CN109309594A - Method, apparatus, equipment and the storage medium of communication equipment power failure analysis - Google Patents
Method, apparatus, equipment and the storage medium of communication equipment power failure analysis Download PDFInfo
- Publication number
- CN109309594A CN109309594A CN201811422339.5A CN201811422339A CN109309594A CN 109309594 A CN109309594 A CN 109309594A CN 201811422339 A CN201811422339 A CN 201811422339A CN 109309594 A CN109309594 A CN 109309594A
- Authority
- CN
- China
- Prior art keywords
- power failure
- information
- model
- communication equipment
- cleaning
- 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.)
- Granted
Links
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
- H04L41/06—Management of faults, events, alarms or notifications
- H04L41/0604—Management of faults, events, alarms or notifications using filtering, e.g. reduction of information by using priority, element types, position or time
-
- 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
- H04L41/0631—Management of faults, events, alarms or notifications using root cause analysis; using analysis of correlation between notifications, alarms or events based on decision criteria, e.g. hierarchy, tree or time analysis
-
- 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
- H04L41/069—Management of faults, events, alarms or notifications using logs of notifications; Post-processing of notifications
Abstract
The present invention provides method, apparatus, equipment and the storage mediums of a kind of analysis of communication equipment power failure.This method comprises: obtaining the log key message after communication equipment fault restores；Log key message is cleaned, acquisition move back take class activity warning information and the corresponding cleaning of cleared alarm information after data set；Extract the characteristic information after cleaning in data set；It is analyzed and processed using characteristic information of the fault analysis model after training to data set after cleaning, takes warning information and power failure probability to obtain to move back.It can be by actively obtaining log key message, and pass through the power failure probability that fault analysis model quickly analyzes communication equipment, improve the analysis efficiency of power failure.
Description
Technical field
The present embodiments relate to the sides that technical field of communication equipment more particularly to a kind of communication equipment power failure are analyzed
Method, device, equipment and storage medium.
Background technique
For the power issue of wireless telecom equipment, the prior art is based primarily upon power & environment supervision system and is analyzed, and shows
Power failure can not be analyzed by power & environment supervision by having 60% communication equipment in net wireless telecom equipment, main still according to resistance in people
Work mode analyzes the result after troubleshooting, and data accuracy is lower in power failure probability analysis, and can not will divide
Analysis result is applied to power failure diagnosing in time.
The fault log of wireless telecom equipment belongs to passive query there is only equipment rear end, and engineer is needed to send out in failure
After life, beaching accommodation side is inquired by command mode, and power failure analysis efficiency is low.
Summary of the invention
The embodiment of the present invention provides method, apparatus, equipment and the storage medium of a kind of communication equipment power failure analysis, solution
The low technical problem of power failure analysis efficiency in communication equipment power failure analysis method in the prior art of having determined.
In a first aspect, the embodiment of the present invention provides a kind of method of communication equipment power failure analysis, comprising:
Obtain the log key message after communication equipment fault restores；
The log key message is cleaned, acquisition, which is moved back, takes class activity warning information and cleared alarm information is corresponding
Cleaning after data set；
Extract the characteristic information after the cleaning in data set；
It is analyzed and processed using characteristic information of the fault analysis model after training to data set after cleaning, to be moved back
Take warning information and power failure probability.
Further, method as described above, using the fault analysis model after training to the feature of data set after cleaning
Information is analyzed and processed, and is moved back before taking warning information and power failure probability with obtaining, further includes:
The fault analysis model is constructed using bayesian algorithm；
The fault analysis model is trained using training set, with the fault analysis model after being trained；
Wherein, include at least in the fault analysis model: the first model and the second model, first model are single small
Area's time out of service and abnormal electrical power supply probabilistic model, second model are multiple cell time out of service and abnormal electrical power supply probabilistic model.
Further, method as described above, the fault analysis model using after training is to data set after cleaning
Characteristic information is analyzed and processed, and is moved back after taking warning information and power failure probability with acquisition, further includes:
Warning information is taken to moving back of getting using Laplacian algorithm and power failure probability is modified.
Further, method as described above, the fault analysis model using after training is to data set after cleaning
Characteristic information is analyzed and processed, and is moved back after taking warning information and power failure probability with acquisition, further includes:
Obtain the historical time and the characteristic information of the communication equipment fault；
It is long according to the historical time determining shortest time apart from current time；
Using preset artificial nerve network model to the power failure probability, the shortest time long and described feature
Information is analyzed and processed, judge power supply whether failure.
Further, method as described above, it is described to judge power supply whether after failure, further includes:
If being greater than or equal to using power failure probability acquired in the fault analysis model after the training preset general
Rate threshold value, and be power failure using judging result acquired in the artificial nerve network model, it is determined that the communication is set
Standby power failure；
If being less than preset probability threshold value using power failure probability acquired in the fault analysis model after the training,
And utilizing judging result acquired in the artificial nerve network model is power supply non-faulting, it is determined that the electricity of the communication equipment
It does not break down in source.
Further, method as described above, further includes:
Obtain the real data of the communication equipment power work；
Judge whether the conclusion that power failure or power supply do not break down is correct according to the real data；
If the conclusion is correct, using data corresponding to the conclusion to after the training fault analysis model and institute
It states artificial nerve network model and carries out model optimization, until the artificial nerve network model reaches preset optimization layer subthreshold
Until.
Further, method as described above, the log key message include following information: cell activity alarm letter
Breath, cell cleared alarm information, cell alarm log information；
Cell alarm log information includes following information: site identity, cell ID, alarm name, alarm time of origin；
The characteristic information includes at least following characteristics information: cell, which is moved back, takes information, log abnormal electrical power supply information, time sequence
Column information, base station and cell association information；
The power failure probability includes at least one of: single subdistrict power failure probability, two cell power supplys events
Hinder probability, multiple cell power failure probability.
Second aspect, the embodiment of the present invention provide a kind of device of communication equipment power failure analysis, comprising:
Module is obtained, for obtaining the log key message after communication equipment fault restores；
Cleaning module, for cleaning to the log key message, acquisition, which is moved back, takes class activity warning information and removing
Data set after the corresponding cleaning of warning information；
Extraction module, for extracting the characteristic information after the cleaning in data set；
Analysis module, for being analyzed using characteristic information of the fault analysis model after training to data set after cleaning
Processing takes warning information and power failure probability to obtain to move back.
Further, device as described above, further includes:
Module is constructed, for constructing the fault analysis model using bayesian algorithm；
Training module, for being trained to the fault analysis model using training set, with the failure after being trained
Analysis model；
Wherein, include at least in the fault analysis model: the first model and the second model, first model are single small
Area's time out of service and abnormal electrical power supply probabilistic model, second model are multiple cell time out of service and abnormal electrical power supply probabilistic model.
Further, device as described above, further includes:
Correction module, for taking warning information to moving back of getting using Laplacian algorithm and power failure probability carries out
Amendment.
Further, device as described above, further includes: determining module and judgment module；
The acquisition module is also used to obtain the historical time and the characteristic information of the communication equipment fault；
The determining module is also used to long apart from the shortest time at current time according to historical time determination；
The judgment module, for using preset artificial nerve network model to the power failure probability, it is described most
Short time is long and the characteristic information is analyzed and processed, judge power supply whether failure.
Further, device as described above, the determining module, if being also used to utilize the accident analysis after the training
Power failure probability acquired in model is greater than or equal to preset probability threshold value, and utilizes the artificial nerve network model institute
The judging result of acquisition is power failure, it is determined that the power failure of the communication equipment；If after the training
Power failure probability acquired in fault analysis model is less than preset probability threshold value, and utilizes the artificial nerve network model
Acquired judging result is power supply non-faulting, it is determined that the power supply of the communication equipment does not break down.
Further, device as described above, further includes: optimization module；
The acquisition module is also used to obtain the real data of the communication equipment power work；
The judgment module is also used to judge that power failure or power supply do not break down according to the real data
Conclusion it is whether correct；
The optimization module, if correct for the conclusion, using data corresponding to the conclusion to the training after
Fault analysis model and the artificial nerve network model carry out model optimization, until the artificial nerve network model reaches
Until preset optimization layer subthreshold.
Further, device as described above, the log key message include following information: cell activity alarm letter
Breath, cell cleared alarm information, cell alarm log information；
Cell alarm log information includes following information: site identity, cell ID, alarm name, alarm time of origin；
The characteristic information includes at least following characteristics information: cell, which is moved back, takes information, log abnormal electrical power supply information, time sequence
Column information, base station and cell association information；
The power failure probability includes at least one of: single subdistrict power failure probability, two cell power supplys events
Hinder probability, multiple cell power failure probability.
The third aspect, the embodiment of the present invention provide a kind of electronic equipment, comprising:
Memory, processor and computer program；
Wherein, the computer program stores in the memory, and is configured as being executed by the processor with reality
Now such as the described in any item methods of first aspect.
Fourth aspect, the embodiment of the present invention provide a kind of readable storage medium storing program for executing, are stored thereon with computer program, the meter
Calculation machine program is executed by processor to realize the method as described in any one of first aspect.
The embodiment of the present invention provides method, apparatus, equipment and the storage medium of a kind of communication equipment power failure analysis, leads to
It crosses and obtains the log key message after communication equipment fault restores；Log key message is cleaned, acquisition, which is moved back, takes class activity
Data set after warning information and the corresponding cleaning of cleared alarm information；Extract the characteristic information after cleaning in data set；It utilizes
Fault analysis model after training is analyzed and processed the characteristic information of data set after cleaning, with obtain move back take warning information and
Power failure probability.It can be by actively obtaining log key message, and communication is quickly analyzed by fault analysis model and is set
Standby power failure probability, improves the analysis efficiency of power failure.
It should be appreciated that content described in foregoing invention content part is not intended to limit the pass of the embodiment of the present invention
Key or important feature, the range being also not intended to limit the invention.Other feature of the invention will become to hold by description below
It is readily understood.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow chart of the method for the communication equipment power failure analysis that the embodiment of the present invention one provides；
Fig. 2 is the flow chart of the method for communication equipment power failure provided by Embodiment 2 of the present invention analysis；
Fig. 3 is the structural schematic diagram of the device for the communication equipment power failure analysis that the embodiment of the present invention three provides；
Fig. 4 is the structural schematic diagram of the device for the communication equipment power failure analysis that the embodiment of the present invention four provides；
Fig. 5 is the structural schematic diagram for the electronic equipment that the embodiment of the present invention five provides.
Specific embodiment
Embodiment of the present invention will be described in more detail below with reference to accompanying drawings.Although being shown in attached drawing of the invention certain
Embodiment, it should be understood that, the present invention can be realized by various forms, and should not be construed as being limited to this
In the embodiment that illustrates, providing these embodiments on the contrary is in order to more thorough and be fully understood by the present invention.It should be understood that
It is that being given for example only property of accompanying drawings and embodiments effect of the invention is not intended to limit protection scope of the present invention.
The specification and claims of the embodiment of the present invention and the term " first " in above-mentioned attached drawing, " second ", "
Three ", the (if present)s such as " 4th " are to be used to distinguish similar objects, without for describing specific sequence or successive time
Sequence.It should be understood that the data used in this way are interchangeable under appropriate circumstances, so as to the embodiment of the present invention described herein as can
The enough sequence implementation with other than those of illustrating or describe herein.In addition, term " includes " and " having " and they
Any deformation, it is intended that cover it is non-exclusive include, for example, containing the process, method of a series of steps or units, being
System, product or equipment those of are not necessarily limited to be clearly listed step or unit, but may include be not clearly listed or
For the intrinsic other step or units of these process, methods, product or equipment.
Embodiment one
Fig. 1 is the flow chart of the method for the communication equipment power failure analysis that the embodiment of the present invention one provides, such as Fig. 1 institute
Show, the executing subject of the present embodiment is the device of communication equipment power failure analysis, the dress of communication equipment power failure analysis
Computer can be integrated in by setting, in the electronic equipments such as laptop or server, then communication equipment power supply provided in this embodiment
The method of accident analysis includes following steps.
Step 101, the log key message after communication equipment fault restores is obtained.
It specifically, can be by sending request command to Element management system (abbreviation EMS), from network element pipe in the present embodiment
The log key message after communication equipment fault restores is obtained in reason system.Day after can also in advance restoring communication equipment fault
Will key message is stored, the log key message after storage region acquisition communication equipment fault recovery, in the present embodiment
This is not construed as limiting.
Wherein, log key message may include: cell activity warning information, cell cleared alarm information, cell alarm
The information such as log information.When cell alarm log information may include: site identity, cell ID, alarm name, alarm generation
Between etc. information.
Step 102, log key message is cleaned, acquisition, which is moved back, takes class activity warning information and cleared alarm information phase
Data set after corresponding cleaning.
Specifically, in the present embodiment, the log key message of acquisition is cleaned, acquisition, which is moved back, takes class activity warning information
With cleared alarm information, wherein moving back and taking class activity warning information and cleared alarm information is to break relevant correspondence to non-super stroboscopic
Information.
Class activity warning information is taken by moving back of breaking of non-super stroboscopic and cleared alarm information constitutes the data set after cleaning.
Step 103, the characteristic information after cleaning in data set is extracted.
Specifically, in the present embodiment, the characteristic information in the data set after cleaning is extracted using feature extraction algorithm.Its
In, characteristic information may include: that cell moves back and takes information, log abnormal electrical power supply information, time serial message, base station and cell association
Information etc..
Step 104, it is analyzed and processed using characteristic information of the fault analysis model after training to data set after cleaning,
Warning information and power failure probability are taken to obtain to move back.
Specifically, in the present embodiment, fault analysis model is constructed first, after then restoring using multiple communication equipment faults
Log key message carry out cleaning and feature extraction after, form the training set of fault analysis model, to fault analysis model into
Row training, the fault analysis model after being trained.
Wherein, fault analysis model can be constructed using bayesian algorithm.
In the present embodiment, power failure probability can be single subdistrict power failure probability, and two cell power failures are general
Rate or multiple cell power failure probability.
The method of communication equipment power failure analysis provided in this embodiment, device, equipment and storage medium, pass through acquisition
Log key message after communication equipment fault recovery；Log key message is cleaned, acquisition, which is moved back, takes class activity alarm letter
Data set after breath cleaning corresponding with cleared alarm information；Extract the characteristic information after cleaning in data set；After training
Fault analysis model the characteristic information of data set after cleaning is analyzed and processed, with obtain move back take warning information and power supply therefore
Hinder probability.It can be by actively obtaining log key message, and pass through the electricity that fault analysis model quickly analyzes communication equipment
Source probability of malfunction improves the analysis efficiency of power failure.
Embodiment two
The flow chart of the method for communication equipment power failure analysis provided by Embodiment 2 of the present invention, as shown in Fig. 2, this reality
The method for applying the communication equipment power failure analysis of example offer is the communication equipment power failure provided in the embodiment of the present invention one
On the basis of the method for analysis, further refinement to step 101- step 104, and further comprise other steps, then this reality
The method for applying the communication equipment power failure analysis of example offer includes the following steps.
Step 201, fault analysis model is constructed using bayesian algorithm.
Further, in this embodiment constructing fault analysis model using bayesian algorithm.Bayesian algorithm can be Piao
Plain bayesian algorithm.
Wherein, NB Algorithm is assumed to be characterized in conditional sampling for classification under conditions of classifying determining,
I.e. formula (1) is set up.
Wherein, x(1),X(2)=x(2),…,X(n)For a division of sample space, p (X(1)) it is more than or equal to zero.
According to Bayes' theorem and the conditional sampling of characteristic of division, it is assumed that there is formula (2) establishment
Formula (2) is using the fault analysis model of bayesian algorithm building, P (Y=ck) according to ckValue, indicate
Power failure probability or non-faulting probability.
Step 202, fault analysis model is trained using training set, with the fault analysis model after being trained.
Wherein, include at least in fault analysis model: the first model and the second model, the first model are that single cell is moved back when taking
Between with abnormal electrical power supply probabilistic model, the second model is multiple cell time out of service and abnormal electrical power supply probabilistic model.
Further, in this embodiment the training sample in training set is the log pass after certain communication equipment fault restores
The characteristic information that key information extracts after being cleaned.
Step 203, the log key message after communication equipment fault restores is obtained.
Further, in this embodiment log key message includes following information: cell activity warning information, cell are clear
Except warning information, cell alarm log information.Cell alarm log information includes following information: site identity, cell ID, announcement
Alert title, alarm time of origin.
Step 204, log key message is cleaned, acquisition, which is moved back, takes class activity warning information and cleared alarm information phase
Data set after corresponding cleaning.
Step 205, the characteristic information after cleaning in data set is extracted.
Further, in this embodiment characteristic information includes at least following characteristics information: cell moves back and takes information, log supplies
Electrical anomaly information, time serial message, base station and cell association information.
Step 206, it is analyzed and processed using characteristic information of the fault analysis model after training to data set after cleaning,
Warning information and power failure probability are taken to obtain to move back.
Further, in this embodiment due to being included at least in fault analysis model: the first model and the second model, the
One model is single cell time out of service and abnormal electrical power supply probabilistic model, and the second model is that multiple cell time out of service and abnormal electrical power supply are general
Rate model.So being analyzed and processed using characteristic information of the fault analysis model after training to data set after cleaning, obtain
Power failure probability include at least one of: single subdistrict power failure probability, two cell power failure probability are multiple
Cell power failure probability.
Step 207, warning information is taken to moving back of getting using Laplacian algorithm and power failure probability is modified.
Further, in this embodiment using Maximum-likelihood estimation after using bayesian algorithm building fault analysis model
It is possible that the case where estimated probability value is zero, this influences whether the calculated result of posterior probability, generates classification inclined
Difference carries out " smooth " in estimated probability in the present embodiment in order to solve this problem, i.e., using Laplacian algorithm to acquisition
To move back and take warning information and power failure probability is modified, Laplacian algorithm is modified substantially assume attribute value and
Class label, which is divided equally, to be distributed, this is equivalent to the priori additionally introduced during Bayesian learning about data.Then in failure point
Correction factor λ is increased in analysis model, then fault analysis model is shown in formula (3).
Wherein, k=1,2 ..., K, j=1,2 ..., n, l=1,2 ..., sj。
Wherein, the value of correction factor λ is 1.
Step 208, the historical time and characteristic information of communication equipment fault are obtained.
Specifically, in the present embodiment, the historical time of communication equipment fault may include going through for the failure occurred in history
The history time, if primary fault occurs, for the historical time of primary fault, if multiple failure, then going through including multiple failure
The history time.
Wherein, characteristic information is the characteristic information in step 205.
Step 209, long according to the historical time determining shortest time apart from current time.
Further, in this embodiment it is determining from the shortest historical time of current time interval in historical time, it calculates
Difference from current time interval shortest historical time and current time, acquisition are long apart from the shortest time at current time.
Step 210, using preset artificial nerve network model to power failure probability, shortest time length and characteristic information
Be analyzed and processed, judge power supply whether failure.
Further, in this embodiment by power failure probability, the shortest time is long and characteristic information is input to artificial neuron
In network model, power failure probability is as the predictive factor in artificial nerve network model, shortest time length and characteristic information
As the predictive variable in artificial nerve network model, artificial nerve network model is according to predictive factor and predictive variable to power supply
Whether faulty judged.
Wherein, artificial neural network network select hidden layer can be 30 layers or more, activation primitive be relu and
Sigmoid cross-reference.
Step 211, if being greater than or equal to using power failure probability acquired in the fault analysis model after training default
Probability threshold value, and be power failure using judging result acquired in artificial nerve network model, it is determined that communication equipment
Power failure, if being less than preset probability threshold using power failure probability acquired in the fault analysis model after training
Value, and be power supply non-faulting using judging result acquired in artificial nerve network model, it is determined that the power supply of communication equipment is not
It breaks down.
Further, in this embodiment being carried out according to the result that fault analysis model and artificial nerve network model export
The final judgement whether power supply breaks down.If being greater than using power failure probability acquired in the fault analysis model after training
Or be equal to preset probability threshold value, and be power failure using judging result acquired in artificial nerve network model, then illustrate
The result direction of two kinds of model analysis is consistent, i.e., failure has occurred in power supply, it is determined that the power failure of communication equipment, if sharp
The power failure probability acquired in the fault analysis model after training is less than preset probability threshold value, and utilizes artificial neural network
Judging result acquired in network model is power supply non-faulting, then illustrates that the result direction of two kinds of model analysis is consistent, i.e., power supply is not
It breaks down, it is determined that the power supply of communication equipment does not break down.
Wherein, preset probability threshold value such as can be able to be 70% or other numerical value, this reality by determining after test of many times
It applies in example and this is not construed as limiting.
Step 212, the real data of communication equipment power work is obtained.
Specifically, in the present embodiment, the real data of power work can be obtained from communication equipment, is wrapped in real data
Include the working condition of each stage power supply, if it breaks down, the information such as time of failure.
Step 213, judge whether the conclusion that power failure or power supply do not break down is correct according to real data.
Further, in this embodiment by step 211 power failure or the conclusion that does not break down of power supply with
Whether the power supply in real data breaks down progress in this regard, whether judgement conclusion is correct.
Step 214, if conclusion is correct, using data corresponding to the conclusion to after training fault analysis model and people
Artificial neural networks model carries out model optimization, until artificial nerve network model reaches preset optimization layer subthreshold.
Wherein, the default optimization layer subthreshold of artificial neural network can be 30 or other numerical value, to this in the present embodiment
It is not construed as limiting.
The method of communication equipment power failure analysis provided in this embodiment, by constructing failure point using bayesian algorithm
Model is analysed, fault analysis model is trained using training set, with the fault analysis model after train, acquisition is communicated and set
Log key message after standby fault recovery, cleans log key message, and acquisition, which is moved back, takes class activity warning information and clear
Except data set after the corresponding cleaning of warning information, the characteristic information after cleaning in data set is extracted, the failure after training is utilized
Analysis model is analyzed and processed the characteristic information of data set after cleaning, takes warning information and power failure is general to obtain to move back
Rate takes warning information to moving back of getting using Laplacian algorithm and power failure probability is modified, can be improved training
The accuracy that fault analysis model afterwards analyzes power failure probability.
The method of communication equipment power failure analysis provided in this embodiment, due to constructing failure using bayesian algorithm
Analysis model is trained fault analysis model using training set, when with fault analysis model after being trained, failure point
Include at least in analysis model: the first model and the second model, the first model are single cell time out of service and abnormal electrical power supply probability mould
Type, the second model is multiple cell time out of service and abnormal electrical power supply probabilistic model, and is divided the power failure of communication equipment
It when analysis, is also extracted single cell and moving back for multiple cell takes information and abnormal electrical power supply information, set so being able to carry out communication in the whole network
The analysis of standby power failure.
The method of communication equipment power failure analysis provided in this embodiment, is utilizing the fault analysis model pair after training
The characteristic information of data set is analyzed and processed after cleaning, is moved back after taking warning information and power failure probability with acquisition, is obtained logical
The historical time and characteristic information for believing equipment fault, it is long according to the historical time determining shortest time apart from current time, it utilizes
Preset artificial nerve network model is analyzed and processed power failure probability, shortest time length and characteristic information, judges electricity
Source whether failure, if using training after fault analysis model acquired in power failure probability be greater than or equal to preset probability
Threshold value, and be power failure using judging result acquired in artificial nerve network model, it is determined that the power supply of communication equipment is sent out
Raw failure, if being less than preset probability threshold value, and benefit using power failure probability acquired in the fault analysis model after training
Judging result acquired in employment artificial neural networks model is power supply non-faulting, it is determined that event does not occur for the power supply of communication equipment
Barrier.It can be according to communication equipment determining under the common dissection of fault analysis model and artificial nerve network model after training
Whether power failure occurs, precision of analysis can be further increased.
The method of communication equipment power failure analysis provided in this embodiment, by the reality for obtaining communication equipment power work
Border data；Judge whether the conclusion that power failure or power supply do not break down is correct according to real data；If conclusion is just
Really, then using data corresponding to the conclusion to after training fault analysis model and artificial nerve network model carry out model it is excellent
Change, until artificial nerve network model reaches preset optimization layer subthreshold, conclusion correctly corresponding data can be used
To after training fault analysis model and artificial nerve network model optimize, and then can make training after accident analysis mould
Type and artificial nerve network model are in the process continued to optimize, and can further increase event with increasing for accident analysis number
Hinder the accuracy of analysis.
Embodiment three
Fig. 3 is the structural schematic diagram of the device for the communication equipment power failure analysis that the embodiment of the present invention three provides, such as Fig. 3
Shown, the device of communication equipment power failure analysis provided in this embodiment includes: to obtain module 31, and cleaning module 32 extracts
Module 33 and analysis module 34.
Wherein, module 31 is obtained, for obtaining the log key message after communication equipment fault restores.Cleaning module 32,
For being cleaned to log key message, acquisition is moved back take class activity warning information and the corresponding cleaning of cleared alarm information after
Data set.Extraction module 33, for extracting the characteristic information after cleaning in data set.Analysis module 34, after using training
Fault analysis model the characteristic information of data set after cleaning is analyzed and processed, with obtain move back take warning information and power supply therefore
Hinder probability.
The device of communication equipment power failure analysis provided in this embodiment can execute the skill of embodiment of the method shown in Fig. 1
Art scheme, it is similar that the realization principle and technical effect are similar, and details are not described herein again.
Fig. 4 is the structural schematic diagram of the device for the communication equipment power failure analysis that the embodiment of the present invention four provides, such as Fig. 4
Communication equipment electricity shown, that the device of communication equipment power failure analysis provided in this embodiment is provided in the embodiment of the present invention three
On the basis of the device of source accident analysis, further, further includes: building module 41, training module 42, correction module 43, really
Cover half block 44, judgment module 45 and optimization module 46.
Further, module 41 is constructed, for constructing fault analysis model using bayesian algorithm.Training module 42 is used
In being trained to fault analysis model using training set, with the fault analysis model after being trained.Wherein, accident analysis mould
Include at least in type: the first model and the second model, the first model are single cell time out of service and abnormal electrical power supply probabilistic model, the
Two models are multiple cell time out of service and abnormal electrical power supply probabilistic model.
Further, correction module 43, for taking warning information and power supply to moving back of getting using Laplacian algorithm
Probability of malfunction is modified.
Further, module 31 is obtained, is also used to obtain the historical time and characteristic information of communication equipment fault.Determine mould
Block 44 is also used to long apart from the shortest time at current time according to historical time determination.Judgment module 45, it is preset for utilizing
Artificial nerve network model is analyzed and processed power failure probability, shortest time length and characteristic information, whether judges power supply
Failure.
Further, it is determined that module 44, if being also used to utilize power failure acquired in the fault analysis model after training
Probability is greater than or equal to preset probability threshold value, and is power supply event using judging result acquired in artificial nerve network model
Barrier, it is determined that the power failure of communication equipment；If general using power failure acquired in the fault analysis model after training
Rate is less than preset probability threshold value, and is power supply non-faulting using judging result acquired in artificial nerve network model, then really
The power supply for determining communication equipment does not break down.
Further, module 31 is obtained, is also used to obtain the real data of communication equipment power work.Judgment module 45,
It is also used to judge according to real data whether the conclusion that power failure or power supply do not break down is correct.Optimization module
46, if correct for conclusion, using data corresponding to the conclusion to the fault analysis model and artificial neural network after training
Network model carries out model optimization, until artificial nerve network model reaches preset optimization layer subthreshold.
Further, in this embodiment log key message includes following information: cell activity warning information, cell are clear
Except warning information, cell alarm log information；Cell alarm log information includes following information: site identity, cell ID, announcement
Alert title, alarm time of origin；Characteristic information includes at least following characteristics information: cell, which is moved back, takes information, log abnormal electrical power supply letter
Breath, time serial message, base station and cell association information；Power failure probability includes at least one of: single subdistrict power supply
Probability of malfunction, two cell power failure probability, multiple cell power failure probability.
The device of communication equipment power failure analysis provided in this embodiment can execute the skill of embodiment of the method shown in Fig. 2
Art scheme, it is similar that the realization principle and technical effect are similar, and details are not described herein again.
Embodiment five
Fig. 5 is the structural schematic diagram for the electronic equipment that the embodiment of the present invention five provides, as shown in figure 5, in the present embodiment, it should
Electronic equipment includes: memory 51, processor 52 and computer program.
Wherein, computer program is stored in memory 51, and is configured as being executed by processor 52 to realize the present invention
The analysis method for the communication equipment power failure that embodiment one provides or communication equipment power supply provided by Embodiment 2 of the present invention event
The analysis method of barrier.
Related description can correspond to the corresponding associated description and effect of the step of referring to Fig. 1 to Fig. 2 and be understood, herein
It does not do and excessively repeats.
Embodiment six
The embodiment of the present invention six provides a kind of computer readable storage medium, is stored thereon with computer program, computer
The analysis method or the present invention that program is executed by processor to realize the communication equipment power failure of the offer of the embodiment of the present invention one
The analysis method for the communication equipment power failure that embodiment two provides.
In several embodiments provided by the present invention, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, the apparatus embodiments described above are merely exemplary, for example, the division of module, only
A kind of logical function partition, there may be another division manner in actual implementation, for example, multiple module or components can combine or
Person is desirably integrated into another system, or some features can be ignored or not executed.Another point, shown or discussed is mutual
Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of device or module
It connects, can be electrical property, mechanical or other forms.
Module may or may not be physically separated as illustrated by the separation member, show as module
Component may or may not be physical module, it can and it is in one place, or may be distributed over multiple networks
In module.Some or all of the modules therein can be selected to achieve the purpose of the solution of this embodiment according to the actual needs.
It, can also be in addition, each functional module in each embodiment of the present invention can integrate in a processing module
It is that modules physically exist alone, can also be integrated in two or more modules in a module.Above-mentioned integrated mould
Block both can take the form of hardware realization, can also realize in the form of hardware adds software function module.
The program code of method for carrying out the present invention can using any combination of one or more programming languages come
It writes.These program codes can be supplied to the place of general purpose computer, special purpose computer or other programmable data processing units
Device or controller are managed, so that program code makes defined in flowchart and or block diagram when by processor or controller execution
Function/operation is carried out.Program code can be executed completely on machine, partly be executed on machine, as stand alone software
Is executed on machine and partly execute or executed on remote machine or server completely on the remote machine to packet portion.
In the context of the present invention, machine readable media can be tangible medium, may include or is stored for
The program that instruction execution system, device or equipment are used or is used in combination with instruction execution system, device or equipment.Machine can
Reading medium can be machine-readable signal medium or machine-readable storage medium.Machine readable media can include but is not limited to electricity
Son, magnetic, optical, electromagnetism, infrared or semiconductor system, device or equipment or above content any conjunction
Suitable combination.The more specific example of machine readable storage medium will include the electrical connection of line based on one or more, portable meter
Calculation machine disk, hard disk, random access memory (RAM), read-only memory (ROM), Erasable Programmable Read Only Memory EPROM (EPROM
Or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage facilities or
Any appropriate combination of above content.
Although this should be understood as requiring operating in this way with shown in addition, depicting each operation using certain order
Certain order out executes in sequential order, or requires the operation of all diagrams that should be performed to obtain desired result.
Under certain environment, multitask and parallel processing be may be advantageous.Similarly, although containing several tools in being discussed above
Body realizes details, but these are not construed as the limitation to the scope of the present disclosure.In the context of individual embodiment
Described in certain features can also realize in combination in single realize.On the contrary, in the described in the text up and down individually realized
Various features can also realize individually or in any suitable subcombination in multiple realizations.
Although having used specific to this theme of the language description of structure feature and/or method logical action, answer
When understanding that theme defined in the appended claims is not necessarily limited to special characteristic described above or movement.On on the contrary,
Special characteristic described in face and movement are only to realize the exemplary forms of claims.
Claims (10)
1. a kind of method of communication equipment power failure analysis characterized by comprising
Obtain the log key message after communication equipment fault restores；
The log key message is cleaned, acquisition, which is moved back, takes class activity warning information and cleared alarm information is corresponding clear
Wash rear data set；
Extract the characteristic information after the cleaning in data set；
It is analyzed and processed using characteristic information of the fault analysis model after training to data set after cleaning, moves back clothes announcement to obtain
Alert information and power failure probability.
2. the method according to claim 1, wherein using the fault analysis model after training to data after cleaning
The characteristic information of collection is analyzed and processed, and is moved back before taking warning information and power failure probability with obtaining, further includes:
The fault analysis model is constructed using bayesian algorithm；
The fault analysis model is trained using training set, with the fault analysis model after being trained；
Wherein, include at least in the fault analysis model: the first model and the second model, first model are that single cell is moved back
Time and abnormal electrical power supply probabilistic model are taken, second model is multiple cell time out of service and abnormal electrical power supply probabilistic model.
3. the method according to claim 1, wherein after the fault analysis model using after training is to cleaning
The characteristic information of data set is analyzed and processed, and is moved back after taking warning information and power failure probability with acquisition, further includes:
Warning information is taken to moving back of getting using Laplacian algorithm and power failure probability is modified.
4. method according to claim 1-3, which is characterized in that the fault analysis model using after training
The characteristic information of data set after cleaning is analyzed and processed, is moved back with acquisition after taking warning information and power failure probability, also
Include:
Obtain the historical time and the characteristic information of the communication equipment fault；
It is long according to the historical time determining shortest time apart from current time；
Using preset artificial nerve network model to the power failure probability, the shortest time long and described characteristic information
Be analyzed and processed, judge power supply whether failure.
5. according to the method described in claim 4, it is characterized in that, whether described judge power supply after failure, further includes:
If being greater than or equal to preset probability threshold using power failure probability acquired in the fault analysis model after the training
Value, and be power failure using judging result acquired in the artificial nerve network model, it is determined that the communication equipment
Power failure；
If being less than preset probability threshold value, and benefit using power failure probability acquired in the fault analysis model after the training
The judging result acquired in the artificial nerve network model is power supply non-faulting, it is determined that the power supply of the communication equipment is not
It breaks down.
6. according to the method described in claim 5, it is characterized by further comprising:
Obtain the real data of the communication equipment power work；
Judge whether the conclusion that power failure or power supply do not break down is correct according to the real data；
If the conclusion is correct, using data corresponding to the conclusion to after the training fault analysis model and the people
Artificial neural networks model carries out model optimization, until the artificial nerve network model reaches preset optimization layer subthreshold and is
Only.
7. method according to claim 1-3, which is characterized in that the log key message includes following letter
Breath: cell activity warning information, cell cleared alarm information, cell alarm log information；
Cell alarm log information includes following information: site identity, cell ID, alarm name, alarm time of origin；
The characteristic information includes at least following characteristics information: cell, which is moved back, takes information, log abnormal electrical power supply information, time series letter
Breath, base station and cell association information；
The power failure probability includes at least one of: single subdistrict power failure probability, and two cell power failures are general
Rate, multiple cell power failure probability.
8. a kind of device of communication equipment power failure analysis characterized by comprising
Module is obtained, for obtaining the log key message after communication equipment fault restores；
Cleaning module, for cleaning to the log key message, acquisition, which is moved back, takes class activity warning information and cleared alarm
Data set after the corresponding cleaning of information；
Extraction module, for extracting the characteristic information after the cleaning in data set；
Analysis module, for being carried out at analysis using characteristic information of the fault analysis model after training to data set after cleaning
Reason takes warning information and power failure probability to obtain to move back.
9. a kind of electronic equipment characterized by comprising
Memory, processor and computer program；
Wherein, the computer program stores in the memory, and is configured as being executed by the processor to realize such as
Method of any of claims 1-7.
10. a kind of readable storage medium storing program for executing, which is characterized in that be stored thereon with computer program, the computer program is processed
Device is executed to realize such as method of any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811422339.5A CN109309594B (en) | 2018-11-27 | 2018-11-27 | Method, device, equipment and storage medium for analyzing power failure of communication equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811422339.5A CN109309594B (en) | 2018-11-27 | 2018-11-27 | Method, device, equipment and storage medium for analyzing power failure of communication equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109309594A true CN109309594A (en) | 2019-02-05 |
| CN109309594B CN109309594B (en) | 2021-11-16 |
Family
ID=65223061
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811422339.5A Active CN109309594B (en) | 2018-11-27 | 2018-11-27 | Method, device, equipment and storage medium for analyzing power failure of communication equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109309594B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111488265A (en) * | 2020-04-27 | 2020-08-04 | 北京奇艺世纪科技有限公司 | Fault prediction method, device, equipment and readable storage medium |
| CN111880981A (en) * | 2020-07-30 | 2020-11-03 | 北京浪潮数据技术有限公司 | Fault repairing method and related device for docker container |
| CN113282000A (en) * | 2021-04-30 | 2021-08-20 | 科华数据股份有限公司 | Fault diagnosis method and device of data center and dynamic loop monitoring system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101917297A (en) * | 2010-08-30 | 2010-12-15 | 烽火通信科技股份有限公司 | Method and system for diagnosing faults of core network based on Bayesian network |
| CN103428737A (en) * | 2012-05-16 | 2013-12-04 | 中兴通讯股份有限公司 | Method and device for confirming event source |
| CN104244300A (en) * | 2013-06-17 | 2014-12-24 | 中国移动通信集团浙江有限公司 | Method and system for achieving base station dynamic environment monitoring |
| US20170078400A1 (en) * | 2012-01-09 | 2017-03-16 | May Patents Ltd. | System and method for server based control |
| CN107248927A (en) * | 2017-05-02 | 2017-10-13 | 华为技术有限公司 | Generation method, Fault Locating Method and the device of fault location model |
| CN108303632A (en) * | 2017-12-14 | 2018-07-20 | 佛山科学技术学院 | Circuit failure diagnosis method based on random forests algorithm |
| CN108763654A (en) * | 2018-05-03 | 2018-11-06 | 国网江西省电力有限公司信息通信分公司 | A kind of electrical equipment fault prediction technique based on Weibull distribution and hidden Semi-Markov Process |
-
2018
- 2018-11-27 CN CN201811422339.5A patent/CN109309594B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101917297A (en) * | 2010-08-30 | 2010-12-15 | 烽火通信科技股份有限公司 | Method and system for diagnosing faults of core network based on Bayesian network |
| US20170078400A1 (en) * | 2012-01-09 | 2017-03-16 | May Patents Ltd. | System and method for server based control |
| CN103428737A (en) * | 2012-05-16 | 2013-12-04 | 中兴通讯股份有限公司 | Method and device for confirming event source |
| CN104244300A (en) * | 2013-06-17 | 2014-12-24 | 中国移动通信集团浙江有限公司 | Method and system for achieving base station dynamic environment monitoring |
| CN107248927A (en) * | 2017-05-02 | 2017-10-13 | 华为技术有限公司 | Generation method, Fault Locating Method and the device of fault location model |
| CN108303632A (en) * | 2017-12-14 | 2018-07-20 | 佛山科学技术学院 | Circuit failure diagnosis method based on random forests algorithm |
| CN108763654A (en) * | 2018-05-03 | 2018-11-06 | 国网江西省电力有限公司信息通信分公司 | A kind of electrical equipment fault prediction technique based on Weibull distribution and hidden Semi-Markov Process |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111488265A (en) * | 2020-04-27 | 2020-08-04 | 北京奇艺世纪科技有限公司 | Fault prediction method, device, equipment and readable storage medium |
| CN111880981A (en) * | 2020-07-30 | 2020-11-03 | 北京浪潮数据技术有限公司 | Fault repairing method and related device for docker container |
| CN113282000A (en) * | 2021-04-30 | 2021-08-20 | 科华数据股份有限公司 | Fault diagnosis method and device of data center and dynamic loop monitoring system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109309594B (en) | 2021-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Interval-valued fuzzy spiking neural P systems for fault diagnosis of power transmission networks | |
| CN109581871B (en) | Industrial control system intrusion detection method of immune countermeasure sample | |
| CN110224160B (en) | Fault diagnosis method for fuel cell system | |
| CN109309594A (en) | Method, apparatus, equipment and the storage medium of communication equipment power failure analysis | |
| CN103617469B (en) | Power system device failure prediction method and system | |
| Le Mortellec et al. | Embedded holonic fault diagnosis of complex transportation systems | |
| CN103995215B (en) | A kind of smart power grid fault diagnostic method based on multi-level feedback adjustment | |
| Wang et al. | Comparison of availability between two systems with warm standby units and different imperfect coverage | |
| CN109964182A (en) | Method and system for vehicle analysis | |
| EP3968479A1 (en) | Systems and methods for automatic power topology discovery | |
| Said et al. | Decentralized fault detection and isolation using bond graph and PCA methods | |
| CN103197168A (en) | Fault diagnosis control method in power system based on event set causal chain | |
| Khorasgani et al. | A methodology for monitoring smart buildings with incomplete models | |
| CN116205265A (en) | Power grid fault diagnosis method and device based on deep neural network | |
| KR20190107523A (en) | System and method for handling network failure using syslog | |
| Harvey et al. | Attention for inference compilation | |
| CN111489539A (en) | Household appliance system fault early warning method, system and device | |
| Bregon et al. | Fault diagnosis in hybrid systems using possible conflicts | |
| KR102107689B1 (en) | Apparatus and method for analyzing cause of network failure | |
| Ahn | Deep learning based anomaly detection for a vehicle in swarm drone system | |
| US11144046B2 (en) | Fault signal recovery apparatus and method | |
| Razavi-Far et al. | Ensemble of extreme learning machines for diagnosing bearing defects in non-stationary environments under class imbalance condition | |
| Berenji et al. | Case-based reasoning for fault diagnosis and prognosis | |
| Ye et al. | The efficiency of detecting the failures and troubleshooting while applying technical diagnostics for multi-computer systems | |
| Sarkar et al. | Spatiotemporal information fusion for fault detection in shipboard auxiliary systems |
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 |