CN109613402A - Detection method of high-resistance grounding fault in distribution network based on wavelet transform and neural network - Google Patents

Detection method of high-resistance grounding fault in distribution network based on wavelet transform and neural network Download PDF

Info

Publication number
CN109613402A
CN109613402A CN201910114060.9A CN201910114060A CN109613402A CN 109613402 A CN109613402 A CN 109613402A CN 201910114060 A CN201910114060 A CN 201910114060A CN 109613402 A CN109613402 A CN 109613402A
Authority
CN
China
Prior art keywords
neural network
neuron
distribution network
power distribution
detection method
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910114060.9A
Other languages
Chinese (zh)
Other versions
CN109613402B (en
Inventor
苏文聪
朱星宇
金涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Fuzhou University
Original Assignee
Fuzhou University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Fuzhou University filed Critical Fuzhou University
Priority to CN201910114060.9A priority Critical patent/CN109613402B/en
Publication of CN109613402A publication Critical patent/CN109613402A/en
Application granted granted Critical
Publication of CN109613402B publication Critical patent/CN109613402B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G01R31/08—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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
    • Y04S10/52—Outage or fault management, e.g. fault detection or location

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
  • Locating Faults (AREA)

Abstract

本发明涉及一种基于小波变换和神经网络的配电网高阻接地故障检测方法，当配电网发生高阻接地故障时，故障电流很小，且常常伴有电弧，普通的零序电流保护难以检测。本发明运用演化的神经网络对传统的检测方法进行改进。演化的神经网络是一种基于动态连接结构的智能系统，能够通过增量学习调整其拓扑结构以融入新信息。本发明利用离散小波变换处理故障信号，将故障信号输入演化的神经网络中，以此检测配电网的高阻接地故障。

The invention relates to a high-resistance grounding fault detection method for a distribution network based on wavelet transform and neural network. When a high-resistance grounding fault occurs in the distribution network, the fault current is very small and often accompanied by arcs, and ordinary zero-sequence current protection Difficult to detect. The present invention uses the evolved neural network to improve the traditional detection method. An evolved neural network is an intelligent system based on a dynamically connected structure, capable of adjusting its topology through incremental learning to incorporate new information. The invention uses discrete wavelet transform to process the fault signal, and inputs the fault signal into the evolved neural network, so as to detect the high-resistance grounding fault of the power distribution network.

Description

Power distribution network high resistance earthing fault detection method based on wavelet transformation and neural network
Technical field
The present invention relates to distribution net work earthing fault detections, and in particular to a kind of distribution based on wavelet transformation and neural network Net high resistance earthing fault detection method.
Background technique
High resistance earthing fault (HIF) refers to that power circuit is situated between by the way that road, soil, branch or cement works object etc. are conductive The ground fault that matter is occurred, it may occur however that in each voltage class, influence power distribution network normal operation.Due to radio frequency The high-impedance behavior of medium, when high resistance earthing fault occurs for power distribution network, fault current very little, and it is frequently accompanied by electric arc, commonly Zero-sequence current protection be difficult to detect.In the high resistance earthing fault of small current neutral grounding system especially resonant earthed system, arc Light ground fault accounts for very big a part.Due to the reason of free of air, the impedance ground variation of arc grounding very greatly, makes existing Protection starts repeatedly, restores, and may result in the protection overstepping of adjacent lines, equipment, can also cause total system overvoltage, And then electrical equipment is damaged, make fault spread, reduces the power supply reliability of power grid.Route operates with failure and may make for a long time Fault point temperature is excessively high, to cause fire, causes the permanent damages of electrical equipment, and step voltage is reachable around grounding point To several kilovolts, power system stability operation and personal safety are seriously threatened.It can be to connect that high impedance ground fault, which is effectively detected, The route selection and positioning got off provide criterion, so the high resistance earthing fault detection of power distribution network is particularly significant.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of power distribution network high resistants based on wavelet transformation and neural network to connect Earth fault detection method can effectively detect power distribution network high resistance earthing fault.
To achieve the above object, the present invention adopts the following technical scheme:
A kind of power distribution network high resistance earthing fault detection method based on wavelet transformation and neural network, comprising the following steps:
Step S1: power distribution network line current signal to be detected is acquired；
Step S2: using db4 small echo as morther wavelet, wavelet transform is carried out to line current signal, and reconstructed Waveform；
Step S3: the neural network parameter of evolution is set；
Step S4: using reconfiguration waveform as the input quantity of the neural network to develop, k-th of mind of input quantity and evolution layer is calculated The error of activation degree and output quantity and desired value through manhatton distance, k-th neuron between member；
Step S5: if the activation degree of k-th of neuron is less than preset threshold or output quantity and the error of desired value is more than One new neuron is placed on a position (k+1), otherwise updates weight by preset threshold；
Step S6: the manhatton distance between o-th of neuron and p-th of neuron is calculatedWith
Step S7: ifWithRespectively less than threshold value Dthr, then the two neurons are polymerize；
Step S8: repeating step S4-S7, until all input quantities are disposed, the neural network output developed As a result, and judging power distribution network to be detected with the presence or absence of high resistance earthing fault according to output.
Further, the wavelet transform sample frequency 10kHz, sampling time 0.5s.
Further, the neural network parameter of the evolution includes neuronal activation level thresholds Athr, error threshold Ethr, two neurons manhatton distance threshold value Dthr, learning rate α1And α2。
Further, the step S4 parameter calculates specifically:
J-th of input quantity IjManhatton distance between k-th of neuron of evolution layer is
The activation degree A of k-th of neuronk=1-Djk, WikFor weight；
Output quantity O1With desired valueError
Further, the step S5 specifically: if AkLess than AthrOr E1More than Ethr, a new neuron is placed In a position (k+1), weight W is otherwise updatedik(t+1)=Wik(t)+α1(Ii(t+1)-Wik(t)) and Wk1(t+1)=Wk1(t)+α2(AkE1)。
Further, describedWithSpecifically:
Wherein Wio、Wip、Wo1And Wp1For weight.
Compared with the prior art, the invention has the following beneficial effects:
1, the topological structure that the neural network that the present invention develops is not fixed, can drill online after carrying out data processing Turn to new structure.
2, the neural network that develops of the present invention can by incremental training Fast Learning, have good abstract ability with And network model is reassembled as to the ability of continually changing environment, wherein structure and parameter adapts to simultaneously.
3, the side that the neural network that the present invention develops passes through addition online and the connection weight for deleting neuron, adjustment neuron The generation that averting a calamity property of formula is forgotten, increases the detection reliability and efficiency of power distribution network high resistance earthing fault.
Detailed description of the invention
Fig. 1 is the method for the present invention flow chart；
Fig. 2 is the structural schematic diagram of the neural network to develop in the embodiment of the present invention.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings and embodiments.
Fig. 1 is please referred to, the present invention provides a kind of based on wavelet transformation and the inspection of the power distribution network high resistance earthing fault of neural network Survey method, comprising the following steps:
Step S1: power distribution network line current signal to be detected is acquired；
Step S2: using db4 small echo as morther wavelet, wavelet transform is carried out to line current signal, and reconstructed Waveform；
Step S3: the neural network parameter of evolution is set；
Step S4: using reconfiguration waveform as the input quantity of the neural network to develop, k-th of mind of input quantity and evolution layer is calculated The error of activation degree and output quantity and desired value through manhatton distance, k-th neuron between member；
Step S5: if the activation degree of k-th of neuron is less than preset threshold or output quantity and the error of desired value is more than One new neuron is placed on a position (k+1), otherwise updates weight by preset threshold；
Step S6: the manhatton distance between o-th of neuron and p-th of neuron is calculatedWith
Step S7: ifWithRespectively less than threshold value Dthr, then the two neurons are polymerize；
Step S8: repeating step S4-S7, until all input quantities are disposed, the neural network output developed As a result, and judging power distribution network to be detected with the presence or absence of high resistance earthing fault according to output.
In the present embodiment, the wavelet transform sample frequency 10kHz, sampling time 0.5s.
In the present embodiment, the neural network parameter of the evolution includes neuronal activation level thresholds Athr, error threshold Ethr, two neurons manhatton distance threshold value Dthr, learning rate α1And α2。
Further, the step S4 parameter calculates specifically:
J-th of input quantity IjManhatton distance between k-th of neuron of evolution layer is
The activation degree A of k-th of neuronk=1-Djk, WikFor weight；
Output quantity O1With desired valueError
In the present embodiment, the step S5 specifically: if AkLess than AthrOr E1More than Ethr, by a new neuron It is placed on a position (k+1), otherwise updates weight Wik(t+1)=Wik(t)+α1(Ii(t+1)-Wik(t)) and Wk1(t+1)=Wk1(t)+α2 (AkE1)。
In the present embodiment, describedWithSpecifically:
Wherein Wio、Wip、Wo1And Wp1For weight.
The foregoing is merely presently preferred embodiments of the present invention, all equivalent changes done according to scope of the present invention patent with Modification, is all covered by the present invention.

Claims (6)

1. a kind of power distribution network high resistance earthing fault detection method based on wavelet transformation and neural network, which is characterized in that including Following steps:
Step S1: power distribution network line current signal to be detected is acquired；
Step S2: using db4 small echo as morther wavelet, wavelet transform is carried out to line current signal, and obtain reconfiguration waveform；
Step S3: the neural network parameter of evolution is set；
Step S4: using reconfiguration waveform as the input quantity of the neural network to develop, input quantity and k-th of neuron of evolution layer are calculated Between manhatton distance, the activation degree of k-th neuron and the error of output quantity and desired value；
Step S5: if it is more than default that the activation degree of k-th of neuron, which is less than preset threshold or output quantity and the error of desired value, One new neuron is placed on a position (k+1), otherwise updates weight by threshold value；
Step S6: the manhatton distance between o-th of neuron and p-th of neuron is calculatedWith
Step S7: ifWithRespectively less than threshold value Dthr, then the two neurons are polymerize；
Step S8: repeating step S4-S7, until all input quantities are disposed, the neural network output developed as a result, And judge power distribution network to be detected with the presence or absence of high resistance earthing fault according to output.
2. the power distribution network high resistance earthing fault detection method based on wavelet transformation and neural network according to claim 1, feature It is: the wavelet transform sample frequency 10kHz, sampling time 0.5s.
3. the power distribution network high resistance earthing fault detection method according to claim 1 based on wavelet transformation and neural network, It is characterized by: the neural network parameter of the evolution includes neuronal activation level thresholds Athr, error threshold Ethr, two minds Manhatton distance threshold value D through memberthr, learning rate α1And α2。
4. the power distribution network high resistance earthing fault detection method according to claim 3 based on wavelet transformation and neural network, It is characterized by: the step S4 parameter calculates specifically:
J-th of input quantity IjManhatton distance between k-th of neuron of evolution layer is
The activation degree A of k-th of neuronk=1-Djk,WikFor weight；
Output quantity O1With desired valueError
5. the power distribution network high resistance earthing fault detection method according to claim 4 based on wavelet transformation and neural network, It is characterized by: the step S5 specifically: if AkLess than AthrOr E1More than Ethr, a new neuron is placed on (k+1) Otherwise a position updates weight Wik(t+1)=Wik(t)+α1(Ii(t+1)-Wik(t)) and Wk1(t+1)=Wk1(t)+α2(AkE1)。
6. the power distribution network high resistance earthing fault detection method according to claim 4 based on wavelet transformation and neural network, It is characterized by: describedWithSpecifically:
Wherein Wio、Wip、Wo1And Wp1For weight.
CN201910114060.9A 2019-02-14 2019-02-14 Detection method of high-resistance grounding fault in distribution network based on wavelet transform and neural network Expired - Fee Related CN109613402B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910114060.9A CN109613402B (en) 2019-02-14 2019-02-14 Detection method of high-resistance grounding fault in distribution network based on wavelet transform and neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910114060.9A CN109613402B (en) 2019-02-14 2019-02-14 Detection method of high-resistance grounding fault in distribution network based on wavelet transform and neural network

Publications (2)

Publication Number Publication Date
CN109613402A true CN109613402A (en) 2019-04-12
CN109613402B CN109613402B (en) 2020-11-27

Family

ID=66019693

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910114060.9A Expired - Fee Related CN109613402B (en) 2019-02-14 2019-02-14 Detection method of high-resistance grounding fault in distribution network based on wavelet transform and neural network

Country Status (1)

Country Link
CN (1) CN109613402B (en)

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113129161A (en) * 2019-12-31 2021-07-16 国网山东省电力公司威海供电公司 Wavelet decomposition and neural network-based animal fault analysis method in overhead power distribution system
CN113325269A (en) * 2021-05-28 2021-08-31 西安交通大学 Distribution network high-resistance fault monitoring method, system, equipment and storage medium
CN114089218A (en) * 2021-10-19 2022-02-25 广东电网有限责任公司东莞供电局 Power distribution network high-resistance grounding fault identification method, device, terminal and medium
CN115494350A (en) * 2022-11-21 2022-12-20 昆明理工大学 A lightning strike fault identification method and system for an AC transmission line
CN118091270A (en) * 2023-12-07 2024-05-28 国网湖北省电力有限公司宜昌供电公司 High impedance fault detection method for feeder terminal unit in distribution automation

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005091458A1 (en) * 2004-03-16 2005-09-29 Abb Technology Ag Digital signal processor implementation of high impedance fault algorithms
CN101975910A (en) * 2010-09-07 2011-02-16 昆明理工大学 Intelligent fault classification and location method for ultra-high voltage direct current transmission line
CN102611140A (en) * 2012-03-23 2012-07-25 合肥工业大学 Grid-connected inverter island detection method based on wavelet transform and neural network
TW201303319A (en) * 2011-07-06 2013-01-16 Univ Nat Taiwan Method and system for fault detection, identification and location in high-voltage power transmission networks
CN103257304A (en) * 2013-04-10 2013-08-21 昆明理工大学 ANN fault line selection method through CWT coefficient RMS in zero-sequence current feature band
CN103729687A (en) * 2013-12-18 2014-04-16 国网山西省电力公司晋中供电公司 Electricity price forecasting method based on wavelet transform and neural network
CN103728535A (en) * 2013-10-28 2014-04-16 昆明理工大学 Extra-high-voltage direct-current transmission line fault location method based on wavelet transformation transient state energy spectrum
CN103884966A (en) * 2014-04-15 2014-06-25 河海大学常州校区 Power distribution network low-current single-phase earth fault positioning method based on neural network
CN105759167A (en) * 2016-01-28 2016-07-13 江苏省电力公司南京供电公司 Wavelet neural network-based distribution network single-phase short circuit line selection method
CN108279364A (en) * 2018-01-30 2018-07-13 福州大学 Wire selection method for power distribution network single phase earthing failure based on convolutional neural networks

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005091458A1 (en) * 2004-03-16 2005-09-29 Abb Technology Ag Digital signal processor implementation of high impedance fault algorithms
CN101975910A (en) * 2010-09-07 2011-02-16 昆明理工大学 Intelligent fault classification and location method for ultra-high voltage direct current transmission line
TW201303319A (en) * 2011-07-06 2013-01-16 Univ Nat Taiwan Method and system for fault detection, identification and location in high-voltage power transmission networks
CN102611140A (en) * 2012-03-23 2012-07-25 合肥工业大学 Grid-connected inverter island detection method based on wavelet transform and neural network
CN103257304A (en) * 2013-04-10 2013-08-21 昆明理工大学 ANN fault line selection method through CWT coefficient RMS in zero-sequence current feature band
CN103728535A (en) * 2013-10-28 2014-04-16 昆明理工大学 Extra-high-voltage direct-current transmission line fault location method based on wavelet transformation transient state energy spectrum
CN103729687A (en) * 2013-12-18 2014-04-16 国网山西省电力公司晋中供电公司 Electricity price forecasting method based on wavelet transform and neural network
CN103884966A (en) * 2014-04-15 2014-06-25 河海大学常州校区 Power distribution network low-current single-phase earth fault positioning method based on neural network
CN105759167A (en) * 2016-01-28 2016-07-13 江苏省电力公司南京供电公司 Wavelet neural network-based distribution network single-phase short circuit line selection method
CN108279364A (en) * 2018-01-30 2018-07-13 福州大学 Wire selection method for power distribution network single phase earthing failure based on convolutional neural networks

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
VASILIOS P.ANDROVITSANEASA等: "Wavelet neural network methodology for ground resistance forecasting", 《ELECTRIC POWER SYSTEMS RESEARCH》 *
刘冻: "基于小波神经网络的输电线路故障类型识别", 《科技信息》 *
刘炳南等: "基于KNN的配电网高阻接地故障识别", 《电气技术》 *
叶倩: "配电网高阻故障检测方法研究", 《大科技》 *

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113129161A (en) * 2019-12-31 2021-07-16 国网山东省电力公司威海供电公司 Wavelet decomposition and neural network-based animal fault analysis method in overhead power distribution system
CN113129161B (en) * 2019-12-31 2025-11-28 国网山东省电力公司威海供电公司 Animal fault analysis method in overhead power distribution system based on wavelet decomposition and neural network
CN113325269A (en) * 2021-05-28 2021-08-31 西安交通大学 Distribution network high-resistance fault monitoring method, system, equipment and storage medium
CN114089218A (en) * 2021-10-19 2022-02-25 广东电网有限责任公司东莞供电局 Power distribution network high-resistance grounding fault identification method, device, terminal and medium
CN115494350A (en) * 2022-11-21 2022-12-20 昆明理工大学 A lightning strike fault identification method and system for an AC transmission line
CN115494350B (en) * 2022-11-21 2023-03-24 昆明理工大学 Alternating current transmission line lightning stroke fault recognition method and system
CN118091270A (en) * 2023-12-07 2024-05-28 国网湖北省电力有限公司宜昌供电公司 High impedance fault detection method for feeder terminal unit in distribution automation

Also Published As

Publication number Publication date
CN109613402B (en) 2020-11-27

Similar Documents

Publication Publication Date Title
CN109613402A (en) Detection method of high-resistance grounding fault in distribution network based on wavelet transform and neural network
CN110082640B (en) Distribution network single-phase earth fault identification method based on long-time memory network
CN108120903B (en) A kind of low-current single-phase earth fault line selection method based on pulse nerve membranous system
Hubana et al. Approach for identification and classification of HIFs in medium voltage distribution networks
CN106990324B (en) Power distribution network ground fault detection and positioning method
CN104730423A (en) Island effect detecting method of grid-connected photovoltaic power system
CN103941156A (en) Multi-message fusion section locating method based on extreme learning machine
Li et al. Physics-informed learning for high impedance faults detection
CN113962097A (en) Transformer substation lightning overvoltage simulation analysis method, device, terminal and medium
CN106970302B (en) Distribution network high-resistance fault location and simulation method based on integrated empirical model decomposition
CN111308263A (en) High-resistance grounding fault detection method for power distribution network
CN112710923B (en) Data-driven single-phase earth fault line selection method based on post-fault steady-state information
CN102879704B (en) Multiparameter data fusion detection method for high resistance ground faults of high-voltage transmission line
Hasabe et al. Detection and classification of faults on 220 KV transmission line using wavelet transform and neural network
CN112557808A (en) Single-phase earth fault positioning method for power distribution network
CN112130027A (en) Active distribution network fault location method based on DC dynamic time bending distance
CN113852044A (en) Current separation leakage protection method with double abrupt change quantities
Dodge et al. A deep learning based lightning location system
CN114910744B (en) High-resistance ground fault detection method based on S transformation and self-adaptive average singular entropy
Geethanjali Combined wavelet transfoms and neural network (WNN) based fault detection and classification in transmission lines
CN119830079A (en) Single-phase earth fault line selection method for low-voltage distribution network based on deep migration learning
Cheong et al. A novel fault location technique based on current signals only for thyristor controlled series compensated transmission lines using wavelet analysis and self organising map neural networks
García et al. Detection of high impedance faults in overhead multi grounded networks
CN113804978B (en) Lightning stroke waveform identification method based on fault indicator
Asbery et al. Electric transmission system fault identification using modular artificial neural networks for single transmission lines

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
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20201127

Termination date: 20220214

CF01 Termination of patent right due to non-payment of annual fee