CN110412431A - A kind of diagnostic method and diagnostic system of the shelf depreciation defect type of power equipment - Google Patents
A kind of diagnostic method and diagnostic system of the shelf depreciation defect type of power equipment Download PDFInfo
- Publication number
- CN110412431A CN110412431A CN201910717073.5A CN201910717073A CN110412431A CN 110412431 A CN110412431 A CN 110412431A CN 201910717073 A CN201910717073 A CN 201910717073A CN 110412431 A CN110412431 A CN 110412431A
- Authority
- CN
- China
- Prior art keywords
- shelf depreciation
- map
- power equipment
- diagnostic
- discharge
- 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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Relating To Insulation (AREA)
Abstract
The invention discloses a kind of power equipment shelf depreciation defect type diagnostic method method and diagnostic system, which includes: that the corresponding Partial Discharge Data of all kinds of shelf depreciation defects of power equipment is extracted from historical data；Partial Discharge Data is handled to obtain the corresponding shelf depreciation map of all kinds of shelf depreciation defects of power equipment；Shelf depreciation diagnostic model is obtained using the shelf depreciation map and shelf depreciation defect type label training depth convolutional neural networks of power equipment；Wherein, the input data of shelf depreciation diagnostic model is the shelf depreciation map of power equipment, and output data is the shelf depreciation diagnostic result of power equipment, and diagnostic result includes whether that there are shelf depreciation and shelf depreciation defect types；It acquires to be input in shelf depreciation diagnostic model to the shelf depreciation map of diagnosing electric power and obtains shelf depreciation diagnostic result.The present invention realizes the automatic diagnosis of electric discharge type by this method, and the reliability of diagnostic result is improved using neural network.
Description
Technical field
The invention belongs to local discharge of electrical equipment detection technique fields, and in particular to a kind of shelf depreciation of power equipment
The diagnostic method and diagnostic system of defect type.
Background technique
With the continuous development that Diagnostic Examination And Repair of Electric Power Facilities works, shelf depreciation detects important as status of electric power
Means are more and more applied in power equipment daily maintenance.The work of major part Partial Discharge Detection still relies on O&M at present
Personnel are completed by live detection instrument, carry out live detection by personnel merely and there is following ask in the actual operation process
Topic not can effectively solve:
1, part local discharge detection device has no data diagnosis function, and detection process needs testing staff voluntarily to data
It is analyzed and determined, and the diagnosis of shelf depreciation generally requires just accomplish accurate judgement by permanent experience accumulation, part
Live detection work of discharging is higher to testing staff's professional skill requirement, is unfavorable for the work of shelf depreciation live detection in power grid base
The development of layer.
2, the diagnostic function carried in the local discharge detection device of part is unable to satisfy the needs of live practical application.It is practical
Although in can more accurately the defects of judgment experiment room type, apply at the scene, since shelf depreciation lacks
The complicated multiplicity of the form of expression of type is fallen into, while there is also a large amount of interference, the included diagnostic function of detection device is difficult accurately to know
Other defect type, can generate erroneous judgement in use, and the judgement and subsequent Strategies of Maintenance to live detection personnel cause to miss
It leads.
Summary of the invention
The object of the present invention is to provide the diagnostic method and diagnostic system of a kind of shelf depreciation defect type of power equipment,
It collects shelf depreciation map by the data that localized sensor carries out shelf depreciation to power equipment, and passes through depth
Habit trained depth convolutional neural networks much of that of playing obtain shelf depreciation defect type diagnostic result, and then realize power equipment
The automatic diagnosis and identification of upper shelf depreciation defect type, the reliability of diagnostic result is improved by training neural network.
On the one hand, the present invention provides a kind of diagnostic method of the shelf depreciation defect type of power equipment, including walks as follows
It is rapid:
Step S1: all kinds of shelf depreciation defects pair of power equipment are extracted from history on-the-spot test and/or disintegration process data
The shelf depreciation map answered；
The shelf depreciation map is for indicating local discharge characteristic, the local discharge characteristic and shelf depreciation defect class
Type is corresponding；
Step S2: shelf depreciation map and shelf depreciation defect type label training using power equipment in step S1
Depth convolutional neural networks obtain shelf depreciation diagnostic model；
Wherein, the input data of shelf depreciation diagnostic model is the shelf depreciation map of power equipment, and output data is electricity
The shelf depreciation diagnostic result of power equipment, the diagnostic result include whether that there are shelf depreciation and shelf depreciation defect classes
Type；
Step S3: the shelf depreciation map of the acquisition to diagnosing electric power；
Step S4: the shelf depreciation map of step S3 is input in the shelf depreciation diagnostic model of step S2 and is obtained
Shelf depreciation diagnostic result.
Further preferably, according to preset ratio by the shelf depreciation map of the power equipment acquired in step 2 and part
Discharge defect type label is divided to obtain training set and verifying collection, and training set concentrates each sample standard deviation to correspond to electric power with verifying
The one group of shelf depreciation map and shelf depreciation defect type label of equipment, wherein utilize the training depth convolution of training set
Neural network obtains the shelf depreciation diagnostic model in step S2.
Further preferably, the shelf depreciation defect type includes electric discharge between metal interface and metal interface, metal
Electric discharge, insulator interface between point discharge, metal interface and insulator interface and the electric discharge between insulator interface, solid insulation with
The creeping discharge of gas-insulated intersection, the creeping discharge of solid insulation and fluid insulation intersection, putting inside solid insulation
Electricity, electric discharge, metal particle electric discharge inside fluid insulation.
Further preferably, shelf depreciation map described in one group of power equipment include at least superfrequency PRPD&PRPS map,
High frequency PRPD&PRPS map, ultrasonic wave PRPD&PRPS map, ultrasonic pulse map.
Further preferably, the local discharge characteristic includes at least shelf depreciation amplitude-phase property, partial discharge pulse
Quantity-phase property, shelf depreciation amplitude versus time feature, partial discharge pulse's quantity-temporal characteristics, shelf depreciation amplitude-arteries and veins
The flight time of punching.
Further preferably, the shelf depreciation map in step S1 is raw using the Partial Discharge Data of partial discharge sensor acquisition
At, the partial discharge sensor includes extra-high video sensor, contact ultrasonic sensor, non-contact ultrasonic sensor, height
Video sensor.
On the other hand, the present invention also provides a kind of diagnostic system of above method, including sequentially connected partial discharge sensor,
Live testing apparatus for local discharge and cloud platform；
Wherein, partial discharge sensor is used to acquire the corresponding Partial Discharge Data of all kinds of shelf depreciations of power equipment；
The live testing apparatus for local discharge includes data acquisition module, aggregation of data processing module, data communication mould
Block, wherein data acquisition module obtains Partial Discharge Data from partial discharge sensor, and the aggregation of data processing module is used for part
Discharge data is handled to obtain the corresponding shelf depreciation map of all kinds of shelf depreciation defect types of power equipment；
The cloud platform, for constructing shelf depreciation diagnostic model and using the shelf depreciation diagnostic model to electric power
The diagnosis of equipment real-time perfoming shelf depreciation defect type, and send diagnostic result to the live testing apparatus for local discharge,
The live testing apparatus for local discharge receives diagnostic result by data communication module.
Further preferably, the partial discharge sensor and the wired or wireless connection of live testing apparatus for local discharge, the office
Portion's electric discharge measuring device with electricity is connect with cloud platform by 4G communication network.
Beneficial effect
1, the present invention obtains shelf depreciation diagnostic model using depth convolutional neural networks training pattern, and which raises electric power
The reliability of apparatus local discharge diagnostic result realizes the shelf depreciation defect type automaticdiagnosis of power equipment, can be with
The technical requirements to live live detection personnel are effectively reduced, make testing staff that local put can be completed without accumulating correlation experience
Electro-detection and diagnostic work work convenient for shelf depreciation live detection in the popularization of base.
2, the real-time diagnosis that the present invention is realized using shelf depreciation diagnostic model can greatly improve shelf depreciation band electric-examination
Efficiency is surveyed, the real-time of field data diagnosis is effectively improved.
3, superfrequency PRPD&PRPS map, high frequency PRPD&PRPS map, ultrasonic wave PRPD& is utilized in the present invention simultaneously
This few class data of PRPS map, ultrasonic pulse map, especially from all kinds of PRPD&PRPS maps be simultaneously by PRPD figure and
PRPS figure carries out joint displaying in a map, can be simultaneously to shelf depreciation amplitude-phase property, partial discharge pulse's number
Amount-phase property, shelf depreciation amplitude versus time feature, partial discharge pulse's quantity-temporal characteristics carry out more intuitive exhibition
Show, more local electric discharge type can be analyzed and determined.The present invention utilizes this few class map simultaneously, selects convolution
Neural network is incorporated into one, and then improves the reliability of diagnostic result.
Detailed description of the invention
Fig. 1 is a kind of configuration diagram of power equipment shelf depreciation defect type diagnostic system provided by the invention；
Fig. 2 is the diagnostic process schematic diagram provided by the invention based on diagnostic system；
Fig. 3 is a kind of flow diagram of power equipment shelf depreciation defect type diagnostic method provided by the invention.
Specific embodiment
Below in conjunction with embodiment, the present invention is described further.
As shown in Figure 1, a kind of power equipment shelf depreciation defect type diagnostic system provided by the invention includes successively connecting
Partial discharge sensor, live testing apparatus for local discharge and the cloud platform connect.Wherein, partial discharge sensor and shelf depreciation band electric-examination
It surveys device wirelessly or non-wirelessly to connect, live testing apparatus for local discharge is connect with cloud platform by 4G network communication.
In the present embodiment, partial discharge sensor includes extra-high video sensor, contact ultrasonic sensor, non-contact ultrasonic
Sensor, high frequency sensors.The present invention acquires Partial Discharge Data using partial discharge sensor, specifically, extra-high video sensor is adopted
Collect the data of superfrequency PRPD/PRPS map, contact ultrasonic, non-contact type ultrasonic sensor acquire ultrasonic wave PRPD/
Data, the data of ultrasonic pulse map of PRPS map, high frequency sensors acquire the data of PRPD/PRPS map, the present invention
All kinds of Partial Discharge Datas are acquired by setting multiclass sensor, and then generates part in live testing apparatus for local discharge and puts
Electrograph spectrum, in the present embodiment, one group of shelf depreciation map include superfrequency PRPD&PRPS map, high frequency PRPD&PRPS map,
Ultrasonic wave PRPD&PRPS map, ultrasonic pulse map.Wherein, PRPD&PRPS map refers to while scheming PRPD figure and PRPS
Joint displaying is carried out in a map, it can be simultaneously to shelf depreciation amplitude-phase property, partial discharge pulse's quantity-phase
Feature, shelf depreciation amplitude versus time feature, partial discharge pulse's quantity-temporal characteristics carry out more intuitive displaying.Pulse diagram
Spectrum refers to shelf depreciation amplitude-pulse flight time.When being detected extremely in the present invention to shelf depreciation, acquire simultaneously
Superfrequency PRPD&PRPS map, high frequency PRPD&PRPS map, ultrasonic wave PRPD&PRPS map, ultrasonic pulse map, by four
It opens map and is combined into a map by preset order, go to identify shelf depreciation defect type using the combination map.
Superfrequency involved in the present invention, high frequency PRPD&PRPS map being capable of effecting reaction metal interface and metal interfaces
Between electric discharge, metal tip electric discharge, the electric discharge between metal interface and insulator interface, between insulator interface and insulator interface
Electric discharge, the creeping discharge of solid insulation and fluid insulation intersection, the electric discharge inside solid insulation, the electric discharge inside fluid insulation
Situation；Ultrasonic wave PRPD&PRPS map being capable of electric discharge, metal tip electric discharge, metal interface between effecting reaction metal interface
Electric discharge, insulator interface between insulator interface and the electric discharge between insulator interface, solid insulation and gas-insulated intersection
The creeping discharge situation of creeping discharge, solid insulation and fluid insulation intersection.Ultrasonic pulse map can effectively reflect gold
Belong to particulate electric discharge situation.By the integrated application of above-mentioned four seed types map, it is different that single means detection shelf depreciation can be made up
The limitation of normal type, realizes all standing of shelf depreciation Exception Type diagnosis.
In the present embodiment, live testing apparatus for local discharge includes data acquisition module, aggregation of data processing module, data
Communication module, data disaply moudle and data memory module.Wherein, data acquisition module obtains part from partial discharge sensor and puts
Electric data.Aggregation of data processing module handles Partial Discharge Data to obtain the shelf depreciation map of power equipment.
Data memory module is used to store the diagnostic result of shelf depreciation map and cloud platform feedback, data disaply moudle
For showing diagnostic result or showing other critical informations.
As shown in Fig. 2, cloud platform, for using a kind of power equipment shelf depreciation defect type diagnostic method method of the present invention
Shelf depreciation diagnostic model is constructed, and the shelf depreciation map obtained in real time is diagnosed to obtain diagnostic result.It will finally examine
Disconnected result is issued to live testing apparatus for local discharge by data communication module, when enabling live live detection personnel the first
Between grasp testing result and equipment running status, carry out data diagnosis without live detection personnel itself, the scene part of reduction is put
Requirement of the electric live detection to personnel technical ability itself, can effectively improve work on the spot efficiency.
As shown in figure 3, a kind of power equipment shelf depreciation defect type diagnostic method provided by the invention, including walk as follows
It is rapid:
Step S1: all kinds of shelf depreciation defects pair of power equipment are extracted from history on-the-spot test and/or disintegration process data
The shelf depreciation map answered；
Wherein, Partial Discharge Data is acquired by partial discharge sensor, and generates shelf depreciation map.Wherein, the present embodiment
In, the shelf depreciation map of the power equipment acquired according to 80%, 10%, 10% ratio cut partition is to correspond to shelf depreciation defect class
Type label obtains training set, verifying collection and test set.Wherein, the corresponding electricity of each sample standard deviation in training set, verifying collection and test set
One group of shelf depreciation map of power equipment and local discharge defect type label.Training set data is for training depth convolutional Neural
Network, verification machine data are used to adjust the network after training set training, and carry out over-fitting adjusting.Test set data are for judging
Obtain the quality of shelf depreciation diagnostic model.In other feasible embodiments, above-mentioned set can be divided according to other ratios,
There is only training set or there is only training set and collection can also be verified simultaneously.The present invention is to this without specifically limiting.
Step S2: the shelf depreciation map and shelf depreciation defect type mark of power equipment in step S1 training set are utilized
It signs training depth convolutional neural networks and obtains shelf depreciation diagnostic model；
Wherein, the input data of shelf depreciation diagnostic model is the shelf depreciation map of power equipment, and output data is electricity
The shelf depreciation diagnostic result of power equipment, the diagnostic result include whether that there are shelf depreciation and shelf depreciation defect classes
Type.The generating process of shelf depreciation diagnostic model includes two large divisions in the present embodiment, is to utilize training set training depth respectively
Secondly convolutional neural networks adjust the neural network trained using verifying collection and obtain shelf depreciation diagnostic model.
Wherein, training depth convolutional neural networks when, setting training 10 groups, every group training 1000 times, every time complete instruct
Convolutional layer utilizes back-propagation algorithm modification weight to update depth convolutional neural networks automatically after white silk, and is trained next time.
Record trains the accuracy rate for obtaining model every time, by the highest model of accuracy rate in the group respectively current group of output, compares 10 groups of moulds
Type takes accuracy rate is highest to be used as final algorithm output model.By taking turns training, accuracy rate is made to reach 98% or more more.Wherein,
Existing training method that the present invention uses trains depth convolutional neural networks, is not related to the tune of the internal logic of its training algorithm
It is whole, therefore to its process without specifically describing.For example: training process is as follows:
When using training set training, in the forward propagation process, the graph data of input passes through the convolution of multilayer convolutional layer
With pondization processing, feature vector is proposed, and feature vector is passed to full articulamentum and obtains classification recognition result；According to Classification and Identification
As a result shelf depreciation diagnostic result is obtained, and by shelf depreciation diagnostic result compared with expected result, carries out back-propagation process,
The depth convolutional Neural net that weight modification is updated is carried out to each layer according to the error of shelf depreciation diagnostic result and desired value
Network, then trained next time.
For example: accuracy verifying being carried out to trained depth convolutional neural networks using training set: verifying being concentrated each
Depth convolutional neural networks after the shelf depreciation map input training of a sample obtain the shelf depreciation diagnosis knot of power equipment
Fruit；And identify whether shelf depreciation diagnostic result is correctly obtained verifying collection based on the corresponding shelf depreciation defect type label of sample
Accuracy rate；Model parameter, which is adjusted, further according to accuracy rate obtains shelf depreciation diagnostic model.
Step S3: Partial Discharge Data of the acquisition to diagnosing electric power, and pre-processed to obtain shelf depreciation map；
Step S4: the shelf depreciation map of step S3 is input in the shelf depreciation diagnostic model of step S2 and obtains part
Discharge diagnostic result.
By the above method, shelf depreciation diagnostic result can be obtained in real time, and transmits and shelf depreciation live detection is sent to fill
It sets, so as to live detection, personnel provide judgment basis in time, and live live detection personnel is enable to grasp detection knot at the first time
Fruit and equipment running status.When applied to diagnostic system, shelf depreciation diagnostic model is applied to cloud platform by the present invention, is encapsulated as
Interface is diagnosed, dynamic base is generated.By calling the interface, the various map files to pass in network provide examines cloud platform
Disconnected service.
It is emphasized that example of the present invention be it is illustrative, without being restrictive, thus the present invention it is unlimited
Example described in specific embodiment, other all obtained according to the technique and scheme of the present invention by those skilled in the art
Embodiment does not depart from present inventive concept and range, and whether modification or replacement, also belong to protection model of the invention
It encloses.
Claims (8)
1. a kind of diagnostic method of the shelf depreciation defect type of power equipment, characterized by the following steps:
Step S1: it is corresponding that all kinds of shelf depreciation defects of power equipment are extracted from history on-the-spot test and/or disintegration process data
Shelf depreciation map；
The shelf depreciation map is for indicating local discharge characteristic, the local discharge characteristic and shelf depreciation defect type phase
It is corresponding；
Step S2: the shelf depreciation map and shelf depreciation defect type label training depth of power equipment in step S1 are utilized
Convolutional neural networks obtain shelf depreciation diagnostic model；
Wherein, the input data of shelf depreciation diagnostic model is the shelf depreciation map of power equipment, and output data sets for electric power
Standby shelf depreciation diagnostic result, the diagnostic result include whether that there are shelf depreciation and shelf depreciation defect types；
Step S3: the shelf depreciation map of the acquisition to diagnosing electric power；
Step S4: the shelf depreciation map of step S3 is input in the shelf depreciation diagnostic model of step S2 and obtains part
Discharge diagnostic result.
2. according to the method described in claim 1, it is characterized by: the power equipment that will be acquired in step 2 according to preset ratio
Shelf depreciation map and shelf depreciation defect type label divided to obtain training set and verifying collection, training set and verifying
Each sample standard deviation is concentrated to correspond to the one group of shelf depreciation map and shelf depreciation defect type label of power equipment, wherein benefit
The shelf depreciation diagnostic model in step S2 is obtained with the training depth convolutional neural networks of training set.
3. according to the method described in claim 1, it is characterized by: the shelf depreciation defect type includes metal interface and gold
Belong to electric discharge, insulator interface and the insulator interface between electric discharge, metal tip electric discharge, metal interface and the insulator interface between interface
Between electric discharge, the creeping discharge of solid insulation and gas-insulated intersection, solid insulation and fluid insulation intersection along face
Electric discharge, the electric discharge inside solid insulation, electric discharge, metal particle electric discharge inside fluid insulation.
4. according to the method described in claim 1, it is characterized by: shelf depreciation map described in one group of power equipment at least wraps
Include superfrequency PRPD&PRPS map, high frequency PRPD&PRPS map, ultrasonic wave PRPD&PRPS map, ultrasonic pulse map.
5. according to the method described in claim 4, it is characterized by: the local discharge characteristic includes at least shelf depreciation width
Value-phase property, partial discharge pulse's quantity-phase property, shelf depreciation amplitude versus time feature, partial discharge pulse's quantity-
Temporal characteristics, shelf depreciation amplitude-pulse flight time.
6. according to the method described in claim 1, it is characterized by: the shelf depreciation map in step S1 is sensed using partial discharge
Device acquisition Partial Discharge Data generate, the partial discharge sensor include extra-high video sensor, contact ultrasonic sensor,
Non-contact ultrasonic sensor, high frequency sensors.
7. a kind of diagnostic system based on any one of claim 1-6 the method, it is characterised in that: including sequentially connected office
Put sensor, live testing apparatus for local discharge and cloud platform；
Wherein, partial discharge sensor is used to acquire the corresponding Partial Discharge Data of all kinds of shelf depreciations of power equipment；
The live testing apparatus for local discharge includes data acquisition module, aggregation of data processing module, data communication module,
In, data acquisition module obtains Partial Discharge Data from partial discharge sensor, and the aggregation of data processing module is used for shelf depreciation
Data are handled to obtain the corresponding shelf depreciation map of all kinds of shelf depreciation defect types of power equipment；
The cloud platform, for constructing shelf depreciation diagnostic model and using the shelf depreciation diagnostic model to power equipment
The diagnosis of real-time perfoming shelf depreciation defect type, and send diagnostic result to the live testing apparatus for local discharge, it is described
Live testing apparatus for local discharge receives diagnostic result by data communication module.
8. diagnostic system according to claim 7, it is characterised in that: the partial discharge sensor and shelf depreciation live detection
The wired or wireless connection of device, the live testing apparatus for local discharge are connect with cloud platform by 4G communication network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910717073.5A CN110412431A (en) | 2019-08-05 | 2019-08-05 | A kind of diagnostic method and diagnostic system of the shelf depreciation defect type of power equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910717073.5A CN110412431A (en) | 2019-08-05 | 2019-08-05 | A kind of diagnostic method and diagnostic system of the shelf depreciation defect type of power equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110412431A true CN110412431A (en) | 2019-11-05 |
Family
ID=68365721
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910717073.5A Pending CN110412431A (en) | 2019-08-05 | 2019-08-05 | A kind of diagnostic method and diagnostic system of the shelf depreciation defect type of power equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110412431A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112034310A (en) * | 2020-07-31 | 2020-12-04 | 国网山东省电力公司东营供电公司 | Partial discharge defect diagnosis method and system for combined electrical appliance |
| CN113055270A (en) * | 2021-03-09 | 2021-06-29 | 山东鲁能软件技术有限公司 | Partial discharge map analysis system, method and device based on artificial neural network |
| CN113378960A (en) * | 2021-06-25 | 2021-09-10 | 海南电网有限责任公司电力科学研究院 | Training method of partial discharge detection model, detection information determining method and device |
| CN113391172A (en) * | 2021-05-31 | 2021-09-14 | 国网山东省电力公司电力科学研究院 | Partial discharge diagnosis method and system based on time sequence integration and used for multi-source ultrasonic detection |
| CN113655348A (en) * | 2021-07-28 | 2021-11-16 | 国网湖南省电力有限公司 | Power equipment partial discharge fault diagnosis method based on deep twin network, system terminal and readable storage medium |
| CN113884844A (en) * | 2021-11-18 | 2022-01-04 | 国网山东省电力公司电力科学研究院 | Transformer partial discharge type identification method and system |
| CN115267462A (en) * | 2022-09-30 | 2022-11-01 | 丝路梵天(甘肃)通信技术有限公司 | Partial discharge type identification method based on self-adaptive label generation |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104569745A (en) * | 2013-10-24 | 2015-04-29 | 国家电网公司 | Partial discharge live detection device and method |
| WO2015070513A1 (en) * | 2013-11-14 | 2015-05-21 | 国家电网公司 | Pattern recognition method for partial discharge of three-phase in one enclosure type ultrahigh voltage gis |
| CN105629144A (en) * | 2016-03-11 | 2016-06-01 | 华乘电气科技（上海）股份有限公司 | High voltage switchgear partial discharge diagnosis method and system based on fingerprint database |
| CN106556781A (en) * | 2016-11-10 | 2017-04-05 | 华乘电气科技（上海）股份有限公司 | Shelf depreciation defect image diagnostic method and system based on deep learning |
| CN109116203A (en) * | 2018-10-31 | 2019-01-01 | 红相股份有限公司 | Power equipment partial discharges fault diagnostic method based on convolutional neural networks |
| CN109685138A (en) * | 2018-12-25 | 2019-04-26 | 东南大学 | A kind of XLPE power cable shelf depreciation kind identification method |
-
2019
- 2019-08-05 CN CN201910717073.5A patent/CN110412431A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104569745A (en) * | 2013-10-24 | 2015-04-29 | 国家电网公司 | Partial discharge live detection device and method |
| WO2015070513A1 (en) * | 2013-11-14 | 2015-05-21 | 国家电网公司 | Pattern recognition method for partial discharge of three-phase in one enclosure type ultrahigh voltage gis |
| CN105629144A (en) * | 2016-03-11 | 2016-06-01 | 华乘电气科技（上海）股份有限公司 | High voltage switchgear partial discharge diagnosis method and system based on fingerprint database |
| CN106556781A (en) * | 2016-11-10 | 2017-04-05 | 华乘电气科技（上海）股份有限公司 | Shelf depreciation defect image diagnostic method and system based on deep learning |
| CN109116203A (en) * | 2018-10-31 | 2019-01-01 | 红相股份有限公司 | Power equipment partial discharges fault diagnostic method based on convolutional neural networks |
| CN109685138A (en) * | 2018-12-25 | 2019-04-26 | 东南大学 | A kind of XLPE power cable shelf depreciation kind identification method |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112034310A (en) * | 2020-07-31 | 2020-12-04 | 国网山东省电力公司东营供电公司 | Partial discharge defect diagnosis method and system for combined electrical appliance |
| CN113055270A (en) * | 2021-03-09 | 2021-06-29 | 山东鲁能软件技术有限公司 | Partial discharge map analysis system, method and device based on artificial neural network |
| CN113391172A (en) * | 2021-05-31 | 2021-09-14 | 国网山东省电力公司电力科学研究院 | Partial discharge diagnosis method and system based on time sequence integration and used for multi-source ultrasonic detection |
| CN113378960A (en) * | 2021-06-25 | 2021-09-10 | 海南电网有限责任公司电力科学研究院 | Training method of partial discharge detection model, detection information determining method and device |
| CN113655348A (en) * | 2021-07-28 | 2021-11-16 | 国网湖南省电力有限公司 | Power equipment partial discharge fault diagnosis method based on deep twin network, system terminal and readable storage medium |
| CN113655348B (en) * | 2021-07-28 | 2023-12-08 | 国网湖南省电力有限公司 | Power equipment partial discharge fault diagnosis method, system terminal and readable storage medium based on deep twin network |
| CN113884844A (en) * | 2021-11-18 | 2022-01-04 | 国网山东省电力公司电力科学研究院 | Transformer partial discharge type identification method and system |
| CN115267462A (en) * | 2022-09-30 | 2022-11-01 | 丝路梵天(甘肃)通信技术有限公司 | Partial discharge type identification method based on self-adaptive label generation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110412431A (en) | A kind of diagnostic method and diagnostic system of the shelf depreciation defect type of power equipment | |
| CN104569688B (en) | A kind of distribution terminal automatization test system and method | |
| CN110411580B (en) | Diagnosis method and diagnosis system for heating defects of electric equipment | |
| CN109477863A (en) | Method and system for the Dynamic Fault-Detection in power network | |
| CN108008287A (en) | Track circuit failure monitoring platform and its monitoring method based on frequency-shift signaling | |
| CN106443326A (en) | Test system and method for fault locator of power distribution network | |
| CN114640173A (en) | Early warning model of transformer and generator based on many characteristic quantities | |
| CN108333547A (en) | A kind of distribution line failure indicator automatic checkout system and method | |
| CN109649432A (en) | Cloud platform rail integrity monitoring systems and method based on guided wave technology | |
| CN109212378A (en) | Middle and low voltage network Fault Locating Method | |
| CN107942210A (en) | The classification of transformer pulse electric current Partial Discharge and denoising method and system | |
| CN107450440A (en) | Intelligent Oil Filling Monitoring and Control System and its monitoring method | |
| WO2023284127A1 (en) | Vibration-signal-based online monitoring system for gil defects, and method | |
| CN108073722A (en) | The automatic check device and method of a kind of newly-built substation boss station figure and model | |
| CN108303962A (en) | A kind of ultrasonic fusing data monitoring device and method | |
| CN110888007B (en) | Electric automobile wireless charging interoperability test evaluation system and method thereof | |
| CN116298684A (en) | Fault studying, judging and positioning system for power distribution network | |
| CN107478997A (en) | A kind of base station storage batteries appraisal procedure and device | |
| CN114062839A (en) | Railway power line fault positioning device and method thereof | |
| CN106526460A (en) | Circuit fault locating method and device | |
| CN105867356B (en) | A kind of water pump controller test system and IP address setting method | |
| CN106324364B (en) | A kind of determination method and system of direct current micromotor product | |
| CN109116118A (en) | A kind of earth-continuity self-operated measuring unit and its application method | |
| CN107507657A (en) | RPI rods position probe coil test system and method | |
| CN107227665A (en) | Gapless track destressing method for monitoring state, apparatus and system |
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: 20191105 |