CN107607302A - Mechanical Failure of HV Circuit Breaker diagnostic method based on multidimensional scaling statistical analysis - Google Patents
Mechanical Failure of HV Circuit Breaker diagnostic method based on multidimensional scaling statistical analysis Download PDFInfo
- Publication number
- CN107607302A CN107607302A CN201710774736.8A CN201710774736A CN107607302A CN 107607302 A CN107607302 A CN 107607302A CN 201710774736 A CN201710774736 A CN 201710774736A CN 107607302 A CN107607302 A CN 107607302A
- Authority
- CN
- China
- Prior art keywords
- vibration signal
- circuit breaker
- statistical analysis
- diagnostic method
- mechanical failure
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
Mechanical Failure of HV Circuit Breaker diagnostic method based on multidimensional scaling statistical analysis：Step 1, by fault simulating test, certain vibration signal waveforms of model primary cut-out under different machine performances is obtained using IEPE acceleration transducers, so as to form the model primary cut-out vibration signal data storehouse；Step 2, according to the frequency bandwidth characteristicses of IEPE acceleration transducers bandpass filtering treatment is carried out to the vibration signal of collection；The distribution situation of step 3, the amplitude of statistics vibration signal；Step 4, using multidimensional pantography by distance matrix A from p dimension drop to p' dimension；Step 5, p' dimension coordinates and the coordinate in basic waveform database of the vibration signal for being measured primary cut-out in real time using SVMs carry out comparison-of-pair sorting, realize the mechanical breakdown condition diagnosing of primary cut-out.Simple, the influence that can effectively avoid signal Topical Dispersion from bringing is calculated, reduces the error rate of fault diagnosis.
Description
Technical field
The present invention relates to a kind of Mechanical Failure of HV Circuit Breaker diagnostic method based on multidimensional scaling statistical analysis.
Background technology
Breaker of the operating voltage in more than 3kV is referred to as primary cut-out, and it is most important in high-tension switch gear
Electrical equipment, play a part of controlling in power network and protect, the major failure of breaker has tripping failure, malfunction failure, absolutely
Reason barrier, current-carrying failure, external force and other failures, in the reason for above-mentioned failure counts, mechanical reason accounts for more than 60%, and machine
The problem of overwhelming majority is operating mechanism in tool failure, thus the High Voltage Circuit Breaker Condition especially machine performance is monitored and
Assess it is particularly significant, for improving and safeguarding that the safe and stable operation of power system is significant.
In the prior art, it is conventional method to be diagnosed using signal progress Mechanical Failure of HV Circuit Breaker, the think of of this method
Lu Wei：To signal after pretreatment, extract included in signal characteristic quantity (vibration event initial time in such as vibration signal,
Vibration signal frequency domain/time-frequency characteristics), realized in conjunction with fault diagnosis algorithm (such as neutral net, SVMs) to high pressure
The diagnosis and identification of circuit breaker failure.Following shortcomings be present in it：
Firstth, existing research is the achievement in the case of carrying out a small amount of fault simulation in laboratory mostly, the applicability of method
It is smaller, it is unfavorable for the condition diagnosing under complex fault in primary cut-out actual motion.
Secondth, the diagnostic method is had a great influence by local signal, because primary cut-out vibration signal has dispersiveness, together
Measurement signal local shape under one machine performance simultaneously differs, and existing method can cause characteristic quantity aliasing, cause to judge by accident.
3rd, the diagnostic method amount of calculation is excessive, the Time-frequency method such as wavelet transformation, empirical mode decomposition, feature extraction
Amount of calculation is very big, and the requirement to hardware such as processors is too high, is not suitable for primary cut-out on-line monitoring.
The content of the invention
In view of the above-mentioned problems, the present invention provides the Mechanical Failure of HV Circuit Breaker diagnosis side based on multidimensional scaling statistical analysis
Method, simple, the influence that can effectively avoid signal Topical Dispersion from bringing is calculated, reduce the error rate of fault diagnosis.
To realize above-mentioned technical purpose and the technique effect, the present invention is achieved through the following technical solutions：
Based on the Mechanical Failure of HV Circuit Breaker diagnostic method of multidimensional scaling statistical analysis, comprise the following steps：
Step 1, by fault simulating test, obtain certain model primary cut-out in difference using IEPE acceleration transducers
Vibration signal waveforms under machine performance, so as to form the model primary cut-out vibration signal data storehouse；
Step 2, according to the frequency bandwidth characteristicses of IEPE acceleration transducers bandpass filtering treatment is carried out to the vibration signal of collection；
The distribution situation of step 3, the amplitude of statistics vibration signal：
3.1st, the vibration signal for acquiring p sub-high pressure breakers, respectively x are set1,x2,...,xp, make D={ x1,x2,...,
xp}；
3.2nd, in statistics collection D acceleration amplitude distributed area：
If the scope of acceleration amplitude is 0~y, y is divided into q sections, is divided into Common q section, count i-th of signal xiThe number x of point in q section respectivelyi1、xi2、...xiq, form
Q characteristic quantity, is designated as xi=(xi1；xi2；...；xiq), xi∈D；
3.3rd, p signal amounts to p × q characteristic quantity, and characteristic quantity is normalized；
3.4th, the Euclidean distance between p signal is calculated, forms distance matrix A, A ∈ Rp×p, i.e. distance matrix A is diagonal
Matrix, wherein, distance matrix A the i-th row, the element dist of jth rowijFor sample xiTo xjDistance；
Step 4, using multidimensional pantography by distance matrix A from p dimension drop to p' dimension；
P' dimension coordinates and the basic ripple of step 5, the vibration signal for being measured primary cut-out in real time using SVMs
Coordinate in graphic data storehouse carries out comparison-of-pair sorting, realizes the mechanical breakdown condition diagnosing of primary cut-out.
It is preferred that in step 3.3, characteristic quantity is normalized with linear function method.
It is preferred that IEPE acceleration transducers use the PCB-352B70 sensors of PCB companies exploitation.
It is preferred that IEPE acceleration transducers are arranged on the position on high voltage circuit breaker closing electromagnet periphery.
It is preferred that p'=2, or, p'=3.
It is preferred that carry out visualization output after distance matrix A is dropped into p' dimensions from p dimensions using multidimensional pantography in step 4.
The beneficial effects of the invention are as follows：
Firstth, the collection approach of this method desired signal is simple, does not influence the structure of primary cut-out.
Secondth, because the extraction process of characteristic quantity is using the method for directly counting amplitude, without multiplication and division computing, thus
Calculate simply, it is low to hardware requirement.
3rd, the Characteristic Extraction method based on statistical analysis, the influence that can effectively avoid signal Topical Dispersion from bringing,
Reduce the error rate of fault diagnosis.
4th, optimized using multidimensional pantography (Multiple Dimensional Scaling, abbreviation MDS), can passed through
The visualization technique effect that intuitively judging characteristic extracts.
Brief description of the drawings
Fig. 1 is the flow chart of Mechanical Failure of HV Circuit Breaker diagnostic method of the present invention based on multidimensional scaling statistical analysis.
Embodiment
Technical solution of the present invention is described in further detail with specific embodiment below in conjunction with the accompanying drawings, so that ability
The technical staff in domain can be better understood from the present invention and can be practiced, but illustrated embodiment is not as the limit to the present invention
It is fixed.
As shown in figure 1, the Mechanical Failure of HV Circuit Breaker diagnostic method based on multidimensional scaling statistical analysis, including following step
Suddenly：
Step 1, by fault simulating test, using IEPE acceleration transducers, (IEPE refers to that one kind is put from carried charge
The acceleration transducer of big device or voltage amplifier) obtain certain vibration signal of model primary cut-out under different machine performances
Waveform, so as to form the model primary cut-out vibration signal data storehouse.
In high-voltage circuit-breaker switching on-off operating process, the collision unshakable in one's determination from electromagnet coil power to divide-shut brake sound, during which
Existing collision and friction several times between each part, shows as the vibration event several times in the time domain in vibration signal,
The differences such as the amplitude of different event, shape, perdurabgility.But the Circuit breaker vibration signal under same model, same mechanical state
Similar, Circuit breaker vibration signal has differences under different machine performances, and this is the basis for carrying out fault diagnosis.
It is preferred that IEPE acceleration transducers are arranged on high voltage circuit breaker closing electromagnet periphery or neighbouring position.IEPE
Acceleration transducer can use the PCB-352B70 sensors of PCB companies exploitation.
Step 2, according to the frequency bandwidth characteristicses of IEPE acceleration transducers bandpass filtering treatment is carried out to the vibration signal of collection.
By taking PCB-352B70 sensors as an example, the frequency ranges (- 3dB) of PCB-352B70 acceleration transducers is 0.4~
20000Hz, time-domain signal is switched into frequency domain with DFT (DFT), retains the composition of 0.4~20000Hz in frequency domain,
Other frequency contents are removed, time-domain signal is returned with inverse discrete Fourier transformer inverse-discrete (IDFT) formula.According to IEPE acceleration
The characteristic of sensor carries out targetedly bandpass filtering.
The distribution situation of step 3, the amplitude of statistics vibration signal：
3.1st, the vibration signal for acquiring p sub-high pressure breakers, respectively x are set1,x2,...,xp, make D={ x1,x2,...,
xp}；
3.2nd, in statistics collection D acceleration amplitude distributed area：
If the scope of acceleration amplitude is 0~y, y is divided into q sections, is divided into Common q section, count i-th of signal xiThe number x of point in q section respectivelyi1、xi2、...xiq, form q
Individual characteristic quantity, is designated as xi=(xi1；xi2；...；xiq), xi∈D；
3.3rd, p signal amounts to p × q characteristic quantity, and characteristic quantity is normalized, such as with linear function method pair
Characteristic quantity is normalized；
3.4th, the Euclidean distance between p signal is calculated, forms distance matrix A, A ∈ Rp×p, i.e. distance matrix A is diagonal
Matrix, wherein, distance matrix A the i-th row, the element dist of jth rowijFor sample xiTo xjDistance；
Step 4, using multidimensional pantography (abbreviation MDS) by distance matrix A from p dimension drop to p' dimension, it is preferable that p'=2, or
Person, p'=3, to visualize output.Order
In formula, disti.Root-mean-square value for i-th of sample to each sample distance；dist.jFor j-th of sample to each sample
The root-mean-square value of distance；dist..The root-mean-square value of distance between total sample；bijFor the i-th row of inner product matrix B jth row
Value.
Inner product matrix B can be obtained according to formula (4).
Eigenvalues Decomposition, B=V Λ V are to matrix BT；Wherein Λ=diag (λ1,λ2,...,λp) it is characterized value composition
Diagonal matrix, λ1≥λ2≥...≥λp, V is characterized vector matrix.
TakeBy the diagonal matrix that p' eigenvalue of maximum is formed in Λ,For the corresponding characteristic vector extracted from V
Matrix.Output matrixThat is matrixFor diagonal matrix, the often row of matrix is the p' dimension seats of a sample
Mark.Finally carry out visualization output.
P' dimension coordinates and the basic ripple of step 5, the vibration signal for being measured primary cut-out in real time using SVMs
Coordinate in graphic data storehouse carries out comparison-of-pair sorting, realizes the mechanical breakdown condition diagnosing of primary cut-out.
In actual moving process, because primary cut-out is complicated, there is multi-source uncertainty in vibration signal, cause letter
Dispersiveness be present in number local feature.Existing diagnostic method is sensitive to signal intensity, and the unstable of local signal leads to examine
Gross differences occur for disconnected result.Although local feature has dispersiveness, the overall shape of Circuit breaker vibration signal, amplitude follow
Relatively stable rule, makees the statistical analysis of globality to vibration signal, investigates the signal of breaker once complete action process
The distribution situation of amplitude, the influence that can effectively avoid signal Topical Dispersion from bringing, reduce the error rate of fault diagnosis.
Inherent characteristic of this method based on acceleration transducer, specific aim filtering is done to vibration signal, remove drift and height
Frequency noise；Then take statistics analysis to the amplitude of signal, using the quantity of the point in different amplitude sections as characteristic quantity to be selected；With
Multidimensional scales (Multiple Dimensional Scaling, abbreviation MDS) analytic approach, to sample distance progress dimensionality reduction and visually
Change；By the low-dimensional coordinate input SVMs (SVM) after dimensionality reduction, Fault Identification diagnosis is carried out.
The beneficial effects of the invention are as follows：
Firstth, the collection approach of this method desired signal is simple, does not influence the structure of primary cut-out.
Secondth, because the extraction process of characteristic quantity is using the method for directly counting amplitude, without multiplication and division computing, thus
Calculate simply, it is low to hardware requirement.
3rd, the Characteristic Extraction method based on statistical analysis, the influence that can effectively avoid signal Topical Dispersion from bringing,
Reduce the error rate of fault diagnosis.
4th, optimized using multidimensional pantography (Multiple Dimensional Scaling, abbreviation MDS), can passed through
The visualization technique effect that intuitively judging characteristic extracts.
The preferred embodiments of the present invention are these are only, are not intended to limit the scope of the invention, it is every to utilize this hair
The equivalent structure that bright specification and accompanying drawing content are made either equivalent flow conversion or to be directly or indirectly used in other related
Technical field, be included within the scope of the present invention.
