CN107340456B - Power distribution network operating condition intelligent identification Method based on multiple features analysis - Google Patents
Power distribution network operating condition intelligent identification Method based on multiple features analysis Download PDFInfo
- Publication number
- CN107340456B CN107340456B CN201710381487.6A CN201710381487A CN107340456B CN 107340456 B CN107340456 B CN 107340456B CN 201710381487 A CN201710381487 A CN 201710381487A CN 107340456 B CN107340456 B CN 107340456B
- Authority
- CN
- China
- Prior art keywords
- signal
- power distribution
- distribution network
- wavelet
- identification
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000004458 analytical method Methods 0.000 title claims abstract description 10
- 239000000284 extract Substances 0.000 claims abstract description 22
- 238000012549 training Methods 0.000 claims abstract description 18
- 230000009466 transformation Effects 0.000 claims abstract description 14
- 230000008569 process Effects 0.000 claims abstract description 12
- 238000003066 decision tree Methods 0.000 claims abstract description 7
- 230000001052 transient effect Effects 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 9
- 230000002159 abnormal effect Effects 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000000737 periodic effect Effects 0.000 claims description 5
- 238000001228 spectrum Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims description 3
- 230000001960 triggered effect Effects 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 2
- 239000004744 fabric Substances 0.000 claims 1
- 238000013528 artificial neural network Methods 0.000 abstract description 18
- 230000003044 adaptive effect Effects 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 15
- 230000005284 excitation Effects 0.000 description 6
- 230000004913 activation Effects 0.000 description 4
- 241001269238 Data Species 0.000 description 2
- 208000025274 Lightning injury Diseases 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 210000002569 neuron Anatomy 0.000 description 2
- 238000000844 transformation Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
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
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
-
- 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
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The present invention discloses a kind of power distribution network operating condition intelligent identification Method based on multiple features analysis, data acquisition is carried out by misoperation operating condition of the online recording system of power distribution network to distribution feeder, using various features extracting method, signal characteristic extracting methods including time domain, frequency domain and wavelet transformation, extract a large amount of signal characteristics, such as current temporary state steady state characteristic；There is artificial neural network (ANN) model of adaptive learning characteristic by training to carry out every a kind of unusual service condition identification, and establish the Classification and Identification process shaped like decision tree, realize that the multi-state of power distribution network effectively identifies.
Description
Technical field
The present invention relates to the abnormalities of power distribution network operating condition to identify field.
Background technique
Power distribution network is the pith in electric system, is the key link for guaranteeing power supply quality and power grid efficient operation.
For ensure the intelligence of distribution net height operation, need to monitor feeder line operation data in real time, the timely early warning of abnormal conditions and
Quickly discovery is handled failure, wherein being the critical function of intelligent distribution network to the identification of feeder line unusual service condition.Power distribution network is caused to present
There are many reason of line is operating abnormally, such as phase fault, single-phase earthing, excitation surge current etc..
It is based on setting threshold value, according to certain of single factors that traditional distribution feeder, which is operating abnormally type identification mostly,
Logical relation is realized.This recognition mode is simple, it is difficult to consider the feeder voltage obtained under unusual service condition, current signal,
Also by a variety of enchancement factors such as system operation mode, abort situation, transition impedance and fault moment, therefore have the defects that certain.
And electric current moment is slightly changed, the unconspicuous operating condition of Feature change, such as single-phase big resistance transient earthing, until being at present
Only, it is equipped with fault detector and the distribution system of failure line selection technology still cannot achieve monitoring and identification to these operating conditions.
Summary of the invention
The purpose of the present invention is to propose to a kind of power distribution network operating condition intelligent identification Methods based on multiple features analysis, pass through distribution
It nets online recording system and data acquisition is carried out to the misoperation operating condition of distribution feeder, using various features extracting method, packet
The signal characteristic extracting methods for including time domain, frequency domain and wavelet transformation extract a large amount of signal characteristics, such as current temporary state steady state characteristic
Deng；Every a kind of unusual service condition identification is carried out by training artificial neural network (ANN) model with adaptive learning characteristic,
And the Classification and Identification process shaped like decision tree is established, realize that the multi-state of power distribution network effectively identifies.
To realize the present invention purpose and the technical solution adopted is that such, a kind of power distribution network work based on multiple features analysis
Condition intelligent identification Method, which comprises the following steps:
1) the online recording system for passing through power distribution network, three-phase synchronous recording is triggered when line failure, obtains A, B, C
Three-phase current recording signal and zero sequence current signal.
2) signal characteristic is extracted using multi-feature extraction method:
2-1) extract temporal signatures
In time domain scale, for by online recording system acquisition to current signal extract each spy as shown in Table 1
Sign amount.
The each characteristic component extracted in 1 time domain scale of table
Wherein, Ip,iIt is the collected current signal of detector,Indicate that A, B, C three-phase, p represent tetra- phase of A, B, C, Z, i generation
The periodic sequence of table recording signal, j represent the sampled point serial number in each period, and N is number of sampling points.
2-2) extract frequency domain character
In frequency domain, the direct current and second harmonic component of steady-state signal, believe using recording after occurring for failure
Number a cycle data of i~(i+m) carry out discrete Fourier transform analysis.By taking a cycle current signal as an example, using in Fu
Leaf series expansion obtains the frequency-domain transform result of second harmonic:
I in formulap,iIt is the collected current signal of detector, p represents tetra- phase of A, B, C, Z, and i is the period sequence of current signal
Column, N are number of sampling points, and n is n-th of sampled point, and j is that imaginary part indicates symbol.
Successively extract characteristic quantity as shown in Table 2: DC component content Idp, second harmonic component I2xWith second harmonic point
Measure content I2xp。
The each characteristic component extracted in 2 frequency domain of table
2-3) feature extracting method based on wavelet transformation analyzes transient signal
The transient signal that the moment occurs for distribution feeder misoperation is extracted with wavelet transformation.By original transient signal point
It is analyzed in solution to J different scale, extracts the low frequency and high fdrequency component of multiple frequency ranges.
2-3-1) collected abnormal signal is decomposed, extracts the high fdrequency component feature of unusual service condition signal.
Wherein AJ(k) the low frequency component coefficient obtained for k time-ofday signals through J rank Wavelet decomposing and recomposing, Di(k) high for the i-th rank
Frequency division coefficient of discharge.For uniform expression, D is usedJ+1(k) A is replacedJ(k), expression formula is converted to
The high fdrequency component feature of unusual service condition signal is Dfp(i), objects of statistics is that abnormal signal front and back half period occurs
The sum of high fdrequency component absolute coefficient, be expressed as
2-3-2) extract wavelet energy entropy and wavelet singular entropy
Wavelet energy entropy and wavelet singular entropy are for indicating within the period that unusual service condition occurs, and signal energy is in different frequencies
The confusion degree of section distribution.
The Energy-Entropy WEE of small echo is defined as follows formula:
Wherein pi=Ei/ E is defined on the signal power spectrum on different scale i time k, Ei(k)=| Di(k)|2,
For all moment on scale i energy and,It is approximately the gross energy of signal.
Coefficient D after wavelet transformation is reconstructedi(k) matrix D of (J+1) × M is constituted(J+1)×M, matrix is carried out odd
Different value is decomposed, and J+1 non-negative singular value σ can be obtainedi, then wavelet singular entropy WAE is defined as follows:
3) the characteristic set extracted according to step 2) is subjected to tree-shaped cluster, more points is established in the form of decision tree
Class identification process.Wherein, short circuit and ground connection belong to fault condition, in all operating conditions, need to short circuit and ground connection two major classes operating condition
Both operating conditions are identified first in accordance with the principle of " can fail to judge and not judge by accident ".Short-circuit (ground connection) will be originally used for by failing to judge
Operating condition is judged as that other operating condition type, the unusual service condition that erroneous judgement will not belong to short-circuit (ground connection) are classified into fault condition.
4) classifier in more Classification and Identification processes is using three layers of ANN model building.
4-1) utilize characteristic and unusual service condition classification composing training data set, training ANN classification device.ANN classification device
Using three_layer planar waveguide, all neuron activation functions in input layer and hidden layer are set as tan-sig function, defeated
Layer activation primitive is set as log-sig function out, carries out two Classification and Identifications, will need the operating condition collection real marking type classified to be
1 and 0, i.e., a kind of unusual service condition and other classification operating conditions that need to be identified.
The performance function that model 4-2) is arranged is mean square error:
Wherein, cost function is equal to cost (hw(xi),yi)=(hw(xi)-yi)2.X=(x1,x2,...,xi) it is input square
Battle array, each column xiFor the duty parameter of one group of input, hw(xi) be i-th of ANN model input output, y is known mark work
Condition type, M are training dataset number.
4-3) in the training of the coefficient of ANN classification device 1 (ground connection) and classifier 2 (short circuit), cost function needs, which follow, to be connect
Weight factor K (K > 1) is added in function in the principle on ground and short trouble " can fail to judge and not judge by accident ", and cost function becomes
cost(hw(xi),yi)=yi×(hw(xi)-1)2+K(1-yi)×hw(xi)2
5) by training with after test, control errors are provided to more Classification and Identification models within the scope of admission threshold
Online recording system carries out the multi-state identification of power distribution network.
The solution have the advantages that unquestionable: it is transported extremely according to the online recording system acquisition feeder line of power distribution network first
Row floor data extracts signal characteristic using the various features extracting method such as time domain, frequency domain and wavelet transformation, and is sent to
Learnt in ANN classification device and trained, is finally that node is established using form of decision tree as the power distribution network of frame using ANN classification device
Multi-state intelligent recognition model can effectively carry out power distribution network unusual service condition identification.
Detailed description of the invention
Fig. 1 excitation surge current operating condition recording electric current；
Fig. 2 polymorphic type operating mode's switch process；
Fig. 3 ground connection operating mode recognition result；
Fig. 4 Short-circuit Working Condition recognition result.
Specific embodiment
Below with reference to embodiment, the invention will be further described, but should not be construed the above-mentioned subject area of the present invention only
It is limited to following embodiments.Without departing from the idea case in the present invention described above, according to ordinary skill knowledge and used
With means, various replacements and change are made, should all include within the scope of the present invention.
A kind of power distribution network operating condition intelligent identification Method based on multiple features analysis, which comprises the following steps:
1) the online recording system for passing through power distribution network, three-phase synchronous recording is triggered when line failure, obtains each phase
Current recording signal and zero sequence current signal.Illustrate by taking Fig. 1 excitation surge current operating condition recording current data as an example comprising A, B, C
Three-phase current and zero sequence (Z) electric current, totally 16 cycles, sample rate 4kHz, each cycle number of sampling points are 82 to every phase current；
2) signal characteristic is extracted using multiple features feature extracting method
2-1) extract temporal signatures
In time domain scale, for online recording system acquisition to current signal according in table 1 calculation method extract
Each characteristic quantity.
The each characteristic component extracted in 1 time domain scale of table
I in upper tablep,iIt is the collected current signal of detector,Indicate that A, B, C three-phase, p represent tetra- phase of A, B, C, Z, i
1~16 period of recording signal is represented, j represents 1~82 sampled point in each period, and N is number of sampling points 82.
2-2) extract frequency domain character
In frequency domain, the direct current and second harmonic component of steady-state signal after occurring for failure, to the 8th of recording signal the
~10 cycle datas carry out discrete Fourier transform analysis.
By taking a cycle current signal as an example, using Fourier expansion, the frequency-domain transform result of second harmonic is obtained:
I in formulap,iIt is the collected current signal of detector, p represents tetra- phase of A, B, C, Z, and i is the period sequence of current signal
Column, N are number of sampling points, and n is n-th of sampled point, and j is that imaginary part indicates symbol.
According to the calculation method in table 2, characteristic quantity is successively extracted: DC component content Idp, second harmonic component I2xWith two
Order harmonic components content I2xp。
The each characteristic component extracted in 2 frequency domain of table
2-3) feature extracting method based on wavelet transformation divides transient signal
Original transient signal is decomposed and is analyzed on J different scale, the low frequency and high frequency of multiple frequency ranges are extracted
Component.The transient signal that the moment occurs for distribution feeder misoperation is extracted with wavelet transformation.
2-3-1) collected abnormal signal is decomposed, extracts the high fdrequency component feature of unusual service condition signal.
Wherein AJ(k) the low frequency component coefficient obtained for k time-ofday signals through J rank Wavelet decomposing and recomposing, Di(k) high for the i-th rank
Frequency division coefficient of discharge.For uniform expression, D is usedJ+1(k) A is replacedJ(k), expression formula is converted to
The high fdrequency component feature of unusual service condition signal is Dfp(i), objects of statistics is that abnormal signal front and back half period occurs
The sum of high fdrequency component absolute coefficient, be expressed as
In order to effectively decomposite the transient signal of distribution feeder unusual service condition, 2~4 rank wavelet transformations are generally used, this
In used 3 rank db5 wavelet transformations.
2-3-2) extract wavelet energy entropy and wavelet singular entropy
Indicate that signal energy is in different frequencies within the period that unusual service condition occurs with wavelet energy entropy and wavelet singular entropy
The confusion degree of section distribution.
The Energy-Entropy WEE of small echo is defined as follows formula,
Wherein pi=Ei/ E is defined on the signal power spectrum on different scale i time k, Ei(k)=| Di(k)|2,
For all moment on scale i energy and,It is approximately the gross energy of signal.
Coefficient D after wavelet transformation is reconstructedi(k) matrix D of (J+1) × M is constituted(J+1)×M, matrix is carried out odd
Different value is decomposed, and J+1 non-negative singular value σ can be obtainedi, then wavelet singular entropy WAE is defined as follows,
According to engineering experience, in order to represent the confusion degree of transient signal when abnormal signal occurs, in small wave energy
It measures in entropy and wavelet singular entropy calculating, the section using signal is the sample point of front and back 10 for detecting singular signal and occurring.
3) the characteristic set extracted according to step 2) is subjected to tree-shaped cluster, more points is established in the form of decision tree
Class identification process.The unusual service condition classification designed at present mainly includes the classifications such as ground connection, short circuit, excitation surge current, lightning stroke, power failure.
Such as Fig. 2, the classifier in identification process is all two classifiers.Wherein, short circuit and ground connection belong to fault condition, in all operating conditions
In, the principle in accordance with " can fail to judge and not judge by accident " is needed to short circuit and ground connection two major classes operating condition, short circuit will be originally used for by, which failing to judge, (connects
Ground) operating condition is judged as other operating condition type, the unusual service condition that erroneous judgement will not belong to short-circuit (ground connection) is classified into fault condition, head
First both operating conditions are identified.
4) classifier in more Classification and Identification processes uses three layers of feed forward-fuzzy control (ANN) model construction.
4-1) utilize characteristic and unusual service condition classification composing training data set, training ANN classification device.ANN classification device
Using three-layer network topological structure, all neuron activation functions in input layer and hidden layer are set as tan-sig function, defeated
Layer activation primitive is set as log-sig function out, is 1 and 0 by the operating condition collection real marking type for needing to classify, i.e., need to identify
A kind of unusual service condition and other classification operating conditions.
4-2) performance function of model is mean square error
Wherein, cost function is equal to cost (hw(xi),yi)=(hw(xi)-yi)2, X=(x1,x2,...,xi) it is input square
Battle array, each column xiFor the duty parameter of one group of input, hw(xi) be i-th of ANN model input output, y is known mark work
Condition type.
Training ANN classification device, uses 2924 groups of floor datas (including all operating conditions that need to be identified), wherein being grounded 522
Group, 236 groups of short circuit 560 groups of excitation surge current, are struck by lightning 601 groups, send a telegram in reply 524 groups, have a power failure 293 groups, 188 groups of other operating conditions, by 7:3
Pro rate training set and test set, training set be 2046 groups, test set be 878 groups, wherein for prevent over-fitting model instruct
Regular terms is added in performance function during white silk, and wherein regularization coefficient is set as 0.00001.
4-3) in ANN classification device 1 (ground connection) and the training of 2 (short circuit) coefficients, cost function needs to follow ground connection and short circuit
Failure can fail to judge the principle that do not judge by accident, and weight factor K (K > 1) is added in function, and ground connection is labeled as 1 with Short-circuit Working Condition, cost
Function becomes,
cost(hw(xi),yi)=yi×(hw(xi)-1)2+K(1-yi)×hw(xi)2
In trained and 2 process of testing classification device 1 and classifier, weight factor is adjusted.Such as Fig. 3 and 4, from training
From the point of view of test result, when weight factor K from 1 increase to 4 during, erroneous judgement error has the decline of certain amplitude, and K is equal to 4
When already close to 0, while total error is also without too big variation.
5) by training with after test, control errors are provided to more Classification and Identification models within the scope of admission threshold
Online recording system carries out the multi-state identification of power distribution network.
Since floor data is that random mixing is reallocated, take the average value of 10 experiments as last result.Table 3 is shown
The training and test error result of multi-state identification model.Multiplexing based on decision-tree model as can be seen from the results
Condition classification process combines the ANN classification device of each operating condition, can control multi-state identification error 6% hereinafter, simultaneously in a model
Weight factor is added, meets the requirement for not judging fault condition by accident as far as possible.So far, trained more Classification and Identification models can provide
The multi-state identification of power distribution network is carried out to online recording system.
3 operating mode's switch error result of table
| Identify operating condition type | Training set error (%) | Test set error (%) |
| Ground connection | 1.47 | 4.56 |
| Short circuit | 1.07 | 2.59 |
| Excitation surge current | 1.04 | 1.79 |
| Lightning stroke | 0.94 | 1.23 |
| Telegram in reply | 0.35 | 0.75 |
| Have a power failure | 0.39 | 0.95 |
| Multi-state | 4.99 | 5.12 |
Claims (1)
1. a kind of power distribution network operating condition intelligent identification Method based on multiple features analysis, which comprises the following steps:
1) the online recording system for passing through power distribution network, three-phase synchronous recording is triggered when line failure, obtains A, B, C three-phase
Current recording signal and zero sequence current signal；
2) signal characteristic is extracted using multi-feature extraction method:
2-1) extract temporal signatures
In time domain scale, for by online recording system acquisition to current signal extract it is as follows:
Period maximum value Imax: Imax,p(i)=max (Ip,i(j))
Period minimum value Imin: Imin,p(i)=min (Ip,i(j))
Periodic Mean Imean:
Periodic variance Ivar:
Period mean square deviation Irms:
Maximum mean square deviation Imar: Imar,p=max (Irms,p(i))
Minimum Mean Square Error Imir: Imir,p=min (Irms,p(i))
The difference maximum value I of root mean squaremard: Imard,p=max (Irms,p(2)-Irms,p(1),...,Irms,p(i+1)-Irms,p(i))
Tri-phase unbalance factor IDUB:
Wherein, Ip,iIt is the collected current signal of detector,Indicate that A, B, C three-phase, p represent tetra- phase of A, B, C, Z, i represents record
The periodic sequence of wave signal, j represent the sampled point serial number in each period, and N is number of sampling points；
2-2) extract frequency domain character
In frequency domain, the direct current and second harmonic component of steady-state signal after occurring for failure, using the i of recording signal
A cycle data of~(i+m) carries out discrete Fourier transform analysis；By taking a cycle current signal as an example, Fourier space is used
Expansion, obtains the frequency-domain transform result of second harmonic:
I in formulap,iIt is the collected current signal of detector, p represents tetra- phase of A, B, C, Z, and i is the periodic sequence of current signal, N
For number of sampling points, n is n-th of sampled point, and j is that imaginary part indicates symbol；
Extract following characteristic quantity:
DC component content Idp:
Second harmonic component I2x:
Second harmonic component content I2xp:
2-3) feature extracting method based on wavelet transformation analyzes transient signal
The transient signal that the moment occurs for distribution feeder misoperation is extracted with wavelet transformation；Original transient signal is decomposed
It is analyzed on J different scale, extracts the low frequency and high fdrequency component of multiple frequency ranges；
2-3-1) collected abnormal signal is decomposed, extracts the high fdrequency component feature of unusual service condition signal:
Wherein AJ(k) the low frequency component coefficient obtained for k time-ofday signals through J rank Wavelet decomposing and recomposing, DiIt (k) is the high frequency division of the i-th rank
Coefficient of discharge；For uniform expression, D is usedJ+1(k) A is replacedJ(k), expression formula is converted to
The high fdrequency component feature of unusual service condition signal is Dfp(i), objects of statistics is the height that front and back half period occurs in abnormal signal
The sum of frequency component absolute coefficient, is expressed as
2-3-2) extract wavelet energy entropy and wavelet singular entropy
Wavelet energy entropy and wavelet singular entropy are for indicating within the period that unusual service condition occurs, and signal energy is in different frequency range point
The confusion degree of cloth；
The Energy-Entropy WEE of small echo is defined as follows formula:
Wherein pi=Ei/ E is defined on the signal power spectrum on different scale i time k, Ei(k)=| Di(k)|2,For ruler
Spend the energy at i upper all moment with,It is approximately the gross energy of signal；
Coefficient D after wavelet transformation is reconstructedi(k) matrix D of (J+1) × M is constituted(J+1)×M, matrix is subjected to singular value point
Solution can obtain J+1 non-negative singular value σi, then wavelet singular entropy WAE is defined as follows:
3) the characteristic set extracted according to step 2) is subjected to tree-shaped cluster, more classification is established in the form of decision tree and are known
Other process；
4) classifier in more Classification and Identification processes is using three layers of ANN model building；
5) by training with after test, control errors are provided to online more Classification and Identification models within the scope of admission threshold
Recording system carries out the multi-state identification of power distribution network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710381487.6A CN107340456B (en) | 2017-05-25 | 2017-05-25 | Power distribution network operating condition intelligent identification Method based on multiple features analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710381487.6A CN107340456B (en) | 2017-05-25 | 2017-05-25 | Power distribution network operating condition intelligent identification Method based on multiple features analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107340456A CN107340456A (en) | 2017-11-10 |
| CN107340456B true CN107340456B (en) | 2019-12-03 |
Family
ID=60219939
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710381487.6A Active CN107340456B (en) | 2017-05-25 | 2017-05-25 | Power distribution network operating condition intelligent identification Method based on multiple features analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107340456B (en) |
Families Citing this family (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107909118B (en) * | 2017-12-11 | 2022-02-22 | 北京映翰通网络技术股份有限公司 | Power distribution network working condition wave recording classification method based on deep neural network |
| CN108154223B (en) * | 2017-12-22 | 2022-04-15 | 北京映翰通网络技术股份有限公司 | Power distribution network working condition wave recording classification method based on network topology and long time sequence information |
| CN108154175B (en) * | 2017-12-22 | 2022-04-15 | 北京映翰通网络技术股份有限公司 | Method for accurately identifying wave recording multiple working conditions of power distribution network |
| CN108279364B (en) * | 2018-01-30 | 2020-01-14 | 福州大学 | Power distribution network single-phase earth fault line selection method based on convolutional neural network |
| CN108508297B (en) * | 2018-04-04 | 2020-12-15 | 杭州家和物联技术有限公司 | Fault arc detection method based on mutation coefficient and SVM |
| CN109844749B (en) * | 2018-08-29 | 2023-06-20 | 区链通网络有限公司 | Node abnormality detection method and device based on graph algorithm and storage device |
| CN109270407B (en) * | 2018-11-16 | 2022-02-25 | 国网山东省电力公司电力科学研究院 | Extra-high voltage direct current transmission line fault reason identification method based on multi-source information fusion |
| CN109271975B (en) * | 2018-11-19 | 2020-08-21 | 燕山大学 | Power quality disturbance identification method based on big data multi-feature extraction collaborative classification |
| CN109738467B (en) * | 2018-12-21 | 2021-09-03 | 国网湖南省电力有限公司 | Component calculation method of zinc oxide resistance card |
| CN109782126B (en) * | 2018-12-27 | 2021-01-01 | 上海交通大学 | Power distribution network early fault detection method based on humanoid concept learning |
| GB2582676B (en) | 2018-12-27 | 2022-11-16 | Univ Shanghai Jiaotong | Early Failure Detection Method for Power Distribution Network Equipment Based On Human-Level Concept Learning |
| CN110110740B (en) * | 2019-03-26 | 2023-01-03 | 中国地质大学（武汉） | Drilling process working condition identification method based on multi-time scale features and neural network |
| CN111930545B (en) * | 2019-05-13 | 2023-11-03 | 中国移动通信集团湖北有限公司 | SQL script processing method, SQL script processing device and SQL script processing server |
| CN111999591B (en) * | 2019-05-27 | 2023-07-21 | 北京映翰通网络技术股份有限公司 | Method for identifying abnormal state of primary equipment of power distribution network |
| CN110308369A (en) * | 2019-07-08 | 2019-10-08 | 北京三清互联科技有限公司 | A kind of the power distribution network intelligence sensor and fault detection algorithm of Convergence gateway function |
| CN110543921A (en) * | 2019-10-14 | 2019-12-06 | 四川大学 | cable early fault identification method based on waveform learning |
| CN111612053B (en) * | 2020-05-14 | 2023-06-27 | 国网河北省电力有限公司电力科学研究院 | Calculation method for reasonable interval of line loss rate |
| CN112611944A (en) * | 2020-11-27 | 2021-04-06 | 武汉三相电力科技有限公司 | Method and system for monitoring and early warning abnormal discharge of direct-current transmission line on line |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6047114A (en) * | 1997-08-05 | 2000-04-04 | Institute For Information Industry | Method of constructing testing procedures for analog circuits by using fault classification tables |
| CN1847867A (en) * | 2006-03-24 | 2006-10-18 | 西南交通大学 | Post-wavelet analysis treating method and device for electric power transient signal |
| CN101545943A (en) * | 2009-05-05 | 2009-09-30 | 昆明理工大学 | Method for fault line selection of cable-wire mixed line of electric distribution network by using wavelet energy relative entropy |
| CN102654553A (en) * | 2012-05-07 | 2012-09-05 | 河南省电力公司南阳供电公司 | Method for line selection and positioning |
| CN103941162A (en) * | 2014-05-12 | 2014-07-23 | 福州大学 | Resonant earthed system fault line selection method utilizing waveform time domain feature clustering |
| CN105182106A (en) * | 2015-07-27 | 2015-12-23 | 国家电网公司 | Three-phase phase-controlled rectifier harmonic characteristic analysis method based on Park transformation |
| CN106017879A (en) * | 2016-05-18 | 2016-10-12 | 河北工业大学 | Universal circuit breaker mechanical fault diagnosis method based on feature fusion of vibration and sound signals |
| CN106405339A (en) * | 2016-11-11 | 2017-02-15 | 中国南方电网有限责任公司 | Power transmission line fault reason identification method based on high and low frequency wavelet feature association |
| CN106443394A (en) * | 2016-11-29 | 2017-02-22 | 西安理工大学 | Local discharge detection method for hydraulic turbine generator based on association probability wavelet transformation |
-
2017
- 2017-05-25 CN CN201710381487.6A patent/CN107340456B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6047114A (en) * | 1997-08-05 | 2000-04-04 | Institute For Information Industry | Method of constructing testing procedures for analog circuits by using fault classification tables |
| CN1847867A (en) * | 2006-03-24 | 2006-10-18 | 西南交通大学 | Post-wavelet analysis treating method and device for electric power transient signal |
| CN100520425C (en) * | 2006-03-24 | 2009-07-29 | 西南交通大学 | Post-wavelet analysis treating method and device for electric power transient signal |
| CN101545943A (en) * | 2009-05-05 | 2009-09-30 | 昆明理工大学 | Method for fault line selection of cable-wire mixed line of electric distribution network by using wavelet energy relative entropy |
| CN102654553A (en) * | 2012-05-07 | 2012-09-05 | 河南省电力公司南阳供电公司 | Method for line selection and positioning |
| CN103941162A (en) * | 2014-05-12 | 2014-07-23 | 福州大学 | Resonant earthed system fault line selection method utilizing waveform time domain feature clustering |
| CN105182106A (en) * | 2015-07-27 | 2015-12-23 | 国家电网公司 | Three-phase phase-controlled rectifier harmonic characteristic analysis method based on Park transformation |
| CN106017879A (en) * | 2016-05-18 | 2016-10-12 | 河北工业大学 | Universal circuit breaker mechanical fault diagnosis method based on feature fusion of vibration and sound signals |
| CN106405339A (en) * | 2016-11-11 | 2017-02-15 | 中国南方电网有限责任公司 | Power transmission line fault reason identification method based on high and low frequency wavelet feature association |
| CN106443394A (en) * | 2016-11-29 | 2017-02-22 | 西安理工大学 | Local discharge detection method for hydraulic turbine generator based on association probability wavelet transformation |
Non-Patent Citations (2)
| Title |
|---|
| 《A laboratory for time-frequency analysis of power quality disturbances using wavelets》;Matilde de Apraiz etc.;《Proceedings of 14th International Conference on Harmonics and Quality of Power - ICHQP 2010》;20101109;第1-5页 * |
| 《煤矿配电网电压暂降源辨识模型》;付华等;《电源学报》;20170223;第1-8页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107340456A (en) | 2017-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107340456B (en) | Power distribution network operating condition intelligent identification Method based on multiple features analysis | |
| CN107909118B (en) | Power distribution network working condition wave recording classification method based on deep neural network | |
| CN109444656A (en) | A kind of inline diagnosis method of deformation of transformer winding position | |
| CN109766952B (en) | Photovoltaic array fault detection method based on partial least square method and extreme learning machine | |
| CN112215722B (en) | Dominant instability mode discrimination model construction method and dominant instability mode discrimination method | |
| CN109633368A (en) | The method of duration power quality disturbances containing distributed power distribution network based on VMD and DFA | |
| CN112147462A (en) | Power transmission line fault identification method based on deep learning | |
| CN109034220A (en) | A kind of intelligent photovoltaic array method for diagnosing faults based on optimal rotation forest | |
| CN109284672A (en) | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method based on PSO-Kmeans algorithm | |
| CN107037327A (en) | Partial discharges fault judges feature extracting method and decision method | |
| CN109039280B (en) | Photovoltaic array fault diagnosis method based on non-principal component data characteristics | |
| CN113011084B (en) | Intelligent identification method for power transmission line fault reasons based on correlation vector machine | |
| CN108595884A (en) | Power system transient stability appraisal procedure and device | |
| CN110766313A (en) | Cable tunnel comprehensive state evaluation method based on operation and maintenance system | |
| CN110458240A (en) | A kind of three-phase bridge rectifier method for diagnosing faults, terminal device and storage medium | |
| CN111999591B (en) | Method for identifying abnormal state of primary equipment of power distribution network | |
| CN110610212A (en) | Fault classification method and fault classification device for transformer of power distribution network | |
| Chakraborty et al. | Random forest based fault classification technique for active power system networks | |
| CN108898182A (en) | A kind of MMC method for diagnosing faults based on core pivot element analysis and support vector machines | |
| Sima et al. | Edge–cloud collaboration detection approach for small-sample imbalanced faults in power lines | |
| CN111062569A (en) | Low-current fault discrimination method based on BP neural network | |
| CN114386024A (en) | Power intranet terminal equipment abnormal attack detection method based on ensemble learning | |
| CN108875796A (en) | Diagnosing failure of photovoltaic array method based on linear discriminant analysis and support vector machines | |
| CN111695543A (en) | Method for identifying hidden danger discharge type of power transmission line based on traveling wave time-frequency characteristics | |
| Aljohani et al. | Smart fault detection and classification for distribution grid hybridizing ST and MLP-NN |
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 | ||
| CB02 | Change of applicant information |
Address after: 100031 Xicheng District West Chang'an Avenue, No. 86, Beijing Applicant after: State Grid Co., Ltd. Applicant after: State Grid Hunan Electric Power Co., Ltd. Applicant after: Power Science Research Institute of Hunan Electric Power Co., Ltd. Applicant after: Chongqing University Address before: 100031 Xicheng District West Chang'an Avenue, No. 86, Beijing Applicant before: State Grid Corporation Applicant before: State Grid Hunan Electric Power Company Applicant before: Electric Power Research Institute, State Grid Hunan Electric Power Company Applicant before: Chongqing University |
|
| CB02 | Change of applicant information | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20191108 Address after: 100033 West Chang'an Avenue, Beijing, No. 86, No. Applicant after: State Grid Co., Ltd. Applicant after: State Grid Hunan Electric Power Co., Ltd. Applicant after: Power Science Research Institute of Hunan Electric Power Co., Ltd. Address before: 100031 Xicheng District West Chang'an Avenue, No. 86, Beijing Applicant before: State Grid Co., Ltd. Applicant before: State Grid Hunan Electric Power Co., Ltd. Applicant before: Power Science Research Institute of Hunan Electric Power Co., Ltd. Applicant before: Chongqing University |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |