CN106707118A - Method and device for identifying partial discharge pattern - Google Patents
Method and device for identifying partial discharge pattern Download PDFInfo
- Publication number
- CN106707118A CN106707118A CN201611141952.0A CN201611141952A CN106707118A CN 106707118 A CN106707118 A CN 106707118A CN 201611141952 A CN201611141952 A CN 201611141952A CN 106707118 A CN106707118 A CN 106707118A
- Authority
- CN
- China
- Prior art keywords
- model
- discharge signal
- local discharge
- pattern
- local
- 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
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
- G01R31/1227—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 of components, parts or materials
Abstract
The invention discloses a method and a device for identifying a partial discharge pattern. The method comprises the steps of establishing a typical partial discharge model, and obtaining the partial discharge signal of each model; according to the partial discharge signal of each model, constructing a high-dimensional random matrix for each model; according to the spectral distribution characteristics of the characteristic root of the high-dimensional random matrix for each model, constructing a knowledge base; based on the knowledge base, determining the pattern of a to-be-identified partial discharge signal based on the proximity algorithm. According to the technical scheme of the invention, the technical problems in the prior art that the identification rate of the partial discharge pattern is high and the anti-interference performance is poor can be solved.
Description
Technical field
The present invention relates to shelf depreciation field, in particular to a kind of Partial Discharge Pattern Recognition Method and device.
Background technology
The insulation defect of very little can change from small to big under charged electric power apparatus running status, may finally induce overall exhausted
Edge failure, causes large-area power-cuts, and loss is brought to national economy, and PD Pattern Recognition to finding that insulation is damaged in time
Bad degree and overhaul of the equipments are of great importance.
With the fast development of intelligent grid and energy internet, power system information data and Condition Monitoring Data are increasingly
Huge, the real-time storage of these data, transmission and treatment are the strong firm guarantees of bulk power grid, and Random Matrices Theory is used as one kind
Big data analysis method, its characteristic root spectrum analysis theory has good characteristic to analyzing big dimension data, in status of electric power
Preliminary Applications are obtained in assessment and abnormality detection.
In recent years, domestic and foreign scholars expand substantial amounts of research work, neutral net, small echo in partial discharge area of pattern recognition
Conversion, fractal theory and hidden Markov model have wide application in PD Pattern Recognition theory, achieve
Significant effect, but, recognition methods of the above based on PRPD collection of illustrative plates has some limitations, i.e. operating frequency phase
Meaning is not very big during shelf depreciation, and in DC transmission system, because the presence of DC equipment is believed without phase
Breath, and influenceed larger by magnitude of voltage using impulse waveform or peak value of pulse as the method for identification feature amount above, it is anti-interference
Property is not strong, is inapplicable for non-stationary discharge signal.
For above-mentioned in the prior art to the problem that pattern-recognition rate is high and anti-interference is poor of shelf depreciation, at present still
Effective solution is not proposed.
The content of the invention
A kind of Partial Discharge Pattern Recognition Method and device are the embodiment of the invention provides, at least to solve in the prior art
To the technical problem that pattern-recognition rate is high and anti-interference is poor of shelf depreciation.
A kind of one side according to embodiments of the present invention, there is provided Partial Discharge Pattern Recognition Method, including：Set up allusion quotation
Type partial discharge model, and obtain the local discharge signal of each model；Local discharge signal according to each model builds each model
Higher-dimension random matrix；The Spectral structure characteristic of the characteristic root of the higher-dimension random matrix according to each model builds knowledge base；Based on knowing
Know storehouse, the pattern of local discharge signal to be identified is determined using nearest neighbor algorithm.
Further, Spectral structure characteristic includes the average of the Spectral structure annulus of the characteristic root of the higher-dimension random matrix of each model
Spectral radius.
Further, knowledge based storehouse, the pattern of local discharge signal to be identified is determined using nearest neighbor algorithm, including：From
Using the K shelf depreciation that Euclidean distance algorithm picks are nearest with the average spectral radius of local discharge signal to be identified in knowledge base
Signal；Determine the pattern of most local discharge signals in K local discharge signal, obtain the mould of local discharge signal to be identified
Formula.
Further, the typical partial discharge model of foundation, and the local discharge signal of each model is obtained, including：Set up allusion quotation
Type partial discharge model, and the local discharge signal of each model is obtained using uhf sensor.
Further, typical partial discharge model includes：Plate plate discharging model, suspension electrode discharging model, bubble-discharge
Model, high voltage plane discharge model, corona discharge model and oil clearance discharging model.
Another aspect according to embodiments of the present invention, additionally provides a kind of PD Pattern Recognition device, including：Obtain
Module, for setting up typical partial discharge model, and obtains the local discharge signal of each model；First builds module, for root
The higher-dimension random matrix of each model is built according to the local discharge signal of each model；Second builds module, for according to each model
The Spectral structure characteristic of the characteristic root of higher-dimension random matrix builds knowledge base；Determining module, for knowledge based storehouse, using neighbouring calculation
Method determines the pattern of local discharge signal to be identified.
Further, Spectral structure characteristic includes the average of the Spectral structure annulus of the characteristic root of the higher-dimension random matrix of each model
Spectral radius.
Further, determining module, including：Choose module, for from knowledge base using Euclidean distance algorithm picks with
K nearest local discharge signal of the average spectral radius of local discharge signal to be identified；Determination sub-module, for determining K office
The pattern of most local discharge signals, obtains the pattern of local discharge signal to be identified in portion's discharge signal.
Further, the specific implementation of acquisition module includes：Typical partial discharge model is set up, and is sensed using hyperfrequency
Device obtains the local discharge signal of each model.
Further, typical partial discharge model includes：Plate plate discharging model, suspension electrode discharging model, bubble-discharge
Model, high voltage plane discharge model, corona discharge model and oil clearance discharging model.
In embodiments of the present invention, by setting up typical partial discharge model, and the local discharge signal of each model is obtained；
Local discharge signal according to each model builds the higher-dimension random matrix of each model；The spy of the higher-dimension random matrix according to each model
The Spectral structure characteristic for levying root builds knowledge base；Knowledge based storehouse, the mould of local discharge signal to be identified is determined using nearest neighbor algorithm
Formula, has reached the purpose of the pattern of determination local discharge signal to be identified, and recognition efficiency and accuracy rate are high, and the used time is short, energy
Enough offer sausage testing result defect recognition greatly helps, and then the pattern solved in the prior art to shelf depreciation is known
The technical problem that rate is not high and anti-interference is poor.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, this hair
Bright schematic description and description does not constitute inappropriate limitation of the present invention for explaining the present invention.In the accompanying drawings：
Fig. 1 is a kind of flow chart of according to embodiments of the present invention 1 Partial Discharge Pattern Recognition Method；
Fig. 2 is according to embodiments of the present invention 1 plate plate discharging model schematic diagram；
Fig. 3 is according to embodiments of the present invention 1 suspension electrode discharging model schematic diagram；
Fig. 4 is according to embodiments of the present invention 1 bubble-discharge model schematic；
Fig. 5 is according to embodiments of the present invention 1 high voltage plane discharge model schematic；
Fig. 6 is according to embodiments of the present invention 1 corona discharge model schematic；
Fig. 7 is according to embodiments of the present invention 1 oil clearance discharging model schematic diagram；
Fig. 8 is the higher-dimension random matrix of according to embodiments of the present invention 1 local discharge signal for plate plate discharging model
Schematic diagram；
Fig. 9 is that the Spectral structure of the characteristic root of the higher-dimension random matrix of according to embodiments of the present invention 1 plate plate discharging model shows
It is intended to；
Figure 10 is the spectrum point of the characteristic root of the higher-dimension random matrix of according to embodiments of the present invention 1 suspension electrode discharging model
Cloth schematic diagram；
Figure 11 is that the Spectral structure of the characteristic root of the higher-dimension random matrix of according to embodiments of the present invention 1 bubble-discharge model shows
It is intended to；
Figure 12 is the spectrum point of the characteristic root of the higher-dimension random matrix of according to embodiments of the present invention 1 high voltage plane discharge model
Cloth schematic diagram；
Figure 13 is that the Spectral structure of the characteristic root of the higher-dimension random matrix of according to embodiments of the present invention 1 corona discharge model shows
It is intended to；
Figure 14 is that the Spectral structure of the characteristic root of the higher-dimension random matrix of according to embodiments of the present invention 1 oil clearance discharging model shows
It is intended to；And
Figure 15 is a kind of structure chart of according to embodiments of the present invention 2 PD Pattern Recognition device.
Specific embodiment
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, is clearly and completely described to the technical scheme in the embodiment of the present invention, it is clear that described embodiment is only
The embodiment of a part of the invention, rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained under the premise of creative work is not made, should all belong to the model of present invention protection
Enclose.
It should be noted that term " first ", " in description and claims of this specification and above-mentioned accompanying drawing
Two " it is etc. for distinguishing similar object, without for describing specific order or precedence.It should be appreciated that so using
Data can exchange in the appropriate case, so as to embodiments of the invention described herein can with except illustrating herein or
Order beyond those of description is implemented.Additionally, term " comprising " and " having " and their any deformation, it is intended that cover
Lid is non-exclusive to be included, for example, the process, method, system, product or the equipment that contain series of steps or unit are not necessarily limited to
Those steps or unit clearly listed, but may include not list clearly or for these processes, method, product
Or other intrinsic steps of equipment or unit.
Embodiment 1
According to embodiments of the present invention, there is provided a kind of embodiment of the method for Partial Discharge Pattern Recognition Method is, it is necessary to illustrate
, can be held in the such as one group computer system of computer executable instructions the step of the flow of accompanying drawing is illustrated
OK, and, although show logical order in flow charts, but in some cases, can be with different from order herein
Perform shown or described step.
Fig. 1 is Partial Discharge Pattern Recognition Method according to embodiments of the present invention, as shown in figure 1, the method is including as follows
Step：
Step S102, sets up typical partial discharge model, and obtain the local discharge signal of each model.
Step S104, the local discharge signal according to each model builds the higher-dimension random matrix of each model.
Step S106, the Spectral structure characteristic of the characteristic root of the higher-dimension random matrix according to each model builds knowledge base.
Step S108, knowledge based storehouse determines the pattern of local discharge signal to be identified using nearest neighbor algorithm.
In embodiments of the present invention, by setting up typical partial discharge model, and the local discharge signal of each model is obtained；
Local discharge signal according to each model builds the higher-dimension random matrix of each model；The spy of the higher-dimension random matrix according to each model
The Spectral structure characteristic for levying root builds knowledge base；Knowledge based storehouse, the mould of local discharge signal to be identified is determined using nearest neighbor algorithm
Formula, has reached the purpose of the pattern of determination local discharge signal to be identified, and recognition efficiency and accuracy rate are high, and the used time is short, energy
Enough offer sausage testing result defect recognition greatly helps, and then the pattern solved in the prior art to shelf depreciation is known
The technical problem that rate is not high and anti-interference is poor.
In a kind of optional embodiment, typical partial discharge model includes：Plate plate discharging model, suspension electrode electric discharge mould
Type, bubble-discharge model, high voltage plane discharge model, corona discharge model and oil clearance discharging model.
Specifically, plate plate discharging model, suspension electrode discharging model, bubble-discharge model, high voltage plane discharge model, electricity
The schematic diagram difference of corona model and oil clearance discharging model is as illustrated in figs. 2-7.
In a kind of optional embodiment, the local discharge signal in step S104 according to each model builds the height of each model
Tieing up the specific implementation of random matrix can be：
Higher-dimension random matrix is built for local discharge signal, it is possible to use the data vector of unit interval Δ tCarry out generator matrix X, its expression formula is shown below, Fig. 8 is directed to the shelf depreciation letter of plate plate discharging model
Number higher-dimension random matrix schematic diagram；
Wherein, X is N × T random matrixes in above formula, can adjust matrix X by changing sample frequency and sampling duration
Ranks number ratio, with ensure matrix X meet Random Matrices Theory analysis requirement.To the product of matrixWherein
Xi is the independent identically distributed non-random square formations of hermitian of N × nSingular value equivalent matrice, then the experience spectrum density of Z will restrain
In：
Work as N, n → ∞ and c=N/n≤1, in the case where complex plane is represented, characteristic root will be (1-c) in inner radiiα/2It is cylindrical
Radius is in 1 annulus.Wherein,It is a Harr matrix.
In a kind of optional embodiment, Spectral structure characteristic includes the spectrum point of the characteristic root of the higher-dimension random matrix of each model
The average spectral radius of cloth annulus.
Specifically, the time series produced by local discharge signal, by being configured to the higher dimensional matrix for meeting certain requirements
Afterwards, its characteristic root is proved in theory to be annularly distributed, now choose the average spectral radius of the Spectral structure annulus of six class discharge waveforms
(MSR, Mean Spectral radius) is the characteristic quantity of electric discharge type identification：
Wherein, | λZ| represent λZIn the radius of complex plane.
In a kind of optional embodiment, step S108, including：
Step S202, uses the averaging spectrum half of Euclidean distance algorithm picks and local discharge signal to be identified from knowledge base
K nearest local discharge signal of footpath.
Step S204, determines the pattern of most local discharge signals in K local discharge signal, obtains part to be identified and puts
The pattern of electric signal.
Specifically, nearest neighbor algorithm i.e. k nearest neighbor algorithm, are a kind of simple machine learning algorithms, that is, give an instruction
Practice data set, to new input example, concentrated in training data and find the K example closest with the example, this K example
Majority belongs to certain class, just the input Exemplary classes to this class.The selection of K values, distance metric and categorised decision rule are
Three fundamentals of the algorithm.Distance metric selects Euclidean distance in this experiment, i.e., for 10 of any two classes signal
The annulus Spectral structure of waveform, the Euclidean distance of its MSR is d (κ1iMSR, κ2iMSR)：
In a kind of optional embodiment, for the MSR of new input waveform, can be in 60 waveforms (per class shelf depreciation
Each ten waveforms of signal) Spectral structure knowledge base in choose the K MSR nearest with it, then see which this K MSR majority belong to
The new waveform being input into, then be classified as such local discharge signal, test result indicate that when K takes 3, recognition effect is best by one class.
Above-described embodiment present invention uses KNN algorithms, recognizes the random matrix Spectral structure of different discharge signals, and then realize
Identification to different type local discharge signal.
In a kind of optional embodiment, step S102, including：Typical partial discharge model is set up, and uses hyperfrequency
Sensor obtains the local discharge signal of each model.
In a kind of optional embodiment, it is possible to use discharging model produces six kinds of local discharge signals, and gathers corresponding
Each 10 groups of ultrahigh-frequency signal, this experiment carries out analysis of spectrum using the multiple pulse waveform of discharge signal, using the order of oscillograph
Drainage pattern can obtain the multiple pulse waveform of discharge signal, and so-called multiple pulse waveform is the multiple pulses of collection and only to electric discharge
Nearby some points are acquired for pulse, and oscillograph is set to (500mv/div, 200ns, 1G/S, acquisition order pattern, 300 sections
Pulse), under setting herein, each pulse collection 200*10-9s*1G/s=2000 point, a complete multiple pulse waveform
300*2000=600000 point is acquired altogether, and the 600*1000 that every kind of partial-discharge ultrahigh-frequency signal is constructed respectively ties up random square
Battle array Z.Fig. 9-14 give plate plate discharging model, suspension electrode discharging model, bubble-discharge model, high voltage plane discharge model,
The characteristic root Spectral structure situation of the higher dimensional matrix Z that corona discharge model and oil clearance discharging model are generated, as can be seen from Figure its
The characteristics of characteristic root is clearly presented annulus and is distributed, and the internal diameter of each annulus is of different sizes, the Spectral structure density in ring also has
Itself the characteristics of.Every class discharge waveform 50 can be respectively gathered, constructing corresponding higher-dimension random matrix carries out analysis of spectrum, and will
K- neighbour's recognizers that the characteristic quantity input of extraction is trained are identified, and recognition result is：For the electric discharge of plate plate, identification 35
Individual, discrimination is 70%, for suspension electrode electric discharge, recognizes 43, and discrimination is 86%, for bubble-discharge, recognizes 44,
Discrimination is 88%, for high voltage plane discharge, recognizes 43, and discrimination is 86%, for corona discharge, recognizes 44, is known
Rate is not 88%, for oil clearance electric discharge, recognizes 38, and discrimination is 76%, from the foregoing, it will be observed that the present invention is to six kinds of common defects
Experimental verification is carried out, average recognition rate provides very big help more than 85% to Site Detection result defect recognition, shortens
Analysis time, improve analysis efficiency and accuracy rate.
Embodiment 2
According to embodiments of the present invention, there is provided a kind of product embodiments of PD Pattern Recognition device, Figure 15 is root
According to the PD Pattern Recognition device of the embodiment of the present invention, as shown in figure 15, the device includes that acquisition module, first build mould
Block, second build module and determining module.
Wherein, acquisition module, for setting up typical partial discharge model, and obtains the local discharge signal of each model；The
One builds module, the higher-dimension random matrix for building each model according to the local discharge signal of each model；Second builds module,
Spectral structure characteristic for the characteristic root of the higher-dimension random matrix according to each model builds knowledge base；Determining module, for being based on
Knowledge base, the pattern of local discharge signal to be identified is determined using nearest neighbor algorithm.
In embodiments of the present invention, typical partial discharge model is set up by acquisition module, and obtains the part of each model
Discharge signal；First builds the higher-dimension random matrix that module builds each model according to the local discharge signal of each model；Second structure
Modeling root tuber builds knowledge base according to the Spectral structure characteristic of the characteristic root of the higher-dimension random matrix of each model；Determining module knowledge based
Storehouse, the pattern of local discharge signal to be identified is determined using nearest neighbor algorithm, has reached the mould of determination local discharge signal to be identified
The purpose of formula, and recognition efficiency and accuracy rate are high, and the used time is short, sausage testing result defect recognition can be provided and greatly helped
Help, and then solve the technical problem that pattern-recognition rate is high and anti-interference is poor in the prior art to shelf depreciation.
In a kind of optional embodiment, Spectral structure characteristic includes the spectrum point of the characteristic root of the higher-dimension random matrix of each model
The average spectral radius of cloth annulus.
In a kind of optional embodiment, determining module, including choose module and determination sub-module.
Wherein, module is chosen, for using Euclidean distance algorithm picks and local discharge signal to be identified from knowledge base
The nearest K local discharge signal of average spectral radius；Determination sub-module, for determining most offices in K local discharge signal
The pattern of portion's discharge signal, obtains the pattern of local discharge signal to be identified.
In a kind of optional embodiment, the specific implementation of acquisition module includes：Typical partial discharge model is set up, and is made
The local discharge signal of each model is obtained with uhf sensor.
In a kind of optional embodiment, typical partial discharge model includes：Plate plate discharging model, suspension electrode electric discharge mould
Type, bubble-discharge model, high voltage plane discharge model, corona discharge model and oil clearance discharging model.
The embodiments of the present invention are for illustration only, and the quality of embodiment is not represented.
In the above embodiment of the present invention, the description to each embodiment all emphasizes particularly on different fields, and does not have in certain embodiment
The part of detailed description, may refer to the associated description of other embodiment.
In several embodiments provided herein, it should be understood that disclosed technology contents, can be by other
Mode is realized.Wherein, device embodiment described above is only schematical, such as division of described unit, Ke Yiwei
A kind of division of logic function, can there is other dividing mode when actually realizing, such as multiple units or component can combine or
Person is desirably integrated into another system, or some features can be ignored, or does not perform.Another, shown or discussed is mutual
Between coupling or direct-coupling or communication connection can be the INDIRECT COUPLING or communication link of unit or module by some interfaces
Connect, can be electrical or other forms.
The unit that is illustrated as separating component can be or may not be it is physically separate, it is aobvious as unit
The part for showing can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On unit.Some or all of unit therein can be according to the actual needs selected to realize the purpose of this embodiment scheme.
In addition, during each functional unit in each embodiment of the invention can be integrated in a processing unit, it is also possible to
It is that unit is individually physically present, it is also possible to which two or more units are integrated in a unit.Above-mentioned integrated list
Unit can both be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
If the integrated unit is to realize in the form of SFU software functional unit and as independent production marketing or use
When, can store in a computer read/write memory medium.Based on such understanding, technical scheme is substantially
The part for being contributed to prior art in other words or all or part of the technical scheme can be in the form of software products
Embody, the computer software product is stored in a storage medium, including some instructions are used to so that a computer
Equipment (can be personal computer, server or network equipment etc.) perform each embodiment methods described of the invention whole or
Part steps.And foregoing storage medium includes：USB flash disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), mobile hard disk, magnetic disc or CD etc. are various can be with store program codes
Medium.
The above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (10)
1. a kind of Partial Discharge Pattern Recognition Method, it is characterised in that including：
Typical partial discharge model is set up, and obtains the local discharge signal of each model；
Local discharge signal according to each model builds the higher-dimension random matrix of each model；
The Spectral structure characteristic of the characteristic root of the higher-dimension random matrix according to each model builds knowledge base；
Based on the knowledge base, the pattern of local discharge signal to be identified is determined using nearest neighbor algorithm.
2. method according to claim 1, it is characterised in that the Spectral structure characteristic include the higher-dimension of each model with
The average spectral radius of the Spectral structure annulus of the characteristic root of machine matrix.
3. method according to claim 2, it is characterised in that based on the knowledge base, determine to wait to know using nearest neighbor algorithm
The pattern of other local discharge signal, including：
From the knowledge base using Euclidean distance algorithm picks and the local discharge signal to be identified average spectral radius most
K near local discharge signal；
Determine the pattern of most local discharge signals in the K local discharge signal, obtain the shelf depreciation letter to be identified
Number pattern.
4. method according to claim 1, it is characterised in that set up typical partial discharge model, and obtain each model
Local discharge signal, including：
Typical partial discharge model is set up, and the local discharge signal of each model is obtained using uhf sensor.
5. the method according to claim any one of 1-4, it is characterised in that the typical partial discharge model includes：Plate
Plate discharging model, suspension electrode discharging model, bubble-discharge model, high voltage plane discharge model, corona discharge model and oil clearance
Discharging model.
6. a kind of PD Pattern Recognition device, it is characterised in that including：
Acquisition module, for setting up typical partial discharge model, and obtains the local discharge signal of each model；
First builds module, the higher-dimension random matrix for building each model according to the local discharge signal of each model；
Second builds module, and the Spectral structure characteristic for the characteristic root of the higher-dimension random matrix according to each model builds knowledge
Storehouse；
Determining module, for based on the knowledge base, the pattern of local discharge signal to be identified being determined using nearest neighbor algorithm.
7. device according to claim 6, it is characterised in that the Spectral structure characteristic include the higher-dimension of each model with
The average spectral radius of the Spectral structure annulus of the characteristic root of machine matrix.
8. device according to claim 7, it is characterised in that the determining module, including：
Module is chosen, for using Euclidean distance algorithm picks and the local discharge signal to be identified from the knowledge base
K nearest local discharge signal of average spectral radius；
Determination sub-module, the pattern for determining most local discharge signals in the K local discharge signal, obtains described treating
Recognize the pattern of local discharge signal.
9. device according to claim 6, it is characterised in that the specific implementation of the acquisition module includes：
Typical partial discharge model is set up, and the local discharge signal of each model is obtained using uhf sensor.
10. the device according to claim any one of 6-9, it is characterised in that the typical partial discharge model includes：Plate
Plate discharging model, suspension electrode discharging model, bubble-discharge model, high voltage plane discharge model, corona discharge model and oil clearance
Discharging model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611141952.0A CN106707118B (en) | 2016-12-12 | 2016-12-12 | Partial Discharge Pattern Recognition Method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611141952.0A CN106707118B (en) | 2016-12-12 | 2016-12-12 | Partial Discharge Pattern Recognition Method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106707118A true CN106707118A (en) | 2017-05-24 |
| CN106707118B CN106707118B (en) | 2019-09-10 |
Family
ID=58937092
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611141952.0A Active CN106707118B (en) | 2016-12-12 | 2016-12-12 | Partial Discharge Pattern Recognition Method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106707118B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108008262A (en) * | 2017-11-29 | 2018-05-08 | 国网上海市电力公司 | A kind of shelf depreciation quick determination method |
| CN108196168A (en) * | 2017-10-27 | 2018-06-22 | 全球能源互联网研究院有限公司 | A kind of partial discharge pulse detecting method, system, terminal and readable storage medium storing program for executing |
| CN108734602A (en) * | 2018-04-09 | 2018-11-02 | 广东电网有限责任公司 | A kind of user's electricity stealing identification system based on higher-dimension random matrix |
| CN109829416A (en) * | 2019-01-28 | 2019-05-31 | 国网山东省电力公司日照供电公司 | A kind of method and system detecting ultrahigh frequency partial discharge signal from random noise |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129017A (en) * | 2010-12-30 | 2011-07-20 | 杭州柯林电力设备有限公司 | Case-based intelligent local discharge fault identification system and identification method |
| CN104316846A (en) * | 2014-08-28 | 2015-01-28 | 国家电网公司 | Identification method, apparatus, and system of partial discharger mode of electrical equipment |
| CN105606966A (en) * | 2015-12-21 | 2016-05-25 | 安徽理工大学 | Partial discharge pattern recognition method based on mixed neural network algorithm |
-
2016
- 2016-12-12 CN CN201611141952.0A patent/CN106707118B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102129017A (en) * | 2010-12-30 | 2011-07-20 | 杭州柯林电力设备有限公司 | Case-based intelligent local discharge fault identification system and identification method |
| CN104316846A (en) * | 2014-08-28 | 2015-01-28 | 国家电网公司 | Identification method, apparatus, and system of partial discharger mode of electrical equipment |
| CN105606966A (en) * | 2015-12-21 | 2016-05-25 | 安徽理工大学 | Partial discharge pattern recognition method based on mixed neural network algorithm |
Non-Patent Citations (2)
| Title |
|---|
| 严英杰等: "基于高维随机矩阵大数据分析模型的输变电设备关键性能评估方法", 《中国电机工程学报》 * |
| 魏大千等: "高维随机矩阵描述下的量测大数据建模与异常数据检测方法", 《中国电机工程学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108196168A (en) * | 2017-10-27 | 2018-06-22 | 全球能源互联网研究院有限公司 | A kind of partial discharge pulse detecting method, system, terminal and readable storage medium storing program for executing |
| CN108008262A (en) * | 2017-11-29 | 2018-05-08 | 国网上海市电力公司 | A kind of shelf depreciation quick determination method |
| CN108008262B (en) * | 2017-11-29 | 2020-07-07 | 国网上海市电力公司 | Rapid partial discharge detection method |
| CN108734602A (en) * | 2018-04-09 | 2018-11-02 | 广东电网有限责任公司 | A kind of user's electricity stealing identification system based on higher-dimension random matrix |
| CN108734602B (en) * | 2018-04-09 | 2021-08-03 | 广东电网有限责任公司 | User electricity stealing behavior identification system based on high-dimensional random matrix |
| CN109829416A (en) * | 2019-01-28 | 2019-05-31 | 国网山东省电力公司日照供电公司 | A kind of method and system detecting ultrahigh frequency partial discharge signal from random noise |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106707118B (en) | 2019-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104849633B (en) | A kind of partial discharge of switchgear mode identification method | |
| CN106707118A (en) | Method and device for identifying partial discharge pattern | |
| CN108805747A (en) | A kind of abnormal electricity consumption user detection method based on semi-supervised learning | |
| CN103675616B (en) | The online partial discharge detection signal recognition method of cable | |
| CN110108992B (en) | Cable partial discharge fault identification method and system based on improved random forest algorithm | |
| CN108573225A (en) | A kind of local discharge signal mode identification method and system | |
| CN106778923A (en) | A kind of Power Quality Disturbance sorting technique and device | |
| Wang et al. | Classification of power quality events using optimal time-frequency representations-part 1: Theory | |
| CN110045209A (en) | Detection method, device, equipment and the readable storage medium storing program for executing of electricity consumption data exception | |
| CN110609477B (en) | Electric power system transient stability discrimination system and method based on deep learning | |
| CN106526434A (en) | Partial discharge mode identifying method and device | |
| CN106443380B (en) | A kind of distribution cable local discharge signal recognition methods and device | |
| CN110261746B (en) | Cable defect detection method based on periodic attenuation characteristics of oscillating wave voltage | |
| CN111103477B (en) | Electrical load monitoring method and system based on 3D convolutional neural network | |
| CN109444667A (en) | Power distribution network initial failure classification method and device based on convolutional neural networks | |
| CN108684043A (en) | The abnormal user detection method of deep neural network based on minimum risk | |
| CN114169249A (en) | Artificial intelligence identification method for high-resistance grounding fault of power distribution network | |
| CN110703006A (en) | Three-phase power quality disturbance detection method based on convolutional neural network | |
| CN113627289A (en) | Deep learning-based power equipment identification method and device | |
| CN109325410A (en) | A kind of EEG signals EEG feature extracting method based on convolutional neural networks | |
| CN113076354A (en) | User electricity consumption data analysis method and device based on non-invasive load monitoring | |
| CN106778692A (en) | A kind of cable local discharge signal recognition method and device based on S-transformation | |
| CN115687902A (en) | New load small sample identification method based on MAML | |
| CN116070384A (en) | Transient stability evaluation method and system based on power grid feature arrangement importance | |
| Mahdi et al. | Defensive islanding using self-organizing maps neural networks and hierarchical clustering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |