Power transmission and transformation project mixed multi-attribute decision-making method based on Mo-EDAS method

Abstract

The invention discloses a power transmission and transformation project mixed multi-attribute decision-making method based on a Mo-EDAS method, and the method comprises the steps: processing quantitative evaluation information through employing an accurate number pattern and random variables according to the difference of the characteristics and types of power transmission and transformation project construction safety risk evaluation indexes, and processing qualitative evaluation information through employing a probability semantic term set; aiming at the condition that the incidence relation between the indexes of the mixed information is more complex, the weight of each index is calculated by adopting a combined weighting method; and respectively calculating an average ideal solution of each module according to a good and bad comparison method of different types of evaluation information, calculating a distance weighting standard value of each alternative scheme according to a distance formula of different types of evaluation information, finally calculating an evaluation score of each alternative scheme, and evaluating and sorting the schemes according to the evaluation scores. The modular idea is adopted to effectively avoid the problem that the calculation is tedious or information loss and loss are easily caused due to the fact that mixed information is converted into information in the same form for comparison in the information aggregation process in the mixed multi-attribute decision making process.

Description

Technical Field

[0001] The present invention relates to the technical field of power transmission and transformation project construction, and particularly relates to a hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method. Background Art

[0002] The construction of power transmission and transformation projects includes civil engineering, electrical engineering, overhead line projects, and cable line projects, involving multiple links such as high-altitude operations, large-scale mechanical operations, and electrical equipment installation. Especially for the construction of renovated and expanded substations and transmission line projects, their construction has characteristics such as being adjacent to live bodies, moving, crossing live lines, crossing highways and railways, crossing rivers and lakes, etc. The construction environment is extremely complex and the safety risks are relatively large. It is a highly comprehensive systematic project. In power transmission and transformation projects, due to the complexity and uncertainty of real decision-making problems, there are often multiple types of attribute values. For example, in the problem of choosing an emergency plan, attributes such as the timeliness of rescue, economic cost, utilization of mechanical equipment, aftermath handling, social impact of disasters, and on-site control of disasters are often considered. Among them, the timeliness of rescue is a quantitative attribute, which can be measured by the time spent from the start to the completion of the rescue operation and can be represented by precise numerical information; the social impact of disasters is an uncertain quantitative attribute and can be represented by random variable-type information; on-site control of disasters is a qualitative attribute, and it is difficult to express the degree of control with precise numbers and is commonly represented by semantic-type information. This multi-attribute decision-making problem that includes both quantitative and qualitative attributes and contains uncertainty conforms to the objective actual needs and has important research significance.

[0003] In the information aggregation process of hybrid multi-attribute decision-making problems, the hybrid information is usually converted into the same form of information for comparison. However, many decision-making problems need to consider different forms of quantitative or qualitative information simultaneously, and information loss problems will occur in the process of converting to the same information for evaluation. Summary of the Invention

[0004] The purpose of the present invention is to provide a hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method, which can objectively and comprehensively evaluate and compare each optimal plan and reasonably select an emergency plan. To achieve the above technical features, the purpose of the present invention is realized as follows:

[0005] A hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method includes:

[0006] According to the different characteristics and types of the construction safety risk evaluation indexes of power transmission and transformation projects, precise numerical types and random variables are used to process quantitative evaluation information, and qualitative evaluation information is processed through a probability semantic term set.

[0007] When the evaluation information is mixed information, considering that the mutual influence relationship between indicators is more complex, the combined weighting method is used to calculate the weights of each indicator;

[0008] The Mo-EDAS method is used to calculate the average ideal solution of each module according to the comparison method of the advantages and disadvantages of different types of evaluation information, calculate the distance weighted standard value of each alternative solution according to the distance formula of different types of evaluation information, finally calculate the evaluation score of each alternative solution, and evaluate and rank the solutions according to the evaluation score.

[0009] When the evaluation information is numerical, the range standardization method is used to standardize the evaluation information;

[0010] Let the numerical evaluation information be y ij , and the standardized evaluation information is The calculation formula is as follows:

[0011] If it is a cost-type attribute, then:

[0012]

[0013] If it is a benefit-type attribute, then:

[0014]

[0015] When the attribute is an uncertain quantitative attribute, the evaluation information adopts a random variable type, and the uncertain quantitative attribute is evaluated by constructing a dominance matrix for the attribute;

[0016] Construct a dominance matrix G for the attribute j =[g iqj n×n , where g iqj represents the probability that for the attribute C j , alternative A i is superior to alternative A q , and the calculation formula of g iqj is as follows:

[0017] If it is a cost-type attribute, then:

[0018]

[0019] If it is a benefit-type attribute, then:

[0020]

[0021] Among them: f i (g ij ) and f q (g qj ) are g i and g qj ​The probability density function. According to the dominance matrix G j =[g iqj n×n Measure the advantages and disadvantages of each solution for the attribute C j to obtain the dominance degree of the attribute C j The dominance degree Among them is an exact number, and the calculation formula is as follows:

[0022]

[0023] Normalize the dominance degree to obtain the normalized dominance degree The calculation formula is as follows:

[0024]

[0025] The elements of the probabilistic linguistic term set are composed of linguistic evaluation values and corresponding probabilities. Among them, the linguistic evaluation values come from the given linguistic term set, and the probability is the preference degree or proportion of the decision maker in favor of a certain linguistic evaluation value. The following is the definition of the probabilistic linguistic term set:

[0026] Definition 1. Let S = {s0,....., s a} be the semantic set, then the probabilistic linguistic term set is:

[0027]

[0028] Among them, L (k) (P (k) ) represents that the probability of the linguistic term set L (k) is P (k) , and #L(p) is the number of semantic sets included in all L(p);

[0029] Definition 2. Let L(p)1 and L(p)2 be two probabilistic linguistic term sets If #L(p)1 > #L(p)2, then #L(p)1 - #L(p)2 semantic sets need to be added to L(p)2 to make the number of elements in L(p)1 and L(p)2 the same. Add the smallest semantic set in L(p)2, and its weight is 0;

[0030] Definition 3. Let the of a probabilistic linguistic term set need to be standardized, and the formula is as follows:

[0031]

[0032] Among them, L (k) (P (k) ) is sorted from small to large;

[0033] Definition 4, and are the k-th elements in L(p)1 and L(p)2 respectively, and P1 (k) and respectively represent and weights. Their distance formula is:

[0034]

[0035] where: and represent the subscripts of L(p)1 and L(p)2 respectively;

[0036] Definition 5. Let L(p) = {L (k) (p (k) )|1, 2,..., #L(p)} be a probabilistic linguistic term set, r(k) be the subscript of L(k), and the scoring function expression of L(p) is:

[0037]

[0038] where For the probabilistic semantic sets L(p)1 and L(p)2, if E(L(p)1) > E(L(p)2), then L(p)1 > L(p)2;

[0039] Definition 6. Let L(p) = {L (k) (p (k) )|1, 2,..., #L(p)} be a probabilistic linguistic term set, and the scoring function The deviation formula of L(p) is:

[0040]

[0041] Let the scoring functions of L(p)1 and L(p)2 be E(L(p)1) and E(L(p)2), and the deviation degrees be σ(L(p)1) and σ(L(p)2) respectively; when E(L(p)1) = E(L(p)2), if σ(L(p)1) < σ(L(p)2), then L(p)1 > L(p)2; if σ(L(p)1) = σ(L(p)2), then L(p)1 = L(p)2; if σ(L(p)1) > σ(L(p)2), then L(p)1 < L(p)2.

[0042] The calculation of the weights of evaluation indicators includes subjective weights, objective weights, and combined weighting.

[0043] The subjective weights are calculated using the BWM method, and the specific steps are as follows:

[0044] S1. Determine risk indicators according to the risk rating index system of power transmission and transformation projects;

[0045] S2. Determine the optimal and worst indicators of the evaluation indicators according to the opinions of the expert group;

[0046] S3. Determine the importance preference from the optimal indicator to other indicators; the decision maker compares the optimal indicator with other indicators pairwise, and uses numbers from 1 to 9 to give the preference relationship. 1 means the two are equally important, and 9 means the optimal indicator is very important relative to the latter, generating a "vector of the optimal indicator to other indicators". Similarly, obtain the preference of the worst indicator relative to other criteria, and generate a "vector of the worst indicator to other indicators";

[0047] S4. Use the linear BWM model to determine the weight of the optimal indicator:

[0048]

[0049] where: w' B is the subjective weight of the optimal indicator; w' w is the subjective weight of the worst indicator; w' j is the subjective weight of the jth indicator.

[0050] The objective weight is calculated by the CRITIC method, and the specific steps are as follows:

[0051] S1. Construct the samples to be evaluated. Assume there are m samples to be evaluated and n evaluation indicators, and establish the initial indicator data matrix;

[0052] S2. Data standardization. To eliminate the influence of different indicator dimensions on the results, perform dimensionless processing on the data to obtain the standardized matrix;

[0053] S3. Calculate the information carrying capacity. The volatility S j of the data can be expressed as:

[0054]

[0055] where: is the mean value of each indicator; b' ij represents the data after standardization;

[0056] S4. Calculate the conflict matrix R and information amount C j of the indicators, and the calculation formula is:

[0057]

[0058] where, r ij represents the element in the i-th row and j-th column of the conflict matrix R.

[0059] The conflict A jand the information carrying capacity C j is expressed as:

[0060]

[0061] C j = S j × A j ; (16)

[0062] S5. Calculate the weight wj’

[0063]

[0064] For the evaluation indicators, which involve multiple aspects such as economy, technology and time, the combined weight comprehensively considers the influence of subjective and objective weights. Through the combined weighting method of BWM-CRITIC, while weakening the subjective randomness of experts, it also reduces the over-reliance on the original data by the objective method, makes up for the defects of the single weighting method, so as to improve the authenticity and scientificity of the weighting of the construction safety evaluation system of power transmission and transformation projects. Finally, the weight calculation formula is as follows:

[0065]

[0066] In the formula is the subjective weight, is the objective weight.

[0067] Suppose there are n schemes A i , m evaluation indicators C j , x ij represents the evaluation information of scheme A i for indicator C j ; m modules M j , M j = (x 1j , x 2j , …, x nj ); T M N represents the module with numerical evaluation information; M P represents the module with probabilistic semantic evaluation information; M R represents the module with random variable evaluation information; After standardizing the evaluation information, when the evaluation information is numerical, the evaluation information of scheme A i for indicator C j is When the evaluation information is probabilistic semantic, the evaluation information of scheme A i for indicator C j is When the evaluation information is random variable type, the evaluation information of scheme A i for indicator C j is The evaluation information matrix after aggregating the evaluation information is as follows:

[0068]

[0069] Wherein: when the evaluation information is of the exact number type, when the evaluation information is of the probability semantics type, when the evaluation information of attribute C j is of the random variable type,

[0070] The Mo-EDAS method includes the following steps:

[0071] S1. Determine the average solution under each attribute, and the calculation formula is as follows:

[0072]

[0073] S2. Calculate the positive distance (PDA) and negative distance (NDA) between each solution and the average solution under each attribute;

[0074] If the attribute is a benefit type attribute, the calculation formula is:

[0075]

[0076] If the attribute is a cost type attribute, then the calculation formula is:

[0077]

[0078] S3. Calculate the weighted positive distance (SP) and weighted negative distance (SN):

[0079]

[0080] S4. Standardize the weighted positive distance and weighted negative distance to obtain the standardized weighted positive distance (NSP) and standardized weighted negative distance (NSN) respectively:

[0081]

[0082] S5. Calculate the final evaluation score:

[0083]

[0084] According to the final score AS i Sort the solutions. The higher the final evaluation score, the better the solution, and the more attention should be paid by the decision maker; on the contrary, it indicates that the solution is not ideal and should be considered for rejection.

[0085] The present invention has the following beneficial effects:

[0086] 1. For qualitative attribute features, the use of probabilistic language term sets for evaluation can well represent the uncertainty and hesitation degree of expert evaluation, and effectively avoid information loss caused by expert preference in expression;

[0087] 2. For quantitative forms of attribute features with uncertainty, random variable type evaluation information is used for evaluation, and information processing is carried out through a dominance matrix, which is simple to operate;

[0088] 3. Through the combined weighting method of BWM-CRITIC, the comprehensive weight of each index is determined to ensure the scientificity and objectivity of the index weight, and the weighted calculation of the standardized decision matrix is carried out to obtain a weighted decision matrix that better conforms to objective conditions as the basis for further evaluation and analysis calculations;

[0089] 4. In view of the information loss caused by information aggregation in the evaluation stage for mixed evaluation information, the Mo-EDAS method is proposed for the evaluation and ranking of solutions, avoiding information loss caused in the information conversion process and simplifying the calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] The present invention will be further described below in conjunction with the drawings and embodiments.

[0091] Figure 1 It is a flowchart of a method for selecting and evaluating emergency material suppliers for power transmission and transformation projects provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0092] The embodiments of the present invention will be further described below in conjunction with the drawings. However, it should not be understood that the scope of the above-mentioned subject matter of the present invention is limited to the following embodiments. Any technology implemented based on the content of the present invention belongs to the scope of the present invention.

[0093] Refer to Figure 1 , a hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method, including:

[0094] According to the different characteristics and types of safety risk evaluation indicators for power transmission and transformation project construction, precise number types and random variables are used to process quantitative evaluation information, and qualitative evaluation information is processed through probabilistic semantic term sets;

[0095] When the evaluation information is mixed information, considering that the mutual influence relationship between indicators is more complex, a combined weighting method is used to calculate the weight of each indicator;

[0096] The Mo-EDAS method is adopted to calculate the average ideal solution of each module according to the superiority and inferiority comparison methods of different types of evaluation information, calculate the distance weighted standard value of each alternative according to the distance formula of different types of evaluation information, finally calculate the evaluation score of each alternative, and evaluate and rank the alternatives according to the evaluation score.

[0097] In the power transmission and transformation project, there are often various types of attribute values. For example, in the problem of selecting an emergency plan, attributes such as the timeliness of rescue, economic cost, utilization of mechanical equipment, aftermath treatment, social impact of disasters, and on-site control of disasters are often considered. The timeliness of rescue is a quantitative attribute, the social impact of disasters is an uncertain quantitative attribute, and on-site control of disasters is a qualitative attribute, etc. This multi-attribute decision-making problem that includes both quantitative and qualitative attributes and also contains uncertainties conforms to the objective actual needs and has important research significance. For the evaluation of qualitative indicators, the probabilistic linguistic term set can accurately express the preference information of experts by assigning corresponding probabilities to the semantic set to express different degrees of preference. Therefore, in this paper, precise numbers and random variables are used to process quantitative evaluation information, and probabilistic language terms are used to process qualitative evaluation information.

[0098] Preferably, when the evaluation information is numerical, the range standardization method is used to standardize the evaluation information;

[0099] Let the numerical evaluation information be y ij , and the standardized evaluation information is The calculation formula is as follows:

[0100] If it is a cost-type attribute, then:

[0101]

[0102] If it is a benefit-type attribute, then:

[0103]

[0104] Furthermore, when the attribute is an uncertain quantitative attribute, the evaluation information adopts the random variable type, and the uncertain quantitative attribute is evaluated by constructing a dominance matrix for the attribute;

[0105] Construct a dominance matrix G for the attribute j =[g iqj n×n , where g iqj represents the probability that for the attribute C j , the alternative A i is superior to the alternative A q , and the calculation formula of g iqj is as follows:

[0106] If it is a cost-type attribute, then: ​

[0107]

[0108] If it is a benefit attribute, then:

[0109]

[0110] Wherein:: f i (g ij ) and f q (g qj ) are the probability density functions of g ij and g qj . According to the dominance matrix G j = [g iqj n×n Measure the advantages and disadvantages of each scheme for the attribute C j , so as to obtain the dominance degree of the attribute C j Where is an exact number, and the calculation formula is as follows:

[0111]

[0112] Normalize the dominance degree to obtain the normalized dominance degree The calculation formula is as follows:

[0113]

[0114] Preferably, in the decision-making problem of the power transmission and transformation project, due to the fuzziness of human thinking and the limitation of the cognition of known information, it is difficult to describe and express some qualitative evaluation indicators with accurate numerical values, resulting in hesitation and fuzziness. To solve the expression of such decision-making information, the present invention applies the probabilistic linguistic term set to process the qualitative evaluation information; the elements of the probabilistic linguistic term set are composed of linguistic evaluation values and corresponding probabilities, where the linguistic evaluation values come from the given linguistic term set, and the probability is the preference degree or proportion of the decision maker in favor of a certain linguistic evaluation value. The following is the definition of the probabilistic linguistic term set:

[0115] Definition 1. Let S = {s0,....., s a} be the semantic set, then the probabilistic linguistic term set is:

[0116]

[0117] Where L (k) (P (k) ) represents that the probability of the linguistic term set L (k) is P (k) , and #L(p) is the number of semantic sets included in all L(p);

[0118] ​Definition 2. Let \(L(p)_1\) and \(L(p)_2\) be two probabilistic linguistic term sets. If \(\#L(p)_1>\#L(p)_2\), then \(\#L(p)_1 - \#L(p)_2\) semantic sets need to be added to \(L(p)_2\) to make the number of elements in \(L(p)_1\) and \(L(p)_2\) the same. Add the smallest semantic set in \(L(p)_2\) with a weight of 0.

[0119] Definition 3. Let a probabilistic linguistic term set It needs to be standardized, and the formula is as follows:

[0120]

[0121] where \(L(k)(P(k))\) is sorted from small to large.

[0122] Definition 4. and are the \(k\)-th elements in \(L(p)_1\) and \(L(p)_2\) respectively, and \(P_1\) (k) and respectively represent and The distance formula between their weights is:

[0123]

[0124] where: and respectively represent the subscripts of \(L(p)_1\) and \(L(p)_2\).

[0125] Definition 5. Let \(L(p)=\{L (k) (p (k) )|1, 2,..., \#L(p)\}\) be a probabilistic linguistic term set, \(r(k)\) be the subscript of \(L(k)\), and the score function expression of \(L(p)\) is:

[0126]

[0127] where For probabilistic semantic sets \(L(p)_1\) and \(L(p)_2\), if \(E(L(p)_1)>E(L(p)_2)\), then \(L(p)_1>L(p)_2\).

[0128] Definition 6. Let \(L(p)=\{L (k) (p (k) )|1, 2,..., \#L(p)\}\) be a probabilistic linguistic term set, and the score function The deviation formula of \(L(p)\) is:

[0129]

[0130] Let the scoring functions of L(p)1 and L(p)2 be E(L(p)1) and E(L(p)2), and the deviation degrees be σ(L(p)1) and σ(L(p)2) respectively; when E(L(p)1) = E(L(p)2), if σ(L(p)1) < σ(L(p)2), then L(p)1 > L(p)2; if σ(L(p)1) = σ(L(p)2), then L(p)1 = L(p)2; if σ(L(p)1) > σ(L(p)2), then L(p)1 < L(p)2.

[0131] Preferably, the calculation of the weights of the evaluation indicators includes subjective weights, objective weights, and combined weighting. The subjective weight is calculated using the BWM method, and the objective weight is calculated using the CRITIC method. By the linear combination weighting method, the impacts of subjective and objective weights are comprehensively considered. While weakening the subjective randomness of experts, it also reduces the over-reliance on the original data by the objective method, makes up for the defects of the single weighting method, and thus improves the authenticity and scientificity of the indicator weighting for the power transmission and transformation project.

[0132] The subjective weight is calculated using the BWM method, and the specific steps are as follows:

[0133] S1. Determine the risk indicators according to the risk rating index system of the power transmission and transformation project;

[0134] S2. Determine the optimal and worst indicators of the evaluation indicators according to the opinions of the expert group;

[0135] S3. Determine the importance preference from the optimal indicator to other indicators; the decision maker compares the optimal indicator with other indicators pairwise and gives the preference relationship using the numbers 1 to 9. 1 means the two are equally important, and 9 means the optimal indicator is very important relative to the latter, generating the "optimal indicator to other indicators" vector. Similarly, obtain the preference of the worst indicator relative to other criteria and generate the "worst indicator to other indicators" vector;

[0136] S4. Use the linear BWM model to determine the weight of the optimal indicator:

[0137]

[0138] where: w' B is the subjective weight of the optimal indicator; w' w is the subjective weight of the worst indicator; w' j is the subjective weight of the j-th indicator.

[0139] The objective weight is calculated using the CRITIC method, and the specific steps are as follows:

[0140] S1. Construct the samples to be evaluated. Assume there are m samples to be evaluated and n evaluation indicators, and establish the initial indicator data matrix;

[0141] S2. Data standardization: To eliminate the influence of different index dimensions on the results, the data is dimensionless processed to obtain a standardized matrix.

[0142] S3. Calculate the information carrying capacity, the volatility S of the data j Can be expressed as:

[0143]

[0144] In the formula: Is the mean of each index; b' ij Represents the data after standardization;

[0145] S4. Calculate the conflict matrix R and information quantity C of the indexes j , and the calculation formula is:

[0146]

[0147] In the formula, r ij Represents the element in the i-th row and j-th column of the conflict matrix R.

[0148] The conflict A of the data j And the information carrying capacity C j Are expressed as:

[0149]

[0150] C j =S j ×A j ; (16)

[0151] S5. Calculate the weight wj'

[0152]

[0153] When using different single weighting methods, the results fluctuate greatly and are easily affected by discrete extreme values or subjective biases. For evaluation indexes, which involve multiple perspectives such as economy, technology and time, their combined weights comprehensively consider the influence of subjective and objective weights. Through the combined weighting method of BWM-CRITIC, while weakening the subjective randomness of experts, it also weakens the over-reliance of the objective method on the original data, makes up for the defects of the single weighting method, so as to improve the authenticity and scientificity of the weighting of the construction safety evaluation system of power transmission and transformation projects. Finally, the weight calculation formula is as follows:

[0154]

[0155] In the formula Is the subjective weight, Is the objective weight.

[0156] Furthermore, the Mo-EDAS method is adopted. First, the evaluation information type of an attribute is determined according to the attribute nature, and the evaluation information matrix is divided into independent modules according to the attributes. Then, the average ideal solution of each module is calculated respectively according to the superiority and inferiority comparison methods of different types of evaluation information. The distance weighted standard value of each alternative solution is calculated according to the distance formula of different types of evaluation information. Finally, the evaluation scores of each alternative solution are calculated, and the solutions are evaluated and sorted according to the evaluation scores. The information modular processing of the Mo-EDAS method not only effectively avoids information loss caused by the need to convert different types of evaluation information into the same type of information, but also can reasonably handle the problem that the mixed evaluation information cannot be converted into the same type and the superiority and inferiority of the solutions cannot be compared.

[0157] Suppose there are n solutions A i , m evaluation indicators C j , x ij represents the evaluation information of solution A i for indicator C j ; m modules M j , M j = (x ij , x 2j , …, x nj ); T ; M N represents the module with numerical evaluation information; M P represents the module with probabilistic semantic evaluation information; M R represents the module with random variable evaluation information; After the evaluation information is standardized, when the evaluation information is numerical, the evaluation information of solution A i for indicator C j is When the evaluation information is probabilistic semantic, the evaluation information of solution A i for indicator C j is When the evaluation information is random variable, the evaluation information of solution A i for indicator C j is The evaluation information matrix after aggregating the evaluation information is as follows:

[0158]

[0159] Among them: when the evaluation information is of exact number type, when the evaluation information is of probabilistic semantic type, when the evaluation information of attribute C j is of random variable type,

[0160] The Mo-EDAS method includes the following steps:

[0161] S1. Determine the average solution for each attribute, and the calculation formula is as follows:

[0162]

[0163] S2. Calculate the positive distance (PDA) and negative distance (NDA) between each solution and the average solution for each attribute;

[0164] If the attribute is a benefit - type attribute, the calculation formula is:

[0165]

[0166] If the attribute is a cost - type attribute, then the calculation formula is:

[0167]

[0168] S3. Calculate the weighted positive distance (SP) and weighted negative distance (SN):

[0169]

[0170] S4. Standardize the weighted positive distance and weighted negative distance to obtain the standardized weighted positive distance (NSP) and standardized weighted negative distance (NSN) respectively:

[0171]

[0172]

[0173] S5. Calculate the final evaluation score:

[0174]

[0175] According to the final score AS i Sort the solutions. The higher the final evaluation score, the better the solution, and the more attention should be paid by decision - makers; on the contrary, it indicates that the solution is not ideal and should be considered for abandonment.

[0176] Taking a certain power transmission and transformation project as an example, the accident site is located in the span between the 6# and 7# poles of the old Changwen Line. This span crosses the operating 10 kV Xiadian Line. When the construction workers used a winch to slacken the old conductor to the ground and were recovering the old conductor, the insulation layer was damaged, resulting in an electric shock accident during the pulling of the old conductor. After preliminary judgment, 4 sets of emergency rescue plans A1, A2, A3, and A4 were formulated. Now, the proposed decision-making model is applied to conduct decision-making analysis on the emergency rescue plans. The emergency hierarchy structure of the power transmission and transformation project of the present invention considers the attribute values of rescue timeliness C1, disaster site control C2, disaster site impact C3, economic cost C4, aftermath handling C5, and mechanical equipment utilization C6. Among them, C1, C3, and C4 are cost-type attributes, and C2, C5, and C6 are benefit-type attributes. The evaluation information of each attribute in the case analysis is shown in Table 1.

[0177] Table 1 Evaluation Information of Attribute Values of Each Plan

[0178]

[0179] According to Table 1, the evaluation information of attribute C1 is (10, 12, 15, 18). Using formula (1), we get

[0180] According to Table 1, the evaluation information of attribute C2 is (S2(0.7), S3(0.3); S2(0.7), S3(0.3); S2(0.8), S3(0.2); S2(0.8), S3(0.2)). Using formulas (7)-(11), we get the standardized probability semantic set. The probability semantic set of attribute C3 is Where:

[0181]

[0182] According to Table 1, the evaluation information of attribute C3 is:

[0183] (N(2, 0.5 2 ), N(2, 0.5 2 ), N(1.5, 0.5 2 ), N(1.5, 0.2 2 )) Using formulas (3)-(6), we get the dominance matrix G1 = [g iq1 3×3 As shown below:

[0184]

[0185] The dominance degree of attribute C3 That is

[0186] ​According to Table 1, the evaluation information of attribute C4 is (20, 16, 12, 10). Using formula (1), we get

[0187] According to Table 1, the evaluation information of attribute C5 is (S2(0.3), S3(0.7); S2(0.5), S3(0.5); S2(0.5), S3(0.5); S2(0.3), S3(0.7)). Using formulas (7)-(11), we obtain the standardized probability semantic set. The probability semantic set of attribute C5 is Where:

[0188]

[0189] According to Table 1, the evaluation information of attribute C6 is (0.85, 0.88, 0.90, 0.95). Using formula (2), we get

[0190] By normalizing the evaluation information, we obtain the evaluation information matrix after aggregating each module, as shown in Table 2.

[0191] Table 2 Evaluation Information after Aggregating Each Module

[0192]

[0193] Determine the average solutions of the 6 attributes according to the data in Table 2. Calculate according to formula (20) and obtain:

[0194] AV1 = 0.53, AV2 = {S2(0.75), S3(0.25)}, AV3 = 0.61, AV4 = 0.55, AV5 = {S2(0.4), S3(0.6)}, AV6 = 0.45.

[0195] Calculate the positive distance (PDA) and negative distance (NDA) between each solution and the average solution under each attribute. Construct the positive distance matrix and negative distance matrix as shown in Tables 3 and 4:

[0196] Table 3 Positive Distance Matrix

[0197]

[0198] Table 4 Negative Distance Matrix

[0199]

[0200] To avoid the weight imbalance caused by a single type of empowerment method, the present invention adopts a linear combination empowerment method to obtain a weight combination with a balanced subjectivity and objectivity. The subjective and objective weights are calculated through formulas (12)-(17), and the combined weight coefficient is taken as a1 = 0.4 and a2 = 0.6. The final weight is calculated through formula (18), and the results are shown in Table 5:

[0201] Table 5 Calculation of the Final Index Weights

[0202] Subjective weight Objective weight Combined weight <![CDATA[Rescue timeliness C1]]> 0.4340 0.2726 0.3327 <![CDATA[Disaster Site Control C2]]> 0.1785 0.1677 0.1720 <![CDATA[Disaster scene impact on C3]]> 0.1285 0.1350 0.1324 <![CDATA[Economic cost C4]]> 0.1071 0.1104 0.1091 <![CDATA[Follow-up processing C5]]> 0.0450 0.2000 0.1380 <![CDATA[The mechanical equipment utilizes C6]]> 0.1069 0.1143 0.1113

[0203] Combined with Table 5, calculate the positive distance (SP) including weights and the negative distance (SN) including weights according to formulas (25)-(26), and the results are shown in Table 6.

[0204] Table 6 Weighted Sum of Positive and Negative Distances

[0205]

[0206]

[0207] According to formulas (27)-(28), calculate the normalization of the positive distance including weights and the negative distance including weights to obtain the normalized positive distance of weights (NSP) and the normalized negative distance of weights (NSN), as shown in Table 7.

[0208] Table 7 Normalized Values of the Weighted Sums of Positive and Negative Distances

[0209] <![CDATA[NSP i > <![CDATA[NSN i > <![CDATA[S1]]> 0.6128 1.0000 <![CDATA[S2]]> 0.2521 0.6830 <![CDATA[S3]]> 0.4511 0.2281 <![CDATA[S4]]> 1.0000 0.4971

[0210] Calculate the final evaluation score through formula (29), and the calculation results are shown in Table 8 below:

[0211] Table 8 Final Calculated Scores

[0212] <![CDATA[A1]]> <![CDATA[A2]]> <![CDATA[A3]]> <![CDATA[A4]]> <![CDATA[AS i > 0.8064 0.4675 0.3396 0.7485

[0213] According to the final score AS i it can be seen that Plan A1 > A4 > A2 > A3. In the case of the present invention, the economic cost of Emergency Plan A5 is relatively high, but it is the best choice. The main reasons are that the rescue timeliness of this emergency plan is the shortest, and it has better disaster site control and aftermath handling. And rescue timeliness has a very high weight in the selection of emergency plans.

Claims

1. A hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method, characterized in that, Including: According to the characteristics and types of construction safety risk assessment indicators for power transmission and transformation projects, precise numerical types and random variables are used to process quantitative evaluation information, and probability semantic term sets are used to process qualitative evaluation information; When the evaluation information is mixed information, considering that the interaction relationship between indicators is more complex, the combined weighting method is used to calculate the weights of each indicator; The Mo-EDAS method is used to calculate the average ideal solution of each module according to the comparison method of the advantages and disadvantages of different types of evaluation information, calculate the distance weighted standard value of each alternative solution according to the distance formula of different types of evaluation information, finally calculate the evaluation scores of each alternative solution, and evaluate and rank the solutions according to the evaluation scores.

2. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 1, wherein: When the evaluation information is numerical, the range standardization method is used to standardize the evaluation information; Let the numerical evaluation information be y ij , and the standardized evaluation information is The calculation formula is as follows: If it is a cost-type attribute, then: If it is a benefit-type attribute, then:

3. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 2, wherein: When the attribute is an uncertain quantitative attribute, the evaluation information adopts a random variable type, and a dominance matrix for the attribute is constructed to evaluate the uncertain quantitative attribute; Construct the dominance matrix G for attributes j = [g iqj n×n , where g iqj represents the probability that for attribute C j , alternative A i is superior to alternative A q . The calculation formula for g iqj is as follows:​ If it is a cost-type attribute, then: If it is a benefit-type attribute, then: where: f i (g ij ) and f q (g qj ) are the probability density functions of g ij and g qj . According to the dominance matrix G j = [g iqj n×n to measure the advantages and disadvantages of each scheme for the attribute C j , so as to obtain the dominance degree of the attribute C j where is an exact number, and the calculation formula is as follows: ​​ Normalize the dominance degree to obtain the normalized dominance degree The calculation formula is as follows:

4. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 3, characterized in that: The elements of the probability language term set are composed of language evaluation values and corresponding probabilities. Among them, the language evaluation values come from the given language term set, and the probability is the preference degree or proportion of the decision maker in favor of a certain language evaluation value. The following is the definition of the probability language term set: Definition 1. Let \(S = \{s_0,\cdots,s\) a \}\) be a semantic set, then the probabilistic linguistic term set is as follows: where L (k) (P (k) ) represents the language term set L (k) has a probability of P (k) , and #L(p) is the number of semantic sets included in all L(p); Definition 2. Let \(L(p)_1\) and \(L(p)_2\) be two probabilistic linguistic term sets. If \(\#L(p)_1>\#L(p)_2\), then \(\#L(p)_1 - \#L(p)_2\) semantic sets need to be added to \(L(p)_2\) to make the number of elements in \(L(p)_1\) and \(L(p)_2\) the same. Add the smallest semantic set in \(L(p)_2\), and its weight is 0. Definition 3. Let a probabilistic linguistic term set which needs to be standardized. The formula is as follows: Among them, L (k) (p (k) ) is sorted from small to large; Definition 4, and are the k-th elements in L(p)1 and L(p)2 respectively, and represent respectively and weights, and their distance formula is: Wherein: and respectively represent the subscripts of L(p)1 and L(p)2; Definition 5. Let \(L(p)=\{L (k) (p (k) )|1, 2,\cdots, \#L(p)\}\) be a probabilistic linguistic term set, \(r(k)\) be the subscript of \(L(k)\), and the scoring function expression of \(L(p)\) is: Among them For probability semantic sets L(p)1 and L(p)2, if E(L(p)1) > E(L(p)2), then L(p)1 > L(p)2; Definition 6. Let \(L(p)=\{L (k) (p (k) )|1, 2,\cdots, \#L(p)\}\) be a probabilistic linguistic term set, and the scoring function The deviation formula of \(L(p)\) is: Let the score functions of L(p)1 and L(p)2 be E(L(p)1) and E(L(p)2), and the deviation degrees be σ(L(p)1) and σ(L(p)2); when E(L(p)1) = E(L(p)2), if σ(L(p)1) < σ(L(p)2), then L(p)1 > L(p)2; if σ(L(p)1) = σ(L(p)2), then L(p)1 = L(p)2; if σ(L(p)1) > σ(L(p)2), then L(p)1 < L(p)2.

5. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 4, characterized in that: The calculation of the weights of evaluation indicators includes subjective weights, objective weights, and combined weights.

6. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 5, wherein: The subjective weight is calculated using the BWM method, and the specific steps are as follows: S1. Determine the risk indicators according to the risk rating index system of the power transmission and transformation project; S2. Determine the optimal and worst indicators of the evaluation indicators according to the opinions of the expert group; S3. Determine the importance preference from the optimal indicator to other indicators; the decision maker compares the optimal indicator with other indicators pairwise, and uses the numbers 1 to 9 to give the preference relationship. 1 means that the two are equally important, and 9 means that the optimal indicator is very important relative to the latter, generating a "vector of the optimal indicator to other indicators". Similarly, the preference of the worst indicator relative to other criteria is obtained, generating a "vector of the worst indicator to other indicators"; S4. Use the linear BWM model to determine the weight of the optimal indicator: where: w' B is the subjective weight of the optimal index; w' w is the subjective weight of the worst indicator; w' j is the subjective weight of the j-th indicator.

7. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 6, characterized in that: The objective weight is calculated using the CRITIC method, and the specific steps are as follows: S1. Construct the samples to be evaluated. Suppose there are m samples to be evaluated and n evaluation indicators, and establish the initial index data matrix; S2. Data standardization. To eliminate the influence of different index dimensions on the results, the data is dimensionless processed to obtain the standardized matrix; S3. Calculate the information carrying capacity and the volatility S of the data j It can be expressed as: In the formula: is the mean value of each index; b' ij represents the data after standardization; S4. Calculate the conflict matrix R and information quantity C of the indicators j , and the calculation formula is as follows: where r ij represents the element in the i-th row and j-th column of the conflict matrix R. Conflict A of data j and information carrying capacity C j It is expressed as: C j = S j × A j ; (16) S5. Calculate the weight wj’ 8. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 7, wherein: For the evaluation indicators, which involve multiple perspectives such as economy, technology, and time, the combined weight comprehensively considers the influence of subjective and objective weights. Through the combined weighting method of BWM-CRITIC, while weakening the subjective randomness of experts, it also reduces the over-reliance on the original data by the objective method, makes up for the defects of the single weighting method, thereby improving the authenticity and scientificity of the weighting of the construction safety evaluation system for power transmission and transformation projects. The final weight calculation formula is as follows: In the formula is the subjective weight, is the objective weight.

9. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 8, characterized in that: There are n solutions A i , and m evaluation indicators C j , x ij represents the evaluation information of solution A i for indicator C j ; m modules M j , M j = (x ij , x 2j , …, x nj ) T ; M N represents the module with numerical evaluation information; M P represents the module with probabilistic semantic evaluation information; M R represents the module with random variable evaluation information; after standardizing the evaluation information, when the evaluation information is numerical, the evaluation information of solution A i for indicator C j is When the evaluation information is probabilistic semantic, the evaluation information of solution A i for indicator C j is When the evaluation information is random variable type, the evaluation information of solution A i for indicator C j is The evaluation information matrix after aggregating the evaluation information is as follows: Wherein: when the evaluation information is of the exact number type, when the evaluation information is of the probabilistic semantic type, when the j evaluation information of attribute C 10. The hybrid multi-attribute decision-making method for power transmission and transformation projects based on the Mo-EDAS method according to claim 9, wherein The Mo-EDAS method includes the following steps: S1. Determine the average solution under each attribute, and the calculation formula is as follows: S2. Calculate the positive distance (PDA) and negative distance (NDA) between each solution and the average solution under each attribute; If the attribute is a benefit-type attribute, the calculation formula is: If the attribute is a cost-type attribute, then the calculation formula is: S3. Calculate the positive distance (SP) with weights and the negative distance (SN) with weights: S4. Standardize the positive distance with weights and the negative distance with weights to obtain the standardized weight positive distance (NSP) and the standardized weight negative distance (NSN) respectively: S5. Calculate the final evaluation score: According to the final score AS i Sort the solutions. The higher the final evaluation score, the better the solution, and the more attention should be paid by decision-makers; on the contrary, it indicates that the solution is not ideal and should be considered for abandonment.

Images