CN107222322A - A kind of communication failure diagnostic method and device - Google Patents
A kind of communication failure diagnostic method and device Download PDFInfo
- Publication number
- CN107222322A CN107222322A CN201610164660.2A CN201610164660A CN107222322A CN 107222322 A CN107222322 A CN 107222322A CN 201610164660 A CN201610164660 A CN 201610164660A CN 107222322 A CN107222322 A CN 107222322A
- Authority
- CN
- China
- Prior art keywords
- failure
- fusion
- possible breakdown
- phenomenon
- conflict
- 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.)
- Pending
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
- H04L41/0631—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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The invention provides a kind of communication failure diagnostic method and device, the existing method for diagnosing faults based on D S evidence theories, when evidences conflict is higher, the problem of fusion diagnosis result is inaccurate are solved.The method of the present invention includes:At least two phenomena of the failure included in fault message are obtained, and obtain the possible breakdown reason corresponding to each phenomenon of the failure;According to the evidence of D S evidence theories and the possible breakdown reason, fusion treatment is carried out to the phenomenon of the failure, draw fusion results and conflict coefficient, the fusion results, which include each possible breakdown reason, turns into the probable value of true fault reason, and the conflict coefficient is used to represent the conflict spectrum between the fusion results and true fault reason;According to the possible breakdown reason, the different information between the phenomenon of the failure is obtained;According to the different information and the conflict coefficient, the fusion results are modified, and fault diagnosis result is drawn according to revised fusion results.
Description
Technical field
The present invention relates to the technical field of converged communication network fault diagnosis, a kind of communication failure diagnostic method is particularly related to
And device.
Background technology
Communication system has structural complexity and running environment particularity, therefore the fault detection and location of communication system,
It is extremely complex, difficult work.Communication failure has at any time when being usually contained in the possibility of generation, communication system failure signal
There is coupling phenomenon between the composition of change, and multiple failures, cause the fault diagnosis difficulty of communication system to increase.It is common at present
The main method of communication system diagnosis has wavelet analysis method, neutral net, comparison method and some optimized algorithms etc..Based on these sides
Easily there is inconsistent situation in diagnostic result produced by method.It is then desired to be examined using certain algorithm communication system failure
Disconnected result carries out conclusion synthesis, to make result consistent.
The classical D-S evidence theory being used widely in data fusion is used as the one of processing unascertained information fusion
Kind of effective ways, but in actual periodic traffic information fusion system, due to environment or human factor can usually make it is some
Equipment exports the information runed counter to actual conditions, and these information often conflict larger with the correct information that miscellaneous equipment is exported.Such as
Where evidence highly conflict under realize the problem of effective integration of multi-source information is one in the urgent need to address.
The content of the invention
It is existing based on D-S cards to solve it is an object of the invention to provide a kind of communication failure diagnostic method and device
According to theoretical method for diagnosing faults, when evidences conflict is higher, the problem of fusion diagnosis result is inaccurate.
To achieve these goals, the invention provides a kind of communication failure diagnostic method, including:
At least two phenomena of the failure included in fault message are obtained, and are obtained corresponding to each phenomenon of the failure
Possible breakdown reason;
According to the evidence of D-S evidence theory and the possible breakdown reason, the phenomenon of the failure is melted
Conjunction is handled, and draws fusion results and conflict coefficient, the fusion results, which include each possible breakdown reason, turns into true fault original
The probable value of cause, the conflict coefficient is used to represent the conflict spectrum between the fusion results and true fault reason;
According to the possible breakdown reason, the different information between the phenomenon of the failure is obtained;
According to the different information and the conflict coefficient, the fusion results are modified, and according to revised
Fusion results draw fault diagnosis result.
Wherein, the evidence and the possible breakdown reason according to D-S evidence theory, it is existing to the failure
As carrying out fusion treatment, fusion results and conflict coefficient are drawn, including:
It is each possible breakdown reason distribution one corresponding to each phenomenon of the failure according to predetermined probabilities allocation rule
Probability values;
According to the evidence of D-S evidence theory and the probability values of each possible breakdown reason, to the event
Hinder phenomenon and carry out fusion treatment, draw the fusion results and the conflict coefficient.
Wherein, the evidence and the probability values of each possible breakdown reason according to D-S evidence theory,
Fusion treatment is carried out to the phenomenon of the failure, the fusion results and the conflict coefficient is drawn, including:
Pass through formulaFusion treatment is carried out to the phenomenon of the failure, drawn described
Fusion results and the conflict coefficient;
Wherein, C ∈ 2U, Ai∈2U, Bj∈2U, φ represents empty set, and U is the set of all possible breakdown reasons, and defines U and be
One identification framework of the possible breakdown reason, 2UThe set constituted for U all subsets, 2U→ [0,1], m is the base on U
This probability assignment function, wherein including the probable value corresponding to the corresponding all possible breakdown reasons of the phenomenon of the failure, m1,
m2According to predetermined probabilities allocation rule it is respectively basic probability assignment function that two phenomena of the failure are distributed for the identification framework,
M (C) represents the fusion results, Represent the conflict coefficient, AiAnd BjRepresent two institutes
State phenomenon of the failure and distinguish corresponding one group of possible breakdown reason.
Wherein, it is described according to the possible breakdown reason, the different information between the phenomenon of the failure is obtained, including:
The basic probability assignment function m of two phenomena of the failure is obtained respectively1And m2;
Different information between described two phenomena of the failure is obtained by equation below:
Wherein, dBPARepresent the difference between described two phenomena of the failure
Different information,2M × 2M matrix is represented, M is that the element in the total number of all possible breakdown reasons, matrix isA and B represent that two phenomena of the failure distinguish corresponding all possible breakdown reasons, and A and B sums are
M。
Wherein, it is described according to the different information and the conflict coefficient, the fusion results are modified, and according to
Revised fusion results draw fault diagnosis result, including:
According to the different information, the conflict coefficient and equation below, conflict coefficient after amendment is obtained;
Represent the conflict coefficient, dBPA(m1,m2) represent two events
Hinder the different information between phenomenon,Represent conflict coefficient after the amendment;
If conflict coefficient is less than predetermined threshold value after the amendment, the fusion results are regard as the fault diagnosis knot
Really, otherwise determine that the fusion results are insincere.
Embodiments of the invention additionally provide a kind of communication failure diagnostic device, including:
First acquisition module, for obtaining at least two phenomena of the failure included in fault message, and obtains each institute
State the possible breakdown reason corresponding to phenomenon of the failure;
Fusion Module, for the evidence according to D-S evidence theory and the possible breakdown reason, to the event
Hinder phenomenon and carry out fusion treatment, draw fusion results and conflict coefficient, the fusion results include each possible breakdown reason into
For the probable value of true fault reason, the conflict coefficient is used to represent rushing between the fusion results and true fault reason
Prominent degree;
Second acquisition module, for according to the possible breakdown reason, obtaining the different information between the phenomenon of the failure;
Correcting module, for according to the different information and the conflict coefficient, being modified to the fusion results, and
Fault diagnosis result is drawn according to revised fusion results.
Wherein, the Fusion Module includes:
Distribution sub module, for according to predetermined probabilities allocation rule, being that each corresponding to each phenomenon of the failure can
Can failure cause one probability values of distribution;
Submodule is merged, for the initial of the evidence according to D-S evidence theory and each possible breakdown reason
Probable value, carries out fusion treatment to the phenomenon of the failure, draws the fusion results and the conflict coefficient.
Wherein, the fusion submodule is specifically for passing through formulaTo the event
Hinder phenomenon and carry out fusion treatment, draw the fusion results and the conflict coefficient;
Wherein, C ∈ 2U, Ai∈2U, Bj∈2U, φ represents empty set, and U is the set of all possible breakdown reasons, and defines U and be
One identification framework of the possible breakdown reason, 2UThe set constituted for U all subsets, 2U→ [0,1], m is the base on U
This probability assignment function, wherein including the probable value corresponding to the corresponding all possible breakdown reasons of the phenomenon of the failure, m1,
m2According to predetermined probabilities allocation rule it is respectively basic probability assignment function that two phenomena of the failure are distributed for the identification framework,
M (C) represents the fusion results, Represent the conflict coefficient, AiAnd BjRepresent two institutes
State phenomenon of the failure and distinguish corresponding one group of possible breakdown reason.
Wherein, second acquisition module includes:
First acquisition submodule, the basic probability assignment function m for obtaining two phenomena of the failure respectively1And m2;
Calculating sub module, for obtaining the different information between described two phenomena of the failure by equation below:
Wherein, dBPARepresent the difference between described two phenomena of the failure
Different information,2M × 2M matrix is represented, M is that the element in the total number of all possible breakdown reasons, matrix isA and B represent that two phenomena of the failure distinguish corresponding all possible breakdown reasons, and A and B sums are
M。
Wherein, the correcting module includes:
Second acquisition submodule, for according to the different information, the conflict coefficient and equation below, obtaining after amendment
Conflict coefficient;
Represent the conflict coefficient, dBPA(m1,m2) represent described in two
Different information between phenomenon of the failure,Represent conflict coefficient after the amendment;
Correct submodule, if for conflict coefficient after the amendment be less than predetermined threshold value, using the fusion results as
The fault diagnosis result, otherwise determines that the fusion results are insincere.
The embodiment of the present invention has the advantages that:
The communication failure diagnostic method of the embodiment of the present invention, obtains at least two failures included in fault message and shows
As, and obtain the possible breakdown reason corresponding to each phenomenon of the failure;According to the evidence of D-S evidence theory and possibility
Failure cause, carries out fusion treatment to phenomenon of the failure, draws fusion results and conflict coefficient;According to possible breakdown reason, obtain
Different information between phenomenon of the failure;According to different information and the conflict coefficient, the fusion results are modified, and root
Fault diagnosis result is drawn according to revised fusion results, the existing fault diagnosis side based on D-S evidence theory is efficiently solved
Method, when evidences conflict is higher, the problem of fusion diagnosis result is inaccurate.
Brief description of the drawings
Fig. 1 is the Troubleshooting Flowchart based on D-S evidence theory in the prior art;
Fig. 2 is the workflow diagram of the communication failure diagnostic method of the embodiment of the present invention;
Fig. 3 is the structured flowchart of the communication failure diagnostic device of the embodiment of the present invention.
Embodiment
To make the technical problem to be solved in the present invention, technical scheme and advantage clearer, below in conjunction with specific implementation
Example and accompanying drawing are described in detail.
It is right first to make those skilled in the art better understood when the communication failure diagnostic method of the embodiment of the present invention
D-S evidence theory is done as described below.
D-S evidence theory is the mapping between proposition and set, mainly by the way that the uncertainty of proposition is switched into set
It is uncertain to be handled.That is the mainly result (evidence) that is occurred according to event, find out event generation main cause (assuming that).Such as
Shown in Fig. 1, in a communications system, failure symptom is that many possible failures are produced, the failure below each failure symptom
There is certain probability of happening, we represent the size of probability with belief function.When failure occurs failure is carried out using D-S theories
Information fusion, show that symptom is belonging respectively to the belief function of certain class failure, communication system is determined further according to the decision criteria of science
The type of failure, is finally sent to communication common carrier Breakdown Maintenance system by the failure determination result drawn and is handled in real time.
If U is the set of all values of variable X, meanwhile, each element mutual exclusion in set U then calls U one of X
Identification framework, is 2 with U power setUConstitute proposition collection.When first prime number in set U is n, the space representated by proposition collection
Size is 2n。
Defining 1 makes U be identification framework Θ, 2UThe set constituted for U all subsets,If set function m:2U→
[0,1] meet:
M (φ)=0
Then m is called the basic probability assignment function on U, and m (A) is referred to as A elementary probability number, represents the accurate trust to A.
One elementary tactics of evidence theory is that evidence set is divided into two or more incoherent parts, and is distinguished using them
Framework of identification is independently judged, then they combined with Dempster rules of combination.
The existing fusion method based on D-S evidence theory, generally measures the conflict between each evidence with conflict coefficient k
Degree, but this method is found under study for action, the conflict coefficient based on classical D-S rules of combination can not characterize each exactly
Conflict spectrum between evidence, and then cause the inaccurate of communication system failure diagnostic result.
Example 1:If identification framework U={ A, B, C }, two basic probability assignment functions of system are as follows:
m1:m1(A)=0.99, m1(B)=0.01;
m2:m2(B)=0.01, m2(C)=0.09;
The m it can be seen from two evidences of system1And m2Between be height conflict.According to D-S rules of combination to two
The result of combining evidences is:
K=0.99, m (A)=m (C)=0, m (B)=1
Although m1And m2Support to B is not high, i.e. the degree of belief of proposition B is not high, but last result is really proposition
B is provided with the degree of belief of maximum after combining, and this is clearly incorrect.
In actual periodic traffic information fusion system, because environment or human factor usually can export some equipment
The information runed counter to actual conditions, these information often conflict larger with the correct information that miscellaneous equipment is exported.How in evidence
The problem of effective integration that multi-source information is realized under highly conflicting is one in the urgent need to address.
Therefore, The embodiment provides a kind of communication failure diagnostic method and device, to solve existing be based on
The method for diagnosing faults of D-S evidence theory, when evidences conflict is higher, the problem of fusion diagnosis result is inaccurate.
First embodiment:
As shown in Fig. 2 the communication failure diagnostic method of the embodiment of the present invention, including:
Step 21:At least two phenomena of the failure included in fault message are obtained, and obtain each phenomenon of the failure
Corresponding possible breakdown reason.
In a particular embodiment of the present invention, after at least two phenomena of the failure in obtaining fault message, according to failure
The failure cause data stored in storehouse, obtain the possible breakdown reason corresponding to each phenomenon of the failure.
Step 22:According to the evidence of D-S evidence theory and the possible breakdown reason, phenomenon of the failure is carried out
Fusion treatment, draws fusion results and conflict coefficient, fusion results, which include each possible breakdown reason, turns into true fault reason
Probable value, conflict coefficient is used to represent conflict spectrum between fusion results and true fault reason.
Here, by the evidence and the possible breakdown reason of D-S evidence theory, fusion results and punching are drawn
Prominent coefficient, but the conflict coefficient can not characterize conflict spectrum between each evidence (phenomenon of the failure) exactly, it is necessary in the punching
Obtain preparing the new conflict coefficient of conflict spectrum between sign each evidence on the basis of prominent coefficient, and fusion results are entered
Row amendment.Because the conflict spectrum between each evidence is bigger, show that the credible result degree after D-S fusions is lower;Each card
Conflict spectrum between is smaller, shows that the credible result degree after D-S fusions is higher, so by entering to D-S fusion results
Row rational modification, it becomes possible to can effectively reduce the influence of " bad evidence " to the final fusion results of system, improves communication system event
Hinder the accuracy of diagnosis.
Step 23:According to the possible breakdown reason, the different information between the phenomenon of the failure is obtained.
Under evidence theory framework, the main cause highly conflicted between each evidence is caused to include:1st, framework of identification is imperfect.
Such as, in a communications system, for the failure that is potentially present of can not completely known to my technical staff, that is, system failure
Storehouse is not special comprehensive.Assuming that having 3 failures A, B, C in fault database, then framework of identification can only be this 3 elementary events
Power set, and actual failure is D, in this case, each device report result of system highly conflicts, and system is possible to
The fusion results that must be made mistake;2nd, the reliability of system equipment itself.Such as equipment is receiving the dry of environment or human factor
Disturb so that the judgement of equipment can result from the situation that actual result is runed counter to, and so be easier to cause the height between each evidence
Conflict.In summary, all evidences conflict degree accurately to be measured, on evidence between otherness be to be ignored
's.In a particular embodiment of the present invention, the different information between the phenomenon of the failure is obtained by distance function.
Step 24:According to the different information and the conflict coefficient, the fusion results are modified, and according to repairing
Fusion results after just draw fault diagnosis result.
According to the different information and the conflict coefficient, the fusion results are modified, and it is soft by matlab
Part is emulated, and output result shows by considering to the otherness between each evidence and compatibility, using improved card
The various PRELIMINARY RESULTSs diagnosed according to theory to communication system failure carry out fusion decision-making according to algorithm, can effectively reduce bad card
According to the influence to diagnostic result, the theoretical deficiencies of classical D-S are compensate for a certain extent, improve communication system failure diagnosis
Accuracy.
The communication failure diagnostic method of the embodiment of the present invention, obtains at least two failures included in fault message and shows
As, and obtain the possible breakdown reason corresponding to each phenomenon of the failure;According to the evidence of D-S evidence theory and possibility
Failure cause, carries out fusion treatment to phenomenon of the failure, draws fusion results and conflict coefficient;According to possible breakdown reason, obtain
Different information between phenomenon of the failure;According to different information and the conflict coefficient, the fusion results are modified, and root
Fault diagnosis result is drawn according to revised fusion results, the existing fault diagnosis side based on D-S evidence theory is efficiently solved
Method, when evidences conflict is higher, the problem of fusion diagnosis result is inaccurate.
Further, the evidence and the possible breakdown reason according to D-S evidence theory, to the event
Hinder phenomenon and carry out fusion treatment, draw fusion results and conflict coefficient, including:
It is each possible breakdown reason distribution one corresponding to each phenomenon of the failure according to predetermined probabilities allocation rule
Probability values.
In a particular embodiment of the present invention, mainly by an identification framework of possible breakdown reason according to predetermined probabilities point
It is that each possible breakdown reason corresponding to each phenomenon of the failure distributes a probability values with rule.
According to the evidence of D-S evidence theory and the probability values of each possible breakdown reason, to the event
Hinder phenomenon and carry out fusion treatment, draw the fusion results and the conflict coefficient.
Specifically, passing through formulaFusion treatment is carried out to the phenomenon of the failure,
Draw the fusion results and the conflict coefficient;
Wherein, C ∈ 2U, Ai∈2U, Bj∈2U, φ represents empty set, and U is the set of all possible breakdown reasons, and defines U and be
One identification framework of the possible breakdown reason, 2UThe set constituted for U all subsets, 2U→ [0,1], m is the base on U
This probability assignment function, wherein including the probable value corresponding to the corresponding all possible breakdown reasons of the phenomenon of the failure, m1,
m2According to predetermined probabilities allocation rule it is respectively basic probability assignment function that two phenomena of the failure are distributed for the identification framework,
M (C) represents the fusion results, Represent the conflict coefficient, AiAnd BjRepresent two institutes
State phenomenon of the failure and distinguish corresponding one group of possible breakdown reason.
Further, it is described according to the possible breakdown reason, obtain the different information between the phenomenon of the failure, bag
Include:
The basic probability assignment function m of two phenomena of the failure is obtained respectively1And m2;
Different information between described two phenomena of the failure is obtained by equation below:
Wherein, dBPARepresent the difference between described two phenomena of the failure
Different information,2M × 2M matrix is represented, M is that the element in the total number of all possible breakdown reasons, matrix isA and B represent that two phenomena of the failure distinguish corresponding all possible breakdown reasons, and A and B sums are
M。
In a particular embodiment of the present invention, if Θ is a standard framework of identification for including M proposition, m1And m2It is to distinguish
Know 2 basic probability assignment BPA on framework Θ, use EPTo represent the space of all subset generations on framework of identification Θ.One
BPA represents to be one in EPUpper coordinate system is m (Ai) vectorial m, and pass through above-mentioned distance function dBPACalculate m1And m2Away from
From the distance function can reasonably represent the otherness between evidence.
Further, it is described according to the different information and the conflict coefficient, the fusion results are modified, and
Fault diagnosis result is drawn according to revised fusion results, including:
According to the different information, the conflict coefficient and equation below, conflict coefficient after amendment is obtained;
Represent the conflict coefficient, dBPA(m1,m2) represent described in two
Different information between phenomenon of the failure,Represent conflict coefficient after the amendment;
If conflict coefficient is less than predetermined threshold value after the amendment, the fusion results are regard as the fault diagnosis knot
Really, otherwise determine that the fusion results are insincere.
In a particular embodiment of the present invention, after amendment the size of conflict coefficient depend on classical prominent coefficient k and evidence away from
From dBPAThe collective effect of two parameters.Only classics conflict coefficient k and evidence distance dBPATwo values of consult volume all than it is larger when,
System can judge that the conflict between evidence is larger by new conflict coefficient;Similarly, classical conflict coefficient k and evidence away from
From dBPAWhen being all zero, show on evidence between do not conflict.
Example 2:If two basic probability assignment functions of identification framework U={ A, B, C, D } system are as follows:
m1{ A, D }=1
m2{ A, B, C }=1
The distance between evidence dBPAWith conflict coefficient k according to formulaResult of calculation
It is as follows:
k12=0;dBPA=0.866
Conflict coefficient is after amendment:
Conflict coefficient illustrates after amendment, two evidence m1And m2Between exist conflict, the direct result one with two evidences
Cause.Analysis example 1 is understood in this way, the distance between evidence dBPAIt is as follows with conflict coefficient k:
k12=0.9999;dBPA=0.99
New conflict coefficient is:
Required result illustrates that the conflict of two evidences is very high, and is characterized unanimously with two evidences.
IfThen illustrate it is complete conflict between evidence.
Below by example 3 a comprehensively contrast will be done to conflict coefficient after amendment and classical conflict coefficient.
Example 3:If identification framework U={ a1,a2,a3,…,a20Two basic probability assignments of system are as follows:
m1{a7}=0.1
m1{ A }=0.9
m2{a1,a2,a3,a4,a5}=1
A is according to { a in formula1, { a1,a2, { a1,a2,a3..., { a1,a2,a3,…,a20Conversion.Table 1 gives classical punching
Prominent coefficient k, evidence distance dBPAWith the change of conflict coefficient after amendment:
For example | dBPA | kd | k |
{a1} | 0.854 | 0.477 | 0.1 |
{a1,a2} | 0.742 | 0.421 | 0.1 |
{a1,…,a3} | 0.609 | 0.354 | 0.1 |
{a1,…,a4} | 0.436 | 0.268 | 0.1 |
{a1,…,a5} | 0.1 | 0.1 | 0.1 |
{a1,…,a6} | 0.4 | 0.25 | 0.1 |
{a1,…,a7} | 0.529 | 0.315 | 0.1 |
{a1,…,a8} | 0.599 | 0.350 | 0.1 |
{a1,…,a9} | 0.648 | 0.374 | 0.1 |
{a1,…,a10} | 0.685 | 0.392 | 0.1 |
{a1,…,a11} | 0.714 | 0.407 | 0.1 |
{a1,…,a12} | 0.737 | 0.418 | 0.1 |
{a1,…,a13} | 0.756 | 0.428 | 0.1 |
{a1,…,a14} | 0.771 | 0.436 | 0.1 |
{a1,…,a15} | 0.785 | 0.443 | 0.1 |
{a1,…,a16} | 0.797 | 0.448 | 0.1 |
{a1,…,a17} | 0.807 | 0.453 | 0.1 |
{a1,…,a18} | 0.816 | 0.458 | 0.1 |
{a1,…,a19} | 0.823 | 0.462 | 0.1 |
{a1,…,a20} | 0.830 | 0.465 | 0.1 |
Table 1
Interpretation of result:As shown in Table 1, in subset A={ a1,…,a5When amendment after conflict coefficient reach minimum value, i.e., respectively
Conflict between individual evidence is minimum.In this example, evidence m1With evidence m2Consider from classical conflict, it is basic between them
It is upper that conflict is not present.New conflict coefficient had both considered the compatibility between evidence it is contemplated that difference between each evidence
Property, caused conflict spectrum change reflecting strictly according to the facts when can subset A be changed is combined reality logical derivation.After amendment
Conflict coefficient span consistent with classical conflict coefficient span is [0,1], and represented meaning is same, and the numerical value is got over
Greatly, the conflict between each evidence is bigger;On the contrary, the numerical value is smaller, the conflict between each evidence is also smaller.
The conflict spectrum between each evidence is generally measured with conflict coefficient k, but this method is found under study for action, base
The conflict spectrum between each evidence can not be characterized exactly in the conflict coefficient of classical D-S rules of combination.The present invention is implemented
Example proposes a kind of method that new conflict coefficient is represented in classical D-S theoretical foundations, and conflict coefficient is represented using this method
To the valid metric of conflict, fusion results are modified.Conflict spectrum between each evidence is bigger, shows that D-S merges it
Credible result degree afterwards is lower;Conflict spectrum between each evidence is smaller, shows that the credible result degree after D-S fusions is got over
Height, so by carrying out rational modification to D-S fusion results, it becomes possible to can effectively reduce " bad evidence " to the final fusion of system
As a result influence.
By considering to the otherness between each evidence and compatibility, using improved evidence theory to communication
The various PRELIMINARY RESULTSs of system fault diagnosis carry out fusion decision-making according to algorithm, can effectively reduce bad evidence to diagnostic result
Influence, the theoretical deficiency of classical D-S is compensate for a certain extent, the accuracy of communication system failure diagnosis is improved.
Second embodiment:
As shown in figure 3, embodiments of the invention additionally provide a kind of communication failure diagnostic device, including:
First acquisition module 31, for obtaining at least two phenomena of the failure included in fault message, and obtains each
Possible breakdown reason corresponding to the phenomenon of the failure;
Fusion Module 32, for the evidence according to D-S evidence theory and the possible breakdown reason, to described
Phenomenon of the failure carries out fusion treatment, draws fusion results and conflict coefficient, and the fusion results include each possible breakdown reason
Probable value as true fault reason, the conflict coefficient is used to represent between the fusion results and true fault reason
Conflict spectrum;
Second acquisition module 33, for according to the possible breakdown reason, obtaining the difference letter between the phenomenon of the failure
Breath;
Correcting module 34, for according to the different information and the conflict coefficient, being modified to the fusion results,
And draw fault diagnosis result according to revised fusion results.
The communication failure diagnostic device of the embodiment of the present invention, the Fusion Module 32 includes:
Distribution sub module 321, for being each corresponding to each phenomenon of the failure according to predetermined probabilities allocation rule
Possible breakdown reason distributes a probability values;
Merge submodule 322, for the evidence according to D-S evidence theory and each possible breakdown reason just
Beginning probable value, carries out fusion treatment to the phenomenon of the failure, draws the fusion results and the conflict coefficient.
The communication failure diagnostic device of the embodiment of the present invention, the fusion submodule 322 is specifically for passing through formulaFusion treatment is carried out to the phenomenon of the failure, the fusion results and the punching is drawn
Prominent coefficient;
Wherein, C ∈ 2U, Ai∈2U, Bj∈2U, φ represents empty set, and U is the set of all possible breakdown reasons, and defines U and be
One identification framework of the possible breakdown reason, 2UThe set constituted for U all subsets, 2U→ [0,1], m is the base on U
This probability assignment function, wherein including the probable value corresponding to the corresponding all possible breakdown reasons of the phenomenon of the failure, m1,
m2According to predetermined probabilities allocation rule it is respectively basic probability assignment function that two phenomena of the failure are distributed for the identification framework,
M (C) represents the fusion results, Represent the conflict coefficient, AiAnd BjRepresent two institutes
State phenomenon of the failure and distinguish corresponding one group of possible breakdown reason.
The communication failure diagnostic device of the embodiment of the present invention, second acquisition module 33 includes:
First acquisition submodule 331, the basic probability assignment function m for obtaining two phenomena of the failure respectively1And m2;
Calculating sub module 332, for obtaining the different information between described two phenomena of the failure by equation below:
Wherein, dBPARepresent the difference between described two phenomena of the failure
Different information,2M × 2M matrix is represented, M is that the element in the total number of all possible breakdown reasons, matrix isA and B represent that two phenomena of the failure distinguish corresponding all possible breakdown reasons, and A and B sums are
M。
The communication failure diagnostic device of the embodiment of the present invention, the correcting module 34 includes:
Second acquisition submodule 341, for according to the different information, the conflict coefficient and equation below, acquisition to be repaiied
Conflict coefficient after just;
Represent the conflict coefficient, dBPA(m1,m2) represent two events
Hinder the different information between phenomenon,Represent conflict coefficient after the amendment;
Submodule 342 is corrected, if being less than predetermined threshold value for conflict coefficient after the amendment, the fusion results are made
For the fault diagnosis result, otherwise determine that the fusion results are insincere.
It should be noted that the device is device corresponding with above method embodiment, own in above method embodiment
Implementation can also reach identical technique effect suitable for the embodiment of the device.
The communication failure diagnostic method and device of the embodiment of the present invention, obtain at least two events included in fault message
Hinder phenomenon, and obtain the possible breakdown reason corresponding to each phenomenon of the failure;According to the evidence of D-S evidence theory and
Possible breakdown reason, carries out fusion treatment to phenomenon of the failure, draws fusion results and conflict coefficient;According to possible breakdown reason,
Obtain the different information between phenomenon of the failure;According to different information and the conflict coefficient, the fusion results are modified,
And fault diagnosis result is drawn according to revised fusion results, efficiently solve the existing failure based on D-S evidence theory and examine
Disconnected method, when evidences conflict is higher, the problem of fusion diagnosis result is inaccurate.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
God is with principle, and any modification, equivalent substitution and improvements made etc. should be included in the scope of the protection.
Claims (10)
1. a kind of communication failure diagnostic method, it is characterised in that including:
At least two phenomena of the failure included in fault message are obtained, and obtain the possibility corresponding to each phenomenon of the failure
Failure cause;
According to the evidence of D-S evidence theory and the possible breakdown reason, the phenomenon of the failure is carried out at fusion
Reason, draws fusion results and conflict coefficient, the fusion results, which include each possible breakdown reason, turns into true fault reason
Probable value, the conflict coefficient is used to represent the conflict spectrum between the fusion results and true fault reason;
According to the possible breakdown reason, the different information between the phenomenon of the failure is obtained;
According to the different information and the conflict coefficient, the fusion results are modified, and according to revised fusion
As a result fault diagnosis result is drawn.
2. communication failure diagnostic method according to claim 1, it is characterised in that the card according to D-S evidence theory
According to rule of combination and the possible breakdown reason, fusion treatment is carried out to the phenomenon of the failure, fusion results and conflict system is drawn
Number, including:
It is that each possible breakdown reason distribution one corresponding to each phenomenon of the failure is initial according to predetermined probabilities allocation rule
Probable value;
It is existing to the failure according to the evidence of D-S evidence theory and the probability values of each possible breakdown reason
As carrying out fusion treatment, the fusion results and the conflict coefficient are drawn.
3. communication failure diagnostic method according to claim 2, it is characterised in that the card according to D-S evidence theory
According to rule of combination and the probability values of each possible breakdown reason, fusion treatment is carried out to the phenomenon of the failure, drawn described
Fusion results and the conflict coefficient, including:
Pass through formulaFusion treatment is carried out to the phenomenon of the failure, the fusion is drawn
And the conflict coefficient as a result;
Wherein, C ∈ 2U, Ai∈2U, Bj∈2U, φ represents empty set, and U is the set of all possible breakdown reasons, and it is described to define U
One identification framework of possible breakdown reason, 2UThe set constituted for U all subsets, 2U→ [0,1], m is substantially general on U
Rate assignment function, wherein including the probable value corresponding to the corresponding all possible breakdown reasons of the phenomenon of the failure, m1, m2For
The identification framework is according to the basic probability assignment function that predetermined probabilities allocation rule is respectively that two phenomena of the failure are distributed, m (C)
The fusion results are represented, Represent the conflict coefficient, AiAnd BjRepresent two events
Hinder phenomenon and distinguish corresponding one group of possible breakdown reason.
4. communication failure diagnostic method according to claim 3, it is characterised in that described former according to the possible breakdown
Cause, obtains the different information between the phenomenon of the failure, including:
The basic probability assignment function m of two phenomena of the failure is obtained respectively1And m2;
Different information between described two phenomena of the failure is obtained by equation below:
Wherein, dBPARepresent the difference letter between described two phenomena of the failure
Breath,2M × 2M matrix is represented, M is that the element in the total number of all possible breakdown reasons, matrix isA and B represent that two phenomena of the failure distinguish corresponding all possible breakdown reasons, and A and B sums are
M。
5. communication failure diagnostic method according to claim 4, it is characterised in that described according to the different information and institute
Conflict coefficient is stated, the fusion results are modified, and fault diagnosis result is drawn according to revised fusion results, is wrapped
Include:
According to the different information, the conflict coefficient and equation below, conflict coefficient after amendment is obtained;
Represent the conflict coefficient, dBPA(m1,m2) represent that two failures show
Different information as between,Represent conflict coefficient after the amendment;
If conflict coefficient is less than predetermined threshold value after the amendment, no using the fusion results as the fault diagnosis result
Then determine that the fusion results are insincere.
6. a kind of communication failure diagnostic device, it is characterised in that including:
First acquisition module, for obtaining at least two phenomena of the failure included in fault message, and obtains each event
Hinder the possible breakdown reason corresponding to phenomenon;
Fusion Module, it is existing to the failure for the evidence according to D-S evidence theory and the possible breakdown reason
As carrying out fusion treatment, fusion results and conflict coefficient are drawn, the fusion results, which include each possible breakdown reason, turns into true
The probable value of real failure cause, the conflict coefficient is used to represent the journey that conflicts between the fusion results and true fault reason
Degree;
Second acquisition module, for according to the possible breakdown reason, obtaining the different information between the phenomenon of the failure;
Correcting module, for according to the different information and the conflict coefficient, being modified to the fusion results, and according to
Revised fusion results draw fault diagnosis result.
7. communication failure diagnostic device according to claim 6, it is characterised in that the Fusion Module includes:
Distribution sub module, for being each possible event corresponding to each phenomenon of the failure according to predetermined probabilities allocation rule
Hinder reason and distribute a probability values;
Submodule is merged, for the evidence according to D-S evidence theory and the probability of each possible breakdown reason
Value, carries out fusion treatment to the phenomenon of the failure, draws the fusion results and the conflict coefficient.
8. communication failure diagnostic device according to claim 7, it is characterised in that the fusion submodule is specifically for logical
Cross formulaTo the phenomenon of the failure carry out fusion treatment, draw the fusion results and
The conflict coefficient;
Wherein, C ∈ 2U, Ai∈2U, Bj∈2U, φ represents empty set, and U is the set of all possible breakdown reasons, and it is described to define U
One identification framework of possible breakdown reason, 2UThe set constituted for U all subsets, 2U→ [0,1], m is substantially general on U
Rate assignment function, wherein including the probable value corresponding to the corresponding all possible breakdown reasons of the phenomenon of the failure, m1, m2For
The identification framework is according to the basic probability assignment function that predetermined probabilities allocation rule is respectively that two phenomena of the failure are distributed, m (C)
The fusion results are represented, Represent the conflict coefficient, AiAnd BjRepresent two events
Hinder phenomenon and distinguish corresponding one group of possible breakdown reason.
9. communication failure diagnostic device according to claim 8, it is characterised in that second acquisition module includes:
First acquisition submodule, the basic probability assignment function m for obtaining two phenomena of the failure respectively1And m2;
Calculating sub module, for obtaining the different information between described two phenomena of the failure by equation below:
Wherein, dBPARepresent the difference letter between described two phenomena of the failure
Breath,2M × 2M matrix is represented, M is that the element in the total number of all possible breakdown reasons, matrix isA and B represent that two phenomena of the failure distinguish corresponding all possible breakdown reasons, and A and B sums are
M。
10. communication failure diagnostic device according to claim 9, it is characterised in that the correcting module includes:
Second acquisition submodule, conflicts for according to the different information, the conflict coefficient and equation below, obtaining after amendment
Coefficient;
Represent the conflict coefficient, dBPA(m1,m2) represent that two failures show
Different information as between,Represent conflict coefficient after the amendment;
Submodule is corrected, if being less than predetermined threshold value for conflict coefficient after the amendment, using the fusion results as described
Fault diagnosis result, otherwise determines that the fusion results are insincere.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610164660.2A CN107222322A (en) | 2016-03-22 | 2016-03-22 | A kind of communication failure diagnostic method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610164660.2A CN107222322A (en) | 2016-03-22 | 2016-03-22 | A kind of communication failure diagnostic method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107222322A true CN107222322A (en) | 2017-09-29 |
Family
ID=59927267
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610164660.2A Pending CN107222322A (en) | 2016-03-22 | 2016-03-22 | A kind of communication failure diagnostic method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107222322A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109460356A (en) * | 2018-10-19 | 2019-03-12 | 中国船舶重工集团公司第七六研究所 | A kind of data fusion method for software fault prediction |
CN110045672A (en) * | 2018-01-15 | 2019-07-23 | 中冶长天国际工程有限责任公司 | A kind of method for diagnosing faults and device of belt feeder |
CN111078456A (en) * | 2019-12-26 | 2020-04-28 | 新奥数能科技有限公司 | Equipment fault diagnosis method and device, computer readable storage medium and electronic equipment |
CN111445037A (en) * | 2020-03-17 | 2020-07-24 | 中国船舶重工集团公司第七0三研究所 | Steam turbine fault diagnosis method, system and equipment based on evidence theory |
CN111521937A (en) * | 2020-04-29 | 2020-08-11 | 粤电集团贵州有限公司 | Method and device for diagnosing starting fault of gas generator |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102208028A (en) * | 2011-05-31 | 2011-10-05 | 北京航空航天大学 | Fault predicting and diagnosing method suitable for dynamic complex system |
CN103120113A (en) * | 2011-11-18 | 2013-05-29 | 江南大学 | Small-sized area irrigation system based on ZigBee |
CN103577707A (en) * | 2013-11-15 | 2014-02-12 | 上海交通大学 | Robot failure diagnosis method achieved by multi-mode fusion inference |
CN103871004A (en) * | 2014-03-31 | 2014-06-18 | 国家电网公司 | Power distribution network failure cause analyzing method based on expert system and D-S evidence theory |
CN104931857A (en) * | 2015-06-25 | 2015-09-23 | 山东大学 | Power distribution network fault locating method based on D-S evidence theory |
CN105373700A (en) * | 2015-10-30 | 2016-03-02 | 哈尔滨工程大学 | Method for mechanical fault diagnosis based on information entropies and evidence theory |
-
2016
- 2016-03-22 CN CN201610164660.2A patent/CN107222322A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102208028A (en) * | 2011-05-31 | 2011-10-05 | 北京航空航天大学 | Fault predicting and diagnosing method suitable for dynamic complex system |
CN103120113A (en) * | 2011-11-18 | 2013-05-29 | 江南大学 | Small-sized area irrigation system based on ZigBee |
CN103577707A (en) * | 2013-11-15 | 2014-02-12 | 上海交通大学 | Robot failure diagnosis method achieved by multi-mode fusion inference |
CN103871004A (en) * | 2014-03-31 | 2014-06-18 | 国家电网公司 | Power distribution network failure cause analyzing method based on expert system and D-S evidence theory |
CN104931857A (en) * | 2015-06-25 | 2015-09-23 | 山东大学 | Power distribution network fault locating method based on D-S evidence theory |
CN105373700A (en) * | 2015-10-30 | 2016-03-02 | 哈尔滨工程大学 | Method for mechanical fault diagnosis based on information entropies and evidence theory |
Non-Patent Citations (1)
Title |
---|
蒋雯: "基于新的证据冲突表示的信息融合方法研究", 《西北工业大学学报》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110045672A (en) * | 2018-01-15 | 2019-07-23 | 中冶长天国际工程有限责任公司 | A kind of method for diagnosing faults and device of belt feeder |
CN109460356A (en) * | 2018-10-19 | 2019-03-12 | 中国船舶重工集团公司第七六研究所 | A kind of data fusion method for software fault prediction |
CN109460356B (en) * | 2018-10-19 | 2021-12-28 | 中国船舶重工集团公司第七一六研究所 | Data fusion method for software fault prediction |
CN111078456A (en) * | 2019-12-26 | 2020-04-28 | 新奥数能科技有限公司 | Equipment fault diagnosis method and device, computer readable storage medium and electronic equipment |
CN111078456B (en) * | 2019-12-26 | 2023-05-16 | 新奥数能科技有限公司 | Device fault diagnosis method, device, computer readable storage medium and electronic device |
CN111445037A (en) * | 2020-03-17 | 2020-07-24 | 中国船舶重工集团公司第七0三研究所 | Steam turbine fault diagnosis method, system and equipment based on evidence theory |
CN111521937A (en) * | 2020-04-29 | 2020-08-11 | 粤电集团贵州有限公司 | Method and device for diagnosing starting fault of gas generator |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107222322A (en) | A kind of communication failure diagnostic method and device | |
CN109522746A (en) | A kind of data processing method, electronic equipment and computer storage medium | |
CN104111920B (en) | A kind of Forecasting Methodology and device based on decision tree | |
CN107704625A (en) | Fields match method and apparatus | |
CN106202886A (en) | Track circuit red band Fault Locating Method based on fuzzy coarse central Yu decision tree | |
CN104021248B (en) | A kind of airborne machinery product FMECA analysis methods | |
CN109060398A (en) | A kind of multi-source information equipment fault diagnosis method | |
CN105677563A (en) | Integrated test sequence generating method based on SysML module diagrams | |
CN106932670A (en) | A kind of distribution power automation terminal method for diagnosing status based on D S evidence theories | |
CN108398939A (en) | A kind of method for diagnosing faults based on DS evidence theories | |
CN108776861A (en) | Railway Communication safety risk estimating method and device | |
CN107124325A (en) | A kind of power communication network operation safety evaluation method and system | |
CN107609325A (en) | The method that fault tree based on SAT solves minimal cut set | |
CN109299846A (en) | A kind of nuclear power plant equipment analysis method for reliability, system and terminal device | |
CN104574141A (en) | Service influence degree analysis method | |
CN109670610A (en) | Fault diagnosis optimization method based on fault propagation analysis | |
CN104181912A (en) | Method for adaptively translating vehicle fault codes by vehicle diagnosing instrument | |
CN111445037B (en) | Steam turbine fault diagnosis method, system and equipment based on evidence theory | |
CN112562870A (en) | Evaluation and early warning system for enhancing mutual trust of doctors and patients and reducing doctor-patient contradiction | |
CN108304831A (en) | A kind of method and device that monitoring worker safety helmet is worn | |
CN103365777A (en) | Test case generation system and system | |
CN106330429A (en) | Generation method and device for S box of SM4 algorithm | |
CN107579844A (en) | It is a kind of that failure method for digging is dynamically associated based on service path and frequency matrix | |
CN110277131A (en) | Method of calibration, terminal device and storage medium based on NAND FLASH memory | |
Holland et al. | The statistical analysis of local structure in social networks |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170929 |