Claims (7)
1. the Mechanical Failure of HV Circuit Breaker diagnostic method based on multidimensional scaling statistical analysis, it is characterised in that including following step
Suddenly：
Step 1, by fault simulating test, obtain certain model primary cut-out in different machineries using IEPE acceleration transducers
Vibration signal waveforms under state, so as to form the model primary cut-out vibration signal data storehouse；
Step 2, according to the frequency bandwidth characteristicses of IEPE acceleration transducers bandpass filtering treatment is carried out to the vibration signal of collection；
The distribution situation of step 3, the amplitude of statistics vibration signal：
3.1st, the vibration signal for acquiring p sub-high pressure breakers, respectively x are set1,x2,...,xp, make D={ x1,x2,...,xp}；
3.2nd, in statistics collection D acceleration amplitude distributed area：
If the scope of acceleration amplitude is 0~y, y is divided into q sections, is divided into Common q section, count i-th of signal xiThe number x of point in q section respectivelyi1、xi2、...xiq, form
Q characteristic quantity, is designated as xi=(xi1；xi2；...；xiq), xi∈D；
3.3rd, p signal amounts to p × q characteristic quantity, and characteristic quantity is normalized；
3.4th, the Euclidean distance between p signal is calculated, forms distance matrix A, A ∈ Rp×p, i.e. distance matrix A is diagonal matrix,
Wherein, distance matrix A the i-th row, the element dist of jth rowijFor sample xiTo xjDistance；
Step 4, using multidimensional pantography by distance matrix A from p dimension drop to p' dimension；
The p' dimension coordinates and basic waveform number of step 5, the vibration signal for being measured primary cut-out in real time using SVMs
Comparison-of-pair sorting is carried out according to the coordinate in storehouse, realizes the mechanical breakdown condition diagnosing of primary cut-out.
2. the Mechanical Failure of HV Circuit Breaker diagnostic method according to claim 1 based on multidimensional scaling statistical analysis, its
It is characterised by, in step 3.3, characteristic quantity is normalized with linear function method.
3. the Mechanical Failure of HV Circuit Breaker diagnostic method according to claim 1 based on multidimensional scaling statistical analysis, its
It is characterised by, IEPE acceleration transducers use the PCB-352B70 sensors of PCB companies exploitation.
4. the Mechanical Failure of HV Circuit Breaker diagnostic method according to claim 3 based on multidimensional scaling statistical analysis, its
It is characterised by, IEPE acceleration transducers are arranged on the position on high voltage circuit breaker closing electromagnet periphery.
5. the Mechanical Failure of HV Circuit Breaker diagnostic method according to claim 3 based on multidimensional scaling statistical analysis, its
It is characterised by, p'=2.
6. the Mechanical Failure of HV Circuit Breaker diagnostic method according to claim 3 based on multidimensional scaling statistical analysis, its
It is characterised by, p'=3.
7. the Mechanical Failure of HV Circuit Breaker diagnostic method based on multidimensional scaling statistical analysis according to claim 5 or 6,
Characterized in that, carry out visualization output after distance matrix A is dropped into p' dimensions from p dimensions using multidimensional pantography in step 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710774736.8A CN107607302B (en) | 2017-08-31 | 2017-08-31 | The Mechanical Failure of HV Circuit Breaker diagnostic method of statistical analysis is scaled based on multidimensional |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710774736.8A CN107607302B (en) | 2017-08-31 | 2017-08-31 | The Mechanical Failure of HV Circuit Breaker diagnostic method of statistical analysis is scaled based on multidimensional |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107607302A true CN107607302A (en) | 2018-01-19 |
| CN107607302B CN107607302B (en) | 2018-07-17 |
Family
ID=61056813
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710774736.8A Active CN107607302B (en) | 2017-08-31 | 2017-08-31 | The Mechanical Failure of HV Circuit Breaker diagnostic method of statistical analysis is scaled based on multidimensional |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107607302B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108955871A (en) * | 2018-08-27 | 2018-12-07 | 河海大学 | A kind of tap switch vibration signal noise-reduction method based on quick kurtogram algorithm |
| CN114936582A (en) * | 2022-06-08 | 2022-08-23 | 华侨大学 | Working modal parameter identification method and system and fault position identification method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100114813A1 (en) * | 2008-10-20 | 2010-05-06 | Zalay Osbert C | Method and rhythm extractor for detecting and isolating rhythmic signal features from an input signal using the wavelet packet transform |
| US20130262087A1 (en) * | 2012-03-29 | 2013-10-03 | Kabushiki Kaisha Toshiba | Speech synthesis apparatus, speech synthesis method, speech synthesis program product, and learning apparatus |
| CN103823180A (en) * | 2014-02-27 | 2014-05-28 | 国家电网公司 | Method for diagnosing mechanical faults of distribution switch |
| CN105136454A (en) * | 2015-10-15 | 2015-12-09 | 上海电机学院 | Wind turbine gear box fault recognition method |
| CN106482937A (en) * | 2016-09-30 | 2017-03-08 | 南方电网科学研究院有限责任公司 | A kind of monitoring method of mechanical state of high-voltage circuit breaker |
| CN107101813A (en) * | 2017-04-26 | 2017-08-29 | 河北工业大学 | A kind of frame-type circuit breaker mechanical breakdown degree assessment method based on vibration signal |
-
2017
- 2017-08-31 CN CN201710774736.8A patent/CN107607302B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100114813A1 (en) * | 2008-10-20 | 2010-05-06 | Zalay Osbert C | Method and rhythm extractor for detecting and isolating rhythmic signal features from an input signal using the wavelet packet transform |
| US20130262087A1 (en) * | 2012-03-29 | 2013-10-03 | Kabushiki Kaisha Toshiba | Speech synthesis apparatus, speech synthesis method, speech synthesis program product, and learning apparatus |
| CN103823180A (en) * | 2014-02-27 | 2014-05-28 | 国家电网公司 | Method for diagnosing mechanical faults of distribution switch |
| CN105136454A (en) * | 2015-10-15 | 2015-12-09 | 上海电机学院 | Wind turbine gear box fault recognition method |
| CN106482937A (en) * | 2016-09-30 | 2017-03-08 | 南方电网科学研究院有限责任公司 | A kind of monitoring method of mechanical state of high-voltage circuit breaker |
| CN107101813A (en) * | 2017-04-26 | 2017-08-29 | 河北工业大学 | A kind of frame-type circuit breaker mechanical breakdown degree assessment method based on vibration signal |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108955871A (en) * | 2018-08-27 | 2018-12-07 | 河海大学 | A kind of tap switch vibration signal noise-reduction method based on quick kurtogram algorithm |
| CN114936582A (en) * | 2022-06-08 | 2022-08-23 | 华侨大学 | Working modal parameter identification method and system and fault position identification method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107607302B (en) | 2018-07-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106482937B (en) | A kind of monitoring method of mechanical state of high-voltage circuit breaker | |
| Abdullah | Ultrafast transmission line fault detection using a DWT-based ANN | |
| CN107817098B (en) | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method | |
| CN110503004B (en) | On-line judging method for operating state of switching power supply | |
| CN104655423B (en) | A kind of Fault Diagnosis of Roller Bearings based on time-frequency domain multi-dimensional vibration Fusion Features | |
| CN104181460B (en) | A kind of permanent magnetic actuator vacuum circuit breaker method for diagnosing faults based on Multi-information acquisition | |
| CN108229544A (en) | A kind of NPC three-level photovoltaic inverters open fault diagnostic method | |
| CN108362488A (en) | OLTC mechanical failure diagnostic methods based on MPE and SVM | |
| CN103308292A (en) | Vacuum breaker mechanical state detecting method based on vibration signal analysis | |
| EP3723218B1 (en) | Method and device for handling direct current arc | |
| CN102998545A (en) | Online transformer winding work state monitoring method | |
| CN102222151B (en) | Analog circuit fault prediction method based on ARMA (Autoregressive Moving Average) | |
| CN108267687B (en) | Based on probability density and the Mechanical Failure of HV Circuit Breaker diagnostic method being locally linear embedding into | |
| CN103823180A (en) | Method for diagnosing mechanical faults of distribution switch | |
| CN108398252A (en) | OLTC mechanical failure diagnostic methods based on ITD and SVM | |
| CN107607302B (en) | The Mechanical Failure of HV Circuit Breaker diagnostic method of statistical analysis is scaled based on multidimensional | |
| CN108205662A (en) | Power transmission line lightning shielding counterattack recognition methods based on Pearson came algorithm | |
| CN105703258A (en) | Action state monitoring system of gas insulated switchgear (GIS) and application method of action state monitoring system | |
| CN104636746B (en) | A kind of GIS device Eigenvalue Extraction Method based on regional area wavelet transformation | |
| CN108344922A (en) | A kind of transmission line of electricity direct lightning strike fault recognition method based on similar differentiation and S-transformation | |
| CN103018537A (en) | Disturbance classification and recognition method for quality of transient electric energy based on CWD Choi-Williams Distribution spectral kurtosis | |
| Yang et al. | Fault diagnosis of circuit breakers based on time–frequency and chaotic vibration analysis | |
| CN105893976A (en) | Parameterization identification method of traveling wave signal | |
| Wan et al. | Fault diagnosis for high voltage circuit breaker based on timing parameters and FCM | |
| CN108919041B (en) | Transformer winding state online monitoring method based on cluster analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |