CN109523101A - A kind of distribution Running State fuzzy synthetic appraisement method - Google Patents
A kind of distribution Running State fuzzy synthetic appraisement method Download PDFInfo
- Publication number
- CN109523101A CN109523101A CN201710842212.8A CN201710842212A CN109523101A CN 109523101 A CN109523101 A CN 109523101A CN 201710842212 A CN201710842212 A CN 201710842212A CN 109523101 A CN109523101 A CN 109523101A
- Authority
- CN
- China
- Prior art keywords
- index
- weight
- running state
- matrix
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000011156 evaluation Methods 0.000 claims abstract description 34
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 7
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 7
- 238000012216 screening Methods 0.000 claims description 27
- 239000011159 matrix material Substances 0.000 claims description 26
- 239000013598 vector Substances 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 9
- 238000013210 evaluation model Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000001419 dependent effect Effects 0.000 claims description 6
- 241000826860 Trapezium Species 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 4
- 238000013459 approach Methods 0.000 claims description 3
- 230000007547 defect Effects 0.000 abstract description 3
- 230000008901 benefit Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 230000001105 regulatory effect Effects 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009885 systemic effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of distribution Running State fuzzy synthetic appraisement method, belong to power distribution network synthesis evaluation field, the present invention provides a kind of distribution Running State model of fuzzy synthetic evaluation based on cooperative game method and trapezoidal cloud model, traditional fuzzy integrated evaluating method is solved to fail to fully consider the defect of randomness and ambiguity, and the boundary problem really up to the mark of interval division, the accuracy that weight can be improved avoids determining the case where weight judge mode can not reflect true distribution Running State.
Description
Technical field
The present invention relates to power distribution network synthesis to evaluate field, be specifically be related to a kind of distribution Running State obscure it is comprehensive
Close evaluation method.
Background technique
Power distribution network is located at electric system end, is the important link that power supply and user are contacted in electric system.Safety,
Reliable power distribution network is to ensure national economy sustainable and stable development and the important material base that living standards of the people are continuously improved.
Scientific and rational distribution network planning is the important prerequisite for ensureing power grid operation.China's power distribution network operation level and resource benefit
It is low with rate, in order to improve the operation conditions of power distribution network, overall merit is carried out to power distribution network first, and then instruct the planning of power distribution network
Construction.
Common integrated evaluating method has Field Using Fuzzy Comprehensive Assessment and analytic hierarchy process (AHP) etc., and the key of Field Using Fuzzy Comprehensive Assessment is
The determination of subordinating degree function, ununified standard.Traditional comprehensive evaluation model is not according to power distribution network characteristic to index
Weight and evaluation method are targetedly analyzed and are handled, and the randomness for fully considering electric distribution network data collection and mould are also failed to
Paste property；The determination of weight is mainly the Evaluation formula for utilizing introducing regulatory factor in traditional evaluation model, and regulatory factor
Value be it is empirically determined, it is subjective, cause weight accuracy to decline, and the weight that the method determines is fixed power
Weight, when certain index values deviate normal value, often indicates power distribution network part in power distribution network synthesis assessment indicator system
Performance decline, need to reinforce patrolling, but may be smaller because of its weight in fixed power judge mode, the overall evaluation or normal,
It cannot reflect the time of day of power distribution network.
Cloud model will realize the conversion between qualitative and quantitative, and common cloud model has normal cloud model, trapezoidal cloud model,
Normal cloud model can only describe the case where desired value is a value, and trapezoidal cloud model can describe desired value not just value
Situation can also describe the case where desired value is a section；Cloud model has become multi-level evaluation, multi objective and evaluation and refers to
Target description has the powerful of the Comprehensive Evaluation Problem of very strong ambiguity and randomness.
Summary of the invention
It is comprehensive that the purpose of the present invention proposes that a kind of distribution Running State based on cooperative game method and trapezoidal cloud model obscures
Evaluation model is closed, traditional fuzzy integrated evaluating method is solved and fails to fully consider defect and the section of randomness and ambiguity
The boundary of division problem really up to the mark, can be improved the accuracy of weight, avoid determining weight and judge mode can not reflecting true distribution
The case where Running State.
The present invention solves the above problems by the following technical programs:
A kind of distribution Running State fuzzy synthetic appraisement method, includes the following steps:
Step 1: being matched according to the building of the purpose of power distribution network, systematicness, operability, independence, conspicuousness and dynamic
Power grid initial indication system；
Step 2: preliminary screening being carried out to initial indication system and obtains preliminary screening index system；
Step 3: postsearch screening being carried out to preliminary screening index system, obtains determining index system；
Step 4: calculating separately to obtain three weights using consistent matrix analytic approach, entropy assessment, anti-entropy assessment；
Step 5: being combined according to these three weights using cooperative game algorithm and seek optimal weights W；
Step 6: statistics calculates terraced cloud model parameter；
Step 7: establishing fuzzy evaluating matrix；
Step 8: design synthesis evaluation model；
Step 9: differentiating state of electric distribution network grade；
Step 10: determining evaluation result using degree of membership maximum principle.
In above scheme, the detailed process of preliminary screening preferably in step 3 are as follows:
The related coefficient between each index is calculated, the index set by preliminary screening shares n (n=57) a index, rij
For the related coefficient between i-th of index and jth index；xit、xjtIt is i-th respectively, t-th of observation (t=of j index
(1,2 ..., k)) then
Wherein, K is positive integer,
Then according to needing to choose critical correlation coefficients M, M=0.9, | rij| > 0.9 illustrates to reflect between two indices
Information repeats, and is considered to delete one of index according to observation etc.；|rij| < 0.9 retains two indices simultaneously；Secondly
Collinearity diagnostics are carried out using variance inflation factor, delete multicollinearity index；The variance inflation factor of i-th of index is denoted as
(VIF)i, i=(1,2 ..., n), R2 be by dependent variable of i-th of index using other n-1 index as independent variable when, dependent variable
To the coefficient of determination that independent variable returns, then own (VIF)iIn maximum value be usually used to as multicollinearity severity
Index, (VIF)iWhen >=10, index has serious multiple correlation, deletes index i.
In above scheme, the detailed process of postsearch screening preferably in step 5 are as follows:
The corresponding related coefficient of index set after screening constitutes correlation matrix R, and the characteristic value of calculating matrix R calculates
Obtain the p eigenvalue λ greater than 01≥λ2≥…≥λp>=0, and acquire corresponding feature vector are as follows: V=(v1,v1,…,vq),
Then p principal component are as follows:
Calculate the variance contribution ratio and accumulative variance contribution ratio of principal component: the wherein value of m are as follows:1,2,
3 ... p, according to retain data information universal criterious, as ρ >=85%, retain before s principal component, that is, s=m, calculating it is main at
Molecular group load matrix B (b1,…,bq)=(bij)q×s, calculation formulaWherein λiIt is characteristic value, ViIt is its correspondence
Feature vector, BiIt is variable X and principal component yiPhase relation ordered series of numbers, wherein load bijIt is variable xiWith principal component yiCorrelation
Coefficient, according to absolute value | bij| screening index, absolute value | bij| show more greatly and principal component yiMore related, corresponding index is answered
The reservation, it is smaller instead to delete.
In above scheme, the algorithm of cooperative game algorithm preferably in step 6 are as follows:
Step 6.1: calculating the consistency related coefficient L of W (i) and W (k_i)i,
Wherein, i is the weight method calculated, and W (i) is i-th kind of weight, Wj(i) method to calculate weight using i-th kind
The weight of j-th of the index calculated, W (k_i) are k-1 kind weight W (1) ..., W (i-1), W (i+1) ..., W other than W (i)
(k) combining weights；K is weight middle finger target number；"-" expression is averaged；
Step 6.2: seek combining weights W':
Step 6.3: using recursive call, weight number of every calling subtracts 1, until the number of weight is equal to 2；
Step 6.4: when weight number is equal to 2,
Wherein, W (1) and W (2) indicates the 1/2nd weight；
Step 6.5: W' being normalized to obtain weight W.
In above scheme, the algorithm of terraced cloud model parameter preferably in step 7 are as follows:
Step 7.1: judge index value x, if x belongs to [Ex1,Ex2], then degree of membership μ=1；If x < Ex1, then Ex=Ex1；
If x > Ex2, then Ex=Ex2；Wherein [Ex1,Ex2] it is desired section, EnFor entropy and HeIt is super entropy for indicating whole quantitative performance；
Step 7.2: calculate degree of membership:
E ` in formulanIt is with En for expectation, HeThe normal random number generated for standard deviation.
In above scheme, preferably fuzzy evaluating matrix in step 8 are as follows:
Wherein, rijFor the related coefficient between i-th of index and jth index.
In above scheme, the process of comprehensive evaluation model is preferably established in step 9:
Step 9.1: determining evaluation indice, if first layer index has m index, be denoted as U={ u1,u2,…,um, according to
Attaching relation is divided into k subset, then U={ U1,U2,…,Uk, umFor index；
Step 9.2: establishing opinion rating collection, opinion rating collection is evaluation criterion set, is equipped with l evaluation criterion, that is, comments
Valence collection V={ v1, v2,…,vl, V={ v1=poor, v2=general, v3=medium, v4=good, v5=outstanding；
Step 9.3: determining index weights, the side combined using the cooperative game method and variable-weight theory that propose in step 6
Method parameter weight；The weight for calculating first layer index, obtains k weight vectorsWherein i=1,2,
3 ..., k；
Step 9.4: determining subordinated-degree matrix, calculated according to the statistics of the historical data of power distribution network index and cloud model parameter
The trapezium cloud model parameter of each index corresponding grade is calculated in method, and is calculated according to trapezoidal cloud generator algorithm each
The degree of membership of grade constitutes subordinated-degree matrix Ri；
Step 9.5: calculating assessment vector；
Step 9.6: calculating comprehensive assessment vector, second layer index weights vector w=[w1,w2,…,wk], the degree of membership of U
Matrix R=[B1,B2,…,Bk], final assessment vector are as follows:
B=wR=[b1,b2,…,bl]
Complete evaluation.
The advantages and effects of the present invention are:
The present invention provides a kind of distribution Running State fuzzy overall evaluation based on cooperative game method and trapezoidal cloud model
Model solves traditional fuzzy integrated evaluating method and fails to fully consider the defect of randomness and ambiguity and interval division
Boundary problem really up to the mark, can be improved the accuracy of weight, avoid determining weight and judge mode can not reflecting true power distribution network operation
The case where state.
Detailed description of the invention
Fig. 1 is evaluation rubric figure of the present invention；
Fig. 2 is power distribution network evaluation of running status index system figure of the present invention；
Specific embodiment
The invention will be further described with reference to embodiments.
A kind of distribution Running State fuzzy synthetic appraisement method, as shown in Figure 1,
Step 1: building power distribution network index system.
Follow the building that following six big principles carry out index system:
(1) purpose, selecting for power distribution network index has to comply with same evaluation purpose, and index can accurately describe
The feature of distribution network system is that guiding carrys out index for selection with purpose；In the present invention with the overall operation state of power distribution network be evaluation
Purpose is guiding to choose evaluation index.
(2) systemic, index system can embody the main feature of power distribution network on the whole, and the hierarchical structure of index wants bright
Really, clear；In order to fully demonstrate the hierarchical structure and globality of power distribution network index system, power distribution network is constructed using analytic hierarchy process (AHP)
Index system.
(3) operability, the base values of the power distribution network of selection must be collectable, and acquisition cost should not be too large,
The complexity of index collection is mainly considered in selective goal.
(4) independence, should be mutually indepedent between each index, but since the complexity of power distribution network index system determines respectively
It is impossible for being completely independent between a index, keeps the independence between each index with maximum likelihood, analysis indexes it
Between correlation method it is relatively independent as far as possible between evaluation index to guarantee, utmostly guarantee to refer to using correlation analysis
Mark independence.
(5) conspicuousness is not that index number is The more the better when constructing index system, and index number is mostly led with regard to meaning
A possibility that causing data redundancy is bigger；The conspicuousness of power distribution network index system is embodied using main critical evaluation index, benefit
Conspicuousness screening is carried out to index with Principal Component Analysis.
(6) dynamic, as the transformation and evaluation purpose of power distribution network change, index system needs to make corresponding tune
It is whole, the index system of building is adjusted by reasonableness test and evaluation feedback.
It follows on the basis of the above principle from safety, quality, economy, intelligence, the feature of environmental protection and sustainability six
A aspect constructs power distribution network synthesis assessment indicator system.1 selection base values collection is as shown in table 2 on principle, and then basis can
Operability carries out preliminary screening to base values collection, then results of preliminary screening such as table 2 carries out the index set after screening related
Property analysis: firstly, calculating the related coefficient between each index, the index set by preliminary screening shares n (n=57) a finger
Mark, rijFor the related coefficient between i-th of index and jth index；xit、xjtIt is i-th respectively, t-th of observation of j index
(t=(1,2 ..., k)) then
Then according to needing to choose critical correlation coefficients M, M=0.9 is taken in this patent, | rij| > 0.9 illustrates two indices
Between the information that reflects repeat, considered to delete one of index according to observation etc.；|rij| < 0.9 retains two simultaneouslyA index；Secondly collinearity diagnostics are carried out using variance inflation factor, deletes multicollinearity index.I-th
The variance inflation factor of a index is denoted as (VIF) i, i=(1,2 ..., n), and it is dependent variable with other n-1 that R2, which is using i-th of index,
When a index is independent variable, the coefficient of determination that dependent variable returns independent variable, then the maximum value in all (VIF) i usually by with
As multicollinearity severity, it is considered that, when (VIF) i >=10, index has serious multiple correlation, deletes
Except index i.Secondly, carrying out conspicuousness screening using Principal Component Analysis to the index set after further screening: firstly, will screening
The corresponding related coefficient of index set afterwards constitutes correlation matrix R, and the p greater than 0 is calculated in the characteristic value of calculating matrix R
A eigenvalue λ1≥λ2>=... p >=0 >=λ, and acquire corresponding feature vector are as follows: V=(v1,v1,…,vq), then p principal component
Are as follows:
It calculates the variance contribution ratio of principal component and tires outCount variance contribution ratio: the wherein value of m are as follows: 1,
2,3 ... p, according to the universal criterious for retaining data information, as ρ >=85%, s principal component, that is, s=m before retaining.Calculate master
Components Factor load matrix B (b1 ..., bq)=(bij) q × s, calculation formulaWherein λ i is characteristic value, and Vi is
Its corresponding feature vector, Bi are the phase relation ordered series of numbers of variable X Yu principal component yi, and wherein load bij is variable xi and principal component
The related coefficient of yi.According to absolute value | bij | screening index.Absolute value | bij | show more greatly it is more related to principal component yi, it is right
The index answered should retain, smaller instead to delete.Index set after above step deletes choosing is as shown in table 2, specifically
Index system is as shown in Fig. 2.
Step 2: parameter weight.
Weight W (1), the W (2), W for calculating separately consistent matrix analytic approach, entropy assessment, anti-entropy assessment in the present invention
(3) it is combined using cooperative game and seeks optimal weights W, be then modified to obtain variable weight W using variable-weight theory, as
Final weight.
It is given a mark according to expert to the relative importance between index, scoring criterion is as shown in table 3, obtains relatively important journey
Spend Judgement Matricies A=(aij) n × n.It enablesObtained consistent matrix B=(bij) n × n, B meet
Bii=1, bij=1/bji, bij=bikbkj；The formula of consistent matrix calculating weight are as follows:
Wherein, (i=1,2,3 ..., n)
WhereinI=1,2,3 ..., n.Weight W (1) is calculated according to formula 6.By the operation number of index
It is obtained weight W (2) according to substitution entropy assessment and anti-entropy assessment calculation formula, W (3).By weight W (1), W (2), W (3)
Determine that the input of the algorithm of combining weights iterates to calculate to obtain combining weights W as cooperative game method.By combining weights W by becoming
Power formula is corrected to obtain final weight W, the input as step 3.Illustrate calculating process by taking safety U1 subordinate's index as an example:
Carrying out relative importance by seven base values of the expert to safety subordinate with table 3 is scale, is obtained
Judgment matrix Au1, its weight, which is calculated, using formula is
Wu1(1)={ wr 11,wr 12,wr 13,wr 14,wr 15,wr 16,wr 17, it is W that weight, which is calculated, using entropy assessmentu1(2)=
{wh 11,wh 12,wh 13,wh 14,wh 15,wh 16,wh 17, it is W that weight, which is calculated, in anti-entropy assessmentu1(3)={ we 11,we 12,we 13,we 14,
we 15,we 16,we 17, cooperative game Evaluation formula algorithm is realized with matlab, by Wu1(1),Wu1(2),Wu1(3) as defeated
Enter, obtains combining weights Wu1={ w11,w12,w13,w14,w15,w16,w17, combining weights Wu1 is passed through into variable weight formulaThe amendment of α=0 is wherein taken to obtain final weight.Similarly obtain the weight of other indexs.
Step 3: evaluation model of the design based on trapezoidal cloud model
It is corresponding that each opinion rating is obtained by the analysis of historical data statistics and index ideal value to power distribution network index
Cloud model parameter, the statistical formula of trapezium cloud model parameter is as follows:
Wherein ExjFor the assembly average of index j, the group number of m achievement data acquisition, xijFor index collection data.Pass through statistics
Obtained cloud model parameter isWherein i is index, and k is opinion rating, then will be to be evaluated
The achievement data of valence is passed to the x condition trapezium cloud model generator of corresponding index respectively (program of trapezium cloud model algorithm is realized)
Degree of membership is calculated, constitutes Membership Vestor Ri and is configured to subordinated-degree matrix R according to the classification situation of index, is counted by step 2
Calculation obtains final weight vectors W.The degree of membership of upper one layer of index is calculated using the formula in fuzzy overall evaluation principle
Final membership vector B={ B1, B2, B3, B4, B5 } is successively calculated in the specific formula of vector such as formula 7, then using most
Big degree of membership principle judgement and evaluation grade instructs further distribution network planning the reason of analysis according to rating level in these level
Draw construction.
The division of evaluate collection is data definition domain to be divided into multiple discrete adjacent sections in fuzzy overall evaluation,
Interval border value determines very much, causes interval border really up to the mark.Synthesis is carried out to power distribution network using traditional fuzzy comprehensive evaluation to comment
When valence, for quantitative targets such as rate of qualified voltage, frequency qualification rates, subordinating degree function is the distribution situation according to index value
Determining, ununified design standard, the method can only consider the ambiguity of index, have ignored randomness.Cloud model is one
Kind realize the uncertainty models of the conversion between qualitativing concept and quantitative value, cloud model there are three numerical characteristic Ex, En,
He, cloud model consider the ambiguity and randomness of index relative status.Normal cloud model can only meet dull index properties,
Not being able to satisfy whole power distribution network index distribution characters, ((it is ideal for such as direct index (index value is the bigger the better), osculant index
Value is median or intermediate section), negative index (value is the smaller the better)).
1 distribution Running State descriptive grade of table
2 power distribution network index system the selection result of table
The preferred embodiment of the present invention has been described in detail above, but the present invention is not limited to embodiment,
Those skilled in the art can also make various equivalent modifications on the premise of not violating the inventive spirit of the present invention
Or replacement, these equivalent variation or replacement are all contained in scope of the present application.
Claims (7)
1. a kind of distribution Running State fuzzy synthetic appraisement method, which comprises the steps of:
Step 1: power distribution network is constructed according to the purpose of power distribution network, systematicness, operability, independence, conspicuousness and dynamic
Initial indication system；
Step 2: preliminary screening being carried out to initial indication system and obtains preliminary screening index system；
Step 3: postsearch screening being carried out to preliminary screening index system, obtains determining index system；
Step 4: calculating separately to obtain three weights using consistent matrix analytic approach, entropy assessment, anti-entropy assessment；
Step 5: being combined according to these three weights using cooperative game algorithm and seek optimal weights W；
Step 6: statistics calculates terraced cloud model parameter；
Step 7: establishing fuzzy evaluating matrix；
Step 8: design synthesis evaluation model；
Step 9: differentiating state of electric distribution network grade；
Step 10: determining evaluation result using degree of membership maximum principle.
2. a kind of distribution Running State fuzzy synthetic appraisement method according to claim 1, it is characterised in that: the step
The detailed process of preliminary screening in rapid 3 are as follows:
The related coefficient between each index is calculated, the index set by preliminary screening shares n (n=57) a index, rijIt is i-th
Related coefficient between a index and jth index；xit、xjtI-th respectively, t-th of observation of j index (t=(1,
2 ..., k)) then
Wherein, K is positive integer,
Then according to needing to choose critical correlation coefficients M, M=0.9, | rij| > 0.9 information for illustrating to reflect between two indices
It repeats, is considered to delete one of index according to observation etc.；|rij| < 0.9 retains two indices simultaneously；Secondly it utilizes
Variance inflation factor carries out collinearity diagnostics, deletes multicollinearity index；The variance inflation factor of i-th of index is denoted as
(VIF)i, i=(1,2 ..., n), R2 be by dependent variable of i-th of index using other n-1 index as independent variable when, dependent variable
To the coefficient of determination that independent variable returns, then own (VIF)iIn maximum value be usually used to as multicollinearity severity
Index, (VIF)iWhen >=10, index has serious multiple correlation, deletes index i.
3. a kind of distribution Running State fuzzy synthetic appraisement method according to claim 1, it is characterised in that: the step
The detailed process of postsearch screening in rapid 5 are as follows:
The corresponding related coefficient of index set after screening constitutes correlation matrix R, and the characteristic value of calculating matrix R is calculated
The p eigenvalue λ greater than 01≥λ2≥…≥λp>=0, and acquire corresponding feature vector are as follows: V=(v1,v1,…,vq), then p
Principal component are as follows:
Calculate the variance contribution ratio and accumulative variance contribution ratio of principal component: the wherein value of m are as follows:
According to the universal criterious for retaining data information, as ρ >=85%, s principal component, that is, s=m before retaining calculates the principal component factor
Load matrix B (b1,…,bq)=(bij)q×s, calculation formulaWherein λiIt is characteristic value, ViIt is its corresponding feature
Vector, BiIt is variable X and principal component yiPhase relation ordered series of numbers, wherein load bijIt is variable xiWith principal component yiRelated coefficient, root
According to absolute value | bij| screening index, absolute value | bij| show more greatly and principal component yiMore related, corresponding index should retain,
It is smaller instead to delete.
4. a kind of distribution Running State fuzzy synthetic appraisement method according to claim 1, it is characterised in that: the step
The algorithm of cooperative game algorithm in rapid 6 are as follows:
Step 6.1: calculating the consistency related coefficient L of W (i) and W (k_i)i,
Wherein, i is the weight method calculated, and W (i) is i-th kind of weight, Wj(i) method to be calculated weight using i-th kind is calculated
J-th of index weight, W (k_i) is k-1 kind weight W (1) ..., W (i-1), W (i+1) ..., W (k) other than W (i)
Combining weights；K is weight middle finger target number；"-" expression is averaged；
Step 6.2: seek combining weights W':
Step 6.3: using recursive call, weight number of every calling subtracts 1, until the number of weight is equal to 2；
Step 6.4: when weight number is equal to 2,
Wherein, W (1) and W (2) indicates the 1/2nd weight；
Step 6.5: W' being normalized to obtain weight W.
5. a kind of distribution Running State fuzzy synthetic appraisement method according to claim 1, it is characterised in that: the step
The algorithm of terraced cloud model parameter in rapid 7 are as follows:
Step 7.1: judge index value x, if x belongs to [Ex1,Ex2], then degree of membership μ=1；If x < Ex1, then Ex=Ex1；If
x>Ex2, then Ex=Ex2；Wherein [Ex1,Ex2] it is desired section, EnFor entropy and HeIt is super entropy for indicating whole quantitative performance；
Step 7.2: calculate degree of membership:
E` in formulanIt is with En for expectation, HeThe normal random number generated for standard deviation.
6. a kind of distribution Running State fuzzy synthetic appraisement method according to claim 1, it is characterised in that: the step
Fuzzy evaluating matrix in rapid 8 are as follows:
Wherein, rijFor the related coefficient between i-th of index and jth index.
7. a kind of distribution Running State fuzzy synthetic appraisement method according to claim 1, it is characterised in that: the step
The process of comprehensive evaluation model is established in rapid 9:
Step 9.1: determining evaluation indice, if first layer index has m index, be denoted as U={ u1,u2,…,um, according to ownership
Relationship is divided into k subset, then U={ U1,U2,…,Uk, umFor index；
Step 9.2: establishing opinion rating collection, opinion rating collection is evaluation criterion set, is equipped with l evaluation criterion, i.e. evaluate collection V
={ v1, v2,…,vl, V={ v1=poor, v2=general, v3=medium, v4=good, v5=outstanding；
Step 9.3: determining index weights, the method meter combined using the cooperative game method and variable-weight theory that propose in step 6
Calculate index weights；The weight for calculating first layer index, obtains k weight vectorsWherein i=1,2,3 ...,
k；
Step 9.4: subordinated-degree matrix is determined, according to the statistic algorithm of the historical data of power distribution network index and cloud model parameter, meter
Calculation obtains the trapezium cloud model parameter of each index corresponding grade, and each grade is calculated according to trapezoidal cloud generator algorithm
Degree of membership constitutes subordinated-degree matrix Ri；
Step 9.5: calculating assessment vector；
Step 9.6: calculating comprehensive assessment vector, second layer index weights vector w=[w1,w2,…,wk], the subordinated-degree matrix R of U
=[B1,B2,…,Bk], final assessment vector are as follows:
B=wR=[b1,b2,…,bl]
Complete evaluation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710842212.8A CN109523101A (en) | 2017-09-18 | 2017-09-18 | A kind of distribution Running State fuzzy synthetic appraisement method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710842212.8A CN109523101A (en) | 2017-09-18 | 2017-09-18 | A kind of distribution Running State fuzzy synthetic appraisement method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109523101A true CN109523101A (en) | 2019-03-26 |
Family
ID=65767597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710842212.8A Pending CN109523101A (en) | 2017-09-18 | 2017-09-18 | A kind of distribution Running State fuzzy synthetic appraisement method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109523101A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766320A (en) * | 2019-10-23 | 2020-02-07 | 北京新机场建设指挥部 | Method and device for evaluating operation safety of airport intelligent power grid |
| CN111401740A (en) * | 2020-03-16 | 2020-07-10 | 国网浙江浙电招标咨询有限公司 | Power grid side energy storage system evaluation system and method |
| CN111724030A (en) * | 2020-05-09 | 2020-09-29 | 中国科学院南京地理与湖泊研究所 | Water quality comprehensive evaluation method, model, device and storage medium |
| CN111815190A (en) * | 2020-07-15 | 2020-10-23 | 国网能源研究院有限公司 | Power grid development diagnosis analysis method and system based on multivariate information deep mining |
| CN112488766A (en) * | 2020-12-09 | 2021-03-12 | 广州品唯软件有限公司 | Page display diagram setting method and device, computer equipment and storage medium |
| CN112633622A (en) * | 2020-09-29 | 2021-04-09 | 国网四川省电力公司信息通信公司 | Intelligent power grid operation index screening method |
-
2017
- 2017-09-18 CN CN201710842212.8A patent/CN109523101A/en active Pending
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110766320A (en) * | 2019-10-23 | 2020-02-07 | 北京新机场建设指挥部 | Method and device for evaluating operation safety of airport intelligent power grid |
| CN111401740A (en) * | 2020-03-16 | 2020-07-10 | 国网浙江浙电招标咨询有限公司 | Power grid side energy storage system evaluation system and method |
| CN111724030A (en) * | 2020-05-09 | 2020-09-29 | 中国科学院南京地理与湖泊研究所 | Water quality comprehensive evaluation method, model, device and storage medium |
| CN111815190A (en) * | 2020-07-15 | 2020-10-23 | 国网能源研究院有限公司 | Power grid development diagnosis analysis method and system based on multivariate information deep mining |
| CN112633622A (en) * | 2020-09-29 | 2021-04-09 | 国网四川省电力公司信息通信公司 | Intelligent power grid operation index screening method |
| CN112633622B (en) * | 2020-09-29 | 2024-02-27 | 国网四川省电力公司信息通信公司 | Smart power grid operation index screening method |
| CN112488766A (en) * | 2020-12-09 | 2021-03-12 | 广州品唯软件有限公司 | Page display diagram setting method and device, computer equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109523101A (en) | A kind of distribution Running State fuzzy synthetic appraisement method | |
| CN108108875A (en) | A kind of distribution Running State fuzzy synthetic appraisement method based on cooperative game method and trapezoidal cloud model | |
| CN106505593B (en) | A kind of analysis of distribution transforming three-phase imbalance and the method for load adjustment based on big data | |
| CN103679544A (en) | Comprehensive assessment method for running of intelligent power distribution network | |
| CN104881609A (en) | Credibility evaluation method of software unit of complex software system | |
| CN102930482A (en) | Investment effect evaluation method for county power distribution networks | |
| CN108197848A (en) | A kind of energy quality comprehensive assessment method and device based on intuitionistic fuzzy theory | |
| CN105160149B (en) | A kind of demand response scheduling evaluation system construction method for simulating regulating units | |
| Wan | Multi-attribute decision making method based on possibility variance coefficient of triangular intuitionistic fuzzy numbers | |
| CN102063657A (en) | Operating level and power supplying capability evaluation method for urban electric distribution network | |
| CN107358332A (en) | A kind of dispatching of power netwoks runs lean evaluation method | |
| CN104881003A (en) | Effectiveness evaluation method for metering production facilities | |
| CN106296019A (en) | Region energy quality comprehensive assessment method and system | |
| CN105335902A (en) | Reliability determining method and device for electric power communication net | |
| CN106485089A (en) | The interval parameter acquisition methods of harmonic wave user's typical condition | |
| Liu et al. | A multistage decision-making method for multi-source information with Shapley optimization based on normal cloud models | |
| Sarkar et al. | A multi‐criteria decision making approach for strategy formulation using Pythagorean fuzzy logic | |
| CN105894212A (en) | Comprehensive evaluation method for coupling and decoupling ring of electromagnetic ring network | |
| CN107292489A (en) | A kind of dispatching of power netwoks runs lean evaluation system | |
| CN111340325A (en) | Method and system for evaluating service level of power transmission and transformation facility based on comprehensive evaluation index | |
| CN114139940A (en) | Generalized demand side resource network load interaction level assessment method based on combined empowerment-cloud model | |
| CN109523102A (en) | A kind of distribution Running State integrated evaluating method | |
| CN106548284A (en) | A kind of adaptive mode massing power grid security Alarm Assessment method towards operation regulation and control | |
| CN104217296A (en) | Listed company performance comprehensive evaluation method | |
| CN104143159B (en) | A kind of power distribution network health degree self-diagnosing method for considering regional disparity |
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 |