CN111865635B - Method and device for determining out-of-limit time of ring network capacity - Google Patents
Method and device for determining out-of-limit time of ring network capacity Download PDFInfo
- Publication number
- CN111865635B CN111865635B CN201910355934.XA CN201910355934A CN111865635B CN 111865635 B CN111865635 B CN 111865635B CN 201910355934 A CN201910355934 A CN 201910355934A CN 111865635 B CN111865635 B CN 111865635B
- Authority
- CN
- China
- Prior art keywords
- network
- peak value
- flow peak
- unit time
- 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
Links
Images
Classifications
-
- 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
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/42—Loop networks
-
- 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
-
- 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
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
-
- 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
- H04L43/08—Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
- H04L43/0876—Network utilisation, e.g. volume of load or congestion level
- H04L43/0882—Utilisation of link capacity
Abstract
The embodiment of the invention discloses a method and a device for determining the out-of-limit time of looped network capacity, wherein the method comprises the steps of carrying out cluster training on the peak value change coefficient of looped network flow in unit time in a preset historical time period based on two classifications to obtain a flow mutation screening model, wherein the two classifications comprise a normal value and a mutation value; correcting a network looped network flow peak value per unit time in a historical investigation period based on a flow mutation screening model; determining a unit time increase coefficient of the network ring network flow peak value based on the corrected unit time network ring network flow peak value in the historical investigation period; and determining a time point when the network ring network flow peak value reaches a preset capacity threshold value based on the current network ring network flow peak value in unit time and the unit time increase coefficient of the network ring network flow peak value. The embodiment of the invention solves the problem that the capacity expansion time cannot be accurately obtained in the prior art.
Description
Technical Field
The embodiment of the invention relates to the technical field of communication, in particular to a method and a device for determining the out-of-limit time of the capacity of a ring network.
Background
At present, the prediction of the capacity of the transmission PTN looped network is generally determined based on manual check, so the prediction result is inaccurate and time-consuming, meanwhile, the existing partial network management prediction scheme cannot accurately inform a specific capacity expansion time point, and only can inform the current occupation condition of the looped network, so that the specific date that the capacity of the looped network is full of 100% cannot be predicted, the capacity expansion of the transmission PTN network is lagged, the capacity expansion is generally arranged only when the load of the looped network is very high, and the capacity expansion cannot be prepared in advance, thereby bringing adverse effects to the development of the 4G mobile internet service.
Disclosure of Invention
The embodiment of the invention provides a method and a device for determining out-of-limit time of capacity of a ring network, and aims to solve the problem that the capacity-expanding time cannot be accurately determined in the prior art.
In order to solve the technical problem, the invention is realized as follows:
in a first aspect, a method for determining a ring network capacity out-of-limit time is provided, including:
performing cluster training on the network looped network flow peak value change coefficient in unit time in a preset historical time period based on two classifications to obtain a flow mutation screening model, wherein the two classifications comprise a normal value and a mutation value;
based on the flow mutation screening model, modifying a mutation value in a network looped network flow peak value per unit time in a historical investigation period;
determining a unit time increase coefficient of the network ring network flow peak value based on the corrected unit time network ring network flow peak value in the historical investigation period;
and determining a time point when the network ring network flow peak value reaches a preset capacity threshold value based on the current network ring network flow peak value in unit time and the unit time increase coefficient of the network ring network flow peak value.
In a second aspect, a device for determining a ring network capacity violation time is provided, including:
the model obtaining unit is used for carrying out clustering training on the network looped network flow peak value change coefficient in unit time in a preset historical time period based on two classifications to obtain a flow mutation screening model, wherein the two classifications comprise a normal value and a mutation value;
the flow peak value correcting unit is used for correcting the sudden change value in the flow peak value of the network ring network per unit time in the historical investigation period based on the flow sudden change screening model;
the growth coefficient determining unit is used for determining the growth coefficient of the network ring network flow peak value in unit time based on the corrected network ring network flow peak value in the historical investigation period;
and the time determining unit is used for determining a time point when the network ring network flow peak value reaches a preset capacity threshold value based on the network ring network flow peak value of the current unit time and the unit time increase coefficient of the network ring network flow peak value.
In a third aspect, a terminal device is also provided, which comprises a processor, a memory and a computer program stored on the memory and executable on the processor, the computer program implementing the steps of the method according to the first aspect when executed by the processor.
In a fourth aspect, there is also provided a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the method according to the first aspect.
In the embodiment of the invention, the method for determining the ring network capacity threshold crossing time corrects the flow peak value of the network ring network at a single time in the historical investigation period through a flow mutation screening model obtained by performing clustering training on the flow peak value change coefficient of the network ring network at the single time in the preset historical time period according to two categories, determines the unit time increase coefficient of the flow peak value of the ring network according to the flow peak value of the network ring network at the single time in the corrected historical investigation period, and determines the time point when the flow peak value of the ring network reaches the preset capacity threshold value according to the flow peak value of the network ring network at the current unit time and the unit time increase coefficient of the flow peak value of the ring network. Therefore, the unit time increase coefficient of the looped network flow peak value determined by the looped network flow peak value per unit time in the corrected historical investigation period can be obtained according to the flow mutation screening model, so that the time point when the looped network flow peak value reaches the preset capacity threshold value can be accurately determined, the problem that the capacity expansion time cannot be accurately obtained in the prior art can be solved, and the problems that time consumption is caused and accuracy is low due to the fact that the looped network capacity is predicted in a manual checking mode in the prior art can be solved.
Drawings
Fig. 1 is a schematic flow chart of a method for determining a ring network capacity violation time according to one embodiment of the present invention;
fig. 2 is a schematic flow chart of a method for determining a ring network capacity violation time according to another embodiment of the present invention;
FIG. 3 is a schematic flow chart diagram of a method for determining the ring network capacity violation time according to another embodiment of the present invention;
fig. 4 is a schematic flow chart of a method for determining a ring network capacity violation time according to another embodiment of the present invention;
fig. 5 is a schematic flow chart of a method for determining a ring network capacity violation time according to a specific embodiment of the present invention;
FIG. 6 is a schematic diagram of a method for determining a ring network capacity violation time according to one embodiment of the present invention;
fig. 7 is a schematic block diagram of a device for determining the ring network capacity out-of-limit time according to an embodiment of 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 some, but not all, embodiments of the present invention. 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.
The technical solutions provided by the embodiments of the present invention are described in detail below with reference to the accompanying drawings.
Fig. 1 is a schematic flowchart of a method for determining an out-of-limit time of a ring network capacity according to an embodiment of the present invention, so as to solve a problem that an expansion time cannot be accurately obtained in the prior art. The method can comprise the following steps:
and 102, performing cluster training on the network looped network flow peak value change coefficient in unit time in a preset historical time period based on two classifications to obtain a flow mutation screening model, wherein the two classifications comprise a normal value and a mutation value. The peak value change coefficient is used to characterize the peak value change amplitude, for example, the peak value change coefficient is determined based on the flow peak value change conditions of two days before and after the current day.
Specifically, PTN network looped network flow data of the last two years can be collected, daily peak value change conditions of the last two years are calculated according to the network looped network flow data, then clustering training is carried out through an artificial intelligence K-means algorithm, and a flow mutation screening model of the current looped network is trained. When the PTN ring network flow number of the last two years is collected, the unit time may correspond to each hour, even each minute, and the like, and the operation may be specifically performed according to specific operation requirements, which is not limited to the manner described in the embodiment of the present invention.
And 104, correcting the mutation value in the network ring network flow peak value per unit time in the historical investigation period based on the flow mutation screening model.
As shown in fig. 2, the operation of correcting the mutation value in the network ring network traffic peak per unit time in the historical investigation period may include:
and 202, determining a sudden change value in the network looped network flow peak value per unit time in the historical investigation period based on the flow sudden change threshold value output by the flow sudden change screening model.
And 204, deleting the mutation value to correct the network ring network flow peak value per unit time in the historical investigation period.
After the flow mutation screening model is obtained, a mutation value in the network looped network flow peak value in unit time in the historical investigation period can be determined and deleted according to the flow mutation threshold value output by the flow mutation screening model, so that the network looped network flow peak value in unit time in the historical investigation period can be corrected, that is, data which cannot be referred to in the network looped network flow peak value in unit time in the historical investigation period can be removed through the flow mutation screening model, a relatively smooth flow peak value growth curve can be obtained, and therefore the unit time growth coefficient of the network looped network flow peak value can be determined in the subsequent steps.
And 106, determining the unit time increase coefficient of the network ring network flow peak value based on the corrected unit time network ring network flow peak value in the historical investigation period. Wherein, unit time may correspond to each day, then the daily average peak flow growth coefficient formula is:
wherein t represents the current day, t-1 represents the previous day, N t Indicates the peak flow rate on day t, N t-1 The peak flow rate on the t-1 th day, Δ N represents the average daily growth factor (i.e., the growth factor per unit time).
the value range of t may indicate a historical investigation period, for example, the current day is advanced by 20 days, and the maximum value does not exceed 30 days. The number of days for acquiring the flow peak of each PTN looped network on the OMC network management system can be the number of days defined by the historical investigation period.
It can be understood that the flow peak value of the network looped network collected in the historical investigation period is led into the artificial intelligence module, the flow mutation screening model is started to perform data smoothing so as to remove the mutation value in the collected looped network flow peak value, the smoothed data is summed, and the ratio of the summed result and the time number (such as 20) of the investigation period after the mutation value is removed is the unit time increase coefficient delta N of the PTN looped network in the historical investigation period. The unit time increase coefficient Δ N in the historical investigation period may be obtained by a method of evaluating the smoothed data by weighted average or mean square error, and the specific implementation process is not set forth herein.
Therefore, the unit time increase coefficient delta N of all PTN looped networks to be inspected in the historical inspection period can be obtained according to the formula (1), so that the unit time bandwidth historical increase rate of each looped network in each area can be calculated, and the unit time increase rate obtained according to the unit time historical increase rate in the subsequent steps and the final bandwidth utilization rate which can be achieved by the looped networks can be obtained.
And 108, determining a time point when the network ring network flow peak value reaches a preset capacity threshold value based on the network ring network flow peak value of the current unit time and the unit time increase coefficient of the network ring network flow peak value.
As shown in fig. 3, determining a time point when the network ring network traffic peak value reaches the preset capacity threshold value based on the current network ring network traffic peak value in unit time and the unit time growth coefficient of the network ring network traffic peak value includes:
and 302, determining the network ring network flow peak value in unit time based on the network ring network flow peak value in the current unit time, the unit time increase coefficient of the network ring network flow peak value and the holiday flow peak value compensation coefficient.
The formula for calculating the network ring network flow peak value in unit time is as follows:
A t1 ＝N t1-1 +ΔN+ΔM (2)
wherein, t 1 Indicating a certain day in the historical investigation period, A t1 Indicating the t-th time in the history investigation period 1 Peak daily flow, N t1-1 Indicating the t-th time in the history investigation period 1 The flow peak value of 1 day, Δ N is a unit time increase coefficient in a historical investigation period, and Δ M is a holiday compensation coefficient.
Specifically, the Δ N of the PTN ring network calculated by the formula (1) is substituted into the formula (2), and the traffic peak value a on the first day in the future is calculated first 1 . Then, the calculated A is calculated 1 Substituting equation (2) again to replace N with a value t1-1 Term, calculate A 2 . And the rest can be analogized until all A in the historical investigation period are calculated to form an array N 0 ＝{A 1 ，A 2 ，A 3 ，......A t1 }. In the calculation of A t1 In the process of (1)The system judges whether the future investigation date covers the holiday, and if the future investigation date covers the holiday, the system extracts the flow increase coefficient corresponding to the same holiday of the PTN looped network history for compensation.
Therefore, the flow peak value of a certain PTN looped network in a future investigation period (namely the investigation period after the current moment) is calculated by substituting the result obtained by the formula (1) into the formula (2) and is displayed in the system, then the specific days that each looped network exceeds a certain utilization rate are counted, and the information that the operation and maintenance personnel screen out the high-frequency occupied looped network is provided, so that the important attention and analysis can be performed.
And 304, determining a time point when the network ring network flow peak value reaches a preset capacity threshold value based on the network ring network flow peak value in unit time.
The calculation formula of the time point when the ring network flow peak value reaches the preset capacity threshold value in the future investigation period is shown as a formula (3):
T＝max{N 0 }≥W×70％ (3)
wherein, T represents an overrun day (an expansion completion day, that is, a time point when a preset capacity threshold is reached), w represents a static capacity, the static capacity of the PTN ring network to be investigated can be obtained through an interface of the ring network in the network management system, and Wx70% can be represented as an early warning value of the capacity upper limit.
By obtaining N 0 A in (1) 1 ，A 2 \8230themaximum value in the equation 8230that is obtained by the system is substituted into the equation (3) to judge max { N } 0 Whether the formula condition is satisfied is judged to be W multiplied by 70%.
And if the flow peak value which enables the formula (3) to be established is obtained, taking out the ring network, and determining the time corresponding to the flow peak value as the time point reaching the preset capacity threshold value. Of course, if the traffic peak value that makes equation (3) true is not obtained, it indicates that the PTN ring network does not exceed the capacity expansion threshold (preset capacity threshold) in the future investigation period. After all the ring networks are subjected to the logic, the PTN ring network needing capacity expansion is formed, the capacity expansion date and the overrun countdown calculated from the current date are presented to operation and maintenance personnel, the capacity expansion date and the capacity expansion countdown can be conveniently and visually observed, and therefore the operation and maintenance personnel can conveniently and timely make a scientific capacity expansion strategy.
In the embodiment of the invention, the method for determining the ring network capacity out-of-limit time corrects the flow peak value of the ring network at a single time in a historical investigation period through a flow mutation screening model obtained by performing cluster training on the flow peak value change coefficient of the ring network at the single time in a preset historical time period according to two categories, determines the unit time increase coefficient of the flow peak value of the ring network according to the flow peak value of the ring network at the single time in the corrected historical investigation period, and determines the time point when the flow peak value of the ring network reaches the preset capacity threshold value according to the flow peak value of the ring network at the current unit time and the unit time increase coefficient of the flow peak value of the ring network. Therefore, the unit time increase coefficient of the looped network flow peak value can be determined according to the unit time network flow peak value in the corrected historical investigation period obtained by the flow mutation screening model, so that the time point when the looped network flow peak value reaches the preset capacity threshold value can be accurately determined, the problem that the capacity expansion time cannot be accurately obtained in the prior art can be solved, and the problems that time consumption is caused and the accuracy is low due to the fact that the looped network capacity is predicted in a manual checking mode in the prior art can be solved.
In the above further embodiment, as shown in fig. 4, the method for determining the ring network capacity out-of-limit time further includes:
and 406, correcting the time point when the network ring network flow peak value reaches the preset capacity threshold value based on the current network ring network flow peak value in unit time and the unit time increase coefficient of the corrected network ring network flow peak value.
Wherein the operation of adjusting the flow mutation screening model may comprise:
if the time point when the current unit time network ring network flow peak value reaches the preset capacity threshold value is earlier than the time point when the network ring network flow peak value reaches the preset capacity threshold value, increasing the flow sudden change threshold value output by the flow sudden change screening model;
and if the time point when the current unit time network ring network flow peak value reaches the preset capacity threshold value is later than the time point when the network ring network flow peak value reaches the preset capacity threshold value, reducing the flow sudden change threshold value output by the flow sudden change screening model.
It can be understood that, after the system determines the capacity expansion date of a certain PTN ring network (i.e. the time point when the flow peak value of the network ring network reaches the preset capacity threshold), since the system collects the occupation status of the ring network in unit time and judges the occupation status, when it is determined that 100% of the flow peak value of the network ring network reaches the preset capacity threshold, the time point when the flow peak value of the current unit time network ring network reaches the preset capacity threshold is compared with the capacity expansion date determined by the system, and if the compared difference value exceeds the preset capacity expansion time deviation threshold, at this time, the artificial intelligent learning module may be triggered to adjust the flow sudden change screening model to change the flow sudden change threshold output by the flow sudden change screening model, thereby re-determining and deleting the sudden change value in the flow peak value of the network ring network in the historical investigation period according to the flow sudden change threshold which is re-output, and modifying the unit time increase coefficient of the flow peak value of the ring network, so that the time point when the flow peak value of the modified ring network reaches the preset capacity threshold is closer to the actual capacity expansion date.
When the sudden change value in the network ring network flow peak value per unit time in the historical investigation period is re-determined according to the flow sudden change threshold value output after the flow sudden change screening model is adjusted, the previously determined sudden change value is probably not the sudden change value any more when the newly determined sudden change value is re-determined after the flow sudden change screening model is adjusted, and therefore the previously determined sudden change value needs to be re-determined.
In any of the above embodiments, the method for determining the ring network capacity out-of-limit time further includes: and displaying the time point when the flow peak value of the ring network reaches the preset capacity threshold value to a user. That is, after the method described in any of the above embodiments is performed on all ring networks, the PTN ring network that needs capacity expansion is formed, the capacity expansion date and the overrun countdown calculated from the current day are presented to the operation and maintenance staff, and the capacity expansion date and the capacity expansion countdown can be visually observed, so that the operation and maintenance staff can make a scientific capacity expansion strategy in time.
With reference to fig. 5 and fig. 6, the implementation process of the method for determining the ring network capacity out-of-limit time according to the embodiment of the present invention may be:
firstly, PTN looped network flow data in the last two years are collected, a daily flow peak value change coefficient is calculated, then clustering training is carried out through a K-means algorithm in an artificial intelligence module, and a flow mutation screening model is trained.
Secondly, in a historical investigation period, collecting a daily flow peak value of the PTN looped network, and then removing a sudden change value in the daily flow peak value of the looped network in the historical investigation period through a K flow sudden change screening model.
Thirdly, after the mutation value is screened out, the daily growth coefficient of the looped network flow peak value in the historical investigation period is calculated.
Fourthly, calculating the daily looped network flow peak value according to the daily increase coefficient of the looped network flow peak value in the historical investigation period and the holiday compensation coefficient.
Fifthly, according to 70% of the ring network occupancy rate investigation threshold w (red and yellow cards can be set to be listed for early warning, 45-70% of the ring network occupancy rate investigation threshold w is yellow cards, and more than 70% of the ring network occupancy rate investigation threshold w is red cards) which is manually set, when the calculated ring network flow peak value of the target day reaches 70% of the ring network occupancy rate investigation threshold w, the target day is the capacity expansion date.
And sixthly, when the difference value between the expansion date obtained by comparison calculation and the expansion date determined by the actual looped network proportion exceeds a preset expansion time deviation threshold value, correcting the flow mutation screening model through the self-learning adjusting module to change the output flow mutation threshold value, and returning to the first step to continue execution to correct the unit time increase coefficient of the looped network flow peak value, so that the time point when the corrected looped network flow peak value reaches the preset capacity threshold value is closer to the actual expansion date.
As shown in fig. 6, the data processed by the artificial intelligence function module is sent to the database for storage, and is called by algorithms 1, 2, and 3, and the ring network to be expanded is calculated after calling, and is stored in the storage table of the super-tolerance ring network details, and the data integration and calling presentation are performed by the front-end web interface.
Therefore, the method provided by the embodiment of the invention carries out clustering training on the collected historical data by adopting a K-means algorithm, establishes the flow mutation screening model, screens out the data of the flow mutation value, namely the non-referential condition through the flow mutation screening model, calculates the full load date of the PTN looped network, and displays the full load date through the front-end BS framework, so that a PTN network manager can prepare in advance and plan in advance.
And the K-means algorithm has artificial intelligence learning capability, after a full load date is displayed, the actual flow occupation condition of the PTN looped network acquired by the system is compared and calculated, when the current day is full, the full load date of the PTN looped network and the current full load date are compared, when a difference threshold value is exceeded, the machine self-learning is started to perform interval dynamic adjustment, a flow mutation screening model is changed, the flow mutation threshold value output by the model is changed, the daily growth coefficient is automatically changed, the full load date calculated for the PTN looped network is closer to the actual condition through continuous dynamic adjustment, the problem that the capacity expansion time cannot be accurately obtained in the prior art is solved, and the problem that time consumption and accuracy are low due to the fact that the looped network capacity is predicted in a manual checking mode in the prior art can be solved.
An embodiment of the present invention further provides a device for determining a ring network capacity violation time, as shown in fig. 7, which may include: a model obtaining unit 702, configured to perform cluster training on a network ring network traffic peak value change coefficient in a predetermined historical time period based on two classifications, so as to obtain a traffic sudden change screening model, where the two classifications include a normal value and a sudden change value; the flow peak value correcting unit 704 is used for correcting the flow peak value of the network ring network per unit time in the historical investigation period based on the flow mutation screening model; a growth coefficient determining unit 706, configured to determine a unit time growth coefficient of a network ring network traffic peak value based on the corrected network ring network traffic peak value per unit time in the history investigation period; the time determining unit 708 is configured to determine, based on the current network ring network traffic peak value in unit time and a unit time growth coefficient of the network ring network traffic peak value, a time point at which the network ring network traffic peak value reaches a preset capacity threshold.
The device for determining the ring network capacity out-of-limit time in the embodiment of the invention corrects the flow peak value of the ring network at a unit time in a historical investigation period through a flow sudden change screening model obtained by performing cluster training on the flow peak value change coefficient of the ring network at the unit time in a preset historical time period based on two classifications by a flow peak value correcting unit 704 according to a model obtaining unit 702, determines the unit time increase coefficient of the flow peak value of the ring network at the unit time in the historical investigation period through an increase coefficient determining unit 706 according to the flow peak value of the ring network at the unit time in the corrected historical investigation period, and determines the time point when the flow peak value of the ring network reaches the preset capacity threshold value by a time determining unit 708 according to the flow peak value of the ring network at the current unit time and the unit time increase coefficient of the flow peak value of the ring network. Therefore, the unit time increase coefficient of the looped network flow peak value can be determined according to the unit time network flow peak value in the corrected historical investigation period obtained by the flow mutation screening model, so that the time point when the looped network flow peak value reaches the preset capacity threshold value can be accurately determined, the problem that the capacity expansion time cannot be accurately obtained in the prior art can be solved, and the problems that time consumption is caused and the accuracy is low due to the fact that the looped network capacity is predicted in a manual checking mode in the prior art can be solved.
In the above embodiment, the apparatus for determining the ring network capacity out-of-limit time further includes: a mutation value determination unit 710, configured to determine a mutation value in a network ring network traffic peak value per unit time in a historical investigation period based on a traffic mutation threshold output by the traffic mutation screening model; the traffic peak correction unit 704 is configured to delete the mutation value to correct the network ring network traffic peak per unit time in the historical investigation period.
After the flow mutation screening model is obtained, a mutation value in the network looped network flow peak value in unit time in the historical investigation period can be determined and deleted according to the output flow mutation threshold value, so that the network looped network flow peak value in unit time in the historical investigation period can be corrected, that is, data which cannot be referred to in the network looped network flow peak value in unit time in the historical investigation period can be removed through the flow mutation screening model, a relatively smooth flow peak value growth curve can be obtained, and therefore the unit time growth coefficient of the network looped network flow peak value can be determined in the subsequent steps.
In the above further embodiment, further comprising: an adjusting unit 712, configured to adjust the sudden flow rate screening model when a target difference value exceeds a preset capacity expansion time deviation threshold value, where the target difference value is a difference value between a time point when a network ring network flow peak value reaches a preset capacity threshold value and a time point when the network ring network flow peak value of a current unit time reaches the preset capacity threshold value; the flow peak value correcting unit 704 is configured to revise the flow peak value of the single time network ring network in the historical investigation period based on the adjusted flow mutation screening model, so as to revise the unit time increase coefficient of the flow peak value of the network ring network based on the revised flow peak value of the single time network ring network in the historical investigation period; and a time correction unit 714, configured to correct a time point at which the network ring network traffic peak value reaches the preset capacity threshold, based on the network ring network traffic peak value in the current unit time and the unit time growth coefficient of the corrected network ring network traffic peak value.
Wherein the adjusting unit 712 is further configured to: if the time point when the current unit time network ring network flow peak value reaches the preset capacity threshold value is earlier than the time point when the network ring network flow peak value reaches the preset capacity threshold value, increasing the flow sudden change threshold value output by the flow sudden change screening model; and if the time point when the current unit time network ring network flow peak value reaches the preset capacity threshold value is later than the time point when the network ring network flow peak value reaches the preset capacity threshold value, reducing the flow sudden change threshold value output by the flow sudden change screening model.
It can be understood that, after the system determines the expansion date of a certain PTN ring network (i.e. the time point when the flow peak value of the network ring network reaches the preset capacity threshold), since the system collects the occupation status of the ring network in unit time and judges the occupation status, when it is determined that 100% of the flow peak value of the network ring network reaches the preset capacity threshold, the time point when the current flow peak value of the ring network in unit time reaches the preset capacity threshold is compared with the expansion date determined by the system, and if the compared difference value exceeds the preset expansion time deviation threshold, at this time, the artificial intelligent learning module may be triggered to adjust the flow sudden change screening model to change the flow sudden change threshold output by the flow sudden change screening model, thereby re-determining and deleting the sudden change value in the flow peak value of the network ring network in unit time in the historical investigation period according to the flow sudden change threshold which is re-output, and modifying the unit time growth coefficient of the flow peak value of the ring network, so that the time point when the flow peak value of the modified network ring network reaches the preset capacity threshold is closer to the actual expansion date.
In any of the above embodiments, the determining apparatus may further include a traffic peak determining unit 716, configured to determine a network ring network traffic peak per unit time based on the current network ring network traffic peak per unit time, a unit time growth coefficient of the network ring network traffic peak, and a holiday traffic peak compensation coefficient; the time determination unit 708 is configured to determine, based on the network ring network traffic peak value in unit time, a time point at which the network ring network traffic peak value reaches the preset capacity threshold. A display unit 718 may also be included for displaying a time point when the peak value of the network ring network traffic reaches a preset capacity threshold value to a user.
That is to say, the network ring network flow peak value in unit time is calculated according to the network ring network flow peak value in current unit time and the unit time increase coefficient of the network ring network flow peak value, the expansion date is determined according to the corresponding time when the calculated network ring network flow peak value reaches the preset capacity threshold value, after all ring networks are subjected to the logic, the PTN ring network needing expansion is formed, the expansion date and the overrun countdown calculated from the current date are displayed to operation and maintenance personnel, the expansion date and the expansion countdown are conveniently and visually observed, and therefore the operation and maintenance personnel can conveniently and timely make a scientific expansion strategy.
An embodiment of the present invention further provides a terminal device, which may include a processor, a memory, and a computer program stored in the memory and capable of running on the processor, where the computer program, when executed by the processor, implements each process of the method embodiments shown in fig. 1 to 5, and can achieve the same technical effect, and details are not repeated here to avoid repetition.
An embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements the processes of the methods shown in fig. 1 to 5, and can achieve the same technical effects, and in order to avoid repetition, the computer program is not described herein again. The computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
It should also be noted that 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. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The above description is only an example of the present invention, and is not intended to limit the present invention. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (7)
1. A method for determining the out-of-limit time of the capacity of a ring network comprises the following steps:
performing cluster training on the network looped network flow peak value change coefficient in unit time in a preset historical time period based on two classifications to obtain a flow mutation screening model, wherein the two classifications comprise a normal value and a mutation value;
correcting a sudden change value in a network ring network flow peak value in unit time in a historical investigation period based on the flow sudden change screening model;
determining a unit time increase coefficient of the network ring network flow peak value based on the corrected unit time network ring network flow peak value in the historical investigation period;
determining a time point when the network ring network flow peak value reaches a preset capacity threshold value based on the network ring network flow peak value in the current unit time and a unit time increase coefficient of the network ring network flow peak value;
the method for correcting the mutation value of the network looped network flow peak value in unit time in the historical investigation period comprises the following steps:
determining a sudden change value in a unit time network looped network flow peak value in a historical investigation period based on a flow sudden change threshold value output by the flow sudden change screening model;
deleting the mutation value to correct the network ring network flow peak value in unit time in the historical investigation period;
based on the current unit time network ring network flow peak value and the unit time increase coefficient of the ring network flow peak value, the time point when the network ring network flow peak value reaches the preset capacity threshold value is determined, which comprises the following steps:
determining a network looped network flow peak value in unit time based on a network looped network flow peak value in current unit time, a unit time increase coefficient of the looped network flow peak value and a holiday flow peak value compensation coefficient;
and determining a time point when the network ring network flow peak value reaches a preset capacity threshold value based on the network ring network flow peak value in unit time.
2. The method of claim 1, further comprising:
when the target difference value exceeds a preset capacity expansion time deviation threshold value, adjusting the flow mutation screening model, wherein the target difference value is the difference value between the time point when the network ring network flow peak value reaches the preset capacity threshold value and the time point when the current unit time network ring network flow peak value reaches the preset capacity threshold value; the time point when the current unit time looped network flow peak value reaches the preset capacity threshold value is the time point when the occupation condition of the looped network reaches 100% in unit time;
based on the adjusted flow mutation screening model, revising the unit time network looped network flow peak value in the historical investigation period, and revising the unit time increase coefficient of the network looped network flow peak value based on the revised unit time network looped network flow peak value in the historical investigation period;
and correcting the time point when the network ring network flow peak value reaches the preset capacity threshold value based on the current network ring network flow peak value in unit time and the unit time increase coefficient of the corrected network ring network flow peak value.
3. The method of claim 2, wherein adjusting the abrupt flow elimination model comprises:
if the time point when the current unit time network ring network flow peak value reaches the preset capacity threshold value is earlier than the time point when the current unit time network ring network flow peak value reaches the preset capacity threshold value, increasing the flow sudden change threshold value output by the flow sudden change screening model;
and if the time point when the current unit time network ring network flow peak value reaches the preset capacity threshold value is later than the time point when the network ring network flow peak value reaches the preset capacity threshold value, reducing the flow sudden change threshold value output by the flow sudden change screening model.
4. The method of claim 1, further comprising:
and displaying the time point when the flow peak value of the network ring network reaches the preset capacity threshold value to a user.
5. A device for determining a capacity violation time of a ring network, comprising:
the device comprises a model obtaining unit, a flow mutation screening module and a flow mutation screening module, wherein the model obtaining unit is used for carrying out clustering training on a unit time network looped network flow peak value change coefficient in a preset historical time period based on two classifications to obtain a flow mutation screening model, and the two classifications comprise a normal value and a mutation value;
the flow peak value correcting unit is used for correcting a sudden change value in a unit time network looped network flow peak value in a historical investigation period based on the flow sudden change screening model;
the growth coefficient determining unit is used for determining a unit time growth coefficient of the network ring network flow peak value based on the corrected unit time network ring network flow peak value in the historical investigation period;
the time determination unit is used for determining a time point when the network ring network flow peak value reaches a preset capacity threshold value based on the network ring network flow peak value of the current unit time and the unit time increase coefficient of the network ring network flow peak value;
further comprising:
the sudden change value determining unit is used for determining a sudden change value in a network looped network flow peak value in unit time in a historical investigation period based on the flow sudden change threshold value output by the flow sudden change screening model;
the flow peak value correction unit is used for deleting the mutation value so as to correct the network ring network flow peak value in unit time in the historical investigation period;
the determining device further comprises a flow peak value determining unit, wherein the flow peak value determining unit is used for determining the network ring network flow peak value in unit time based on the network ring network flow peak value in the current unit time, the unit time increase coefficient of the network ring network flow peak value and the holiday flow peak value compensation coefficient;
and the time determining unit is used for determining the time point when the network ring network flow peak value reaches the preset capacity threshold value based on the network ring network flow peak value in unit time.
6. The apparatus of claim 5, further comprising:
the adjusting unit is used for adjusting the flow mutation screening model when a target difference value exceeds a preset capacity expansion time deviation threshold value, wherein the target difference value is a difference value between a time point when a network ring network flow peak value reaches a preset capacity threshold value and a time point when the network ring network flow peak value of the current unit time reaches the preset capacity threshold value; the time point when the current network looped network flow peak value reaches the preset capacity threshold value is the time point when the occupation condition of the looped network in unit time reaches 100%;
the flow peak value correction unit is used for revising the flow peak value of the network ring network in unit time in the historical investigation period based on the adjusted flow mutation screening model so as to revise the unit time increase coefficient of the flow peak value of the network ring network based on the revised flow peak value of the network ring network in unit time in the historical investigation period;
and the time correction unit is used for correcting the time point when the flow peak value of the network ring network reaches the preset capacity threshold value based on the current flow peak value of the network ring network in unit time and the unit time growth coefficient of the corrected flow peak value of the network ring network.
7. A computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium, which computer program, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910355934.XA CN111865635B (en) | 2019-04-29 | 2019-04-29 | Method and device for determining out-of-limit time of ring network capacity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910355934.XA CN111865635B (en) | 2019-04-29 | 2019-04-29 | Method and device for determining out-of-limit time of ring network capacity |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111865635A CN111865635A (en) | 2020-10-30 |
| CN111865635B true CN111865635B (en) | 2022-11-22 |
Family
ID=72966151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910355934.XA Active CN111865635B (en) | 2019-04-29 | 2019-04-29 | Method and device for determining out-of-limit time of ring network capacity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111865635B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113839835B (en) * | 2021-09-27 | 2023-09-26 | 长沙理工大学 | Top-k flow accurate monitoring system based on small flow filtration |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103200114A (en) * | 2013-04-18 | 2013-07-10 | 湖北邮电规划设计有限公司 | Metropolitan area network planning method |
| CN105580407A (en) * | 2014-12-05 | 2016-05-11 | 华为技术有限公司 | Network resource deployment method and device |
| CN105681063A (en) * | 2014-11-18 | 2016-06-15 | 中国移动通信集团北京有限公司 | Method and apparatus for monitoring network index |
| WO2017084723A1 (en) * | 2015-11-20 | 2017-05-26 | Telefonaktiebolaget Lm Ericsson (Publ) | Traffic steering between radio access network nodes |
| CN107231264A (en) * | 2017-07-25 | 2017-10-03 | 北京百度网讯科技有限公司 | For the method and apparatus for the capacity for managing Cloud Server |
| CN109302719A (en) * | 2018-11-05 | 2019-02-01 | 北京市天元网络技术股份有限公司 | A kind of LTE cell capacity prediction analysis method and device |
| CN109600757A (en) * | 2018-11-29 | 2019-04-09 | 南京亚信软件有限公司 | Prediction technique, device, computer equipment and the storage medium of base station dilatation |
-
2019
- 2019-04-29 CN CN201910355934.XA patent/CN111865635B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103200114A (en) * | 2013-04-18 | 2013-07-10 | 湖北邮电规划设计有限公司 | Metropolitan area network planning method |
| CN105681063A (en) * | 2014-11-18 | 2016-06-15 | 中国移动通信集团北京有限公司 | Method and apparatus for monitoring network index |
| CN105580407A (en) * | 2014-12-05 | 2016-05-11 | 华为技术有限公司 | Network resource deployment method and device |
| WO2017084723A1 (en) * | 2015-11-20 | 2017-05-26 | Telefonaktiebolaget Lm Ericsson (Publ) | Traffic steering between radio access network nodes |
| CN107231264A (en) * | 2017-07-25 | 2017-10-03 | 北京百度网讯科技有限公司 | For the method and apparatus for the capacity for managing Cloud Server |
| CN109302719A (en) * | 2018-11-05 | 2019-02-01 | 北京市天元网络技术股份有限公司 | A kind of LTE cell capacity prediction analysis method and device |
| CN109600757A (en) * | 2018-11-29 | 2019-04-09 | 南京亚信软件有限公司 | Prediction technique, device, computer equipment and the storage medium of base station dilatation |
Non-Patent Citations (3)
| Title |
|---|
| "Forecasting short-term data center network traffic load with convolutional neural networks";Alberto Mozo，Bruno Ordozgoiti，Sandra Gomez-Canaval;《hrrps://jouranls.plos.org/plosone/article?id=10.1371/jounrnal.pone.0191939》;20180206;全文 * |
| "Logistic回归模型及其应用";常振海，刘薇;《延边大学学报(自然科学版)》;20120320;全文 * |
| "网络规划阶段RNC及载波扩容预测方法研究";袁栋;《移动通信》;20150630;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111865635A (en) | 2020-10-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101267362B (en) | A dynamic identification method and its device for normal fluctuation range of performance normal value | |
| CN106126391A (en) | System monitoring method and apparatus | |
| WO2021129086A1 (en) | Traffic prediction method, device, and storage medium | |
| CN110995477A (en) | Early warning processing method, device and equipment based on dynamic threshold and storage medium | |
| CN105989441A (en) | Model parameter adjustment method and device | |
| CN111294812B (en) | Resource capacity-expansion planning method and system | |
| US20110085649A1 (en) | Fluctuation Monitoring Method that Based on the Mid-Layer Data | |
| CN106708692A (en) | Method and device for building filtering alarm model, filtering alarm method and device and electronic equipment | |
| CN111865635B (en) | Method and device for determining out-of-limit time of ring network capacity | |
| CN105721194A (en) | Intelligent positioning system of faults and hidden dangers of mobile network | |
| CN106682159A (en) | Threshold configuration method | |
| CN112787878B (en) | Network index prediction method and electronic equipment | |
| CN113055923B (en) | Mobile network traffic prediction method, device and equipment | |
| CN111901134B (en) | Method and device for predicting network quality based on recurrent neural network model (RNN) | |
| CN111881961A (en) | Power distribution network fault risk grade prediction method based on data mining | |
| CN111143070A (en) | Resource scheduling method and device | |
| CN109299975B (en) | Object characteristic parameter determination method and device, electronic equipment and readable storage medium | |
| CN110135627B (en) | Water resource optimization method and device | |
| CN116545867A (en) | Method and device for monitoring abnormal performance index of network element of communication network | |
| CN108521435B (en) | Method and system for user network behavior portrayal | |
| CN116912016A (en) | Bill auditing method and device | |
| CN112559737B (en) | Node classification method and system of knowledge graph | |
| CN111368859A (en) | Complaint early warning processing method and device | |
| CN107316056B (en) | Automatic evaluation system and automatic evaluation method for network security level | |
| CN113723692A (en) | Data processing method, apparatus, device, medium, and program product |
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 |