CN106250353A - A kind of entropy weight computational methods and Multiobjective Decision Making Method - Google Patents
A kind of entropy weight computational methods and Multiobjective Decision Making Method Download PDFInfo
- Publication number
- CN106250353A CN106250353A CN201610651441.7A CN201610651441A CN106250353A CN 106250353 A CN106250353 A CN 106250353A CN 201610651441 A CN201610651441 A CN 201610651441A CN 106250353 A CN106250353 A CN 106250353A
- Authority
- CN
- China
- Prior art keywords
- target
- entropy
- omega
- solution
- represent
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/15—Correlation function computation including computation of convolution operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Computational Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of entropy weight computational methods and device, the method includes: obtain the evaluation index being evaluated object, and the entropy of each evaluation index;Calculate the entropy weight of each evaluation index according to the following formula: Wherein, ωejRepresent the entropy weight of evaluation index j, HjRepresent the entropy of evaluation index j,Representing the average of all entropy not being 1, m represents the quantity of evaluation index, j=1,2 ..., m.Thus, power all can be rationally composed when entropy is bigger than normal or less than normal, solve the entropy when all evaluation indexes that entropy assessment in prior art exists level off to 1 time, the minute differences of entropy can cause the change at double of corresponding entropy weight, and then the problem causing entropy weight unreasonable distribution.It addition, the embodiment of the present invention additionally provides a kind of Multiobjective Decision Making Method realized based on technique scheme and device, simple and efficient, it is especially suitable for running online.
Description
Technical field
The present invention relates to entropy assessment technical field, more particularly, it relates to a kind of entropy weight computational methods and multiobjective decision-making
Method.
Background technology
Entropy (Entropy) itself is thermodynamics concept, and Shannon (C.E.Shannon) was introduced in theory of information later, composes
Give the concept of entropy broad sense.According to the thought of entropy, it is to determine decision accuracy that people obtain the quality and quantity of information in decision-making
One of key factor with reliability.Entropy assessment is to utilize entropy can be formed with the feature of the useful information amount that metric data is provided
A kind of method of Objective Weight.In the case of given evaluation object collection and various evaluation index value all determine, from information angle
From the point of view of Du, the weight using entropy assessment to give certain index represents this index and passes to the size of policymaker's quantity of information, i.e. one finger
The difference degree being marked between each evaluation object is the biggest, and the quantity of information that it comprises is the biggest, and its entropy is the least, therefore this index
The weight obtained just should be the biggest, and it is the biggest on the impact of assessment result certainly.
Assuming to be evaluated in the appraisement system of object at m item evaluation index, n, iotave evaluation matrix is Dnm, to this matrix
It is standardized process and obtains Standard Process Rnm.According to the definition of entropy, determine the entropy of evaluation index j:
Wherein, HjRepresent the entropy of evaluation index j, 0≤Hj≤ 1, for making lnfijMeaningful, it is assumed that to work as fijWhen=0,
fijlnfij=0;I=1,2 ..., n;J=1,2 ..., m.
The entropy weight of entropy Calculation Estimation index j of Utilization assessment index j:
Wherein, ωejRepresent the entropy weight of evaluation index j, 0≤ωej≤ 1, and
According to entropy weight calculating formula above, as the entropy H of all evaluation indexesj→ 1 (j=1,2 ..., time m), entropy
Minute differences will cause the change at double of corresponding entropy weight, such as: when evaluation index entropy vector be respectively (0.9999,
0.9998,0.9997) and time (0.9000,0.8000,0.7000), although the difference between entropy is different, but the entropy of both correspondences
Weight vector is (0.1667,0.3333,0.5000), and this is the most irrational.And rational pass between entropy weight and entropy
System should be: one is that different evaluation index entropy difference means that its quantity of information provided is essentially identical the most at most, then corresponding entropy
Power also should be essentially identical;Two is that the quantity of information that different entropy vector representation provides is different, therefore should have different entropy weight to
Amount.
In sum, in prior art entropy assessment exist when the entropy of all evaluation indexes level off to 1 time, entropy small
Difference can cause the change at double of corresponding entropy weight, and then the problem causing entropy weight unreasonable distribution.
Summary of the invention
It is an object of the invention to provide a kind of entropy weight computational methods and device, to solve what entropy assessment in prior art existed
When the entropy of all evaluation indexes level off to 1 time, the minute differences of entropy can cause the change at double of corresponding entropy weight, and then cause
The problem of entropy weight unreasonable distribution;It addition, present invention also offers a kind of Multiobjective Decision Making Method and device.
To achieve these goals, the present invention provides following technical scheme:
A kind of entropy weight computational methods, including:
Obtain the evaluation index being evaluated object, and the entropy of each described evaluation index;
Calculate the entropy weight of each described evaluation index according to the following formula:
Wherein, ωejRepresent the entropy weight of evaluation index j, HjRepresent the entropy of evaluation index j,Represent all be not 1 entropy
The average of value, m represents the quantity of described evaluation index, j=1,2 ..., m.
A kind of Multiobjective Decision Making Method, including:
Obtain and advance with the Pareto disaggregation that multi-objective Algorithm obtains, and described Pareto disaggregation is standardized place
Reason, obtains Standard Process, and in this matrix, every a line represents a solution, total n solution, and every string represents a target, total m
Individual target, each solution has m the desired value corresponding with m target;
Calculate the entropy of each target in described Standard Process, and utilize the entropy weight of the following equation each target of calculating:
Wherein, ωejRepresent the entropy weight of target j, HjRepresent the entropy of target j,For the average of all entropy not being 1, m
The quantity of expression target, j=1,2 ..., m;
Entropy weight based on each target and desired value corresponding to each solution calculate the comprehensive assessment value of each solution, and choose
Solution corresponding to big comprehensive assessment value is last solution.
Preferably, entropy weight based on each target and desired value corresponding to each solution calculate the comprehensive assessment value of each solution,
Including:
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the target solving i correspondence target j in described Standard Process
Value.
Preferably, entropy weight based on each target and desired value corresponding to each solution calculate the comprehensive assessment value of each solution,
Including:
Obtain the subjective weight given for each target of external world's input, and calculate combining of each target according to the following formula
Conjunction weight:
Wherein, ωjRepresent the comprehensive weight of target j, ωsjRepresent the subjective weight of target j;
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the target solving i correspondence target j in described Standard Process
Value.
Preferably, it is standardized described Pareto disaggregation processing, obtains Standard Process, including:
If described Multiobjective Decision Making Method specially maximizes optimization problem, the most according to the following formula to described Pareto
Disaggregation is standardized processing:
If described Multiobjective Decision Making Method specially minimizes optimization problem, the most according to the following formula to described Pareto
Disaggregation is standardized processing:
Wherein, rijRepresent the desired value solving i correspondence target j in Standard Process, xijRepresent that Pareto solution is concentrated and solve i pair
Answer the desired value of target j, 1≤i≤n, 1≤j≤m.
A kind of entropy weight calculates device, including:
Data acquisition module, for obtaining the evaluation index being evaluated object, and the entropy of each described evaluation index;
Entropy weight computing module, for calculating the entropy weight of each described evaluation index according to the following formula:
Wherein, ωejRepresent the entropy weight of evaluation index j, HjRepresent the entropy of evaluation index j,Represent all be not 1 entropy
The average of value, m represents the quantity of described evaluation index, j=1,2 ..., m.
A kind of multiobjective decision-making device, including:
Matrix disposal module, advances with, for obtaining, the Pareto disaggregation that multi-objective Algorithm obtains, and tires out described handkerchief
Torr disaggregation is standardized processing, and obtains Standard Process, and in this matrix, every a line represents a solution, total n solution, every string
Representing a target, total m target, each solution has m the desired value corresponding with m target;
Entropy weight acquisition module, for calculating the entropy of each target in described Standard Process, and utilizes following equation meter
Calculate the entropy weight of each target:
Wherein, ωejRepresent the entropy weight of target j, HjRepresent the entropy of target j,For the average of all entropy not being 1, m
The quantity of expression target, j=1,2 ..., m;
Last solution acquisition module, the desired value corresponding for entropy weight based on each target and each solution calculates each solution
Comprehensive assessment value, and solution corresponding to comprehensive assessment value choosing maximum be last solution.
Preferably, last solution acquisition module includes:
First assessment unit, for calculating the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the target solving i correspondence target j in described Standard Process
Value.
Preferably, last solution acquisition module includes:
Second assessment unit, calculates combining of each solution for the desired value that entropy weight based on each target and each solution are corresponding
Close assessed value, including:
Obtain the subjective weight given for each target of external world's input, and calculate combining of each target according to the following formula
Conjunction weight:
Wherein, ωjRepresent the comprehensive weight of target j, ωsjRepresent the subjective weight of target j;
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the target solving i correspondence target j in described Standard Process
Value.
Preferably, matrix disposal module includes:
First processing unit, if being used for solving maximizing optimization problem for described multiobjective decision-making device, then according to
Described Pareto disaggregation is standardized processing by following equation:
Second processing unit, if minimizing optimization problem for described multiobjective decision-making device for solution, then according to
Described Pareto disaggregation is standardized processing by following equation:
Wherein, rijRepresent the desired value solving i correspondence target j in Standard Process, xijRepresent that Pareto solution is concentrated and solve i pair
Answer the desired value of target j, 1≤i≤n, 1≤j≤m.
The invention provides a kind of entropy weight computational methods and device, wherein, the method includes: obtains and is evaluated commenting of object
Valency index, and the entropy of each described evaluation index;Calculate the entropy weight of each described evaluation index according to the following formula:
Wherein, ωejRepresent the entropy weight of evaluation index j, HjRepresent the entropy of evaluation index j,Represent all be not 1 entropy
The average of value, m represents the quantity of described evaluation index, j=1,2 ..., m.Technique scheme disclosed by the invention, in background
Technology adds in the single tax power mode of the formula of original calculating entropy weight tax the power mode, i.e. ω of a kind of complementationej1With
ωej2;And use the mean control two kinds of the entropy of each evaluation index to compose the ratio shared by power mode, i.e. use each evaluation index
The average of entropy as weight so that the ω when entropy is close to 0ej2Shared ratio reduces, when entropy is close to 1
ωej1Proportion reduces, and then the scheme calculating entropy weight in the present invention all can rationally be composed when entropy is bigger than normal or less than normal
Power, solve the entropy when all evaluation indexes that entropy assessment in prior art exists level off to 1 time, the minute differences meeting of entropy
Cause the change at double of corresponding entropy weight, and then the problem causing entropy weight unreasonable distribution.It addition, the embodiment of the present invention additionally provides
A kind of Multiobjective Decision Making Method and device, wherein, the method includes: obtains and advances with the Pareto solution that multi-objective Algorithm obtains
Collection, and be standardized described Pareto disaggregation processing, obtaining Standard Process, in this matrix, every a line represents a solution,
Total n solution, every string represents a target, and total m target, each solution has m the desired value corresponding with m target;
Calculate the entropy of each target in described Standard Process, and utilize the entropy weight of the following equation each target of calculating:
Wherein, ωejRepresent the entropy weight of target j, HjRepresent the entropy of target j,For the average of all entropy not being 1, m
The quantity of expression target, j=1,2 ..., m;Entropy weight based on each target and desired value corresponding to each solution calculate each solution
Comprehensive assessment value, and solution corresponding to comprehensive assessment value choosing maximum be last solution.The technique scheme letter that the present invention provides
Single and efficient, it is especially suitable for running online.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
In having technology to describe, the required accompanying drawing used is briefly described, it should be apparent that, the accompanying drawing in describing below is only this
Inventive embodiment, for those of ordinary skill in the art, on the premise of not paying creative work, it is also possible to according to
The accompanying drawing provided obtains other accompanying drawing.
The flow chart of a kind of entropy weight computational methods that Fig. 1 provides for the embodiment of the present invention;
Fig. 2 calculates the structural representation of device for a kind of entropy weight that the embodiment of the present invention provides;
The flow chart of a kind of Multiobjective Decision Making Method that Fig. 3 provides for the embodiment of the present invention;
The structural representation of a kind of multiobjective decision-making device that Fig. 4 provides for the embodiment of the present invention.
Detailed description of the invention
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Describe, it is clear that described embodiment is only a part of embodiment of the present invention rather than whole embodiments wholely.Based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under not making creative work premise
Embodiment, broadly falls into the scope of protection of the invention.
Refer to Fig. 1, it illustrates a kind of entropy weight computational methods that the embodiment of the present invention provides, following step can be included
Rapid:
S11: obtain the evaluation index being evaluated object, and the entropy of each evaluation index.
It should be noted that the entropy being evaluated object, evaluation index and evaluation index obtained in this step can be with
Corresponding contents involved in background technology is identical, and specifically, iotave evaluation matrix can be Dnm, wherein, every a line represents
One is evaluated object, and total n is evaluated object, and every string represents an evaluation index, and total m item evaluation index is each
It is evaluated the corresponding each evaluation index of object and is respectively provided with the value of correspondence, be each element in iotave evaluation matrix;Comment original
Valency matrix is standardized process and obtains Standard Process Rnm, RnmIn element and DnmIn element implication identical, difference only exist
In each element has been carried out standardization.Therefore, it can calculate each evaluation index according to the formula in background technology
Entropy, to calculate the entropy weight of corresponding evaluation index based on this entropy.
S12: calculate the entropy weight of each evaluation index according to the following formula:
Wherein, ωejRepresent the entropy weight of evaluation index j, HjRepresent the entropy of evaluation index j,Represent all be not 1 entropy
The average of value, m represents the quantity of evaluation index, j=1,2 ..., m.Owing to this formula is used for calculating entropy weight, therefore, in the accompanying drawings
Above-mentioned formula is represented by this title of entropy weight computing formula.
It should be noted that the formula of above-mentioned calculating entropy weight meets the requirement that the biggest entropy weight of entropy is the least, meet simultaneously with
Lower requirement:
0≤ωej≤1
Due to ωej2At HjThe difference that when → 0, entropy is small can cause entropy weight change at double, just with ωej1Formed mutually
Benefit relation, uses the average of the entropy of each evaluation index simultaneouslyAs weight, the ω when entropy is close to 0 can be madeej2Shared
Ratio reduce, the ω when entropy is close to 1ej1Proportion reduce, the formula of the most above-mentioned calculating entropy weight bigger than normal at entropy or
Power all can be rationally composed, to avoid the entropy weight deformity when entropy is in interval two ends time less than normal.
Technique scheme disclosed by the invention, the single tax power side of the formula of the most original calculating entropy weight
Tax the power mode, i.e. ω of a kind of complementation is added in formulaej1And ωej2;And use the mean control two of the entropy of each evaluation index
Plant the ratio shared by tax power mode, i.e. use the average of entropy of each evaluation index as weight so that when entropy is close to 0
ωej2Shared ratio reduces, the ω when entropy is close to 1ej1Proportion reduces, and then makes to calculate in the present invention entropy weight
Scheme all can rationally compose power when entropy is bigger than normal or less than normal, solve that entropy assessment in prior art exists when all evaluation indexes
Entropy level off to 1 time, the minute differences of entropy can cause the change at double of corresponding entropy weight, and then cause entropy weight unreasonable distribution
Problem.
It addition, for the reasonability verifying technique scheme that the embodiment of the present invention provides, use different entropy classes
Type is tested, and the technique scheme (B) using the technical scheme (A) in background technology, the embodiment of the present invention to provide calculates
Entropy weight, the results are shown in Table 1.
Table 1 background technology scheme and the contrast of the present invention program result of calculation
As shown in Table 1, when entropy vector is bigger than normal or less than normal, the technical scheme that the embodiment of the present invention provides gives more
Moderate and rational entropy weight vector;Particularly, for entropy vector (1.0000,0.5000,0.1000), the embodiment of the present invention carries
The technical scheme of confession has obtained having more the entropy weight vector of discrimination compared to the technical scheme in background technology.
Refer to Fig. 2, it illustrates the flow chart of a kind of Multiobjective Decision Making Method that the embodiment of the present invention provides, can wrap
Include following steps:
S21: obtain and advance with the Pareto disaggregation that multi-objective Algorithm obtains, and Pareto disaggregation is standardized place
Reason, obtains Standard Process, and in this matrix, every a line represents a solution, total n solution, and every string represents a target, total m
Individual target, each solution has m the desired value corresponding with m target.
Wherein it is desired to explanation, evolution multi-objective Algorithm is utilized to obtain Pareto disaggregation Xnm, wherein, every a line represents
One solution, total n solution, every string represents a target, has m target, and it is individual that each solution has the m corresponding with m target
Desired value, the most corresponding Pareto disaggregation X of each desired valuenmIn an element;Respective party disclosed in this step and prior art
Case principle is consistent, does not repeats them here.To Pareto disaggregation XnmIt is standardized processing, obtains Standard Process Rnm, wherein,
Each element and Pareto disaggregation X in Standard ProcessnmIn the implication of each element consistent, differ only in each element
Carry out standardization.
S22: calculate the entropy of each target in Standard Process, and utilize the entropy weight of the following equation each target of calculating:
Wherein, ωejRepresent the entropy weight of target j, HjRepresent the entropy of target j,For the average of all entropy not being 1, m
The quantity of expression target, j=1,2 ..., m.Owing to this formula is used for calculating entropy weight, therefore, entropy weight computing formula is used in the accompanying drawings
This title represents above-mentioned formula.
Wherein, the entropy calculating each evaluation index in Standard Process in the entropy of each target and background technology is calculated
Computing formula used is identical, and the implication differing only in parameter is different, specifically, determines each target according to the following formula
Entropy:
Wherein, HjRepresent the entropy of target j, 0≤Hj≤ 1, for making lnfijMeaningful, it is assumed that to work as fijWhen=0, fijln fij
=0;I=1,2 ..., n;J=1,2 ..., m.
S23: entropy weight based on each target and desired value corresponding to each solution calculate the comprehensive assessment value of each solution, and select
The solution taking maximum comprehensive assessment value corresponding is last solution.
Entropy weight based on each target and desired value corresponding to each solution calculate the comprehensive assessment value of each solution, and then determine
Go out last solution, i.e. concentrated by Pareto solution and select last solution, thus complete Multiobjective Decision Making Method.
It should be noted that existing Multiobjective Decision Making Method concentrates the process choosing last solution complicated in Pareto solution,
And needing the participation of policymaker and parameter to arrange so that last solution selection course is difficult to on-line operation.Disclosure
Technique scheme in, by the embodiment of the present invention provide above-mentioned entropy weight computational methods be applied in Multiobjective Decision Making Method,
First Pareto disaggregation is standardized processing and being calculated the entropy of each target, then uses the embodiment of the present invention to provide
Above-mentioned entropy weight computational methods calculate the entropy weight of each target, finally entropy weight and standardized desired value according to each target obtain respectively
The comprehensive assessment value of individual solution, select the maximum solution of comprehensive assessment value as last solution, thus fast and effectively by Pareto disaggregation
In get last solution, i.e. the technique scheme that the present invention provides is simple and efficiently, and, above-mentioned skill disclosed by the invention
Art scheme can select last solution for policymaker automatically according to normalized matrix, it is not necessary to policymaker participates in, and is especially suitable for transporting online
OK.
A kind of Multiobjective Decision Making Method that the embodiment of the present invention provides, entropy weight based on each target and each solution are corresponding
Desired value calculates the comprehensive assessment value of each solution, may include that
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the desired value solving i correspondence target j in Standard Process.
When policymaker is not required to be the setting that each target carries out subjective weight, can directly determine the entropy weight of each target
Be its weight, and then utilize the comprehensive assessment value of each solution of above-mentioned calculating, thus rapidly and efficiently get last solution.
A kind of Multiobjective Decision Making Method that the embodiment of the present invention provides, entropy weight based on each target and each solution are corresponding
Desired value calculates the comprehensive assessment value of each solution, may include that
Obtain the subjective weight given for each target of external world's input, and calculate combining of each target according to the following formula
Conjunction weight:
Wherein, ωjRepresent the comprehensive weight of target j, ωsjRepresent the subjective weight of target j;
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the desired value solving i correspondence target j in Standard Process.
When policymaker needs the setting that each target carries out subjective weight, each target of its input can be obtained
Subjective weight, utilizes such scheme to calculate the comprehensive weight of each target, and then calculates combining of each solution based on this comprehensive weight
Close assessed value, it is seen then that technical scheme disclosed by the invention can allow policymaker to set the subjective weight to each target, then
Decision making process then need not policymaker's operation, therefore can conveniently realize online operation, improve the efficiency of decision-making further.
A kind of Multiobjective Decision Making Method that the embodiment of the present invention provides, is standardized Pareto disaggregation processing, obtains
Standard Process, may include that
If Multiobjective Decision Making Method specially maximizes optimization problem, the most according to the following formula Pareto disaggregation is carried out
Standardization:
If Multiobjective Decision Making Method specially minimizes optimization problem, the most according to the following formula Pareto disaggregation is carried out
Standardization:
Wherein, rijRepresent the desired value solving i correspondence target j in Standard Process, xijRepresent that Pareto solution is concentrated and solve i pair
Answer the desired value of target j, 1≤i≤n, 1≤j≤m.
It should be noted that the implication phase of largest optimization problem and minimum optimization problem corresponding title with prior art
With, do not repeat them here;It addition, aforesaid way is the most only standardized one of mode of process, other can carry out standard
The technical scheme that change processes is the most all within protection scope of the present invention.
It addition, for the advantage verifying a kind of Multiobjective Decision Making Method that the embodiment of the present invention provides, carry out following test.
Wherein, table 2 give this example use Pareto disaggregation:
With document " Multi-objective optimization and decision making for power
dispatch of a large-scale integrated energy system with distributed DHCs
Embedded " in as a example by the Pareto disaggregation of 39 solution 5 targets that draws of multiobjective decision-making, use technical scheme and suitable
Sequence structure Evaluation Method (Preference Ranking Organization Method for Enrichment
Evaluation, PROMETHEE) respectively this disaggregation is carried out decision-making, and the trap queuing result obtained is contrasted, such as table
Shown in 2, wherein technical solution of the present invention correspondence sequence 1, sequential organization Evaluation Method correspondence sequence 2.Visible, two kinds of methods give
Roughly the same decision scheme, all selecting the 19th to solve is last solution, this demonstrates technical solution of the present invention in multiobjective decision-making
The effectiveness of middle selection last solution, and in implementation process, technical solution of the present invention is simpler relative to sequential organization Evaluation Method soon
Speed.And document " Multi-objective optimization and decision making for power dispatch
Of a large-scale integrated energy system with distributed DHCs embedded " use
The method of evidential reasoning selects last solution, but due to the complexity of the method, five solutions that only have chosen overstriking in table 2 are entered
Go assessment, and owing to target 1 and target 2 are used to the target that optimizes, decision making process can be not involved in, only use rear three mesh
Mark carries out evidential reasoning and rear three targets with the addition of subjective weights omegasj=[0.2,0.4,0.4].In order to push away with evidence
Reason method is made comparisons, and technical solution of the present invention the most only uses these five rear three targets solved to carry out decision-making, and the result obtained is such as
Shown in table 3, wherein, Evidential reasoning algorithm correspondence sequence 3.
Visible, three kinds of methods give identical sequencing schemes, but assess compared to Evidential reasoning algorithm and sequential organization
Method, technical scheme disclosed by the invention has directly utilized the good and bad information that desired value itself comprises, the most efficiently, has been highly suitable for
Line runs.
Table 2 two schemes is to 39 sequencing schemes solving 5 target disaggregation
3 three kinds of schemes of table are to 5 sequencing schemes solving 3 target disaggregation
Corresponding with said method, the embodiment of the present invention additionally provides a kind of entropy weight and calculates device, as it is shown on figure 3, permissible
Including:
Data acquisition module 11, for obtaining the evaluation index being evaluated object, and the entropy of each evaluation index;
Entropy weight computing module 12, for calculating the entropy weight of each evaluation index according to the following formula:
Wherein, ωejRepresent the entropy weight of evaluation index j, HjRepresent the entropy of evaluation index j,Represent all be not 1 entropy
The average of value, m represents the quantity of evaluation index, j=1,2 ..., m.
A kind of entropy weight that the embodiment of the present invention provides calculates the explanation of relevant portion in device and refers to the embodiment of the present invention
In a kind of entropy weight computational methods provided, the detailed description of corresponding part, does not repeats them here.
Corresponding with said method, the embodiment of the present invention additionally provides a kind of multiobjective decision-making device, as shown in Figure 4, and can
To include:
Matrix disposal module 21, advances with, for obtaining, the Pareto disaggregation that multi-objective Algorithm obtains, and to Pareto
Disaggregation is standardized processing, and obtains Standard Process, and in this matrix, every a line represents a solution, total n solution, every string generation
One target of table, total m target, each solution has m the desired value corresponding with m target;
Entropy weight acquisition module 22, for calculating the entropy of each target in Standard Process, and utilizes following equation to calculate
The entropy weight of each target:
Wherein, ωejRepresent the entropy weight of target j, HjRepresent the entropy of target j,For the average of all entropy not being 1, m
The quantity of expression target, j=1,2 ..., m;
Last solution acquisition module 23, the desired value corresponding for entropy weight based on each target and each solution calculates each solution
Comprehensive assessment value, and solution corresponding to comprehensive assessment value choosing maximum be last solution.
A kind of multiobjective decision-making device that the embodiment of the present invention provides, last solution acquisition module may include that
First assessment unit, for calculating the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the desired value solving i correspondence target j in Standard Process.
A kind of multiobjective decision-making device that the embodiment of the present invention provides, last solution acquisition module may include that
Second assessment unit, calculates combining of each solution for the desired value that entropy weight based on each target and each solution are corresponding
Close assessed value, including:
Obtain the subjective weight given for each target of external world's input, and calculate combining of each target according to the following formula
Conjunction weight:
Wherein, ωjRepresent the comprehensive weight of target j, ωsjRepresent the subjective weight of target j;
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the desired value solving i correspondence target j in Standard Process.
A kind of multiobjective decision-making device that the embodiment of the present invention provides, matrix disposal module includes:
First processing unit, if being used for solving to maximize optimization problem, then according to following for multiobjective decision-making device
Pareto disaggregation is standardized processing by formula:
Second processing unit, if minimizing optimization problem, then according to following for multiobjective decision-making device for solution
Pareto disaggregation is standardized processing by formula:
Wherein, rijRepresent the desired value solving i correspondence target j in Standard Process, xijRepresent that Pareto solution is concentrated and solve i pair
Answer the desired value of target j, 1≤i≤n, 1≤j≤m.
In a kind of multiobjective decision-making device that the embodiment of the present invention provides, the explanation of relevant portion refers to present invention enforcement
In a kind of Multiobjective Decision Making Method that example provides, the detailed description of corresponding part, does not repeats them here.
Described above to the disclosed embodiments, makes those skilled in the art be capable of or uses the present invention.To this
The multiple amendment of a little embodiments will be apparent from for a person skilled in the art, and generic principles defined herein can
With without departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention will not be limited
It is formed on the embodiments shown herein, and is to fit to consistent with principles disclosed herein and features of novelty the widest
Scope.
Claims (10)
1. entropy weight computational methods, it is characterised in that including:
Obtain the evaluation index being evaluated object, and the entropy of each described evaluation index;
Calculate the entropy weight of each described evaluation index according to the following formula:
Wherein, ωejRepresent the entropy weight of evaluation index j, HjRepresent the entropy of evaluation index j,Represent all be not 1 entropy
Average, m represents the quantity of described evaluation index, j=1,2 ..., m.
2. a Multiobjective Decision Making Method, it is characterised in that including:
Obtain and advance with the Pareto disaggregation that multi-objective Algorithm obtains, and be standardized described Pareto disaggregation processing,
Obtaining Standard Process, in this matrix, every a line represents a solution, total n solution, and every string represents a target, total m
Target, each solution has m the desired value corresponding with m target;
Calculate the entropy of each target in described Standard Process, and utilize the entropy weight of the following equation each target of calculating:
Wherein, ωejRepresent the entropy weight of target j, HjRepresent the entropy of target j,For the average of all entropy not being 1, m represents
The quantity of target, j=1,2 ..., m;
Entropy weight based on each target and desired value corresponding to each solution calculate the comprehensive assessment value of each solution, and choose maximum
Solution corresponding to comprehensive assessment value is last solution.
Method the most according to claim 2, it is characterised in that entropy weight based on each target and target corresponding to each solution
Value calculates the comprehensive assessment value of each solution, including:
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the desired value solving i correspondence target j in described Standard Process.
Method the most according to claim 2, it is characterised in that entropy weight based on each target and target corresponding to each solution
Value calculates the comprehensive assessment value of each solution, including:
Obtain the subjective weight given for each target of external world's input, and calculate the synthetic weights of each target according to the following formula
Weight:
Wherein, ωjRepresent the comprehensive weight of target j, ωsjRepresent the subjective weight of target j;
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the desired value solving i correspondence target j in described Standard Process.
Method the most according to claim 2, it is characterised in that be standardized described Pareto disaggregation processing, obtain
Standard Process, including:
If described Multiobjective Decision Making Method specially maximizes optimization problem, the most according to the following formula to described Pareto disaggregation
It is standardized processing:
If described Multiobjective Decision Making Method specially minimizes optimization problem, the most according to the following formula to described Pareto disaggregation
It is standardized processing:
Wherein, rijRepresent the desired value solving i correspondence target j in Standard Process, xijRepresent that Pareto solution is concentrated and solve i correspondence target
The desired value of j, 1≤i≤n, 1≤j≤m.
6. an entropy weight calculates device, it is characterised in that including:
Data acquisition module, for obtaining the evaluation index being evaluated object, and the entropy of each described evaluation index;
Entropy weight computing module, for calculating the entropy weight of each described evaluation index according to the following formula:
Wherein, ωejRepresent the entropy weight of evaluation index j, HjRepresent the entropy of evaluation index j,Represent all be not 1 entropy
Average, m represents the quantity of described evaluation index, j=1,2 ..., m.
7. a multiobjective decision-making device, it is characterised in that including:
Matrix disposal module, advances with, for obtaining, the Pareto disaggregation that multi-objective Algorithm obtains, and to described Pareto solution
Collection is standardized processing, and obtains Standard Process, and in this matrix, every a line represents a solution, total n solution, and every string represents
One target, total m target, each solution has m the desired value corresponding with m target;
Entropy weight acquisition module, for calculating the entropy of each target in described Standard Process, and it is every to utilize following equation to calculate
The entropy weight of individual target:
Wherein, ωejRepresent the entropy weight of target j, HjRepresent the entropy of target j,For the average of all entropy not being 1, m represents
The quantity of target, j=1,2 ..., m;
Last solution acquisition module, the desired value corresponding for entropy weight based on each target and each solution calculates the comprehensive of each solution
Assessed value, and solution corresponding to comprehensive assessment value choosing maximum be last solution.
Device the most according to claim 7, it is characterised in that last solution acquisition module includes:
First assessment unit, for calculating the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the desired value solving i correspondence target j in described Standard Process.
Device the most according to claim 7, it is characterised in that last solution acquisition module includes:
Second assessment unit, the desired value corresponding for entropy weight based on each target and each solution calculates comprehensively commenting of each solution
Valuation, including:
Obtain the subjective weight given for each target of external world's input, and calculate the synthetic weights of each target according to the following formula
Weight:
Wherein, ωjRepresent the comprehensive weight of target j, ωsjRepresent the subjective weight of target j;
Calculate the comprehensive assessment value of each solution according to the following formula:
Wherein, uiRepresent the comprehensive assessment value solving i, rijRepresent the desired value solving i correspondence target j in described Standard Process.
Device the most according to claim 7, it is characterised in that matrix disposal module includes:
First processing unit, if being used for solving to maximize optimization problem, then according to following for described multiobjective decision-making device
Described Pareto disaggregation is standardized processing by formula:
Second processing unit, if minimizing optimization problem, then according to following for described multiobjective decision-making device for solution
Described Pareto disaggregation is standardized processing by formula:
Wherein, rijRepresent the desired value solving i correspondence target j in Standard Process, xijRepresent that Pareto solution is concentrated and solve i correspondence target
The desired value of j, 1≤i≤n, 1≤j≤m.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610651441.7A CN106250353A (en) | 2016-08-10 | 2016-08-10 | A kind of entropy weight computational methods and Multiobjective Decision Making Method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610651441.7A CN106250353A (en) | 2016-08-10 | 2016-08-10 | A kind of entropy weight computational methods and Multiobjective Decision Making Method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106250353A true CN106250353A (en) | 2016-12-21 |
Family
ID=58079081
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610651441.7A Pending CN106250353A (en) | 2016-08-10 | 2016-08-10 | A kind of entropy weight computational methods and Multiobjective Decision Making Method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106250353A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112381377A (en) * | 2020-11-10 | 2021-02-19 | 国网上海市电力公司 | Power grid connection mode evaluation method, device, equipment and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101540018A (en) * | 2009-01-23 | 2009-09-23 | 浙江师范大学 | Method and system for evaluating degree of application of enterprise information technology |
CN103454528A (en) * | 2013-08-30 | 2013-12-18 | 华南理工大学 | Method for detecting and identifying electric system element fault based on form singular entropy |
CN104967629A (en) * | 2015-07-16 | 2015-10-07 | 网宿科技股份有限公司 | Network attack detection method and apparatus |
CN105261013A (en) * | 2015-09-25 | 2016-01-20 | 孙高磊 | Comprehensive evaluation method and evaluation system for scanned image quality |
CN105373880A (en) * | 2015-10-13 | 2016-03-02 | 国网福建省电力有限公司 | Power grid engineering project scheme feedback evaluation and scheme decision system |
CN105469196A (en) * | 2015-11-18 | 2016-04-06 | 山东科技大学 | Comprehensive evaluation method and comprehensive evaluation system for evaluating mine construction project process |
-
2016
- 2016-08-10 CN CN201610651441.7A patent/CN106250353A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101540018A (en) * | 2009-01-23 | 2009-09-23 | 浙江师范大学 | Method and system for evaluating degree of application of enterprise information technology |
CN103454528A (en) * | 2013-08-30 | 2013-12-18 | 华南理工大学 | Method for detecting and identifying electric system element fault based on form singular entropy |
CN104967629A (en) * | 2015-07-16 | 2015-10-07 | 网宿科技股份有限公司 | Network attack detection method and apparatus |
CN105261013A (en) * | 2015-09-25 | 2016-01-20 | 孙高磊 | Comprehensive evaluation method and evaluation system for scanned image quality |
CN105373880A (en) * | 2015-10-13 | 2016-03-02 | 国网福建省电力有限公司 | Power grid engineering project scheme feedback evaluation and scheme decision system |
CN105469196A (en) * | 2015-11-18 | 2016-04-06 | 山东科技大学 | Comprehensive evaluation method and comprehensive evaluation system for evaluating mine construction project process |
Non-Patent Citations (3)
Title |
---|
南铁雷: "基于熵权和多目标优化的投标报价方案优选", 《价值工程》 * |
郭金维等: "一种改进的多目标决策指标权重计算方法", 《西安电子科技大学学报(自然科学版)》 * |
陈深等: "基于 Pareto 最优的电力***多目标优化研究", 《黑龙江电力》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112381377A (en) * | 2020-11-10 | 2021-02-19 | 国网上海市电力公司 | Power grid connection mode evaluation method, device, equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103426120A (en) | Medium and low voltage power distribution network comprehensive evaluation method based on reliability | |
CN106250438A (en) | Based on random walk model zero quotes article recommends method and system | |
CN107590623A (en) | A kind of cloud manufacturing service preferred method of combination for considering region logistics service capability | |
CN110276528A (en) | A kind of control system of power plant information security risk evaluation system | |
CN103793747B (en) | A kind of sensitive information template construction method in network content security management | |
Chen et al. | A hybrid MCDM model for new product development: Applied on the Taiwanese LiFePO4 industry | |
Samanlioglu et al. | A fuzzy AHP-VIKOR approach for evaluation of educational use simulation software packages | |
Cheng et al. | MCDM aggregation model using situational ME-OWA and ME-OWGA operators | |
CN106295909A (en) | For calculating data processing method and the device of the electrical network project indicator | |
Sotoudeh-Anvari et al. | A new data envelopment analysis in fully fuzzy environment on the base of the degree of certainty of information | |
CN107273361A (en) | The word computational methods and its device closed based on the general type-2 fuzzy sets of broad sense | |
CN106250353A (en) | A kind of entropy weight computational methods and Multiobjective Decision Making Method | |
Sajid | Machine learned artificial neural networks vs linear regression: A case of chinese carbon emissions | |
CN107009359A (en) | Reliability fuzzy method for predicting based on welding robot system distinguishing hierarchy | |
CN102495944B (en) | Time series forecasting method and equipment and system adopting same | |
Cayir Ervural | A combined methodology for evaluation of electricity distribution companies in Turkey | |
CN109447498B (en) | Rail transit system cost association multi-domain influence element big data driving analysis method | |
Zahedi-Seresht et al. | A new method for ranking distribution companies with several scenarios data by using DEA/MADM | |
Chima et al. | Artificial Neural Network Application in Prediction–A Review | |
CN110162704A (en) | More scale key user extracting methods based on multiple-factor inheritance algorithm | |
Mohammadi et al. | Selecting adequate security mechanisms in e-business processes using fuzzy TOPSIS | |
Wang et al. | A Coevolution Algorithm Based on Spatial Division and Hybrid Matching Strategy | |
Capitanescu et al. | Assessing the uses of NLP-based surrogate models for solving expensive multi-objective optimization problems: application to potable water chains | |
CN106952122A (en) | A kind of vehicle evaluation method and system | |
Abdullah | A fuzzy decision making method in developing environmental performance index |
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: 20161221 |