CN109492857A - A kind of distribution network failure risk class prediction technique and device - Google Patents
A kind of distribution network failure risk class prediction technique and device Download PDFInfo
- Publication number
- CN109492857A CN109492857A CN201811088010.XA CN201811088010A CN109492857A CN 109492857 A CN109492857 A CN 109492857A CN 201811088010 A CN201811088010 A CN 201811088010A CN 109492857 A CN109492857 A CN 109492857A
- Authority
- CN
- China
- Prior art keywords
- distribution network
- factor
- power distribution
- risk
- data
- 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
- 238000009826 distribution Methods 0.000 title claims abstract description 238
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000000977 initiatory effect Effects 0.000 claims abstract description 41
- 238000012706 support-vector machine Methods 0.000 claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 32
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 238000004422 calculation algorithm Methods 0.000 claims description 26
- 239000011159 matrix material Substances 0.000 claims description 25
- 238000004140 cleaning Methods 0.000 claims description 11
- 230000008569 process Effects 0.000 claims description 11
- 230000002068 genetic effect Effects 0.000 claims description 8
- 238000003064 k means clustering Methods 0.000 claims description 8
- 238000005457 optimization Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000013480 data collection Methods 0.000 claims description 4
- 238000010606 normalization Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 238000012545 processing Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000013459 approach Methods 0.000 description 3
- 238000012502 risk assessment Methods 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000003860 storage Methods 0.000 description 2
- 238000004804 winding Methods 0.000 description 2
- 206010063659 Aversion Diseases 0.000 description 1
- 241001269238 Data Species 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000004141 dimensional analysis Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000002360 explosive Substances 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Economics (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Strategic Management (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Marketing (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Biomedical Technology (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Genetics & Genomics (AREA)
- Educational Administration (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Physiology (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
Abstract
The present invention provides a kind of distribution network failure risk class prediction technique and devices to determine power distribution network risk major influence factors data set first according to the data of the power distribution network initial effects factor of acquisition；Then according to power distribution network risk major influence factors data set, the optimal influence factor data set of power distribution network risk is determined using gray relative analysis method；Finally use the optimal influence factor data set Training Support Vector Machines of power distribution network risk, according to the SVM prediction distribution network failure risk class after training, present invention determine that using gray relative analysis method when the optimal influence factor of power distribution network risk, the relevance between power distribution network initial effects factor data is considered, distribution network failure risk profile accuracy and forecasting efficiency are substantially increased.
Description
Technical field
The present invention relates to distribution network technology fields, and in particular to a kind of distribution network failure risk class prediction technique and dress
It sets.
Background technique
The safe operation of power distribution network is the important ring of entire electric power netting safe running, is that the current power supply system that improves runs water
Flat key link.Statistics show about 80% power outage be as caused by distribution system failure, power distribution network fortune
Row risk be to fault signature attribute, fault severity level, break down when environmental factor comprehensive statement.Therefore, to distribution
Failure risk grade present in net operational process is effectively predicted, weak link is found out, and takes risk prevention system measure in time,
As current urgent problem to be solved.
Distribution network topology is complicated, and device category is numerous, is distributed relative distribution, and equipment running status is easy by outside
The influence of factor, and the randomness that failure occurs is strong, therefore, it is difficult to realize Risk-warning by analysis mechanism.Traditional power distribution network
Risk assessment predicted impact factor source is single, and data volume is few, and cannot reflect comprehensively from various dimensions such as time, spaces leads to distribution
Net failure influence factor that may be present.Existing power distribution network risk assessment early warning is mainly partial to establish grinding for power failure risk indicator
Study carefully, it is less to the association Journal of Sex Research between the failure risk influence factor selection method and influence factor of field of distribution network, and
Existing feature selection approach can only often select the big influence factor of weight, cannot effectively reject redundancy factor.Based on probability
The power distribution network risk assessment prediction technique of statistics, although can reflect that power grid overall risk is horizontal, obtained prediction result cannot
It for operations staff's discovery, finds out risk reason and risk source provides foundation, has one to the formulation of risk-aversion control measure
Fixed limitation.
With constantly improve for the systems such as power information acquisition, power distribution automation, power distribution network basic data, operation data and
There is explosive growth in management data, and the big datas characteristics such as magnanimity, logic complexity and data item complexity, the prior art is gradually presented
The prediction of middle distribution network failure risk class is generally basede on prediction algorithm (including the neural network, Kalman of data mining technology
Filter method, time series method and wavelet analysis etc.) it realizes, although having been widely used for power grid research and engineering in practice,
It is that prediction lacks the analysis of power distribution network venture influence correlate in the process, causes the accuracy of prediction low.
Summary of the invention
In order to overcome the above-mentioned power distribution network venture influence correlate that lacks in the prior art to analyze caused power distribution network event
Hinder the low deficiency of risk profile accuracy, the present invention provides a kind of distribution network failure risk class prediction technique and device, first root
According to the data of the power distribution network initial effects factor of acquisition, power distribution network risk major influence factors data set is determined；Then basis is matched
Power grid risk major influence factors data set determines the optimal influence factor data of power distribution network risk using gray relative analysis method
Collection；The optimal influence factor data set Training Support Vector Machines of power distribution network risk are finally used, according to the support vector machines after training
Distribution network failure risk class is predicted, present invention determine that using grey correlation analysis when the optimal influence factor of power distribution network risk
Method, it is contemplated that the relevance between power distribution network initial effects factor data improves distribution network failure risk profile accuracy.
In order to achieve the above-mentioned object of the invention, the present invention adopts the following technical scheme that:
On the one hand, the present invention provides a kind of distribution network failure risk class prediction technique, comprising:
According to the data of the power distribution network initial effects factor of acquisition, power distribution network risk major influence factors data set is determined；
According to power distribution network risk major influence factors data set, determine that power distribution network risk is optimal using gray relative analysis method
Influence factor data set；
Using the optimal influence factor data set Training Support Vector Machines of power distribution network risk, according to the support vector machines after training
Predict distribution network failure risk class.
Power distribution network initial effects factor include natural cause, external force factor, apparatus factor, system factor, operation factors and
Time factor.
The data of the power distribution network initial effects factor are obtained by following manner；
The external force factor, apparatus factor, system factor, operation factors and time factor data from power distribution network message tube
Distribution automation system and/or intelligent common monitoring system of distribution transformer in reason system obtain；
The data of the natural cause are obtained from the power grid Meteorological Information System in distribution network information management system.
According to the power distribution network initial effects factor data of acquisition, power distribution network risk major influence factors data set is determined, wrap
It includes:
The data of the power distribution network initial effects factor of acquisition are pre-processed；
According to pretreated influence factor data, power distribution network risk major influence factors are determined using ReliefF algorithm
Data set.
The data of the power distribution network initial effects factor of acquisition are pre-processed, comprising:
Power distribution network initial effects factor data is cleaned using k-means clustering algorithm, the distribution after being cleaned
Net influence factor data；
Power distribution network influence factor data after cleaning are normalized, pretreated influence factor number is obtained
According to.
The data set according to power distribution network risk major influence factors determines power distribution network wind using gray relative analysis method
The data set of the optimal influence factor in danger, comprising:
Successively select the data of any major influence factors in power distribution network risk major influence factors as reference sequences,
Remaining influence factor determines the matrix of differences of reference sequences and subsequence as subsequence；
Resolution ratio is set, and according to the matrix of differences of resolution ratio and reference sequences and subsequence determine it is main influence because
Plain degree of association matrix；
Correlation threshold is set, and the major influence factors in major influence factors degree of association matrix more than correlation threshold are picked
It removes, obtains the optimal influence factor data set of power distribution network risk.
The SVM prediction distribution network failure risk class according to after training, comprising:
The support vector machines after training is optimized using genetic algorithm, the support vector machines after being optimized；
According to fault outage frequency and loss of outage load accumulated value, distribution network failure risk class is divided into general wind
Danger, moderate risk and serious risk；
Distribution network failure risk class is predicted according to the support vector machines after optimization.
On the other hand, the present invention also provides a kind of distribution network failure risk class prediction meanss, comprising:
First determining module determines power distribution network risk master for the data according to the power distribution network initial effects factor of acquisition
Want influence factor data set；
Second determining module is used for according to power distribution network risk major influence factors data set, using gray relative analysis method
Determine the optimal influence factor data set of power distribution network risk；
Prediction module, for using the optimal influence factor data set Training Support Vector Machines of power distribution network risk, according to training
SVM prediction distribution network failure risk class afterwards.
The power distribution network initial effects factor include natural cause, external force factor, apparatus factor, system factor, operation because
Element and time factor.
First determining module is specifically used for:
From in distribution network information management system distribution automation system and/or intelligent common monitoring system of distribution transformer obtain
The external force factor, apparatus factor, system factor, operation factors and time factor data；
The data of the natural cause are obtained from the power grid Meteorological Information System in distribution network information management system.
First determining module includes:
Pretreatment unit pre-processes the data of the power distribution network initial effects factor of acquisition；
Determination unit, for determining power distribution network risk using ReliefF algorithm according to pretreated influence factor data
The data set of major influence factors.
The pretreatment unit includes:
Cleaning unit is obtained for being cleaned using k-means clustering algorithm to power distribution network initial effects factor data
Power distribution network influence factor data after cleaning；
Normalization unit is pre-processed for the power distribution network influence factor data after cleaning to be normalized
Influence factor data afterwards.
Second determining module is specifically used for:
Successively select the data of any major influence factors in power distribution network risk major influence factors as reference sequences,
Remaining influence factor determines the matrix of differences of reference sequences and subsequence as subsequence；
Resolution ratio is set, and according to the matrix of differences of resolution ratio and reference sequences and subsequence determine it is main influence because
Plain degree of association matrix；
Correlation threshold is set, and the major influence factors in major influence factors degree of association matrix more than correlation threshold are picked
It removes, obtains the optimal influence factor data set of power distribution network risk.
The prediction module includes:
Optimize unit, for being optimized using genetic algorithm to the support vector machines after training, the branch after being optimized
Hold vector machine；
Division unit is used for according to fault outage frequency and loss of outage load accumulated value, by distribution network failure risk etc.
Grade is divided into average risk, moderate risk and serious risk；
Predicting unit, for being predicted according to the support vector machines after optimization distribution network failure risk class.
Compared with the immediate prior art, technical solution provided by the invention is had the advantages that
In distribution network failure risk class prediction technique provided by the invention, first according to the power distribution network initial effects of acquisition because
The data of element, determine power distribution network risk major influence factors data set；Then according to power distribution network risk major influence factors data
Collection, determines the optimal influence factor data set of power distribution network risk using gray relative analysis method；It is finally optimal using power distribution network risk
Influence factor data set Training Support Vector Machines, according to the SVM prediction distribution network failure risk class after training, originally
It invents and uses gray relative analysis method when determining the optimal influence factor of power distribution network risk, it is contemplated that power distribution network initial effects factor
Relevance between data improves distribution network failure risk profile accuracy；
Distribution network failure risk class prediction meanss provided by the invention include the first determining module, the second determining module and
Prediction module, the first determining module determine power distribution network risk master for the data according to the power distribution network initial effects factor of acquisition
Want influence factor data set；Second determining module, for being closed using grey according to power distribution network risk major influence factors data set
Connection analytic approach determines the optimal influence factor data set of power distribution network risk；Prediction module, for using the optimal influence of power distribution network risk
Factor data collection Training Support Vector Machines, according to the SVM prediction distribution network failure risk class after training, determination is matched
Gray relative analysis method is used when the optimal influence factor data set of power grid risk, it is contemplated that power distribution network initial effects factor data
Between relevance, improve distribution network failure risk profile accuracy；
The present invention has comprehensively considered natural cause, apparatus factor, system factor and the time of distribution network failure risk generation
The power distribution networks risk initial effects factor such as factor, reflecting comprehensively from various dimensions leads to distribution network failure venture influence that may be present
Factor；
The present invention is classified more according to power distribution network risk initial effects factor and pretreated influence factor data, and using
ReliefF algorithm determines power distribution network venture influence principal element, improves distribution network failure risk class forecasting efficiency；
The present invention determines the incidence relation between major influence factors using gray relative analysis method, can reject redundancy shadow
The factor of sound, obtains optimal influence factor, provides basis for the accuracy of prediction result；
The present invention uses the optimal influence factor data Training Support Vector Machines of the optimal influence factor of power distribution network risk, utilizes
Support vector machines after genetic algorithm optimization training, and according to the SVM prediction power distribution network risk class after training, it mentions
The high classifying quality of support vector machines, to improve the accuracy of power distribution network risk class prediction；
The present invention is suitable for the distribution network failure risk class prediction under big data framework, provides for power distribution network operation and maintenance
Effective foundation.
Detailed description of the invention
Fig. 1 is distribution network failure risk class prediction technique flow chart in the embodiment of the present invention.
Specific embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings.
The embodiment of the present invention provides a kind of distribution network failure risk class prediction technique, and specific flow chart is as shown in Figure 1, tool
Body process is as follows:
S101: according to the data of the power distribution network initial effects factor of acquisition, power distribution network risk major influence factors number is determined
According to collection；
S102: according to power distribution network risk major influence factors data set, power distribution network wind is determined using gray relative analysis method
The optimal influence factor data set in danger；
S103: the optimal influence factor data set Training Support Vector Machines of power distribution network risk are used, according to the support after training
Vector machine predicts distribution network failure risk class.
Above-mentioned power distribution network initial effects factor include natural cause, external force factor, apparatus factor, system factor, operation because
Element and time factor.
Natural cause therein includes gas epidemic disaster, filth, thunder and lightning and icing；
External force factor therein include tree line contradiction, vehicle lance, it is mechanical accidentally hit, municipal construction and overhead line foreign matter it is short
Road；
Apparatus factor therein includes winding insulation dampness, insulation ag(e)ing, winding deformation and transformer oil；
System factor therein includes transformer overload, overload, ferro-resonance over-voltage, turn-to-turn short circuit, phase fault
And open circuit；
Operation factors therein include maloperation, distributed generation resource switching reconciliation Loop Closing Operation；
Time factor therein includes working day, festivals or holidays and guarantor's energized period.
Specific power distribution network risk initial effects factor is as shown in table 1:
Table 1
The data of above-mentioned power distribution network initial effects factor are obtained by following manner；
Wherein the data of external force factor, apparatus factor, system factor, operation factors and time factor are from power distribution network message tube
Distribution automation system and/or intelligent common monitoring system of distribution transformer in reason system obtain；
Wherein the data of natural cause are obtained from the power grid Meteorological Information System in distribution network information management system.
In above-mentioned S101, according to the power distribution network initial effects factor data of acquisition, determine power distribution network risk mainly and influence because
Plain data set, detailed process is as follows:
The data of the power distribution network initial effects factor of acquisition are pre-processed；
According to pretreated influence factor data, power distribution network risk major influence factors are determined using ReliefF algorithm
Data set.
The data of the power distribution network initial effects factor of acquisition are pre-processed, comprising:
Power distribution network initial effects factor data is cleaned using k-means clustering algorithm, the distribution after being cleaned
Net influence factor data；
Above-mentioned K-means clustering algorithm is a kind of simple, efficient clustering algorithm.It is the similitude handle according to data
Data set divides in groups, and similar to each other with the object in cluster, the object in different clusters is different.Use cluster CiCentroid (distribution
To the mean value of the object of the cluster) represent the cluster, cluster CiQuality can use (the cluster C that is deteriorated in clusteriIn all objects and centroid ciIt
Between error quadratic sum) measurement, be defined asdist(p,ci) indicate object p ∈ CiWith this
Cluster represents CiDifference.Assuming that the nearest center for arriving object p is cp, cpBe assigned to cpAverage distance l between objectcp, definition ratio
RateOutliers are judged according to ratio R and are rejected.
Power distribution network influence factor data after cleaning are normalized, pretreated influence factor number is obtained
According to.
Standardization and sliding-model control are carried out to sample data, data is made to fall into lesser common section.By venture influence
The original value of the numerical attribute (for example, temperature) of factor section label (for example, -10~0,0~10 etc.) or concepts tab (example
Such as, low, high) replacement.By the data for venture influence factor of standardizing, transforms the data into and normalize to identical section model
In enclosing.
The gray relative analysis method used in above-mentioned S102 be by analyze and determine influence degree between each factor or
It is a kind of analysis method that several sub- factors (subsequence) carry out the contribution degree of main factor (auxiliary sequence).It is with by certain
The curve of special parameter composition visually indicates the feature and development process of things, passes through certain particular difference value between calculated curve
The degree of association acquired analyzes its inner link, the degree of approximation as the development process for measuring these things.Thus according to matching
The data set of power grid risk major influence factors determines the number of the optimal influence factor of power distribution network risk using gray relative analysis method
It is specific as follows according to collection:
Successively select the data of any major influence factors in power distribution network risk major influence factors as reference sequences,
Remaining influence factor determines the matrix of differences of reference sequences and subsequence as subsequence；
It is arranged resolution ratio (0.5 can be set as), and true according to the matrix of differences of resolution ratio and reference sequences and subsequence
Determine major influence factors degree of association matrix；
Correlation threshold is set, and the major influence factors in major influence factors degree of association matrix more than correlation threshold are picked
It removes, obtains the optimal influence factor data set of power distribution network risk.
In above-mentioned S103, according to the SVM prediction distribution network failure risk class after training, comprising:
The support vector machines after training is optimized using genetic algorithm, the support vector machines after being optimized；
According to fault outage frequency and loss of outage load accumulated value, distribution network failure risk class is divided into general wind
Danger, moderate risk and serious risk；
Distribution network failure risk class is predicted according to the support vector machines after optimization.
Above-mentioned support vector machines (support vector machines, SVM) is substantially a kind of best supervised
Disaggregated model is practised, learning machine generalization ability is improved by seeking structuring least risk, realizes empiric risk and fiducial range
Minimum, obtain the purpose of good statistical law.Major advantage is: improving Generalization Capability, solves the problems, such as higher-dimension and non-linear
Problem avoids neural network structure selection and local minimum point's problem.Make power distribution network risk class using support vector cassification
Type can be predicted, and comprehensively consider the influence factor feature of power distribution network risk, the type that quantitative response failure risk occurs, for
Power grid risk pre-control provides effective foundation.
Distribution network failure risk class prediction technique provided in an embodiment of the present invention may be right from the summary of multiple dimensional analysis
The factor that distribution network failure risk impacts；Based on big data preconditioning technique, to the power distribution network venture influence data of acquisition
Data cleansing and normalized are carried out, and rejects outliers with K-means clustering algorithm；Utilize more classification ReliefF
Algorithm calculates the feature weight of failure risk influence factor, determines major influence factors, and carry out influence factor by weight sequencing；
It is rejected superfluous using the incidence relation between grey correlation analysis various risks influence factor by calculation risk degree of association matrix
Optimal influence factor is obtained after remaining influence factor；Consider that fault outage situation and severity divide risk class, using support to
Amount machine predicts power distribution network risk class, and optimizes processing to support vector machines using genetic algorithm, is desirably to obtain
The power distribution network risk class classifier met the requirements, is effectively predicted power distribution network risk class.
Embodiment 2
The embodiment of the present invention 2 provides a kind of distribution network failure risk class prediction meanss, specifically includes:
First determining module therein determines power distribution network for the data according to the power distribution network initial effects factor of acquisition
Risk major influence factors data set；
Second determining module therein is used for according to power distribution network risk major influence factors data set, using grey correlation
Analytic approach determines the optimal influence factor data set of power distribution network risk；
Prediction module therein, for using the optimal influence factor data set Training Support Vector Machines of power distribution network risk, root
According to the SVM prediction distribution network failure risk class after training.
Above-mentioned power distribution network initial effects factor include natural cause, external force factor, apparatus factor, system factor, operation because
Element and time factor.
Above-mentioned first determining module is specifically used for:
From in distribution network information management system distribution automation system and/or intelligent common monitoring system of distribution transformer obtain
The external force factor, apparatus factor, system factor, operation factors and time factor data；
The data of the natural cause are obtained from the power grid Meteorological Information System in distribution network information management system.
Above-mentioned first determining module includes:
Pretreatment unit pre-processes the data of the power distribution network initial effects factor of acquisition；
Determination unit, for determining power distribution network risk using ReliefF algorithm according to pretreated influence factor data
The data set of major influence factors.
Above-mentioned pretreatment unit includes:
Cleaning unit is obtained for being cleaned using k-means clustering algorithm to power distribution network initial effects factor data
Power distribution network influence factor data after cleaning；
Normalization unit is pre-processed for the power distribution network influence factor data after cleaning to be normalized
Influence factor data afterwards.
Above-mentioned second determining module is specifically used for:
Successively select the data of any major influence factors in power distribution network risk major influence factors as reference sequences,
Remaining influence factor determines the matrix of differences of reference sequences and subsequence as subsequence；
Resolution ratio is set, and according to the matrix of differences of resolution ratio and reference sequences and subsequence determine it is main influence because
Plain degree of association matrix；
Correlation threshold is set, and the major influence factors in major influence factors degree of association matrix more than correlation threshold are picked
It removes, obtains the optimal influence factor data set of power distribution network risk.
Above-mentioned prediction module includes:
Optimize unit, for being optimized using genetic algorithm to the support vector machines after training, the branch after being optimized
Hold vector machine；
Division unit is used for according to fault outage frequency and loss of outage load accumulated value, by distribution network failure risk etc.
Grade is divided into average risk, moderate risk and serious risk；
Predicting unit, for being predicted according to the support vector machines after optimization distribution network failure risk class.
For convenience of description, each section of apparatus described above is divided into various modules with function or unit describes respectively.
Certainly, each module or the function of unit can be realized in same or multiple softwares or hardware when implementing the application.
It should be understood by those skilled in the art that, embodiments herein can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the application
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the application, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The application is referring to method, the process of equipment (system) and computer program product according to the embodiment of the present application
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Finally it should be noted that: the above embodiments are merely illustrative of the technical scheme of the present invention and are not intended to be limiting thereof, institute
The those of ordinary skill in category field can still modify to a specific embodiment of the invention referring to above-described embodiment or
Equivalent replacement, these are applying for this pending hair without departing from any modification of spirit and scope of the invention or equivalent replacement
Within bright claims.
Claims (14)
1. a kind of distribution network failure risk class prediction technique characterized by comprising
According to the data of the power distribution network initial effects factor of acquisition, power distribution network risk major influence factors data set is determined；
According to power distribution network risk major influence factors data set, the optimal influence of power distribution network risk is determined using gray relative analysis method
Factor data collection；
Using the optimal influence factor data set Training Support Vector Machines of power distribution network risk, according to the SVM prediction after training
Distribution network failure risk class.
2. distribution network failure risk class prediction technique according to claim 1, which is characterized in that the power distribution network is initial
Influence factor includes natural cause, external force factor, apparatus factor, system factor, operation factors and time factor.
3. distribution network failure risk class prediction technique according to claim 2, which is characterized in that the power distribution network is initial
The data of influence factor are obtained by following manner；
The external force factor, apparatus factor, system factor, operation factors and time factor data from distribution network information management system
Distribution automation system and/or intelligent common monitoring system of distribution transformer in system obtain；
The data of the natural cause are obtained from the power grid Meteorological Information System in distribution network information management system.
4. distribution network failure risk class prediction technique according to claim 1, which is characterized in that described according to acquisition
Power distribution network initial effects factor data determines power distribution network risk major influence factors data set, comprising:
The data of the power distribution network initial effects factor of acquisition are pre-processed；
According to pretreated influence factor data, the number of power distribution network risk major influence factors is determined using ReliefF algorithm
According to collection.
5. distribution network failure risk class prediction technique according to claim 4, which is characterized in that described to match to acquisition
The data of power grid initial effects factor are pre-processed, comprising:
Power distribution network initial effects factor data is cleaned using k-means clustering algorithm, the power distribution network shadow after being cleaned
Ring factor data；
Power distribution network influence factor data after cleaning are normalized, pretreated influence factor data are obtained.
6. distribution network failure risk class prediction technique according to claim 1, which is characterized in that described according to power distribution network
The data set of risk major influence factors determines the data of the optimal influence factor of power distribution network risk using gray relative analysis method
Collection, comprising:
Successively select the data of any major influence factors in power distribution network risk major influence factors as reference sequences, remaining
Influence factor determines the matrix of differences of reference sequences and subsequence as subsequence；
Resolution ratio is set, and determines that major influence factors are closed according to the matrix of differences of resolution ratio and reference sequences and subsequence
Connection degree matrix；
Correlation threshold, and the major influence factors rejecting that in major influence factors degree of association matrix will be more than correlation threshold are set,
Obtain the optimal influence factor data set of power distribution network risk.
7. distribution network failure risk class prediction technique according to claim 1, which is characterized in that it is described according to training after
SVM prediction distribution network failure risk class, comprising:
The support vector machines after training is optimized using genetic algorithm, the support vector machines after being optimized；
According to fault outage frequency and loss of outage load accumulated value, by distribution network failure risk class be divided into average risk,
Moderate risk and serious risk；
Distribution network failure risk class is predicted according to the support vector machines after optimization.
8. a kind of distribution network failure risk class prediction meanss characterized by comprising
First determining module determines the main shadow of power distribution network risk for the data according to the power distribution network initial effects factor of acquisition
Ring factor data collection；
Second determining module, for being determined using gray relative analysis method according to power distribution network risk major influence factors data set
The optimal influence factor data set of power distribution network risk；
Prediction module, for using the optimal influence factor data set Training Support Vector Machines of power distribution network risk, after training
SVM prediction distribution network failure risk class.
9. distribution network failure risk class prediction meanss according to claim 8, which is characterized in that the power distribution network is initial
Influence factor includes natural cause, external force factor, apparatus factor, system factor, operation factors and time factor.
10. distribution network failure risk class prediction meanss according to claim 9, which is characterized in that described first determines
Module is specifically used for:
From in distribution network information management system distribution automation system and/or intelligent common monitoring system of distribution transformer obtain described in
External force factor, apparatus factor, system factor, operation factors and time factor data；
The data of the natural cause are obtained from the power grid Meteorological Information System in distribution network information management system.
11. distribution network failure risk class prediction meanss according to claim 8, which is characterized in that described first determines
Module includes:
Pretreatment unit pre-processes the data of the power distribution network initial effects factor of acquisition；
Determination unit, for determining that power distribution network risk is main using ReliefF algorithm according to pretreated influence factor data
The data set of influence factor.
12. distribution network failure risk class prediction meanss according to claim 11, which is characterized in that the pretreatment is single
Member includes:
Cleaning unit is cleaned for being cleaned using k-means clustering algorithm to power distribution network initial effects factor data
Power distribution network influence factor data afterwards；
Normalization unit obtains pretreated for the power distribution network influence factor data after cleaning to be normalized
Influence factor data.
13. distribution network failure risk class prediction meanss according to claim 8, which is characterized in that described second determines
Module is specifically used for:
Successively select the data of any major influence factors in power distribution network risk major influence factors as reference sequences, remaining
Influence factor determines the matrix of differences of reference sequences and subsequence as subsequence；
Resolution ratio is set, and determines that major influence factors are closed according to the matrix of differences of resolution ratio and reference sequences and subsequence
Connection degree matrix；
Correlation threshold, and the major influence factors rejecting that in major influence factors degree of association matrix will be more than correlation threshold are set,
Obtain the optimal influence factor data set of power distribution network risk.
14. distribution network failure risk class prediction meanss according to claim 8, which is characterized in that the prediction module
Include:
Optimize unit, for being optimized to the support vector machines after training using genetic algorithm, the support after being optimized to
Amount machine；
Division unit, for according to fault outage frequency and loss of outage load accumulated value, distribution network failure risk class to be drawn
It is divided into average risk, moderate risk and serious risk；
Predicting unit, for being predicted according to the support vector machines after optimization distribution network failure risk class.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811088010.XA CN109492857A (en) | 2018-09-18 | 2018-09-18 | A kind of distribution network failure risk class prediction technique and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811088010.XA CN109492857A (en) | 2018-09-18 | 2018-09-18 | A kind of distribution network failure risk class prediction technique and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109492857A true CN109492857A (en) | 2019-03-19 |
Family
ID=65690495
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811088010.XA Pending CN109492857A (en) | 2018-09-18 | 2018-09-18 | A kind of distribution network failure risk class prediction technique and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109492857A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222743A (en) * | 2019-05-23 | 2019-09-10 | 山西金晖隆开关有限公司 | The method for generating plan of power grid accident |
| CN110533213A (en) * | 2019-07-08 | 2019-12-03 | 广东工业大学 | Transmission line of electricity defect Risk Modeling and its prediction technique based on support vector machines |
| CN110929853A (en) * | 2019-12-11 | 2020-03-27 | 国网河南省电力公司洛阳供电公司 | Power distribution network line fault prediction method based on deep learning |
| CN111369168A (en) * | 2020-03-18 | 2020-07-03 | 武汉大学 | Associated feature selection method suitable for multiple regulation and control operation scenes of power grid |
| CN111597549A (en) * | 2020-04-17 | 2020-08-28 | 国网浙江省电力有限公司湖州供电公司 | Network security behavior identification method and system based on big data |
| CN111626360A (en) * | 2020-05-28 | 2020-09-04 | 新智数字科技有限公司 | Method, device, equipment and storage medium for detecting fault type of boiler |
| CN112488871A (en) * | 2020-10-23 | 2021-03-12 | 广西电网有限责任公司电力科学研究院 | Method and system for eliminating redundant data of original input features of power grid |
| CN112881827A (en) * | 2021-01-12 | 2021-06-01 | 新疆大学 | Oil-immersed transformer fault diagnosis method based on improved grey correlation analysis |
| CN113222356A (en) * | 2021-04-21 | 2021-08-06 | 长沙电力职业技术学院 | Safety risk assessment method and system for rural power grid emergency power supply equipment |
| CN113344448A (en) * | 2021-07-02 | 2021-09-03 | 广东电网有限责任公司 | Distribution network line risk level assessment method |
| CN113450014A (en) * | 2021-07-14 | 2021-09-28 | 上海海事大学 | Equipment operation dynamic risk classification method |
| CN114970774A (en) * | 2022-07-29 | 2022-08-30 | 国网经济技术研究院有限公司 | Intelligent transformer fault prediction method and device |
| CN115640880A (en) * | 2022-09-30 | 2023-01-24 | 海南电网有限责任公司 | Weak link early warning method based on support vector machine |
| CN117895661A (en) * | 2024-03-14 | 2024-04-16 | 国网山西省电力公司太原供电公司 | Power distribution network control method and system combined with risk analysis |
| CN117895661B (en) * | 2024-03-14 | 2024-05-24 | 国网山西省电力公司太原供电公司 | Power distribution network control method and system combined with risk analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345639A (en) * | 2013-07-01 | 2013-10-09 | 河海大学常州校区 | Support vector machine high-voltage circuit breaker fault diagnosis method based on fuzzy clustering |
| CN103810533A (en) * | 2013-08-15 | 2014-05-21 | 国家电网公司 | Cloud-model-based power distribution network fault risk identification method |
| CN107832870A (en) * | 2017-10-16 | 2018-03-23 | 国家电网公司 | Project of transmitting and converting electricity cost forecasting method based on grey correlation analysis and SVMs |
| CN107958354A (en) * | 2018-01-08 | 2018-04-24 | 国网能源研究院有限公司 | A kind of analysis method of power grid layer utilization rate of equipment and installations major influence factors |
| CN108009695A (en) * | 2016-10-29 | 2018-05-08 | 南京理工大学 | A kind of methods of risk assessment for taking into account index related active distribution network operation |
-
2018
- 2018-09-18 CN CN201811088010.XA patent/CN109492857A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103345639A (en) * | 2013-07-01 | 2013-10-09 | 河海大学常州校区 | Support vector machine high-voltage circuit breaker fault diagnosis method based on fuzzy clustering |
| CN103810533A (en) * | 2013-08-15 | 2014-05-21 | 国家电网公司 | Cloud-model-based power distribution network fault risk identification method |
| CN108009695A (en) * | 2016-10-29 | 2018-05-08 | 南京理工大学 | A kind of methods of risk assessment for taking into account index related active distribution network operation |
| CN107832870A (en) * | 2017-10-16 | 2018-03-23 | 国家电网公司 | Project of transmitting and converting electricity cost forecasting method based on grey correlation analysis and SVMs |
| CN107958354A (en) * | 2018-01-08 | 2018-04-24 | 国网能源研究院有限公司 | A kind of analysis method of power grid layer utilization rate of equipment and installations major influence factors |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222743A (en) * | 2019-05-23 | 2019-09-10 | 山西金晖隆开关有限公司 | The method for generating plan of power grid accident |
| CN110533213A (en) * | 2019-07-08 | 2019-12-03 | 广东工业大学 | Transmission line of electricity defect Risk Modeling and its prediction technique based on support vector machines |
| CN110929853A (en) * | 2019-12-11 | 2020-03-27 | 国网河南省电力公司洛阳供电公司 | Power distribution network line fault prediction method based on deep learning |
| CN111369168B (en) * | 2020-03-18 | 2022-07-05 | 武汉大学 | Associated feature selection method suitable for multiple regulation and control operation scenes of power grid |
| CN111369168A (en) * | 2020-03-18 | 2020-07-03 | 武汉大学 | Associated feature selection method suitable for multiple regulation and control operation scenes of power grid |
| CN111597549A (en) * | 2020-04-17 | 2020-08-28 | 国网浙江省电力有限公司湖州供电公司 | Network security behavior identification method and system based on big data |
| CN111626360A (en) * | 2020-05-28 | 2020-09-04 | 新智数字科技有限公司 | Method, device, equipment and storage medium for detecting fault type of boiler |
| CN111626360B (en) * | 2020-05-28 | 2023-11-17 | 新奥新智科技有限公司 | Method, apparatus, device and storage medium for detecting boiler fault type |
| CN112488871A (en) * | 2020-10-23 | 2021-03-12 | 广西电网有限责任公司电力科学研究院 | Method and system for eliminating redundant data of original input features of power grid |
| CN112881827A (en) * | 2021-01-12 | 2021-06-01 | 新疆大学 | Oil-immersed transformer fault diagnosis method based on improved grey correlation analysis |
| CN113222356A (en) * | 2021-04-21 | 2021-08-06 | 长沙电力职业技术学院 | Safety risk assessment method and system for rural power grid emergency power supply equipment |
| CN113344448A (en) * | 2021-07-02 | 2021-09-03 | 广东电网有限责任公司 | Distribution network line risk level assessment method |
| CN113450014A (en) * | 2021-07-14 | 2021-09-28 | 上海海事大学 | Equipment operation dynamic risk classification method |
| CN113450014B (en) * | 2021-07-14 | 2024-04-05 | 上海海事大学 | Equipment operation dynamic risk classification method |
| CN114970774A (en) * | 2022-07-29 | 2022-08-30 | 国网经济技术研究院有限公司 | Intelligent transformer fault prediction method and device |
| CN115640880A (en) * | 2022-09-30 | 2023-01-24 | 海南电网有限责任公司 | Weak link early warning method based on support vector machine |
| CN117895661A (en) * | 2024-03-14 | 2024-04-16 | 国网山西省电力公司太原供电公司 | Power distribution network control method and system combined with risk analysis |
| CN117895661B (en) * | 2024-03-14 | 2024-05-24 | 国网山西省电力公司太原供电公司 | Power distribution network control method and system combined with risk analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109492857A (en) | A kind of distribution network failure risk class prediction technique and device | |
| CN108375715A (en) | A kind of distribution network line fault risk day prediction technique and system | |
| CN108537394A (en) | A kind of intelligent grid actual time safety method for early warning and device | |
| CN110232499A (en) | A kind of power distribution network information physical side method for prewarning risk and system | |
| CN109872003A (en) | Obj State prediction technique, system, computer equipment and storage medium | |
| CN104200288A (en) | Equipment fault prediction method based on factor-event correlation recognition | |
| CN108964023B (en) | Short-term prediction method and system for bus voltage situation of power grid | |
| CN111967738B (en) | Risk early warning method, system and medium of power grid information energy fusion system | |
| CN108665181A (en) | A kind of appraisal procedure and device of distribution network reliability | |
| CN109344990A (en) | A kind of short-term load forecasting method and system based on DFS and SVM feature selecting | |
| Ünal et al. | Big data-driven detection of false data injection attacks in smart meters | |
| Hou et al. | Spatial distribution assessment of power outage under typhoon disasters | |
| de Assis et al. | Optimal allocation of remote controlled switches in radial distribution systems | |
| Amini et al. | Electrical energy systems resilience: A comprehensive review on definitions, challenges, enhancements and future proceedings | |
| Gao et al. | Real-time long-term voltage stability assessment based on eGBDT for large-scale power system with high renewables penetration | |
| Shuai et al. | Analysis and identification of power blackout-sensitive users by using big data in the energy system | |
| Hijazi et al. | Transfer learning for transient stability predictions in modern power systems under enduring topological changes | |
| WO2021130298A1 (en) | System, apparatus and method for managing energy consumption at a technical installation | |
| Stamatescu et al. | Deep and efficient impact models for edge characterization and control of energy events | |
| Daraghmi et al. | Accurate and time‐efficient negative binomial linear model for electric load forecasting in IoE | |
| Guan et al. | Enhancing security and resilience of bulk power systems via multisource big data learning | |
| CN113052250A (en) | Decision support method, system, device and medium based on meteorological disaster | |
| Ji et al. | Risk index early-warning of smart grid based on neural network | |
| Dongli et al. | Research on fault risk ranking and screening of distribution network based on uncertainty theory | |
| Ji et al. | Research on risk indicator system of smart distribution grid based on LVQ neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |