CN104849620A - Grounding grid fault diagnosis method based on BP neural network - Google Patents
Grounding grid fault diagnosis method based on BP neural network Download PDFInfo
- Publication number
- CN104849620A CN104849620A CN201510281775.5A CN201510281775A CN104849620A CN 104849620 A CN104849620 A CN 104849620A CN 201510281775 A CN201510281775 A CN 201510281775A CN 104849620 A CN104849620 A CN 104849620A
- Authority
- CN
- China
- Prior art keywords
- neural network
- node
- excitation
- test point
- branch road
- 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
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 40
- 238000003745 diagnosis Methods 0.000 title claims abstract description 25
- 238000000034 method Methods 0.000 title claims abstract description 14
- 230000005284 excitation Effects 0.000 claims abstract description 32
- 230000003278 mimic effect Effects 0.000 claims description 18
- 239000007787 solid Substances 0.000 claims description 4
- 230000007797 corrosion Effects 0.000 description 5
- 238000005260 corrosion Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 239000003638 chemical reducing agent Substances 0.000 description 2
- 239000004020 conductor Substances 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000004807 localization Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 208000032368 Device malfunction Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
Landscapes
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
The invention discloses a grounding grid fault diagnosis method based on a BP neural network, which comprises the following steps: simulating a grounding network to be diagnosed into a planar grid structure, selecting a pair of adjacent nodes as excitation nodes, and forming m accessible test points by a plurality of pairs of adjacent nodes; establishing a grounding network analog circuit, applying direct current power supply excitation, sequentially selecting and disconnecting a pair of adjacent accessible test points from the accessible test points to measure and construct a plurality of groups of test data and train a BP neural network; and applying direct-current power supply excitation to the excitation node of the grounding network to be diagnosed, generating an input vector according to the voltage of the accessible test point, inputting the input vector into the BP neural network, and diagnosing the fault branch of the grounding network to be diagnosed according to the output vector. The invention can provide guarantee for the safe operation of the grounding grid, and has the advantages of accurate and reliable fault positioning, small field workload, high diagnosis speed, high diagnosis efficiency, small field workload, simple operation of fault diagnosis, easy popularization and convenient popularization.
Description
Technical field
The present invention relates to the Fault Diagnosis for Grounding Grids technology in electric system, be specifically related to a kind of Fault Diagnosis for Grounding Grids method based on BP neural network.
Background technology
The safe operation of grounded screen is as the key factor ensureing safe operation of power system, and its Anticorrosion Problems has become the problem that electric power enterprise needs solution badly.Grounded screen, as transformer station's AC/DC equipment ground and thunder proof protection ground in electric system, plays an important role to the safe operation of electric system.When grounded screen existing defects, when electric system generation ground short circuit defect or when being struck by lightning, the current potential of equipment ground point and the partial potential difference on earth's surface all can raise extremely, directly jeopardize life and the device security of operations staff.The conductor of grounded screen is embedded in underground, often make performance depreciation because of soil corrosion for many years, when causing system earth short circuit, earth potential is abnormal raises or skewness, serious threat operations staff safety, also secondary device be may destroy because of counterattack or cable skin circulation, detection or opertaing device malfunction, tripping and expansion accident caused.Earth mat ground connection performance is generally judged indirectly by the size of stake resistance, but Grounding Grid situation cannot be understood, and earth mat conductor corrode even breakpoint time, stake resistance still may be normal, if ground network ground resistance to be found is defective or after causing accident, dig on a wide area searches grounded screen breakpoint and corrosion section again, be then the method that blindness is large, workload is large, also affect Operation of Electric Systems.
Summary of the invention
The technical problem to be solved in the present invention is: the complicacy when grounded screen fault for prior art checks, uncertainty, the corrosion of friction reducer and pollution problem, provide a kind of and can provide safeguard for the safe operation of grounded screen, localization of fault accurately and reliably, work on the spot amount is less, diagnosis speed is fast, diagnosis efficiency is high, work on the spot amount is less, fault diagnosis simple to operate, be easy to the Fault Diagnosis for Grounding Grids method based on BP neural network promoted.
In order to solve the problems of the technologies described above, the technical solution used in the present invention is:
Based on a Fault Diagnosis for Grounding Grids method for BP neural network, step comprises:
1) when not considering electric capacity, solid netted follow-up disconnecting earth mat is modeled as the plane net grating texture comprising n node and b bar branch road, from n node, select a pair adjacent node as excitation node, from n node, select multipair adjacent node, obtain altogether m can and test point;
2) grounded screen mimic channel is set up based on described plane net grating texture, excitation node for described grounded screen mimic channel applies direct supply excitation, then successively from m can and test point select disconnection a pair adjacent can and test point, measure respectively can and the voltage of test point, according to respectively can and the voltage of test point generate input vector X{x
1, x
2..., x
m, element x
irepresent i-th can and the voltage of test point, the output vector O{O that record is corresponding
1, O
2... O
b, if jth bar branch road disconnects, then element O in output vector O
jvalue be 1, all the other elements are 0, described input vector X and output vector O forms one group of test data, finally obtains organizing test data more;
3) nodes of initialization BP neural network input layer be can and test point quantity m, output layer nodes be the branch road quantity b of plane net grating texture, by described many group test data training BP neural networks;
4) apply direct supply excitation for follow-up disconnecting earth mat excitation node, to detect in follow-up disconnecting earth mat m can and test point voltage and generate input vector X{x
1, x
2..., x
m, input vector X is inputted the BP neural network after training, judges output vector O{O
1, O
2... O
bthe branch road of intermediate value corresponding to the element of 1 be fault branch.
Preferably, described step 2) in the grounded screen mimic channel set up based on plane net grating texture be the pure resistance mimic channel of plane net trellis of b bar branch road composition, and each horizontal branches in b bar branch road is in series with a resistance, a node in the corresponding described plane net grating texture of tie point between each group adjacent legs.
Preferably, described step 2) and step 3) in encourage node apply direct supply excitation size of current be 10A.
Preferably, the BP neural network in described step 3) is three layers of BP neural network.
The Fault Diagnosis for Grounding Grids method that the present invention is based on BP neural network has following advantage:
1, the present invention is by being modeled as solid netted follow-up disconnecting earth mat the plane net grating texture comprising n node and b bar branch road when not considering electric capacity, then grounded screen mimic channel is set up based on plane net grating texture, carry out fault simulation obtain many group test datas and train BP neural network as expert system, then to detect in follow-up disconnecting earth mat m can and test point voltage and generate input vector, input vector X is inputted the BP neural network after training, judge that the branch road of output vector intermediate value corresponding to the element of 1 is as fault branch, therefore, it is possible to utilize the self-learning function of BP neural network, directly utilize m in follow-up disconnecting earth mat can and the voltage of test point carry out quick diagnosis and to be out of order branch road, excavation is not needed to search grounded screen breakpoint and corrosion section, effectively can solve the complicacy during grounded screen fault inspection of prior art, uncertain, the corrosion of friction reducer and pollution problem, can provide safeguard for the safe operation of grounded screen, there is localization of fault advantage accurately and reliably.
2, of the present invention for the excitation of follow-up disconnecting earth mat excitation node applying direct supply, to detect in follow-up disconnecting earth mat m can and test point voltage after, input vector X is inputted the BP neural network after training, the diagnostic result of fault branch can be obtained fast, there is the advantage that diagnosis speed is fast, diagnosis efficiency is high.
3, most of step of the present invention is the step not needing to complete at the scene, only apply direct supply excitation for follow-up disconnecting earth mat excitation node, to detect in follow-up disconnecting earth mat m can and the voltage of test point just need measuring voltage data at the scene, therefore the work on the spot amount of whole fault diagnosis is less.
4, the present invention is based on plane net grating texture and set up grounded screen mimic channel, excitation node for grounded screen mimic channel applies direct supply excitation, when diagnosing at the scene simultaneously, direct supply excitation is applied for follow-up disconnecting earth mat excitation node, owing to adopting direct supply excitation, therefore do not need the reactance characteristic considering grounded screen during grounded screen mimic channel, therefore, it is possible to greatly simplify the structure of grounded screen mimic channel, greatly simplifie the operation of fault diagnosis, be easy to promote.
Accompanying drawing explanation
Fig. 1 is the basic skills schematic flow sheet of the embodiment of the present invention.
Fig. 2 is the plane net trellis structural representation obtained in the embodiment of the present invention.
Fig. 3 is the circuit theory schematic diagram of the grounded screen mimic channel set up in the embodiment of the present invention.
Embodiment
Hereafter there to be certain follow-up disconnecting earth mat of 25 isolated nodes, 40 branch roads in electric system, the Fault Diagnosis for Grounding Grids method that the present invention is based on BP neural network is further described.
As shown in Figure 1, the present embodiment comprises based on the step of the Fault Diagnosis for Grounding Grids method of BP neural network:
1) when not considering electric capacity, solid netted follow-up disconnecting earth mat is modeled as the plane net grating texture (as shown in Figure 2) comprising n (n=25) individual node #1 ~ #24 and b (b=40) bar branch road, from n node, select a pair adjacent node as excitation node, from n node, select multipair adjacent node, obtain altogether m can and test point; In the present embodiment, concrete #17 and #18 a pair adjacent node of selecting is as excitation node, selects #3 and #4, #7 and #8, #13 and #14, #18 and #23 totally four pairs of adjacent nodes from n node, obtain altogether 8 can and test point;
2) grounded screen mimic channel is set up based on plane net grating texture, excitation node for grounded screen mimic channel applies direct supply excitation, then successively from m can and test point select disconnection a pair adjacent can and test point, measure respectively can and the voltage of test point, according to respectively can and the voltage of test point generate input vector X{x
1, x
2..., x
m, element x
irepresent i-th can and the voltage of test point, the output vector O{O that record is corresponding
1, O
2... O
b, if jth bar branch road disconnects, then element O in output vector O
jvalue be 1, all the other elements are 0, input vector X and output vector O forms one group of test data, finally obtains organizing test data more;
3) nodes of initialization BP neural network input layer be can and test point quantity m, output layer nodes be the branch road quantity b of plane net grating texture, test data training BP neural network will be organized more;
4) apply direct supply excitation for follow-up disconnecting earth mat excitation node, to detect in follow-up disconnecting earth mat m can and test point voltage and generate input vector X{x
1, x
2..., x
m, input vector X is inputted the BP neural network after training, judges output vector O{O
1, O
2... O
bthe branch road of intermediate value corresponding to the element of 1 be fault branch.
As shown in Figure 3, described step 2) in the grounded screen mimic channel set up based on plane net grating texture be the pure resistance mimic channel of plane net trellis of b bar branch road composition, and each horizontal branches in b bar branch road is in series with a resistance, a node in the corresponding described plane net grating texture of tie point between each group adjacent legs.Such as, be in series with a resistance between node #1 and node #2, between node #2 and node #3, be in series with a resistance etc., the like.
In the present embodiment, step 2) and step 3) in encourage node to apply direct supply excitation size of current be 10A.
In the present embodiment, the BP neural network in step 3) is three layers of BP neural network.BP neural network is the neural network algorithm of current widespread use, and the three layers of BP neural network comprising input layer, hidden layer and output layer are the simplest BP neural network, have and realize simple advantage.
It should be noted that, in the present embodiment, train that BP neural network mainly comprises netinit, hidden layer export calculate, output layer exports calculatings, error calculation, right value update, threshold value renewal, iteration terminates the steps such as judgement.But, above-mentioned netinit, hidden layer export calculate, output layer exports calculatings, error calculation, right value update, threshold value renewal, to terminate the steps such as judgement be the known technology means of training BP neural network at present to iteration, therefore to repeat no more in the present embodiment.
The above is only the preferred embodiment of the present invention, protection scope of the present invention be not only confined to above-described embodiment, and all technical schemes belonged under thinking of the present invention all belong to protection scope of the present invention.It should be pointed out that for those skilled in the art, some improvements and modifications without departing from the principles of the present invention, these improvements and modifications also should be considered as protection scope of the present invention.
Claims (4)
1., based on a Fault Diagnosis for Grounding Grids method for BP neural network, it is characterized in that step comprises:
1) when not considering electric capacity, solid netted follow-up disconnecting earth mat is modeled as the plane net grating texture comprising n node and b bar branch road, from n node, select a pair adjacent node as excitation node, from n node, select multipair adjacent node, obtain altogether m can and test point;
2) grounded screen mimic channel is set up based on described plane net grating texture, excitation node for described grounded screen mimic channel applies direct supply excitation, then successively from m can and test point select disconnection a pair adjacent can and test point, measure respectively can and the voltage of test point, according to respectively can and the voltage of test point generate input vector X{x
1, x
2..., x
m, element x
irepresent i-th can and the voltage of test point, the output vector O{O that record is corresponding
1, O
2... O
b, if jth bar branch road disconnects, then element O in output vector O
jvalue be 1, all the other elements are 0, described input vector X and output vector O forms one group of test data, finally obtains organizing test data more;
3) nodes of initialization BP neural network input layer be can and test point quantity m, output layer nodes be the branch road quantity b of plane net grating texture, by described many group test data training BP neural networks;
4) apply direct supply excitation for follow-up disconnecting earth mat excitation node, to detect in follow-up disconnecting earth mat m can and test point voltage and generate input vector X{x
1, x
2..., x
m, input vector X is inputted the BP neural network after training, judges output vector O{O
1, O
2... O
bthe branch road of intermediate value corresponding to the element of 1 be fault branch.
2. the Fault Diagnosis for Grounding Grids method based on BP neural network according to claim 1, it is characterized in that: described step 2) in the grounded screen mimic channel set up based on plane net grating texture be the pure resistance mimic channel of plane net trellis of b bar branch road composition, and each horizontal branches in b bar branch road is in series with a resistance, a node in the corresponding described plane net grating texture of tie point between each group adjacent legs.
3. the Fault Diagnosis for Grounding Grids method based on BP neural network according to claim 1 and 2, is characterized in that: described step 2) and step 3) in encourage node apply direct supply excitation size of current be 10A.
4. the Fault Diagnosis for Grounding Grids method based on BP neural network according to claim 3, is characterized in that, the BP neural network in described step 3) is three layers of BP neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510281775.5A CN104849620A (en) | 2015-05-28 | 2015-05-28 | Grounding grid fault diagnosis method based on BP neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510281775.5A CN104849620A (en) | 2015-05-28 | 2015-05-28 | Grounding grid fault diagnosis method based on BP neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104849620A true CN104849620A (en) | 2015-08-19 |
Family
ID=53849407
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510281775.5A Pending CN104849620A (en) | 2015-05-28 | 2015-05-28 | Grounding grid fault diagnosis method based on BP neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104849620A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105242173A (en) * | 2015-09-14 | 2016-01-13 | 吉林大学 | Frequency domain electromagnetic method-based grounding grid fault automatic diagnosis method |
| CN107561407A (en) * | 2017-08-18 | 2018-01-09 | 国网辽宁省电力有限公司电力科学研究院 | A kind of distributed grounding net of transformer substation detection and appraisal procedure |
| CN108416103A (en) * | 2018-02-05 | 2018-08-17 | 武汉大学 | A kind of method for diagnosing faults of electric automobile of series hybrid powder AC/DC convertor |
| CN109239533A (en) * | 2018-11-16 | 2019-01-18 | 国网山东省电力公司电力科学研究院 | A kind of Fault Locating Method of the extra high voltage direct current transmission line based on artificial neural network |
| CN112055284A (en) * | 2019-06-05 | 2020-12-08 | 北京地平线机器人技术研发有限公司 | Echo cancellation method, neural network training method, apparatus, medium, and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0829480A (en) * | 1994-07-14 | 1996-02-02 | Tokyo Electric Power Co Inc:The | Method for specifying accident section of transmission line and situation of accident |
| CN103439631A (en) * | 2013-08-12 | 2013-12-11 | 国家电网公司 | Method and system for detecting corrosion state of grounding grid |
| CN104299032A (en) * | 2014-08-25 | 2015-01-21 | 国家电网公司 | Method for predicating corrosion rate of soil of transformer substation grounding grid |
-
2015
- 2015-05-28 CN CN201510281775.5A patent/CN104849620A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0829480A (en) * | 1994-07-14 | 1996-02-02 | Tokyo Electric Power Co Inc:The | Method for specifying accident section of transmission line and situation of accident |
| CN103439631A (en) * | 2013-08-12 | 2013-12-11 | 国家电网公司 | Method and system for detecting corrosion state of grounding grid |
| CN104299032A (en) * | 2014-08-25 | 2015-01-21 | 国家电网公司 | Method for predicating corrosion rate of soil of transformer substation grounding grid |
Non-Patent Citations (2)
| Title |
|---|
| 刘庆珍等: "电力系统接地网故障的无伤检测方法", 《高电压技术》 * |
| 易志鹏: "基于神经网络的接地网故障诊断研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105242173A (en) * | 2015-09-14 | 2016-01-13 | 吉林大学 | Frequency domain electromagnetic method-based grounding grid fault automatic diagnosis method |
| CN107561407A (en) * | 2017-08-18 | 2018-01-09 | 国网辽宁省电力有限公司电力科学研究院 | A kind of distributed grounding net of transformer substation detection and appraisal procedure |
| CN108416103A (en) * | 2018-02-05 | 2018-08-17 | 武汉大学 | A kind of method for diagnosing faults of electric automobile of series hybrid powder AC/DC convertor |
| CN109239533A (en) * | 2018-11-16 | 2019-01-18 | 国网山东省电力公司电力科学研究院 | A kind of Fault Locating Method of the extra high voltage direct current transmission line based on artificial neural network |
| CN112055284A (en) * | 2019-06-05 | 2020-12-08 | 北京地平线机器人技术研发有限公司 | Echo cancellation method, neural network training method, apparatus, medium, and device |
| CN112055284B (en) * | 2019-06-05 | 2022-03-29 | 北京地平线机器人技术研发有限公司 | Echo cancellation method, neural network training method, apparatus, medium, and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104849620A (en) | Grounding grid fault diagnosis method based on BP neural network | |
| Farajollahi et al. | Deployment of fault indicator in distribution networks: A MIP-based approach | |
| CN103605042B (en) | Fault Diagnosis for Grounding Grids method based on APSO algorithm | |
| CN105044559B (en) | The subregion method for diagnosing faults of grounding net of transformer substation | |
| CN101493491B (en) | Substation ground network defect synthesis diagnosis method and diagnosis system thereof | |
| CN106463945B (en) | The system and method for fault section in polyphase electric power network for identification | |
| CN102221644B (en) | System and method for online monitoring grounding networks of power stations and substations | |
| CN104599193A (en) | Rule base based single-phase ground fault positioning method of distribution network | |
| CN107561407B (en) | Distributed transformer substation grounding grid detection and evaluation method | |
| CN114236288B (en) | Fault positioning method based on power transmission line | |
| CN107271853A (en) | Electrical power distribution automatization system distribution low current grounding localization method and system | |
| CN105445615A (en) | Corrosion diagnosis method for transformer station grounding network | |
| CN109061391B (en) | Power grid fault diagnosis method and system based on computer vision tidal current diagram | |
| CN107797028A (en) | A kind of distribution arc grounding fault recognition method | |
| CN109613402A (en) | Power distribution network high resistance earthing fault detection method based on wavelet transformation and neural network | |
| Hutchins | Geomagnetically induced currents and their effect on power systems | |
| CN103412190A (en) | Switch-class device state evaluation method based on parameter on-line identification | |
| CN112395788A (en) | Personal safety accurate evaluation method for distribution network neutral point grounding type reconstruction | |
| Harrou et al. | Online model-based fault detection for grid connected PV systems monitoring | |
| Alilouch et al. | Intelligent Relay Based on Artificial Neural Networks ANN for Transmission Line | |
| CN104777399A (en) | Three-dimensional digitalized monitoring device and method for substation grounding grid | |
| CN114386235A (en) | Cable armor layer stray current calculation method, device, equipment and storage medium | |
| Sreewirote et al. | Classification of fault type on loop-configuration transmission system using support vector machine | |
| Dawalibi et al. | Mitigation of transformer saturation due to neutral HVDC currents | |
| CN105259474A (en) | Online fault range finding method of T-connection cable |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate 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: 20150819 |