CN109598282A - Aerial drainage induces hydro-structure damage diagnosis method and device - Google Patents
Aerial drainage induces hydro-structure damage diagnosis method and device Download PDFInfo
- Publication number
- CN109598282A CN109598282A CN201811208354.XA CN201811208354A CN109598282A CN 109598282 A CN109598282 A CN 109598282A CN 201811208354 A CN201811208354 A CN 201811208354A CN 109598282 A CN109598282 A CN 109598282A
- Authority
- CN
- China
- Prior art keywords
- damage
- evidence
- variety
- aerial drainage
- hydro
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a kind of aerial drainages to induce hydro-structure damage diagnosis method and device, wherein this method comprises: a variety of damage criterions of acquisition aerial drainage structure；It is identified according to every kind of damage criterion of a variety of damage criterions and carries out non-destructive tests, to obtain a variety of initial damage recognition results；It is associated according to a variety of initial damage recognition results and decision level fusion, to obtain last diagnostic result by data fusion.This method proposes the blending decision method of multi objective, can carry out integrated treatment to a variety of uncertain informations, accurately reflect the operating status of aerial drainage structure, the position that quickly discovery and identification of damage occur.
Description
Technical field
The present invention relates to hydropower technology field, in particular to a kind of aerial drainage induces hydro-structure damage diagnosis method and dress
It sets.
Background technique
China is the first big country of world's water power, built or in the high dam built, and height of dam, discharge, flood discharge power are all
More than the current highest level in the world.These engineerings have that " head is high, flow is big, flood discharge power is big, river valley is narrow, geology mostly
The features such as complicated condition ", when discharge structure flood discharge, the release floodwatering flow of strong turbulent fluctuation often carries huge energy, if
Design and operation aspect processing are careless, it is most likely that the judder for causing works even results in structural damage, thus shadow
The safe operation for ringing engineering, causes huge economic loss and social loss, will also seriously endanger the life security of the people.
Aerial drainage structure is easy to produce judder under the action of water flow load, and leads to fatigue rupture, according to statistics nearly 1/3 aerial drainage
There is different degrees of destruction in structure, therefore, diagnose to military service aerial drainage structural health operating condition very necessary.
Aerial drainage structure is in During Process of Long-term Operation, produced various damages under the action of hydrodynamic force, temperature and environmental load
It can all be embodied in structural vibration response and its modal parameter with defect, therefore, the work condition for carrying out aerial drainage structure is known
It is not monitored with dynamical health, is to analyze its damage status advantage means.The monitoring of aerial drainage structure is mainly answered with deformation, stress
Static, quasi-static monitoring based on change, seepage flow and seepage pressure, gap opening degree and temperature etc., the visualizer of static or quasi-static monitoring
Device is typically all to be influenced in construction time pre-buried aerial drainage structural body by datum mark variation and instrument monitoring range, cannot be fine
The degree of impairment occurred in ground reflection aerial drainage structure operational process.With working frequency, vibration amplitude, coefficient of kurtosis, the coefficient of skew,
Dynamic realtime monitoring based on amplitude ratio coefficient etc. compensates for the deficiencies of static monitoring techniques to a certain extent, but existing dynamic
Monitoring index cannot be well reflected operating status locating for structure.For example, due to the interference by noise, to aerial drainage arrangement works
Often there is a certain error for the calculating of frequency, prevents it from the working condition that accurately reflects aerial drainage structure, in early injury,
Its vibration amplitude does not exceed the upper limit value of permission generally, in addition, coefficient of kurtosis and coefficient of skew reflection be vibration data with
The departure degree of normal distribution, and the catastrophe of amplitude ratio coefficient reflection vibration, test prove, after damage occurs, structure
Vibratory response usually can still meet normal distribution, and the damage of aerial drainage structure is often Fatigue Failure Process, in response not
Apparent jumping phenomenon can occur, therefore, These parameters are not intuitive to degree of impairment, sensitive.
In the related art, each damage criterion can under certain condition carry out working condition locating for structure certain
Judgement, while part index number also can indicate that the damage position of structure.But due to by sensor self reason, work item
The influence of part, ambient noise and analysis method precision etc., single damage criterion are easy to produce the judging result of mistake.
Summary of the invention
The present invention is directed to solve at least some of the technical problems in related technologies.
For this purpose, an object of the present invention is to provide a kind of aerial drainages to induce hydro-structure damage diagnosis method, this method
It proposes the blending decision method of multi objective, integrated treatment can be carried out to a variety of uncertain informations, accurately reflect aerial drainage structure
Operating status, the position that quickly discovery and identification of damage occur.
It is another object of the present invention to propose a kind of aerial drainage induction hydro-structure damage diagnostic device.
In order to achieve the above objectives, one aspect of the present invention embodiment proposes a kind of aerial drainage induction hydro-structure damage diagnosis side
Method, comprising the following steps: a variety of damage criterions of acquisition aerial drainage structure；According to every kind of damage criterion of a variety of damage criterions
Identification carries out non-destructive tests, to obtain a variety of initial damage recognition results；It is carried out according to a variety of initial damage recognition results
Association and decision level fusion, to obtain last diagnostic result by data fusion.
The aerial drainage of the embodiment of the present invention induces hydro-structure damage diagnosis method, by a variety of damages for acquiring aerial drainage structure
Index, and non-destructive tests are carried out to every kind of damage criterion identification and obtain initial damage recognition result, it is determined using the fusion of multi objective
Plan method obtains last diagnostic as a result, it is possible to carry out integrated treatment to a variety of uncertain informations, accurately reflects the fortune of aerial drainage structure
Row state, the position that quickly discovery and identification of damage occur.
In addition, aerial drainage according to the above embodiment of the present invention induce hydro-structure damage diagnosis method can also have it is following
Additional technical characteristic:
Further, in one embodiment of the invention, after a variety of damage criterions for acquiring the aerial drainage structure,
Further include: noise reduction, modal idenlification and/or feature extraction are filtered to a variety of damage criterions, it is more to obtain that treated
Kind damage data, to obtain institute according to a variety of damage datas and the corresponding damnification recognition method of preset every kind of damage criterion
State a variety of initial damage recognition results.
Further, in one embodiment of the invention, described to be carried out according to a variety of initial damage recognition results
Association and decision level fusion further comprise: by every kind of initial damage recognition result of a variety of initial damage recognition results
As evidence, to obtain average evidence；It is determined according to the Jousselme distance between every evidence and the average evidence described every
The weight coefficient of kind initial damage recognition result；Every evidence is weighted and averaged, to be obtained by Murphy improved method
Last diagnostic result.
Wherein, the calculation formula of the average evidence are as follows:
Wherein, U is the damage reason location index group number for health monitoring, mi(Aj) be each index recognition result；
The calculation formula of the Jousselme distance are as follows:
Wherein, M is average evidence mean value,Indicate two vectorsWithBetween inner product,
Indicate to
AmountInner product,Indicate two vectorsWithInner product；
Weighted average formula are as follows:
Wherein, ωiFor the weight of each evidence, U is the damage reason location index group number for health monitoring.
Further, in one embodiment of the invention, described to be averaged between evidence according to every evidence described
Jousselme distance determines the weight coefficient of every kind of initial damage recognition result, further comprises:
Define degree of belief function, the degree of belief function are as follows:
Wherein, DiFor the recognition result of each damage criterion and the Jousselme distance of average evidence mean value M；
The degree of belief function is normalized, to obtain the weight of every evidence, the weight formula
Are as follows:
Wherein, TiFor degree of belief function, U is the damage reason location index group number for health monitoring.
In order to achieve the above objectives, another aspect of the present invention embodiment proposes a kind of aerial drainage induction hydro-structure damage diagnosis
Device, comprising: acquisition module, for acquiring a variety of damage criterions of aerial drainage structure；Identification module, for according to a variety of damages
The every kind of damage criterion identification for hurting index carries out non-destructive tests, to obtain a variety of initial damage recognition results；Diagnostic module is used for
It is associated according to a variety of initial damage recognition results and decision level fusion, to obtain last diagnostic knot by data fusion
Fruit.
The aerial drainage of the embodiment of the present invention induces hydro-structure and damages diagnostic device, by a variety of damages for acquiring aerial drainage structure
Index, and non-destructive tests are carried out to every kind of damage criterion identification and obtain initial damage recognition result, it is determined using the fusion of multi objective
Plan method obtains last diagnostic as a result, it is possible to carry out integrated treatment to a variety of uncertain informations, accurately reflects the fortune of aerial drainage structure
Row state, the position that quickly discovery and identification of damage occur.
In addition, aerial drainage according to the above embodiment of the present invention induce hydro-structure damage diagnostic device can also have it is following
Additional technical characteristic:
Further, in one embodiment of the invention, the acquisition module is further used for a variety of damages
Index is filtered noise reduction, modal idenlification and/or feature extraction, with a variety of damage datas that obtain that treated, according to
A variety of damage datas and the corresponding damnification recognition method of preset every kind of damage criterion obtain a variety of initial damage identification knots
Fruit.
Further, in one embodiment of the invention, be further used for will be described a variety of initial for the diagnostic module
Every kind of initial damage recognition result of non-destructive tests result is as evidence, to obtain average evidence, and according to every evidence and institute
The weight coefficient that the Jousselme distance between average evidence determines every kind of initial damage recognition result is stated, and then to every
Evidence is weighted and averaged, to obtain last diagnostic result by Murphy improved method.
Wherein, the calculation formula of the average evidence are as follows:
Wherein, U is the damage reason location index group number for health monitoring, mi(Aj) be each index recognition result；
The calculation formula of the Jousselme distance are as follows:
Wherein, M is average evidence mean value,Indicate two vectorsWithBetween inner product,
Indicate to
AmountInner product,Indicate two vectorsWithInner product；
Weighted average formula are as follows:
Wherein, ωiFor the weight of each evidence, U is the damage reason location index group number for health monitoring.
Further, in one embodiment of the invention, described to be averaged between evidence according to every evidence described
Jousselme distance determines the weight coefficient of every kind of initial damage recognition result, further comprises:
Define degree of belief function, the degree of belief function are as follows:
Wherein, DiFor the recognition result of each damage criterion and the Jousselme distance of average evidence mean value M；
The degree of belief function is normalized, to obtain the weight of every evidence, the weight formula
Are as follows:
Wherein, TiFor degree of belief function, U is the damage reason location index group number for health monitoring.
The additional aspect of the present invention and advantage will be set forth in part in the description, and will partially become from the following description
Obviously, or practice through the invention is recognized.
Detailed description of the invention
Above-mentioned and/or additional aspect and advantage of the invention will become from the following description of the accompanying drawings of embodiments
Obviously and it is readily appreciated that, in which:
Fig. 1 is to induce hydro-structure damage diagnosis method flow chart according to the aerial drainage of one embodiment of the invention；
Fig. 2 is to induce hydro-structure damage diagnosis method decision level fusion flow chart according to the aerial drainage of the embodiment of the present invention；
Fig. 3 is to induce hydro-structure according to the aerial drainage of one embodiment of the invention to damage diagnostic device structural schematic diagram.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to is used to explain the present invention, and is not considered as limiting the invention.
Describe with reference to the accompanying drawings the aerial drainage proposed according to embodiments of the present invention induce hydro-structure damage diagnosis method and
Device describes the aerial drainage proposed according to embodiments of the present invention with reference to the accompanying drawings first and induces hydro-structure damage diagnosis method and dress
It sets.
Fig. 1 is that the aerial drainage of one embodiment of the invention induces hydro-structure damage diagnosis method flow chart.
As shown in Figure 1, the aerial drainage induce hydro-structure damage diagnosis method the following steps are included:
In step s101, a variety of damage criterions of aerial drainage structure are acquired.
Wherein, after a variety of damage criterions of acquisition aerial drainage structure, further includes: be filtered drop to a variety of damage criterions
It makes an uproar, modal idenlification and/or feature extraction, with a variety of damage datas that obtain that treated, according to a variety of damage datas and default
The corresponding damnification recognition method of every kind of damage criterion obtain a variety of initial damage recognition results.
In step s 102, it is identified according to every kind of damage criterion of a variety of damage criterions and carries out non-destructive tests, it is more to obtain
Kind initial damage recognition result.
It will be appreciated that preliminary recognition result is obtained after handling damage criterion collected in previous step, then
By the non-destructive tests of this step, the non-destructive tests result that next step is used is obtained.
In step s 103, it is associated according to a variety of initial damage recognition results and decision level fusion, to pass through data
Fusion obtains last diagnostic result.
Further, in one embodiment of the invention, by every kind of initial damage of a variety of initial damage recognition results
Recognition result is as evidence, to obtain average evidence；It is determined according to the Jousselme distance between every evidence and average evidence every
The weight coefficient of kind initial damage recognition result；Every evidence is weighted and averaged, to be obtained by Murphy improved method
Last diagnostic result.
Wherein, the calculation formula of average evidence are as follows:
Wherein, U is the damage reason location index group number for health monitoring, mi(Aj) be each index recognition result；
The calculation formula of Jousselme distance are as follows:
Wherein, M is average evidence mean value,Indicate two vectorsWithBetween inner product,
Indicate to
AmountInner product,Indicate two vectorsWithInner product；
Weighted average formula are as follows:
Wherein, ωiFor the weight of each evidence, U is the damage reason location index group number for health monitoring.
Further, in one embodiment of the invention, according to the Jousselme between every evidence and average evidence away from
The weight coefficient of every kind from determination initial damage recognition result further comprises:
Define degree of belief function, degree of belief function are as follows:
Wherein, DiFor the recognition result of each damage criterion and the Jousselme distance of average evidence mean value M；
Degree of belief function is normalized, to obtain the weight of every evidence, weight formula are as follows:
Wherein, TiFor degree of belief function, U is the damage reason location index group number for health monitoring.
Hydro-structure damage diagnosis method is induced to aerial drainage of the invention below by specific embodiment to be described in detail.
It should be noted that data fusion is an information processing system, in conjunction with the data got from multiple information sources and
Relevant information in Relational database obtains accuracy more higher than single piece of information source and inference.Decision level fusion is highest
The fusion of level, for aerial drainage monitoring structural health conditions field, decision level fusion can integrate the identification of multiple damage monitoring indexs
As a result, to improve the accuracy rate of non-destructive tests.As shown in Fig. 2, firstly, using n kind damnification recognition method respectively to measured data
It is analyzed and processed, including filtering noise reduction, modal idenlification and feature extraction etc., obtains the preliminary judgement of every kind of damnification recognition method
Then conclusion judges to carry out data fusion to various preliminary judging results by relevant treatment, decision level fusion, final to obtain
Comprehensive inferred results, make highest level decision.
The embodiment of the present invention is related to D-S evidence theory, wherein D-S evidence theory basic conception are as follows:
If Ω is framework of identification, all propositions can be indicated with the subset of Ω, power set 2ΩAll subsets comprising Ω
And empty set.If A is an element in Ω, Basic probability assignment function is defined in 2ΩTo a mapping m of [0,1], and it is full
Foot:
Then claiming m is 2ΩOn probability distribution function, m (A) is known as elementary probability number.
It is as follows to define belief function Bel (lower limit function): 2Ω→ [0,1], and:
Then Bel (A) is known as the trust value of event A, indicates that evidence is genuine trusting degree, the trust value of empty set to event A
It is 0.
It enablesIndicate to be false trusting degree to event A, then likelihood function Pl is defined as follows: 2Ω→ [0,1], and
Meet:
In formula, Pl (A) indicates that the trusting degree to event A for non-vacation, Bel (A) and Pl (A) have following relationship:
Pl(A)≥Bel(A) (4)
If Bel1And Bel2It is two belief functions on framework of identification Ω, m1And m2Respectively their elementary probability
Partition function, element are respectively Ai(i=1,2,3 ..., k) and Bj(j=1,2,3 ..., l), definition conflict factor k:
So defined function M:2Ω→ [0,1] is m1And m2Belief assignment after synthesis
For multiple probability distribution functions fusion can by merge two-by-two carry out, final fusion results with merge it is suitable
Sequence is unrelated.
That is, the value for the factor k that conflicts is bigger for the evidence in D-S evidence theory, show each card
Conflict spectrum is higher between.As k=1, evidence can not be synthesized, and as k → 1, can obtain mistake to high combination of conflicting evidence
As a result.Due to the influence of each damage criterion self character and measurement noise, so that there is centainly inclined in the recognition result of each index
Difference.Under conditions of damage criterion limited amount used, it is easy to the judging result to make mistake.Therefore, it is intended to manage D-S evidence
By being applied in the Comprehensive Evaluation of aerial drainage structure dynamics health monitoring, need to carry out a combination thereof rule certain improvement.
Further, the embodiment of the present invention improves the rule of combination in D-S evidence theory, it is assumed that Ω is one complete
Framework of identification includes N number of different proposition two-by-two, m in frameiIt is its corresponding Basic probability assignment function, and meets:
Then Basic probability assignment function m1And m2Jousselme distance can indicate are as follows:
In formula,
Indicate two vectorsWithBetween inner product:
Due to being influenced by sensor self-condition, noise and damage criterion accuracy of identification, the knowledge of each damage criterion
Effect of the other result played in final decision is different, i.e., different weight coefficients should be endowed in decision.By each damage
The recognition result of index regards an evidence as, and the average value for calculating each evidence is defined as average evidence.If an evidence
Jousselme distance between average evidence is smaller, then it is assumed that the evidence supported by other most of evidences, confidence level compared with
Height should assign biggish weight coefficient in fusion process, be acted on enhancing the evidence in fusion decision process, conversely, such as
For the Jousselme of one evidence of fruit and average evidence apart from larger, then the confidence level of the evidence is lower, should assign relatively small
Weight coefficient, to weaken it to the possible negative effect of the result of decision.Therefore, the improvement D-S that the embodiment of the present invention proposes
The basic ideas of evidence theory are: the confidence level of evidence is judged by the Jousselme distance of each evidence and average evidence,
Weight coefficient is introduced, each evidence is weighted and averaged, the improved method for recycling Murphy to propose is merged, wherein tool
Body step are as follows:
(1) assume the damage reason location index for having U group for health monitoring, the recognition result of each index is respectively mi(Aj),
Aj(j=1,2,3 ..., 2N) it is all positions for being likely to occur damage.The mean value M for calculating U group recognition result may be expressed as:
(2) every evidence is calculated:
(3) if DiIt is worth larger, then it is assumed that the evidence is unreliable, and degree of belief is relatively low；On the contrary, DiIt is worth smaller, the then evidence
Degree of belief it is higher.Define degree of belief function TiAre as follows:
(4) to degree of belief function TiIt is normalized, the weight ω of each evidence can be obtainedi:
(5) each evidence is weighted and averaged, obtains average weighted evidence m*(Aj):
(6) the evidence m after being averaged according to the method for Murphy proposition is diffusion-weighted*(Aj), since system has U damage to refer to
Mark, therefore will be weighted and averaged Evidence Combination Methods U-l times.
An example is introduced below, and implementation of the present invention is verified by comparison Dempster method and Murphy improved method
The validity of example method.Table 1 is the probability value of every evidence in example, gives m1~m5Totally 5 evidences, wherein m (A), m
(B) and m (C) be respectively recognition result A, B and C elementary probability value.
Table 1
In case where 5 evidences of application carry out fusion calculations, briefly in plaintext method calculating process: count first
Calculate the mean value M=[0.448 of 5 evidences；0.32；0.232]；Secondly the Jousselme distance D of each evidence and mean value M is calculated
=[0.1042；0.5266；0.1907；0.1618；0.1230]；Then the degree of belief function T=[9.5941 of each evidence is calculated；
1.8991；5.2440；6.1823；8.1314]；And then obtain the weight coefficient w=[0.3090 of each evidence；0.0612；
0.1689；0.1991；0.2619]；M*=[0.5213 is obtained after being weighted and averaged to each evidence；0.2291；0.2496]；Finally
According to Murphy method, m* is combined 4 times, the result of decision [0.9601 to the end is obtained；0.0157；0.0241].Table 2 is various melts
The calculated result of conjunction method, by various method fusion calculations, the results are shown in Table 2.
Table 2
It can be seen that Dempster method from the result in table and give the result of decision of mistake, mistake Producing reason
It is evidence m2The probability value m of middle result A2(A)=0, conflict with other four groups of evidence generations, although evidence m1、m3、m4、m5All it is
Support result A, but Dempster method can not still obtain correctly as a result, therefore this method is not suitable for clashing
Evidence carry out decision calculating.Murphy improved method is by being averaged n all evidences to offset error proof
It influences, n-1 fusion calculation directly is carried out to the result after being averaged, since this method does not account for the correlation between each evidence
Property, so system needs enough evidences that could effectively inhibit the influence of error proof.In the example above, when system is collected
When to 4 evidences, correct recognition result is just provided.The improved method that the embodiment of the present invention proposes, has fully considered each evidence
Between correlation, only just give correct judging result with first three evidence, improve the fusion results in evidences conflict
Reliability and reasonability.
The aerial drainage proposed according to embodiments of the present invention induces hydro-structure damage diagnosis method, by passing through acquisition aerial drainage knot
A variety of damage criterions of structure, and non-destructive tests are carried out to the identification of every kind of damage criterion and obtain initial damage recognition result, using more
The blending decision method of index obtains last diagnostic as a result, it is possible to carry out integrated treatment to a variety of uncertain informations, accurately reflects
The operating status of aerial drainage structure, the position that quickly discovery and identification of damage occur.
The aerial drainage proposed according to embodiments of the present invention referring next to attached drawing description induces hydro-structure damage diagnostic device.
Fig. 3 is that the aerial drainage of one embodiment of the invention induces hydro-structure damage diagnostic device structural schematic diagram.
As shown in figure 3, it includes: acquisition module 100, identification module that the aerial drainage, which induces hydro-structure damage diagnostic device 10,
200 and diagnostic module 300.
Wherein, a variety of damage criterion identification modules 200 that acquisition module 100 is used to acquire aerial drainage structure are used for according to more
Every kind of damage criterion identification of kind damage criterion carries out non-destructive tests, to obtain a variety of initial damage recognition results.Diagnostic module
300 according to a variety of initial damage recognition results for being associated and decision level fusion, finally to be examined by data fusion
Disconnected result.The diagnostic device 10 of the embodiment of the present invention can carry out integrated treatment to a variety of uncertain informations, accurately reflect aerial drainage
The operating status of structure, the position that quickly discovery and identification of damage occur.
Further, in one embodiment of the invention, acquisition module is further used for carrying out a variety of damage criterions
Noise reduction, modal idenlification and/or feature extraction are filtered, with a variety of damage datas that obtain that treated, according to a variety of damage datas
Damnification recognition method corresponding with preset every kind of damage criterion obtains a variety of initial damage recognition results.
Further, in one embodiment of the invention, diagnostic module is further used for identifying a variety of initial damages
As a result every kind of initial damage recognition result is as evidence, to obtain average evidence, and according between every evidence and average evidence
Jousselme distance determine the weight coefficient of every kind of initial damage recognition result, and then every evidence is weighted and averaged,
To obtain last diagnostic result by Murphy improved method.
Wherein, the calculation formula of average evidence are as follows:
Wherein, U is the damage reason location index group number for health monitoring, mi(Aj) be each index recognition result；
The calculation formula of Jousselme distance are as follows:
Wherein, M is average evidence mean value,Indicate two vectorsWithBetween inner product,
Indicate to
AmountInner product,Indicate two vectorsWithInner product；
Weighted average formula are as follows:
Wherein, ωiFor the weight of each evidence, U is the damage reason location index group number for health monitoring.
Further, in one embodiment of the invention, according to the Jousselme between every evidence and average evidence away from
The weight coefficient of every kind from determination initial damage recognition result further comprises:
Define degree of belief function, degree of belief function are as follows:
Wherein, DiFor the recognition result of each damage criterion and the Jousselme distance of average evidence mean value M；
Degree of belief function is normalized, to obtain the weight of every evidence, weight formula are as follows:
Wherein, TiFor degree of belief function, U is the damage reason location index group number for health monitoring.
It should be noted that the aforementioned explanation for inducing hydro-structure damage diagnosis method embodiment to aerial drainage is also suitable
In the device of the embodiment, details are not described herein again.
The aerial drainage proposed according to embodiments of the present invention induces hydro-structure damage diagnostic device, passes through acquisition aerial drainage structure
A variety of damage criterions, and non-destructive tests are carried out to every kind of damage criterion identification and obtain initial damage recognition result, utilize multi objective
Blending decision method obtain last diagnostic as a result, it is possible to a variety of uncertain informations carry out integrated treatment, accurately reflect aerial drainage
The operating status of structure, the position that quickly discovery and identification of damage occur.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the present invention, the meaning of " plurality " is at least two, such as two, three
It is a etc., unless otherwise specifically defined.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or spy described in conjunction with this embodiment or example
Point is included at least one embodiment or example of the invention.In the present specification, schematic expression of the above terms are not
It must be directed to identical embodiment or example.Moreover, particular features, structures, materials, or characteristics described can be in office
It can be combined in any suitable manner in one or more embodiment or examples.In addition, without conflicting with each other, the skill of this field
Art personnel can tie the feature of different embodiments or examples described in this specification and different embodiments or examples
It closes and combines.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of aerial drainage induces hydro-structure damage diagnosis method, which comprises the following steps:
Acquire a variety of damage criterions of aerial drainage structure；
It is identified according to every kind of damage criterion of a variety of damage criterions and carries out non-destructive tests, to obtain a variety of initial damage identifications
As a result；And
It is associated according to a variety of initial damage recognition results and decision level fusion, finally to be examined by data fusion
Disconnected result.
2. aerial drainage according to claim 1 induces hydro-structure damage diagnosis method, which is characterized in that let out described in the acquisition
After a variety of damage criterions of flow structure, further includes:
Noise reduction, modal idenlification and/or feature extraction are filtered to a variety of damage criterions, with a variety of damages that obtain that treated
Hurt data, it is described more to be obtained according to a variety of damage datas and the corresponding damnification recognition method of preset every kind of damage criterion
Kind initial damage recognition result.
3. aerial drainage according to claim 1 induces hydro-structure damage diagnosis method, which is characterized in that described according to
A variety of initial damage recognition results are associated and decision level fusion, further comprise:
Using every kind of initial damage recognition result of a variety of initial damage recognition results as evidence, to obtain average evidence；
Every kind of initial damage recognition result is determined according to the Jousselme distance between every evidence and the average evidence
Weight coefficient；
Every evidence is weighted and averaged, to obtain last diagnostic result by Murphy improved method.
4. aerial drainage according to claim 3 induces hydro-structure damage diagnosis method, which is characterized in that wherein,
The calculation formula of the average evidence are as follows:
Wherein, U is the damage reason location index group number for health monitoring, mi(Aj) be each index recognition result；
The calculation formula of the Jousselme distance are as follows:
Wherein, M is average evidence mean value,Indicate two vectorsWithBetween inner product,
Indicate vectorInner product,Indicate two vectorsWithInner product；
Weighted average formula are as follows:
Wherein, ωiFor the weight of each evidence, U is the damage reason location index group number for health monitoring.
5. aerial drainage according to claim 3 or 4 induces hydro-structure damage diagnosis method, which is characterized in that the basis
Jousselme distance between every evidence and the average evidence determines the weight system of every kind of initial damage recognition result
Number further comprises:
Define degree of belief function, the degree of belief function are as follows:
Wherein, DiFor the recognition result of each damage criterion and the Jousselme distance of average evidence mean value M；
The degree of belief function is normalized, to obtain the weight of every evidence, the weight formula are as follows:
Wherein, TiFor degree of belief function, U is the damage reason location index group number for health monitoring.
6. a kind of aerial drainage, which induces hydro-structure, damages diagnostic device characterized by comprising
Acquisition module, for acquiring a variety of damage criterions of aerial drainage structure；
Identification module carries out non-destructive tests for identifying according to every kind of damage criterion of a variety of damage criterions, more to obtain
Kind initial damage recognition result；And
Diagnostic module, for being associated according to a variety of initial damage recognition results and decision level fusion, to pass through data
Fusion obtains last diagnostic result.
7. aerial drainage according to claim 6, which induces hydro-structure, damages diagnostic device, which is characterized in that the acquisition module
It is further used for being filtered noise reduction, modal idenlification and/or feature extraction to a variety of damage criterions, to obtain, treated
A variety of damage datas, to be obtained according to a variety of damage datas and the corresponding damnification recognition method of preset every kind of damage criterion
A variety of initial damage recognition results.
8. aerial drainage according to claim 6, which induces hydro-structure, damages diagnostic device, which is characterized in that the diagnostic module
It is further used for using every kind of initial damage recognition result of a variety of initial damage recognition results as evidence, to be averaged
Evidence, and determine that every kind of initial damage identification is tied according to the Jousselme distance between every evidence and the average evidence
The weight coefficient of fruit, and then every evidence is weighted and averaged, to obtain last diagnostic result by Murphy improved method.
9. aerial drainage according to claim 8, which induces hydro-structure, damages diagnostic device, which is characterized in that wherein,
The calculation formula of the average evidence are as follows:
Wherein, U is the damage reason location index group number for health monitoring, mi(Aj) be each index recognition result；
The calculation formula of the Jousselme distance are as follows:
Wherein, M is average evidence mean value,Indicate two vectorsWithBetween inner product,
Indicate vectorInner product,Indicate two vectorsWithInner product；
Weighted average formula are as follows:
Wherein, ωiFor the weight of each evidence, U is the damage reason location index group number for health monitoring.
10. aerial drainage according to claim 8 or claim 9, which induces hydro-structure, damages diagnostic device, which is characterized in that the basis
Jousselme distance between every evidence and the average evidence determines the weight system of every kind of initial damage recognition result
Number further comprises:
Define degree of belief function, the degree of belief function are as follows:
Wherein, DiFor the recognition result of each damage criterion and the Jousselme distance of average evidence mean value M；
The degree of belief function is normalized, to obtain the weight of every evidence, the weight formula are as follows:
Wherein, TiFor degree of belief function, U is the damage reason location index group number for health monitoring.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811208354.XA CN109598282A (en) | 2018-10-17 | 2018-10-17 | Aerial drainage induces hydro-structure damage diagnosis method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811208354.XA CN109598282A (en) | 2018-10-17 | 2018-10-17 | Aerial drainage induces hydro-structure damage diagnosis method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109598282A true CN109598282A (en) | 2019-04-09 |
Family
ID=65957549
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811208354.XA Pending CN109598282A (en) | 2018-10-17 | 2018-10-17 | Aerial drainage induces hydro-structure damage diagnosis method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109598282A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298143A (en) * | 2019-07-30 | 2019-10-01 | 中冶建筑研究总院有限公司 | A kind of prestressing force truss string structure damnification recognition method based on two stages data fusion |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102841140A (en) * | 2012-08-17 | 2012-12-26 | 天津大学 | Monitoring method for damages of drainage structure based on stochastic resonance |
| CN106154182A (en) * | 2016-08-26 | 2016-11-23 | 上海电力学院 | A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory |
-
2018
- 2018-10-17 CN CN201811208354.XA patent/CN109598282A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102841140A (en) * | 2012-08-17 | 2012-12-26 | 天津大学 | Monitoring method for damages of drainage structure based on stochastic resonance |
| CN106154182A (en) * | 2016-08-26 | 2016-11-23 | 上海电力学院 | A kind of based on the lithium battery method for diagnosing faults improving D S evidence theory |
Non-Patent Citations (1)
| Title |
|---|
| 李成业: "泄流结构水利拍振机理及动态健康监测技术研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110298143A (en) * | 2019-07-30 | 2019-10-01 | 中冶建筑研究总院有限公司 | A kind of prestressing force truss string structure damnification recognition method based on two stages data fusion |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106769052B (en) | A kind of mechanical system rolling bearing intelligent failure diagnosis method based on clustering | |
| CN104712542B (en) | A kind of reciprocating compressor sensitive features based on Internet of Things are extracted and method for diagnosing faults | |
| CN102870057B (en) | Plant diagnosis device, diagnosis method, and diagnosis program | |
| KR101101974B1 (en) | System for fault detection and diagnosis of aircraft engine and method thereof | |
| CN110823576B (en) | Mechanical anomaly detection method based on generation of countermeasure network | |
| CN108332970A (en) | A kind of Method for Bearing Fault Diagnosis based on LS-SVM and D-S evidence theory | |
| JP6200833B2 (en) | Diagnostic equipment for plant and control equipment | |
| CN106022366B (en) | A kind of rotating machinery method for diagnosing faults based on neighbour's evidence fusion | |
| US20210397175A1 (en) | Abnormality detection device, abnormality detection method, and program | |
| Silva et al. | Damage‐sensitive feature extraction with stacked autoencoders for unsupervised damage detection | |
| CN108022058A (en) | A kind of wind energy conversion system state reliability estimation method | |
| CN109932179A (en) | A kind of rolling bearing fault testing method based on the reconstruct of DS Adaptive spectra | |
| CN108804740A (en) | Long distance pipeline pressure monitoring method based on integrated improvement ICA-KRR algorithms | |
| CN110704911A (en) | Bridge damage cross positioning method based on cluster structure similarity | |
| CN113757093B (en) | Flash steam compressor unit fault diagnosis method | |
| CN110044620A (en) | A kind of Fault Diagnosis of Roller Bearings based on analysis of vibration signal | |
| CN106872172A (en) | The method for real time discriminating and system of Aero Engine Testing security parameter monitoring | |
| CN109799804A (en) | A kind of diagnosis algorithm appraisal procedure and system based on random fault injection | |
| Zhao et al. | Design of a novel knowledge-based fault detection and isolation scheme | |
| CN109598282A (en) | Aerial drainage induces hydro-structure damage diagnosis method and device | |
| CN110486350A (en) | Fault Diagnosis Method of Electro-hydraulic and device, storage medium and electronic equipment | |
| JP6685124B2 (en) | Diagnostic device and diagnostic method | |
| CN110826600B (en) | Engine surge prediction method based on adaptive resonance network online incremental learning | |
| Farrar et al. | Vibration-based damage detection in rotating machinery | |
| CN115563868A (en) | Fault diagnosis method and device for oil circuit system of aviation alternating-current generator |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190409 |
|
| WD01 | Invention patent application deemed withdrawn after publication |