CN108169643A - A kind of method and system for cable local discharge pattern-recognition - Google Patents
A kind of method and system for cable local discharge pattern-recognition Download PDFInfo
- Publication number
- CN108169643A CN108169643A CN201810088179.9A CN201810088179A CN108169643A CN 108169643 A CN108169643 A CN 108169643A CN 201810088179 A CN201810088179 A CN 201810088179A CN 108169643 A CN108169643 A CN 108169643A
- Authority
- CN
- China
- Prior art keywords
- matrix
- data
- feature value
- characteristic value
- recognition
- 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
- 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
- G01R31/12—Testing dielectric strength or breakdown voltage ; Testing or monitoring effectiveness or level of insulation, e.g. of a cable or of an apparatus, for example using partial discharge measurements; Electrostatic testing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Relating To Insulation (AREA)
Abstract
The invention discloses a kind of method and systems for cable local discharge pattern-recognition, data volume needed for calculating can be reduced, shorten recognition time, the discharge characteristic of different shelf depreciations can be effectively obtained simultaneously, under the premise of characteristic parameter is not lost, more significant characteristic value is obtained from discreet portions, improves Classification and Identification rate.This method includes:For each group of data, the data in the wherein m period are taken, are superimposed in one cycle to form the spectrum data of every group of data;To 360 ° of progress phase window divisions in the period as unit of equal angular, n equidistant phase windows are obtained;Spectrum data in calculating cycle corresponds to the characteristic value in each phase window, obtains fisrt feature value matrix;Dimension-reduction treatment is carried out to the fisrt feature value matrix of acquisition, obtains second feature value matrix;Classification and Identification is carried out to the characteristic value in second feature value matrix by pattern recognition classifier device, obtains the corresponding classification results of each sample signal.
Description
Technical field
The present invention relates to power cable partial discharge monitoring technical fields more particularly to one kind to be used for cable local discharge mould
The method and system of formula identification.
Background technology
Since cross-inked polyethylene power cable (XLPE) has preferable electric property and a heat resistance, and its light weight,
Installation is easily and laying is convenient, therefore is largely used for Urban Underground grid power transmission.A large amount of uses of power cable so that therewith
Mating electric cable fitting has also obtained vigorous growth.However, the XLPE power cables in actual motion are due to being installed
The factors such as technique, laying environment, external force destruction, service condition influence, and insulation defect or even dielectric breakdown accident is caused constantly to be sent out
It is raw, wherein being more using the ratio of cable intermediate joint and terminals attachment insulation fault.It generally believes both at home and abroad to XLPE electric power
Cable and its best approach of attachment insulation status evaluation are to carry out partial discharge monitoring, while can carry out pattern-recognition to it
And classification.
Identify and assorting process in, select that appropriate discharge mode feature is extremely important, and result will also directly affect
To the discrimination of grader.For now, in shelf depreciation identification more commonly used characteristic parameter have degree of skewness, steepness,
The statistical natures parameters such as peak value number, initial discharge phase and positive and negative half-wave related coefficient, these parameters are all based on entirely putting
Electric period or the relationship of positive and negative half period.Meanwhile the calculating of these statistical nature parameters needs to draw based on a large amount of data
Picture, obtain its statistical nature parameter and be also required to largely calculate so that local discharge characteristic extraction calculation amount it is huge, speed
Degree is slower.Lead to not obtain significant characteristic value from local discreet portions, obtain moreover, also existing to lose due to characteristic parameter
The more low technical problem of characteristic value accuracy.
Invention content
An object of the present invention at least that, for how to overcome the above-mentioned problems of the prior art, provide one kind
For the method and system of cable local discharge pattern-recognition, data volume needed for calculating can be reduced, shortens recognition time, simultaneously
The discharge characteristic of different shelf depreciations can be effectively obtained, under the premise of characteristic parameter is not lost, is obtained from discreet portions
More significant characteristic value improves Classification and Identification rate.
To achieve these goals, the technical solution adopted by the present invention includes following aspects.
A kind of method for cable local discharge pattern-recognition, including:Obtain the p sample letter of cable local discharge
Number corresponding p groups data;For each group of data, the data in the wherein m period are taken, superposition is every to form in one cycle
The spectrum data of group data;To 360 ° of progress phase window divisions in the period as unit of equal angular, n are obtained at equal intervals
Phase window;Spectrum data in calculating cycle corresponds to the characteristic value in each phase window, obtains fisrt feature value matrix;To obtaining
The fisrt feature value matrix taken carries out dimension-reduction treatment, obtains second feature value matrix;By pattern recognition classifier device to the second spy
Characteristic value in value indicative matrix carries out Classification and Identification, obtains the corresponding classification results of each sample signal.
In conclusion by adopting the above-described technical solution, the present invention at least has the advantages that:
By being overlapped to phase cycling signal, region segmentation is then angularly carried out, it is more aobvious to each extracted region
The characteristic value of work carries out dimension-reduction treatment, the several spies having compared with high-class ability of extraction after the characteristic value for obtaining higher dimensional
Parameter is levied, for local discharge signal type identification, therefore can effectively be obtained while huge data volume is not needed to
Under the premise of characteristic parameter is not lost, more significant feature is obtained from discreet portions for the discharge characteristic of different shelf depreciations
Value improves the discrimination of shelf depreciation pattern;It so that can be according to cable accessory corresponding with shelf depreciation pattern typical case
Defect to take measures in time, avoids the generation of dielectric breakdown accident or reduces its coverage.
Description of the drawings
Fig. 1 is the flow chart of the method for cable local discharge pattern-recognition according to embodiments of the present invention.
Fig. 2 is the schematic diagram of the spectrum data of every group of data of composition according to embodiments of the present invention.
Fig. 3 is the schematic diagram of acquisition fisrt feature value matrix according to embodiments of the present invention.
Fig. 4 is the structure diagram of the system for cable local discharge pattern-recognition according to embodiments of the present invention.
Specific embodiment
With reference to the accompanying drawings and embodiments, the present invention will be described in further detail, so that the purpose of the present invention, technology
Scheme and advantage are more clearly understood.It should be appreciated that specific embodiment described herein is only to explain the present invention, and do not have to
It is of the invention in limiting.
Fig. 1 shows the flow chart of the method for cable local discharge pattern-recognition according to embodiments of the present invention.It should
The method of embodiment includes the following steps:
Step 101:Obtain the corresponding p groups data of p sample signal of cable local discharge
It specifically, can be by partial discharge monitoring equipment (for example, the oscillation wave test based on pulse current method principle is set
It is standby) directly it is connect with monitored XLPE power cables, local discharge signal is taken by amplifying, filtering, adopting according to what is measured
The signal processings such as sample, coding generate corresponding data.In other embodiments, it can also be read from memory saved
Data.Wherein, p is positive integer, for example, 100.
Step 102:For each group of data, the data in the wherein m period are taken, superposition is every to form in one cycle
The spectrum data of group data
As shown in Fig. 2, the data investigation in the 1st period to m-th of period is formed spectrum data, m is positive integer, such as
It is 10.
Step 103:To 360 ° of progress phase window divisions in the period as unit of equal angular, n are obtained at equal intervals
Phase window
Step 104, the spectrum data in calculating cycle corresponds to the characteristic value in each phase window, obtains the First Eigenvalue
Matrix
Specifically, as shown in figure 3, for i-th of phase window, the corresponding peak charge q of spectrum data is calculatedmax(i), it puts down
Equal charge qmn(i) and the quantity q of partial discharge pulsen(i) as Statistical Operator.And then phase each in a cycle can be passed through
Corresponding 3 Statistical Operators of position window, obtain the eigenvalue matrix of 1 × 3n of spectrum data:
[(qmax(1),qmn(1),qn(1))1…(qmax(n),qmn(n),qn(n))1]1×3n;
Further, for the corresponding p groups data of p sample signal, the eigenvalue matrix of p × 3n can be obtained as
One eigenvalue matrix:
Step 105:Dimension-reduction treatment is carried out to the fisrt feature value matrix of acquisition, obtains second feature value matrix
For example, Multidimensional Scaling (Multidimensional Scaling, MDS), principal component analysis may be used
(Principal Component Analysis, PCA), factorial analysis (Factor Analysis) and independent component analysis
The methods of (Independent Component Analysis, ICA), carries out dimension-reduction treatment to obtain second feature value matrix.
Step 106:Classification and Identification is carried out to the characteristic value in second feature value matrix by pattern recognition classifier device, is obtained
The corresponding classification results of each sample signal
Specifically, BP neural network, support vector machines (Support Vector Machine, SVM) etc. may be used
Forming types recognition classifier, this can be identical with traditional pattern recognition classifier device based on statistical nature parameter.Also,
In preferred embodiment, will Second Eigenvalue square can be obtained according to (such as 60%) a part of in the corresponding data of sample signal
Battle array will obtain second feature value matrix for identifying for training mode recognition classifier according to remaining part.
Although it is special to contain a large amount of more specific higher-dimensions of more details by the fisrt feature value matrix constructed by phase division
Value indicative, but Classification and Identification ability is remained while data volume is reduced by the second feature value matrix constructed by dimension-reduction treatment
Higher characteristic value, the higher characteristic value of classification capacity can be obtained by being combined by dimensionality reduction with phase division, reduced pattern and known
Data volume to be processed needed for other grader, so as to improve classification speed and recognition accuracy.
Hereafter by taking MDS carries out dimension-reduction treatment as an example, dimension-reduction treatment flow according to embodiments of the present invention is carried out specifically
It is bright.
First, based on the fisrt feature value matrix that p sample signal obtains in above-described embodiment, the First Eigenvalue square is calculated
Euclidean distance in battle array between characteristic value, defining matrix Δ is:
Wherein, σijFor ith feature value xiWith with j-th of characteristic value xjThe distance between;
Wherein, dimensions (if for example, with 15 ° divide phase windows, d 72) of the d for fisrt feature value matrix, riqFor i-th of spy
Q-th of characteristic value, r in value indicative vectorjqFor q-th of characteristic value in j-th of feature value vector.
Then, singular value decomposition is carried out to the dual centralization matrix of matrix Δ, obtains the Second Eigenvalue square after dimensionality reduction
Battle array X.
Wherein, the dual centralization matrix Δ of matrix Δ is expressed as:
Wherein,Δ(2)=σij 2, E is unit matrix, and p is characterized the quantity of value vector.
Therefore, the expression formula that can obtain the dual centralization matrix Δ of matrix Δ is:
Since matrix Δ is symmetrical and positive semi-definite matrix, carrying out singular value decomposition to dual centralization matrix Δ can be with table
Up to for:
Wherein, Λ be matrix Δ eigenvalue cluster into diagonal matrix, U be Δ feature vector.To the feature of matrix Δ
Value carries out descending sequence, (for example, k has been able to characteristic feature value well for 10) maximum characteristic value k before selection
Feature vector corresponding with them.
And then according to formulaThe second feature value matrix X after dimensionality reduction can be obtained.
In a specific embodiment, 200 sample signals for including four kinds of cable accessory typical defects are obtained and are corresponded to
200 groups of data.The data in 10 periods in every group of data are taken, are superimposed in one cycle to form the collection of illustrative plates of every group of data
Data;To 360 ° of progress phase window divisions in a cycle as unit of 15 °, 24 equidistant phase windows are obtained, are calculated
Characteristic value of the spectrum data in each phase window obtains 1 × 72 characteristic value, by the first of 200 groups of data acquisitions 200 × 72
Eigenvalue matrix.Dimension-reduction treatment further is carried out to it using MDS methods to obtain second feature value matrix.In other embodiment
In, it can also be by the part (such as 120 groups) in 200 groups of data for training mode recognition classifier, and by 80 groups of data
For identifying.By setting the dimension of second feature value matrix during dimension-reduction treatment, different discriminations can be obtained.
The following table 1 shows the dimension for using BP neural network forming types identification separator and second feature value matrix for 2
In the case of~20, by above-mentioned 200 sample signals correspond to four kinds of cable accessory typical defect types (the uneven A of cable fracture,
Major insulation incised wound B, semi-conductive layer damage C, bubble-discharge D) discrimination and overall discrimination result.
Table 1
The following table 2 show use SVM forming types identification separator and second feature value matrix dimension for 2~20 feelings
Under condition, 200 sample signals are corresponded to the discrimination of four kinds of cable accessory typical defect types and the result of overall discrimination.
Table 2
As can be seen that by the above method of the present invention, it is higher resolution can not only to be extracted in numerous characteristic values
Characteristic value, 6 dimension~7 dimension when, just can be stabilized to more than 90% discrimination, so as to which calculation amount be greatly reduced,
Recognition speed is improved, and pattern-recognition separator is trained according to part sample signal can further improve identification
Rate.The following table 3 shows, the overall discrimination after the stabilization that method according to the above embodiment of the present invention obtains and traditional use
Identification after the statistical natures value stabilizations such as degree of skewness, steepness, peak value number, initial discharge phase and positive and negative half-wave related coefficient
The comparing result of rate.
Table 3
Referring to Fig. 4, a kind of system for cable local discharge pattern-recognition is additionally provided according to an embodiment of the invention,
It includes:Display 601, input-output equipment 602, at least one processor 603 and at least one processor 603 are logical
The memory 604 and the power-supply device 605 for power supply for believing connection;
Wherein, display 601 is used to show classification results;Input-output equipment 602 is corresponding for input sample signal
Data;The memory 604 is stored with the instruction that can be performed by least one processor 603, described instruction by it is described extremely
A few processor 603 performs, so that the method that at least one processor 603 is able to carry out aforementioned any embodiment.
The present invention also provides non-transitory computer-readable mediums, and including the instruction compiled on it, described instruction is used for
Perform the embodiment of aforementioned either method.Computer-readable medium may include any medium, can be read by signal processing apparatus
Go out the middle code for performing and being stored thereon, such as floppy disk, CD, tape or hard disk drive.Such code can include object
Code, source code and/or binary code.The code is usually number, is generally used for handling by traditional numerical data
Processor (such as microprocessor, microcontroller or logic circuit, such as programmable gate array, programmable logic circuit/device or special collection
Into circuit [ASIC]).
It should be appreciated that in various embodiments of the present invention, the size of the serial number of above-mentioned each process is not meant to perform
The priority of sequence, the execution sequence of each process should be determined with its function and internal logic, without the reality of the reply embodiment of the present invention
It applies process and forms any restriction.
It will be appreciated by those skilled in the art that:Program can be passed through by realizing all or part of step of above method embodiment
Relevant hardware is instructed to complete, aforementioned program can be stored in computer read/write memory medium, which is performing
When, perform step including the steps of the foregoing method embodiments;And aforementioned storage medium includes:Movable storage device, read-only memory
The various media that can store program code such as (Read Only Memory, ROM), magnetic disc or CD.
When the above-mentioned integrated unit of the present invention is realized in the form of SFU software functional unit and be independent product sell or
In use, it can also be stored in a computer read/write memory medium.Based on such understanding, the skill of the embodiment of the present invention
Art scheme substantially in other words can be embodied the part that the prior art contributes in the form of software product, the calculating
Machine software product is stored in a storage medium, is used including some instructions so that a computer equipment (can be personal
Computer, server or network equipment etc.) perform all or part of each embodiment the method for the present invention.It is and aforementioned
Storage medium include:The various media that can store program code such as movable storage device, ROM, magnetic disc or CD.
The detailed description of the above, the only specific embodiment of the invention rather than limitation of the present invention.The relevant technologies
The technical staff in field is in the case of the principle and range for not departing from the present invention, various replacements, modification and the improvement made
It should all be included in the protection scope of the present invention.
Claims (10)
- A kind of 1. method for cable local discharge pattern-recognition, which is characterized in that the method includes:Obtain the corresponding p groups data of p sample signal of cable local discharge;For each group of data, take in the wherein m period Data, superposition is in one cycle to form the spectrum data of every group of data;To in the period as unit of equal angular 360 ° of progress phase window divisions, obtain n equidistant phase windows;Spectrum data in calculating cycle corresponds to each phase window Interior characteristic value obtains fisrt feature value matrix;Dimension-reduction treatment is carried out to the fisrt feature value matrix of acquisition, obtains second feature Value matrix;Classification and Identification is carried out to the characteristic value in second feature value matrix by pattern recognition classifier device, obtains each sample The corresponding classification results of signal.
- 2. according to the method described in claim 1, it is characterized in that, the method includes:By the 1st period to the 10th period Data investigation form spectrum data.
- 3. according to the method described in claim 1, it is characterized in that, the equal angular is 15 °.
- 4. according to the method described in claim 1, it is characterized in that, the method includes:For i-th of phase window, figure is calculated The corresponding peak charge q of modal datamax(i), mean charge qmn(i) and the quantity q of partial discharge pulsen(i) it is calculated as statistics Son;By corresponding 3 Statistical Operators of phase window each in a cycle, the characteristic value of 1 × 3n of spectrum data is obtained:[(qmax(1),qmn(1),qn(1))1 … (qmax(n),qmn(n),qn(n))1]1×3n。
- 5. according to the method described in claim 4, it is characterized in that, the fisrt feature value matrix is the characteristic value square of p × 3n Battle array:
- 6. according to the method described in claim 1, it is characterized in that, the method includes:Using Multidimensional Scaling, principal component One or more of analysis, factorial analysis and Independent Component Analysis carry out dimension-reduction treatment to obtain Second Eigenvalue square Battle array.
- 7. according to the method described in claim 6, it is characterized in that, the dimension-reduction treatment includes:Calculate fisrt feature value matrix Euclidean distance between middle characteristic value carries out singular value decomposition to obtain matrix Δ, to the dual centralization matrix of matrix Δ, obtains Second feature value matrix X after dimensionality reduction.
- 8. the method according to the description of claim 7 is characterized in that the dual centralization matrix Δ of the matrix Δ is expressed as:Wherein, p is characterized the quantity of value vector;Wherein, d is The dimension of fisrt feature value matrix, riqFor q-th of characteristic value in ith feature value vector, rjqFor in j-th of feature value vector Q-th of characteristic value;Dual centralization matrix Δ progress singular value decomposition is expressed as:Wherein, Λ be dual centralization matrix Δ eigenvalue cluster into diagonal matrix, U be Δ feature vector;To matrix Δ Characteristic value carry out descending sequence, the characteristic value feature vector corresponding with them of k maximum before selection;And then according to formulaObtain the second feature value matrix X after dimensionality reduction.
- 9. method according to any one of claim 1 to 8, which is characterized in that the method includes:Using BP nerve nets Network or support vector machines carry out forming types recognition classifier.
- 10. a kind of system for cable local discharge pattern-recognition, which is characterized in that the system comprises:Display, input Output equipment, at least one processor, the memory being connect at least one processor communication and the electricity for power supply Source device;Wherein, the display is used to show classification results;The input-output equipment is used for the corresponding number of input sample signal According to;The memory is stored with the instruction that can be performed by least one processor, and described instruction is by least one place It manages device to perform, so that at least one processor is able to carry out according to claim 1 to 9 any one of them method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810088179.9A CN108169643A (en) | 2018-01-30 | 2018-01-30 | A kind of method and system for cable local discharge pattern-recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810088179.9A CN108169643A (en) | 2018-01-30 | 2018-01-30 | A kind of method and system for cable local discharge pattern-recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108169643A true CN108169643A (en) | 2018-06-15 |
Family
ID=62512515
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810088179.9A Pending CN108169643A (en) | 2018-01-30 | 2018-01-30 | A kind of method and system for cable local discharge pattern-recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108169643A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109145762A (en) * | 2018-07-27 | 2019-01-04 | 西南石油大学 | A kind of cable accessory Recognition of Partial Discharge based on mathematical morphology and fractal theory |
CN109188244A (en) * | 2018-09-03 | 2019-01-11 | 长沙学院 | Based on the diagnostic method for failure of switch current circuit for improving FastICA |
CN109342909A (en) * | 2018-12-14 | 2019-02-15 | 中国测试技术研究院电子研究所 | A kind of cable accessory Partial Discharge Pattern Recognition Method based on SLLE |
CN110261746A (en) * | 2019-07-08 | 2019-09-20 | 清华大学深圳研究生院 | Electric cable stoppage detection method based on oscillating wave voltage periodic attenuation characteristic |
CN111398755A (en) * | 2020-04-21 | 2020-07-10 | 武汉朕泰智能科技有限公司 | Cable partial discharge waveform extraction method based on short-time FFT (fast Fourier transform) segmentation technology |
CN111766487A (en) * | 2020-07-31 | 2020-10-13 | 南京南瑞继保电气有限公司 | Cable partial discharge defect type identification method based on multiple quality characteristic quantities |
CN111796173A (en) * | 2020-08-13 | 2020-10-20 | 广东电网有限责任公司 | Partial discharge pattern recognition method, computer device, and storage medium |
CN112285494A (en) * | 2020-09-16 | 2021-01-29 | 北京博研中能科技有限公司 | Power cable partial discharge mode recognition analysis system |
CN112578241A (en) * | 2020-12-07 | 2021-03-30 | 国网天津市电力公司电力科学研究院 | High-noise tolerance characteristic extraction system and method for cable joint partial discharge classification |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103076547A (en) * | 2013-01-24 | 2013-05-01 | 安徽省电力公司亳州供电公司 | Method for identifying GIS (Gas Insulated Switchgear) local discharge fault type mode based on support vector machines |
CN103323749A (en) * | 2013-05-16 | 2013-09-25 | 上海交通大学 | Multi-classifier information fusion partial discharge diagnostic method |
CN103558529A (en) * | 2013-11-14 | 2014-02-05 | 国家电网公司 | Method for pattern recognition of three-phase drum-sharing type ultrahigh voltage GIS partial discharge |
CN106255203A (en) * | 2016-09-19 | 2016-12-21 | 哈尔滨工业大学 | The localization method of terminal RSRP disparity compensation based on MDS |
CN107167716A (en) * | 2017-07-11 | 2017-09-15 | 国网福建省电力有限公司泉州供电公司 | A kind of shelf depreciation default kind identification method and device |
-
2018
- 2018-01-30 CN CN201810088179.9A patent/CN108169643A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103076547A (en) * | 2013-01-24 | 2013-05-01 | 安徽省电力公司亳州供电公司 | Method for identifying GIS (Gas Insulated Switchgear) local discharge fault type mode based on support vector machines |
CN103323749A (en) * | 2013-05-16 | 2013-09-25 | 上海交通大学 | Multi-classifier information fusion partial discharge diagnostic method |
CN103558529A (en) * | 2013-11-14 | 2014-02-05 | 国家电网公司 | Method for pattern recognition of three-phase drum-sharing type ultrahigh voltage GIS partial discharge |
CN106255203A (en) * | 2016-09-19 | 2016-12-21 | 哈尔滨工业大学 | The localization method of terminal RSRP disparity compensation based on MDS |
CN107167716A (en) * | 2017-07-11 | 2017-09-15 | 国网福建省电力有限公司泉州供电公司 | A kind of shelf depreciation default kind identification method and device |
Non-Patent Citations (3)
Title |
---|
应高亮等: ""10kV交联聚乙烯电缆接头缺陷局部放电特性的研究"", 《浙江电力》 * |
牛海清等: ""奇异值分解在电缆局部放电信号模式识别中的应用"", 《华南理工大学学报( 自然科学版)》 * |
董超俊等: "《城市区域智能交通控制模型与算法》", 30 April 2015, 华南理工大学出版社 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109145762A (en) * | 2018-07-27 | 2019-01-04 | 西南石油大学 | A kind of cable accessory Recognition of Partial Discharge based on mathematical morphology and fractal theory |
CN109145762B (en) * | 2018-07-27 | 2022-02-01 | 西南石油大学 | Cable accessory partial discharge identification method based on mathematical morphology and fractal theory |
CN109188244B (en) * | 2018-09-03 | 2020-11-03 | 长沙学院 | Switching current circuit fault diagnosis method based on improved FastICA |
CN109188244A (en) * | 2018-09-03 | 2019-01-11 | 长沙学院 | Based on the diagnostic method for failure of switch current circuit for improving FastICA |
CN109342909A (en) * | 2018-12-14 | 2019-02-15 | 中国测试技术研究院电子研究所 | A kind of cable accessory Partial Discharge Pattern Recognition Method based on SLLE |
CN109342909B (en) * | 2018-12-14 | 2021-02-23 | 中国测试技术研究院电子研究所 | SLLE-based cable accessory partial discharge mode identification method |
CN110261746A (en) * | 2019-07-08 | 2019-09-20 | 清华大学深圳研究生院 | Electric cable stoppage detection method based on oscillating wave voltage periodic attenuation characteristic |
CN110261746B (en) * | 2019-07-08 | 2021-08-24 | 清华大学深圳研究生院 | Cable defect detection method based on periodic attenuation characteristics of oscillating wave voltage |
CN111398755A (en) * | 2020-04-21 | 2020-07-10 | 武汉朕泰智能科技有限公司 | Cable partial discharge waveform extraction method based on short-time FFT (fast Fourier transform) segmentation technology |
CN111766487A (en) * | 2020-07-31 | 2020-10-13 | 南京南瑞继保电气有限公司 | Cable partial discharge defect type identification method based on multiple quality characteristic quantities |
CN111796173A (en) * | 2020-08-13 | 2020-10-20 | 广东电网有限责任公司 | Partial discharge pattern recognition method, computer device, and storage medium |
CN111796173B (en) * | 2020-08-13 | 2022-01-21 | 广东电网有限责任公司 | Partial discharge pattern recognition method, computer device, and storage medium |
CN112285494A (en) * | 2020-09-16 | 2021-01-29 | 北京博研中能科技有限公司 | Power cable partial discharge mode recognition analysis system |
CN112578241A (en) * | 2020-12-07 | 2021-03-30 | 国网天津市电力公司电力科学研究院 | High-noise tolerance characteristic extraction system and method for cable joint partial discharge classification |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108169643A (en) | A kind of method and system for cable local discharge pattern-recognition | |
CN111428748B (en) | HOG feature and SVM-based infrared image insulator identification detection method | |
CN103076547B (en) | Method for identifying GIS (Gas Insulated Switchgear) local discharge fault type mode based on support vector machines | |
Peng et al. | Random forest based optimal feature selection for partial discharge pattern recognition in HV cables | |
Contin et al. | Digital detection and fuzzy classification of partial discharge signals | |
US20230300159A1 (en) | Network traffic anomaly detection method and apparatus, and electronic apparatus and storage medium | |
CN104951990B (en) | Power grid data processing method and equipment | |
CN109102508B (en) | Method for identifying insulation defects of alternating current cable based on partial discharge image characteristics | |
CN104155585A (en) | GIS partial discharge type identification method based on GK fuzzy clustering | |
CN109342909B (en) | SLLE-based cable accessory partial discharge mode identification method | |
Guo et al. | Fault diagnosis based on multiscale texture features of cable terminal on EMU of high-speed railway | |
US11061915B2 (en) | System and method for anomaly characterization based on joint historical and time-series analysis | |
CN107831409A (en) | The method and method for detecting abnormality of superfrequency partial discharge detection TuPu method parameter extraction | |
Chang et al. | Application of pulse sequence partial discharge based convolutional neural network in pattern recognition for underground cable joints | |
CN112367400A (en) | Intelligent inspection method and system for power internet of things with edge cloud coordination | |
Liao et al. | Fusion of infrared-visible images in UE-IoT for fault point detection based on GAN | |
CN112381800B (en) | Wire diameter abnormity identification method and device, electronic equipment and computer readable storage medium | |
Hassan et al. | Feature extraction of partial discharges during multiple simultaneous defects in low-voltage electric machines | |
CN111563556A (en) | Transformer substation cabinet equipment abnormity identification method and system based on color gradient weight | |
Chen et al. | Improved YOLOv5s model for key components detection of power transmission lines. | |
CN110189031A (en) | A kind of power distribution network diagnosis index classification method based on regression analysis on factors | |
Ye et al. | A corona recognition method based on visible light color and machine learning | |
CN113945862A (en) | Method, device and equipment for identifying high-resistance grounding fault of power distribution network | |
Chang et al. | Assessment of the insulation status aging in power cable joints using support vector machine | |
CN116204846A (en) | Method for rapidly positioning abnormal sensor data of power distribution network based on visible graph |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180615 |