CN107817098B - A kind of Mechanical Failure of HV Circuit Breaker diagnostic method - Google Patents
A kind of Mechanical Failure of HV Circuit Breaker diagnostic method Download PDFInfo
- Publication number
- CN107817098B CN107817098B CN201710949455.1A CN201710949455A CN107817098B CN 107817098 B CN107817098 B CN 107817098B CN 201710949455 A CN201710949455 A CN 201710949455A CN 107817098 B CN107817098 B CN 107817098B
- Authority
- CN
- China
- Prior art keywords
- signal
- voltage circuitbreaker
- sample
- failure
- voltage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/327—Testing of circuit interrupters, switches or circuit-breakers
- G01R31/3271—Testing of circuit interrupters, switches or circuit-breakers of high voltage or medium voltage devices
- G01R31/3275—Fault detection or status indication
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Electric Properties And Detecting Electric Faults (AREA)
Abstract
The present invention is a kind of Mechanical Failure of HV Circuit Breaker diagnostic method, its main feature is that: including to high-voltage circuitbreaker vibration signals collecting, the processing of high-voltage circuitbreaker vibration signal temporal segmentation, after extracting Tsallis entropy feature to splitting signal, high-voltage circuit-breaker status is diagnosed using the stratification hybrid classifer being made of OCSVM and ELM.Compared with traditional multi-categorizer, due to the stratification hybrid classifer of use, it can be avoided and fault sample is mistakenly identified as normal condition, improve machine state identification effect, and multi-level hybrid classifer can accurately identify no training sample malfunction, find emerging failure in electrical equipment.It is reasonable with methodological science, the advantages that strong applicability, accuracy is high, and effect is good.
Description
Technical field
The present invention is a kind of Mechanical Failure of HV Circuit Breaker diagnostic method, is applied to Mechanical Failure of HV Circuit Breaker state and exists
Radiodiagnosis x and unknown failure type identification.
Background technique
High-voltage circuitbreaker (Highvoltage circuitbreaker, HVCB) mechanical breakdown inline diagnosis is breaker peace
The important prerequisite of row for the national games finds security risk present in breaker operation, repair and maintenance is carried out to fault point in time, to dimension
Shield power system security stabilization is of great significance.With the arrival of Internet era, people monitor and examine for electrical equipment
The requirement for repairing technology is higher and higher, and traditional detection technique can not reach requirement at many aspects, it may appear that maintenance is insufficient or
Person is phenomena such as blindness is repaired, such that resource serious waste, and result is also unsatisfactory.So in electrical equipment
Maintenance in, there is necessity with on-line monitoring and Condition-Based Maintenance Technology, and era development paces are complied with, wherein important
A step be fault diagnosis.Usual breaker mechanical fault diagnosis generally comprises signal processing, feature extraction and fault diagnosis three
A link.In the prior art, signal processing method carries out time-domain and frequency-domain analysis to original signal, and extracts time and frequency zone information.
It mainly include Empirical mode decomposition, dynamic time warping, short-time energy analytic approach, wavelet packet analysis method, zero-phase filtering
Method etc..Although achieving some achievements, there is certain deficiency, such as: easily there is envelope, owes envelope in empirical mode decomposition
And the problems such as modal overlap；The variation that dynamic time warping and short-time energy analytic approach only occur signal at any time is more
It is sensitive；When wavelet packet etc. is traditional-frequency analysis method do not have adaptivity, it has difficulties in the selection of basic function, and the time
It cannot be considered in terms of with frequency resolution.Meanwhile above method treatment process is more complex, time complexity is high, improves the relevant technologies
Application cost and industrialization difficulty.Common vibration signal characteristics include temporal signatures, frequency domain character and time and frequency zone feature.
Although existing feature accurately can effectively describe different conditions signal, the failure of high-voltage circuitbreaker vibration signal is special
It is wide to levy frequency domain distribution, and is installed etc. and to influence in real work, it is difficult to from specific frequency domain extraction correlated characteristic.Fault diagnosis
Common method include neural network, support vector machine etc., extreme learning machine etc..Neural network and support vector machines can identify
Known multiclass state, and accuracy rate is higher.But since Mechanical Failure of HV Circuit Breaker type is more, action frequency is few, and therefore
It is higher to hinder sample experiment procurement cost, thus is difficult to obtain the training sample comprising all fault types.In practical applications, it adopts
It is easily missed when with more classification classifier identification UNKNOWN TYPE malfunctions such as neural network or support vector machines and minor failure
Sentence, i.e., UNKNOWN TYPE failure is mistakenly identified as to the known fault type of normal condition or mistake.
Summary of the invention
The purpose of the present invention is to provide a kind of scientific and reasonable, strong applicabilities, and accuracy is high, the good high-voltage circuitbreaker of effect
Mechanical failure diagnostic method.
The purpose of the present invention is realized by the following technical means: a kind of Mechanical Failure of HV Circuit Breaker diagnosis side
Method, it is characterized in that: it includes carrying out temporal segmentation processing to original signal, after extracting Tsallis entropy feature to splitting signal, benefit
High-voltage circuit-breaker status is diagnosed with the stratification hybrid classifer being made of OCSVM and ELM, specifically includes following step
It is rapid:
1) high-voltage circuitbreaker vibration signals collecting:
Vibration signal acquisition system is built using piezoelectric acceleration transducer and data collecting card, signal sampling rate is every
Seconds 25600 points, the co-ordinate zero point of sampled signal be high-voltage circuitbreaker will action moment, i.e. trigger card is that data acquisition card is sent
Signal acquisition instruction, samples 150 milliseconds altogether；
2) high-voltage circuitbreaker vibration signal temporal segmentation is handled:
Based on high-voltage circuitbreaker original vibration signal, triggering command is received to vibration signal with operating mechanism under normal condition
When a length of temporal segmentation scale when significant change occurs for amplitude, and step-length is 130 sampled points, and signal is divided into 29 sections, point
Analysis total length of data is 3770 sampled points, carries out same scale to signal after each high-voltage circuitbreaker movement acquisition vibration signal
Temporal segmentation processing；
3) high-voltage circuitbreaker vibration signal characteristics extract:
After carrying out temporal segmentation processing to the high-voltage circuitbreaker vibration signal of acquisition, Tsallis entropy is extracted to every segment signal
Feature constitutes the feature vector of malfunction identification, and since temporal segmentation divides the signal into 29 sections, feature vector length is
29 dimensions, Tsallis entropy calculation formula are as follows:
Wherein HαIndicate the feature vector that every section of Tsallis entropy is constituted, piFor the energy of segment signal every after temporal segmentation
Amount accounts for the ratio of signal gross energy, takes adjustability coefficients α=0.4, then Hα=[Hα(p1) ..., Hα(p29)], i.e. HαFor high pressure open circuit
The feature vector of device fault diagnosis, α is non-extensive parameter, p in formulaiIt is the probability density distribution of stochastic variable i, and
IfFor the gross energy of signal x (t), whereinFor i-th section of energy, y (j) is the of signal
I sections of original time domain splitting signals, i=1 ... 29, since signal temporal segmentation is 29 sections, so n=29, if pi=E (i)/E and
4) stratification hybrid classifer identifies high-voltage circuit-breaker status:
Field is monitored in mechanical state of high-voltage circuit breaker, high-voltage circuitbreaker normal condition sample is easy to obtain, with single point
Class device can identify mechanical abnormal state, different with all kinds of different samples of comprehensive identification from traditional classifier, one-class classifier
It identifies that target is as far as possible to come out improper specimen discerning, that is, avoids for malfunction being mistakenly identified as normal condition, it is therefore, single
Class classifier is suitable for requiring the HVCB machine performance of high reliability to monitor field, for the mechanical event of multi classifier identification HVCB
Barrier can not identify specific fault type, adopt though misrecognition and OCSVM, which easily occurs, can accurately monitor mechanical malfunction
It is further accurately identified on the basis of avoiding the misrecognition of malfunction with OCSVM and ELM building stratification hybrid classifer
Without training sample unknown failure type；Each training sample passes through the available one 29 dimension input feature value sample of step 3)
Collection contains normal shape to high-voltage circuitbreaker difference operating status using the sample set as the input variable of stratification hybrid classifer
State and abnormal condition will carry out feature vector calculating respectively, and the one-class support vector machines mould under different conditions is respectively trained
Type；With breaker different faults state feature vector training ELM model；When needing to diagnose the failure of high-voltage circuitbreaker, will adopt
The diagnostic sample collected obtains input vector by step 3), is first input into normal condition OCSVM model and judges high-voltage circuitbreaker
Whether break down；If it is determined that sample, then is input to ELM model again and judges which occurs for high-voltage circuitbreaker by malfunction
Kind failure；After the completion of diagnosis, then sample is input in the OCSVM model of corresponding failure state, judges whether malfunction is known
It is incorrect, it is otherwise no training sample unknown failure type.
A kind of Mechanical Failure of HV Circuit Breaker diagnostic method of the invention, firstly, receiving triggering letter with standard signal vibration
Number to vibration start the period be time scale, to original signal carry out temporal segmentation；Secondly, being asked respectively segment signal each after segmentation
Temporal signatures are taken, vibration signal characteristics vector is constituted；On the basis of analyzing different characteristic classification capacity, determine that Tsallis entropy is special
Sign is used as fault diagnosis feature, constitutes temporal aspect vector；Finally, building is based on one-class support vector machines (One-class
Support Vector Machine, OCSVM) with the mixing of extreme learning machine (Extreme Learning Machine, ELM)
Normal condition, known fault conditions and unknown failure state are effectively distinguished and identified to classifier, when can be avoided-frequency handle when
High-frequency information is lost, guarantees the integrality of characteristic information, meanwhile, a large amount of signal processing time is saved, with traditional more classification
Device is compared, and due to the stratification hybrid classifer that the present invention uses, can be avoided fault sample being mistakenly identified as normal condition, mention
High machine state identification effect, and multi-level hybrid classifer can accurately identify no training sample malfunction, find
Emerging failure in electrical equipment.It is reasonable with methodological science, the advantages that strong applicability, accuracy is high, and effect is good.
Detailed description of the invention
Fig. 1 is a kind of Mechanical Failure of HV Circuit Breaker diagnostic method flow chart；
Fig. 2 is high-voltage circuitbreaker normal state signal temporal segmentation figure；
Fig. 3 is high-voltage circuitbreaker iron core bite status signal temporal segmentation figure；
Fig. 4 is high-voltage circuitbreaker loosened screw status signal temporal segmentation figure；
Fig. 5 is high-voltage circuitbreaker lack of lubrication status signal temporal segmentation figure；
Fig. 6 is that high-voltage circuitbreaker normally shows schematic diagram with three kinds of malfunction Tsallis entropy features.
Specific embodiment
The present invention is described further in the following with reference to the drawings and specific embodiments.
A kind of Mechanical Failure of HV Circuit Breaker diagnostic method of the invention, it includes carrying out at temporal segmentation to original signal
Reason, after extracting Tsallis entropy feature to splitting signal, using the stratification hybrid classifer being made of OCSVM and ELM to high pressure
Circuit-breaker status is diagnosed, specifically includes the following steps:
1) high-voltage circuitbreaker vibration signals collecting:
9234 data collecting card of NI produced using piezoelectric acceleration transducer and American National instrument (NI) company,
Vibration signal acquisition system is built, signal sampling rate is 25600 points per second, and the co-ordinate zero point of sampled signal will for high-voltage circuitbreaker
Action moment (trigger card is that data acquisition card sends signal acquisition instruction) is wanted, samples 150 milliseconds altogether；
2) high-voltage circuitbreaker vibration signal temporal segmentation is handled:
Based on high-voltage circuitbreaker original vibration signal, triggering command is received to vibration signal with operating mechanism under normal condition
When a length of temporal segmentation scale when significant change occurs for amplitude, and step-length is 130 sampled points, and signal is divided into 29 sections, point
Analysis total length of data is 3770 sampled points, carries out same scale to signal after each high-voltage circuitbreaker movement acquisition vibration signal
Temporal segmentation processing；
3) high-voltage circuitbreaker vibration signal characteristics extract:
After carrying out temporal segmentation processing to the high-voltage circuitbreaker vibration signal of acquisition, Tsallis entropy is extracted to every segment signal
Feature constitutes the feature vector of malfunction identification, and since temporal segmentation divides the signal into 29 sections, feature vector length is
29 dimensions, Tsallis entropy calculation formula are as follows:
Wherein HαIndicate every section of Tsallis entropy, piEnergy for segment signal every after temporal segmentation accounts for signal gross energy
Ratio takes adjustability coefficients α=0.4, then H=[Hα(p1) ..., Hα(p29)], i.e. H be Fault Diagnosis for HV Circuit Breakers feature to
Amount,
α is non-extensive parameter, p in formulaiIt is the probability density distribution of stochastic variable i, andIfFor letter
The gross energy of number x (t), whereinFor i-th section of energy, y (i) is i-th section of original time domain point of signal
Cut signal, i=1 ... 29, since signal temporal segmentation is 29 sections, so n=29, if pi=E (i)/E and
4) stratification hybrid classifer identifies high-voltage circuit-breaker status:
Field is monitored in mechanical state of high-voltage circuit breaker, high-voltage circuitbreaker normal condition sample is easy to obtain, with single point
Class device can identify mechanical abnormal state, different with all kinds of different samples of comprehensive identification from traditional classifier, one-class classifier
It identifies that target is as far as possible to come out improper specimen discerning, that is, avoids for malfunction being mistakenly identified as normal condition, it is therefore, single
Class classifier is suitable for requiring the HVCB machine performance of high reliability to monitor field, for the mechanical event of multi classifier identification HVCB
Barrier can not identify specific fault type, adopt though misrecognition and OCSVM, which easily occurs, can accurately monitor mechanical malfunction
It is further accurately identified on the basis of avoiding the misrecognition of malfunction with OCSVM and ELM building stratification hybrid classifer
Without training sample unknown failure type；Each training sample passes through the available one 29 dimension input feature value sample of step 3)
Collection contains normal shape to high-voltage circuitbreaker difference operating status using the sample set as the input variable of stratification hybrid classifer
State and abnormal condition will carry out feature vector calculating respectively, and the one-class support vector machines mould under different conditions is respectively trained
Type；With breaker different faults state feature vector training ELM model；When needing to diagnose the failure of high-voltage circuitbreaker, will adopt
The diagnostic sample collected obtains input vector by step 3), is first input into normal condition OCSVM model and judges high-voltage circuitbreaker
Whether break down；If it is determined that sample, then is input to ELM model again and judges which occurs for high-voltage circuitbreaker by malfunction
Kind failure；After the completion of diagnosis, then sample is input in the OCSVM model of corresponding failure state, judges whether malfunction is known
It is incorrect, it is otherwise no training sample unknown failure type.
Specific embodiment: referring to Fig.1, a kind of Mechanical Failure of HV Circuit Breaker diagnostic method of specific embodiment, it includes
Temporal segmentation processing is carried out to original signal, after extracting Tsallis entropy feature to splitting signal, is formed using by OCSVM and ELM
Stratification hybrid classifer high-voltage circuit-breaker status is diagnosed, specifically includes the following steps:
A, high-voltage circuitbreaker vibration signals collecting:
9234 data collecting card of NI produced using piezoelectric acceleration transducer and American National instrument (NI) company,
Vibration signal acquisition system is built, signal sampling rate is 25600 points per second, and the co-ordinate zero point of sampled signal will for high-voltage circuitbreaker
Action moment (trigger card is that data acquisition card sends signal acquisition instruction) is wanted, samples 150 milliseconds altogether；As Figure 2-Figure 5, root
According to action situation of the high-voltage circuitbreaker under different conditions, the present embodiment acquires the vibration letter of the high-voltage circuitbreaker under four kinds of states
Number, the vibration signal including normal condition and three kinds of malfunctions (iron core bite, loosened screw refer to lack of lubrication)；
B, high-voltage circuitbreaker vibration signal temporal segmentation is handled:
Vibration signal temporal segmentation with operating mechanism under normal condition receive triggering command to vibration signal amplitude occur it is bright
When a length of temporal segmentation scale when aobvious variation, is divided into 29 sections for signal, from four kinds of state segmentation figures it can be seen that in difference
Duan Zhong, there are difference between signal, and can directly retain original signal analysis the integrality of information；
C, high-voltage circuitbreaker vibration signal characteristics extract:
To temporal segmentation signal extraction Tsallis entropy and constitutive characteristic vector, four kinds of different conditions entropy features performances such as Fig. 6
Shown, the differentiation degree as can be seen from the figure between the feature between different conditions is high；
D, stratification hybrid classifer identifies high-voltage circuit-breaker status:
Feature vector is input in stratification hybrid classifer, high-voltage circuit-breaker status is effectively identified, first by
OCSVM0Single classifier judges whether high-voltage circuitbreaker breaks down, if a failure occurs then with ELM multi-categorizer to failure classes
Type is effectively identified, judges that kth class failure occurs for high-voltage circuitbreaker.Classifier will be without training sample unknown failure in order to prevent
Type identification is known fault type, then uses OCSVMkJudge that malfunction identifies whether correctly, is otherwise unknown failure class
Type；
E, high-voltage circuit-breaker status recognition effect:
High-voltage circuit-breaker status is identified into experimental analysis using MATLAB software, although ELM can be to as seen from Table 1
Know that fault type is correctly identified, but unknown failure type can be carried out to be mistakenly identified as normal condition.It can from table 2
O-E (OCSVM-ELM), will be without training although can make up for it the shortcomings that unknown state is mistakenly identified as normal condition by ELM out
Sample unknown failure type is mistakenly identified as known fault type.O-E-O (OCSVM-ELM-OCSVM) energy as can be seen from Table 3
It is enough that the recognition result of O-E is corrected, accurately identify unknown failure.
1 ELM of table is containing no training sample type fault recognition result
2 O-E of table is containing no training sample type fault recognition result
3 O-E-O of table is containing no training sample type fault recognition result
A specific embodiment of the invention is not exhaustive, and any simple duplication and the improvement without creativeness belong to
In the range of the claims in the present invention protection.
Claims (1)
1. a kind of Mechanical Failure of HV Circuit Breaker diagnostic method, it is characterized in that: it includes carrying out at temporal segmentation to original signal
Reason, after extracting Tsallis entropy feature to splitting signal, using the stratification hybrid classifer being made of OCSVM and ELM to high pressure
Circuit-breaker status is diagnosed, specifically includes the following steps:
1) high-voltage circuitbreaker vibration signals collecting:
Vibration signal acquisition system is built using piezoelectric acceleration transducer and data collecting card, signal sampling rate is per second
25600 points, the co-ordinate zero point of sampled signal be high-voltage circuitbreaker will action moment, i.e. trigger card is that data acquisition card sends letter
Number acquisition instructions sample 150 milliseconds altogether；
2) high-voltage circuitbreaker vibration signal temporal segmentation is handled:
Based on high-voltage circuitbreaker original vibration signal, triggering command is received to vibration signal amplitude with operating mechanism under normal condition
When a length of temporal segmentation scale when significant change occurs, step-length is 130 sampled points, and signal is divided into 29 sections, analyzes number
It is 3770 sampled points according to total length, same scale time domain is carried out to signal after each high-voltage circuitbreaker movement acquisition vibration signal
Dividing processing；
3) high-voltage circuitbreaker vibration signal characteristics extract:
After carrying out temporal segmentation processing to the high-voltage circuitbreaker vibration signal of acquisition, Tsallis entropy feature is extracted to every segment signal,
The feature vector for constituting malfunction identification, since temporal segmentation divides the signal into 29 sections, feature vector length is 29 dimensions,
Tsallis entropy calculation formula are as follows:
Wherein HαIndicate the feature vector that every section of Tsallis entropy is constituted, piEnergy for segment signal every after temporal segmentation accounts for
The ratio of signal gross energy takes adjustability coefficients α=0.4, then Hα=[Hα(p1) ..., Hα(p29)], i.e. HαFor high-voltage circuitbreaker event
Hinder the feature vector of diagnosis, α is non-extensive parameter, p in formulaiIt is the probability density distribution of stochastic variable i, andIfFor the gross energy of signal x (t), whereinFor i-th section of energy, y (j) is the i-th of signal
Section original time domain splitting signal, i=1 ... 29, since signal temporal segmentation is 29 sections, so n=29, if pi=E (i)/E and
4) stratification hybrid classifer identifies high-voltage circuit-breaker status:
Field is monitored in mechanical state of high-voltage circuit breaker, high-voltage circuitbreaker normal condition sample is easy to obtain, with single classifier
Mechanical abnormal state can be identified, from traditional classifier different, identification of one-class classifier with all kinds of different samples of comprehensive identification
Target is as far as possible to come out improper specimen discerning, that is, avoids for malfunction being mistakenly identified as normal condition, therefore, single class point
Class device is suitable for requiring the HVCB machine performance of high reliability to monitor field, identifies HVCB mechanical breakdown for multi classifier,
Though misrecognition and OCSVM, which easily occurs, can accurately monitor mechanical malfunction, it can not identify specific fault type, use
OCSVM and ELM building stratification hybrid classifer further accurately identifies nothing on the basis of avoiding the misrecognition of malfunction
Training sample unknown failure type；Each training sample passes through the available one 29 dimension input feature value sample of step 3)
Collection contains normal shape to high-voltage circuitbreaker difference operating status using the sample set as the input variable of stratification hybrid classifer
State and abnormal condition will carry out feature vector calculating respectively, and the one-class support vector machines mould under different conditions is respectively trained
Type；With breaker different faults state feature vector training ELM model；When needing to diagnose the failure of high-voltage circuitbreaker, will adopt
The diagnostic sample collected obtains input vector by step 3), is first input into normal condition OCSVM model and judges high-voltage circuitbreaker
Whether break down；If it is determined that sample, then is input to ELM model again and judges which occurs for high-voltage circuitbreaker by malfunction
Kind failure；After the completion of diagnosis, then sample is input in the OCSVM model of corresponding failure state, judges whether malfunction is known
It is incorrect, it is otherwise no training sample unknown failure type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710949455.1A CN107817098B (en) | 2017-10-12 | 2017-10-12 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710949455.1A CN107817098B (en) | 2017-10-12 | 2017-10-12 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107817098A CN107817098A (en) | 2018-03-20 |
| CN107817098B true CN107817098B (en) | 2019-06-28 |
Family
ID=61608290
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710949455.1A Active CN107817098B (en) | 2017-10-12 | 2017-10-12 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107817098B (en) |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109100646A (en) * | 2018-08-17 | 2018-12-28 | 国网江苏省电力有限公司检修分公司 | A kind of Fault Diagnosis for HV Circuit Breakers method |
| CN110824344A (en) * | 2019-10-22 | 2020-02-21 | 国网浙江省电力有限公司检修分公司 | High-voltage circuit breaker state evaluation method based on vibration signal short-time energy-entropy ratio and DTW |
| CN110796120A (en) * | 2019-11-21 | 2020-02-14 | 杭州电力设备制造有限公司 | Time domain feature-based circuit breaker mechanical fault XGboost diagnosis method |
| CN111504627B (en) * | 2020-05-07 | 2022-04-29 | 山东泰开高压开关有限公司 | Method for detecting defective parts of circuit breaker |
| CN112183590A (en) * | 2020-09-14 | 2021-01-05 | 浙江大学 | Transformer fault diagnosis method based on Oneclass SVM algorithm |
| CN112557004B (en) * | 2020-11-23 | 2022-06-14 | 广西电网有限责任公司电力科学研究院 | Method for detecting mechanical fault of high-voltage circuit breaker |
| CN113341306B (en) * | 2021-04-19 | 2022-11-11 | 云南电网有限责任公司临沧供电局 | High-voltage circuit breaker state evaluation system and method based on vibration signals |
| CN113315099B (en) * | 2021-05-20 | 2022-10-21 | 国网宁夏电力有限公司检修公司 | Fault early warning method, medium and system for converter station alternating current filter circuit breaker |
| CN113341813B (en) * | 2021-06-11 | 2022-11-01 | 上海天麦能源科技有限公司 | Urban gas medium-low pressure pipe network detection method and system |
| CN113705637B (en) * | 2021-08-12 | 2024-04-02 | 西安交通大学 | Method, system, equipment and readable storage medium for detecting mechanical faults of circuit breaker |
| CN113739909B (en) * | 2021-09-08 | 2024-08-06 | 上海海事大学 | Ship rotary machine fault diagnosis method based on full-scale time domain averaging |
| CN113884292B (en) * | 2021-10-09 | 2023-07-21 | 西安西电开关电气有限公司 | SVDD-based mechanical fault diagnosis method and system for switchgear |
| CN114662548B (en) * | 2022-04-12 | 2023-06-20 | 安徽中安昊源电力科技有限公司 | Circuit breaker diagnosis method and system based on abnormal actions |
| CN117591876A (en) * | 2023-11-17 | 2024-02-23 | 广东电网有限责任公司 | Breaker fault diagnosis method, device, equipment and storage medium |
| CN117970091A (en) * | 2023-12-29 | 2024-05-03 | 广东电网有限责任公司江门供电局 | Fault monitoring method based on multi-classification algorithm |
| CN117872123A (en) * | 2024-01-24 | 2024-04-12 | 广东电网有限责任公司江门供电局 | High-voltage circuit breaker fault diagnosis method based on mechanical vibration signals |
| CN118070111B (en) * | 2024-04-24 | 2024-08-02 | 青岛理工大学 | Automobile health monitoring method based on multi-layer parallel multi-neural network fusion |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103308292A (en) * | 2013-05-16 | 2013-09-18 | 国家电网公司 | Vacuum breaker mechanical state detecting method based on vibration signal analysis |
| CN103575525A (en) * | 2013-11-18 | 2014-02-12 | 东南大学 | Intelligent diagnosis method for mechanical fault of circuit breaker |
| CN105241643A (en) * | 2015-09-19 | 2016-01-13 | 东北电力大学 | High-voltage circuit breaker mechanical state monitoring method employing HS transformation and single-type support vector machine |
| CN106199412A (en) * | 2016-07-01 | 2016-12-07 | 太原理工大学 | A kind of permanent magnet mechanism high-pressure vacuum breaker method of fault pattern recognition |
| CN106482937A (en) * | 2016-09-30 | 2017-03-08 | 南方电网科学研究院有限责任公司 | Method for monitoring mechanical state of high-voltage circuit breaker |
-
2017
- 2017-10-12 CN CN201710949455.1A patent/CN107817098B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103308292A (en) * | 2013-05-16 | 2013-09-18 | 国家电网公司 | Vacuum breaker mechanical state detecting method based on vibration signal analysis |
| CN103575525A (en) * | 2013-11-18 | 2014-02-12 | 东南大学 | Intelligent diagnosis method for mechanical fault of circuit breaker |
| CN105241643A (en) * | 2015-09-19 | 2016-01-13 | 东北电力大学 | High-voltage circuit breaker mechanical state monitoring method employing HS transformation and single-type support vector machine |
| CN106199412A (en) * | 2016-07-01 | 2016-12-07 | 太原理工大学 | A kind of permanent magnet mechanism high-pressure vacuum breaker method of fault pattern recognition |
| CN106482937A (en) * | 2016-09-30 | 2017-03-08 | 南方电网科学研究院有限责任公司 | Method for monitoring mechanical state of high-voltage circuit breaker |
Non-Patent Citations (2)
| Title |
|---|
| 基于声振联合特征熵的断路器故障诊断方法;赵书涛等;《华北电力大学学报》;20161130;第43卷(第6期);全文 |
| 基于振动信号特征的高压断路器机械故障诊断技术研究;钟建英;《高压电器》;20130930;第49卷(第9期);全文 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107817098A (en) | 2018-03-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107817098B (en) | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method | |
| CN110926778B (en) | Mechanical fault diagnosis method for gas insulated switchgear assembly based on abnormal vibration | |
| CN106482937B (en) | Method for monitoring mechanical state of high-voltage circuit breaker | |
| CN106443335A (en) | Lightning stroke fault identification method and system | |
| CN109270441A (en) | High-voltage circuitbreaker separating brake characterisitic parameter online test method based on vibration signal | |
| CN105528741B (en) | Circuit breaker state identification method based on multi-signal feature fusion | |
| CN106646096A (en) | Transformer fault classification and identification method based on vibration analysis method | |
| CN105547465A (en) | Transformer vibration signal winding state feature extraction method | |
| CN103175897B (en) | A kind of high-speed switch hurt recognition methods based on vibration signal end-point detection | |
| CN107367687A (en) | A kind of high-voltage breaker spring fault degree detection method and device | |
| CN105241643B (en) | Mechanical state of high-voltage circuit breaker monitoring method with one-class support vector machines is converted using HS | |
| CN109375010A (en) | Winding in Power Transformer method for diagnosing faults based on vibration distribution characteristic model | |
| CN107907320A (en) | The diagnostic method and device of the machine performance feature of on-load tap changers of transformers | |
| CN105676120A (en) | High voltage circuit breaker synchronism testing device and method based on abrupt change of vibration energy | |
| CN104636746B (en) | A kind of GIS device Eigenvalue Extraction Method based on regional area wavelet transformation | |
| CN108362488A (en) | OLTC mechanical failure diagnostic methods based on MPE and SVM | |
| CN111678699A (en) | Early fault monitoring and diagnosing method and system for rolling bearing | |
| CN112083328A (en) | Fault diagnosis method, system and device for high-voltage circuit breaker | |
| CN112557966A (en) | Transformer winding looseness identification method based on local mean decomposition and support vector machine | |
| CN113551895A (en) | Comprehensive intelligent diagnosis method for mechanical fault of on-load tap-changer | |
| CN108267687B (en) | Based on probability density and the Mechanical Failure of HV Circuit Breaker diagnostic method being locally linear embedding into | |
| CN113901862A (en) | Electromagnetic repulsion mechanism fault monitoring method, system, equipment and readable storage medium | |
| CN110426192A (en) | Acoustic fingerprint detection system and method for circuit breaker | |
| CN113884292B (en) | SVDD-based mechanical fault diagnosis method and system for switchgear | |
| CN107607302B (en) | The Mechanical Failure of HV Circuit Breaker diagnostic method of statistical analysis is scaled based on multidimensional |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221116 Address after: 130021 no.4799, Renmin Street, Chaoyang District, Changchun City, Jilin Province Patentee after: ECONOMIC TECHNOLOGY RESEARCH INSTITUTE OF STATE GRID JILIN ELECTRIC POWER CO.,LTD. Patentee after: Jilin Province Beitian Gong Software Development Co.,Ltd. Address before: 132012, Changchun Road, Jilin, Jilin, 169 Patentee before: NORTHEAST DIANLI University |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20221221 Address after: 132012, Changchun Road, Jilin, Jilin, 169 Patentee after: NORTHEAST DIANLI University Patentee after: ECONOMIC TECHNOLOGY RESEARCH INSTITUTE OF STATE GRID JILIN ELECTRIC POWER CO.,LTD. Patentee after: Jilin Province Beitian Gong Software Development Co.,Ltd. Address before: 130021 no.4799, Renmin Street, Chaoyang District, Changchun City, Jilin Province Patentee before: ECONOMIC TECHNOLOGY RESEARCH INSTITUTE OF STATE GRID JILIN ELECTRIC POWER CO.,LTD. Patentee before: Jilin Province Beitian Gong Software Development Co.,Ltd. |