CN109522964A - Clustering method, device, adjusting device and the computer storage medium of virtual plant - Google Patents
Clustering method, device, adjusting device and the computer storage medium of virtual plant Download PDFInfo
- Publication number
- CN109522964A CN109522964A CN201811418832.XA CN201811418832A CN109522964A CN 109522964 A CN109522964 A CN 109522964A CN 201811418832 A CN201811418832 A CN 201811418832A CN 109522964 A CN109522964 A CN 109522964A
- Authority
- CN
- China
- Prior art keywords
- node
- leader
- leader node
- distance
- follows
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
Abstract
The embodiment of the invention discloses a kind of clustering method of virtual plant, device, adjusting device and computer readable storage mediums.It include: that leader node is determined by default cluster numerical value in each node of electric system, wherein using the node in all nodes in addition to leader node as following node；Being calculated separately using the first preset algorithm each described follows node to the first distance of each leader node；Optional one follows node, and node corresponding leader node of the shortest distance into the first distance of each leader node is followed described in selection；It follows node clustering to the leader node chosen for described, follows node clustering to finish until all.It realizes flexibility, operational efficiency and the response efficiency for improving virtual plant, to reduce the dependence to conventional Power Generation Mode, reduces the economic cost and Environmental costs for maintaining stability of power system.
Description
Technical field
The present invention relates to virtual plant resource cluster technical field more particularly to a kind of clustering methods of virtual plant, dress
It sets, adjusting device and computer readable storage medium.
Background technique
Virtual plant (Virtual Power Plant) is to gather Distributed-generation equipment, energy storage device and controllable burden
It is combined, is externally presented as that an entirety participates in one of electricity market and assisted hatching energy polymerized form.
With the development of smart grid, more stringent requirements are proposed for the response reliability of response speed and resource to flexibility resource,
But different types of its characteristic of flexibility resource is different.The development of power supply and load so that the following power distribution network show it is distributed small
The features such as capacity, resource category multiplicity, bi-directional current, interactivity, virtual plant technology effectively connect distributed resource and electricity
Force system realizes resource consolidation and distribution, takes into account power network safety operation and electricity market dual role.Wherein, to resource
Classify, targetedly service, has important meaning to the flexibility and operational efficiency response efficiency that improve virtual plant
Justice.The type for the flexibility resource that different virtual plants is integrated is different, quantity is different, this will affect the polymerization of virtual plant
Characteristic, to influence the ancillary service that virtual plant can participate in.
By the cluster to virtual plant built-in flexibility resource, the angle run from system can be improved virtual plant and be
Power grid provides the quality of service, is conducive to increase its income from flexibility resource individual angle.Electric power networks scale is different,
Distributed clustering method based on multi-agent network (Multi-agent system) is suitable for the subregion of any rank.Mesh
Before, the flexibility resource distribution formula clustering method research for considering topological structure of electric is less.
Summary of the invention
The embodiment of the present invention provides clustering method, device, adjusting device and the computer-readable storage of a kind of virtual plant
Medium realizes flexibility, operational efficiency and the response efficiency for improving virtual plant, thus reduce to conventional Power Generation Mode according to
Rely, reduces the economic cost and Environmental costs for maintaining stability of power system.
In a first aspect, the embodiment of the invention provides a kind of clustering methods of virtual plant, this method comprises:
Leader node is determined by default cluster numerical value in each node of electric system, wherein will remove in all nodes
Node outside leader node, which is used as, follows node；
Being calculated separately using the first preset algorithm each described follows node to the first distance of each leader node；
Optional one follows node, and node most short distance into the first distance of each leader node is followed described in selection
From corresponding leader node；
It follows node clustering to the leader node chosen for described, follows node clustering to finish until all.
Second aspect, the embodiment of the invention also provides a kind of clustering apparatus of virtual plant, which includes:
Leader node obtains module, for determining leader's section by default cluster numerical value in each node of electric system
Point, wherein using the node in all nodes in addition to leader node as following node；
Shortest distance determining module each described follows node to each institute for calculating separately using the first preset algorithm
State the first distance of leader node；
Leader node selecting module follows node for optional one, follows node to each leader described in selection
The corresponding leader node of the shortest distance in the first distance of node；
Cluster module follows node until all for following node clustering to the leader node chosen for described
Cluster finishes.
The third aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer
Program, the program realize the clustering method of virtual plant provided in an embodiment of the present invention when being executed by processor.
Fourth aspect processor and is stored in the embodiment of the invention also provides a kind of adjusting device, including memory
On reservoir and it can realize when the computer program of processor operation, the processor execute the computer program such as the present invention
The clustering method for the virtual plant that embodiment provides.
The embodiment of the present invention determines leader node by default cluster numerical value in each node of electric system, wherein will
Node in all nodes in addition to leader node calculates separately each described follow as following node, using the first preset algorithm
For node to the first distance of each leader node, optional one follows node, follows node to each described described in selection
The corresponding leader node of the shortest distance in the first distance of leader node follows node clustering to the leader chosen for described
Node follows node clustering to finish until all.Realize flexibility, operational efficiency and the response efficiency for improving virtual plant, from
And the dependence to conventional Power Generation Mode is reduced, reduce the economic cost and Environmental costs for maintaining stability of power system.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of the clustering method of virtual plant provided in an embodiment of the present invention；
Fig. 2 is the flow diagram of the clustering method of another virtual plant provided in an embodiment of the present invention；
Fig. 3 is the flow diagram that leader node provided in an embodiment of the present invention determines method；
Fig. 4 is a kind of structural schematic diagram of the clustering apparatus of virtual plant provided in an embodiment of the present invention；
Fig. 5 is a kind of structural schematic diagram of adjusting device provided in an embodiment of the present invention.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Fig. 1 is a kind of flow diagram of the clustering method of virtual plant provided in an embodiment of the present invention, and this method can be with
It is executed by adjusting device provided in an embodiment of the present invention, which can be used software and/or hardware mode and realize, the party
Method specifically comprises the following steps:
Step 110 by default cluster numerical value determines leader node in each node of electric system, wherein will own
Node in node in addition to leader node, which is used as, follows node；
The present invention is based on the consistency algorithms of the discrete amount of offsetting, and are realized in the power system based on flexibility resource
The electrical characteristic of virtual plant clusters.When there are widely distributed, when the biggish virtual plant of quantity, the present invention can in electric power networks
The category division that flexibility resource is carried out according to flexibility resource composition and Integrated Trait is realized in whole network in a distributed manner,
To provide basis for its functional distribution, classifies to flexibility resource, be allowed to pointedly provide auxiliary for electric system
Service.Each node need to only acquire local information and neighbor information in power grid, compared to centralized clustering algorithm, reduce to letter
The requirement of breath amount, while also protecting the privacy of user.Present invention employs distributed methods, it is only necessary to local power information
The dispatch command for realizing global optimization is made with operating parameter.
Before resource cluster, it is necessary first to determine leader's section according to default cluster numerical value in each node of electric system
Point, remaining node are used as and follow node.In the present embodiment, the method for determining leader node can be divided into two kinds, and one is random
Optionally, one is be calculated by distribution K-means++ algorithm.Specifically: if default cluster numerical value is k, a kind of method
It is that as leader node, remaining node is used as and follows node random optionally k node in each node of electric system；It is another
Method is that the local intelligent body of each node in electric system executes distribution K-means++ algorithm, is gathered according to specified presetting
Class numerical value k determines k node as leader node, and for remaining node as following node, the specific step principle of this method will be
It is illustrated in latter embodiments, wouldn't explain herein.
Step 120 is calculated separately using the first preset algorithm and each described follows node to each leader node
First distance；
In the present embodiment, first preset algorithm is the minimum consistency algorithm of discrete offset amount, the sheet of each node
The minimum consistency algorithm that ground intelligent body executes discrete offset amount calculates each most short distance for following node to each leader node
From the shortest distance is the first distance.
It should be noted that before calculating each shortest distance for following node to each leader node, according to electric power
The characteristic of system topology information and its flexibility resource, the local intelligent body of each node calculate the electricity between connected node
Gas characteristic distance.The classification needs of flexibility resource are considered the following factors:
Responding ability: ability (the up-regulation section c including adjusting loadu, lower section cd) and climbing capacity r etc.；
Features of response: main includes the speed s, duration T of prior notice etc. of resource response；
Responsiveness: including response times, the adjustable period maintains duration etc.；
Illustratively, since each component in the feature vector of resource faces standardized problem, i.e. big point of variance
Amount will occupy bigger weight.To solve this problem, the calculating of variance is introduced.Each component is standardized, such as:
After this operation, original feature vector (cu,cd, r, s, T) becomeAssuming that existing, there are two empty
Quasi- power plant, respectivelyWithDefine the distance of the two are as follows:
Above-mentioned distance can be understood as being exactly the electric characteristic distance between connected node.
Wherein, the minimum consistency algorithm concrete principle of discrete offset amount is as follows:
Assuming that the number of nodes in electric power networks topology is N, point set N can be divided into two point sets: S1 is the point of leader node
Collection, S2 is the point set for following node.The quantity of state of leader node remains unchanged, and is not influenced, is followed by other node state amounts
The quantity of state of node is determined by the quantity of state of leader node.Point set NiIt is the point set of node i adjacent node, xiFor the state of node
Value.
It is to be understood that the electric characteristic in obtaining electric system between connected node executes discrete offset after
When the minimum consistency algorithm of amount calculates each shortest distance for following node to each leader node, due to can exist some with
There is no direct-connected relationship with node and leader node, therefore one or more transfer connecting nodes will necessarily be passed through, thus it is direct-connected having
Select the route of wherein shortest path to calculate the most short distance for following node to leader node between the node of relationship every time
From.
Step 130, optional one follow node, follow node to the first distance of each leader node described in selection
The corresponding leader node of the middle shortest distance；
Step 140 follows node clustering to the leader node chosen for described, follows node clustering complete until all
Finish.
In the present embodiment, based on the first distance that step 120 is calculated, if some follows node to save to i-th of leader
The first distance of point is minimum (i=1,2 ..., k), then follows node clustering to the leader node this, follow node until all
All cluster finishes.Illustratively, such as: only there are two leader node 1 and 2, follow node a to the shortest distance of leader node 1
It is 10, the shortest distance for following node a to leader node 2 is 12, then node a will be followed to cluster to leader node 1, that is to say, that
Node a and leader node 1 is followed to be classified as one kind.Similarly, and so on, follow node to gather using the above method all
Class, until all node clustering is followed to finish, it can be understood as all to follow node all to determine corresponding to be classified as one with it
The leader node of class.
The technical solution of the present embodiment determines leader node by default cluster numerical value in each node of electric system,
Wherein, it using the node in all nodes in addition to leader node as node is followed, is calculated separately using the first preset algorithm each
The first distance that node is followed to each leader node, optional one follows node, follows node to arrive described in selection
The corresponding leader node of the shortest distance in the first distance of each leader node follows node clustering to selection for described
The leader node follows node clustering to finish until all.Realize flexibility, operational efficiency and the response for improving virtual plant
Efficiency reduces the economic cost and Environmental costs for maintaining stability of power system to reduce the dependence to conventional Power Generation Mode.
Fig. 2 is the flow diagram of the clustering method of another virtual plant provided in an embodiment of the present invention, referring to fig. 2,
This method further comprises following steps:
Step 210, in each classification, respectively using each node as sub- leader node, remaining node is followed as son
Node；
Step 220 is calculated separately each son using the first preset algorithm and follows node to the corresponding sub- leader
The second distance of node；
Step 230, calculate the second distance and value, and by described and value as the sub- leader node third away from
From；
Step 240, the third distance, and will wherein minimum third be saved apart from corresponding node as class leader
Point；
In the present embodiment, after cluster is completed, need to determine all kinds of leader nodes again in every class.It is exemplary
, if there is 6 nodes in electric system, nodal scheme is respectively 1,2,3,4,5 and 6, and leader node is 1 and 3, and cluster is completed
Later the result is that 1,2 and 4 is a kind of, 3,5 and 6 is a kind of.
Illustratively, in 1,2 and 4 this kind, 1 is used as sub- leader node, and 2 and 4 are used as son to follow node, 2,4 nodes
The minimum consistency algorithm that local intelligent body executes discrete offset amount calculates the shortest distance to sub- leader node 1, if S21 and
S41 enables S1=S21+S41, wherein the S21 and S41 is second distance, and S1 is third distance, the method for calculating S21 and S41
It is identical as the method for first distance is calculated in above-described embodiment step 120；2 are used as sub- leader node, and 1 and 4 are used as son to follow section
Point, the minimum consistency algorithm that 1,4 node local intelligent bodies execute discrete offset amount calculate the most short distance to sub- leader node 2
From enabling S2=S12+S42 if S12 and S42, wherein the S12 and S42 is second distance, and S2 is third distance；4 conducts
Sub- leader node, 1 and 2 are used as son to follow node, and 1,2 node local intelligent bodies execute the minimum consistency algorithm of discrete offset amount
The shortest distance for calculating sub- leader node 4 enables S4=S14+S24 if S14 and S24, wherein the S14 and S24 is the
Two distances, S4 are third distance；Based on S1, S2 and S4 obtained above, compare the size of three, if S1 is minimum, 1 node is
1, the class leader node in 2 and 4 this kind.
Similarly, in 3,5 and 6 this kind, class leader node is determined using above-mentioned identical method.
In the present embodiment, after obtaining third distance, the local intelligent body of node executes discrete minimum consistency algorithm
Find the smallest node of distance value in every class, the node is as new leader node, i.e., class leader node described herein, and wide
It broadcasts to all nodes.Wherein, discrete minimum consistency algorithm is as follows:
After the completion of broadcast, discrete minimum consistency algorithm reaches stationary value, the state value of the intelligent body of each node
It is as follows:
Xi(∞)=min { x1(0),K,xi(0), K, xn(0)}·1
Wherein, X (∞)=[x1(∞),K,xn(∞)]T；1=[1, K, 1]T, xiFor the state value of node.
Step 250 judges whether the class leader node and the leader node are consistent；
If step 260, consistent, end of clustering is determined；
If step 270, inconsistent, using the class leader node as leader node, remaining node is as following node；
Step 120 is returned and executes described calculated separately using the first preset algorithm and each described follow node to each institute
The step of stating the first distance of leader node.
Further, if class leader node mutually relatively before leader node change, using class leader node as newly
Leader node, remaining node return to step 120 as following node, until class leader node is identical as leader node, then
End of clustering.
Illustratively, above-mentioned to assume that 1 and 3 nodes are leader node, if 1 node is the class leader in 1,2 and 4 this kind
Node, 3 nodes are the class leader node in 3,5 and 6 this kind, then determine end of clustering；If 2 nodes are in 1,2 and 4 this kind
Class leader node, 5 nodes are the class leader node in 3,5 and 6 this kind, then using 2 nodes and 5 nodes as leader node,
120 are returned to step, is clustered again.
The technical solution of the present embodiment determines leader node after the completion of cluster again in each classification, if leader's section
Point is unchanged, then end of clustering, in this way, the accuracy of cluster is increased, so that the result of cluster more tallies with the actual situation.
Fig. 3 is the flow diagram that leader node provided in an embodiment of the present invention determines method, referring to Fig. 3, this method packet
Include following steps:
Step 310, an optional leader node and the default centroid vector of deposit in each node of electric system, remaining section
Point is used as and follows node；
Step 320, the electric characteristic parameter for obtaining the leader node, and the electric characteristic parameter is broadcasted to described
Follow node；
In the present embodiment, if default cluster numerical value is k, the vector that centroid vector is a k row 1 column is preset, if just opening
Optional 1 node begin as leader node, then the nodal scheme of 1 node is stored in default centroid vector, has been had selected to indicate
One leader node.Electric characteristic parameter can be understood as the feature vector value in above-described embodiment, the local intelligence of leader node
Energy body executes discrete most homogeneous algorithm and broadcasts default centroid vector, in this way, each follow node to both know about 1
Node is leader node.
Wherein, discrete most homogeneous algorithm is as follows:
After the completion of broadcast, discrete most homogeneous algorithm reaches stationary value, the state value of the intelligent body of each node
It is as follows:
Xi(∞)=max { x1(0),K,xi(0), K, xn(0)}·1
Wherein, X (∞)=[x1(∞),K,xn(∞)]T；1=[1, K, 1]T。
Follow node to the electric characteristic distance of the leader node described in step 330, calculating are each；
Step 340 is calculated using the second preset algorithm and each described follows node to the network topology of the leader node
Direct-connected most short electric characteristic distance；
In the present embodiment, second preset algorithm is discrete minimum consistency algorithm, has been done in the above-described embodiments
It introduces.If there is 6 nodes, nodal scheme is respectively 1,2,3,4,5 and 6,1 to be chosen to be leader node, and network is opened up
The electric characteristic distance for flutterring connected two node can be calculated, because 2,3,4,5,6 nodes and 1 node are not necessarily net
Network topology is directly connected to, therefore needs to calculate 2,3,4,5,6 nodes to 1 node there are the most short electrical of the direct-connected route of network topology
Characteristic distance.It is to be understood that obtaining 5 diValue, i indicate the label of 2,3,4,5,6 nodes.
It should be noted that in network topological diagram, if with a line to connect between 1 node and 2 nodes, then it represents that
1 node and 2 nodes are that network topology is direct-connected.
Step 350 determines random number range according to the most short electric characteristic distance；
Step 360 each described follows node to generate corresponding random number, wherein the random number of generation is described
In random number range；
The random number range is (0, di*di).It is each that node is followed to generate corresponding random number, as long as meeting generation
Random number is (0, di*di) in.
Step 370 will generate that largest random number is corresponding follows node as new leader node, and be stored in described default
Centroid vector；
The size of more each random number for following node to generate, will wherein largest random number be corresponding follows node conduct
New leader node, and it is stored in default centroid vector, expression has had selected a leader node.
Step 380 updates the default centroid vector, and updated default centroid vector is broadcasted to all nodes；
Step 390 judges whether the quantity of the leader node reaches default cluster numerical value by presetting centroid vector；
If step 3100 reaches, determine that leader node acquisition finishes；
If step 330, not up to, returns and execute described calculate and each described follow node to the electricity of the leader node
The step of gas characteristic distance.
In the present embodiment, in the new leader node of determination, and after being deposited into default centroid vector, update default mass center to
Amount, and updated default centroid vector is broadcasted to all nodes, in this way, other follow node to be known that new centroid vector
Which is.The information for being stored with all leader nodes due to presetting centroid vector, therefore mesh is known that by default centroid vector
Before have determined that several leader nodes, and then compared with default cluster numerical value, can know whether and obtain specified neck
Lead node.
Further, if determining leader node quantity has reached default cluster numerical value at present, determine that leader node obtains
It takes complete；Otherwise, it returns and executes described calculate and each described follow node to the step of the electric characteristic distance of the leader node
Suddenly, continue to obtain leader node.
The technical solution of the present embodiment teaches the method for determining leader node with distribution K-means++ algorithm, should
Method introduces the mode of Distributed Parallel Computing on the basis of traditional K-means++ algorithm, optimizes in the present invention and determines
The method of leader node improves the precision of cluster to reduce the amount of work of Distributed Cluster calculating.
Fig. 4 is a kind of structural schematic diagram of the clustering apparatus of virtual plant provided in an embodiment of the present invention, which is applicable in
In the clustering method for executing the virtual plant that any embodiment of that present invention provides, as shown in figure 4, the device includes: leader node
Obtain module 410, shortest distance determining module 420, leader node selecting module 430 and cluster module 440.
Leader node obtains module 410, for determining leader by default cluster numerical value in each node of electric system
Node, wherein using the node in all nodes in addition to leader node as following node；
Shortest distance determining module 420 each described follows node to respectively for calculating separately using the first preset algorithm
The first distance of a leader node；
Leader node selecting module 430 follows node for optional one, follows node to each neck described in selection
Lead the corresponding leader node of the shortest distance in the first distance of node；
Cluster module 440 follows section until all for following node clustering to the leader node chosen for described
Point cluster finishes.
The clustering apparatus of virtual plant provided in this embodiment passes through default cluster numerical value in each node of electric system
Determine leader node, wherein using the node in all nodes in addition to leader node as node is followed, using the first preset algorithm
Calculate separately it is each it is described follow node to the first distance of each leader node, optional one follows node, selection institute
It states and follows node corresponding leader node of the shortest distance into the first distance of each leader node, follow node for described
Cluster follows node clustering to finish to the leader node chosen until all.Realize the flexibility for improving virtual plant, fortune
Line efficiency and response efficiency, to reduce dependence to conventional Power Generation Mode, reduce maintain stability of power system it is economical at
Sheet and Environmental costs.
On the basis of the above embodiments, further includes:
Class leader node obtains module, for obtaining the class leader node in each classification when clustering completion；
Judge whether the class leader node and the leader node are consistent；
If consistent, end of clustering is determined.
On the basis of the above embodiments, further includes:
If inconsistent, using the class leader node as leader node, remaining node is used as and follows node；
It returns and executes described calculate separately using the first preset algorithm and each described follow node to each leader's section
The step of first distance of point.
On the basis of the above embodiments, class leader node acquisition module includes:
In each classification, respectively using each node as sub- leader node, remaining node follows node as son；
Calculating separately each son using the first preset algorithm follows node to the of the corresponding sub- leader node
Two distances；
Calculate the second distance and value, and by third distance described and that value is as the sub- leader node；
Compare the third distance, and will wherein minimum third apart from corresponding node as class leader node.
On the basis of the above embodiments, leader node acquisition module 410 includes:
An optional leader node and be stored in default centroid vector in each node of electric system, remaining node be used as with
With node；
The electric characteristic parameter of the leader node is obtained, and the electric characteristic parameter is broadcasted to described and follows section
Point；
It calculates and each described follows node to the electric characteristic distance of the leader node；
It is direct-connected most that each network topology for following node to the leader node is calculated using the second preset algorithm
Short electric characteristic distance；
New leader node obtains module, for determining new leader node according to the most short electric characteristic distance, and deposits
Enter the default centroid vector；
The default centroid vector is updated, and updated default centroid vector is broadcasted to all nodes；
Judge whether the quantity of the leader node reaches default cluster numerical value by presetting centroid vector；
If reaching, determine that leader node acquisition finishes；
If not up to, return execute it is described calculate it is each it is described follow node to the leader node electric characteristic away from
From the step of.
On the basis of the above embodiments, new leader node acquisition module includes:
Random number range is determined according to the most short electric characteristic distance；
Node is followed to generate corresponding random number described in each, wherein the random number of generation is in the random number model
In enclosing；
To generate that largest random number is corresponding follows node as new leader node, and be stored in the default mass center to
Amount.
On the basis of the above embodiments, first preset algorithm is the minimum consistency algorithm of discrete offset amount；Institute
Stating the second preset algorithm is discrete minimum consistency algorithm.
The embodiment of the invention provides a kind of computer readable storage mediums, are stored thereon with computer program, the program
The clustering method of the virtual plant provided such as all embodiments of the invention is realized when being executed by processor: that is, the program is located
Reason device is realized when executing: determining leader node by default cluster numerical value in each node of electric system, wherein by all sections
Node in point in addition to leader node is calculated separately using the first preset algorithm and each described node is followed to arrive as following node
The first distance of each leader node, optional one follows node, and node is followed to save described in selection to each leader
Point first distance in the corresponding leader node of the shortest distance, by it is described follow node clustering to choose the leader node,
Node clustering is followed to finish until all.
It can be using one or any combination of at least two computer-readable media.Computer-readable medium can be
Computer-readable signal media or computer readable storage medium.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example (non exhaustive list) of calculation machine readable storage medium storing program for executing includes: being electrically connected with one or at least two conducting wires
Connect, portable computer diskette, hard disk, random access memory (RAM), read-only memory (ROM), erasable type may be programmed it is read-only
Memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory
Part or above-mentioned any appropriate combination.In this document, computer readable storage medium, which can be, any include or stores
The tangible medium of program, the program can be commanded execution system, device or device use or in connection.
Computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
Wherein carry computer-readable program code.The data-signal of this propagation can take various forms, including --- but
It is not limited to --- electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be
Any computer-readable medium other than computer readable storage medium, which can send, propagate or
Transmission is for by the use of instruction execution system, device or device or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In --- wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
The computer for executing operation of the present invention can be write with one or more programming languages or combinations thereof
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
It further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partially executes or executed on remote computer or adjusting device completely on the remote computer on the user computer.
In situations involving remote computers, remote computer can pass through the network of any kind --- including local area network (LAN)
Or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as utilize Internet service
Provider is connected by internet).
Fig. 5 is a kind of structural schematic diagram of adjusting device provided in an embodiment of the present invention, which can collect cost hair
The clustering apparatus for the virtual plant that bright embodiment provides.Referring to Fig. 5, adjusting device 500 may include: memory 510, processor
520 and the computer program that is stored on memory 510 and can be run in processor 520, the processor 520 execute the meter
The clustering method of virtual plant as described in the embodiments of the present invention is realized when calculation machine program.
Adjusting device provided in an embodiment of the present invention determines neck by default cluster numerical value in each node of electric system
Lead node, wherein using the node in all nodes in addition to leader node as node is followed, count respectively using the first preset algorithm
Calculate it is each it is described follow node to the first distance of each leader node, optional one follows node, follows described in selection
Node corresponding leader node of the shortest distance into the first distance of each leader node, follows node clustering extremely for described
The leader node chosen follows node clustering to finish until all.Realize the flexibility for improving virtual plant, operational efficiency
And response efficiency reduces the economic cost and ring for maintaining stability of power system to reduce the dependence to conventional Power Generation Mode
Border cost.
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of clustering method of virtual plant characterized by comprising
Leader node is determined by default cluster numerical value in each node of electric system, wherein by all nodes except leader
Node outside node, which is used as, follows node；
Being calculated separately using the first preset algorithm each described follows node to the first distance of each leader node；
Optional one follows node, and node shortest distance pair into the first distance of each leader node is followed described in selection
The leader node answered；
It follows node clustering to the leader node chosen for described, follows node clustering to finish until all.
2. the method according to claim 1, wherein following node clustering to the leader section chosen for described
Point, until it is all follow node clustering to finish after, further includes:
When clustering completion, the class leader node in each classification is obtained；
Judge whether the class leader node and the leader node are consistent；
If consistent, end of clustering is determined.
3. according to the method described in claim 2, it is characterized in that, judging whether are the class leader node and the leader node
After consistent, the clustering method of the virtual plant further include:
If inconsistent, using the class leader node as leader node, remaining node is used as and follows node；
It returns and executes described calculate separately using the first preset algorithm and each described follow node to each leader node
The step of first distance.
4. according to the method described in claim 2, it is characterized in that, obtaining the class leader in each classification when clustering completion
Node, comprising:
In each classification, respectively using each node as sub- leader node, remaining node follows node as son；
Using the first preset algorithm calculate separately each son follow node to the corresponding sub- leader node second away from
From；
Calculate the second distance and value, and by third distance described and that value is as the sub- leader node；
Compare the third distance, and will wherein minimum third apart from corresponding node as class leader node.
5. the method according to claim 1, wherein passing through default cluster numerical value in each node of electric system
Determine leader node, comprising:
Centroid vector is preset in an optional leader node and deposit in each node of electric system, remaining node is used as and follows section
Point；
The electric characteristic parameter of the leader node is obtained, and the electric characteristic parameter is broadcasted to described and follows node；
It calculates and each described follows node to the electric characteristic distance of the leader node；
The direct-connected most short electricity of each network topology for following node to the leader node is calculated using the second preset algorithm
Gas characteristic distance；
New leader node is determined according to the most short electric characteristic distance, and is stored in the default centroid vector；
The default centroid vector is updated, and updated default centroid vector is broadcasted to all nodes；
Judge whether the quantity of the leader node reaches default cluster numerical value by presetting centroid vector；
If reaching, determine that leader node acquisition finishes；
If not up to, returning and executing described calculate and each described follow node to the electric characteristic distance of the leader node
Step.
6. according to the method described in claim 5, it is characterized in that, determining new leader according to the most short electric characteristic distance
Node, and it is stored in the default centroid vector, comprising:
Random number range is determined according to the most short electric characteristic distance；
Node is followed to generate corresponding random number described in each, wherein the random number of generation is in the random number range；
It will generate that largest random number is corresponding follows node as new leader node, and be stored in the default centroid vector.
7. method according to claim 1-6, which is characterized in that first preset algorithm is discrete offset amount
Minimum consistency algorithm；Second preset algorithm is discrete minimum consistency algorithm.
8. a kind of clustering apparatus of virtual plant characterized by comprising
Leader node obtains module, for determining leader node by default cluster numerical value in each node of electric system,
In, using the node in all nodes in addition to leader node as following node；
Shortest distance determining module each described follows node to each neck for calculating separately using the first preset algorithm
Lead the first distance of node；
Leader node selecting module follows node for optional one, follows node to each leader node described in selection
First distance in the corresponding leader node of the shortest distance；
Cluster module follows node clustering until all for following node clustering to the leader node chosen for described
It finishes.
9. a kind of computer storage medium, is stored thereon with computer program, which is characterized in that when the program is executed by processor
Realize the clustering method of the virtual plant as described in any in claim 1-7.
10. a kind of adjusting device, which is characterized in that including memory, processor and storage are on a memory and can be in processor
The computer program of operation, the processor are realized as described in any in claim 1-7 when executing the computer program
The clustering method of virtual plant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811418832.XA CN109522964B (en) | 2018-11-26 | 2018-11-26 | Clustering method and device for virtual power plant, regulation and control equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811418832.XA CN109522964B (en) | 2018-11-26 | 2018-11-26 | Clustering method and device for virtual power plant, regulation and control equipment and computer storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109522964A true CN109522964A (en) | 2019-03-26 |
| CN109522964B CN109522964B (en) | 2023-03-31 |
Family
ID=65778808
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811418832.XA Active CN109522964B (en) | 2018-11-26 | 2018-11-26 | Clustering method and device for virtual power plant, regulation and control equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109522964B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111245026A (en) * | 2020-03-09 | 2020-06-05 | 国网冀北电力有限公司 | Virtual power plant regulation and control method, system and equipment |
| CN111382939A (en) * | 2020-03-06 | 2020-07-07 | 国网冀北电力有限公司 | Virtual power plant resource optimal configuration method, device and equipment |
| CN111815018A (en) * | 2020-05-29 | 2020-10-23 | 国网冀北电力有限公司计量中心 | Optimal scheduling method and device for virtual power plant |
| CN112929272A (en) * | 2021-03-24 | 2021-06-08 | 国网上海市电力公司 | Segment routing information transmission method, device, electronic equipment, system and storage medium |
| CN112965945A (en) * | 2019-12-13 | 2021-06-15 | 阿里巴巴集团控股有限公司 | Data storage method and device, electronic equipment and computer readable medium |
| CN113507141A (en) * | 2021-07-08 | 2021-10-15 | 国网上海市电力公司 | Virtual power plant equivalent closed-loop control method and system, electronic equipment and storage medium |
| CN114123175A (en) * | 2021-11-15 | 2022-03-01 | 东北大学 | Power distribution network distributed control method considering generalized demand side resource alliance characteristics |
| CN117154747A (en) * | 2023-10-30 | 2023-12-01 | 江苏益捷思信息科技有限公司 | Virtual power plant demand response system and method based on multi-tenant technology |
| CN117498467A (en) * | 2024-01-02 | 2024-02-02 | 国网浙江省电力有限公司宁波供电公司 | Energy coordination control distribution method and system based on multi-level virtual power plant |
| CN117498467B (en) * | 2024-01-02 | 2024-05-03 | 国网浙江省电力有限公司宁波供电公司 | Energy coordination control distribution method and system based on multi-level virtual power plant |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030204625A1 (en) * | 2002-04-29 | 2003-10-30 | Harris Corporation | Hierarchical mobile ad-hoc network and methods for performing reactive routing therein using ad-hoc on-demand distance vector routing (AODV) |
| CN1925437A (en) * | 2005-08-29 | 2007-03-07 | 国际商业机器公司 | System and method for detecting status changes in a network |
| US20110055140A1 (en) * | 2009-09-01 | 2011-03-03 | Oracle International Corporation | Expediting k-means cluster analysis data mining using subsample elimination preprocessing |
| CN105354393A (en) * | 2015-12-02 | 2016-02-24 | 上海核工程研究设计院 | System and method for realizing nuclear plant three-dimensional digital map |
| CN105656040A (en) * | 2016-03-22 | 2016-06-08 | 国网山东省电力公司潍坊供电公司 | Power grid restoring capacity calculation method considering black-start process |
| CN107093150A (en) * | 2017-03-13 | 2017-08-25 | 国网安徽省电力公司电力科学研究院 | Active power distribution network assemblage classification method based on equiva lent impedance |
| CN107330809A (en) * | 2017-04-12 | 2017-11-07 | 广东电网有限责任公司电力调度控制中心 | Black starting-up partition method based on semi-supervised spectral clustering |
| CN108574300A (en) * | 2018-04-19 | 2018-09-25 | 中国科学院电工研究所 | A kind of distributed photovoltaic power generation power rating method of estimation based on similitude polymerization |
| CN108599238A (en) * | 2018-04-26 | 2018-09-28 | 国家电网公司 | Consider the virtual plant distributed generation resource planing method of wind energy and solar energy complementation |
| CN108629375A (en) * | 2018-05-08 | 2018-10-09 | 广东工业大学 | Power customer sorting technique, system, terminal and computer readable storage medium |
-
2018
- 2018-11-26 CN CN201811418832.XA patent/CN109522964B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030204625A1 (en) * | 2002-04-29 | 2003-10-30 | Harris Corporation | Hierarchical mobile ad-hoc network and methods for performing reactive routing therein using ad-hoc on-demand distance vector routing (AODV) |
| CN1925437A (en) * | 2005-08-29 | 2007-03-07 | 国际商业机器公司 | System and method for detecting status changes in a network |
| US20110055140A1 (en) * | 2009-09-01 | 2011-03-03 | Oracle International Corporation | Expediting k-means cluster analysis data mining using subsample elimination preprocessing |
| CN105354393A (en) * | 2015-12-02 | 2016-02-24 | 上海核工程研究设计院 | System and method for realizing nuclear plant three-dimensional digital map |
| CN105656040A (en) * | 2016-03-22 | 2016-06-08 | 国网山东省电力公司潍坊供电公司 | Power grid restoring capacity calculation method considering black-start process |
| CN107093150A (en) * | 2017-03-13 | 2017-08-25 | 国网安徽省电力公司电力科学研究院 | Active power distribution network assemblage classification method based on equiva lent impedance |
| CN107330809A (en) * | 2017-04-12 | 2017-11-07 | 广东电网有限责任公司电力调度控制中心 | Black starting-up partition method based on semi-supervised spectral clustering |
| CN108574300A (en) * | 2018-04-19 | 2018-09-25 | 中国科学院电工研究所 | A kind of distributed photovoltaic power generation power rating method of estimation based on similitude polymerization |
| CN108599238A (en) * | 2018-04-26 | 2018-09-28 | 国家电网公司 | Consider the virtual plant distributed generation resource planing method of wind energy and solar energy complementation |
| CN108629375A (en) * | 2018-05-08 | 2018-10-09 | 广东工业大学 | Power customer sorting technique, system, terminal and computer readable storage medium |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112965945A (en) * | 2019-12-13 | 2021-06-15 | 阿里巴巴集团控股有限公司 | Data storage method and device, electronic equipment and computer readable medium |
| CN111382939A (en) * | 2020-03-06 | 2020-07-07 | 国网冀北电力有限公司 | Virtual power plant resource optimal configuration method, device and equipment |
| CN111245026A (en) * | 2020-03-09 | 2020-06-05 | 国网冀北电力有限公司 | Virtual power plant regulation and control method, system and equipment |
| CN111815018A (en) * | 2020-05-29 | 2020-10-23 | 国网冀北电力有限公司计量中心 | Optimal scheduling method and device for virtual power plant |
| CN112929272B (en) * | 2021-03-24 | 2024-01-26 | 国网上海市电力公司 | Segment routing information transmission method, device, electronic equipment, system and storage medium |
| CN112929272A (en) * | 2021-03-24 | 2021-06-08 | 国网上海市电力公司 | Segment routing information transmission method, device, electronic equipment, system and storage medium |
| CN113507141A (en) * | 2021-07-08 | 2021-10-15 | 国网上海市电力公司 | Virtual power plant equivalent closed-loop control method and system, electronic equipment and storage medium |
| CN113507141B (en) * | 2021-07-08 | 2024-04-09 | 国网上海市电力公司 | Virtual power plant equivalent closed-loop control method, system, electronic equipment and storage medium |
| CN114123175A (en) * | 2021-11-15 | 2022-03-01 | 东北大学 | Power distribution network distributed control method considering generalized demand side resource alliance characteristics |
| CN114123175B (en) * | 2021-11-15 | 2023-11-07 | 东北大学 | Distributed control method for power distribution network considering generalized demand side resource alliance characteristics |
| CN117154747B (en) * | 2023-10-30 | 2023-12-26 | 江苏益捷思信息科技有限公司 | Virtual power plant demand response system and method based on multi-tenant technology |
| CN117154747A (en) * | 2023-10-30 | 2023-12-01 | 江苏益捷思信息科技有限公司 | Virtual power plant demand response system and method based on multi-tenant technology |
| CN117498467A (en) * | 2024-01-02 | 2024-02-02 | 国网浙江省电力有限公司宁波供电公司 | Energy coordination control distribution method and system based on multi-level virtual power plant |
| CN117498467B (en) * | 2024-01-02 | 2024-05-03 | 国网浙江省电力有限公司宁波供电公司 | Energy coordination control distribution method and system based on multi-level virtual power plant |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109522964B (en) | 2023-03-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109522964A (en) | Clustering method, device, adjusting device and the computer storage medium of virtual plant | |
| CN107679658B (en) | Power transmission network planning method under high-proportion clean energy access | |
| CN108470239A (en) | The active distribution network multi objective layered programming method of meter and demand side management and energy storage | |
| CN109583665A (en) | A kind of unmanned plane charging tasks dispatching method in wireless sensor network | |
| CN109687458A (en) | Consider the space truss project method of regional distribution network risk tolerance difference | |
| CN102953936B (en) | Unit commitment for wind power generation | |
| CN109066660A (en) | A kind of power distribution network congestion management and decreasing loss method and apparatus based on optimal reconfiguration | |
| CN107039984A (en) | A kind of reactive voltage partition method | |
| CN108649605A (en) | A kind of grid-connected allowed capacity planing methods of DER based on the double-deck scene interval trend | |
| CN108769843B (en) | Power optical fiber home-in EPON (Ethernet passive optical network) optimization method and system | |
| CN111900719A (en) | Power grid adequacy evaluation method, device and system considering flexible controllable load | |
| CN110740394B (en) | Power optical transmission network optimization method and system based on site classification business classification | |
| CN103595652B (en) | The stage division of QoS efficiency in a kind of powerline network | |
| CN113132437A (en) | CDN scheduling method, system, device and storage medium | |
| Griffone et al. | Applications of clustering techniques to the definition of the bidding zones | |
| CN109165718A (en) | Network reconstruction method and system based on paralleling ant cluster algorithm | |
| CN107911763B (en) | Intelligent power distribution and utilization communication network EPON network planning method based on QoS | |
| CN105591389B (en) | Distributed power supply voltage control method based on hybrid hierarchical multi-agent system | |
| CN115360717B (en) | Power distribution network scheduling method, system, equipment and medium for comprehensive energy | |
| CN108493908B (en) | A kind of locality protection partition method considering communication constraint | |
| CN110046766A (en) | A kind of cooperative optimization method of virtual plant and power distribution network | |
| KR102090319B1 (en) | method for competitive pricing mechanism for demand side management of energy system | |
| CN109688068A (en) | Network load balancing method and device based on big data analysis | |
| CN110198543A (en) | Network resource planning method, apparatus, computer-readable medium and electronic equipment | |
| CN110392409B (en) | WMSNs multipath QoS routing method, system and storage medium based on distribution communication 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |