation model and the maximum number of users in a preset time period of the target network element.

5. The method for generating a network element operation scheme according to claim 4, wherein the determining whether the capacity of the target network element is too high according to the maximum license capacity utilization rate within the preset time period of the target network element specifically includes:

judging whether the maximum license capacity utilization rate in a preset time period of a target network element is larger than a first preset value;

And determining that the capacity of the target network element is too high in response to the maximum license capacity utilization being greater than a first preset value.

6. The method for generating a network element operation scheme according to claim 4, wherein the generating the optimal operation scheme of the target network element according to the optimal operation scheme recommendation model and the maximum number of users in the target network element preset time period in response to the target network element capacity being too high specifically comprises:

Judging whether the maximum number of users in a preset time period of a target network element is smaller than a splitting threshold of the target network element or not;

Generating a target network topological graph, a capacity expansion threshold and a capacity expansion scheme based on an optimal operation scheme recommendation model to serve as an optimal operation scheme of the target network element in response to the fact that the maximum number of users in a preset time period of the target network element is smaller than a splitting threshold of the target network element;

And generating a target network topological graph and a newly-built network element scheme as optimal operation schemes of the target network element based on the optimal operation scheme recommendation model in response to the maximum number of users in a preset time period of the target network element being greater than or equal to a splitting threshold of the target network element.

7. The method according to claim 4, further comprising, after the generating the optimal operation plan of the target network element according to the capacity prediction data and the optimal operation plan recommendation model:

generating an optimal operation scheme report based on an optimal operation scheme and a preset report structure frame;

performing page presentation on network element capacity data, capacity prediction data and an optimal operation scheme report of a target network element, and storing the page presentation in a database;

And responding to the excessive capacity of the target network element, and carrying out push early warning on the capacity prediction data and the optimal operation scheme.

8. A device for generating a network element operation scheme, comprising: the system comprises an acquisition module, a prediction module and a generation module,

The acquisition module is used for acquiring the network element full data and extracting the network element capacity data from the network element full data, wherein the network element capacity data comprises: license capacity utilization, current number of users and split threshold,

The prediction module is connected with the acquisition module and is used for carrying out capacity prediction on the target network element according to the network element capacity data and the capacity prediction model to obtain capacity prediction data, wherein the capacity prediction data comprises: the maximum license capacity utilization and the maximum number of users after a preset period of time by the target network element,

And the generating module is connected with the prediction module and is used for generating the optimal operation scheme of the target network element according to the capacity prediction data and the optimal operation scheme recommendation model.

9. An electronic device comprising a memory and a processor, the memory having stored therein a computer program, the processor being arranged to run the computer program to implement a method of generating a network element operating scheme as claimed in any one of claims 1 to 7.

10. A computer readable storage medium having stored thereon a computer program, which when executed by a processor, implements a method of generating a network element operating scheme according to any of claims 1 to 7.