CN114553733B - Intelligent gateway monitoring management system and method based on artificial intelligence - Google Patents

Intelligent gateway monitoring management system and method based on artificial intelligence Download PDF

Info

Publication number
CN114553733B
CN114553733B CN202210440949.8A CN202210440949A CN114553733B CN 114553733 B CN114553733 B CN 114553733B CN 202210440949 A CN202210440949 A CN 202210440949A CN 114553733 B CN114553733 B CN 114553733B
Authority
CN
China
Prior art keywords
data
gateway
time
value
flow
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210440949.8A
Other languages
Chinese (zh)
Other versions
CN114553733A (en
Inventor
张林利
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu Electric Nanny Electric Service Co ltd
Original Assignee
Jiangsu Electric Nanny Electric Service Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jiangsu Electric Nanny Electric Service Co ltd filed Critical Jiangsu Electric Nanny Electric Service Co ltd
Priority to CN202210440949.8A priority Critical patent/CN114553733B/en
Publication of CN114553733A publication Critical patent/CN114553733A/en
Application granted granted Critical
Publication of CN114553733B publication Critical patent/CN114553733B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/06—Generation of reports
    • H04L43/067—Generation of reports using time frame reporting
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/06—Management of faults, events, alarms or notifications
    • H04L41/0681—Configuration of triggering conditions
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/147—Network analysis or design for predicting network behaviour
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/04—Processing captured monitoring data, e.g. for logfile generation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/06—Generation of reports
    • H04L43/062—Generation of reports related to network traffic
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00—Arrangements for monitoring or testing data switching networks
    • H04L43/16—Threshold monitoring
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Mining & Analysis (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Computer And Data Communications (AREA)

Abstract

The invention discloses an intelligent gateway monitoring and management system and method based on artificial intelligence.A gateway data association item integration module analyzes corresponding sets of various data of gateway data, judges whether the different sets have association or not, and integrates the sets with the association to respectively obtain each association data set corresponding to the gateway data association item; the gateway associated data intelligent analysis module clusters each associated data group and analyzes the cluster to obtain a first change rate, a second change rate and a third change rate corresponding to each category; the gateway traffic data prediction module predicts gateway traffic data by combining each associated data group corresponding to the gateway data association item at the current time and the associated data group in the historical data; and the gateway traffic data calibration module calibrates the prediction result of the gateway traffic data according to the prediction result of the whole gateway traffic data in the region.

Description

Intelligent gateway monitoring management system and method based on artificial intelligence
Technical Field
The invention relates to the technical field of gateway systems, in particular to an intelligent gateway monitoring and management system and method based on artificial intelligence.
Background
With the rapid development of computer technology, people have more and more extensive application to networks, and people need to convert information through a gateway when obtaining flow information through the network, so that the effective monitoring of the service condition of a user network can be realized by monitoring the condition of flow data corresponding to each request in the gateway.
The existing intelligent monitoring system for the gateway has a great disadvantage that the total amount of the traffic used by the user in a certain period of time in the historical data can be monitored only in a statistical summation mode, the monitored historical data is inherent, and the traffic use condition of the user in a certain period of time in the future cannot be effectively predicted.
In view of the above, there is a need for an intelligent gateway monitoring and management system and method based on artificial intelligence.
Disclosure of Invention
The invention aims to provide an intelligent gateway monitoring and management system and method based on artificial intelligence, so as to solve the problems in the background technology.
In order to solve the technical problems, the invention provides the following technical scheme: an intelligent gateway monitoring management system based on artificial intelligence, comprising:
the gateway data acquisition module is used for asynchronously acquiring gateway data, writing an acquisition result into a first log, analyzing the content of the first log, and respectively extracting various data of the gateway data to obtain a corresponding set of the various data of the gateway data;
the gateway data association item integration module analyzes sets corresponding to various data of gateway data, judges whether the elements in different sets have association or not, and integrates the elements through the set elements with the association to respectively obtain each association data set corresponding to the gateway data association item;
the gateway associated data intelligent analysis module is used for clustering and analyzing each associated data group to obtain a first change rate, a second change rate and a third change rate corresponding to each category;
the gateway traffic data prediction module is used for predicting gateway traffic data by combining each associated data set corresponding to the gateway data association item at the current time and the associated data set in the historical data;
an early warning module which compares the prediction result of the gateway flow data from the gateway flow data prediction module with a threshold value,
when the prediction result is more than or equal to the threshold value, the early warning module gives an alarm to the user,
when the prediction result is smaller than the threshold value, the early warning module does not give an alarm to the user;
when gateway traffic data is predicted, a first predicted value W1 is obtained through the second change rate and the third change rate, and a second predicted value W2 is obtained through the first change rate, so that a final predicted value W of the gateway traffic data is { W1, W2} max, wherein { W1, W2} max represents a maximum value of W1 and W2.
The invention realizes the monitoring of the traffic use condition in the gateway through the cooperative cooperation of all the modules, simultaneously predicts the traffic use condition in the gateway at the next stage according to the monitored historical data, and pre-warns the user in advance according to the prediction result to ensure the normal use of the corresponding traffic data in the gateway.
Furthermore, the gateway data acquisition module asynchronously acquires flow data, the contents corresponding to different flow data are independent of each other, one flow data corresponds to one request, one request corresponds to one software interface, and one software interface can correspond to multiple requests;
the gateway data items include: the size of each piece of flow data, the request time corresponding to each piece of flow data, and a software interface corresponding to the request corresponding to each piece of flow data;
the gateway data acquisition module analyzes the first log content once every other first unit time;
the gateway data acquisition module records the size of each piece of flow data corresponding to the first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a flow value data set A, and records the value corresponding to the nth element in the flow value data set A as An;
the gateway data acquisition module records request time corresponding to each piece of flow data corresponding to first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a request time set B, and records a value corresponding to the nth element in the request time set B as Bn;
the gateway data acquisition module records the software interfaces corresponding to the requests corresponding to each flow data of each flow data corresponding to the first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a software interface set C, and records the value corresponding to the nth element in the software interface set C as Cn;
the number of elements corresponding to the flow value data set A, the request time set B and the software interface set C is equal to the number of flow data corresponding to the first unit time in the first log content analyzed by the gateway data acquisition module;
the gateway data acquisition module also monitors the running state of each software interface in real time so as to obtain the running state time curve of each software interface, each running state time curve represents the change of the running state of the corresponding interface software along with time, the running state comprises an opening state and a closing state,
and the value of the running state time curve corresponding to the opening state is marked as 1, and the value of the running state time curve corresponding to the closing state is marked as 0.
In the gateway data, each request sent by each software interface corresponds to one piece of flow data, and each piece of flow data is independent, so that the gateway data is analyzed, the analyzed data needs to be refined, and the condition of each piece of flow data corresponding to each request of each software interface is locked, and the size of each piece of flow data, the request time corresponding to each piece of flow data, and the software interface corresponding to each request of each piece of flow data are further obtained; setting a gateway data acquisition module to analyze the first log content once every first unit time, so as to ensure the frequency of analyzing the gateway data and also to lock the range corresponding to the analyzed data (the condition of the flow data corresponding to the first unit time in the first log content analyzed last time) when analyzing the gateway data each time; obtaining a flow value data set A, a request time set B and a software interface set C, wherein the flow value data set A, the request time set B and the software interface set C are used for uniformly storing and managing the acquired data and simultaneously quickly obtaining a corresponding associated data set during data analysis; obtaining the running state time curve of each software interface, so as to obtain the relation between the running state of each software interface and the time, and further quickly counting the total running time corresponding to the specified software interface in the first unit time; the values of the operating state time curve corresponding to the operating state are set to be 1 and 0, so as to clearly and intuitively reflect the operating state corresponding to the specified time of the specified software interface (1 represents an open state, and 0 represents a closed state).
Further, the method for judging whether the association exists between the elements in different sets by the gateway data association integration module comprises the following steps:
s1.1, acquiring a flow value data set A, a request time set B and a software interface set C corresponding to the latest analysis of a first log content in a gateway data acquisition module;
s1.2, extracting a value An1 corresponding to the n1 th element in the A, extracting a value Bn2 corresponding to the n2 th element in the B and extracting a value Cn3 corresponding to the n3 th element in the C;
s1.3, comparing the size relationship among n1, n2 and n3,
when the element values of the sets corresponding to the two sets are equal to each other between n1 and n2, between n1 and n3, or between n2 and n3, the element values of the sets corresponding to the two sets are judged to have relevance,
when the values of the elements in the sets corresponding to n1, n2 and n3 are not equal to each other in the ranges of n1 and n2, n1 and n3 or n2 and n3, judging that no relevance exists among the values of the elements in the sets corresponding to n1, n2 and n 3;
the method for obtaining each associated data group corresponding to the gateway data associated item by the gateway data associated item integration module comprises the following steps:
s2.1, acquiring a flow value data set A, a request time set B and a software interface set C corresponding to the last analysis of the first log content in a gateway data acquisition module;
s2.2, extracting a value Ai corresponding to the ith element in the A;
s2.3, extracting a value Bi1 corresponding to the element with the relevance to Ai in B and extracting a value Ci2 corresponding to the element with the relevance to Ai in C;
and S2.4, obtaining the ith associated data group Zi corresponding to the latest analysis of the first log content in the gateway data acquisition module, wherein Zi is [ Ai, Bi1, Ci2 ].
Further, when the gateway associated data intelligent analysis module clusters each associated data group, the number of the categories is the same as the number of the software interface categories corresponding to each element in the software interface set C, and each associated data group with the same software interface is divided into the same category,
recording the m-th element in the k-th associated data group in the j class
Figure GDA0003693575180000041
J is more than or equal to 0 and less than or equal to x, and x is the number of the software interface types corresponding to each element in the software interface set C.
The gateway associated data intelligent analysis module divides associated data groups with the same software interface into the same class, and is used for analyzing the flow use condition corresponding to each software interface in the gateway subsequently, predicting the flow use condition corresponding to each software interface, and summarizing and accumulating predicted values corresponding to each software interface so as to predict gateway data.
Further, the method for the gateway associated data intelligent analysis module to obtain the first change rate corresponding to each category includes the following steps:
s3.1, obtaining the sum of the corresponding values of the 1 st element in each associated data group in the jth class to obtain the total flow of the software interface corresponding to the jth class corresponding to the first unit time before the first log content analyzed last time
Figure GDA0003693575180000042
The above-mentioned
Figure GDA0003693575180000043
Wherein,
Figure GDA0003693575180000044
represents the value corresponding to the 1 st element in the kth associated data group in the jth class, and k1j represents the associated data in the jth classThe total number of groups;
s3.2, acquiring total flow corresponding to the first unit time in the first log content analyzed by the software interface corresponding to the jth class at the first k2 times
Figure GDA0003693575180000051
S3.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring total flow corresponding to the previous first unit time in the first k 2-time analyzed first log content by the software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure GDA0003693575180000052
K2 is more than or equal to 1 and less than or equal to k3, p is more than or equal to 0 and less than or equal to p1, k3 is a first preset value, and p1 is a second preset value;
s3.4, acquiring corresponding total flow when k2 is different values in the previous p days
Figure GDA0003693575180000053
Maximum value of (1), is noted
Figure GDA0003693575180000054
S3.5, when p is judged to be different values, respectively corresponding
Figure GDA0003693575180000055
Whether or not it is meaningful to have the information,
when in use
Figure GDA0003693575180000056
Then, it is determined
Figure GDA0003693575180000057
Meaningless, and
Figure GDA0003693575180000058
wherein g represents a pair
Figure GDA0003693575180000059
The normalized process equation of (a) is,
when in use
Figure GDA00036935751800000510
Time, determine
Figure GDA00036935751800000511
Is significant in that
Figure GDA00036935751800000512
Figure GDA00036935751800000513
S3.6, obtaining a first change rate corresponding to the jth category
Figure GDA00036935751800000514
The above-mentioned
Figure GDA00036935751800000515
Wherein,
Figure GDA00036935751800000516
is composed of
Figure GDA00036935751800000517
The coefficient of adjustment of (a) is,
Figure GDA00036935751800000518
both alpha 1 and beta 1 are constant values and
Figure GDA00036935751800000519
in the process of acquiring the first change rate corresponding to each category by the gateway associated data intelligent analysis module, the total flow corresponding to the first unit time before the last analysis of the software interface corresponding to the jth category in the first log content is acquired
Figure GDA00036935751800000520
Is to solve each kind of software interfaceThe total flow corresponding to the previous first unit time in the analyzed first log content is used as a data analysis unit, and the total flow corresponding to each software interface is further analyzed from two angles of history P, k2, so that a first change rate corresponding to each category is obtained; obtaining
Figure GDA00036935751800000521
The reason is that k2 in the previous p days are different values, so that the previous p days in the historical data correspond to a plurality of total flows, during data prediction, the worst case of the possibly occurring events needs to be considered, and the prediction result can achieve the purpose of early warning, and the worst case of the previous p days in the historical data corresponding to a plurality of total flows is each k2 bit different value
Figure GDA00036935751800000522
Maximum value of
Figure GDA00036935751800000523
Computing
Figure GDA0003693575180000061
In order to obtain the rate of change of the total flow rate in a first unit time; is provided with a pair
Figure GDA0003693575180000062
Is to avoid the settlement result
Figure GDA0003693575180000063
A meaningless situation occurs, so that a prediction result is meaningless, and the final prediction result of the gateway data is influenced; when the first change rate is calculated, p is set as p1-1 because the value range of p is 0 ≦ p1, and p +1 occurs in the calculation process, and the upper limit of p at this time can be deduced as p is p1-1 through 0 ≦ p +1 ≦ p 1; setting when calculating the first rate of change
Figure GDA0003693575180000064
Adjustment coefficient of
Figure GDA0003693575180000065
Is due to
Figure GDA0003693575180000066
The method includes the steps that the increase change rate of total flow of software interfaces corresponding to the jth class corresponding to the jth day corresponding to the previous p +1 day is set for the previous p day, the first change rate is obtained relative to the increase change rate of the total flow of software interfaces corresponding to the jth class corresponding to the current time, the flow use condition of the software interfaces per se has large fluctuation, therefore, certain deviation exists between the corresponding increase change rate in historical data and the first change rate needing to be obtained, the first change rate needs to be corrected to be used as reference data for obtaining the first change rate, corresponding adjusting coefficients are set for the corresponding increase change rate in the historical data, and the purpose of calibrating each corresponding increase change rate in the historical data is achieved
Figure GDA0003693575180000067
Corresponding adjustment coefficient is
Figure GDA0003693575180000068
The calibrated growth rate of change is
Figure GDA0003693575180000069
Figure GDA00036935751800000610
Corresponding to the rate of change of growth from historical data
Figure GDA00036935751800000611
Obtaining the increase change rate of the total flow of the software interfaces corresponding to the jth class corresponding to the current time), and obtaining a first change rate corresponding to the jth class by means of averaging according to the increase change rate of the total flow of the software interfaces corresponding to the jth classes after calibration, wherein the first change rates obtained by the means are obtained by referring to the current time, and relatively speaking, the first change rates are more accurate, so that the prediction result of gateway flow data is more accurate(ii) a When obtaining the adjustment coefficient, considering a factor of a time difference between a time corresponding to an increase change rate and a current time, in general, the larger the time difference is, the smaller the referential significance of the corresponding increase change rate on obtaining the first change rate is, the more conservative prediction needs to be performed, and then the size of the adjustment coefficient needs to be continuously adjusted according to the length of the time difference, α 1 is the adjustment coefficient corresponding to the increase change rate corresponding to the jth category on the day, and the adjustment coefficient is set
Figure GDA00036935751800000612
The method is used for determining an adjusting value corresponding to the change condition of an adjusting coefficient along with the time difference, the whole adjusting value is in a descending trend and is reduced along with the increase of the time difference, and the beta 1 reflects the size degree of the adjusting value; the whole descending trend is set to reduce the influence of the growth change rate with large time difference on the acquired first change rate, so that the calibrated value of the growth change rate with large time difference is continuously set, the result of the first change rate can be reduced to a certain extent, the interference degree of the first change rate with the growth change rate with large time difference is reduced, the acquired first change rate is deviated from a conservative value, and the finally estimated interference degree of the gateway flow data with the history data with large time difference is reduced.
Further, the method for the gateway associated data intelligent analysis module to obtain the second change rate corresponding to each category includes the following steps:
s4.1, obtaining the total time length corresponding to the time curve median value of the operation state of the software interface corresponding to the jth class is 1, and obtaining the total operation time length corresponding to the first unit time before the software interface corresponding to the jth class in the first log content analyzed last time
Figure GDA0003693575180000071
S4.2, acquiring the total operation duration corresponding to the first unit time in the first k2 times of analyzed first log content of the software interface corresponding to the jth class
Figure GDA0003693575180000072
S4.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring the total operating duration corresponding to the previous first unit time in the first k 2-time analyzed first log content of the software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure GDA0003693575180000073
S4.4, acquiring corresponding total operation time when k2 is different in the previous p days
Figure GDA0003693575180000074
Maximum value of (1), is noted
Figure GDA0003693575180000075
S4.5, when p is judged to be different values, respectively corresponding
Figure GDA0003693575180000076
Whether or not it is meaningful to have the information,
when in use
Figure GDA0003693575180000077
Then, it is determined
Figure GDA0003693575180000078
Meaningless, and
Figure GDA0003693575180000079
Figure GDA00036935751800000710
wherein g1 represents a pair
Figure GDA00036935751800000711
The normalized process equation of (a) is,
when the temperature is higher than the set temperature
Figure GDA00036935751800000712
Time, determine
Figure GDA00036935751800000713
Is significant in that
Figure GDA00036935751800000714
Figure GDA00036935751800000715
S4.6, obtaining a second change rate corresponding to the jth category
Figure GDA00036935751800000716
The described
Figure GDA00036935751800000717
Wherein,
Figure GDA00036935751800000718
is composed of
Figure GDA00036935751800000719
The coefficient of adjustment of (a) is,
Figure GDA00036935751800000720
both alpha 2 and beta 2 are constant values and
Figure GDA00036935751800000721
in the process of obtaining the second change rate corresponding to each category by the gateway associated data intelligent analysis module, the total operation time corresponding to the first unit time before the software interface corresponding to the jth category in the first log content analyzed last time is obtained first
Figure GDA0003693575180000081
The total operation time length corresponding to each software interface is analyzed from two perspectives of history P, k2 by taking the total operation time length corresponding to the first unit time before the first log content analyzed each time of each type of corresponding software interface as a data analysis unit, and then the total operation time length corresponding to each software interface is obtainedA second rate of change to each category; obtaining
Figure GDA0003693575180000082
The reason is that since k2 is different values in the previous p days, the previous p days in the historical data correspond to a plurality of total operating durations, during data prediction, the worst case of the possibly occurring events needs to be considered, and then the prediction result can achieve the purpose of early warning, and the worst case in the previous p days in the historical data corresponding to a plurality of total operating durations is each time when the k2 bits are different values
Figure GDA0003693575180000083
Maximum value of
Figure GDA0003693575180000084
Computing
Figure GDA0003693575180000085
The change rate of the total operation time length in the first unit time is obtained; is provided with a pair
Figure GDA0003693575180000086
Is to avoid the settlement result
Figure GDA0003693575180000087
A meaningless situation occurs, so that a prediction result is meaningless, and the final prediction result of gateway data is influenced; setting an adjustment factor when calculating the first rate of change
Figure GDA0003693575180000088
Acting on the adjustment coefficient
Figure GDA0003693575180000089
Has the same effect as that of (1) and also has the effect of adjusting the calibration
Figure GDA00036935751800000810
For is to
Figure GDA00036935751800000811
Adjustment calibration is performed).
Further, the method for the gateway associated data intelligent analysis module to obtain the third change rate corresponding to each category includes the following steps:
s5.1, acquiring request time corresponding to the 2 nd element in each associated data group in the j category, calculating the time difference between two adjacent request times,
the time difference between the request time corresponding to the v +1 th correlated data set in the j-th class and the request time corresponding to the v-th correlated data set in the j-th class is denoted as tv,
acquiring a flow value corresponding to the 1 st element in the v-th associated data group in the j type and recording the flow value as
Figure GDA00036935751800000812
When v is calculated to be different values respectively,
Figure GDA00036935751800000813
quotient to tv
Figure GDA00036935751800000814
Further, the flow consumption value of the software interface corresponding to the jth class in the first unit time before the first log content analyzed last time is obtained
Figure GDA00036935751800000815
The above-mentioned
Figure GDA00036935751800000816
Wherein k1j represents the total number of associated data groups in the j-th class;
s5.2, acquiring total flow corresponding to the first unit time in the first log content analyzed by the software interface corresponding to the jth class at the first k2 times
Figure GDA00036935751800000817
S5.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring total flow corresponding to previous first unit time in first k2 times of analyzed first log content by a software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure GDA0003693575180000091
K3 is more than or equal to 1 and less than or equal to k2, p is more than or equal to 0 and less than or equal to p1, k3 is a first preset value, and p1 is a second preset value;
s5.4, obtaining corresponding total flow when k2 is different in the previous p days
Figure GDA0003693575180000092
Maximum value of (1), is recorded as
Figure GDA0003693575180000093
S5.5, when p is judged to be different values, respectively corresponding
Figure GDA0003693575180000094
Whether or not it is meaningful to have the information,
when in use
Figure GDA0003693575180000095
Then, it is determined
Figure GDA0003693575180000096
Meaningless, and
Figure GDA0003693575180000097
Figure GDA0003693575180000098
wherein g2 represents a pair
Figure GDA0003693575180000099
The normalized processing equation of (a) is,
when the temperature is higher than the set temperature
Figure GDA00036935751800000910
Time, judge
Figure GDA00036935751800000911
Is significant in that
Figure GDA00036935751800000912
Figure GDA00036935751800000913
S5.6, obtaining a third change rate corresponding to the jth category
Figure GDA00036935751800000914
The above-mentioned
Figure GDA00036935751800000915
Wherein,
Figure GDA00036935751800000916
is composed of
Figure GDA00036935751800000917
The coefficient of adjustment of (a) is,
Figure GDA00036935751800000918
both alpha 3 and beta 3 are constant and
Figure GDA00036935751800000919
further, the method for predicting gateway traffic data by the gateway traffic data prediction module includes the following steps:
s6.1, obtaining a first predicted value W1 of the gateway flow data in the first unit time before in the first log content analyzed next time based on the current time,
when j is 0, the W1 is 0,
when j ≠ 0, the
Figure GDA00036935751800000920
S6.2, obtaining a second predicted value W2 of the gateway flow data in the previous first unit time in the first log content analyzed next time based on the current time,
when j is 0, the W2 is 0,
when j ≠ 0, the
Figure GDA00036935751800000921
And S6.3, obtaining a final predicted value W of the gateway traffic data in the first unit time in the first log content analyzed next time based on the current time, wherein W is { W1, W2} max.
When the gateway flow data prediction module predicts gateway flow data, two prediction modes are adopted, and two prediction results are screened to obtain a final prediction value of the gateway flow data; when the first predicted value W1 is obtained, the prediction is performed by the second change rate and the third change rate, and the prediction is analyzed from two angles of the increase of the service time of the software interface and the increase of the flow consumption value of the software interface per unit time; when the second predicted value W2 is obtained, the prediction is performed by the first change rate, and the analysis is made from the perspective of the increase of the total flow used by the software interface; because the first change rate, the second change rate and the third change rate all adopt a mode of adjusting coefficients in the process of obtaining, interference caused by an increase change rate corresponding to historical data with a large time difference is reduced, and then the obtained first change rate, the second change rate and the third change rate belong to conservative values, that is, the obtained values are possibly slightly smaller than actual data, therefore, when obtaining a final predicted value, a mode of selecting a maximum value (W ═ W1, W2 ═ max) is adopted to obtain a predicted result, and then deviation between the predicted value and the actual value is reduced, and the technical effect of reducing errors is achieved.
An intelligent gateway monitoring management method based on artificial intelligence, the method comprises the following steps:
s1, asynchronously collecting gateway data through a gateway data collection module, writing a collection result into a first log, analyzing the content of the first log, and respectively extracting various data of the gateway data to obtain a corresponding set of the various data of the gateway data;
s2, analyzing sets corresponding to various data of gateway data through a gateway data association item integration module, judging whether the elements in different sets have relevance, and integrating through the set elements with relevance to respectively obtain each association data set corresponding to the gateway data association item;
s3, clustering and analyzing each associated data group through a gateway associated data intelligent analysis module to obtain a first change rate, a second change rate and a third change rate corresponding to each category;
s4, predicting gateway traffic data by combining each associated data set corresponding to the gateway data association item at the current time and the associated data set in the historical data through a gateway traffic data prediction module;
s5, in the early warning module, the gateway traffic data prediction module compares the prediction result of the gateway traffic data with the threshold value,
when the prediction result is more than or equal to the threshold value, the early warning module gives an alarm to the user,
and when the prediction result is smaller than the threshold value, the early warning module does not give an alarm to the user.
Compared with the prior art, the invention has the following beneficial effects: the invention uses the artificial intelligence technology, acquires and analyzes the flow use conditions corresponding to different software in the historical data of the gateway, and further respectively obtains the total operation time length change rate, the unit time flow consumption value change rate and the total flow change rate of different software interfaces in unit time, and further accurately predicts the flow use conditions and the gateway flow data corresponding to different software interfaces of the gateway after unit time, thereby achieving the technical effect of early warning and realizing the effective monitoring and management of the gateway.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a schematic structural diagram of an intelligent gateway monitoring and management system based on artificial intelligence according to the present invention;
fig. 2 is a schematic flowchart of a method for obtaining a first change rate corresponding to each category by a gateway association data intelligent analysis module in the intelligent gateway monitoring and management system based on artificial intelligence according to the present invention;
fig. 3 is a schematic flowchart of a method for obtaining a second change rate corresponding to each category by a gateway association data intelligent analysis module in the intelligent gateway monitoring and management system based on artificial intelligence according to the present invention;
fig. 4 is a schematic flow chart of an intelligent gateway monitoring and management method based on artificial intelligence according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1-4, the present invention provides a technical solution: an intelligent gateway monitoring and management system based on artificial intelligence, comprising:
the gateway data acquisition module is used for asynchronously acquiring gateway data, writing an acquisition result into a first log, analyzing the content of the first log, and respectively extracting various data of the gateway data to obtain a corresponding set of the various data of the gateway data;
the gateway data association item integration module analyzes corresponding sets of gateway data items, judges whether the elements in different sets have association or not, and integrates the set elements with the association to respectively obtain each association data set corresponding to the gateway data association item;
the gateway associated data intelligent analysis module is used for clustering and analyzing each associated data group to obtain a first change rate, a second change rate and a third change rate corresponding to each category;
the gateway traffic data prediction module is used for predicting gateway traffic data by combining each associated data set corresponding to the gateway data association item at the current time and the associated data set in the historical data;
an early warning module which compares the prediction result of the gateway flow data from the gateway flow data prediction module with a threshold value,
when the prediction result is more than or equal to the threshold value, the early warning module gives an alarm to the user,
when the prediction result is smaller than the threshold value, the early warning module does not give an alarm to the user;
when gateway traffic data is predicted, a first predicted value W1 is obtained through the second change rate and the third change rate, and a second predicted value W2 is obtained through the first change rate, so that a final predicted value W of the gateway traffic data is { W1, W2} max, wherein { W1, W2} max represents a maximum value of W1 and W2.
The invention realizes the monitoring of the traffic use condition in the gateway through the cooperative cooperation of all the modules, simultaneously predicts the traffic use condition in the gateway at the next stage according to the monitored historical data, and pre-warns the user in advance according to the prediction result to ensure the normal use of the corresponding traffic data in the gateway.
The gateway data acquisition module asynchronously acquires flow data, contents corresponding to different flow data are mutually independent, one flow data corresponds to one request, one request corresponds to one software interface, and one software interface can correspond to multiple requests;
the gateway data items include: the size of each piece of flow data, the request time corresponding to each piece of flow data, and a software interface corresponding to the request corresponding to each piece of flow data;
the gateway data acquisition module analyzes the first log content once every other first unit time;
the gateway data acquisition module records the size of each piece of flow data corresponding to the first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a flow value data set A, and records the value corresponding to the nth element in the flow value data set A as An;
the gateway data acquisition module records request time corresponding to each piece of flow data corresponding to first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a request time set B, and records a value corresponding to the nth element in the request time set B as Bn;
the gateway data acquisition module records the software interfaces corresponding to the requests corresponding to each flow data of each flow data corresponding to the first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a software interface set C, and records the value corresponding to the nth element in the software interface set C as Cn;
the number of elements corresponding to the flow value data set A, the request time set B and the software interface set C is equal to the number of flow data corresponding to the first unit time in the first log content analyzed by the gateway data acquisition module;
the gateway data acquisition module also monitors the running state of each software interface in real time so as to obtain the running state time curves of each software interface, each running state time curve represents the change condition of the running state of the corresponding interface software along with time, the running state comprises an opening state and a closing state,
and the value of the running state time curve corresponding to the opening state is marked as 1, and the value of the running state time curve corresponding to the closing state is marked as 0.
In the gateway data, each request sent by each software interface corresponds to one piece of flow data, and each piece of flow data is independent, so that the gateway data is analyzed, the analyzed data needs to be refined, and the condition of each piece of flow data corresponding to each request of each software interface is locked, and the size of each piece of flow data, the request time corresponding to each piece of flow data, and the software interface corresponding to each request of each piece of flow data are further obtained; the gateway data acquisition module is arranged to analyze the first log content once every first unit time, so as to ensure the frequency of analyzing the gateway data and lock the range corresponding to the analyzed data (the traffic data condition corresponding to the first unit time in the first log content analyzed last time) each time the gateway data is analyzed; obtaining a flow value data set A, a request time set B and a software interface set C, wherein the flow value data set A, the request time set B and the software interface set C are used for uniformly storing and managing the acquired data and simultaneously quickly obtaining a corresponding associated data set during data analysis; obtaining an operation state time curve of each software interface so as to obtain the relationship between the operation state of each software interface and time, and further quickly counting the total operation time corresponding to the specified software interface in the first unit time; the values of the operating state time curve corresponding to the operating state are set to be 1 and 0, so as to clearly and intuitively reflect the operating state corresponding to the specified time of the specified software interface (1 represents an open state, and 0 represents a closed state).
The method for judging whether the relevance exists between the elements in different sets by the gateway data association item integration module comprises the following steps:
s1.1, acquiring a flow value data set A, a request time set B and a software interface set C corresponding to the last analysis of a first log content in a gateway data acquisition module;
s1.2, extracting a value An1 corresponding to the n1 th element in the A, extracting a value Bn2 corresponding to the n2 th element in the B and extracting a value Cn3 corresponding to the n3 th element in the C;
s1.3, comparing the size relationship among n1, n2 and n3,
when the values of the elements in the sets corresponding to the n1 and the n2, the n1 and the n3, or the n2 and the n3 are equal, the element values in the sets corresponding to the equal two are judged to have relevance,
when the conditions of equality do not exist between n1 and n2, between n1 and n3, or between n2 and n3, judging that the element values in the sets corresponding to n1, n2 and n3 do not have relevance;
the method for obtaining each associated data group corresponding to the gateway data associated item by the gateway data associated item integration module comprises the following steps:
s2.1, acquiring a flow value data set A, a request time set B and a software interface set C corresponding to the latest analysis of a first log content in a gateway data acquisition module;
s2.2, extracting a value Ai corresponding to the ith element in the A;
s2.3, extracting a value Bi1 corresponding to the element with the relevance to Ai in the B and extracting a value Ci2 corresponding to the element with the relevance to Ai in the C;
and S2.4, obtaining the ith associated data group Zi corresponding to the latest analysis of the first log content in the gateway data acquisition module, wherein Zi is [ Ai, Bi1, Ci2 ].
When the gateway associated data intelligent analysis module clusters each associated data group, the number of the categories is the same as the number of the software interface categories corresponding to each element in the software interface set C, and each associated data group with the same corresponding software interface is divided into the same category,
recording the m-th element in the k-th associated data group in the j class
Figure GDA0003693575180000141
J is more than or equal to 0 and less than or equal to x, and x is the number of the software interface types corresponding to each element in the software interface set C.
The gateway associated data intelligent analysis module divides associated data groups with the same software interface into the same class, and is used for analyzing the flow use condition corresponding to each software interface in the gateway subsequently, predicting the flow use condition corresponding to each software interface, and summarizing and accumulating predicted values corresponding to each software interface so as to predict gateway data.
The method for obtaining the first change rate corresponding to each category by the gateway associated data intelligent analysis module comprises the following steps:
s3.1, obtaining the sum of the corresponding values of the 1 st element in each associated data group in the jth class to obtain the total flow of the software interface corresponding to the jth class corresponding to the first unit time before the first log content analyzed last time
Figure GDA0003693575180000142
The described
Figure GDA0003693575180000143
Wherein,
Figure GDA0003693575180000144
representing the value corresponding to the 1 st element in the kth associated data group in the jth class, and k1j representing the total number of the associated data groups in the jth class;
s3.2, acquiring total flow corresponding to the first unit time in the first log content analyzed by the software interface corresponding to the jth class at the first k2 times
Figure GDA0003693575180000151
S3.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring total flow corresponding to previous first unit time in first k2 times of analyzed first log content by a software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure GDA0003693575180000152
K2 is more than or equal to 1 and less than or equal to k3, p is more than or equal to 0 and less than or equal to p1, k3 is a first preset value, and p1 is a second preset value;
s3.4, acquiring corresponding total flow when k2 is different values in the previous p days
Figure GDA0003693575180000153
Maximum value of (1), is noted
Figure GDA0003693575180000154
S3.5, when p is judged to be different values, respectively corresponding
Figure GDA0003693575180000155
Whether or not it is meaningful to have the information,
when in use
Figure GDA0003693575180000156
Then determine
Figure GDA0003693575180000157
Meaningless, and
Figure GDA0003693575180000158
wherein g represents a pair
Figure GDA0003693575180000159
The normalized processing equation of (a) is,
when in use
Figure GDA00036935751800001510
Time, determine
Figure GDA00036935751800001511
Is significant in that
Figure GDA00036935751800001512
Figure GDA00036935751800001513
S3.6, obtaining a first change rate corresponding to the jth category
Figure GDA00036935751800001514
The above-mentioned
Figure GDA00036935751800001515
Wherein,
Figure GDA00036935751800001516
is composed of
Figure GDA00036935751800001517
The coefficient of adjustment of (a) is,
Figure GDA00036935751800001518
both alpha 1 and beta 1 are constant and
Figure GDA00036935751800001519
in this embodiment, if the gateway has only one software interface, i.e., j equals 1, and k3 equals 2, p1 equals 2,
and is
Figure GDA00036935751800001520
And α 1 ═ 1.1, β 1 ═ 2,
since 600 > 500, then
Figure GDA00036935751800001521
Since 0 < 400, then
Figure GDA00036935751800001522
Since 0 equals 0, then
Figure GDA00036935751800001523
Because of
Figure GDA00036935751800001524
Then
Figure GDA00036935751800001525
Because of
Figure GDA0003693575180000161
Meaningless, then
Figure GDA0003693575180000162
Then the
Figure GDA0003693575180000163
Figure GDA0003693575180000164
A first rate of change of the software interface
Figure GDA0003693575180000165
In the process of obtaining the first change rate corresponding to each category by the gateway associated data intelligent analysis module, the total flow corresponding to the first unit time before the software interface corresponding to the jth category in the first log content analyzed last time is obtained first
Figure GDA0003693575180000166
The total flow corresponding to each software interface in the first unit time before in the first log content analyzed each time by each type of corresponding software interface is taken as a data analysis unit, and the total flow corresponding to each software interface is further analyzed from two angles of history P, k2, so that a first change rate corresponding to each type is further obtained; obtaining
Figure GDA0003693575180000167
The reason is that k2 in the previous p days are different values, so that the previous p days in the historical data correspond to a plurality of total flows, during data prediction, the worst case of the possibly occurring events needs to be considered, and the prediction result can achieve the purpose of early warning, and the worst case of the previous p days in the historical data corresponding to a plurality of total flows is each k2 bit different value
Figure GDA0003693575180000168
Maximum value of
Figure GDA0003693575180000169
Computing
Figure GDA00036935751800001610
Is to obtain the change rate of the total flow in the first unit time(ii) a Is provided with a pair
Figure GDA00036935751800001611
Is to avoid the settlement result
Figure GDA00036935751800001612
A meaningless situation occurs, so that a prediction result is meaningless, and the final prediction result of the gateway data is influenced; when the first change rate is calculated, p is set to p1-1 because p has a value range of 0 ≦ p1, and p +1 occurs in the calculation process, and the upper limit of p at this time can be derived to be p1-1 by 0 ≦ p +1 ≦ p 1; setting when calculating the first rate of change
Figure GDA00036935751800001613
Coefficient of regulation of
Figure GDA00036935751800001614
Is due to
Figure GDA00036935751800001615
The method includes the steps that (1) an increase change rate of total flow of software interfaces corresponding to a jth class corresponding to a jth day in the previous p day relative to a previous p +1 day is obtained, the first change rate is obtained relative to the increase change rate of the total flow of software interfaces corresponding to the jth class corresponding to the current time, and the flow use condition of the software interfaces has large fluctuation, so that certain deviation exists between the corresponding increase change rate in historical data and a first change rate needing to be obtained, and corresponding adjusting coefficients are set for the corresponding increase change rate in the historical data so as to calibrate each corresponding increase change rate in the historical data ((the method includes the steps of (1) obtaining the total flow of the software interfaces corresponding to the jth class corresponding to the current time, (b) obtaining the total flow of the software interfaces according to the first change rate, and setting corresponding adjusting coefficients for the corresponding increase change rate in the historical data so as to calibrate each corresponding increase change rate in the historical data: (b)
Figure GDA00036935751800001616
Corresponding adjustment coefficient is
Figure GDA00036935751800001617
The calibrated growth rate of change is
Figure GDA00036935751800001618
Figure GDA00036935751800001619
Corresponding to the rate of change of growth from historical data
Figure GDA00036935751800001620
Obtaining the increase change rate of the total flow of the software interfaces corresponding to the jth class corresponding to the current time), and obtaining a first change rate corresponding to the jth class by means of averaging according to the increase change rate of the total flow of the software interfaces corresponding to the jth classes after calibration, wherein the first change rates obtained by the method are obtained by referring to the current time, and relatively speaking, the method is more accurate, and further the prediction result of gateway flow data is more accurate.
The method for obtaining the second change rate corresponding to each category by the gateway associated data intelligent analysis module comprises the following steps:
s4.1, obtaining the total time length corresponding to the condition that the median of the operating state time curve of the software interface corresponding to the jth class is 1, and obtaining the total operating time length corresponding to the first unit time before the software interface corresponding to the jth class in the first log content analyzed last time
Figure GDA0003693575180000171
S4.2, acquiring the total operation time length corresponding to the first unit time before the software interface corresponding to the jth class in the first log content analyzed at the first k2 times
Figure GDA0003693575180000172
S4.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring the total operation time length corresponding to the first unit time in the first k 2-time analyzed first log content of the software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure GDA0003693575180000173
S4.4, acquiring corresponding total operation time when k2 is different in the previous p days
Figure GDA0003693575180000174
Maximum value of (1), is recorded as
Figure GDA0003693575180000175
S4.5, when p is judged to be different values, respectively corresponding
Figure GDA0003693575180000176
Whether or not it is meaningful to have the information,
when in use
Figure GDA0003693575180000177
Then, it is determined
Figure GDA0003693575180000178
Not meaningfully, and
Figure GDA0003693575180000179
Figure GDA00036935751800001710
wherein g1 represents a pair
Figure GDA00036935751800001711
The normalized processing equation of (a) is,
when in use
Figure GDA00036935751800001712
Time, determine
Figure GDA00036935751800001713
Is significant in that
Figure GDA00036935751800001714
Figure GDA00036935751800001715
S4.6, obtaining a second change rate corresponding to the jth category
Figure GDA00036935751800001716
The described
Figure GDA00036935751800001717
Wherein,
Figure GDA0003693575180000181
is composed of
Figure GDA0003693575180000182
The coefficient of adjustment of (a) is,
Figure GDA0003693575180000183
both alpha 2 and beta 2 are constant and
Figure GDA0003693575180000184
in this embodiment, if the gateway has only one software interface, i.e., j equals 1, and k3 equals 2, p1 equals 2,
and is provided with
Figure GDA0003693575180000185
And α 1 ═ 1.01, β 1 ═ 10,
since 2400 > 2100, then
Figure GDA0003693575180000186
Since 0 equals 0, then
Figure GDA0003693575180000187
Since 2000 > 0, then
Figure GDA0003693575180000188
Because of
Figure GDA0003693575180000189
Meaningless, then
Figure GDA00036935751800001810
Because of
Figure GDA00036935751800001811
Then the
Figure GDA00036935751800001812
Figure GDA00036935751800001813
Figure GDA00036935751800001814
A second rate of change of the software interface
Figure GDA00036935751800001815
In the process of obtaining the second change rate corresponding to each category by the gateway associated data intelligent analysis module, the total operation time corresponding to the first unit time before the software interface corresponding to the jth category in the first log content analyzed last time is obtained first
Figure GDA00036935751800001816
The total operating time length corresponding to the first unit time in the first log content analyzed each time by the software interface corresponding to each type is taken as a data analysis unit, and the total operating time length corresponding to each software interface is further analyzed from the two aspects of history P, k2, so that a second change rate corresponding to each type is obtained; obtaining
Figure GDA00036935751800001817
The reason is that since k2 is different in the previous p days, the previous p days in the historical data correspond to a plurality of total operating durations, and during data prediction, the worst case of possible events needs to be considered, and then the predicted result is obtainedIf the current time is not more than k2, the early warning can be achieved, and the worst condition of the total operation time corresponding to the previous p days in the historical data is each time when the k2 bit is different
Figure GDA00036935751800001818
Maximum value of
Figure GDA00036935751800001819
Computing
Figure GDA00036935751800001820
The change rate of the total operation time length in the first unit time is obtained; is provided with a pair
Figure GDA00036935751800001821
For avoiding settlement results, equation g1
Figure GDA00036935751800001822
A meaningless situation occurs, so that a prediction result is meaningless, and the final prediction result of gateway data is influenced; setting an adjustment factor when calculating the first rate of change
Figure GDA0003693575180000191
Acting on the adjustment coefficient
Figure GDA0003693575180000192
Has the same effect of adjusting the calibration (
Figure GDA0003693575180000193
For is to
Figure GDA0003693575180000194
Adjustment calibration is performed).
The method for obtaining the third change rate corresponding to each category by the gateway associated data intelligent analysis module comprises the following steps:
s5.1, acquiring request time corresponding to the 2 nd element in each associated data group in the j category, calculating the time difference between two adjacent request times,
let tv be the time difference between the request time corresponding to the v +1 th correlated data set in the jth class and the request time corresponding to the v th correlated data set in the jth class,
acquiring a flow value corresponding to the 1 st element in the v-th associated data group in the j type and recording the flow value as
Figure GDA0003693575180000195
When v is calculated to be different values respectively,
Figure GDA0003693575180000196
quotient to tv
Figure GDA0003693575180000197
Further, the flow consumption value per unit time corresponding to the software interface corresponding to the j-th class in the first log content analyzed last time in the previous first unit time is obtained
Figure GDA0003693575180000198
The above-mentioned
Figure GDA0003693575180000199
Wherein k1j represents the total number of associated data groups in the j-th class;
s5.2, acquiring total flow corresponding to the first unit time in the first log content analyzed by the software interface corresponding to the jth class at the first k2 times
Figure GDA00036935751800001910
S5.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring total flow corresponding to previous first unit time in first k2 times of analyzed first log content by a software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure GDA00036935751800001911
K3 is more than or equal to 1 and less than or equal to k2, p is more than or equal to 0 and less than or equal to p1, and k3 is a first preset valueValue p1 is a second preset value;
s5.4, obtaining corresponding total flow when k2 is different in the previous p days
Figure GDA00036935751800001912
Maximum value of (1), is recorded as
Figure GDA00036935751800001913
S5.5, when p is judged to be different values, respectively corresponding
Figure GDA00036935751800001914
Whether or not it makes sense to determine whether,
when the temperature is higher than the set temperature
Figure GDA00036935751800001915
Then determine
Figure GDA00036935751800001916
Not meaningfully, and
Figure GDA00036935751800001917
Figure GDA00036935751800001918
wherein g2 represents a group
Figure GDA00036935751800001919
The normalized process equation of (a) is,
when the temperature is higher than the set temperature
Figure GDA00036935751800001920
Time, judge
Figure GDA00036935751800001921
Is significant in that
Figure GDA00036935751800001922
Figure GDA0003693575180000201
S5.6, obtaining a third change rate corresponding to the jth category
Figure GDA0003693575180000202
The above-mentioned
Figure GDA0003693575180000203
Wherein,
Figure GDA0003693575180000204
is composed of
Figure GDA0003693575180000205
The coefficient of adjustment of (a) is,
Figure GDA0003693575180000206
both alpha 3 and beta 3 are constant values and
Figure GDA0003693575180000207
the method for predicting the gateway traffic data by the gateway traffic data prediction module comprises the following steps:
s6.1, obtaining a first predicted value W1 of the gateway flow data in the first unit time before in the first log content analyzed next time based on the current time,
when j is 0, the W1 is 0,
when j ≠ 0, the
Figure GDA0003693575180000208
S6.2, obtaining a second predicted value W2 of the gateway flow data in the previous first unit time in the first log content analyzed next time based on the current time,
when j is 0, the W2 is 0,
when j ≠ 0, the
Figure GDA0003693575180000209
And S6.3, obtaining a final predicted value W of the gateway traffic data in the first unit time in the first log content analyzed next time based on the current time, wherein W is { W1, W2} max.
In this embodiment, if the gateway has only one software interface, i.e., j equals 1, and k3 equals 2, p1 equals 2,
and is
Figure GDA00036935751800002010
And a first rate of change of the software interface
Figure GDA00036935751800002011
0.275, second rate of change of software interface
Figure GDA00036935751800002013
A third rate of change of the software interface of-0.48
Figure GDA00036935751800002012
Is 0.3;
then the first predicted value W1 ═ 1+0.3 ═ 0.5 ═ 1-0.48 ═ 2400 ═ 811.2;
the second predicted value W2 ═ (1+0.275) × 600 ═ 765;
because 811.2 is greater than 765, the material,
therefore, the final predicted value W is { W1, W2} max is 811.2.
When the gateway flow data prediction module predicts the gateway flow data, two prediction modes are adopted, and two prediction results are screened to obtain a final prediction value of the gateway flow data; when the first predicted value W1 is obtained, the prediction is performed by the second change rate and the third change rate, and the prediction is analyzed from two points of view of the increase of the service time of the software interface and the increase of the flow consumption value of the software interface per unit time; when the second predicted value W2 is obtained, the prediction is performed by the first change rate, and the second predicted value W2 is analyzed from the viewpoint of the increase of the total flow rate used by the software interface.
An intelligent gateway monitoring and management method based on artificial intelligence, the method comprises the following steps:
s1, asynchronously collecting gateway data through a gateway data collection module, writing a collection result into a first log, analyzing the content of the first log, and respectively extracting various data of the gateway data to obtain a corresponding set of the various data of the gateway data;
s2, analyzing sets corresponding to various data of gateway data through a gateway data association item integration module, judging whether the elements in different sets have association, and integrating through the set elements with association to respectively obtain each association data set corresponding to the gateway data association item;
s3, clustering and analyzing each associated data group through a gateway associated data intelligent analysis module to obtain a first change rate, a second change rate and a third change rate corresponding to each category;
s4, predicting gateway traffic data by combining each associated data set corresponding to the gateway data association item at the current time and the associated data set in the historical data through a gateway traffic data prediction module;
s5, in the early warning module, the gateway traffic data prediction module compares the prediction result of the gateway traffic data with the threshold value,
when the prediction result is more than or equal to the threshold value, the early warning module gives an alarm to the user,
and when the prediction result is smaller than the threshold value, the early warning module does not give an alarm to the user.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that changes may be made in the embodiments and/or equivalents thereof without departing from the spirit and scope of the invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (5)

1. An intelligent gateway monitoring and management system based on artificial intelligence is characterized by comprising:
the gateway data acquisition module is used for asynchronously acquiring gateway data, writing an acquisition result into a first log, analyzing the content of the first log, and respectively extracting various data of the gateway data to obtain a corresponding set of the various data of the gateway data;
the gateway data association item integration module analyzes corresponding sets of gateway data items, judges whether the elements in different sets have association or not, and integrates the set elements with the association to respectively obtain each association data set corresponding to the gateway data association item;
the gateway associated data intelligent analysis module is used for clustering and analyzing each associated data group to obtain a first change rate, a second change rate and a third change rate corresponding to each category, the first change rate represents the increase change rate of the total flow in first unit time, the second change rate represents the change rate of the total operation duration in first unit time, and the third change rate represents the change rate of the flow consumption value in corresponding unit time in first unit time;
the gateway traffic data prediction module predicts gateway traffic data by combining each associated data set corresponding to the gateway data association item at the current time and the associated data set in the historical data;
the early warning module compares the prediction result of the gateway flow data from the gateway flow data prediction module with a threshold value,
when the prediction result is more than or equal to the threshold value, the early warning module gives an alarm to the user,
when the prediction result is smaller than the threshold value, the early warning module does not give an alarm to the user;
when gateway traffic data are predicted, a first predicted value W1 is obtained through a second change rate and a third change rate, a second predicted value W2 is obtained through the first change rate, and a final predicted value W of the gateway traffic data is obtained, wherein the final predicted value W is { W1, W2} max, and the { W1, W2} max represents the maximum value of W1 and W2;
the gateway data acquisition module asynchronously acquires flow data, contents corresponding to different flow data are mutually independent, one flow data corresponds to one request, one request corresponds to one software interface, and one software interface can correspond to multiple requests;
the gateway data items include: the size of each piece of flow data, the request time corresponding to each piece of flow data, and a software interface corresponding to the request corresponding to each piece of flow data;
the gateway data acquisition module analyzes the first log content once every other first unit time;
the gateway data acquisition module records the size of each piece of flow data corresponding to the first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a flow value data set A, and records the value corresponding to the nth element in the flow value data set A as An;
the gateway data acquisition module records request time corresponding to each piece of flow data corresponding to first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a request time set B, and records a value corresponding to the nth element in the request time set B as Bn;
the gateway data acquisition module records the software interfaces corresponding to the requests corresponding to each piece of flow data corresponding to the first unit time in the analyzed first log content into a blank set one by one according to the sequence of the analyzed flow data to obtain a software interface set C, and records the value corresponding to the nth element in the software interface set C as Cn;
the number of elements corresponding to the flow value data set A, the request time set B and the software interface set C is equal to the number of flow data corresponding to the first unit time in the first log content analyzed by the gateway data acquisition module;
the gateway data acquisition module also monitors the running state of each software interface in real time so as to obtain the running state time curve of each software interface, each running state time curve represents the change of the running state of the corresponding interface software along with time, the running state comprises an opening state and a closing state,
the value of the running state time curve corresponding to the opening state is marked as 1, and the value of the running state time curve corresponding to the closing state is marked as 0;
the method for judging whether the relevance exists between the elements in different sets by the gateway data association item integration module comprises the following steps:
s1.1, acquiring a flow value data set A, a request time set B and a software interface set C corresponding to the last analysis of a first log content in a gateway data acquisition module;
s1.2, extracting a value An1 corresponding to the n1 th element in the A, extracting a value Bn2 corresponding to the n2 th element in the B and extracting a value Cn3 corresponding to the n3 th element in the C;
s1.3, comparing the size relation among n1, n2 and n3,
when the values of the elements in the sets corresponding to the n1 and the n2, the n1 and the n3, or the n2 and the n3 are equal, the element values in the sets corresponding to the equal two are judged to have relevance,
when the values of the elements in the sets corresponding to n1, n2 and n3 are not equal to each other in the ranges of n1 and n2, n1 and n3 or n2 and n3, judging that no relevance exists among the values of the elements in the sets corresponding to n1, n2 and n 3;
the method for obtaining each associated data group corresponding to the gateway data associated item by the gateway data associated item integration module comprises the following steps:
s2.1, acquiring a flow value data set A, a request time set B and a software interface set C corresponding to the last analysis of the first log content in a gateway data acquisition module;
s2.2, extracting a value Ai corresponding to the ith element in the A;
s2.3, extracting a value Bi1 corresponding to the element with the relevance to Ai in B and extracting a value Ci2 corresponding to the element with the relevance to Ai in C;
s2.4, obtaining an ith associated data group Zi corresponding to the latest analysis first log content in the gateway data acquisition module, wherein Zi is [ Ai, Bi1, Ci2 ];
when the gateway associated data intelligent analysis module clusters each associated data group, the number of the categories is the same as the number of the software interface categories corresponding to each element in the software interface set C, and each associated data group with the same corresponding software interface is divided into the same category,
recording the m-th element in the k-th associated data group in the j-th class as
Figure FDA0003693575170000031
J is more than or equal to 0 and less than or equal to x, and x is the number of the software interface types corresponding to each element in the software interface set C;
the method for obtaining the first change rate corresponding to each category by the gateway associated data intelligent analysis module comprises the following steps:
s3.1, obtaining the sum of the corresponding values of the 1 st element in each associated data group in the j category to obtain the total flow of the software interface corresponding to the j category corresponding to the first unit time before the first log content analyzed last time
Figure FDA0003693575170000032
The above-mentioned
Figure FDA0003693575170000033
Wherein,
Figure FDA0003693575170000034
representing the value corresponding to the 1 st element in the kth associated data group in the jth class, wherein k1j represents the total number of the associated data groups in the jth class;
s3.2, acquiring total flow corresponding to the first unit time in the first log content analyzed by the software interface corresponding to the jth class at the first k2 times
Figure FDA0003693575170000035
S3.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring total flow corresponding to previous first unit time in first k2 times of analyzed first log content by a software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure FDA0003693575170000036
K3 is more than or equal to 1 and less than or equal to k2, p is more than or equal to 0 and less than or equal to p1, k3 is a first preset value, and p1 is a second preset value;
s3.4, acquiring corresponding total flow when k2 is different values in the previous p days
Figure FDA0003693575170000037
Maximum value of (1), is recorded as
Figure FDA0003693575170000038
S3.5, when p is judged to be different values, respectively corresponding
Figure FDA0003693575170000041
Whether or not it is meaningful to have the information,
when in use
Figure FDA0003693575170000042
Then, it is determined
Figure FDA0003693575170000043
Meaningless, and
Figure FDA0003693575170000044
wherein g represents a pair
Figure FDA0003693575170000045
The normalized processing equation of (a) is,
when in use
Figure FDA0003693575170000046
Time, judge
Figure FDA0003693575170000047
Is significant in that
Figure FDA0003693575170000048
Figure FDA0003693575170000049
S3.6, obtaining a first change rate corresponding to the jth category
Figure FDA00036935751700000410
The above-mentioned
Figure FDA00036935751700000411
Wherein,
Figure FDA00036935751700000412
is composed of
Figure FDA00036935751700000413
The coefficient of adjustment of (a) is,
Figure FDA00036935751700000414
both alpha 1 and beta 1 are constant values and
Figure FDA00036935751700000415
2. the system according to claim 1, wherein said system comprises: the method for obtaining the second change rate corresponding to each category by the gateway associated data intelligent analysis module comprises the following steps:
s4.1, obtaining the total time length corresponding to the condition that the median of the operating state time curve of the software interface corresponding to the jth class is 1, and obtaining the total operating time length corresponding to the first unit time before the software interface corresponding to the jth class in the first log content analyzed last time
Figure FDA00036935751700000416
S4.2, acquiring the total operation time length corresponding to the first unit time before the software interface corresponding to the jth class in the first log content analyzed at the first k2 times
Figure FDA00036935751700000417
S4.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring the total operating duration corresponding to the previous first unit time in the first k 2-time analyzed first log content of the software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure FDA00036935751700000418
S4.4, acquiring corresponding total operation time when k2 is different in the previous p days
Figure FDA00036935751700000419
Maximum value of (1), is noted
Figure FDA00036935751700000420
S4.5, when p is judged to be different values, respectively corresponding
Figure FDA00036935751700000421
Whether or not it is meaningful to have the information,
when in use
Figure FDA00036935751700000422
Then determine
Figure FDA00036935751700000423
Not meaningfully, and
Figure FDA00036935751700000424
Figure FDA0003693575170000051
wherein g1 represents a pair
Figure FDA0003693575170000052
The normalized processing equation of (a) is,
when in use
Figure FDA0003693575170000053
Time, judge
Figure FDA0003693575170000054
Is significant in that
Figure FDA0003693575170000055
Figure FDA0003693575170000056
S4.6, obtaining a second change rate corresponding to the jth category
Figure FDA0003693575170000057
The described
Figure FDA0003693575170000058
Wherein,
Figure FDA0003693575170000059
is composed of
Figure FDA00036935751700000510
The coefficient of adjustment of (a) is,
Figure FDA00036935751700000511
both alpha 2 and beta 2 are constant values and
Figure FDA00036935751700000512
3. the system according to claim 2, wherein said system comprises: the method for obtaining the third change rate corresponding to each category by the gateway associated data intelligent analysis module comprises the following steps:
s5.1, acquiring the request time corresponding to the 2 nd element in each associated data group in the j class, calculating the time difference between two adjacent request times,
let tv be the time difference between the request time corresponding to the v +1 th correlated data set in the jth class and the request time corresponding to the v th correlated data set in the jth class,
acquiring a flow value corresponding to the 1 st element in the v-th associated data group in the j-th class and recording the flow value as
Figure FDA00036935751700000513
When v is calculated to be different values respectively,
Figure FDA00036935751700000514
quotient of tv
Figure FDA00036935751700000515
Further obtaining the first analyzed latest time of the software interface corresponding to the jth classThe consumption value of the flow in the unit time corresponding to the first unit time in the log content
Figure FDA00036935751700000516
The described
Figure FDA00036935751700000517
Wherein k1j represents the total number of the associated data groups in the j-th class;
s5.2, acquiring total flow corresponding to the first unit time in the first k2 times of first log content analyzed by the software interface corresponding to the jth class
Figure FDA00036935751700000518
S5.3, taking the time point which is the same as the current time in the previous p days as a reference point,
acquiring total flow corresponding to previous first unit time in first k2 times of analyzed first log content by a software interface corresponding to the jth class corresponding to the reference point in the previous p days
Figure FDA00036935751700000519
K3 is more than or equal to 1 and less than or equal to k2, p is more than or equal to 0 and less than or equal to p1, k3 is a first preset value, and p1 is a second preset value;
s5.4, obtaining corresponding total flow when k2 is different in the previous p days
Figure FDA0003693575170000061
Maximum value of (1), is noted
Figure FDA0003693575170000062
S5.5, when p is judged to be different values, respectively corresponding
Figure FDA0003693575170000063
Whether or not it makes sense to determine whether,
when the temperature is higher than the set temperature
Figure FDA0003693575170000064
Then, it is determined
Figure FDA0003693575170000065
Meaningless, and
Figure FDA0003693575170000066
Figure FDA0003693575170000067
wherein g2 represents a group
Figure FDA0003693575170000068
The normalized processing equation of (a) is,
when the temperature is higher than the set temperature
Figure FDA0003693575170000069
Time, judge
Figure FDA00036935751700000610
Is significant in that
Figure FDA00036935751700000611
Figure FDA00036935751700000612
S5.6, obtaining a third change rate corresponding to the jth category
Figure FDA00036935751700000613
The above-mentioned
Figure FDA00036935751700000614
Wherein,
Figure FDA00036935751700000615
is composed of
Figure FDA00036935751700000616
The coefficient of adjustment of (a) is,
Figure FDA00036935751700000617
both alpha 3 and beta 3 are constant values and
Figure FDA00036935751700000618
4. the system according to claim 3, wherein the system comprises: the method for predicting the gateway traffic data by the gateway traffic data prediction module comprises the following steps:
s6.1, obtaining a first predicted value W1 of the gateway flow data in the first unit time before in the first log content analyzed next time based on the current time,
when j is 0, said W1 is 0,
when j ≠ 0, the
Figure FDA00036935751700000619
S6.2, obtaining a second predicted value W2 of the gateway flow data in the previous first unit time in the first log content analyzed next time based on the current time,
when j is 0, the W2 is 0,
when j ≠ 0, the
Figure FDA00036935751700000620
And S6.3, obtaining a final predicted value W of the gateway traffic data in the first unit time in the first log content analyzed next time based on the current time, wherein the W is { W1, W2} max.
5. The intelligent gateway monitoring and managing method based on artificial intelligence of the intelligent gateway monitoring and managing system based on artificial intelligence of any one of the application claims 1-4, characterized in that: the method comprises the following steps:
s1, asynchronously collecting gateway data through a gateway data collection module, writing the collection result into a first log, analyzing the content of the first log, and respectively extracting various data of the gateway data to obtain a corresponding set of the various data of the gateway data;
s2, analyzing sets corresponding to various data of gateway data through a gateway data association item integration module, judging whether the elements in different sets have association, and integrating through the set elements with association to respectively obtain each association data set corresponding to the gateway data association item;
s3, clustering and analyzing each associated data group through a gateway associated data intelligent analysis module to obtain a first change rate, a second change rate and a third change rate corresponding to each category;
s4, predicting gateway traffic data by combining each associated data group corresponding to the gateway data association item at the current time and associated data groups in historical data through a gateway traffic data prediction module;
s5, in the early warning module, the gateway traffic data prediction module compares the prediction result of the gateway traffic data with the threshold value,
when the prediction result is more than or equal to the threshold value, the early warning module gives an alarm to the user,
and when the prediction result is smaller than the threshold value, the early warning module does not give an alarm to the user.
CN202210440949.8A 2022-04-26 2022-04-26 Intelligent gateway monitoring management system and method based on artificial intelligence Active CN114553733B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210440949.8A CN114553733B (en) 2022-04-26 2022-04-26 Intelligent gateway monitoring management system and method based on artificial intelligence

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210440949.8A CN114553733B (en) 2022-04-26 2022-04-26 Intelligent gateway monitoring management system and method based on artificial intelligence

Publications (2)

Publication Number Publication Date
CN114553733A CN114553733A (en) 2022-05-27
CN114553733B true CN114553733B (en) 2022-07-26

Family

ID=81667733

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210440949.8A Active CN114553733B (en) 2022-04-26 2022-04-26 Intelligent gateway monitoring management system and method based on artificial intelligence

Country Status (1)

Country Link
CN (1) CN114553733B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN118487868B (en) * 2024-07-11 2024-10-01 云御科技(珠海)有限公司 Traffic safety analysis method, device, platform and storage medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10320644B1 (en) * 2015-09-14 2019-06-11 Amazon Technologies, Inc. Traffic analyzer for isolated virtual networks
CN110086649A (en) * 2019-03-19 2019-08-02 深圳壹账通智能科技有限公司 Detection method, device, computer equipment and the storage medium of abnormal flow
CN113591950A (en) * 2021-07-19 2021-11-02 中国海洋大学 Random forest network traffic classification method, system and storage medium

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111212337B (en) * 2019-12-27 2022-02-25 中国联合网络通信集团有限公司 Port classification method and device
CN112949739A (en) * 2021-03-17 2021-06-11 中国电子科技集团公司第二十九研究所 Information transmission scheduling method and system based on intelligent traffic classification

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10320644B1 (en) * 2015-09-14 2019-06-11 Amazon Technologies, Inc. Traffic analyzer for isolated virtual networks
CN110086649A (en) * 2019-03-19 2019-08-02 深圳壹账通智能科技有限公司 Detection method, device, computer equipment and the storage medium of abnormal flow
CN113591950A (en) * 2021-07-19 2021-11-02 中国海洋大学 Random forest network traffic classification method, system and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
互联网流量分类中流量特征研究;刘珍等;《计算机应用研究》;20170131;第34卷(第01期);第8-13页 *

Also Published As

Publication number Publication date
CN114553733A (en) 2022-05-27

Similar Documents

Publication Publication Date Title
CN111356148B (en) Method and related equipment for realizing network optimization
Tulone et al. An energy-efficient querying framework in sensor networks for detecting node similarities
CN108415789B (en) Node fault prediction system and method for large-scale hybrid heterogeneous storage system
WO2021217865A1 (en) Method and apparatus for locating root cause of alarm, computer device, and storage medium
CN114553733B (en) Intelligent gateway monitoring management system and method based on artificial intelligence
CN113159615A (en) Intelligent information security risk measuring system and method for industrial control system
CN113159374B (en) Data-driven urban traffic flow rate mode identification and real-time prediction early warning method
CN112363896A (en) Log anomaly detection system
WO2024066331A1 (en) Network abnormality detection method and apparatus, electronic device, and storage medium
CN118378199A (en) Real-time anomaly detection method in big data analysis platform
CN110490486A (en) A kind of enterprise&#39;s big data management system
US8471712B2 (en) Detecting abnormal time intervals
CN111984514A (en) Prophet-bLSTM-DTW-based log anomaly detection method
CN117828371B (en) Intelligent analysis method for business information of comprehensive operation and maintenance platform
CN117667585B (en) Operation and maintenance efficiency evaluation method and system based on operation and maintenance quality management database
US20230409422A1 (en) Systems and Methods for Anomaly Detection in Multi-Modal Data Streams
Wang et al. A state monitoring method of gas regulator station based on evidence theory driven by time-domain information
CN114819685A (en) Intelligent management system and method for attendance information of public security personnel
CN118426318B (en) Internet of things service self-adaptive system based on process mining
CN117689070B (en) Ammeter management system based on internet of things equipment
CN114647386B (en) Big data distributed storage method based on artificial intelligence
CN118503894B (en) Lithium battery quality detection system based on process index data analysis
CN118014288B (en) Building design flow dynamic management method based on BIM
CN115063259B (en) Heating system sensitivity analysis method and system based on trusted load interval
CN112330003B (en) Periodic capacity data prediction method, system and storage medium based on bidirectional cyclic neural network

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant