CN105956757A - Comprehensive evaluation method for sustainable development of smart power grid based on AHP-PCA algorithm - Google Patents

Comprehensive evaluation method for sustainable development of smart power grid based on AHP-PCA algorithm Download PDF

Info

Publication number
CN105956757A
CN105956757A CN201610268511.0A CN201610268511A CN105956757A CN 105956757 A CN105956757 A CN 105956757A CN 201610268511 A CN201610268511 A CN 201610268511A CN 105956757 A CN105956757 A CN 105956757A
Authority
CN
China
Prior art keywords
evaluation
factor
ahp
pca
index
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
CN201610268511.0A
Other languages
Chinese (zh)
Inventor
陈海波
陈丽霞
凌平
陈靖文
施侠
孙弢
方陈
徐舒玮
张宇
姜山
柳劲松
冯冬涵
施勇
郑健
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.)
Shanghai Jiaotong University
State Grid Shanghai Electric Power Co Ltd
Original Assignee
Shanghai Jiaotong University
State Grid Shanghai Electric Power Co Ltd
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 Shanghai Jiaotong University, State Grid Shanghai Electric Power Co Ltd filed Critical Shanghai Jiaotong University
Priority to CN201610268511.0A priority Critical patent/CN105956757A/en
Publication of CN105956757A publication Critical patent/CN105956757A/en
Pending legal-status Critical Current

Links

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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Abstract

The invention provides a comprehensive evaluation method for sustainable development of smart power grid based on AHP-PCA algorithm. In the first k months before the evaluation method is put into use, based on the analytic hierarchy process (AHP), all the evaluation factors are statically weighted and are scored according to the operation of the smart power grid so as to establish the evaluation index WAHP of AHP. After K-month operation of the evaluation method, based on the principal component analysis, all the evaluation factors were dynamically weighted from k + 1 month, and the effective principal components are sorted out to establish the evaluation index WPCA of PCA. The evaluation index WPCA of PCA is then combined with the evaluation index WAHP of AHP to serve as the evaluation index WAHP for evaluating the sustainable development of the smart power grid. The invention combines the analytic hierarchy process and the principal component analysis method to solve the problem of error caused by subjective difference in smart power grid evaluation and the problem that the principal components are void of meaning despite contained information so as to provide an effective management scheme for the sustainable development of a smart power grid.

Description

Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method
Technical field
The present invention relates to the technology of a kind of intelligent grid evaluation areas, specifically a kind of based on AHP (AnalyticHierarchyProcess, step analysis) method and PCA (PrincipalComponentAnalysis, main constituent Analyze) the intelligent grid Comprehensive Evaluation of Sustainable Development method of method.
Background technology
Intelligent grid has become the development trend of coming few decades power system, and its capability of sustainable development is to weigh electrical network The core element of development degree, therefore the Comprehensive Appraisal Study of intelligent grid sustainable development is for the construction of intelligent grid Significant with development.
It is the more commonly used analysis method in the intelligent grid middle-level analysis of assessment and principal component analysis.Can hold to meet The needs of continuous Development Assessment, it is desirable to the evaluation index of used analysis method has ageing, and single employing step analysis Or principal component analysis is difficult to.
Analytic hierarchy process (AHP) be a kind of qualitative and quantitatively combine, systematization, the analysis method of stratification, be systematic analysis Important tool.But analytic hierarchy process (AHP) quantitative data is few, qualitative index is many, subjective factors impact is big, continuous along with intelligent grid Development and reform, the tax power of expert can no longer effective property；And selecting index quantity more time, to journey important between each two index The judgement of degree arises that difficulty, even the concordance of Mode of Level Simple Sequence and total sequence can be produced impact, make consistency check Can not pass through.
PCA is the thought utilizing dimensionality reduction, and multiple original variables are converted into the aggregate variable of negligible amounts, With the most information in reflection original variable.Principal component analysis needs the support of mass data, and otherwise dimensionality reduction is likely to result in Main constituent cannot provide the explanation meeting real background and meaning, causes having quantity of information in vain and without physical meaning, but intelligent grid Initial operating stage data volume is less, it is clear that cannot meet analysis needs.
Through the retrieval of prior art is found, Chinese patent literature CN105429133A, open (bulletin) day 2016.03.23, a kind of energy the Internet comprehensive assessment based on intelligent grid innovative demonstration district and the side of distributing rationally are disclosed Method, comprises the following steps: step 1, foundation are based on intelligent grid innovative demonstration district energy the Internet assessment indicator system；Step 2, According to the dependency between each index, use ISM that index is carried out layered shaping；Step 3, to solve each index corresponding Weight factor；Step 4, each index is carried out data prediction: step 5, intelligent grid innovative demonstration district is comprehensively commented Valency；Step 6, foundation evaluation result are optimized configuration one to demonstration area intelligent grid.But this technology Main Basis level Analytic process, has stronger subjectivity.
Chinese patent literature CN105303468A, open (bulletin) day 2016.02.03, discloses a kind of based on main one-tenth The intelligent grid of segregation alanysis builds integrated evaluating method, comprises the following steps: step 1, sets up or selects universally recognized Intelligent grid builds System of Comprehensive Evaluation；Step 2, achievement data is standardized process；Step 3, set up index number The eigenvalue according to correlation matrix and solving this matrix and characteristic vector, generate main constituent expression formula；Step 4, calculate main one-tenth Divide variance contribution ratio and cumulative proportion in ANOVA, determine main constituent number；Step 5, construct comprehensive main constituent evaluation index function, Provide the comprehensive evaluation result of intelligent grid development construction level；Step 6, set up main constituent factor loads matrix, to Intelligent electric Net comprehensive evaluation index carries out cluster analysis.But this technology can be used for the data volume of analysis relatively at intelligent grid first stage of construction Few, it is difficult to meet the analysis needs of the method.
Summary of the invention
The present invention is directed to deficiencies of the prior art, it is proposed that a kind of intelligent grid based on AHP-PCA method can Sustainable development integrated evaluating method, it is possible to solve error problem and main constituent that in intelligent grid assessment, subjective differences is brought empty There is a quantity of information and without the problem of physical meaning, the sustainable development for intelligent grid provides suggestion with guidance program preferably to promote Enter the sustainable development of intelligent grid.
The present invention is achieved by the following technical solutions,
Front k month that the present invention comes into operation in evaluation methodology, due to data deficiencies, puts aside use principal component analysis Based on analytic hierarchy process (AHP), method, now carries out that all factors of evaluation carry out static tax and weighs, and according to the ruuning situation of intelligent grid Each factor of evaluation is given a mark, builds AHP evaluation number WAHP；After evaluation methodology runs k month, from+1 month beginning base of kth In PCA, all factors of evaluation are carried out dynamic weight index, filter out effective main constituent, build PCA evaluation number WPCA, And with AHP evaluation number WAHPIt is combined, as intelligent grid sustainable development appraisal index WAHP, as shown in Figure 1；
Described WAHPScore and its respective weights sum of products for factor of evaluation each in step analysis；
Described WPCAFor main constituent effective in principal component analysis and its variance contribution ratio sum of products；
Described WAHP-PCAIn, after evaluation methodology runs K month, as K≤k, WAHPWeight be 1, WPCAWeight It is 0；As K ＞ k, WAHPWeight beWPCAThe weight of result is
Described analytic hierarchy process (AHP) builds WAHP, comprise the following steps:
S11, set up intelligent grid sustainable development evaluation index system according to hierarchy Model, including: destination layer, standard Then layer and solution layer；
S12, to factor development of judgment matrix in rule layer, obtain the sequencing weight of factor in rule layer；Then to solution layer In factor of evaluation carry out the structure of same sequence judgment matrix, and carry out the single sequence of same sequence evaluation factor, obtain this In sequence, factor of evaluation is based on the importance ranking weights of factor in rule layer；
S13, the result of the same single sequence of sequence evaluation factor is carried out single sequence consistency check, meets consistency check , carry out total sequence of the whole factor of evaluation of solution layer, otherwise repeat step S12, until single by same sequence evaluation factor Sequence consistency check；
S14, the same single sequence of sequence evaluation factor is by, after consistency check, carrying out the total of the whole factor of evaluation of solution layer Sort and always sort consistency check, and the sequence in this, as the whole factor of evaluation of solution layer meeting consistency check is weighed Value, otherwise repeats step S13, until by the whole factor of evaluation of solution layer single sequence consistency check, obtaining with factor of evaluation afterwards Combination is divided to obtain WAHP。
Described destination layer is intelligent grid sustainable development appraisal index WAHP-PCA；
Described rule layer includes: economic index, social index and Environmental index；
The solution layer sequence that described economic index is corresponding, including following factor of evaluation: net coal consumption rate, comprehensive line loss Rate, average peak-valley ratio, stored energy capacitance ratio, clean energy resource installation proportion and synthesis desulfurating efficiency；
The solution layer sequence that described social index is corresponding, including following factor of evaluation: power supply reliability, per capita electricity consumption Amount, stored energy capacitance ratio, net factory conditioning unit scale, distributed power source proportion, intelligent meter meter popularity rate and electric automobile permeability；
The solution layer sequence that described Environmental index is corresponding, including following factor of evaluation: net coal consumption rate, electric automobile Occupation rate, clean energy resource installation proportion, distributed energy proportion and synthesis desulfurating efficiency.
Described judgment matrix is A=(aij)n×n, wherein:The a as i=jij=1, aijRepresent same sequence Middle any two factor of evaluation liWith ljThe importance compared, n is the exponent number of judgment matrix A.
Described sequencing weight refers to determining judgment matrix A eigenvalue of maximum λmaxOn the basis of, it is normalized The characteristic vector W obtained；Described this feature vector W be in this level same sequence evaluation factor based on last layer time factor Importance ranking weights.
Described same sequence evaluation factor single sequence consistency check, refers to when same sequence evaluation factor concordance ratio RateTime, then it is assumed that judgment matrix A has satisfied concordance, otherwise needs to readjust in judgment matrix A Part aijValue, wherein:For single sequence coincident indicator；RIFor single sequence random index, with rank N is relevant for number, is used for weighing CISize；Due to λmaxContinuous print depends on aij, therefore use λmax-n weighs differing of judgment matrix A Cause degree；Work as λmaxDuring-n=0, it is judged that matrix A has a concordance completely, and λmaxThe difference of-n is the biggest, CIThe biggest, it is judged that matrix A discordance is the most serious.
The whole factor of evaluation of the described solution layer consistency check that always sorts to destination layer, refers to whole factor concordance RatioTime, total hierarchial sorting has satisfied concordance, otherwise needs to readjust those consistent A in the judgment matrix A that sex rate is highijValue, wherein: Q is the number of rule layer factor, aqFor rule layer factor q to target The sequence of layer,For the coincident indicator that always sorts, CIQ () is the solution layer Mode of Level Simple Sequence one to rule layer factor q Cause property index；For the random index that always sorts, RIWhat q () was solution layer to rule layer factor q is random consistent Property index.
Described PCA builds WPCA, comprise the following steps:
S21, according to intelligent grid sustainable development evaluation index system, i.e. acquisition scheme when evaluation methodology brings into operation The sample data of each factor of evaluation in Ceng, and be standardized processing to the sample data of K month, obtain normalized matrix, disappear Except the inconsistent problem causing being difficult to compare of each factor of evaluation dimension；
S22, the sample data gathered is analyzed, sets up covariance matrix, and covariance matrix is carried out eigenvalue and The process of characteristic vector, obtains the variance of each main constituent；
S23, calculate main constituent variance contribution ratio, choose effective main constituent, to reduce dimension；Check effective main constituent Dependency, if dependency is zero, then generate main constituent composite evaluation function, otherwise, return S21Recalculate each index number According to standardized value；
S24, main constituent composite evaluation function is carried out anti-standardization, obtains WPCA。
Described normalized matrix isBy to sample data matrixZ-score method standardization is used to obtain, wherein: the mathematics phase after standardization Hope as E (Zi)=0, variance is D (Zi)=1, wherein: ZiRepresent the sample data after all normalization of i-th evaluation index；
Described sample data matrix X represents that in intelligent grid sustainable development evaluation index system, total p evaluation refers to Mark, acquires t data sample in K month.
Described Z-score method standardization is realized by formula calculated below:
Described correlation matrix is R=(rij)p×p, each element z in reaction normal matrix ZiAnd zjMutual pass System, wherein: correlation coefficientAnd covariance cov (zi,zj)=E{ [zi-E(zi)] [zj-E(zj)]=E (zi,zj)-E(zi)E(zj)=E (zi,zj), i.e. after Z-score method standardization, correlation coefficient square Battle array R, equal to covariance matrix Σ, therefore carries out eigenvalue and characteristic vector solves and i.e. can obtain main one-tenth covariance matrix Σ Divide yeVariance, judge each main constituent dependency with this.
The process that described covariance matrix Σ solves eigenvalue and characteristic vector is as follows: because covariance matrix Σ is for real Symmetrical matrix, must orthogonal similarity in diagonal matrix, then have BTΣ B=Λ=diag (λ1,λ2,...λp), make Y=BTZ, then have cov (YYT)=cov (BTZZTB)=BTcov(ZZT) B=Λ=diag (λ1,λ2,...λp), finally give covarianceVariance D (the y of i.e. e main constituente)=λe, wherein: yeAnd yfRepresent e main one-tenth respectively Point, the f main constituent；Variance D (the y of e main constituente) and eigenvalue λeEqual, then comprise yeAt the interior main one-tenth of any two / orthogonal.
Described variance contribution ratio refers to that main constituent carries the percentage ratio of original indication information amount, main constituent yeVariance tribute The rate of offering isFront m main constituent contribution rate of accumulative total of variance is:In actual applications, based on experience value Method, when main constituent contribution rate of accumulative total of variance ρ reaches 90%, can cast out remaining main constituent, m main constituent before only retaining, institute Front m the main constituent stated is effective main constituent.
Described covariance matrix Σ has m the eigenvalue more than zero and meets λ1≥λ2≥...≥λm>=0, eigenvalue pair The specification features vector answered is C=(c1,c2,...cm), then m main constituent is expressed as:
It is abbreviated as Y=CTZm, wherein: Y is referred to as main constituent Factor load-matrix, ZmFor m main constituent normalized matrix.
The dependency of described front m main constituent passes through correlation matrixCarry out Inspection, meets RmExplanation before between m main constituent dependency be zero, i.e. the whole dimensionality reduction restructuring procedure of index system achieves Decorrelation completely.
The described main constituent composite evaluation function that front m dependency is zero is f=w1y1+w2y2+...+wmym, can be The evaluation information amount of the big degree ground former index system of sign, and greatly reduce the scale of original index system, by quantity Numerous indexs is attributed to single main constituent comprehensive evaluation index, embodies principal component analysis in intellectual technology appraisement system Terseness, substantivity and operability.
Described anti-standardization refers to Z-score method standardization is carried out inverse operation, i.e.Instead Standardization uses average and the standard deviation of evaluation number based on analytic hierarchy process (AHP) so that WPCAWith WAHPThere is comparability.
Technique effect
Compared with prior art, the present invention is according to analytic hierarchy process (AHP) and the feature of PCA complementation, it is proposed that knot Closing analytic hierarchy process (AHP) and the integrated evaluating method of PCA, Primary Stage Data amount uses more targeted level time less Analytic process, the data volume of collection the most progressively relies on based on the entitled PCA of data；Mistake in both transition Cheng Zhong, completes and is composed the power smooth conversion to dynamic weight index by static state so that intelligent grid sustainability development index is evaluated more Reasonability, science and stability, the scale construction for intelligent grid provides foundation with operation.
Accompanying drawing explanation
Fig. 1 is to set up intelligent grid sustainable development appraisal index flow chart in the present invention；
Fig. 2 is the hierarchy Model figure of intelligent grid sustainable development appraisal index in the present invention；
Fig. 3 is evaluation index principal component analysis characteristic root distribution rubble figure in embodiment 1；
Fig. 4 is the intelligent grid sustainable development appraisal index result figure of embodiment 1.
Detailed description of the invention
Elaborating embodiments of the invention below, the present embodiment is carried out under premised on technical solution of the present invention Implement, give detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following enforcement Example.
Embodiment 1
Step S11, as in figure 2 it is shown, the present embodiment is according to the needs of intelligent grid Sustainable Development Evaluation, with intelligent grid Sustainable development appraisal index is as destination layer, using economy, social and Environmental as rule layer, including net coal consumption rate Rate, comprehensive line loss per unit, average peak-valley ratio, stored energy capacitance ratio, clean energy resource installation proportion, power supply reliability, per capita household electricity consumption, Net factory conditioning unit scale, distributed power source proportion, intelligent meter meter popularity rate, electric automobile permeability and synthesis desulfurating efficiency exist Interior many factors is as solution layer.
Described net coal consumption rate refers to the gross coal consumption rate grams of every kilowatt hour delivery；
Described comprehensive line loss per unit refers to the ratio of operation of power networks consume and delivery；
Described average peak-valley ratio refers to the ratio of the electrical network peak load difference with minimum load and peak load；
Described stored energy capacitance ratio refers to that energy-storage system installed capacity accounts for the ratio of total installation of generating capacity；
Described clean energy resource installation proportion refers to that clean energy resource installed capacity accounts for the ratio of total installation of generating capacity；
Described power supply reliability refers to that the power-on time after rejecting the average power off time of user accounts for the ratio of power supply total time Example；
Described per capita household electricity consumption is embodied by the average power off time of user, the power failure total time of the most all users and user The ratio of sum；
Described net factory conditioning unit scale refers to that the generating set quantity being configured with net factory conditioning unit scale accounts for all The ratio of generating set quantity；
Described distributed power source proportion refers to that distributed power source capacity accounts for the ratio of total installation of generating capacity；
Described intelligent meter meter popularity rate refers to be equipped with the number of users of intelligent meter meter and accounts for the ratio of total number of users；
Described electric automobile permeability refers to that electric automobile is held quantity and accounted for the ratio of automobile total quantity；
Described synthesis desulfurating efficiency refers to the product of desulfurization equipment operational percentage and desulfurization degree, wherein: desulfurization equipment puts into operation Rate is the desulfurization equipment operation time to account for generating set to run the ratio of time.
Described judgment matrix A=(aij)n×nIn, aijValue is as shown in table 1:
Table 1aijValue
Wherein: xi/xjJudge to obtain the intermediate value value 2,4,6,8 respectively of the above results two-by-two.
Step S12, according to judgment matrix A=(aij)n×n, in described rule layer, factor is weighed relative to the sequence of destination layer Value is as shown in table 2:
The each factor of table 2 rule layer is relative to the judgment matrix of destination layer and each factor sequencing weight
Further, the judgment matrix based on the corresponding factor of rule layer of same sequence evaluation factor in solution layer is obtained with each Factor sequencing weight, as shown in table 3～table 5:
Table 3 judgment matrix based on economy and each factor of evaluation sequencing weight
Table 4 is based on social judgment matrix and each factor of evaluation sequencing weight
Table 5 is based on Environmental judgment matrix and each factor of evaluation sequencing weight
Step S13, carrying out the Mode of Level Simple Sequence consistency check of table 2～table 5, assay is as shown in table 7, meanwhile, in order to Weigh Consistency Ratio CRSize, the random index R to each exponent numberICarry out value, as shown in table 6:
Table 6 random index RIValue
Table 7 Mode of Level Simple Sequence consistency check
Mode of Level Simple Sequence Consistency Ratio CR＜ 0.10, the most single sequence judgment matrix A have satisfied concordance.
Step S14, then carry out the solution layer total hierarchial sorting consistency check to destination layer, according in table 2 and table 7 Data, try to achieve Rf=0.0339 ＜ 0.10, meets consistency check.
After completing total sequence consistency check of whole factor of evaluation, each factor of evaluation is relative to the weights such as table of destination layer Shown in 8:
Table 8 factor of evaluation is relative to the weights of destination layer
Giving a mark these factors of evaluation, intelligent grid Sustainable Development Evaluation index is divided into two according to its Variation Features Class: stable type index and development-oriented index；The feature of stable type index is that current situation is the most ripe, in the coming years or even ten Will not produce in several years and change largely, for this type of index when grade classification, using average national level as score Intermediate value, such as comprehensive line loss per unit, synthesis desulfurating efficiency etc.；And the technology that the feature of development-oriented index is this field is sent out the most rapidly Exhibition, will may have bigger development within the coming years；In order to ensure the appraisement system effectiveness in following a period of time, such as electricity The average national level of the development-oriented indexs such as electrical automobile permeability will be less than score intermediate value, and the grade of score value is divided into 1,3,5,7,9 Totally five grades, 2,4,6,8 is the intermediate value of score, the most as shown in table 9:
Table 9 factor of evaluation data Score Lists
As shown in table 10-1 and table 10-2, with the somewhere electrical network data instance of continuous 25 months, it is carried out based on level The achievement data score value of analytic process judges and the calculating of evaluation number, and the result obtained is as shown in table 11:
Continuous 25 the month by month correlation evaluation index initial datas of table 10-1 somewhere electrical network
Continuous 25 the month by month correlation evaluation index initial datas of table 10-2 somewhere electrical network
Continuous 25 months data scores of table 11 somewhere electrical network and intelligent grid sustainable development based on analytic hierarchy process (AHP) refer to Number
Table 12 is obtained after above-mentioned intelligent grid sustainable development appraisal index based on analytic hierarchy process (AHP) is analyzed:
Table 12 evaluation number based on analytic hierarchy process (AHP)
Step S21, the electric network data of continuous to table 10 25 months carries out principal component analysis, because of factor of evaluation each in solution layer Dimension different, it is impossible to carry out the analysis of main constituent, therefore first have to initial data is standardized, use Z-score method Carry out data process, obtain standardization electric network data, as shown in table 13-1 and table 13-2:
Continuous 25 the month by month correlation evaluation index quasi-ization data of table 13-1 somewhere electrical network
Continuous 25 the month by month correlation evaluation index quasi-ization data of table 13-2 somewhere electrical network
Step S22, table 13 Plays data are solved main constituent eigenvalue, obtain table 14:
The main constituent eigenvalue distribution of table 14 evaluation index
Step S23, rule of thumb method, the main constituent chosen meets information quantity requirement when adding up variance contribution ratio more than 90%, And the variance contribution ratio that first principal component and Second principal component, add up has reached 96.65%, meet minimal information amount requirement, therefore Extract front two main constituents to be analyzed as effective main constituent；It is illustrated in figure 3 the characteristic root distribution of 12 main constituents Rubble figure, the characteristic root of first principal component is in absolute advantages, owing to main constituent characteristic root is equal to its variance, therefore first principal component Can farthest explain original index, characterize original evaluation and test data volume to greatest extent；From the beginning of the 3rd main constituent, its side Difference the most gradually tends to null value, illustrates that rear 10 main constituents have lost meaning for characterizing original appraisement system index.
Main constituent Factor load-matrix is as follows:
Table 15 evaluation index main constituent Factor load-matrix
Evaluation index Factor loading on first principal component Factor loading on Second principal component,
Power supply reliability 0.091 -0.130
Per capita household electricity consumption 0.002 0.501
Stored energy capacitance ratio 0.103 0.021
Net factory conditioning unit scale 0.103 0.024
Distributed power source proportion 0.100 0.029
Intelligent meter meter popularity rate 0.102 0.021
Electric automobile permeability 0.103 0.021
Net coal consumption rate -0.103 -0.021
Clean energy resource installation proportion 0.103 0.019
Comprehensive line loss per unit -0.103 -0.021
Average peak-valley ratio -0.019 0.492
Synthesis desulfurating efficiency 0.103 0.022
According to table 15, the first principal component after principal component analysis can be obtained and Second principal component, expression formula is:
y1=0.091x1+0.002x2+0.103x3+0.103x4
+0.100x5+0.102x6+0.103x7-0.103x8,
+0.103x9-0.103x10-0.019x11+0.103x12
y2=-0.103x1+0.501x2+0.021x3+0.024x4
+0.029x5+0.021x6+0.021x7-0.021x8,
+0.019x9-0.021x10+0.492x11+0.022x12
Wherein, x1、x2、...x12Represent the factor of evaluation normal data after standardization.
Check that the dependency of first and second main constituent obtains covariance matrix ∑ as shown in table 16:
Table 16 extracts main constituent covariance matrix ∑
Be can be seen that between two effective main constituents of extraction orthogonal, thus by the covariance matrix ∑ shown in table Checking principal component analysis flow process is correct.
In order to calculate intelligent grid sustainability development index based on PCA, it is necessary first to calculate main constituent and divide The evaluation of estimate of analysis method, therefore builds main constituent function；The characteristic root being learnt first principal component by table 14 is 9.653, the second main one-tenth The characteristic root divided is 1.945, then when generating comprehensive main constituent evaluation function, with reference to principal component weight computing formula, calculates Shared by first principal component and Second principal component, weight such as formula is:
Obtain comprehensive main constituent evaluation function: y=w1y1+w2y2=0.832y1+0.167y2；
Therefore, the sustainable development of 25th month comprehensive main constituent evaluation of estimate is as shown in table 17:
25th month sustainable development comprehensive main constituent evaluation of estimate of table 17
Step S25, by the evaluation of estimate of Principal Component Analysis Method in table 17 with the average of evaluation number based on AHP in table 12 with Standard deviation carries out anti-standardization as benchmark, obtains intelligent grid sustainable development appraisal index based on PCA, As shown in table 18:
Table 18 intelligent grid based on PCA sustainable development appraisal index
The intelligent grid sustainable development based on PCA in other months can be obtained with same method step Exhibition evaluation number.
W according to continuous 25 months be the previously calculatedAHPAnd WPCA, take k=11, obtain WAHP-PCA, as shown in table 19:
Table 19 intelligent grid sustainable development appraisal index
It should be noted that WPCANeed monthly to carry out calculating to update.
In order to preferably study W in table 19AHP、WPCAAnd WAHP-PCABetween relation, the development of three kinds of evaluation numbers is become Change process is mapped, as shown in Figure 4:
At 1-11 month period, WAHPCurve and WAHP-PCACurve co-insides；
From 12nd month, the introducing of PCA made WPCATo WAHP-PCAProduce certain impact: shadow originally Ring less, WAHPStill to WAHP-PCAPlay a leading role；After a period of time, WPCATo WAHP-PCAImpact constantly increase, WAHP-PCABent Line is constantly to WPCADraw close.

Claims (10)

1. an intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method, it is characterised in that in evaluation methodology The front k come into operation month, carry out that all factors of evaluation are carried out static tax based on analytic hierarchy process (AHP) and weigh, and according to intelligent grid Ruuning situation each factor of evaluation is given a mark, build AHP evaluation number WAHP；After evaluation methodology runs k month, from kth+1 Within individual month, start based on PCA and all factors of evaluation are carried out dynamic weight index, filter out effective main constituent, build PCA and comment Valency index WPCA, and with AHP evaluation number WAHPIt is combined, as intelligent grid sustainable development appraisal index WAHP；
Described WAHPScore and its respective weights sum of products for factor of evaluation each in step analysis；
Described WPCAFor main constituent effective in principal component analysis and its variance contribution ratio sum of products；
Described WAHP-PCAIn, after evaluation methodology runs K month, as K≤k, WAHPWeight be 1, WPCAWeight be 0；When During K ＞ k, WAHPWeight beWPCAThe weight of result is
Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method the most according to claim 1, its feature It is that described analytic hierarchy process (AHP) builds WAHP, comprise the following steps:
S11, set up intelligent grid sustainable development evaluation index system according to hierarchy Model, including: destination layer, rule layer And solution layer；
S12, to factor development of judgment matrix in rule layer, obtain the sequencing weight of factor in rule layer；Then in solution layer Factor of evaluation carries out the structure of same sequence judgment matrix, and carries out the single sequence of same sequence evaluation factor, obtains this sequence Middle factor of evaluation is based on the importance ranking weights of factor in rule layer；
S13, the result of the same single sequence of sequence evaluation factor is carried out single sequence consistency check, meets consistency check, enter Total sequence of the whole factor of evaluation of row solution layer, otherwise repeats step S12, until by the same single sequence of sequence evaluation factor one Cause is checked；
S14, the same single sequence of sequence evaluation factor, by after consistency check, carries out total sequence of the whole factor of evaluation of solution layer And the consistency check that always sorts, meet the sequencing weight in this, as the whole factor of evaluation of solution layer of consistency check, Otherwise repeat step S13, until by the whole factor of evaluation of solution layer single sequence consistency check, afterwards with factor of evaluation score In conjunction with obtaining WAHP。
Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method the most according to claim 2, its feature It is that described destination layer is intelligent grid sustainable development appraisal index WAHP-PCA；Described rule layer includes: economy refers to Mark, social index and Environmental index；The solution layer sequence that described economic index is corresponding, including following factor of evaluation: Net coal consumption rate, comprehensive line loss per unit, average peak-valley ratio, stored energy capacitance ratio, clean energy resource installation proportion and synthesis desulfurating efficiency； The solution layer sequence that described social index is corresponding, including following factor of evaluation: power supply reliability, per capita household electricity consumption, energy storage Capacity Ratio, net factory conditioning unit scale, distributed power source proportion, intelligent meter meter popularity rate and electric automobile permeability；Described The solution layer sequence that Environmental index is corresponding, including following factor of evaluation: net coal consumption rate, electric automobile occupation rate, cleaning energy Source installation proportion, distributed energy proportion and synthesis desulfurating efficiency.
Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method the most according to claim 2, its feature It is that described judgment matrix is A=(aij)n×n, wherein:The a as i=jij=1, aijRepresent any in same sequence Two factors of evaluation liAnd ljThe importance compared, n is the exponent number of judgment matrix A.
5. according to the intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method described in claim 2 or 4, its Feature is, described sequencing weight refers to determining judgment matrix A eigenvalue of maximum λmaxOn the basis of, it is normalized The characteristic vector W obtained；Described this feature vector W be in this level same sequence evaluation factor based on last layer time factor Importance ranking weights.
Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method the most according to claim 2, its feature It is, described same sequence evaluation factor single sequence consistency check, refers to when same sequence evaluation factor Consistency RatioTime, it is believed that judgment matrix A has satisfied concordance, otherwise needs to readjust a in judgment matrix Aij's Value, wherein:For single sequence coincident indicator；RIIt is for single sequence random index, relevant with exponent number n, For weighing CISize.
Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method the most according to claim 2, its feature It is that the whole factor of evaluation of described solution layer always sorts consistency check, refers to whole factor Consistency RatioTime, total hierarchial sorting has satisfied concordance, otherwise needs to readjust those concordance ratios A in the judgment matrix A that rate is highijValue, wherein: Q is the number of rule layer factor, aqFor rule layer factor q to destination layer Sequence,For the coincident indicator that always sorts, CIQ () is the solution layer Mode of Level Simple Sequence concordance to rule layer factor q Index；For the random index that always sorts, RIQ () is that the random concordance of rule layer factor q is referred to by solution layer Mark.
Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method the most according to claim 1, its feature It is that described PCA builds WPCA, comprise the following steps:
S21, according to intelligent grid sustainable development evaluation index system, when evaluation methodology brings into operation i.e. acquisition scheme layer The sample data of each factor of evaluation, and be standardized processing to the sample data of K month, obtain normalized matrix, eliminate each The inconsistent problem causing being difficult to compare of factor of evaluation dimension；
S22, the sample data gathered is analyzed, sets up covariance matrix, and covariance matrix is carried out eigenvalue and feature The process of vector, obtains the variance of each main constituent；
S23, check the dependency of effective main constituent, if dependency is zero, then calculate main constituent variance contribution ratio, to reduce variable Dimension, obtains effective main constituent, generates main constituent composite evaluation function, otherwise, returns S21Recalculate each achievement data Standardized value；
S24, main constituent composite evaluation function is carried out anti-standardization, obtains WPCA；
Described normalized matrix isBy to sample data matrixZ-score method standardization is used to obtain, wherein: the mathematic expectaion after standardization For E (Zi)=0, variance is D (Zi)=1, ZiRepresent the sample data after all normalization of i-th evaluation index；
Described sample data matrix X represents total p evaluation index, K in intelligent grid sustainable development evaluation index system Within individual month, acquire t data sample；
Described dependency is zero correlation matrix referring to front m main constituentIts In: correlation matrix R is equal to covariance matrix ∑, and correlation matrix is R=(rij)p×p, correlation coefficient
Covariance in described covariance matrix ∑Variance D (the y of i.e. e main constituente) and feature Value λeEqual, wherein: yeAnd yfRepresent the e main constituent, the f main constituent respectively；
Described effective main constituent refers to so that accumulative variance contribution ratio ρ reaches front m the main constituent of 90%, wherein:Single main constituent y in described front m main constituenteVariance contribution ratio be
Described anti-standardization refers to Z-score method standardization is carried out inverse operation, i.e.Wherein, Anti-standardization formula uses average and the standard deviation of evaluation number based on analytic hierarchy process (AHP) so that PCA evaluation number is commented with AHP Valency index has comparability.
Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method the most according to claim 8, its feature It is that described Z-score method standardization is realized by formula calculated below:
Intelligent grid Comprehensive Evaluation of Sustainable Development method based on AHP-PCA method the most according to claim 8, it is special Levying and be, described covariance matrix Σ has m the eigenvalue more than zero and meets λ1≥λ2≥…≥λm>=0, eigenvalue is corresponding Specification features vector is C=(c1,c2,...cm), then m main constituent is expressed asLetter It is written as Y=CTZm, wherein: Y is referred to as main constituent Factor load-matrix, ZmFor m main constituent normalized matrix, front m dependency is The main constituent composite evaluation function of zero is f=w1y1+w2y2+...+wmym。
CN201610268511.0A 2016-04-27 2016-04-27 Comprehensive evaluation method for sustainable development of smart power grid based on AHP-PCA algorithm Pending CN105956757A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610268511.0A CN105956757A (en) 2016-04-27 2016-04-27 Comprehensive evaluation method for sustainable development of smart power grid based on AHP-PCA algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610268511.0A CN105956757A (en) 2016-04-27 2016-04-27 Comprehensive evaluation method for sustainable development of smart power grid based on AHP-PCA algorithm

Publications (1)

Publication Number Publication Date
CN105956757A true CN105956757A (en) 2016-09-21

Family

ID=56915996

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610268511.0A Pending CN105956757A (en) 2016-04-27 2016-04-27 Comprehensive evaluation method for sustainable development of smart power grid based on AHP-PCA algorithm

Country Status (1)

Country Link
CN (1) CN105956757A (en)

Cited By (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106709821A (en) * 2017-03-10 2017-05-24 东北电力大学 Main component analysis-based power supply reliability evaluation method of medium-voltage power distribution network
CN106899035A (en) * 2017-03-15 2017-06-27 中国电力科学研究院 Inversion system participates in the appraisal procedure and device of grid-connected rear power distribution network Operating ettectiveness
CN107633354A (en) * 2017-09-08 2018-01-26 国家电网公司 A kind of integrated evaluating method stood with straight-flow system running status health degree
CN107705019A (en) * 2017-09-30 2018-02-16 安徽四创电子股份有限公司 A kind of smart city Huimin service level evaluation method
CN107833149A (en) * 2017-10-24 2018-03-23 国家电网公司 A kind of power network development Development stage method and system based on more discriminant criterions
CN107895227A (en) * 2017-11-02 2018-04-10 上海电力学院 A kind of integrated evaluating method of mobile welding robot operating scheme
CN108665090A (en) * 2018-04-06 2018-10-16 东北电力大学 Urban distribution network saturation load forecasting method based on principal component analysis Yu Verhulst models
CN109307854A (en) * 2018-10-29 2019-02-05 中国电力科学研究院有限公司 A kind of electric energy metering device comprehensive estimation method and system
CN109543737A (en) * 2018-11-15 2019-03-29 国网四川省电力公司信息通信公司 A kind of information system health degree appraisal procedure based on FAHP_FCA combination weighting
CN111582571A (en) * 2020-04-30 2020-08-25 中国电力科学研究院有限公司 Power grid operation situation sensing method and system with model driving and data driving integrated
CN111612371A (en) * 2020-05-29 2020-09-01 国网河北省电力有限公司电力科学研究院 Intelligent ammeter quality evaluation method based on analytic hierarchy process
CN112784486A (en) * 2021-01-20 2021-05-11 中国石油大学(华东) Injection-production relation optimization method based on heterogeneous flow field representation
CN112907032A (en) * 2021-01-21 2021-06-04 中国人民解放军火箭军工程大学 Efficiency-based evaluation method for large electromagnetic ejection device
CN114723283A (en) * 2022-04-07 2022-07-08 中山大学 Ecological bearing capacity remote sensing evaluation method and device for urban group

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106709821A (en) * 2017-03-10 2017-05-24 东北电力大学 Main component analysis-based power supply reliability evaluation method of medium-voltage power distribution network
CN106899035B (en) * 2017-03-15 2021-04-30 中国电力科学研究院有限公司 Method and device for evaluating operation efficiency of power distribution network after inverter system participates in grid connection
CN106899035A (en) * 2017-03-15 2017-06-27 中国电力科学研究院 Inversion system participates in the appraisal procedure and device of grid-connected rear power distribution network Operating ettectiveness
CN107633354A (en) * 2017-09-08 2018-01-26 国家电网公司 A kind of integrated evaluating method stood with straight-flow system running status health degree
CN107705019A (en) * 2017-09-30 2018-02-16 安徽四创电子股份有限公司 A kind of smart city Huimin service level evaluation method
CN107833149A (en) * 2017-10-24 2018-03-23 国家电网公司 A kind of power network development Development stage method and system based on more discriminant criterions
CN107833149B (en) * 2017-10-24 2022-02-25 国家电网公司 Power grid development period division method and system based on multiple discrimination indexes
CN107895227A (en) * 2017-11-02 2018-04-10 上海电力学院 A kind of integrated evaluating method of mobile welding robot operating scheme
CN108665090A (en) * 2018-04-06 2018-10-16 东北电力大学 Urban distribution network saturation load forecasting method based on principal component analysis Yu Verhulst models
CN108665090B (en) * 2018-04-06 2021-06-18 东北电力大学 Urban power grid saturation load prediction method based on principal component analysis and Verhulst model
CN109307854A (en) * 2018-10-29 2019-02-05 中国电力科学研究院有限公司 A kind of electric energy metering device comprehensive estimation method and system
CN109543737A (en) * 2018-11-15 2019-03-29 国网四川省电力公司信息通信公司 A kind of information system health degree appraisal procedure based on FAHP_FCA combination weighting
CN111582571A (en) * 2020-04-30 2020-08-25 中国电力科学研究院有限公司 Power grid operation situation sensing method and system with model driving and data driving integrated
CN111612371A (en) * 2020-05-29 2020-09-01 国网河北省电力有限公司电力科学研究院 Intelligent ammeter quality evaluation method based on analytic hierarchy process
CN112784486A (en) * 2021-01-20 2021-05-11 中国石油大学(华东) Injection-production relation optimization method based on heterogeneous flow field representation
CN112907032A (en) * 2021-01-21 2021-06-04 中国人民解放军火箭军工程大学 Efficiency-based evaluation method for large electromagnetic ejection device
CN114723283A (en) * 2022-04-07 2022-07-08 中山大学 Ecological bearing capacity remote sensing evaluation method and device for urban group

Similar Documents

Publication Publication Date Title
CN105956757A (en) Comprehensive evaluation method for sustainable development of smart power grid based on AHP-PCA algorithm
Vučijak et al. Multicriteria decision making in selecting best solid waste management scenario: a municipal case study from Bosnia and Herzegovina
Boggia et al. Measuring sustainable development using a multi-criteria model: A case study
Khuong et al. Analyzing drivers of renewable energy development in Southeast Asia countries with correlation and decomposition methods
Currim et al. A comparative evaluation of multiattribute consumer preference models
Haas et al. Compensating for non-homogeneity in decision-making units in data envelopment analysis
Esteller-Moré et al. Tax setting in a federal system: The case of personal income taxation in Canada
van Vliet et al. Revisiting Kappa to account for change in the accuracy assessment of land-use change models
Hatefi et al. A common weight MCDA–DEA approach to construct composite indicators
Banker et al. Analysis of trends in technical and allocative efficiency: An application to Texas public school districts
Didenko et al. Analysis of rural areas development of the region using the ADL-model
CN106339779A (en) Evaluation method for distributed power supply configuration scheme in power distribution network
Read et al. Stakeholder-driven multi-attribute analysis for energy project selection under uncertainty
CN106548272A (en) A kind of electric automobile fills the evaluation methodology of facility combination property soon
Sun et al. Fuzzy set-based risk evaluation model for real estate projects
Wilhelmsson A method to derive housing sub‐markets and reduce spatial dependency
Günay et al. Measurement of financial efficiency and productivity of Turkish Public Universities by using non-parametric methods
CN108805471A (en) Evaluation method for water resources carrying capacity based on the analysis of hybrid system interactively
Foliente et al. Modelling building stock energy use and carbon emission scenarios
CN106056274A (en) Power construction main body benefit analysis method based on PCA-DEA two-dimensional comprehensive evaluation model
Ding et al. Comparison of the applicability of city-level building energy consumption quota methods
Alexandri et al. The impact of climate change mitigation policies on European labour markets
Retzlaff The use of aggregate data in comparative political analysis
CN105844373A (en) Target business district strength evaluation system and method
CN112633762A (en) Building energy efficiency obtaining method and equipment

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20160921