CN103093097A - Electrical power system fragile section identification method based on normalized-cut - Google Patents

Electrical power system fragile section identification method based on normalized-cut Download PDF

Info

Publication number
CN103093097A
CN103093097A CN201310014905XA CN201310014905A CN103093097A CN 103093097 A CN103093097 A CN 103093097A CN 201310014905X A CN201310014905X A CN 201310014905XA CN 201310014905 A CN201310014905 A CN 201310014905A CN 103093097 A CN103093097 A CN 103093097A
Authority
CN
China
Prior art keywords
vector
node
sigma
matrix
power system
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
Application number
CN201310014905XA
Other languages
Chinese (zh)
Inventor
赵金利
李媛媛
贾宏杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tianjin University
Original Assignee
Tianjin University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tianjin University filed Critical Tianjin University
Priority to CN201310014905XA priority Critical patent/CN103093097A/en
Publication of CN103093097A publication Critical patent/CN103093097A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention relates to the technical field of electrical power system static status safety analysis and on-line monitoring, and provides a quick method of identifying an electrical power system fragile section. An electrical power system fragile section identification method based on normalized-cut has important guiding significance for the establishment of an effective prevention and control measure and an islanding control strategy. The electrical power system fragile section identification method based on the normalized-cut comprises a first step of establishing a suitable electrical power system graph theory model, a second step of establishing a laplacian matrix and a similarity matrix of the electrical power system graph theory model, a third step of forming corresponding split guiding vectors, searching corresponding sections of the split guiding vectors in a heuristic mode, a forth step of building an objective function regarding to the split guiding vectors based on the normalized-cut mode, calculating the objective function of each of the split guiding vectors, and a fifth step of screening a split guiding vector which enables the objective function to obtain the minimum value, determining the section corresponding to the split guiding vector to be a fragile section. The electrical power system fragile section identification method based on the normalized-cut is mainly applied to the electrical power system static status safety analysis and on-line monitoring.

Description

The electric system vulnerable profile recognition method of cutting based on standard
Technical field
The present invention relates to Contingency Analysis of Power Systems and in-service monitoring technical field, specifically, relate to the electric system vulnerable profile recognition method of cutting based on standard.
Background technology
In order to guarantee the safe and stable operation of electric system, need real-time analysis and monitor some key characters that characterize security of system.Key sections is a kind of important form of characterization system security.Therefore for many years, key sections is one of focus of electrical network analysis and dispatcher's concern always, and its recognition methods has also obtained very large development in the evolution of electrical network.
Recognition methods for key sections in electric system at present mainly comprises following three classes:
1. based on the transmission cross-section recognition methods of interregional interconnection.These class methods are usually according to the regional feature of system, affiliated administrative region, and the interconnection interregional in conjunction with operations staff's experience artificial selection forms transmission cross-section.At the initial stage of Power System Interconnection, the net connection between each zone is weaker, and this moment, electric network composition was still more clear, utilizes the method identification transmission cross-section simple, and fast, practical application is more.But along with the scale of interconnected network increases day by day, interregional contact strengthens gradually, and the method for operation is more flexible and changeable, and the transmission cross-section that this method identifies might be also insufficient and accurate.
2. the transmission cross-section recognition methods of testing based on branch breaking.These class methods are intended to be subjected to the larger line combination of influence on tidal flow to consist of section by cut-offfing the test of overload branch road, filtering out.This class section exists the threat that suffers than the megatrend transfer after the branch road overload trip, have the potential safety hazard that causes the cascading trip fault.But the transmission cross-section that the method identifies depends on the selection of cut-offfing branch road, selects different branch roads to cut-off test, may obtain different transmission cross-sections, has significant limitation.
3. based on the key sections recognition methods of topological structure.These class methods are from the viewpoint of complex network, electric system is converted into the topological model that is formed by point and limit, and then utilize Complex Networks Theory identification by the section that the minority circuit forms, and will make during these line failures that between most of nodes, shortest path becomes large, make system's fragility that becomes.But these class methods only relate to the physical arrangement of system, do not take into account the impact of trend, and along with the variation of power system operating mode and the adjustment of trend on a large scale, the key sections that the method identifies might be also insufficient and accurate.
When said method was applied to the more flexible and changeable interconnected network of the method for operation nowadays, the section that identifies all may be inaccurate.Exist a kind of in electric system and the closely-related fragile section of electrical network Real-time Power Flow distribution, this section is comprised of the weak circuit of contact on physical arrangement, and these circuits trend conveying in system simultaneously is larger.Very easily destroy the passway for transmitting electricity of electrical network when the line failure in this section, cause that in system, trend shifts, the safety and stability problems such as branch road overload easily occur, jeopardize the security of system operation.
Summary of the invention
The present invention is intended to overcome the deficiencies in the prior art, a kind of fast method of identifying the electric system vulnerable section is provided, has great importance to formulating effective prevention and control measure and off-the-line control strategy, for achieving the above object, the technical scheme that the present invention takes is, electric system vulnerable profile recognition method based on standard is cut comprises the steps:
The first step: according to the needs of fragile profile recognition, calculate the electrical network Real-time Power Flow, utilize based on the electric system branch road of electrical distance and trend distribution and compose the power method, set up applicable electric system graph theory model;
Second step: set up this abstract model Laplacian Matrix and degree matrix, and calculate Laplacian Matrix about minimum non-zero eigenwert and the characteristic of correspondence vector thereof of degree matrix;
The 3rd step: the direction that provides according to this proper vector is divided into two groups with the node in system, forms corresponding cutting apart and indicates vector, searches for the section of this vector correspondence heuristicly;
The 4th step: a little less than contact on the circuit physical arrangement in fragile section, and trend is carried larger characteristics in system, builds the objective function of indicating vector about cutting apart based on the standard blanking method, and each is cut apart indication vector calculation target function value;
The 5th step: the screening objective function of sening as an envoy to is obtained the indication vector cut apart of minimum value, and the section of this vector correspondence is exactly fragile section.
The electric system branch road that utilization distributes based on electrical distance and trend is composed the power method and is specially: the graph theory model G=(V that contains n node and m bar limit, E, W), V is vertex set, and the bus in corresponding real system, generator, load, E are the limit collection, the branch road of corresponding connection bus, comprise circuit and transformer, W is weight matrix, weighting matrix elements w ijBe connected node v iAnd v jThe weight on limit; Utilization is composed the power method based on the electric system branch road of electrical distance and trend distribution branch road is composed power:
w ij = y ij | P ij | - - - ( 1 )
y ijThe line admittance of connected node i and node j, P ijTo flow to the meritorious trend of ground state of node j from node i.
Calculating Laplacian Matrix is specially about minimum non-zero eigenwert and the characteristic of correspondence vector thereof of spending matrix: the degree of node i is:
d i = Σ j = 1 n w ij - - - ( 2 )
The degree matrix D is with d 1, nBe the diagonal matrix of diagonal element, Laplacian Matrix L is:
L=D-W (3)
Method for Solving Generalized Eigenproblem:
Lz=λDz (4)
λ representation feature value, the characteristic of correspondence vector is z, minimum non-zero eigenwert characteristic of correspondence vector v FExpression.
Cut apart the indication vector:
x=(x 1,x 2,…x n) T∈R n (5)
x i = 1 , i ∈ A 1 - 1 , i ∈ A 2 - - - ( 6 )
A 1And A 2Represent that respectively will scheme G is divided into the node that two subgraphs comprise, and cuts apart the situation of cutting apart that the indication vector x has indicated figure, order
k=v F +α (7)
α be one by-∞ to+number that ∞ changes, form the corresponding indication vector x of cutting apart according to the symbol of element of k:
x i = 1 , k i &GreaterEqual; 0 - 1 , k i < 0 - - - ( 8 )
Get rid of the situations that in x, element all equates, be i.e. nonsensical cutting apart, can be formed to many n-1 and cut apart the indication vector x.
The 4th step further was specially:
Fragile section C should satisfy as much as possible:
min &Sigma; l ij &Element; C y ij , max &Sigma; l ij &Element; C | P ij | - - - ( 9 )
l ijThe circuit of expression connected node i and node j, y ijCircuit l ijAdmittance, suppose that circuit can't harm, P ijCircuit l ijThe meritorious trend of ground state, (9) first of formulas are revised as:
min &Sigma; l ij &Element; C y ij &Sigma; i &Element; A 1 , t &Element; V y it &Sigma; j &Element; A 2 , t &Element; V y jt ( 10 )
Consider on this basis active power, require the meritorious larger of circuit transmission, build about cutting apart the objective function of indication vector:
min &Sigma; i &Element; A 1 , j &Element; A 2 y ij | P ij | ( x i - x j ) 2 &Sigma; i &Element; A 1 , t &Element; V y it | P ij | &Sigma; j &Element; A 2 , t &Element; V y jt | P ij | - - - ( 11 )
With x iBe relaxed to real number field, according to Rayleigh theorem, x=v FThe time, formula (11) obtains minimum value; With gained x difference substitution formula (11) calculating target function value in the 3rd step.
Technical characterstic of the present invention and effect:
By method provided by the present invention, utilization is composed the power method based on the electric system branch road of electrical distance and trend distribution, set up and be applicable to the nowadays more flexible and changeable interconnected network electric system graph theory model of the method for operation, and then utilize the standard blanking method to build about cutting apart the objective function of indication vector, rapidly and accurately in recognition system by a little less than contact on physical arrangement, and trend is carried the fragile section that larger circuit forms in system.The attention degree of raising to these sections has great importance to formulating effective prevention and control measure and off-the-line control strategy.
Description of drawings
The electric system vulnerable profile recognition method flow diagram of cutting based on standard that Fig. 1 is proposed by the invention.
Fig. 2 IEEE 30 node system figure and the fragile section that identifies.
Embodiment
The present invention is intended to a kind of approximation method fast, from the fragile section of this class of the whole aspect identification of electrical network.Be the combination of some circuits due to section, identify namely np complete problem of fragile section, simultaneously because actual electric network is huge, the section quantity that comprises is also very large, is difficult to obtain all sections.In recent years, existing proposition utilizes the spectrogram opinion to realize spectral clustering analysis, Grid division, the power transmission network subregion of Voltage Stability Analysis and the determining of key sections of Power grid structure, and these achievements in research show that the spectrogram opinion provides a kind of approximate data of Simple fast for np complete problem.
The purpose of this invention is to provide a kind of fast method of identifying the electric system vulnerable section, set up the graph theory abstract model of electric system in the method, utilize the standard blanking method of spectral graph theory, a little less than contacting from the whole aspect identification of electrical network by physical arrangement, and trend is carried larger circuit in system, consist of the fragile section in electric system, have great importance to formulating effective prevention and control measure and off-the-line control strategy.
The present invention proposes a kind of electric system vulnerable profile recognition method of cutting based on standard, is used for the safe in-service monitoring of power system static, and its step is as follows:
The first step: according to the needs of fragile profile recognition, calculate the electrical network Real-time Power Flow, utilize based on the electric system branch road of electrical distance and trend distribution and compose the power method, set up applicable electric system graph theory model;
Second step: set up this abstract model Laplacian Matrix and degree matrix, and calculate Laplacian Matrix about minimum non-zero eigenwert and the characteristic of correspondence vector thereof of degree matrix;
The 3rd step: the direction that provides according to this proper vector is divided into two groups with the node in system, forms corresponding cutting apart and indicates vector, searches for the section of this vector correspondence heuristicly;
The 4th step: a little less than contact on the circuit physical arrangement in fragile section, and trend is carried larger characteristics in system, builds the objective function of indicating vector about cutting apart based on the standard blanking method, and each is cut apart indication vector calculation target function value;
The 5th step: the screening objective function of sening as an envoy to is obtained the indication vector cut apart of minimum value, and the section of this vector correspondence is exactly fragile section.
The present invention according to the needs of fragile profile recognition, calculates the electric network swim distribution situation in step 1, utilize based on the electric system branch road of electrical distance and trend distribution and compose the power method, sets up applicable electric system graph theory model.
Contain the graph theory model G=(V, E, W) on n node and m bar limit, V is vertex set, and the bus in corresponding real system, generator, load, E are the limit collection, and the branch road of corresponding connection bus comprises circuit and transformer, and W is weight matrix, weighting matrix elements w ijBe connected node v iAnd v jThe weight on limit.Utilization is composed the power method based on the electric system branch road of electrical distance and trend distribution branch road is composed power:
w ij = y ij | P ij | - - - ( 1 )
y ijThe line admittance of connected node i and node j, P ijTo flow to the meritorious trend of ground state of node j from node i.Suppose that circuit can't harm.Admittance is less, and the physical connection of circuit is more weak.The meritorious trend absolute value of circuit is larger, and the rear impact on system of breaking down is larger.Therefore, little weight means weak and this circuit important role in trend distributes of internodal physical connection.Weight w ijVariation with Line Flow changes, and it has reflected the power characteristic of this circuit, and also the variation with the line reactance parameter changes, and has reflected simultaneously the network characteristic of this circuit.
In step 2, set up Laplacian Matrix and the degree matrix of electric system graph theory abstract model, and calculate Laplacian Matrix about minimum non-zero eigenwert and the characteristic of correspondence vector thereof of degree matrix.
The degree of node i is:
d i = &Sigma; j = 1 n w ij - - - ( 2 )
The degree matrix D is with d 1... d nDiagonal matrix for diagonal element.Laplacian matrix of graphs L is:
L=D-W (3)
Method for Solving Generalized Eigenproblem:
Lz=λDz (4)
λ representation feature value, the characteristic of correspondence vector is z.Use v FExpression minimum non-zero eigenwert characteristic of correspondence vector.
In step 3, the direction that provides according to this proper vector is divided into two groups with the node in system, forms corresponding cutting apart and indicates vector, searches for the section of this vector correspondence heuristicly.
Cut apart the indication vector:
x=(x 1,x 2,…x n) T∈R n (5)
x i = 1 , i &Element; A 1 - 1 , i &Element; A 2 - - - ( 6 )
A 1And A 2Represent that respectively will scheme G is divided into the node that two subgraphs comprise.Cut apart the situation of cutting apart that the indication vector x has indicated figure.Order
k=v F+α (7)
α be one by-∞ to+number that ∞ changes.Form the corresponding indication vector x of cutting apart according to the symbol of element of k:
x i = 1 , k i &GreaterEqual; 0 - 1 , k i < 0 - - - ( 8 )
Get rid of the situations that in x, element all equates, be i.e. nonsensical cutting apart, can be formed to many n1 and cut apart the indication vector x.
In step 4, a little less than contact on the circuit physical arrangement in fragile section, and trend is carried larger characteristics in system, builds the objective function of indicating vector about cutting apart based on the standard blanking method, and each is cut apart indication vector calculation target function value.
If interregional grid structure a little less than, and the trend on cut set is very little, multiple failure on this section occurs, the general extensive transfer that can not cause trend, the safety and stability problem be difficult for to occur, otherwise, when carrying the line failure than megatrend in system, very easily cause the extensive transfer of trend, cause the safety and stability problem.Therefore fragile section C should satisfy as much as possible:
min &Sigma; l ij &Element; C y ij , max &Sigma; l ij &Element; C | P ij | - - - ( 9 )
l ijThe circuit of expression connected node i and node j, y ijCircuit l ijAdmittance, suppose that circuit can't harm, P ijCircuit l ijThe meritorious trend of ground state.Line admittance is less, shows that circuit is longer, and internodal physical connection is more weak, the admittance of cut set and Characterized the connection tightness degree of subsystem.
Figure BDA00002739878900055
Larger, show that the meritorious trend of carrying on circuit in cut set is larger, the effect of cut set on trend distributes is more important.Utilize the standard blanking method can fast searching to making two contacts between subgraph weak, and subgraph internal connection cut set closely.According to the objective function of standard blanking method, for the section of avoiding identifying comprises the generator outlet circuit, (9) first of formulas are revised as:
min &Sigma; l ij &Element; C y ij &Sigma; i &Element; A 1 , t &Element; V y it &Sigma; j &Element; A 2 , t &Element; V y jt ( 10 )
Consider on this basis active power, require the meritorious larger of circuit transmission, build about cutting apart the objective function of indication vector:
min &Sigma; i &Element; A 1 , j &Element; A 2 y ij | P ij | ( x i - x j ) 2 &Sigma; i &Element; A 1 , t &Element; V y it | P ij | &Sigma; j &Element; A 2 , t &Element; V y jt | P ij | - - - ( 11 )
With x iBe relaxed to real number field, according to Rayleigh theorem, x=v FThe time, formula (11) obtains minimum value.With gained x difference substitution formula (11) calculating target function value in the 3rd step.Be the combination of some circuits due to section, have 2 for the system that m bar branch road is arranged m-1 section, and the method that the present invention proposes is according to v FThe direction of proper vector is searched for section heuristicly, with 2 m-1 section is reduced to n-1, greatly reduces computation complexity.
In step 5, the screening objective function of sening as an envoy to is obtained the indication vector cut apart of minimum value, and the section of this vector correspondence is exactly fragile section.
By method provided by the present invention, utilization is composed the power method based on the electric system branch road of electrical distance and trend distribution, set up and be applicable to the nowadays more flexible and changeable interconnected network electric system graph theory model of the method for operation, and then utilize the standard blanking method to build about cutting apart the objective function of indication vector, rapidly and accurately in recognition system by a little less than contact on physical arrangement, and trend is carried the fragile section that larger circuit forms in system.The attention degree of raising to these sections has great importance to formulating effective prevention and control measure and off-the-line control strategy.
Further illustrate the present invention below in conjunction with the drawings and specific embodiments.
The fragile fractograph analysis of embodiment 1:IEEE30 node example system
The first step is calculated the electric network swim distribution situation, utilizes based on the electric system branch road of electrical distance and trend distribution and composes the power method, sets up applicable electric system weighted graph model.Table 1 has been listed the node output power in IEEE 30 node system embodiment, and line parameter circuit value and trend that table 2 has been listed IEEE 30 node systems distribute, and table 3 has been listed the weight of circuit in the graph model.
The node output power of table 1IEEE 30 node systems
Figure BDA00002739878900062
Line parameter circuit value and the trend of table 2IEEE 30 node systems distribute
Figure BDA00002739878900063
The weight of circuit in table 3IEEE 30 node system graph models
Figure BDA00002739878900072
Figure BDA00002739878900081
Second step is set up Laplacian Matrix and the degree matrix of electric system graph theory abstract model, and calculates Laplacian Matrix about minimum non-zero eigenwert and the characteristic of correspondence vector thereof of degree matrix.
Laplacian Matrix is about the minimum non-zero eigenvalue λ=0.0415 of degree matrix, characteristic of correspondence vector v F=[0.0651-0.0630-0.0607-0.0571-0.0623-0.0497-0.0577-0.0497-0.0548-0.0185-0.0572-0.0342-0.0357-0.0298-0.0261-0.0287-0.0248-0.0343-0.0353-0.0336 0.0658 0.11690.0641 0.1210 0.1214 0.1267 0.0979-0.0451 0.1111 0.1121] T
The 3rd step, in the direction that minimum non-zero eigenwert characteristic of correspondence vector provides, formation system all nodes corresponding cut apart the indication vector.
At first, make x 0Be complete 1 column vector of 30 dimensions, h=1; Then, find out v FIn the row at the little element of h place, make x h-1The element of middle corresponding row is-1, forms x h, h=h+1 repeats this step, until h=30 forms altogether 29 and cuts apart the indication vector.For example, find out v FMiddle least member-0.0651 place behavior 1 makes x 0The element of middle corresponding row is-1, forms to cut apart the indication vector x 1Find out v FIn the second little element-0.0630 place behavior 2, make x 1The element of middle corresponding row is-1, forms to cut apart the indication vector x 2, until form x 29
The 4th step: build about cutting apart the indication vector x hObjective function:
min &Sigma; i &Element; A 1 , j &Element; A 2 y ij | P ij | ( x i h - x j h ) 2 &Sigma; i &Element; A 1 , t &Element; V y it | P ij | &Sigma; j &Element; A 2 , t &Element; V y jt | P ij | - - - ( 11 )
And each is cut apart indication vector calculation target function value, as table 4.
The node output power of table 4 IEEE 30 node systems
Figure BDA00002739878900083
The 5th step: the screening objective function of sening as an envoy to is obtained the indication vector cut apart of minimum value, and the section of this vector correspondence is exactly fragile section.
The minimum value of objective function is 0.0631, h=21, and the corresponding indication vector of cutting apart is x 21=[1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1-1 111111 1-1 1 1] TThe section of this vector correspondence is by 4 circuit 10-21,10-22, and 15-23,28-27 forms, and 9 nodes (21,22,23,24,25,26,27,29,30) and 21 nodes that are left are separated.Dotted line red in Fig. 2 is the fragile section that this method identifies in IEEE 30 node systems.
The mean value of the meritorious absolute value of system line is 8.39MW, analyzes the mean value of active power of gained section take this method as 11.70MW.All the mean value than system line active power absolute value is large for the active power absolute value that transmits on all formation circuits of this section simultaneously, and average reactance is the average reactance 0.2p.u. greater than whole circuits, illustrates that the gained section also forms by connecting weak circuit.Above sample calculation analysis proof this method can effectively identify the fragile section that the heavier circuit of load forms by physical connection is weak.

Claims (5)

1. an electric system vulnerable profile recognition method of cutting based on standard, is characterized in that, comprises the steps:
The first step: according to the needs of fragile profile recognition, calculate the electrical network Real-time Power Flow, utilize based on the electric system branch road of electrical distance and trend distribution and compose the power method, set up applicable electric system graph theory model;
Second step: set up this abstract model Laplacian Matrix and degree matrix, and calculate Laplacian Matrix about minimum non-zero eigenwert and the characteristic of correspondence vector thereof of degree matrix;
The 3rd step: the direction that provides according to this proper vector is divided into two groups with the node in system, forms corresponding cutting apart and indicates vector, searches for the section of this vector correspondence heuristicly;
The 4th step: a little less than contact on the circuit physical arrangement in fragile section, and trend is carried larger characteristics in system, builds the objective function of indicating vector about cutting apart based on the standard blanking method, and each is cut apart indication vector calculation target function value;
The 5th step: the screening objective function of sening as an envoy to is obtained the indication vector cut apart of minimum value, and the section of this vector correspondence is exactly fragile section.
2. electric system vulnerable profile recognition method of cutting based on standard as claimed in claim 1, it is characterized in that, utilize the electric system branch road that distributes based on electrical distance and trend to compose the power method and be specially: the graph theory model G=(V, the E that contain n node and m bar limit, W), V is vertex set, and the bus in corresponding real system, generator, load, E are the limit collection, the branch road of corresponding connection bus, comprise circuit and transformer, W is weight matrix, weighting matrix elements w ijBe connected node v iAnd v jThe weight on limit; Utilization is composed the power method based on the electric system branch road of electrical distance and trend distribution branch road is composed power:
w ij = y ij | P ij | - - - ( 1 )
y ijThe line admittance of connected node i and node j, P ijTo flow to the meritorious trend of ground state of node j from node i.
3. electric system vulnerable profile recognition method of cutting based on standard as claimed in claim 1, is characterized in that, calculate Laplacian Matrix and be specially about minimum non-zero eigenwert and the characteristic of correspondence vector thereof of spending matrix: the degree of node i is:
d i = &Sigma; j = 1 n w ij - - - ( 2 )
The degree matrix D is with d 1... d nBe the diagonal matrix of diagonal element, Laplacian Matrix L is:
L=D-W (3)
Method for Solving Generalized Eigenproblem:
Lz=λDz (4)
λ representation feature value, the characteristic of correspondence vector is z, minimum non-zero eigenwert characteristic of correspondence vector v FExpression.
4. electric system vulnerable profile recognition method of cutting based on standard as claimed in claim 1, is characterized in that, cuts apart the indication vector:
Cut apart the indication vector:
x=(x 1,x 2,…x n) T∈R n (5)
x i = 1 , i &Element; A 1 - 1 , i &Element; A 2 - - - ( 6 )
A 1And A 2Represent that respectively will scheme G is divided into the node that two subgraphs comprise, and cuts apart the situation of cutting apart that the indication vector x has indicated figure, order
k=v F+α (7)
α be one by-∞ to+number that ∞ changes, form the corresponding indication vector x of cutting apart according to the symbol of element of k:
x i = 1 , k i &GreaterEqual; 0 - 1 , k i < 0 - - - ( 8 )
Get rid of the situations that in x, element all equates, be i.e. nonsensical cutting apart, can be formed to many n-1 and cut apart the indication vector x.
5. electric system vulnerable profile recognition method of cutting based on standard as claimed in claim 1, is characterized in that, the 4th step further was specially:
Fragile section C should satisfy as much as possible:
min &Sigma; l ij &Element; C y ij , max &Sigma; l ij &Element; C | P ij | - - - ( 9 )
l ijThe circuit of expression connected node i and node j, y ijCircuit l ijAdmittance, suppose that circuit can't harm, P ijCircuit l ijThe meritorious trend of ground state, (9) first of formulas are revised as:
min &Sigma; l ij &Element; C y ij &Sigma; i &Element; A 1 , t &Element; V y it &Sigma; j &Element; A 2 , t &Element; V y jt - - - ( 10 )
Consider on this basis active power, require the meritorious larger of circuit transmission, build about cutting apart the objective function of indication vector:
min &Sigma; i &Element; A 1 , j &Element; A 2 y ij | P ij | ( x i - x j ) 2 &Sigma; i &Element; A 1 , t &Element; V y it | P ij | &Sigma; j &Element; A 2 , t &Element; V y jt | P ij | - - - ( 11 )
With x iBe relaxed to real number field, according to Rayleigh theorem, x=v FThe time, formula (11) obtains minimum value; With gained x difference substitution formula (11) calculating target function value in the 3rd step.
CN201310014905XA 2013-01-15 2013-01-15 Electrical power system fragile section identification method based on normalized-cut Pending CN103093097A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310014905XA CN103093097A (en) 2013-01-15 2013-01-15 Electrical power system fragile section identification method based on normalized-cut

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310014905XA CN103093097A (en) 2013-01-15 2013-01-15 Electrical power system fragile section identification method based on normalized-cut

Publications (1)

Publication Number Publication Date
CN103093097A true CN103093097A (en) 2013-05-08

Family

ID=48205656

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310014905XA Pending CN103093097A (en) 2013-01-15 2013-01-15 Electrical power system fragile section identification method based on normalized-cut

Country Status (1)

Country Link
CN (1) CN103093097A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103259265A (en) * 2013-06-06 2013-08-21 山东大学 Power network partitioning and detached island identification method of intelligent power distribution network based on reachability matrix
CN103368176A (en) * 2013-07-22 2013-10-23 国家电网公司 Method for searching for dangerous power transmission sections
CN104269846A (en) * 2014-09-26 2015-01-07 华北电力大学 Power grid fault partitioning method
CN105356458A (en) * 2015-11-20 2016-02-24 广东电网有限责任公司电力调度控制中心 Grid safety stability margin description method based on weak section
CN108090667A (en) * 2017-12-13 2018-05-29 国网山东省电力公司枣庄供电公司 Regional new energy receives partition method, apparatus and system
CN108923427A (en) * 2018-08-13 2018-11-30 哈尔滨工程大学 A kind of reconstructing method of the ship power distribution network based on queue intelligent algorithm
CN109508734A (en) * 2018-10-26 2019-03-22 北京交通大学 Road network ability bottleneck identification method based on spectral clustering
CN109510245A (en) * 2019-01-03 2019-03-22 东北电力大学 A kind of electric system Coherent Generator Group discrimination method based on figure segmentation
CN112989129A (en) * 2021-03-18 2021-06-18 国家电网有限公司 Power transmission network key perception section identification method and system based on graph theory
CN117741345A (en) * 2023-12-21 2024-03-22 暨南大学 Method and system for quickly searching key power transmission sections of power grid

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
YUANYUAN LI ET.AL: "《Innovative Smart Grid Technologies-Asia》", 24 May 2012 *
曹一家 等: "基于复杂网络理论的大型电力***脆弱线路辨识", 《电力自动化设备》 *

Cited By (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103259265A (en) * 2013-06-06 2013-08-21 山东大学 Power network partitioning and detached island identification method of intelligent power distribution network based on reachability matrix
CN103368176A (en) * 2013-07-22 2013-10-23 国家电网公司 Method for searching for dangerous power transmission sections
CN103368176B (en) * 2013-07-22 2015-04-29 国家电网公司 Method for searching for dangerous power transmission sections
CN104269846A (en) * 2014-09-26 2015-01-07 华北电力大学 Power grid fault partitioning method
CN104269846B (en) * 2014-09-26 2016-08-17 华北电力大学 Electric network fault partition method
CN105356458A (en) * 2015-11-20 2016-02-24 广东电网有限责任公司电力调度控制中心 Grid safety stability margin description method based on weak section
CN105356458B (en) * 2015-11-20 2018-02-06 广东电网有限责任公司电力调度控制中心 A kind of electricity net safety stable nargin based on weak section describes method
CN108090667A (en) * 2017-12-13 2018-05-29 国网山东省电力公司枣庄供电公司 Regional new energy receives partition method, apparatus and system
CN108923427A (en) * 2018-08-13 2018-11-30 哈尔滨工程大学 A kind of reconstructing method of the ship power distribution network based on queue intelligent algorithm
CN108923427B (en) * 2018-08-13 2020-11-03 哈尔滨工程大学 Method for reconstructing ship power distribution network based on queue intelligent algorithm
CN109508734A (en) * 2018-10-26 2019-03-22 北京交通大学 Road network ability bottleneck identification method based on spectral clustering
CN109508734B (en) * 2018-10-26 2022-03-11 北京交通大学 Road network capacity bottleneck identification method based on spectral clustering
CN109510245A (en) * 2019-01-03 2019-03-22 东北电力大学 A kind of electric system Coherent Generator Group discrimination method based on figure segmentation
CN112989129A (en) * 2021-03-18 2021-06-18 国家电网有限公司 Power transmission network key perception section identification method and system based on graph theory
CN112989129B (en) * 2021-03-18 2024-05-24 国家电网有限公司 Power transmission network key perception section identification method and system based on graph theory
CN117741345A (en) * 2023-12-21 2024-03-22 暨南大学 Method and system for quickly searching key power transmission sections of power grid

Similar Documents

Publication Publication Date Title
CN103093097A (en) Electrical power system fragile section identification method based on normalized-cut
CN104617574B (en) A kind of appraisal procedure of power system load region Transient Voltage Stability
CN102934312B (en) Energy production system and control thereof
US20200212710A1 (en) Method for predicting operation state of power distribution network with distributed generations based on scene analysis
Xue et al. Impact of transmission tower-line interaction to the bulk power system during hurricane
CN100438251C (en) A large electric system vulnerable line identifying method
CN104318347A (en) Power transmission line icing state assessment method based on information fusion of multiple sensors
CN103606109B (en) A kind of safe operation of electric network risk integrative assessment method based on evaluation object
CN104392083A (en) Analysis method for power outage risk and reliability based on full-web topology
CN105354675A (en) Key transmission section identification-based cascading failure analysis method for alternating current/direct current power network
CN104715423A (en) Method for assessing risk and reliability of power distribution network
CN106327071A (en) Power line communication risk analysis method and power line communication risk analysis system
CN103246939A (en) Security and stability margin based on-line identification method for power network operating safety risk incidents
CN102930408B (en) A kind of 750kV electric grid secondary equipment state appraisal procedure based on information fusion
CN115545479A (en) Method and device for determining important nodes or important lines of power distribution network
CN106570582A (en) Method and system for building transmission line dancing tripping risk prediction network model
Garau et al. ICT reliability modelling in co-simulation of smart distribution networks
CN105046011A (en) Rapid analysis method used for state of electric device and based on topological computation of distributed power grid
CN111062569A (en) Low-current fault discrimination method based on BP neural network
CN108596450B (en) Power grid risk early warning method and system
Dong et al. Fault diagnosis and classification in photovoltaic systems using scada data
Hu et al. Operational reliability evaluation method based on big data technology
CN103001220B (en) Optimal matching loop power based method for identifying topological structures of power distribution network
CN111191867B (en) Reliability evaluation method for complex network of power system
CN106159940A (en) PMU optimum points distributing method based on network load specificity analysis

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C02 Deemed withdrawal of patent application after publication (patent law 2001)
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20130508