CN110533265A - A kind of power distribution network lightning stroke method for early warning and power distribution network are struck by lightning prior-warning device - Google Patents
A kind of power distribution network lightning stroke method for early warning and power distribution network are struck by lightning prior-warning device Download PDFInfo
- Publication number
- CN110533265A CN110533265A CN201910892080.9A CN201910892080A CN110533265A CN 110533265 A CN110533265 A CN 110533265A CN 201910892080 A CN201910892080 A CN 201910892080A CN 110533265 A CN110533265 A CN 110533265A
- Authority
- CN
- China
- Prior art keywords
- data
- data matrix
- power distribution
- distribution network
- sample point
- 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
- 208000025274 Lightning injury Diseases 0.000 title claims abstract description 84
- 238000000034 method Methods 0.000 title claims abstract description 27
- 239000011159 matrix material Substances 0.000 claims abstract description 178
- 238000012706 support-vector machine Methods 0.000 claims abstract description 11
- 239000002689 soil Substances 0.000 claims description 9
- 238000001514 detection method Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 8
- 230000005540 biological transmission Effects 0.000 description 6
- 230000033001 locomotion Effects 0.000 description 6
- 239000013598 vector Substances 0.000 description 3
- 230000005611 electricity Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01W—METEOROLOGY
- G01W1/00—Meteorology
- G01W1/10—Devices for predicting weather conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Environmental & Geological Engineering (AREA)
- Software Systems (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Atmospheric Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Life Sciences & Earth Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Biodiversity & Conservation Biology (AREA)
- Ecology (AREA)
- Environmental Sciences (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Quality & Reliability (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application provides a kind of power distribution network lightning stroke method for early warning and power distribution network lightning stroke prior-warning device, this method comprises: obtaining historical data, and according to the historical data, establishes the first data matrix;First data matrix is about subtracted, the second data matrix is obtained;Second data matrix is trained using support vector machines, obtains power distribution network lightning stroke Early-warning Model;The second microclimate data are obtained in real time, and according to the second microclimate data obtained in real time, establish third data matrix;The third data matrix is input in the power distribution network lightning stroke Early-warning Model, lightning stroke warning information is obtained.In this way, having fully considered that the correlation between the multivariable for causing power distribution network to be struck by lightning, lightning stroke early warning effect are preferable since the power distribution network lightning stroke Early-warning Model established covers a variety of variables.
Description
Technical field
This application involves technical field of power systems more particularly to a kind of power distribution network lightning stroke method for early warning and power distribution network thunder
Hit prior-warning device.
Background technique
Power distribution network is run often by adverse weather conditions such as wind, rain, thunder and lightning or ice and snow.Wherein, thunder and lightning has become
Influence one of the principal element of power distribution network safe operation.How it is monitored and early warning becomes particularly significant.
Lightning stroke is strong electric discharge phenomena in nature, causes lightning stroke to occur many because being known as.It is first exactly meteorological condition,
Followed by environmental parameter locating for power transmission line corridor.For example, the landform in shaft tower location, landforms, periphery vegetation etc..Cause
There are also the associated body attributes of transmission line of electricity for the factor that thunder and lightning occurs.For example, shaft tower exhales title height and voltage class etc..
It in the related technology, is that transmission line lightning stroke early warning is realized in the movement tendency prediction based on thundercloud group, according only to thundercloud
Short-term thundercloud group motion profile prediction model is established in the movement of group, and when thundercloud group is close to power transmission line corridor, starting lightning stroke is in advance
It is alert.Such method is only analyzed using unitary variant, and lightning stroke early warning effect is poor.
Summary of the invention
This application provides a kind of power distribution network lightning stroke method for early warning and power distribution network lightning stroke prior-warning devices, to solve related skill
It in art, is only analyzed using unitary variant, the poor problem of lightning stroke early warning effect.
On the one hand, the application provides a kind of power distribution network lightning stroke method for early warning, comprising:
Historical data is obtained, and according to the historical data, establishes the first data matrix, wherein the historical data packet
Include the electric power line pole tower elevation data, the first microclimate data, shaft tower terrain data, relief data, shaft tower of the power distribution network
Periphery surface data, soil regime data and shaft tower, which are exhaled, claims high data;
First data matrix is about subtracted, the second data matrix is obtained;
Second data matrix is trained using support vector machines, obtains power distribution network lightning stroke Early-warning Model;
The second microclimate data are obtained in real time, and according to the second microclimate data obtained in real time, establish third data square
Battle array;
The third data matrix is input in the power distribution network lightning stroke Early-warning Model, lightning stroke warning information is obtained.
Optionally, before the step of about being subtracted described to first data matrix, obtaining the second data matrix, institute
State method further include:
First data matrix is normalized, the 4th data matrix is obtained;
It is described that first data matrix is about subtracted, obtain the second data matrix, comprising:
4th data matrix is about subtracted, second data matrix is obtained.
Optionally, described that 4th data matrix is about subtracted, obtain second data matrix, comprising:
Calculate the Euclidean distance in multiple sample points that the 4th data matrix is included between any two sample point;
According to the Euclidean distance between any two sample point being calculated, included in the 4th data matrix
The smallest K first sample point of Euclidean distance between sample point is searched in multiple sample points, wherein K is integer, and K >=2;
According to the K first sample point, the 4th data matrix is about subtracted, obtains the second data square
Battle array.
Optionally, described according to the K first sample point, the 4th data matrix is about subtracted, described in acquisition
Second data matrix, comprising:
It determines in multiple sample points that the 4th data matrix is included generic second in preceding K the second sample points
The destination number of sample point;
Judge whether the destination number is less than δ, wherein δ is integer, and 0 < δ≤K/2;
In the case where the destination number is less than δ, multiple sample points included in the 4th data matrix are deleted
In preceding K the second sample points.
Optionally, described according to the K first sample point, the 4th data matrix is about subtracted, described in acquisition
Second data matrix, comprising:
It detects in the K first sample point with the presence or absence of identical first sample point;
There are in the case where identical first sample point in the K first sample point, the 4th data square is deleted
The K first sample point in multiple sample points that battle array is included, obtains second data matrix.
On the other hand, the application also provides a kind of power distribution network lightning stroke prior-warning device, comprising:
First acquisition module establishes the first data matrix for obtaining historical data, and according to the historical data,
In, the historical data includes the electric power line pole tower elevation data of the power distribution network, the first microclimate data, shaft tower ground figurate number
According to, relief data, shaft tower periphery surface data, soil regime data and shaft tower exhale and claim high data;
About subtract module, for about being subtracted to first data matrix, obtains the second data matrix;
Training module obtains power distribution network lightning stroke for being trained using support vector machines to second data matrix
Early-warning Model;
Second obtains module, for obtaining the second microclimate data in real time, and according to the second microclimate number obtained in real time
According to establishing third data matrix;
Input module obtains thunder for the third data matrix to be input in the power distribution network lightning stroke Early-warning Model
Hit warning information.
Optionally, the power distribution network lightning stroke prior-warning device further include:
Normalized module obtains the 4th data matrix for first data matrix to be normalized;
The module that about subtracts is specifically used for about subtracting the 4th data matrix, obtains second data matrix.
Optionally, the module that about subtracts includes:
Computational submodule, for calculating any two sample point in multiple sample points that the 4th data matrix is included
Between Euclidean distance;
Submodule is searched, the Euclidean distance between any two sample point being calculated for basis, the described 4th
The smallest K first sample point of Euclidean distance between sample point is searched in multiple sample points that data matrix is included, wherein
K is integer, and K >=2;
About subtract submodule, for about being subtracted to the 4th data matrix according to the K first sample point, obtains
Second data matrix.
Optionally, the submodule that about subtracts includes:
Determination unit, for determining preceding K the second sample points in multiple sample points that the 4th data matrix is included
In generic the second sample point destination number;
Judging unit, for judging whether the destination number is less than δ, wherein δ is integer, and 0 < δ≤K/2;
First deletes unit, for deleting institute in the 4th data matrix in the case where the destination number is less than δ
The preceding K in multiple sample points the second sample points for including.
It is optionally, described about to subtract submodule further include:
Detection unit, for detecting in the K first sample point with the presence or absence of identical first sample point;
Second deletes unit, for, there are in the case where identical first sample point, deleting in the K first sample point
The K first sample point in the multiple sample points for being included except the 4th data matrix obtains the second data square
Battle array.
From the above technical scheme, the application provides a kind of power distribution network lightning stroke method for early warning and power distribution network lightning stroke early warning
Device, which comprises obtain historical data, and according to the historical data, establish the first data matrix, wherein described
Historical data includes the electric power line pole tower elevation data of the power distribution network, the first microclimate data, shaft tower terrain data, landforms
Data, shaft tower periphery surface data, soil regime data and shaft tower, which are exhaled, claims high data;First data matrix is carried out about
Subtract, obtains the second data matrix;Second data matrix is trained using support vector machines, it is pre- to obtain power distribution network lightning stroke
Alert model;The second microclimate data are obtained in real time, and according to the second microclimate data obtained in real time, establish third data square
Battle array;The third data matrix is input in the power distribution network lightning stroke Early-warning Model, lightning stroke warning information is obtained.In this way, can
To establish power distribution network lightning stroke Early-warning Model, the second microclimate data can be obtained in real time, and according to obtain in real time second micro- gas
Image data establishes third data matrix.And then third data matrix can be input in power distribution network lightning stroke Early-warning Model, it obtains
Lightning stroke warning information.Since the power distribution network lightning stroke Early-warning Model established covers a variety of variables, fully considers and caused distribution
The correlation between the multivariable of lightning stroke is netted, lightning stroke early warning effect is preferable.
Detailed description of the invention
In order to illustrate more clearly of the technical solution of the application, letter will be made to attached drawing needed in the embodiment below
Singly introduce, it should be apparent that, for those of ordinary skills, without creative efforts, also
Other drawings may be obtained according to these drawings without any creative labor.
Fig. 1 is a kind of flow chart of power distribution network lightning stroke method for early warning provided by the present application;
Fig. 2 is a kind of structure chart of power distribution network lightning stroke prior-warning device provided by the present application;
Fig. 3 is the structure chart of another power distribution network lightning stroke prior-warning device provided by the present application;
Fig. 4 is the structure chart of another power distribution network lightning stroke prior-warning device provided by the present application;
Fig. 5 is the structure chart of another power distribution network lightning stroke prior-warning device provided by the present application;
Fig. 6 is the structure chart of another power distribution network lightning stroke prior-warning device provided by the present application.
Specific embodiment
Embodiment will be illustrated in detail below, the example is illustrated in the accompanying drawings.In the following description when referring to the accompanying drawings,
Unless otherwise indicated, the same numbers in different drawings indicate the same or similar elements.Implement described in following embodiment
Mode does not represent all embodiments consistent with the application.It is only and be described in detail in claims, the application
The example of the consistent system and method for some aspects.
It is a kind of flow chart of power distribution network lightning stroke method for early warning provided by the present application referring to Fig. 1, Fig. 1.As shown in Figure 1, packet
Include following steps:
Step 101 obtains historical data, and according to the historical data, establishes the first data matrix, wherein described to go through
History data include the electric power line pole tower elevation data of the power distribution network, the first microclimate data, shaft tower terrain data, landforms number
According to, shaft tower periphery surface data, soil regime data and shaft tower exhale and claim high data.
In a step 101, available historical data, and according to historical data, establish the first data matrix X ∈ RD,
Middle D is data dimension.Wherein, historical data includes the electric power line pole tower elevation data of power distribution network, the first microclimate data, bar
Tower terrain data, relief data, shaft tower periphery surface data, soil regime data and shaft tower, which are exhaled, claims high data.Above-mentioned history number
Historical data accessed under normal condition is according to for power distribution network.
Step 102 about subtracts first data matrix, obtains the second data matrix.
In a step 102, the first data matrix can about be subtracted, obtains the second data matrix.
Optionally, before the step of about being subtracted described to first data matrix, obtaining the second data matrix, institute
State method further include:
First data matrix is normalized, the 4th data matrix is obtained;
It is described that first data matrix is about subtracted, obtain the second data matrix, comprising:
4th data matrix is about subtracted, second data matrix is obtained.
Further, first the first data matrix can be normalized, obtains the 4th data matrix.For example, can
With by the first data matrix by filling up by hand or the methods of mean value is filled up fills up missing data, and to the first data matrix into
The 4th data matrix is formed after row normalized.Normalized is that instigate the mean value of each process variable be zero, and variance is
1.4th data matrix can beN is the sample number of the 4th data matrix.
Next, can about be subtracted to the 4th data matrix, the second data matrix is obtained.It should be noted that can be with
The 4th data matrix is about subtracted using nearest neighbor algorithm (k-Nearest Neighbor, KNN), obtains the second data matrix.
Optionally, described that 4th data matrix is about subtracted, obtain second data matrix, comprising:
Calculate the Euclidean distance in multiple sample points that the 4th data matrix is included between any two sample point;
According to the Euclidean distance between any two sample point being calculated, included in the 4th data matrix
The smallest K first sample point of Euclidean distance between sample point is searched in multiple sample points, wherein K is integer, and K >=2;
According to the K first sample point, the 4th data matrix is about subtracted, obtains the second data square
Battle array.
Further, it can calculate in multiple sample points that the 4th data matrix is included between any two sample point
Euclidean distance.For example,WhereinThe feature vector of example, yi∈ γ=
{c1,c2,···,ckBe example classification.
The calculation formula of Euclidean distance between any two sample point such as formula (1):
Next, can be according to the Euclidean distance between any two sample point being calculated, in the 4th data matrix
The smallest K first sample point of Euclidean distance between sample point is searched in the multiple sample points for being included.Wherein, K is integer,
And K >=2.It is then possible to about be subtracted according to K first sample point to the 4th data matrix, the second data matrix is obtained.
Optionally, described according to the K first sample point, the 4th data matrix is about subtracted, described in acquisition
Second data matrix, comprising:
It determines in multiple sample points that the 4th data matrix is included generic second in preceding K the second sample points
The destination number of sample point;
Judge whether the destination number is less than δ, wherein δ is integer, and 0 < δ≤K/2;
In the case where the destination number is less than δ, multiple sample points included in the 4th data matrix are deleted
In preceding K the second sample points.
It can determine in multiple sample points that the 4th data matrix is included generic second in preceding K the second sample points
The destination number of sample point, and then may determine that whether the destination number is less than δ.Wherein, δ is integer, and 0 < δ≤K/2.Upper
Destination number is stated less than preceding K second in the case where δ, can deleting in multiple sample points included in the 4th data matrix
Sample point.
Optionally, described according to the K first sample point, the 4th data matrix is about subtracted, described in acquisition
Second data matrix, comprising:
It detects in the K first sample point with the presence or absence of identical first sample point;
There are in the case where identical first sample point in the K first sample point, the 4th data square is deleted
The K first sample point in multiple sample points that battle array is included, obtains second data matrix.
It can also detect in K first sample point with the presence or absence of identical first sample point.It is deposited in K first sample point
In the case where identical first sample point, K first in multiple sample points that the 4th data matrix is included can be deleted
Sample point obtains the second data matrix.
Step 103 is trained second data matrix using support vector machines, obtains power distribution network lightning stroke early warning mould
Type.
In step 103, it can use support vector machines (Support Vector Machine, SVM) to the second data
Matrix is trained, and obtains power distribution network lightning stroke Early-warning Model.
For example, the second data matrix is (x1,y1), i=1,2 ..., l, x ∈ Rn, y ∈ { 0,1 }, building optimal classification it is super flat
Face ω x+b=0.For ensure classifying face can two class data separate and make the distance between two class supporting vectors maximum,
It is asked to meet formula (2), can obtain class interval is 2/ | | ω | |.Then the problem of constructing optimal separating hyper plane is converted into following band
The minimum problems of constraint, and Lagrangian Lagrange function is introduced, such as formula (3):
yi[(ωgxi)+b] >=1, i=1,2, L, l (2)
Wherein ai> 0 is Lagrange coefficient.Constrained optimization problem determines by the saddle point condition of Lagrange function, and
And it is 0 that the solution of optimization problem, which meets at saddle point to the local derviation of ω and b,.
Dual problem (a=(a as shown in formula (4) is converted by the quadratic programming problem1,a2,…,al)):
By calculating, best initial weights vector sum is optimal to be biased respectively formula (6), formula (7):
WhereinOptimal separating hyper plane is ω*·x+b*=0, and optimal classification function such as formula (8) institute
Show:
Wherein, x ∈ Rn。
At this point, SVM training terminates, the parameter value for the Early-warning Model that power distribution network can be struck by lightning is saved.
Step 104 obtains the second microclimate data in real time, and according to the second microclimate data obtained in real time, establishes the
Three data matrixes.
At step 104, the second microclimate data can be obtained in real time, and according to the second microclimate number obtained in real time
According to establishing third data matrix.
The third data matrix is input in the power distribution network lightning stroke Early-warning Model by step 105, obtains lightning stroke early warning
Information.
In step 105, third data matrix can be input in power distribution network lightning stroke Early-warning Model, obtains lightning stroke early warning
Information.
It should be noted that being that the movement tendency prediction based on thundercloud group realizes that transmission line lightning stroke is pre- in the related technology
Alert, short-term thundercloud group motion profile prediction model is established in the movement according only to thundercloud group, when thundercloud is rolled into a ball close to transmission line of electricity
Starting lightning stroke early warning when corridor.Such method is only analyzed using unitary variant, and lightning stroke early warning effect is poor.
And in this application, it can establish power distribution network lightning stroke Early-warning Model, the second microclimate data can be obtained in real time, and
According to the second microclimate data obtained in real time, third data matrix is established.And then third data matrix can be input to and be matched
Power grid is struck by lightning in Early-warning Model, obtains lightning stroke warning information.By established power distribution network lightning stroke Early-warning Model cover it is a variety of
Variable has fully considered that the correlation between the multivariable for causing power distribution network to be struck by lightning, lightning stroke early warning effect are preferable.
A kind of power distribution network lightning stroke method for early warning provided by the present application.Historical data is obtained, and according to the historical data, is built
Vertical first data matrix, wherein the historical data includes the electric power line pole tower elevation data of the power distribution network, first micro- gas
Image data, shaft tower terrain data, relief data, shaft tower periphery surface data, soil regime data and shaft tower, which are exhaled, claims high data;It is right
First data matrix is about subtracted, and the second data matrix is obtained;Using support vector machines to second data matrix into
Row training obtains power distribution network lightning stroke Early-warning Model;The second microclimate data are obtained in real time, and according to obtain in real time second micro- gas
Image data establishes third data matrix;The third data matrix is input in the power distribution network lightning stroke Early-warning Model, is obtained
Lightning stroke warning information.In this way, can establish power distribution network lightning stroke Early-warning Model, the second microclimate data, and root can be obtained in real time
The the second microclimate data obtained when factually, establish third data matrix.And then third data matrix can be input to distribution
In net lightning stroke Early-warning Model, lightning stroke warning information is obtained.Since the power distribution network lightning stroke Early-warning Model established covers a variety of changes
Amount has fully considered that the correlation between the multivariable for causing power distribution network to be struck by lightning, lightning stroke early warning effect are preferable.
Referring to fig. 2, Fig. 2 is a kind of structure chart of power distribution network lightning stroke prior-warning device provided by the present application.As shown in Fig. 2, matching
Power grid lightning stroke prior-warning device 200 includes the first acquisition module 201, about subtracts module 202, the acquisition module of training module 203, second
204 and input module 205, in which:
First acquisition module 201 establishes the first data matrix for obtaining historical data, and according to the historical data,
Wherein, the historical data includes the electric power line pole tower elevation data of the power distribution network, the first microclimate data, shaft tower landform
Data, relief data, shaft tower periphery surface data, soil regime data and shaft tower, which are exhaled, claims high data;
About subtract module 202, for about being subtracted to first data matrix, obtains the second data matrix;
Training module 203 obtains power distribution network thunder for being trained using support vector machines to second data matrix
Hit Early-warning Model;
Second obtains module 204, for obtaining the second microclimate data in real time, and according to the second microclimate obtained in real time
Data establish third data matrix;
Input module 205 is obtained for the third data matrix to be input in the power distribution network lightning stroke Early-warning Model
Lightning stroke warning information.
Optionally, the prior-warning device as shown in figure 3, power distribution network is struck by lightning further include:
Normalized module 206 obtains the 4th data square for first data matrix to be normalized
Battle array;
The module 202 that about subtracts is specifically used for about subtracting the 4th data matrix, obtains the second data square
Battle array.
Optionally, as shown in figure 4, the module 202 that about subtracts includes:
Computational submodule 2021, for calculating any two sample in multiple sample points that the 4th data matrix is included
Euclidean distance between this point;
Submodule 2022 is searched, the Euclidean distance between any two sample point being calculated for basis, described
The smallest K first sample point of Euclidean distance between sample point is searched in multiple sample points that 4th data matrix is included,
Wherein, K is integer, and K >=2;
About subtract submodule 2023, for about being subtracted to the 4th data matrix according to the K first sample point,
Obtain second data matrix.
Optionally, as shown in figure 5, the submodule 2023 that about subtracts includes:
Determination unit 20231, for determining preceding K the second samples in multiple sample points that the 4th data matrix is included
The destination number of the second generic sample point in this point;
Judging unit 20232, for judging whether the destination number is less than δ, wherein δ is integer, and 0 < δ≤K/2;
First deletes unit 20233, for deleting the 4th data square in the case where the destination number is less than δ
Preceding K the second sample points in multiple sample points included in battle array.
Optionally, as shown in fig. 6, described about subtract submodule 2023 further include:
Detection unit 20234, for detecting in the K first sample point with the presence or absence of identical first sample point;
Second deletes unit 20235, for the case where there are identical first sample points in the K first sample point
Under, the K first sample point in multiple sample points that the 4th data matrix is included is deleted, second number is obtained
According to matrix.
Power distribution network lightning stroke prior-warning device 200 can be realized power distribution network lightning stroke prior-warning device in the embodiment of the method for Fig. 1 and realize
Each process, to avoid repeating, which is not described herein again.And power distribution network lightning stroke prior-warning device 200 may be implemented to establish power distribution network
It is struck by lightning Early-warning Model, can obtain the second microclimate data in real time, and according to the second microclimate data obtained in real time, establish the
Three data matrixes.And then third data matrix can be input in power distribution network lightning stroke Early-warning Model, obtain lightning stroke warning information.
Since the power distribution network lightning stroke Early-warning Model established covers a variety of variables, the multivariable for causing power distribution network to be struck by lightning has been fully considered
Between correlation, lightning stroke early warning effect it is preferable.
Similar portion cross-reference between embodiment provided by the present application, specific embodiment provided above is only
It is several examples under the total design of the application, does not constitute the restriction of the application protection scope.For those skilled in the art
For member, any other embodiment expanded without creative efforts according to application scheme all belongs to
In the protection scope of the application.
Claims (10)
- The method for early warning 1. a kind of power distribution network is struck by lightning characterized by comprisingHistorical data is obtained, and according to the historical data, establishes the first data matrix, wherein the historical data includes institute State electric power line pole tower elevation data, the first microclimate data, shaft tower terrain data, the relief data, shaft tower periphery of power distribution network Surface data, soil regime data and shaft tower, which are exhaled, claims high data;First data matrix is about subtracted, the second data matrix is obtained;Second data matrix is trained using support vector machines, obtains power distribution network lightning stroke Early-warning Model;The second microclimate data are obtained in real time, and according to the second microclimate data obtained in real time, establish third data matrix;The third data matrix is input in the power distribution network lightning stroke Early-warning Model, lightning stroke warning information is obtained.
- 2. the method as described in claim 1, which is characterized in that about subtracted described to first data matrix, obtained Before the step of second data matrix, the method also includes:First data matrix is normalized, the 4th data matrix is obtained;It is described that first data matrix is about subtracted, obtain the second data matrix, comprising:4th data matrix is about subtracted, second data matrix is obtained.
- 3. method according to claim 2, which is characterized in that it is described that 4th data matrix is about subtracted, obtain institute State the second data matrix, comprising:Calculate the Euclidean distance in multiple sample points that the 4th data matrix is included between any two sample point;According to the Euclidean distance between any two sample point being calculated, the 4th data matrix included it is multiple The smallest K first sample point of Euclidean distance between sample point is searched in sample point, wherein K is integer, and K >=2;According to the K first sample point, the 4th data matrix is about subtracted, obtains second data matrix.
- 4. method as claimed in claim 3, which is characterized in that it is described according to the K first sample point, to the 4th number About subtracted according to matrix, obtain second data matrix, comprising:Determine the second sample generic in preceding K the second sample points in multiple sample points that the 4th data matrix is included The destination number of point;Judge whether the destination number is less than δ, wherein δ is integer, and 0 < δ≤K/2;In the case where the destination number is less than δ, delete in multiple sample points included in the 4th data matrix Preceding K the second sample points.
- 5. method as claimed in claim 3, which is characterized in that it is described according to the K first sample point, to the 4th number About subtracted according to matrix, obtain second data matrix, comprising:It detects in the K first sample point with the presence or absence of identical first sample point;There are in the case where identical first sample point in the K first sample point, the 4th data matrix institute is deleted The K first sample point in the multiple sample points for including obtains second data matrix.
- The prior-warning device 6. a kind of power distribution network is struck by lightning characterized by comprisingFirst acquisition module establishes the first data matrix, wherein institute for obtaining historical data, and according to the historical data State historical data include the electric power line pole tower elevation data of the power distribution network, the first microclimate data, shaft tower terrain data, Looks data, shaft tower periphery surface data, soil regime data and shaft tower, which are exhaled, claims high data;About subtract module, for about being subtracted to first data matrix, obtains the second data matrix;Training module obtains power distribution network lightning stroke early warning for being trained using support vector machines to second data matrix Model;Second obtains module, for obtaining the second microclimate data in real time, and according to the second microclimate data obtained in real time, builds Vertical third data matrix;It is pre- to obtain lightning stroke for the third data matrix to be input in the power distribution network lightning stroke Early-warning Model for input module Alert information.
- The prior-warning device 7. power distribution network as claimed in claim 6 is struck by lightning, which is characterized in that the power distribution network lightning stroke prior-warning device is also Include:Normalized module obtains the 4th data matrix for first data matrix to be normalized;The module that about subtracts is specifically used for about subtracting the 4th data matrix, obtains second data matrix.
- The prior-warning device 8. power distribution network as claimed in claim 7 is struck by lightning, which is characterized in that the module that about subtracts includes:Computational submodule, for calculating in multiple sample points that the 4th data matrix is included between any two sample point Euclidean distance;Submodule is searched, the Euclidean distance between any two sample point being calculated for basis, in the 4th data The smallest K first sample point of Euclidean distance between sample point is searched in multiple sample points that matrix is included, wherein K is Integer, and K >=2;About subtract submodule, for about being subtracted to the 4th data matrix, described in acquisition according to the K first sample point Second data matrix.
- The prior-warning device 9. power distribution network as claimed in claim 8 is struck by lightning, which is characterized in that the submodule that about subtracts includes:Determination unit, it is same in preceding K the second sample points for determining in multiple sample points that the 4th data matrix is included The destination number of second sample point of classification;Judging unit, for judging whether the destination number is less than δ, wherein δ is integer, and 0 < δ≤K/2;First deletes unit, for deleting included in the 4th data matrix in the case where the destination number is less than δ Multiple sample points in preceding K the second sample points.
- The prior-warning device 10. power distribution network as claimed in claim 8 is struck by lightning, which is characterized in that described about to subtract submodule further include:Detection unit, for detecting in the K first sample point with the presence or absence of identical first sample point;Second deletes unit, for, there are in the case where identical first sample point, deleting institute in the K first sample point The K first sample point in multiple sample points that the 4th data matrix is included is stated, second data matrix is obtained.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910892080.9A CN110533265A (en) | 2019-09-20 | 2019-09-20 | A kind of power distribution network lightning stroke method for early warning and power distribution network are struck by lightning prior-warning device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910892080.9A CN110533265A (en) | 2019-09-20 | 2019-09-20 | A kind of power distribution network lightning stroke method for early warning and power distribution network are struck by lightning prior-warning device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110533265A true CN110533265A (en) | 2019-12-03 |
Family
ID=68669391
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910892080.9A Pending CN110533265A (en) | 2019-09-20 | 2019-09-20 | A kind of power distribution network lightning stroke method for early warning and power distribution network are struck by lightning prior-warning device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110533265A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111275193A (en) * | 2020-01-15 | 2020-06-12 | 杭州华网信息技术有限公司 | National power grid lightning stroke prediction method |
CN113295935A (en) * | 2021-06-01 | 2021-08-24 | 兰州资源环境职业技术学院 | Lightning stroke risk assessment method based on high-precision lightning positioning technology |
CN115099531A (en) * | 2022-08-19 | 2022-09-23 | 国网江苏省电力有限公司苏州供电分公司 | Power transmission line lightning stroke early warning method and system based on support vector machine |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104268576A (en) * | 2014-10-11 | 2015-01-07 | 国家电网公司 | Electric system transient stability classification method based on TNN-SVM |
CN106771847A (en) * | 2016-11-21 | 2017-05-31 | 国网福建省电力有限公司厦门供电公司 | A kind of 35kV power distribution networks transmission line lightning stroke Risk Forecast Method |
CN107085755A (en) * | 2017-05-15 | 2017-08-22 | 内蒙古电力(集团)有限责任公司 | A kind of photovoltaic plant short term power Forecasting Methodology and system |
CN107784015A (en) * | 2016-08-30 | 2018-03-09 | 中国电力科学研究院 | A kind of Data Reduction method based on the online historical data of power system |
CN107832875A (en) * | 2017-10-27 | 2018-03-23 | 云南电网有限责任公司电力科学研究院 | A kind of transmission line lightning stroke failure prediction method of improved adaptive GA-IAGA Support Vector Machines Optimized |
CN109543870A (en) * | 2018-05-28 | 2019-03-29 | 云南大学 | A kind of electric power line pole tower lightning stroke method for early warning keeping embedded mobile GIS based on neighborhood |
-
2019
- 2019-09-20 CN CN201910892080.9A patent/CN110533265A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104268576A (en) * | 2014-10-11 | 2015-01-07 | 国家电网公司 | Electric system transient stability classification method based on TNN-SVM |
CN107784015A (en) * | 2016-08-30 | 2018-03-09 | 中国电力科学研究院 | A kind of Data Reduction method based on the online historical data of power system |
CN106771847A (en) * | 2016-11-21 | 2017-05-31 | 国网福建省电力有限公司厦门供电公司 | A kind of 35kV power distribution networks transmission line lightning stroke Risk Forecast Method |
CN107085755A (en) * | 2017-05-15 | 2017-08-22 | 内蒙古电力(集团)有限责任公司 | A kind of photovoltaic plant short term power Forecasting Methodology and system |
CN107832875A (en) * | 2017-10-27 | 2018-03-23 | 云南电网有限责任公司电力科学研究院 | A kind of transmission line lightning stroke failure prediction method of improved adaptive GA-IAGA Support Vector Machines Optimized |
CN109543870A (en) * | 2018-05-28 | 2019-03-29 | 云南大学 | A kind of electric power line pole tower lightning stroke method for early warning keeping embedded mobile GIS based on neighborhood |
Non-Patent Citations (1)
Title |
---|
聂海福: "全景数据驱动的输电线路杆塔雷击预警模型研究及***设计", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111275193A (en) * | 2020-01-15 | 2020-06-12 | 杭州华网信息技术有限公司 | National power grid lightning stroke prediction method |
CN113295935A (en) * | 2021-06-01 | 2021-08-24 | 兰州资源环境职业技术学院 | Lightning stroke risk assessment method based on high-precision lightning positioning technology |
CN115099531A (en) * | 2022-08-19 | 2022-09-23 | 国网江苏省电力有限公司苏州供电分公司 | Power transmission line lightning stroke early warning method and system based on support vector machine |
CN115099531B (en) * | 2022-08-19 | 2023-01-31 | 国网江苏省电力有限公司苏州供电分公司 | Power transmission line lightning stroke early warning method and system based on support vector machine |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110533265A (en) | A kind of power distribution network lightning stroke method for early warning and power distribution network are struck by lightning prior-warning device | |
US10891481B2 (en) | Automated detection of features and/or parameters within an ocean environment using image data | |
CN110149237B (en) | Hadoop platform computing node load prediction method | |
US20170261646A1 (en) | Self-correcting multi-model numerical rainfall ensemble forecasting method | |
CN102072992B (en) | Detailed terrain data-based transmission line failure-shielding and lightning-protection performance evaluation method | |
Mohandes et al. | Wind speed extrapolation using machine learning methods and LiDAR measurements | |
CN104463700A (en) | Power transmission line tower lightning strike risk evaluation method based on data mining technology | |
CN105488528A (en) | Improved adaptive genetic algorithm based neural network image classification method | |
CN104200081B (en) | The forecasting procedure and system of landfall typhoon characterization factor based on historical data | |
Chou | Modelling watershed scale soil loss prediction and sediment yield estimation | |
CN108898247A (en) | A kind of power grid Rainfall Disaster Risk Forecast Method, system and storage medium | |
CN109901020A (en) | The fault recognition method for electric transmission line of sparse autocoder optimization neural network | |
CN108564263A (en) | One kind is for the disaster-stricken prediction technique of electric power line pole tower under squall line environment | |
CN104039008B (en) | A kind of hybrid locating method | |
CN115359366A (en) | Remote sensing image target detection method based on parameter optimization | |
CN103345552A (en) | Method and device for assessing reliability of power ICT communication network | |
Lu et al. | An integrated damage modeling and assessment framework for overhead power distribution systems considering tree-failure risks | |
CN109284877A (en) | Based on AIGA-WLSSVM Buried Pipeline rate prediction method | |
CN113313342A (en) | Method and system for analyzing power grid equipment fault probability caused by multiple natural disasters | |
CN109523776A (en) | Public transport operation plan generating means, method and system | |
CN116415495A (en) | Lightning trip-out rate acquisition method and equipment for complex terrain and meteorological environment | |
CN116091801A (en) | Rainfall image similarity searching method based on deep learning | |
CN113872524B (en) | Method and device for positioning low-efficiency group strings and computer storage medium | |
CN114611777A (en) | Method for predicting reservoir warehousing flow of flood season based on stacked long and short memory networks | |
CN115130661A (en) | Model parameter optimization method, and time sequence data anomaly detection method and device |
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: 20191203 |