CN107330610A - A kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating - Google Patents
A kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating Download PDFInfo
- Publication number
- CN107330610A CN107330610A CN201710507406.2A CN201710507406A CN107330610A CN 107330610 A CN107330610 A CN 107330610A CN 201710507406 A CN201710507406 A CN 201710507406A CN 107330610 A CN107330610 A CN 107330610A
- Authority
- CN
- China
- Prior art keywords
- index
- value
- power network
- power
- saving
- 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 118
- 230000008901 benefit Effects 0.000 title claims abstract description 37
- 238000011156 evaluation Methods 0.000 claims abstract description 69
- 238000012545 processing Methods 0.000 claims abstract description 20
- 239000006185 dispersion Substances 0.000 claims abstract description 12
- 239000000203 mixture Substances 0.000 claims abstract description 8
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 7
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 6
- 238000010276 construction Methods 0.000 claims abstract description 5
- 238000002203 pretreatment Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 56
- 229910052799 carbon Inorganic materials 0.000 claims description 34
- 238000000926 separation method Methods 0.000 claims description 25
- 230000008569 process Effects 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000013459 approach Methods 0.000 claims description 7
- 239000003086 colorant Substances 0.000 claims description 7
- 239000013598 vector Substances 0.000 claims description 7
- 238000011425 standardization method Methods 0.000 claims description 5
- 230000001747 exhibiting effect Effects 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 239000000470 constituent Substances 0.000 claims description 3
- 230000001788 irregular Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000005303 weighing Methods 0.000 claims description 3
- 238000013210 evaluation model Methods 0.000 abstract description 9
- 238000004458 analytical method Methods 0.000 abstract description 4
- 238000011161 development Methods 0.000 description 18
- 238000012216 screening Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 6
- 238000010219 correlation analysis Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 230000035699 permeability Effects 0.000 description 3
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 230000036541 health Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 230000002040 relaxant effect Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000003556 assay Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000003862 health status Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000010561 standard procedure Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 230000000007 visual effect 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
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/82—Energy audits or management systems therefor
-
- 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
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 invention discloses a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating, it is characterised in that comprises the following steps：S1, achievement data pre-treatment step, including unification processing and nondimensionalization processing to data；S2, index related process step, the principal component of each system is extracted using PCA, isolates after first principal component, obtains composition coefficient of every two-level index to first principal component, retain and constitute the big two-level index of coefficient, delete and constitute the small two-level index of coefficient；S3, agriculture products weight step, including power is assigned according to the relative importance of index and power is assigned according to the dispersion degree of achievement data；S4, construction aggregate model step, using linear weighted function synthesis, with the linear weighting function of each desired value as comprehensive evaluation model, make to be able to linear compensation between each evaluation index；S5, evaluation result displaying step, analysis displaying is carried out to intermediate result and final result.
Description
Technical field
The present invention relates to Electric Power Network Planning and operation field, especially a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating.
Background technology
The involved index of power network energy-saving low-carbon development covers a wide range, and species is more, and required data volume is big, and respectively refers to
Correlation is there may be between mark.Complicated index system meets the comprehensive requirement of benefit evaluation, but policymaker is easily fallen into
Enter the mire of mass data, it is impossible to accurately and efficiently hold the development of power network energy-saving low-carbon benefit on the whole.
The content of the invention
It is an object of the invention to provide a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating, it can be broken away from aid decision making person
To the careful analysis of power network energy-saving low-carbon each side, the development of power network whole energy low-carbon is examined closely from a higher angle
Situation, holds the development progress of energy-saving low-carbon, and the energy-saving low-carbon development strategy for formulation next step provides sound assurance.
To achieve the above object, the present invention uses following technical proposals：
A kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating, comprises the following steps：
S1, achievement data pre-treatment step, including unification processing and nondimensionalization processing to data；
S2, index related process step, the principal component of each system is extracted using PCA, the first master is isolated
After composition, composition coefficient of every two-level index to first principal component is obtained, retains and constitutes the big two-level index of coefficient, delete structure
Into the small two-level index of coefficient；
S3, agriculture products weight step, including according to the relative importance of index assign power and according to achievement data from
The degree of dissipating assigns power；
S4, construction aggregate model step, using linear weighted function synthesis, with the linear weighting function of each desired value as comprehensive
Evaluation model is closed, makes to be able to linear compensation between each evaluation index；
S5, evaluation result displaying step, the tri-layer constituted using tri-colour LED method, radar map method and quantitatively evaluating value
Evaluation result methods of exhibiting.
Further, the unification processing to data specifically includes following steps：
If referring to x for an achievement data, following unification processing is done to cost type index and interval type index, can be turned to
Profit evaluation model index,
For cost type index, order
x*=M-x or
In formula, permissions or Greatest lower bound of the M for index x.
For interval type index, order
In formula, [q1,q2] interval for index x best stabilized, M, m are respectively index x permission bound.
Further, the nondimensionalization processing specifically includes following steps：All indexs have been handled by unification,
The object of nondimensionalization is only large index, if a certain index xj(j=1,2 ..., are m) large index, its observation is
{xj| i=1,2 ..., n；J=1,2 ..., m },
Standardization method
In formula,For the index sample value of nondimensionalization,sj(j=1,2 ..., m) it is respectively i-th index observation sample
This average value and mean square deviation.
Further, the step of PCA is as follows：
S201, provided with there is n two-level index under certain first class index, each two-level index has m electric network data sample (
By standardization), can obtain electric network data sample matrix is：
X=(Xij)m×nI=1,2 ..., m；J=1,2 ..., n
Wherein XijRepresent the achievement data of jth i-th of power network of index；
S202, the covariance matrix R for obtaining according to power network standardized data matrix X data sample, the matrix can reflect
Correlation between each achievement data, wherein, Rij(i, j=1,2 ..., n) it is target variable XiWith XjCoefficient correlation, R is
Real symmetric matrix (i.e. Rij=Rji), therefore triangle element thereon or lower triangle element need to be only calculated, its calculation formula is：
S203, the characteristic root λ for obtaining covariance matrix RiAnd by order arrangement from big to small, solve the spy corresponding to it
Levy vectorial li(i=1,2 ..., n), characteristic value is the variance of each principal component, and its size reflects the influence power of each principal component.
Principal component ZiContribution rate be：
Contribution rate of accumulative total is：
General selection contribution rate of accumulative total reaches the principal component corresponding to 85~95% characteristic value；
S204, by eigenvalue λiCorresponding characteristic vector liCan in the hope of correspondence principal component sampled data values, to i-th
Electric network data sample, the sample value that can try to achieve its each constituent is：
Further, it is described that power is assigned according to the relative importance of index and power shape is assigned according to the dispersion degree of achievement data
It is specially into combination weights method：If pj、qjObtained power is respectively judged based on index relative importance and data dispersion degree
Weight coefficient, then have ωj=k1pj+k2qjIn the weight coefficient obtained for combination weights, formula, k1、k2For undetermined constant, k need to be met1
＞ 0, k2＞ 0 and k1+k2=1.
Further, it is described that analytic hierarchy process (AHP) is used according to the relative importance of index tax power, comprise the following steps that：
S301, each relevant factor resolved into two levels according to different attribute from top to down：Upper strata is target
Layer, decision objective can be the energy-saving low-carbon benefit of every first class index or power network；Lower floor is indicator layer, is to wait to assign comprising factor
The indices of power；In indicator layer, each single index is done into multilevel iudge two-by-two on evaluating the importance degree of target, can be with
Judgment matrix A is obtained, specifically compares yardstick as shown in table 1, thus for a system containing n index, one can be formed
The judgment matrix of individual n × n ranks；
Yardstick table is compared in the classification of the AHP methods of table 1
| Assignment | Meaning |
| 1 | Index xk-1With xkCompare, no less important |
| 3 | Index xk-1With xkCompare, the former is somewhat more important than the latter |
| 5 | Index xk-1With xkCompare, the former is more obvious than the latter important |
| 7 | Index xk-1With xkCompare, the former is stronger than the latter important |
| 9 | Index xk-1With xkCompare, the former is more extremely important than the latter |
| 2,4,6,8 | The median of above-mentioned adjacent judgement |
| It is reciprocal | aji=1/aij |
S302, the judgement quality to matrix carry out consistency check；Weigh the quantitative index of the inconsistent degree of judgment matrix
As coincident indicator C.I., its calculation formula is：
In formula, λmaxFor the eigenvalue of maximum of judgment matrix；
S303, introducing Aver-age Random Consistency Index R.I., are modified to C.I., are calculated by a large amount of random samplings
Sample average to R.I. is as shown in table 2,
The Aver-age Random Consistency Index R.I. values of table 2
| n | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| R.I. | 0 | 0.5419 | 0.8931 | 1.1185 | 1.2494 | 1.3450 | 1.4200 |
| n | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
| R.I. | 1.4616 | 1.4874 | 1.5156 | 1.5405 | 1.5583 | 1.5779 | 1.5894 |
It will determine that the ratio between coincident indicator C.I. and same order Aver-age Random Consistency Index R.I. of matrix are referred to as random one
Sex ratio C.R. is caused, when
When, it is believed that the inconsistency of judgment matrix can receive, and otherwise be accomplished by carrying out assignment again to judgment matrix, directly
To condition for consistence is met, the judgment matrix for meeting condition for consistence, by the characteristic vector corresponding to its eigenvalue of maximum
Normalization is the weight coefficient of each evaluation index.
Further, it is described that analytic hierarchy process (AHP) is used according to the relative importance of index tax power, comprise the following steps that：
S311, most important index is selected in all indexs, make number one, be designated as x1, then from remaining index
Most important one is selected, second is come, is designated as x2, by that analogy, finally give a unique index of order relation important
Property sequencing table, is designated as X；
S312, expert are to adjacent evaluation index x in Xk-1With xkThe ratio between relative importance judged, r can be usedkCome
Represent：
rk=pk-1/pk, k=2,3 ..., n
In formula, pkFor the corresponding weight of kth evaluation index, r in index set XkImplication as shown in table 3,
The G-1 methods of table 3 compare yardstick table
| Assignment | Meaning |
| 1.0 | Index xk-1With xkCompare, no less important |
| 1.2 | Index xk-1With xkCompare, the former is somewhat more important than the latter |
| 1.4 | Index xk-1With xkCompare, the former is more obvious than the latter important |
| 1.6 | Index xk-1With xkCompare, the former is stronger than the latter important |
| 1.8 | Index xk-1With xkCompare, the former is more extremely important than the latter |
| 1.1,1.3,1.5,1.7 | The median of above-mentioned adjacent judgement |
Determine the order relation of evaluation index and to its importance degree assignment after, can be calculated by equation below and obtain each
The weight of item index：
ωk-1=rkωk, k=n, n-1 ..., 2.
Further, power is assigned according to the dispersion degree of achievement data and uses entropy assessment, comprised the following steps that：
Under S331, calculating jth index, the feature proportion of i-th of power network object：
S332, the entropy for calculating jth index：
In formula, k=1/lnm；
S333, the difference property coefficient for calculating jth index：
gj=1-ej；
S334, the weight coefficient for determining jth index
Further, tri-colour LED method specifically includes following steps：S501, determine in indicator lamp color separation lower threshold value and
Average value, finds the single index for evaluating area in other interlocal best levels and worst level, with true respectively first
Determine the upper lower threshold value of indicator lamp color separation；The average level of the upper single index is as average value using in big region, for region electricity
For net, optimal, the most bad value of the single index score in region-wide all prefecture-level power networks can be regard as indicator lamp color separation
Upper and lower threshold value, the average value of each department power network single index as indicator lamp color separation average value；
S502, indicator lamp color separation standard is determined, on indicator lamp color separation is determined on the basis of lower threshold value and average value, with
The midpoint of average value and upper threshold value is upper line of demarcation, using the midpoint of average value and lower threshold value as lower line of demarcation, obtains indicator lamp point
Colour standard；
S503, the indicator light colors for evaluating single index, compare These parameters color separation standard, can quickly and conveniently give
Go out the indicator light colors of single index, be green and healthy index when the value of single index is higher than upper line of demarcation；Work as single index
Value between up and down between line of demarcation when, be blue normal index；It is red police when the value of single index is less than lower line of demarcation
Ring mark.
Further, radar method specifically includes following steps：
S511, all indexs are converted into the consistent large index of type, and handled by nondimensionalization will be each
Index is converted into the dimensionless number between [0,1]；
S512, office's circle, draw p bar rays, every ray represents an evaluation index, each bar ray and circle from the center of circle
The intersection point in week represents the optimal level of the upper limit, the i.e. index of each desired value；
S513, the numerical value according to each index after type unification, nondimensionalization conversion, are marked each on every ray
The data point of index；
S514, the point that index value is represented on adjacent two rays is connected with straight line, ultimately form one it is in irregular shape
Polygon, the polygon be reflect comprehensive evaluation result radar map.
The beneficial effects of the invention are as follows,
1st, the present invention can eliminate difference of each achievement data in property and the order of magnitude by pretreatment, will " raw number
According to " " ripe data " are converted to, it is that the later stage overall merit of each index is ready.
2nd, original index system can be reassembled into one group of new mutual nothing by the present invention using PCA
Some overall targets closed, reflect as much as possible while can therefrom take out several less overall targets according to actual needs
The information of primal variable, so as to reduce the correlation between each index.Index system after screening is more simplified, index it
Between correlation be weakened severely, the information content that each index is covered will be greatly increased, and this aspect will cause follow-up comment
Valency process is simpler to be understood, while also increasing the confidence level of evaluation result.
3rd, the present invention assigns power and assign power according to the dispersion degree of achievement data according to the relative importance of index forms comprehensive
Close enabling legislation and avoid the shadow for easily being lacked correlation experience and knowledge by policymaker according to the entitled method of relative importance degree
Loud defect；It it also avoid have ignored the influence of policymaker's subjective factor completely according to the entitled method of data degree of scatter, with
The defect that the original intention of many Process of Comprehensive Assessment is not inconsistent.
4th, grid company can intuitively obtain all by tri-colour LED when using System of Comprehensive Evaluation
The health information of single index, so that weight lifting keeps away the main lance gently, with invigorating large enterprises while relaxing control over small ones paid close attention in current electric grid evolution
Shield and outstanding problem, i.e. red early warning index, while taking into account blue normal index.It is truly realized power network energy-saving low-carbon development management
With a definite target in view, suit the remedy to the case, so as to improve power network energy-saving low-carbon development level conscientiously.
5th, by making the radar map of indices, not only can intuitively assessment object aggregate level, also may be used
To identify the balanced development situation of each single index at a glance, it is easy to find the weakness of impact evaluation object integral level
Link；Both it can be used for the across comparison of integral level between contemporaneity, different assessment objects, and can be used for same assessment
Object, the longitudinal direction contrast in different times integral level.
Brief description of the drawings
Fig. 1 is this power network energy-saving and emission-reduction benefit method for quantitatively evaluating flow chart；
Fig. 2 is AHP methods layering schematic diagram；
Fig. 3 is tri-colour LED method color separation standard drawing figure；
Fig. 4 is the radar map for multiple attribute synthetical evaluation.
Embodiment
As shown in figure 1, a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating, the overall merit flow of power network energy-saving low-carbon
It is main to include such as the next stage：
S1, achievement data pre-treatment step
Existing very big, minimal type index, there is interval type again in power network energy-saving low-carbon benefit comprehensive evaluation model index system
Index.Before overall merit is carried out to each index, it is necessary to each index first is carried out into unification processing, otherwise can not just be specified comprehensive
Close the optimization value direction of evaluation result so that the quality of power network energy-saving low-carbon benefit can not be judged according to comprehensive evaluation result.
In addition, the unit and the order of magnitude of each single index are often different in index system so that there is incommensurability between each index, because
This needs makees appropriate nondimensionalization processing to each index, otherwise may occur in follow-up overall merit " big number eats decimal "
Phenomenon, cause the evaluation conclusion of mistake.Unification processing and nondimensionalization are handled to " data prediction " for being collectively referred to as index
Stage.The form of data has been unified in pretreatment, and " the raw data " that are got by power network actual count are converted to " the ripe number of standardization
According to ", the calculating available for next step is analyzed.
S2, index related process step
One of structure principle of index system is to avoid the juxtaposition between index, has each index good
It is representative.Therefore the correlation between index is removed, invalid index and redundancy index is weeded out for simplifying index system, simplifying and count
Calculation process, optimizing evaluation result suffer from important meaning.However, the key element of influence power network energy-saving low-carbon development is numerous, respectively
There may be inner link between element, and in order to ensure that designed index system can reflect that power network energy-saving low-carbon is imitated comprehensively
Benefit, needs to design more index again.In the power network energy-saving low-carbon benefit System of Comprehensive Evaluation set up, between each index
Certain dependency relation is there may be, i.e., the information of multiple index reflections has certain overlapping, such as main power-saving technology permeability
When index becomes excellent, the reduction of electric network synthetic Network Loss Rate will be caused to a certain extent, but can not therefore just by alternative one from
Index system is deleted, because power-saving technology permeability index also reflects each grid company main body and made in terms of energy-saving low-carbon
Effort, and the influence factor for integrating Network Loss Rate is far above the permeability one of power-saving technology.For the power network section by screening
Energy low-carbon benefit core assessment indicator system, although may still suffer from certain correlation between each index, degree is very big
Ground weakens, and each index passes through well-chosen, typically no longer needs to carry out correlation processing.
In this regard, need to carry out correlation processing to data, can be by original index system weight using PCA
Combination nova is into one group of new mutual unrelated some overall target, while can therefrom take out according to actual needs several less
Overall target reflects the information of primal variable as much as possible, so as to reduce the correlation between each index.
S3, agriculture products weight step
The determination of index weights is the key problem of integrated evaluating method.Each index is to power network energy-saving low-carbon in index system
The embodiment degree of benefit difference, therefore suitable weight coefficient should be assigned respectively, to embody different indexs to final evaluation
As a result percentage contribution.
According to the difference of focus, the determination method of index weights coefficient can be divided into three classes：
1. weight coefficient is determined according to the relative importance degree of index, its determination approach can be divided into objective approach and master again
The major class of sight approach two.Objective approach is to calculate weight coefficient according to structure proportion, mechanism formation of system etc., is not related to based on people
Sight factor.Because system is influenceed in the process of running or by external environment influence, or by subjective factor in real system, more commonly
The quality of the relative importance degree or each object of weight coefficient, i.e. policymaker first to each index is determined by subjective approach
Judgement is compared, weight coefficient is calculated further according to comparison information, conventional method includes method of characteristic, analytic hierarchy process (AHP), G1-
Method, G2- methods etc..
2. assigned and weighed according to the dispersion degree of achievement data.In the overall merit to n object, if a certain item index
Degree of fluctuation is very small, then even if being assigned to higher weights, and its influence to final appraisal results is still smaller.In order to which protrusion compares
The difference of each scheme, it is to avoid determine to be disturbed by human factor during weight coefficient, can provide information content according to each index
Size determines corresponding weight coefficient, and conventional method includes range method, entropy assessment etc..
3. the combination weights combined both.The major class method of the above two respectively has length：It is entitled according to relative importance degree
Method reflects the subjective intuition of policymaker, and but easily being lacked correlation experience and knowledge by policymaker is influenceed；According to data point
Dissipate the entitled method of degree and be based entirely on mathematical theory and method, result of calculation is objective and accurate, but have ignored policymaker completely
The influence of subjective factor, is not inconsistent with the original intention of many Process of Comprehensive Assessment.It therefore, it can consider to combine both the above method,
So that identified weight coefficient embodies the subjective information of policymaker and the objective information of data distribution simultaneously.
The information of first class index covering is relatively broad, it is impossible to obtains direct data and supports, therefore is adapted to using subjective tax
Quan Fa, this problem will be assigned to first class index using the G-1 methods without test and judge matrix uniformity and weighed.Two grades and three-level are referred to
Mark, because of directly supporting for data, in order to make full use of the comparison information that each sample data dispersion degree is brought, simultaneously
Also for assign actual conditions and expert opinion that power result meets power network, tax is integrated using what G-1 methods and entropy assessment were combined
Power method is to two grades and three-level Index Weights.Using PCA to index system carry out correlation analysis after, due to be from
Some principal components have been extracted in former achievement data, therefore entitled object is just into this some principal component.If without phase
The processing of closing property, then it is still former indices to assign power object.
S4, construction aggregate model step
, need to be according to the reality of each evaluation index for the energy-saving low-carbon benefit of comprehensively multiple power network objects of systematic analysis technique
Border influences, it is determined that corresponding weight vectors, and select suitable mathematical method construction composite evaluation function (i.e. aggregate model) meter
Comprehensive evaluation value is calculated, and multiple power network objects are ranked up by the size of value compared.
S5, evaluation result displaying step
The purpose for setting up comprehensive evaluation model is that the development for power network energy-saving low-carbon provides guidance and advised, yet with
Index number in index system is more, and indices do not have clear and definite reference threshold, and Grid manager is merely by numerous and disorderly
Index value collection can not intuitively hold the energy-saving low-carbon benefit situation of power network.In this regard, the methods of exhibiting for needing design appropriate will be commented
Valency result shows Grid manager visual in imagely.
Pointer type unification
Index in designed power network energy-saving low-carbon assessment indicator system has following three types：Profit evaluation model：Always expect
The value of index is the bigger the better；Cost type：Always it is expected that the value of index is the smaller the better；Interval type：Always expect taking for index
Value was both not too big, also should not too it is small preferably, that is, take appropriate median to be best.If referring to x for an achievement data, to cost type
Index and interval type index do following unification processing, can be turned to profit evaluation model index.
For cost type index, order
x*=M-x
Or
In formula, permissions or Greatest lower bound of the M for index x.
For interval type index, order
In formula, [q1,q2] interval for index x best stabilized, M, m are respectively index x permission bound.
Handled by above unification, cost type index and interval type index can be converted into profit evaluation model index.Herein
On the basis of, it may be determined that each level comprehensive evaluation result of index system is also that value is the bigger the better.
Indices non-dimension
The nondimensionalization of index is called the standardization for doing achievement data, standardization, is to eliminate original by mathematic(al) manipulation
The influence of index dimension.The method of indices non-dimension is more, and conventional standardization method and extreme value facture are entered at this
Row is introduced, and other methods also include linear scale method, normalized method, vectorial laws for criterion etc..
Think that all indexs have been handled by unification herein, the object of nondimensionalization is only large index.If certain
One index xj(j=1,2 ..., m) are large index, its observation is { xj| i=1,2 ..., n；J=1,2 ..., m }.
Standardization method
In formula,For the index sample value of nondimensionalization,sj(j=1,2 ..., m) it is respectively i-th index observation sample
This average value and mean square deviation.
The nondimensionalization sample value obtained by standardization method, which has just, negative, therefore it is big not to be suitable for requirement desired value
Information Entropy, the Random geometric sery method of average in zero etc..Sample mean is 0, and variance is 1.In addition, the sample interval that processing is obtained is not
It is determined that, i.e., it cannot be guaranteed that the nondimensionalization result to any initial data is in the span of a determination.
Extreme value facture
In formula, Mj、mjRespectively index xjMaximum, the minimum value of observation sample.The nondimensionalization obtained by extreme value facture
Sample maximum is 1, and minimum value is 0.
The uniformity of achievement data is particularly important, and the data of all types of indexs are handled by above-mentioned standard method
Afterwards, unified the form of data, eliminate its in nature with the difference in expressing information mode, can be straight in follow-up calculating
Connect substitution, compare.
Achievement data correlation analysis
Correlation between index is difficult directly to be judged by subjective understanding, therefore with the inherence between real data
Contact to carry out objective screening, it is more relatively reliable than carrying out subjective screening by expertise.PCA is statistics
It is middle that higher dimensional space variable index is converted into the conventional data processing method of lower dimensional space variable index.Designed evaluation index it
Between be commonly present a certain degree of correlation, in multiple variable indexs of research object, original is replaced with a few generalized variable
Higher-dimension variable is to reach the purpose of assay problem.This small number of index comprehensive former research object information as much as possible is to subtract
It is mutually independent between the distortion and loss, and index of few information so that be not in that repetition is commented in Process of Comprehensive Assessment
The phenomenon of valency.
In general, the senior index (such as first class index) in index system all recommends to form by weighing and considering,
And negligible amounts, its each single item index all contains sizable information content, and index system can all be caused by removing any one
Imperfection, therefore for senior index without screening, the rudimentary index that screening operation can directly quantify mainly for those.
For example in the presence of an index system being made up of two-stage index, only its two-level index is screened, and regarding each first class index
Corresponding two-level index is an independent system.
First with PCA, the principal component of each system is extracted.Usual first principal component can include the system
Most information, and can reflect the integrated information of system, and other principal components then comprehensive letter without reflected sample
Breath, and simply represent these information characteristics in a certain respect, to simplify calculating, it can only isolate first principal component.Separation
Go out after first principal component, obtain composition coefficient of every two-level index to first principal component.Constitute coefficient bigger, illustrate this
Two-level index is bigger for the information contribution rate of corresponding first class index, therefore should retain and constitute the big two-level index of coefficient,
Delete and constitute the small two-level index of coefficient.In general, if a certain item two-level index is small to the composition coefficient of first principal component
When 0.1, you can think that influence of the index to whole index system is very little, should give deletion.
The step of PCA, is as follows：
Provided with there is n two-level index under certain first class index, each two-level index has m electric network data sample (to pass through
Standardization), can obtain electric network data sample matrix is：
X=(Xij)m×nI=1,2 ..., m；J=1,2 ..., n
Wherein XijRepresent the achievement data of jth i-th of power network of index.
The covariance matrix R of data sample is obtained according to power network standardized data matrix X, the matrix can reflect each index
Correlation between data.Wherein, Rij(i, j=1,2 ..., n) it is target variable XiWith XjCoefficient correlation.R is real symmetrical square
Battle array (i.e. Rij=Rji), therefore triangle element thereon or lower triangle element need to be only calculated, its calculation formula is：
Obtain covariance matrix R characteristic root λiAnd by order arrangement from big to small, solve feature corresponding to it to
Measure li(i=1,2 ..., n).Characteristic value is the variance of each principal component, and its size reflects the influence power of each principal component.It is main into
Divide ZiContribution rate be：
Contribution rate of accumulative total is：
General selection contribution rate of accumulative total reaches the principal component corresponding to 85~95% characteristic value.
By eigenvalue λiCorresponding characteristic vector liCan in the hope of correspondence principal component sampled data values.To i-th of power network number
According to sample, the sample value that can try to achieve its each constituent is：
For correlation analysis, if it find that the coefficient correlation between a certain index and remaining multiple index is all very big, then
It is considered that existence information is overlapped between the index and other indexs, it is a redundancy index, should generally deletes.But
It is if the result of correlation analysis and principal component analysis are on the contrary, should be for example a certain based on the evaluation result of principal component analysis
Index occupies very big proportion in the composition of first principal component, then even if it and multiple index linear correlations, also unsuppressible-suppression.
Index system after screening is more simplified, and the correlation between index is weakened severely, and each index is covered
The information content of lid will be greatly increased, and this aspect will cause follow-up evaluation procedure is simpler to understand, be commented while also increasing
The confidence level of valency result.
Index Weights method
In order to embody difference of each index to the horizontal significance level of power network energy-saving low-carbon, appropriate weight should be assigned respectively.
Illustrate first herein according to the entitled analytic hierarchy process (AHP) of index relative importance (Analytical Hierarchy
Process, AHP), G-1 methods and the Computing Principle according to the entitled entropy assessment of achievement data dispersion degree, the two is tied afterwards
Close, as combination weights method.
Analytic hierarchy process (AHP) (AHP)
Analytic hierarchy process (AHP) be by U.S. Pittsburg college professor Satie proposed in early 1970s it is a kind of it is qualitative and
The level weight method of decision analysis being quantitatively combined.What is analysed in depth to relation between power network energy-saving low-carbon target and each index
On the basis of, as shown in Fig. 2 each relevant factor is resolved into two layers from top to down according to different attribute with reference to AHP methods
It is secondary：Upper strata is destination layer, and decision objective can be the energy-saving low-carbon benefit of every first class index or power network；Lower floor is indicator layer,
It is to treat entitled indices comprising factor.
In indicator layer, each single index is done into multilevel iudge two-by-two on evaluating the importance degree of target, can be obtained
Judgment matrix A, specifically compares yardstick as shown in table 1.Thus for a system containing n index, can be formed a n ×
The judgment matrix of n ranks.
Preferable judgment matrix should meet condition for consistence.So-called condition for consistence refers to the element tool in judgment matrix A
There is transitivity, that is, have ready conditions aijajk=aik；I, j, k=1,2 ..., n is set up.However, being limited by judge condition, actual judgement
Matrix can not usually meet condition for consistence.In this regard, needing the judgement quality to matrix to carry out consistency check.
Yardstick table is compared in the classification of table 1AHP methods
| Assignment | Meaning |
| 1 | Index xk-1With xkCompare, no less important |
| 3 | Index xk-1With xkCompare, the former is somewhat more important than the latter |
| 5 | Index xk-1With xkCompare, the former is more obvious than the latter important |
| 7 | Index xk-1With xkCompare, the former is stronger than the latter important |
| 9 | Index xk-1With xkCompare, the former is more extremely important than the latter |
| 2,4,6,8 | The median of above-mentioned adjacent judgement |
| It is reciprocal | aji=1/aij |
Weighing the quantitative index of the inconsistent degree of judgment matrix turns into coincident indicator C.I., and its calculation formula is：
In formula, λmaxFor the eigenvalue of maximum of judgment matrix.
The inconsistency of judgment matrix is related to its exponent number.In fact, the exponent number of judgment matrix is bigger, compare two-by-two between element
Relatively judge just to be more difficult to reach uniformity.In order to obtain the critical value to the applicable consistency check of different rank judgment matrix,
Also need to consider uniformity and the relation of matrix exponent number.Therefore, introducing Aver-age Random Consistency Index R.I., C.I. is repaiied
Just.The sample average for obtaining R.I. by a large amount of random samplings calculating is as shown in table 2.
The Aver-age Random Consistency Index R.I. values of table 2
| n | 2 | 3 | 4 | 5 | 6 | 7 | 8 |
| R.I. | 0 | 0.5419 | 0.8931 | 1.1185 | 1.2494 | 1.3450 | 1.4200 |
| n | 9 | 10 | 11 | 12 | 13 | 14 | 15 |
| R.I. | 1.4616 | 1.4874 | 1.5156 | 1.5405 | 1.5583 | 1.5779 | 1.5894 |
It will determine that the ratio between coincident indicator C.I. and same order Aver-age Random Consistency Index R.I. of matrix are referred to as random one
Cause sex ratio C.R..When
When, it is believed that the inconsistency of judgment matrix can receive, and otherwise be accomplished by judgment matrix
Assignment again is carried out, until meeting condition for consistence.
Judgment matrix for meeting condition for consistence, be by the characteristic vector normalization corresponding to its eigenvalue of maximum
The weight coefficient of each evaluation index.
G-1 methods
During using AHP methods, the inconsistency of judgment matrix can have a strong impact on the result of calculation of index weights.Meanwhile, with
The increase of evaluation index number, the amount of calculation of judgment matrix can also be doubled and redoubled.Therefore, can use without test and judge matrix
The G-1 methods of uniformity.
The key of G-1 methods is ranked up for the importance of each index.Most important finger is selected in all indexs first
Mark, makes number one, is designated as x1, most important one is then selected from remaining index, second is come, is designated as x2, with such
Push away, finally give the unique index importance sequencing table of an order relation, be designated as X.
Expert is to adjacent evaluation index x in Xk-1With xkThe ratio between relative importance judged, r can be usedkCarry out table
Show：
rk=pk-1/pk, k=2,3 ..., n
In formula, pkFor the corresponding weight of kth evaluation index in index set X.rkImplication it is as shown in table 3.
The G-1 methods of table 3 compare yardstick table
| Assignment | Meaning |
| 1.0 | Index xk-1With xkCompare, no less important |
| 1.2 | Index xk-1With xkCompare, the former is somewhat more important than the latter |
| 1.4 | Index xk-1With xkCompare, the former is more obvious than the latter important |
| 1.6 | Index xk-1With xkCompare, the former is stronger than the latter important |
| 1.8 | Index xk-1With xkCompare, the former is more extremely important than the latter |
| 1.1,1.3,1.5,1.7 | The median of above-mentioned adjacent judgement |
Determine the order relation of evaluation index and to its importance degree assignment after, can be calculated by equation below and obtain each
The weight of item index：
ωk-1=rkωk, k=n, n-1 ..., 2
It is to the entitled basic step of indices using G-1 methods above.
Entropy assessment
Entropy assessment is a kind of size of the information content provided according to indices observation come the side of agriculture products weight
Method.Entropy is a concept in thermodynamics, introduces information theory by Shen Nong earliest.According to the definition of information theory, comentropy then reflects
The disordering degree of information, its value is smaller, and the information utility value provided is bigger.If system is likely to be at a variety of different conditions,
And the probability that every kind of state occurs is respectively pi(i=1,2 ..., m), then the entropy of the system is just defined as：
As can be seen from the above equation, the probability occurred when the various states of system is identical, i.e. pi=1/m, (i=1,
2 ..., when m), the entropy of the system is maximum, and the information utility value now provided from the system to integrated decision-making person is minimum.
The basic thought of entropy assessment is：If a certain index entropy is smaller, illustrate the degree of variation of the achievement data sequence
It is larger, effect of the evaluation index for whole assessment models should be paid attention to, its weight also should be larger, otherwise should just reduce its weight
Coefficient.
It is still based on the electric network data sample matrix shown in formula to be analyzed, it is assumed that the sample data of each power network is equal under each index
More than zero, then the step of determining each weight coefficient based on entropy assessment is as follows：
Calculate under jth index, the feature proportion of i-th of power network object：
Calculate the entropy of jth index：
In formula, k=1/lnm.
Calculate the difference property coefficient of jth index：
gj=1-ej
Determine the weight coefficient of jth index
Combination weights method
The different information that based on entropy assessment sample data can be made full use of to be provided is can be seen that from above calculating process
System is compared, but due to have ignored the knowledge and experience of people, finally obtain each index weight coefficient may with it is pre-
The result first estimated is far from each other, and the tax power result of the enabling legislation such as AHP, G-1 method meets the expection of policymaker, but fails fully
The different information provided using sample data.In this regard, two class methods can be combined by more than, combination weights method is formed.If pj、
qjObtained weight coefficient is respectively judged based on index relative importance and data dispersion degree, then is claimed
ωj=k1pj+k2qjThe weight coefficient obtained for combination weights.In formula, k1、k2For undetermined constant, k need to be met1＞ 0,
k2＞ 0 and k1+k2=1.
Construct aggregate model
It is workable because weigthed sums approach Computing Principle is simple, and the correlation index has been handled,
Therefore embodiment selects linear weighted function aggregate model.With the linear weighting function of each desired value as comprehensive evaluation model, it can make
Linear compensation is able between each evaluation index.Influence of the design of weight coefficient to evaluation result is obvious, the larger desired value of weight
Larger is acted on to overall target.This method is suitable for situation separate between each evaluation index, and each index is not completely independent
When by due to the repetition of information between each index so that evaluation result can not objectively reflect reality.
Evaluation result shows step
According to the hierarchical structure of designed index system, the present embodiment is proposed by tri-colour LED method, radar map method and amount
Change the tri-layer evaluation result methods of exhibiting that evaluation of estimate is constituted：Tri-colour LED method is the difference according to each index value, respectively
The different instruction color of mark, the good and bad situation of single index is showed with succinct, intuitive way；Radar map method be by
Many index value is shown in the way of radar map, and comprehensive development situation of the power network on energy-saving low-carbon can be intuitively embodied comprehensively,
When evaluation index number is more, it first can do certain aggregative weighted via foregoing integrated evaluating method and handle, obtain number less
Comprehensive evaluation index, such as every first class index in foundation power network energy-saving low-carbon System of Comprehensive Evaluation, may be such that and paint
The more clear readability of the radar map made；Quantitative evaluation result is the final comprehensive evaluation result to all desired values to be evaluated,
Further weighted comprehensive is done on the basis of indices shown in radar map, the final of power network energy-saving low-carbon benefit can be obtained
Quantitatively evaluating value, the comparison of different power network sample energy-saving low-carbon benefit states of development can be realized based on the quantized value, so as to be
The development plan of power network and decision-making provide reference.
Quantitative evaluation result is the scoring to power network sample energy-saving low-carbon benefit, can pass through the tax to first class index value
Power synthesis is obtained, and can realize that power network development and the transverse and longitudinal of benefit situation compare based on the evaluation of estimate, its application method is more
Intuitively, assign power and aggregation method is also as good as with low-level index, will not be repeated here.Separately below to tri-colour LED method and thunder
Da Tufa Computing Principle and application method is discussed in detail.
Tri-colour LED method
The general principle of tri-colour LED method is same with other areas in a wider context with the objective area single index
Index of classification is compared, and the single index value in all regions calculated in measurement period is carried out into ranking；According to objective area
The good and bad situation of index comparing result in a wider context, using green, yellow, red three kinds of colors indicator lamp, intuitively
Single index is shown in health status, normal condition or the state of alert.
The evaluation rubric of tri-colour LED method：
1. lower threshold value and average value in indicator lamp color separation are determined.The single index for evaluating area is found respectively first at it
Its interlocal best level and worst level, to determine the upper lower threshold value of indicator lamp color separation；Referred to upper individual event in big region
Target average level is used as average value., can be by region-wide (province) all prefecture-level power networks for region (province) power network
Optimal, the most bad value of the single index score is used as the upper and lower threshold value of indicator lamp color separation, each province (area) the power network single index
Average value as indicator lamp color separation average value.
2. indicator lamp color separation standard is determined.On indicator lamp color separation is determined on the basis of lower threshold value and average value, with flat
The midpoint of average and upper threshold value is upper line of demarcation, using the midpoint of average value and lower threshold value as lower line of demarcation, obtains indicator lamp color separation
Standard, as shown in Figure 3.
3. the indicator light colors of single index are evaluated.These parameters color separation standard is compared, list can be quickly and conveniently provided
The indicator light colors of item index.It is green and healthy index when the value of single index is higher than upper line of demarcation；When the value of single index
It is blue normal index when between line of demarcation up and down；It is that red alert refers to when the value of single index is less than lower line of demarcation
Mark.Accordingly, grid company can intuitively obtain all lists when using System of Comprehensive Evaluation by tri-colour LED
The health information of item index, so that weight lifting keeps away the principal contradiction gently, with invigorating large enterprises while relaxing control over small ones paid close attention in current electric grid evolution
And outstanding problem, i.e. red early warning index, while taking into account blue normal index.Being truly realized power network energy-saving low-carbon development management has
Put arrow, suit the remedy to the case so that conscientiously improve power network energy-saving low-carbon development level.
Radar map method
Radar map method is a kind of multivariable comparative analysis, Comprehensive Assessment Technology, because the figure likeness in form that the technology is used is led
Boat radar display screen on figure and gain the name.Traditional radar map method is typical pattern evaluation method, and as a kind of qualitative
Method is applied to overall merit, by first drawing the radar map of evaluation object indices, then all kinds of by compareing by estimator
Typical radar figure or the radar map for obtaining the assessment object in history, qualitatively provide the comprehensive evaluation result of evaluation object.
The implementing procedure of conventional radar figure method is as follows：
1. all indexs are converted into the consistent large index of type, and handled by nondimensionalization by each index
It is converted into the dimensionless number between [0,1]；
2. office justifies, and p bar rays are drawn from the center of circle, and every ray represents an evaluation index, each bar ray and circumference
Intersection point represent each desired value the upper limit, the i.e. index optimal level；
3. the numerical value according to each index after type unification, nondimensionalization conversion, marks each finger on every ray
Target data point；
4. the point that index value is represented on adjacent two rays is connected with straight line, ultimately form one it is in irregular shape many
Side shape, as shown in figure 4, the polygon is the radar map for reflecting comprehensive evaluation result.
Power network energy-saving low-carbon index system scale is more huge, and bottom index number is more.In order to utilize radar well
The figure advantage directly perceived for showing multiple index evaluation result, only carries out radar map displaying to relative small number of first class index.Commenting
In valency method, each bottom index (three-level or two-level index) data standardization is subjected to, the data obtained is in [0,1]
It is interval.Further, tax power is carried out to indexs at different levels, then the evaluation of estimate of first class index can be obtained by the synthesis of each bottom desired value.
The evaluation of estimate for the first class index so tried to achieve meets the use condition of radar map still in the range of [0,1].In addition, also needing root
Resulting first class index value is modified according to the weight of first class index, will each first class index value be multiplied by its weight, obtain
Finally should be in numerical value shown in radar map.
On conventional radar figure, the distance in each index data point and the center of circle represents the good and bad situation of each index；Radar map
More convergence is cylindrical, illustrates that the level of aggregation for assessing object is higher；The shape of radar map more levels off to circle, represents and assesses object
Indices development is more balanced.By making the radar map of indices, not only can intuitively assessment object totality
Level, the balanced development situation of each single index can also be identified at a glance, it is easy to find that impact evaluation object is overall
The weak link of level；Both it can be used for the across comparison of integral level between contemporaneity, different assessment objects, can also use
In same assessment object, the longitudinal direction contrast in different times integral level.
Although above-mentioned the embodiment of the present invention is described with reference to accompanying drawing, not to present invention protection model
The limitation enclosed, one of ordinary skill in the art should be understood that on the basis of technical scheme those skilled in the art are not
Need to pay various modifications or deform still within protection scope of the present invention that creative work can make.
Claims (10)
1. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating, it is characterised in that comprise the following steps：
S1, achievement data pre-treatment step, including unification processing and nondimensionalization processing to data；
S2, index related process step, the principal component of each system is extracted using PCA, first principal component is isolated
Afterwards, composition coefficient of every two-level index to first principal component is obtained, retains and constitutes the big two-level index of coefficient, delete and constitute system
The small two-level index of number；
S3, agriculture products weight step, including power is assigned and according to the discrete journey of achievement data according to the relative importance of index
Degree assigns power；
S4, construction aggregate model step, using linear weighted function synthesis, are commented with the linear weighting function of each desired value as synthesis
Valency model, makes to be able to linear compensation between each evaluation index；
S5, evaluation result displaying step, the tri-layer evaluation constituted using tri-colour LED method, radar map method and quantitatively evaluating value
As a result methods of exhibiting.
2. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 1, it is characterised in that described to data
Unification processing specifically include following steps：
If referring to x for an achievement data, following unification processing is done to cost type index and interval type index, benefit can be turned to
Type index,
For cost type index, order
x*=M-x or
In formula, permissions or Greatest lower bound of the M for index x.
For interval type index, order
In formula, [q1,q2] interval for index x best stabilized, M, m are respectively index x permission bound.
3. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 1, it is characterised in that the dimensionless
Change processing specifically includes following steps：All indexs have been handled by unification, and the object of nondimensionalization is only that large refers to
Mark, if a certain index xj(j=1,2 ..., m) are large index, its observation is { xj| i=1,2 ..., n；J=1,2 ...,
M },
Standardization method
In formula,For the index sample value of nondimensionalization,sj(j=1,2 ..., m) it is respectively i-th index observation sample
Average value and mean square deviation.
4. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 1, it is characterised in that the principal component
The step of analytic approach, is as follows：
S201, provided with there is n two-level index under certain first class index, each two-level index has m electric network data sample (to pass through
Standardization), can obtain electric network data sample matrix is：
X=(Xij)m×nI=1,2 ..., m；J=1,2 ..., n
Wherein XijRepresent the achievement data of jth i-th of power network of index；
S202, the covariance matrix R for obtaining according to power network standardized data matrix X data sample, the matrix can reflect each finger
The correlation between data is marked, wherein, Rij(i, j=1,2 ..., n) it is target variable XiWith XjCoefficient correlation, R is real symmetrical
Matrix (i.e. Rij=Rji), therefore triangle element thereon or lower triangle element need to be only calculated, its calculation formula is：
S203, the characteristic root λ for obtaining covariance matrix RiAnd by order arrangement from big to small, solve feature corresponding to it to
Measure li(i=1,2 ..., n), characteristic value is the variance of each principal component, and its size reflects the influence power of each principal component.It is main into
Divide ZiContribution rate be：
Contribution rate of accumulative total is：
General selection contribution rate of accumulative total reaches the principal component corresponding to 85~95% characteristic value；
S204, by eigenvalue λiCorresponding characteristic vector liCan in the hope of correspondence principal component sampled data values, to i-th of power network
Data sample, the sample value that can try to achieve its each constituent is：
5. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 1, it is characterised in that described according to finger
Target relative importance assigns power and forms combination weights method according to the dispersion degree of achievement data tax power：If pj、qjPoint
Obtained weight coefficient Wei not be judged based on index relative importance and data dispersion degree, then has ωj=k1pj+k2qjTo be comprehensive
Close in the weight coefficient that the power of tax is obtained, formula, k1、k2For undetermined constant, k need to be met1＞ 0, k2＞ 0 and k1+k2=1.
6. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 5, it is characterised in that described according to finger
Target relative importance assigns power and uses analytic hierarchy process (AHP), comprises the following steps that：
S301, each relevant factor resolved into two levels according to different attribute from top to down：Upper strata is destination layer, certainly
Plan target can be the energy-saving low-carbon benefit of every first class index or power network；Lower floor is indicator layer, is to treat entitled comprising factor
Indices；In indicator layer, each single index is done into multilevel iudge two-by-two on evaluating the importance degree of target, can be obtained
Judgment matrix A, specifically compare yardstick as shown in table 1, thus for a system containing n index, can be formed a n ×
The judgment matrix of n ranks；
Yardstick table is compared in the classification of the AHP methods of table 1
S302, the judgement quality to matrix carry out consistency check；Weighing the quantitative index of the inconsistent degree of judgment matrix turns into
Coincident indicator C.I., its calculation formula is：
In formula, λmaxFor the eigenvalue of maximum of judgment matrix；
S303, introducing Aver-age Random Consistency Index R.I., are modified to C.I., are calculated and obtained by a large amount of random samplings
R.I. sample average is as shown in table 2,
The Aver-age Random Consistency Index R.I. values of table 2
It will determine that the ratio between coincident indicator C.I. and same order Aver-age Random Consistency Index R.I. of matrix are referred to as random uniformity
Ratio C.R., when
When, it is believed that the inconsistency of judgment matrix can receive, and otherwise be accomplished by carrying out assignment again to judgment matrix, until full
Sufficient condition for consistence, the judgment matrix for meeting condition for consistence, by the characteristic vector normalizing corresponding to its eigenvalue of maximum
Change the weight coefficient of as each evaluation index.
7. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 5, it is characterised in that described according to finger
Target relative importance assigns power and uses analytic hierarchy process (AHP), comprises the following steps that：
S311, most important index is selected in all indexs, make number one, be designated as x1, then selected most from remaining index
Important one, comes second, is designated as x2, by that analogy, finally give the unique index importance sequence of an order relation
Table, is designated as X；
S312, expert are to adjacent evaluation index x in Xk-1With xkThe ratio between relative importance judged, r can be usedkCarry out table
Show：
rk=pk-1/pk, k=2,3 ..., n
In formula, pkFor the corresponding weight of kth evaluation index, r in index set XkImplication as shown in table 3,
The G-1 methods of table 3 compare yardstick table
Determine the order relation of evaluation index and to its importance degree assignment after, can be calculated by equation below and obtain items and refer to
Target weight：
ωk-1=rkωk, k=n, n-1 ..., 2.
8. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 5, it is characterised in that according to index number
According to dispersion degree assign power use entropy assessment, comprise the following steps that：
Under S331, calculating jth index, the feature proportion of i-th of power network object：
S332, the entropy for calculating jth index：
In formula, k=1/lnm；
S333, the difference property coefficient for calculating jth index：
gj=1-ej；
S334, the weight coefficient for determining jth index
9. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 1, it is characterised in that three color refers to
Show that lamp method specifically includes following steps：
S501, lower threshold value and average value in indicator lamp color separation are determined, find the single index for evaluating area respectively first at it
Its interlocal best level and worst level, to determine the upper lower threshold value of indicator lamp color separation；Referred to upper individual event in big region
Target average level is as average value, for regional power grid, can refer to the individual event in region-wide all prefecture-level power networks
Optimal, the most bad value of score is marked as the upper and lower threshold value of indicator lamp color separation, the average value conduct of each department power network single index
The average value of indicator lamp color separation；
S502, indicator lamp color separation standard is determined, on indicator lamp color separation is determined on the basis of lower threshold value and average value, with average
The midpoint of value and upper threshold value is upper line of demarcation, using the midpoint of average value and lower threshold value as lower line of demarcation, obtains indicator lamp color separation mark
It is accurate；
S503, the indicator light colors for evaluating single index, compare These parameters color separation standard, can quickly and conveniently provide list
The indicator light colors of item index, are green and healthy index when the value of single index is higher than upper line of demarcation；When the value of single index
It is blue normal index when between line of demarcation up and down；It is that red alert refers to when the value of single index is less than lower line of demarcation
Mark.
10. a kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating as claimed in claim 1, it is characterised in that the radar
Method specifically includes following steps：
S511, all indexs are converted into the consistent large index of type, and are handled by nondimensionalization by each index
It is converted into the dimensionless number between [0,1]；
S512, office's circle, draw p bar rays, every ray represents an evaluation index, each bar ray and circumference from the center of circle
Intersection point represents the optimal level of the upper limit of each desired value, the i.e. index；
S513, the numerical value according to each index after type unification, nondimensionalization conversion, mark each index on every ray
Data point；
S514, the point that index value is represented on adjacent two rays is connected with straight line, ultimately form one it is in irregular shape many
Side shape, the polygon is the radar map for reflecting comprehensive evaluation result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710507406.2A CN107330610A (en) | 2017-06-28 | 2017-06-28 | A kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710507406.2A CN107330610A (en) | 2017-06-28 | 2017-06-28 | A kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107330610A true CN107330610A (en) | 2017-11-07 |
Family
ID=60197178
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710507406.2A Pending CN107330610A (en) | 2017-06-28 | 2017-06-28 | A kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107330610A (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109034649A (en) * | 2018-08-14 | 2018-12-18 | 中国南方电网有限责任公司超高压输电公司 | A kind of energy-saving power transmission network benefit comprehensive estimation method |
| CN109064020A (en) * | 2018-08-02 | 2018-12-21 | 中邮建技术有限公司 | A kind of Energy efficiency evaluation method and apparatus |
| CN109147081A (en) * | 2018-09-03 | 2019-01-04 | 深圳市智物联网络有限公司 | A kind of equipment operation stability analysis method and system |
| CN109377034A (en) * | 2018-10-11 | 2019-02-22 | 国网新疆电力有限公司信息通信公司 | Risk profiles appraisal procedure based on smart grid information communication system |
| CN109492889A (en) * | 2018-08-31 | 2019-03-19 | 国网电力科学研究院(武汉)能效测评有限公司 | A kind of method for building up and system of energy conservation service project influence property Integrated Evaluation Model |
| CN109636189A (en) * | 2018-12-13 | 2019-04-16 | 国网山东省电力公司经济技术研究院 | A kind of Transmission Expansion Planning in Electric schemes evaluation method and system based on weight comprehensive assessment |
| CN109636126A (en) * | 2018-11-19 | 2019-04-16 | 南方电网能源发展研究院有限责任公司 | Quantization method, device, equipment and the storage medium of increment power distribution network investment decision |
| CN109657967A (en) * | 2018-12-13 | 2019-04-19 | 国网山东省电力公司经济技术研究院 | A kind of confirmation method and system of Transmission Expansion Planning in Electric evaluating indexesto scheme weight |
| CN109727059A (en) * | 2018-11-20 | 2019-05-07 | 北京云和互动信息技术有限公司 | A kind of evaluation method and system based on big data |
| CN111191944A (en) * | 2020-01-02 | 2020-05-22 | 中国科学院地理科学与资源研究所 | County development activity evaluation method and problem county classification statistical method |
| CN111967777A (en) * | 2020-08-19 | 2020-11-20 | 国网河南省电力公司经济技术研究院 | Comprehensive evaluation method for energy storage and promotion of clean energy utilization value |
| CN112766785A (en) * | 2021-01-28 | 2021-05-07 | 中国人寿保险股份有限公司上海数据中心 | Quality evaluation method, system, device and storage medium for insurance data |
| CN114219245A (en) * | 2021-12-02 | 2022-03-22 | 国网浙江省电力有限公司 | Village power index evaluation method and device based on big data and storage medium |
| CN115481906A (en) * | 2022-09-22 | 2022-12-16 | 中航机载系统共性技术有限公司 | Production line evaluation method, device, equipment and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104376413A (en) * | 2014-11-11 | 2015-02-25 | 国家电网公司 | Power grid planning scheme evaluation system based on analytic hierarchy process and data envelopment analysis |
| US20160226757A1 (en) * | 2015-01-29 | 2016-08-04 | Dell Products L.P. | System and method for energy cost aware routing |
-
2017
- 2017-06-28 CN CN201710507406.2A patent/CN107330610A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104376413A (en) * | 2014-11-11 | 2015-02-25 | 国家电网公司 | Power grid planning scheme evaluation system based on analytic hierarchy process and data envelopment analysis |
| US20160226757A1 (en) * | 2015-01-29 | 2016-08-04 | Dell Products L.P. | System and method for energy cost aware routing |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109064020A (en) * | 2018-08-02 | 2018-12-21 | 中邮建技术有限公司 | A kind of Energy efficiency evaluation method and apparatus |
| CN109034649A (en) * | 2018-08-14 | 2018-12-18 | 中国南方电网有限责任公司超高压输电公司 | A kind of energy-saving power transmission network benefit comprehensive estimation method |
| CN109492889A (en) * | 2018-08-31 | 2019-03-19 | 国网电力科学研究院(武汉)能效测评有限公司 | A kind of method for building up and system of energy conservation service project influence property Integrated Evaluation Model |
| CN109147081A (en) * | 2018-09-03 | 2019-01-04 | 深圳市智物联网络有限公司 | A kind of equipment operation stability analysis method and system |
| CN109147081B (en) * | 2018-09-03 | 2021-02-26 | 深圳市智物联网络有限公司 | Equipment operation stability analysis method and system |
| CN109377034A (en) * | 2018-10-11 | 2019-02-22 | 国网新疆电力有限公司信息通信公司 | Risk profiles appraisal procedure based on smart grid information communication system |
| CN109636126A (en) * | 2018-11-19 | 2019-04-16 | 南方电网能源发展研究院有限责任公司 | Quantization method, device, equipment and the storage medium of increment power distribution network investment decision |
| CN109727059A (en) * | 2018-11-20 | 2019-05-07 | 北京云和互动信息技术有限公司 | A kind of evaluation method and system based on big data |
| CN109657967A (en) * | 2018-12-13 | 2019-04-19 | 国网山东省电力公司经济技术研究院 | A kind of confirmation method and system of Transmission Expansion Planning in Electric evaluating indexesto scheme weight |
| CN109636189A (en) * | 2018-12-13 | 2019-04-16 | 国网山东省电力公司经济技术研究院 | A kind of Transmission Expansion Planning in Electric schemes evaluation method and system based on weight comprehensive assessment |
| CN111191944A (en) * | 2020-01-02 | 2020-05-22 | 中国科学院地理科学与资源研究所 | County development activity evaluation method and problem county classification statistical method |
| CN111967777A (en) * | 2020-08-19 | 2020-11-20 | 国网河南省电力公司经济技术研究院 | Comprehensive evaluation method for energy storage and promotion of clean energy utilization value |
| CN111967777B (en) * | 2020-08-19 | 2023-10-27 | 国网河南省电力公司经济技术研究院 | Comprehensive evaluation method for energy storage and clean energy utilization value promotion |
| CN112766785A (en) * | 2021-01-28 | 2021-05-07 | 中国人寿保险股份有限公司上海数据中心 | Quality evaluation method, system, device and storage medium for insurance data |
| CN112766785B (en) * | 2021-01-28 | 2024-03-15 | 中国人寿保险股份有限公司上海数据中心 | Quality evaluation method, system, equipment and storage medium for insurance data |
| CN114219245A (en) * | 2021-12-02 | 2022-03-22 | 国网浙江省电力有限公司 | Village power index evaluation method and device based on big data and storage medium |
| CN114219245B (en) * | 2021-12-02 | 2023-11-14 | 国网浙江省电力有限公司 | Rural power index evaluation method and device based on big data and storage medium |
| CN115481906A (en) * | 2022-09-22 | 2022-12-16 | 中航机载系统共性技术有限公司 | Production line evaluation method, device, equipment and medium |
| CN115481906B (en) * | 2022-09-22 | 2024-04-16 | 中航机载系统共性技术有限公司 | Production line evaluation method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107330610A (en) | A kind of power network energy-saving and emission-reduction benefit method for quantitatively evaluating | |
| CN110008442A (en) | A kind of pilot training's control method based on pilot's comprehensive ability evaluation | |
| CN103679558B (en) | Electric automobile charging and conversion electric station fire risk data evaluation method | |
| CN105512465B (en) | Based on the cloud platform safety quantitative estimation method for improving VIKOR methods | |
| CN101853290A (en) | Meteorological service performance evaluation method based on geographical information system (GIS) | |
| CN104217122A (en) | Tunnel construction process safety evaluation method based on multi-element information early warning system | |
| Alqahtani et al. | Artificial neural networks incorporating cost significant items towards enhancing estimation for (life-cycle) costing of construction projects | |
| Lahdelma et al. | Stochastic multicriteria acceptability analysis (SMAA) | |
| Bag | A short review on structural equation modeling: applications and future research directions | |
| CN109146158A (en) | A kind of Alpine Meadow ecosystem health analysis method, computer | |
| CN104239712A (en) | Real-time evaluation method for anti-interference performance of radar | |
| CN109636189A (en) | A kind of Transmission Expansion Planning in Electric schemes evaluation method and system based on weight comprehensive assessment | |
| CN107295537A (en) | A kind of method and system for wireless sensor network reliability of testing and assessing | |
| CN106970986A (en) | Urban waterlogging influence degree method for digging and system based on deep learning | |
| Mustajoki et al. | Using intervals for global sensitivity and worst-case analyses in multiattribute value trees | |
| CN108805471A (en) | Evaluation method for water resources carrying capacity based on the analysis of hybrid system interactively | |
| CN116665489A (en) | Method for identifying congestion area of airway network | |
| Amarullah et al. | Planning decision support system using building mall AHP (Analytical Hierarchy Process) | |
| Hatefi | A typology scheme for the criteria weighting methods in MADM | |
| Yang et al. | Ecological risk assessment based on road network development analysis of Xiamen city, China | |
| CN112508336A (en) | Space and environment efficiency correlation measurement method based on structural equation model | |
| CN116109212A (en) | Airport operation efficiency evaluation index design and monitoring method | |
| Yazdi et al. | Evaluating the performance of oil and gas companies by an extended balanced scorecard and the hesitant fuzzy best-worst method | |
| CN108710987A (en) | Attribute decision making method | |
| Ahmad et al. | Comparative Analysis of Intrusion Detection Approaches |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171107 |
|
| RJ01 | Rejection of invention patent application after publication |