Secondary equipment evaluation index weighting method based on double entropy correction G1 method

Abstract

The invention discloses a secondary equipment evaluation index weighting method based on a double entropy correction G1 method, which considers the influence of expert scoring on the field of secondary equipment evaluation, establishes a secondary equipment evaluation index weighting model of the entropy correction G1 method, and can be used for realizing the calculation of a secondary equipment evaluation index weight coefficient. The method comprises the following steps: firstly, constructing a secondary equipment health level evaluation index system based on existing secondary equipment related evaluation regulations and the like in China; secondly, analyzing the rationality of the secondary equipment evaluation index importance expert scoring result in combination with a relative entropy principle; the evaluation index entropy is used as an objective factor to correct the G1 subjective weighting method; and finally, integrating the corrected subjective score value and objective weighting value to obtain the final index weight of the evaluation index. The method is suitable for performing weighting on the secondary equipment evaluation indexes, and the weights of the secondary equipment evaluation indexes can be calculated more accurately, so that the method can be used for guiding work such as state evaluation of the secondary equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of power system secondary equipment evaluation technology, and particularly relates to a secondary equipment evaluation index weighting method based on a double entropy value correction G1 method. BACKGROUND

[0002] The rapid development of smart grids brings more opportunities and challenges to secondary equipment, and also puts forward higher requirements for the safe and stable operation of secondary equipment itself. As the basis of condition-based maintenance, equipment state evaluation can provide basis and guidance for it, and the determination of evaluation index weight is a key step to realize the comprehensive evaluation of secondary equipment. The importance of different indexes relative to the comprehensive evaluation result is different, and a scientific and reasonable evaluation index weighting method is of great significance to improve the accuracy of the comprehensive evaluation result of secondary equipment.

[0003] On the one hand, there is a lot of subjective randomness in the process of giving the index two-by-two comparison importance ratio by expert subjective scoring, and the occurrence of human error will have a negative impact on the index weighting result, resulting in a decrease in the credibility of the comprehensive evaluation result. On the other hand, some scholars have proposed that only the importance ranking relationship given by the expert is retained, and the importance degree ratio between indexes is represented by the entropy value of the evaluation index, that is, the entropy value is used to correct the G1 subjective weighting method. This method can reduce the influence of subjective randomness on the comprehensive evaluation to a certain extent, but its shortcomings are also obvious, that is, whether the original scoring result given by the expert is reasonable or not is discarded, which weakens the role of expert experience in index weighting.

[0004] In view of the above problems, the present application provides a secondary equipment evaluation index weighting method based on a double entropy value correction G1 method to realize the calculation of the evaluation index weight coefficient. The expert scoring result of the importance of the secondary equipment evaluation index is corrected based on the relative entropy to improve the rationality of the scoring result. Then, the G1 subjective weighting method is further corrected by using the generalized entropy value, which reduces the subjective randomness while taking into account the importance of the index experience. SUMMARY

[0005] The purpose of the present application is to provide a secondary equipment evaluation index weighting method based on a double entropy value correction G1 method, which considers the influence of expert scoring on the field of secondary equipment evaluation, establishes an entropy value correction G1 method for the secondary equipment evaluation index weighting model, and can be used to realize the calculation of the secondary equipment evaluation index weight coefficient, and has strong practical value.

[0006] The purpose of the present application is achieved by the following technical solutions:

[0007] A secondary equipment evaluation index weighting method based on a double entropy value correction G1 method, the method comprising:

[0008] Step 1, based on the existing secondary equipment related evaluation procedures in China, combined with the actual statistical data of secondary equipment operation and maintenance, a secondary equipment health level evaluation index system is constructed;

[0009] Step 2, based on the relative entropy value, the rationality of the actual expert score is analyzed and corrected, so as to correct the subjective weight;

[0010] Step 3, the G1 method based on the improved entropy value is used to correct the objective index weight;

[0011] Step 4, the final index weight of the evaluation index is obtained by comprehensively correcting the subjective score value and the objective weight value.

[0012] It can be seen from the above technical scheme provided by the present application that the above method is simple to calculate, can accurately determine the weight value of the secondary equipment evaluation index, and has strong practical value. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the technical scheme of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0014] Figure 1 The secondary equipment health level evaluation index system provided for the embodiments of the present application;

[0015] Figure 2 The index weight method flowchart based on entropy correction provided for the embodiments of the present application; DETAILED DESCRIPTION

[0016] The technical scheme in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0017] The embodiments of the present application will be further described in detail below with reference to the drawings. The method comprises:

[0018] Step 1, based on the existing secondary equipment related evaluation procedures in China, combined with the actual statistical data of secondary equipment operation and maintenance, a secondary equipment health level evaluation index system is constructed;

[0019] In the step 1, the specific process is:

[0020] Referring to the existing secondary equipment related evaluation regulations in China, following the principles of typicality, availability, evaluability and simplicity, combined with the actual statistical data of secondary equipment operation and maintenance, the secondary equipment health level evaluation index system is constructed as shown in Table 1 and Table 2. Figure 1 and Table 1.

[0021] Table 1 Secondary equipment evaluation index system

[0022]

[0023]

[0024] Step 2, based on the relative entropy value, the rationality of the actual expert score is analyzed and corrected, so as to correct the subjective weight;

[0025] In the step 2, as shown in Table 2, the specific process is: Figure 2

[0026] (1) Invite n experts A i =(A1,A2,…,A n ) to score the importance of m indexes in the evaluation index set I j =(I1,I2,…,I m ) relative to the corresponding index of the previous level by 1-9 scoring method, to obtain the original score matrix

[0027] P=[p ij ] n×m (1)

[0028] Wherein, p ij is the importance score of expert A i to index I j .

[0029] As shown in Table 2 and Table 3, the scoring results of 5 experts for Table 1 index are as follows:

[0030] Table 2 Expert score table of primary evaluation index

[0031]

[0032] Table 3 Expert score table of secondary evaluation index

[0033]

[0034] (2) Unitize the elements in the original score matrix:

[0035]

[0036] The preference utility vector of expert A i can be expressed as ​

[0037] (3) Record the entire score of the expert group A g The group preference utility vector of the group is P g =(P g1 ,P g2 ,…,P gm ), and the group preference utility vector of the expert group A g is taken as a reference, which should satisfy the minimization of the individual preference utility vector of the expert, so the relative entropy nonlinear programming with constraint conditions is constructed accordingly:

[0038]

[0039] Wherein, the smaller the relative entropy, the higher the rationality of the expert's score. The importance group score value of the expert group for the indicators in the evaluation index set is obtained by solving the Lagrange function:

[0040]

[0041] Further, the score level vector of the expert A i is defined as

[0042]

[0043] Wherein, e ij =1-Δd j -Δl j , represents the score level of the individual expert A i about the indicator I j , Δd j represents the difference of the importance score value of the indicator; Δl j represents the difference of the importance ranking position of the indicator.

[0044] The score entropy H i is taken as the measure

[0045]

[0046] (4) Based on the relative entropy, the score level and rationality of the individual expert are analyzed, the score results in the expert score matrix P that reduce the rationality are removed, and the score results of the remaining experts are analyzed again until the requirements are met.

[0047] For 5 scoring experts, the above method is iterated to obtain the original results of the importance group score of the experts, which satisfy the iteration error ε=0.0001; further calculation obtains the rationality analysis results of the importance score of each expert, and the poor expert score in the rationality analysis results is removed, and the original expert score table is corrected. Finally, the index weight is shown in Table 4 and Table 5.

[0048] Table 4 Weights of first-level evaluation indicators

[0049]

[0050] Table 5 Weights of Secondary Evaluation Indicators

[0051]

[0052] Step 3: Use the G1 method based on entropy to modify the weighting of objective indicators;

[0053] In step 3, Figure 2 As shown, the specific process is:

[0054] Calculate the index entropy value e of l evaluation objects after normalization j :

[0055]

[0056] As shown in Table 6, the entropy values ​​of the above secondary equipment indicators are:

[0057] Table 6 Evaluation index entropy value table

[0058]

[0059] Furthermore, based on the revised expert group scoring results, the indicators are sorted according to the score size to obtain the ranking relationship of the evaluation indicators' importance:

[0060] {I1≥I2≥…≥I m} (9)

[0061] Among them, “≥” means that the importance score of the previous indicator is not lower than that of the next indicator.

[0062] Calculate adjacent evaluation index I j-1 and I j The relative importance ratio r j :

[0063]

[0064] Among them, e j and p j I j The entropy value and the reasonable score value of the expert group. Then the indicator I m The weight β m for:

[0065]

[0066] The weights of the remaining indicators can be calculated by the following formula:

[0067] β j-1 =r j β m j=m,m-1,...,2(12)

[0068] According to the above method, the intra-layer weights of the three first-level indicators relative to the comprehensive evaluation target can be calculated as follows: 0.591, 0.251 and 0.158, and the intra-layer weights of each second-level indicator relative to the first-level indicator β j , as shown in Table 7.

[0069] Table 7 Evaluation index weight calculation results

[0070]

[0071] Step 4: Comprehensively combine the corrected subjective scoring value and objective weighting value to obtain the final indicator weight of the evaluation indicator.

[0072] In step 4, Figure 2 As shown, the specific process is:

[0073] The relative entropy corrects the importance of the evaluation index results of the expert group to obtain the subjective score value, the evaluation index entropy value improves the G1 method to obtain the objective index weight value, and the comprehensive index I is obtained. j The inter-layer weight relative to the evaluation target is:

[0074] ω j =α j β j j=1,2,...,m(13)

[0075] Among them, α j For indicator I j The weight of the corresponding first-level indicator relative to the evaluation target; β j For indicator I j The weight relative to the corresponding first-level indicator.

[0076] As shown in Table 8, the inter-layer weights ω of the above-mentioned secondary indicators relative to the comprehensive evaluation target are: j ,

[0077] Table 8 Evaluation index weight calculation results

[0078]

[0079] It should be noted that the contents not described in detail in the embodiments of the present invention belong to the prior art known to those skilled in the art.

[0080] The actual example shows that the problem of the non-independent signals in the monitoring system can make the quantitative calculation of the reliability of the substation monitoring system more accurate, and thus can be used for guiding the analysis design, operation and maintenance of the substation monitoring system.

[0081] The above merely provides the preferred embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of the changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for assigning weights to secondary equipment evaluation indicators based on the double entropy value modified G1 method, characterized in that: The method comprises: Step 1: Based on the existing secondary equipment evaluation procedures in my country and combined with the actual statistical data of secondary equipment operation and maintenance, a secondary equipment health level evaluation index system is constructed; Step 2: Based on the relative entropy value, analyze and correct the rationality of the actual expert ratings, thereby correcting the subjective weights; Step 3: Use the G1 method based on entropy to modify the weighting of objective indicators; Step 4: Comprehensively combine the corrected subjective scoring value and objective weighting value to obtain the final indicator weight of the evaluation indicator.

2. The secondary equipment evaluation index weighting method according to claim 1 is characterized in that: In step 1, the process of constructing the secondary equipment health level evaluation index system is specifically as follows: By consulting the existing evaluation procedures for secondary equipment in my country, adhering to the principles of typicality, accessibility, evaluability and simplicity, and combining the actual statistical data of secondary equipment operation and maintenance, an evaluation index system for the health level of secondary equipment was constructed.

3. The secondary equipment evaluation index weighting method according to claim 1 is characterized in that: In step 2, the process of analyzing and correcting the rationality of the actual expert ratings, thereby correcting the subjective weights, is specifically as follows: (1) Invite n experts to score the importance of the indicators in the evaluation indicator set relative to the corresponding indicators in the previous level using a 1-9 scoring method to obtain the original scoring matrix; (2) Normalize the elements in the original scoring matrix to obtain expert A i 's preference utility vector; (3) According to the fact that the group preference utility vector of the expert group should satisfy the minimization of the individual preference utility vector of the expert, a relative entropy nonlinear programming with constraints is constructed, and the importance group score of the expert group for the indicators in the evaluation indicator set is obtained by solving the Lagrange function; (4) Further, the expert's rating level vector is defined and the rating entropy is calculated as a measure. (5) Based on relative entropy, the individual expert scoring levels and rationality are analyzed, the scoring results that reduce rationality in the expert scoring matrix are eliminated, and the remaining expert scoring results are re-analyzed until the requirements are met.

4. The secondary equipment evaluation index weighting method according to claim 1 is characterized in that: In step 3, the process of correcting the weighting of objective indicators using the G1 method improved based on entropy is specifically as follows: First, the entropy value of the normalized evaluation object is calculated. Then, based on the revised expert group scoring results, the indicators are sorted according to the score size to obtain the ranking relationship of the evaluation indicator importance, thereby calculating the relative importance ratio between adjacent evaluation indicators. Finally, the weight of the indicator is calculated based on this.

5. The secondary equipment evaluation index weighting method according to claim 1 is characterized in that: In step 4, the process of comprehensively modifying the subjective scoring value and the objective weighting value to obtain the final indicator weight of the evaluation indicator is specifically as follows: The relative entropy corrects the importance of the evaluation index results of the expert group to obtain the subjective score value, and the evaluation index entropy improves the G1 method to obtain the objective index weight value, and the two are combined to obtain the index I j The inter-layer weight relative to the evaluation target is: oh j =a j b j j=1,2,…,m Among them, α j For indicator I j The weight of the corresponding first-level indicator relative to the evaluation target; β j For indicator I j The weight relative to the corresponding first-level indicator.

Images