CN105046582A - Convenient power grid security risk evaluation method - Google Patents

Abstract

The present invention discloses a convenient power grid security risk evaluation method. The method comprises: performing layering on risk factors by using an analytic hierarchy process, and calculating a weight of influence of each risk factor on a system by using the analytic hierarchy process; and then using a D-S evidence theory to correct weight determination by the analytic hierarchy process, combining evidences by using an evidence theory combination rule, and performing ranking on security and effectiveness of each risk factor according to a combined basic reliability distribution function to finally obtain the security and effectiveness level of the system. The method is accurate and effective, and is convenient in evaluation.

Description

Grid security risk assessment method easily

The application is application number: the divisional application of 201510412517.6, the applying date: 2015.7.14, title " grid security risk assessment method ".

Technical field

The present invention relates to a kind of grid security risk assessment method.

Background technology

In recent years, the sustainable development of China's economic is filled with new vitality to the development of power industry, and need for electricity constantly increases, and electric network composition and scale strengthen day by day, and the complex characteristics of electric system also shows become clear day by day.This has higher requirement and challenges to the security and stability control of modern power systems.Under the large good prospect greatly developing intelligent grid, people are to ensureing that the safe and stable operation of electric system is more confident.But people have in the face of a fact---electric system large-area power-cuts risk exists all the time.By the analysis to a series of large-area power-cuts events in recent years, traditional safety evaluation method (comprising deterministic stability analysis and probabilistic appraisal procedure) all Shortcomings parts can be found out.Therefore, except continuing to improve except existing various safety and risk analytical approach, also more complete power system security risk evaluation system must be sought from practical angle.

Summary of the invention

The object of the present invention is to provide one grid security risk assessment method accurately and effectively.

Technical solution of the present invention is:

A kind of grid security risk assessment method, is characterized in that: comprise the following steps:

Step 1: the division carrying out on level to causing the evaluation index of power grid security, is divided into destination layer, rule layer, indicator layer; Determine the membership between layer and layer element;

Step 2: quantize evaluation index, structure multilevel iudge matrix, using last layer time certain element as comparison criterion, compares scale C with one _ijrepresent the relative importance of i-th element and a jth element in this level; C _ijvalue rule is:

C ij	Implication
		1	Index i and index j is of equal importance
3	Index i is more important a little than index j
		5	Index i is obviously more important than index j
7	Index i is much more important than index j
		9	Index i is extremely more important than index j
2，4，6，8	Between above-mentioned judgement
		Each inverse	Plan compares: the ratio of index j and index i importance: C ij＝1/C ji

Set up multilevel iudge Matrix C

Step 3: carry out consistency desired result to the judgment matrix C of structure, formula is as follows:

$CR=\frac{{\mathrm{\λ}}_{\max }-n}{RI(n-1)}$

Wherein, λ _maxbe the eigenvalue of maximum of judgment matrix, calculated by approximate root method; RI is Aver-age Random Consistency Index:

As CR < 0.1, think that matrix has comparatively satisfied consistance, go to step 6; If not just carry out reparation judgment matrix, go to step 4;

Step 4: repair judgment matrix

Step 4.1: by the Elements C in matrix _ijdivided by (wherein $\stackrel{\&OverBar;}{C_{ij}}=\frac{1}{n}{\&Sigma;}_{k=1}^n{C}_{ik}{C}_{kj}$ ), make variable $S_{ij}=C_{ij}/\stackrel{\&OverBar;}{C_{ij}};$

Step 4.2: if S _ij<1, and C _ij=9, then do not calculate deviation distance d _ijif, S _ij>1, and C _ij=(1/9), then do not calculate deviation distance d _ij, other situation all calculates deviation distance d _ij;

Step 4.3: compare maximum d _ij, and record the sequence number i of element and the value of j, get in 1 ~ 9 scale closest to C _ij/ S _ijnumber replace Elements C _ij;

Step 4.4: check the consistance after adjustment, if inconsistent, repeats above step by the matrix after adjustment;

Step 5: consider the impact of next hierarchical elements on last layer minor element, must calculate the combining weights of k+1 layer element, its computing formula is as follows:

${\mathrm{\&omega;}}^{k+1,k-1}={({\mathrm{\&omega;}}_1^{k+1,k},{\mathrm{\&omega;}}_2^{k+1,k}\mathrm{...}{\mathrm{\&omega;}}_n^{k+1,k})}^T{\left({\mathrm{\&omega;}}^{k,k-1}\right)}^T$

Wherein: ω ^{k+1, k-1}for the upper element of kth+1 layer is relative to the combining weights of element on k-1 layer; for the upper element of kth+1 layer is relative to the weight of element on k layer;

Step 6: determine to evaluate collection, make X={X ₁, X ₂x _kbe model Comment gathers, wherein X ₁, X ₂x _kfor concrete comment;

F (X)={ F (X ₁), F (X ₂), F (X ₃), F (X ₄), F (X ₅) the fuzzy evaluation value that provides for fuzzy evaluation collection, F (X _i) (i=1,2 ... 5) for corresponding to fuzzy evaluation value X _iassessed value, its scope is 0<=F (X _i) <=1;

Risk comment	Risk indicator fuzzy set	Value-at-risk
			Excessive risk	[0.8,1.0]	0.9
High risk	[0.6,0.8]	0.7
			Medium risk	[0.4,0.6]	0.5
Comparatively low-risk	[0.2,0.4]	0.3
			Low-risk	[0.0,0.2]	0.1

Step 7: the basic brief inference function m calculating indices according to assessment result in conjunction with combining weights _n(X _i), be included in Θ by unknown message, the basic brief inference value of burnt first Θ is simultaneously:

$m_n\left(\mathrm{\&Theta;}\right)=1-{\mathrm{\&Sigma;}}_{i=1}^n{m}_n\left(X_i\right);$

Step 8: according to Combination Rules of Evidence Theory, obtains the basic brief inference function of the indices synthesized, and carries out size sequence by its value; Suppose 2 evidence E under identification framework Θ ₁and E ₂, its corresponding basic brief inference function is m ₁and m ₂, burnt unit is respectively X _iand Y _j, utilize composition rule

In formula: it represents the conflict spectrum between each evidence; The like, can merge each bar evidence theory, obtain the basic brief inference value after merging, realize the rational evaluation to each index;

Step 9: according to ordering relation, can obtain the magnitude relationship of indices safety and effectiveness intuitively and classification, thus realize the safety and effectiveness evaluation to whole system; Index is divided by step theory, is divided into 5 ranks: 1. excellent; 2. good; 3. in; 4. pass; 5. poor; On the basis of the basic brief inference function of each index of aforementioned calculating, adopt credible degree recognition criterion, Assessment for classification is carried out to indices; Arranging reliability is θ (θ > 0.5), and its value is generally 0.6,0.7,0.8 and 0.9; Order the proportion μ shared in systems in which according to the basic brief inference value of indices _iwith magnitude relationship, judge the rank that each index belongs to.

The present invention is accurate and effective, convenient; By analyzing the every definition of grid security risk assessment method, the risk factors obtained System's composition threatens mainly can be divided into the out-of-limit rate of equipment operation ratio, equipment, equipment deficiency rate and relay protection to change.Then adopt analytical hierarchy process to carry out distinguishing hierarchy to risk factors, and calculate risk factors to the weight shared by systematic influence.The weight of D-S evidence theory modification level fractional analysis is finally utilized to determine, and use Combination Rules of Evidence Theory to merge each bar evidence, basic brief inference function according to merging carries out classification to the safety and effectiveness of each risk factors, finally draws the safety and effectiveness rank of system.

Grid security risk assessment method uses analytical hierarchy process to carry out distinguishing hierarchy, and calculates risk factors to the weight shared by systematic influence.The weight of D-S evidence theory modification level fractional analysis is utilized to determine, and use Combination Rules of Evidence Theory to merge each bar evidence, basic brief inference function according to merging carries out classification to the safety and effectiveness of each risk factors, finally draws the safety and effectiveness rank of system.

Accompanying drawing explanation

Below in conjunction with drawings and Examples, the invention will be further described.

Fig. 1 is present system block diagram.

Fig. 2 is power grid security risk assessment hierarchical chart.

Embodiment

By analyzing the data contained by the systems such as OPEN3000, OMS, carry out the safe operation situation of reflecting regional transformer station.

Assess the security risk index of transformer station mainly: the equipment operation ratio (switch junctions row, equipment are stopped transport) of transformer station; The out-of-limit rate of equipment (trend out-of-limit, voltage out-of-limit); Equipment deficiency rate (electrical equipment, automation equipment, transformer substation video) and relay protection change.Device status data in the basis EMS of wherein equipment operation ratio, the out-of-limit and out-of-limit data of remote signalling of the equipment in the basis EMS of the out-of-limit rate of equipment, the equipment deficiency in the basis OMS that equipment deficiency rate and relay protection change and relay protection change data.

This platform mainly carries out risk assessment to 220KV transformer station, 110KV transformer station, 35KV transformer station and responsible consumer.Guarantor's risk of responsible consumer is comprehensively analyzed according to the risk assessment mark of relevant transformer station.Wherein relevant one-level transformer station, accounting 70%, relevant secondary substation, accounting 30%.

For different transformer stations, as follows to the selection of index:

(1) 220kv system transformer station:

Equipment operation ratio: switch junctions row, equipment stop transport (interconnection, main transformer, bus)

The out-of-limit rate of equipment: trend is out-of-limit, voltage out-of-limit

Equipment deficiency rate: electrical equipment, automation equipment, transformer substation video

Relay protection changes

(2) 110KV system transformer station:

Equipment operation ratio: the mode of connection, equipment stop transport (interconnection, main transformer, bus)

The out-of-limit rate of equipment: trend is out-of-limit, voltage out-of-limit

Equipment deficiency rate: electrical equipment, automation equipment, transformer substation video

Relay protection changes

(3) 35KV system transformer station:

Equipment operation ratio: the mode of connection, equipment stop transport (interconnection, main transformer, bus)

The out-of-limit rate of equipment: trend is out-of-limit, voltage out-of-limit

Equipment deficiency rate: electrical equipment, automation equipment, transformer substation video

Relay protection changes

2. the method adopting analytical hierarchy process (AHP) and D-S evidence theory to be combined with each other is assessed power grid security risk.Analytical hierarchy process is in order to determine the weight of each assessment factor, and the weight of D-S evidence theory modification level fractional analysis is determined.Level methods of risk assessment, particular content is as follows:

Step 1: the division carrying out on level to causing the evaluation index of power grid security, is divided into destination layer, rule layer, indicator layer.Determine the membership between layer and layer element, as shown in Figure 1.

Step 2: quantize evaluation index, structure multilevel iudge matrix, using last layer time certain element as comparison criterion, compares scale C with one _ijrepresent the relative importance of i-th element and a jth element in this level.C _ijvalue rule is as shown in table 1.

C ij	Implication
		1	Index i and index j is of equal importance
3	Index i is more important a little than index j
		5	Index i is obviously more important than index j

7	Index i is much more important than index j
		9	Index i is extremely more important than index j
2，4，6，8	Between above-mentioned judgement
		Each inverse	Plan compares: the ratio of index j and index i importance: C ij＝1/C ji

Table 1 Elements C _ijvalue rule

Set up multilevel iudge Matrix C

Step 3: carry out consistency desired result to the judgment matrix C of structure, formula is as follows:

$CR=\frac{{\mathrm{\&lambda;}}_{\max }-n}{RI(n-1)}$

Wherein, λ _maxbe the eigenvalue of maximum of judgment matrix, calculated by approximate root method.RI is Aver-age Random Consistency Index, can be inquired about and can be obtained by table 2

The table 2 consistency check table of comparisons

As CR < 0.1, think that matrix has comparatively satisfied consistance, go to step 6.If not just carry out reparation judgment matrix, go to step 4.

Step 4: repair judgment matrix

Step 4.1: by the Elements C in matrix _ijdivided by (wherein $\stackrel{\&OverBar;}{C_{ij}}=\frac{1}{n}{\&Sigma;}_{k=1}^n{C}_{ik}{C}_{kj}$ ), make variable $S_{ij}=C_{ij}/\stackrel{\&OverBar;}{C_{ij}};$

Step 4.2: if S _ij<1, and C _ij=9, then do not calculate deviation distance d _ijif, S _ij>1, and C _ij=(1/9), then do not calculate deviation distance d _ij, other situation all calculates deviation distance d _ij;

Step 4.3: compare maximum d _ij, and record the sequence number i of element and the value of j, get in 1 ~ 9 scale closest to C _ij/ S _ijnumber replace Elements C _ij;

Step 4.4: check the consistance after adjustment, if inconsistent, repeats above step by the matrix after adjustment.

Step 5: consider the impact of next hierarchical elements on last layer minor element, must calculate the combining weights of k+1 layer element, its computing formula is as follows:

${\mathrm{\&omega;}}^{k+1,k-1}={({\mathrm{\&omega;}}_1^{k+1,k},{\mathrm{\&omega;}}_2^{k+1,k}...{\mathrm{\&omega;}}_n^{k+1,k})}^T{\left({\mathrm{\&omega;}}^{k,k-1}\right)}^T$

Wherein: ω ^{k+1, k-1}for the upper element of kth+1 layer is relative to the combining weights of element on k-1 layer; for the upper element of kth+1 layer is relative to the weight of element on k layer.

Step 6: determine to evaluate collection, analyzes data by expert according to professional knowledge and provides, make X={X ₁, X ₂x _kbe model Comment gathers, wherein X ₁, X ₂x _kfor concrete comment, as shown in table 3.F (X)={ F (X ₁), F (X ₂), F (X ₃), F (X ₄), F (X ₅) the fuzzy evaluation value that provides for fuzzy evaluation collection, F (X _i) (i=1,2 ... 5) for corresponding to fuzzy evaluation value X _iassessed value, its scope is 0<=F (X _i) <=1.

Risk comment	Risk indicator fuzzy set	Value-at-risk
			Excessive risk	[0.8,1.0]	0.9
High risk	[0.6,0.8]	0.7

Medium risk	[0.4,0.6]	0.5
			Comparatively low-risk	[0.2,0.4]	0.3
Low-risk	[0.0,0.2]	0.1

Table 3 risk assessment collection

Step 7: the basic brief inference function m calculating indices according to the assessment result of expert in conjunction with combining weights _n(X _i), be included in Θ by unknown message, the basic brief inference value of burnt first Θ is simultaneously: $m_n\left(\mathrm{\&Theta;}\right)=1-{\mathrm{\&Sigma;}}_{i=1}^n{m}_n\left(X_i\right).$

Step 8: according to Combination Rules of Evidence Theory, obtains the basic brief inference function of the indices synthesized, and carries out size sequence by its value.Suppose 2 evidence E under identification framework Θ ₁and E ₂, its corresponding basic brief inference function is m ₁and m ₂, burnt unit is respectively X _iand Y _j, utilize composition rule

In formula: it represents the conflict spectrum between each evidence.The like, can merge each bar evidence theory, obtain the basic brief inference value after merging, realize the rational evaluation to each index.

Step 9: according to ordering relation, can obtain the magnitude relationship of indices safety and effectiveness intuitively and classification, thus realize the safety and effectiveness evaluation to whole system.Index is divided by step theory, is divided into 5 ranks: 1. excellent; 2. good; 3. in; 4. pass; 5. poor.On the basis of the basic brief inference function of each index of aforementioned calculating, adopt credible degree recognition criterion, Assessment for classification is carried out to indices.Arranging reliability is θ (θ > 0.5), and its value is generally 0.6,0.7,0.8 and 0.9.Order the proportion μ shared in systems in which according to the basic brief inference value of indices _iwith magnitude relationship, judge the rank that each index belongs to.

Claims (1)

1. a grid security risk assessment method easily, is characterized in that: comprise the following steps:

Step 1: the division carrying out on level to causing the evaluation index of power grid security, is divided into destination layer, rule layer, indicator layer; Determine the membership between layer and layer element;

Step 2: quantize evaluation index, structure multilevel iudge matrix, using last layer time certain element as comparison criterion, compares scale C with one _ijrepresent the relative importance of i-th element and a jth element in this level; C _ijvalue rule is:

C _ij Implication 1 Index i and index j is of equal importance 3 Index i is more important a little than index j 5 Index i is obviously more important than index j 7 Index i is much more important than index j 9 Index i is extremely more important than index j 2，4，6，8 Between above-mentioned judgement Each inverse Plan compares: the ratio of index j and index i importance: C _ij＝1/C _ji

Set up multilevel iudge Matrix C

Step 3: carry out consistency desired result to the judgment matrix C of structure, formula is as follows:

$CR=\frac{{\mathrm{\&lambda;}}_{\max }-n}{RI(n-1)}$

Wherein, λ _maxbe the eigenvalue of maximum of judgment matrix, calculated by approximate root method; RI is Aver-age Random Consistency Index:

As CR < 0.1, think that matrix has comparatively satisfied consistance, go to step 6; If not just carry out reparation judgment matrix, go to step 4;

Step 4: repair judgment matrix

Step 4.1: by the Elements C in matrix _ijdivided by (wherein $\stackrel{\&OverBar;}{C_{ij}}=\frac{1}{n}{\&Sigma;}_{k=1}^n{C}_{ik}{C}_{kj}$ ), make variable $S_{ij}=C_{ij}/\stackrel{\&OverBar;}{C_{ij}};$

Step 4.2: if S _ij<1, and C _ij=9, then do not calculate deviation distance d _ijif, S _ij>1, and C _ij=(1/9), then do not calculate deviation distance d _ij, other situation all calculates deviation distance d _ij;

Step 4.3: compare maximum d _ij, and record the sequence number i of element and the value of j, get in 1 ~ 9 scale closest to C _ij/ S _ijnumber replace Elements C _ij;

Step 4.4: check the consistance after adjustment, if inconsistent, repeats above step by the matrix after adjustment;

Step 5: consider the impact of next hierarchical elements on last layer minor element, must calculate the combining weights of k+1 layer element, its computing formula is as follows:

${\mathrm{\&omega;}}^{k+1,k-1}={({\mathrm{\&omega;}}_1^{k+1,k},{\mathrm{\&omega;}}_2^{k+1,k}\mathrm{...}{\mathrm{\&omega;}}_n^{k+1,k})}^T{\left({\mathrm{\&omega;}}^{k,k-1}\right)}^T$

Wherein: ω ^{k+1, k-1}for the upper element of kth+1 layer is relative to the combining weights of element on k-1 layer; for the upper element of kth+1 layer is relative to the weight of element on k layer;

Step 6: determine to evaluate collection, make X={X ₁, X ₂x _kbe model Comment gathers, wherein X ₁, X ₂x _kfor concrete comment;

F (X)={ F (X ₁), F (X ₂), F (X ₃), F (X ₄), F (X ₅) the fuzzy evaluation value that provides for fuzzy evaluation collection, F (X _i) (i=1,2 ... 5) for corresponding to fuzzy evaluation value X _iassessed value, its scope is 0<=F (X _i) <=1;

Risk comment Risk indicator fuzzy set Value-at-risk Excessive risk [0.8,1.0] 0.9 High risk [0.6,0.8] 0.7 Medium risk [0.4,0.6] 0.5 Comparatively low-risk [0.2,0.4] 0.3 Low-risk [0.0,0.2] 0.1

Step 7: the basic brief inference function m calculating indices according to assessment result in conjunction with combining weights _n(X _i), be included in Θ by unknown message, the basic brief inference value of burnt first Θ is simultaneously:

$m_n\left(\mathrm{\&Theta;}\right)=1-{\&Sigma;}_{i=1}^n{m}_n\left(X_i\right);$

Step 8: according to Combination Rules of Evidence Theory, obtains the basic brief inference function of the indices synthesized, and carries out size sequence by its value; Suppose 2 evidence E under identification framework Θ ₁and E ₂, its corresponding basic brief inference function is m ₁and m ₂, burnt unit is respectively X _iand Y _j, utilize composition rule

In formula: it represents the conflict spectrum between each evidence; The like, can merge each bar evidence theory, obtain the basic brief inference value after merging, realize the rational evaluation to each index;

Step 9: according to ordering relation, can obtain the magnitude relationship of indices safety and effectiveness intuitively and classification, thus realize the safety and effectiveness evaluation to whole system; Index is divided by step theory, is divided into 5 ranks: 1. excellent; 2. good; 3. in; 4. pass; 5. poor; On the basis of the basic brief inference function of each index of aforementioned calculating, adopt credible degree recognition criterion, Assessment for classification is carried out to indices; Arranging reliability is θ (θ > 0.5), and its value is generally 0.6,0.7,0.8 and 0.9; Order the proportion μ shared in systems in which according to the basic brief inference value of indices _iwith magnitude relationship, judge the rank that each index belongs to;

Described evaluation index to 35KV system transformer station is:

Equipment operation ratio: the mode of connection, equipment are stopped transport;

The out-of-limit rate of equipment: trend is out-of-limit, voltage out-of-limit;

Equipment deficiency rate: electrical equipment, automation equipment, transformer substation video;

Relay protection changes.