Offshore wind turbine collision damage evaluation method based on improved weighting method

Abstract

The invention discloses an offshore wind turbine collision damage evaluation method based on an improved weighting method, and relates to the technical field of ocean engineering, and the method comprises the following steps: S1, obtaining damage data and processing the data: constructing a ship collision offshore wind turbine finite element simulation model, simulating the collision under different collision conditions, and obtaining the damage data; extracting a damage index and constructing a mathematical model between the collision damage factor and a damage evaluation index; s2, distributing each collision damage factor and each damage evaluation index weight: correcting the weight by using an analytic hierarchy process of objective information correction, selecting and rejecting weight calculation by using a fuzzy inverse equation method, determining a plurality of groups of possible weight vectors as a standby selection set of the fuzzy inverse equation method through subjective and objective combination, solving a Nash equilibrium solution of the plurality of groups of weight vectors, and calculating a collision damage evaluation index weight according to the Nash equilibrium solution of the plurality of groups of weight vectors; the optimal weight is solved; and S3, implementation of an evaluation model: determining a fan damage grade division basis and corresponding maintenance measures, defining a membership function, constructing a second-order comprehensive evaluation matrix, and outputting a damage degree through fuzzy comprehensive evaluation.

Description

Technical Field

[0001] The present invention relates to the technical field of ocean engineering, and more specifically, it relates to a method for evaluating the collision damage of an offshore wind turbine by improving the weight assignment method. Background Art

[0002] As an important part of clean energy, offshore wind power occupies an increasingly important position in the global energy structure. Due to its unique operating environment and working characteristics, offshore wind turbines are prone to being affected by accidental events such as ship collisions, resulting in structural damage and increased maintenance costs. Therefore, damage safety assessment and monitoring of offshore wind turbines are the keys to ensuring their stable operation and reducing operation risks. Currently, the damage evaluation of offshore wind turbines mainly relies on strain gauges, vibration sensors, etc. to obtain the operation data of the wind turbines and extract characteristic values, and uses artificial intelligence, machine learning algorithms, etc. for structural health monitoring to judge the damage condition and even the location. Or based on the model modification method, the damage parameters of the finite element model are iteratively updated to make it match the test data after actual damage, so as to realize damage identification. However, these methods have complex systems, are accompanied by high costs, and have high requirements for the accuracy and real-time of data.

[0003] After an offshore wind turbine suffers accidental events such as ship collisions, it is often difficult to accurately evaluate the overall damage degree through traditional methods within the allowable range of a single index. Because there are complex relationships between damage factors such as collision speed, ship mass, and offset collision angle and the response of the wind turbine structure, and the actual damage situation may involve combinations of multiple damage modes and degrees, and it is fuzzy to judge the damage degree, all of which pose challenges to accurately evaluating the damage level. In addition, the evaluation results are significantly affected by the index weights, and the determination of weights should follow certain principles. The method for determining weights should have reasonable mathematical logic, contain clear practical significance, and meet the need for accurately evaluating the damage degree. Summary of the Invention

[0004] Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide a method for evaluating the collision damage of an offshore wind turbine by improving the weight assignment method, which has the advantages of reasonable mathematical logic and accuracy.

[0005] To achieve the above purpose, the present invention provides the following technical solution: A method for evaluating the collision damage of an offshore wind turbine by improving the weight assignment method, comprising the following steps:

[0006] S1. Obtaining damage data and data processing: Constructing a finite element simulation model of a ship colliding with an offshore wind turbine, simulating collisions under different collision conditions, extracting damage indicators, and constructing a mathematical model between collision damage factors and damage evaluation indicators;

[0007] S2. Assign weights to each collision damage factor and each damage evaluation index: Use the analytic hierarchy process (AHP) corrected by objective information to correct the weights, and use the fuzzy inverse equation method to calculate and select the weights. Combine subjective and objective factors to determine multiple possible weight vectors as the alternative set of the fuzzy inverse equation method, and obtain the Nash equilibrium solution of multiple weight vectors to solve the optimal weights;

[0008] Among them, the corrected weight judgment index is:

[0009]

[0010] In the formula, a i,j is the data in the i-th row and j-th column of the judgment matrix, E i is the information entropy of the damage data corresponding to index i, std i is the standard deviation of the damage data corresponding to index i, R i,j is the correlation coefficient of index i to index j, W i is the weight vector corresponding to index i;

[0011] S3. Implementation of the evaluation model: Determine the basis for classifying the damage level of the fan and its corresponding maintenance measures, define the membership function, construct a second-order comprehensive evaluation matrix, and output the damage degree through fuzzy comprehensive evaluation.

[0012] Furthermore, in step S2, in the analytic hierarchy process, for each possible situation of each importance level, construct its judgment matrix and conduct a consistency test, retain those that pass the consistency test, and discard those that cannot pass.

[0013] Furthermore, in step S2, take the geometric mean of each damage evaluation index in the judgment matrix as the basis for subjective weight assignment, and calculate the objective data of each index as the correction term for the geometric mean. The characteristics of the objective data include:

[0014] 1) Information entropy, and its calculation method is:

[0015] Among them,

[0016] In the formula, e j is the information entropy of the j-th evaluation index, k represents the normalization of the information entropy, p ij represents the proportion of the i-th damage evaluation index under the j-th damage evaluation index, n is the order of one of the damage evaluation indexes, x ij is the data in the i-th row and j-th column of the comprehensive evaluation matrix, and m is the order of the other damage evaluation index;

[0017] 4) 2) Standard deviation, and its calculation method is:

[0018]

[0019] Wherein, μ is the data mean value, and x i is the data corresponding to the i-th damage evaluation index;

[0020] 5) Pearson correlation coefficient, and its calculation method is:

[0021]

[0022] Wherein, R ij is the correlation coefficient between two damage evaluation indexes, Cov(X,Y) is the covariance of two data, and s X , s Y is the standard deviation of two data.

[0023] Furthermore, the method for solving the optimal weight is:

[0024]

[0025] Wherein, ω is the optimal weight, is the weight vector set, and β k is the weight coefficient;

[0026] When ,

[0027] the corresponding linear equation set is:

[0028]

[0029] In formula (5), β k is the weight coefficient, is the weight vector set, ω i is the optimal weight of index i; in formula (6), ω G is the optimal weight of index G, is the weight vector of index G in the weight vector set, and β G is the weight coefficient of index G; the optimal linear combinations β1, β2,..., β n can be obtained from formula (6), and after normalization, the optimal weight is output according to formula (4).

[0030] Furthermore, in step S1, the collision damage factors include collision speed, ship mass, and offset collision angle; the damage evaluation indexes include area damage rate, natural frequency of the fan, maximum displacement of the tower top during the collision, and maximum displacement of the pile foundation mud surface.

[0031] Furthermore, in step S1, constructing the mathematical model between the collision damage factors and the damage evaluation indexes includes the following steps:

[0032] First, interpolate the collision damage data and use the binary polynomial equation shown in Equation (1) to fit the damage index under different collision speeds and ship masses:

[0033] G(x) = g1x 3 + g2y 3 + g3x 2 y + g4y 2 x + g5x 2 + g6y 2 + g7xy + g8x + g9y + g 10 (1)

[0034] In the formula, x represents the ship mass, y represents the collision speed, and g1, g2... g 10 represent fitting coefficients;

[0035] Secondly, use the tangent function shown in Equation (2) to fit the damage index under different offset collision angles:

[0036] f(x) = a·arctan(bx 2 + cx + d) + e (2)

[0037] In the formula, x represents the offset collision angle; a, b, c, d, and e are all fitting coefficients;

[0038] Furthermore, in step S3, the membership function is a Z-type membership function:

[0039]

[0040] When the index value is positively correlated with the damage degree, M1(x) is adopted; when the index value is negatively correlated with the damage degree, M2(x) is adopted.

[0041] Furthermore, in step S3, the Monte Carlo simulation method is used to process the output result of the membership function of a single damage evaluation index, generate the damage membership vector of this index value, merge all the damage membership vectors of the index values to obtain the first-order fuzzy comprehensive evaluation matrix R1, the weight of the first-order fuzzy evaluation matrix is A1, and the first-order fuzzy comprehensive evaluation result B1 can be obtained by performing a fuzzy transformation on the first-order fuzzy comprehensive evaluation matrix R1 and the weight A1:

[0042]

[0043] Take the B1 calculated for each factor in the collision damage factor set as the row vector of the second-order fuzzy comprehensive evaluation matrix to obtain the second-order fuzzy comprehensive evaluation matrix R2. The weight vector of the collision damage factor set is A2, and the second-order comprehensive evaluation result B2 can be obtained by performing a fuzzy transformation on the second-order fuzzy comprehensive evaluation matrix R2 and the weight vector A2 of the collision damage factor set:

[0044]

[0045] Further, in step S3, A1 = [K1, K2, K3, K4], where K1, K2, K3, and K4 respectively represent the membership degrees of the four indicators of the area damage rate, the natural frequency of the wind turbine, the maximum displacement at the top of the tower barrel, and the maximum displacement of the pile foundation mud surface in the comprehensive fuzzy evaluation; A2 = [P1, P2], where P1 and P2 respectively represent the membership degrees of the coupling factor of the collision speed & ship mass and the offset collision angle in the comprehensive evaluation process.

[0046] In summary, the present invention has the following beneficial effects: The method of the present invention is based on the fuzzy theory, uses the fuzzy comprehensive evaluation method to establish a damage model for ships colliding with offshore wind turbines, uses multiple damage indicators and damage factors as model input information, comprehensively considers the combined effect of each index information in the damage evaluation process, and successfully realizes the purpose of using low-complexity calculations to output relatively accurate damage evaluation results; The method of the present invention improves the weight determination method in the comprehensive evaluation process, combines subjective experience judgment with multiple groups of objective data information, and effectively processes the fuzzy points in the subjective judgment process, with clear mathematical and physical meanings, which is beneficial to improving the accuracy of the evaluation model. Description of the Drawings

[0047] Figure 1 It is a schematic diagram of the weight distribution method of the method provided by the present invention. Detailed Embodiments

[0048] The following further describes the present invention in detail with reference to embodiments.

[0049] Currently, a relatively efficient method for evaluating the collision damage of offshore wind turbines is to construct a damage evaluation model, establish a collision damage factor set and a damage evaluation index set through methods such as the fuzzy comprehensive evaluation method and the topsis comprehensive evaluation method, and construct a damage evaluation system through several damage indicators that can be simply observed. In the process of constructing the damage evaluation system, in order to make the evaluation results more reasonable, it is necessary to assign weights to each damage evaluation index to determine its contribution degree to the damage evaluation results.

[0050] To overcome the deficiencies of the prior art, the present invention provides a method for evaluating the collision damage of offshore wind turbines with an improved weight assignment method, including the following steps:

[0051] S1. Obtain damage data and data processing: Construct a finite element simulation model of a ship colliding with an offshore wind turbine, simulate collisions under different collision conditions, extract damage indicators, and construct a mathematical model between the collision damage factors and the damage evaluation indicators;

[0052] In this embodiment, the collision damage factors are the collision speed, the ship mass, and the offset collision angle; the damage evaluation indicators are the area damage rate, the natural frequency of the fan, the maximum displacement at the top of the tower during the collision, and the maximum displacement of the pile foundation mud surface;

[0053] Among them, the calculation method of the area damage rate is as follows: whether the equivalent plastic strain (PEEQ) of the grid element is 0 during the collision is used to judge whether the grid element is damaged, and the sum of the areas of all damaged elements is divided by the total area to obtain the area damage rate; the calculation method of the natural frequency of the fan is to export the grid elements after the collision damage deformation for modal analysis, and the first-order modal frequency obtained is the natural frequency of the fan; the maximum displacement at the top of the tower during the collision and the maximum displacement of the pile foundation mud surface are directly extracted from the simulation results;

[0054] Constructing a mathematical model between the collision damage factors and the damage evaluation indicators includes the following steps:

[0055] A. First, interpolate the collision damage data, and use the binary polynomial equation shown in Equation (1) to fit the damage indicators under different collision speeds and ship masses:

[0056] G(x) = g1x 3 + g2y 3 + g3x 2 y + g4y 2 x + g5x 2 + g6y 2 + g7xy + g8x + g9y + g 10 (1)

[0057] In the formula, x represents the ship mass, y represents the collision speed, and g1, g2... g 10 represents the fitting coefficient

[0058] B. Secondly, use the tangent function shown in Equation (2) to fit the damage indicators under different offset collision angles:

[0059] f(x) = a·arctan(bx 2 + cx + d) + e (2)

[0060] In the formula, x represents the offset collision angle; a, b, c, d, and e are all fitting coefficients.

[0061] S2. Assign weights to each collision damage factor and each damage evaluation indicator: Use the analytic hierarchy process corrected by objective information to correct the weights, use the fuzzy inverse equation method to calculate and select the weights, combine the subjective and objective to determine multiple possible weight vectors as the alternative set of the fuzzy inverse equation method, and obtain the Nash equilibrium solution of multiple weight vectors to solve the optimal weights;

[0062] In the analytic hierarchy process of this embodiment, judgment matrices are constructed for various possible situations of each level of importance and consistency tests are carried out. Those passing the consistency test are retained, while those failing are discarded. For the often ambiguous parts of the importance levels in the conventional judgment matrices, this method retains the controversial parts and lists multiple possibilities to increase the accuracy of the evaluation results.

[0063] In this example, the geometric mean of each damage evaluation index in the judgment matrix is used as the basis for subjective weighting, and the objective data of each index is calculated as a correction term for the geometric mean. The characteristics of the objective data include:

[0064] 1) Information entropy, and its calculation method is:

[0065] Where,

[0066] In the formula, e j is the information entropy of the jth evaluation index, k represents the normalization of the information entropy, p ij represents the proportion of the ith damage evaluation index under the jth damage evaluation index, n is the order of one of the damage evaluation indexes, x ij is the data in the ith row and jth column of the comprehensive evaluation matrix, and m is the order of the other damage evaluation index;

[0067] 2) Standard deviation, and its calculation method is:

[0068]

[0069] In the formula, μ is the data mean, and x i is the data corresponding to the ith damage evaluation index;

[0070] 3) Pearson correlation coefficient, and its calculation method is:

[0071]

[0072] In the formula, R ij is the correlation coefficient between two damage evaluation indexes, Cov(X, Y) is the covariance of the two data, s X , s Y are the standard deviations of the two data.

[0073] The greater the information entropy, the higher the uncertainty of the indicator, the smaller the amount of information provided, and the relatively smaller its weight should be. Conversely, the smaller the information entropy, the greater the weight of the indicator. In a multi-indicator evaluation scheme, when the standard deviation of an indicator is large, it means that the performance of different collision conditions under this indicator varies greatly. Therefore, this indicator is more important for the evaluation process. The Pearson correlation coefficient can be used to identify and analyze the correlation between two or more indicators to determine whether there is overlap in the information provided by each indicator, which can reduce redundancy in the model and avoid repeated consideration of the same information during the evaluation process;

[0074] The modified weight determination index is:

[0075]

[0076] In the formula, α i，j is the data in the i-th row and j-th column of the judgment matrix, E i is the information entropy of the damage data corresponding to indicator i, std i is the standard deviation of the damage data corresponding to indicator i, R i，j is the correlation coefficient of indicator i to indicator j, W i is the weight vector corresponding to indicator i;

[0077] The method for solving the optimal weight is:

[0078]

[0079] In the formula, ω is the optimal weight, is the set of weight vectors, β k is the weight coefficient;

[0080] When the corresponding linear equations are:

[0081] In formula (5), β

[0082]

[0083] is the weight coefficient, k is the set of weight vectors, ω is the optimal weight of indicator i; In formula (6), ω i is the optimal weight of indicator G, G is the optimal weight of indicator G, is the weight vector of indicator G in the set of weight vectors, β G is the weight coefficient of indicator G;

[0084] From formula (6), the optimal linear combinations β1, β2,..., β n can be obtained. After normalizing them, the optimal weight is output according to formula (4);

[0085] When the formula (5) is satisfied, ω can be made to minimize the deviation from the remaining weights; through the above settings, the deviation of the weighted sum can be minimized, so as to achieve consistency among different weights. By obtaining the Nash equilibrium solution of multiple groups of weight vectors as the optimal weights, large differences among the weights of each group caused by subjective ambiguity can be avoided.

[0086] S3. Implementation of the evaluation model: Determine the basis for classifying the damage level of the fan and its corresponding maintenance measures, define the membership function, construct a second-order comprehensive evaluation matrix, and output the damage degree through fuzzy comprehensive evaluation. In this embodiment, the basis for classifying the damage level of the fan and its corresponding maintenance measures are shown in Table 1.

[0087] Table 1 Basis for classifying the damage level of the fan and its corresponding maintenance measures

[0088]

[0089] In this embodiment, the membership function is a Z-type membership function:

[0090]

[0091] When the index value is positively correlated with the damage degree, M1(x) is adopted; when the index value is negatively correlated with the damage degree, M2(x) is adopted.

[0092] In this embodiment, the Monte Carlo simulation method is used to process the output results of the single-index membership function. Taking the output membership degree M as the reference value, a gain coefficient K is randomly generated. The distribution of K satisfies a normal distribution with M as the mean and a standard deviation σ = 0.5M, as shown in formula (19):

[0093]

[0094] By statistically analyzing the frequencies of the product K*M of the gain coefficient and the single-index membership degree value in each membership degree interval, the membership degree of the damage degree corresponding to each damage level under the current index value is obtained; after determining the single-index membership function, the corresponding membership degree of the damage degree M can be given according to the index value. Subsequently, the Monte Carlo simulation method is used for 1000 random simulations. Finally, the number of simulation points falling in the membership degree intervals corresponding to each damage level is counted, and a 1×5 current index value damage membership degree vector is generated; the four vectors are combined to obtain a first-order fuzzy comprehensive evaluation matrix R1; the weight of the first-order fuzzy evaluation matrix is A1, a1 = [K1, K2, K3, K4], where K1, K2, K3, K4 respectively represent the importance membership degrees of the four indexes of the area damage rate, the natural frequency of the fan, the maximum displacement at the top of the tower barrel, and the maximum displacement of the pile foundation mud surface in the comprehensive fuzzy evaluation;

[0095] Perform a fuzzy transformation on the first-order fuzzy comprehensive evaluation matrix R1 and the weight A1 to obtain the first-order fuzzy comprehensive evaluation result B1:

[0096]

[0097] Take the B1 obtained by calculating each factor in the set of collision damage factors as the row vector of the second-order fuzzy comprehensive evaluation matrix to obtain the second-order fuzzy comprehensive evaluation matrix R2; the second-order fuzzy comprehensive evaluation matrix R2 is a 2×5 matrix, and its weight vector is A2, A2 = [P1, P2], where P1 and P2 respectively represent the membership degrees of the collision speed & ship mass coupling factor and the offset collision angle in the comprehensive evaluation process; perform a fuzzy transformation on the second-order fuzzy comprehensive evaluation matrix R2 and the factor set weight vector A2 to obtain the second-order comprehensive evaluation result B2:

[0098]

[0099] B2 is a 1×5 matrix, and each element in it represents the membership degree of the damage degree corresponding to the comments "complete destruction", "severe damage", "general damage", "minor damage", and "no damage" under the current collision conditions. Finally, according to the principle of maximum membership degree, it is considered that the comment with the highest membership degree is the actual damage degree of the wind turbine, and corresponding maintenance measures are taken accordingly.

[0100] The present invention constructs a two-level fuzzy comprehensive evaluation model for offshore wind turbine collision damage based on fuzzy theory. By establishing clear criteria for determining damage levels and determining the corresponding membership degree ranges of damage degrees, the constructed comprehensive evaluation model can well handle the fuzziness of damage degrees, output accurate damage levels, and thus reasonably arrange corresponding maintenance measures; at the same time, it improves the existing subjective and objective fusion weighting method, combines subjective and objective information, and solves the Nash equilibrium solution of multiple groups of weight vectors generated due to the fuzziness of subjective judgment with the game theory combined weighting method to obtain the best weight that can reflect both expert experience and objective data information.

[0101] This specific embodiment is only an explanation of the present invention and does not limit the present invention. Those skilled in the art can make modifications to this embodiment without creative contributions according to needs after reading this specification, but as long as it is within the scope of the claims of the present invention, it is protected by the patent law.

Claims

1. An evaluation method for the collision damage of an offshore wind turbine with an improved weighting method, characterized in that, It includes the following steps: S1. Obtain damage data and data processing: Construct a finite element simulation model of a ship colliding with an offshore wind turbine, simulate collisions under different collision conditions, extract damage indicators, and construct a mathematical model between collision damage factors and damage evaluation indicators; S2. Assign weights to each collision damage factor and each damage evaluation indicator: Use the analytic hierarchy process modified by objective information to correct the weights, use the fuzzy inverse equation method to calculate and select the weights, combine subjective and objective factors to determine multiple possible weight vectors as the alternative set of the fuzzy inverse equation method, and obtain the Nash equilibrium solution of multiple groups of weight vectors to solve the optimal weights; Among them, the modified weight determination index is: Where α i，j is the data at the i-th row and j-th column of the judgment matrix, E i is the information entropy of the damage data corresponding to index i, std i is the standard deviation of the damage data corresponding to index i, R i，j is the correlation coefficient of index i with respect to index j, W i is the weight vector corresponding to index i; S3. Realization of the evaluation model: Determine the basis for dividing the damaged levels of the wind turbine and their corresponding maintenance measures, define the membership function, construct a second-order comprehensive evaluation matrix, and output the damage degree through fuzzy comprehensive evaluation.

2. For the offshore wind turbine collision damage evaluation method with the improved weight assignment method according to claim 1, construct its judgment matrix and conduct a consistency test, retain those that pass the consistency test, and discard those that cannot pass.

3. The method for evaluating the collision damage of an offshore wind turbine using the improved empowerment method according to claim 1, characterized in that, In step S2, take the geometric mean of each damage evaluation indicator in the judgment matrix as the basis for subjective weight assignment, and calculate the objective data of each indicator as the correction term for the geometric mean. The characteristics of the objective data include: 1) Information entropy, and its calculation method is: Among them, where, e j is the information entropy of the j-th evaluation index, k represents the normalization of the information entropy, p ij represents the proportion of the i-th damage evaluation index under the j-th damage evaluation index, n is the order of one of the damage evaluation indices, x ij is the data in the i-th row and j-th column of the comprehensive evaluation matrix, and m is the order of the other damage evaluation index; 2) Standard deviation, and its calculation method is: where μ is the data mean, and x i is the data corresponding to the i-th damage evaluation index; 3) Pearson correlation coefficient, and its calculation method is: where R ij is the correlation coefficient between two damage evaluation indices, Cov(X, Y) is the covariance of two data, s X , s Y are the standard deviations of two data.

4. The method for evaluating the collision damage of an offshore wind turbine by the improved empowerment method according to claim 3, characterized in that, The method for solving the optimal weights is: where ω is the optimal weight, is the weight vector set, and β k is the weight coefficient; When then The corresponding linear equations are: In formula (5), β k is the weight coefficient, is the weight vector set, ω i is the optimal weight of index i; in formula (6), ω G is the optimal weight of index G, is the weight vector of index G in the weight vector set, β G is the weight coefficient of index G; the optimal linear combinations β1, β2,..., β n can be obtained from formula (6), and the optimal weights are output according to formula (4) after normalization.

5. The method for evaluating the collision damage of an offshore wind turbine by the improved empowerment method according to claim 1, characterized in that, In step S1, the collision damage factors include collision speed, ship mass, and offset collision angle; the damage evaluation indicators include area damage rate, natural frequency of the wind turbine, maximum displacement of the top of the tower during the collision, and maximum displacement of the pile foundation mud surface.

6. The method for evaluating the collision damage of an offshore wind turbine by the improved empowerment method according to claim 5, characterized in that In step S1, constructing a mathematical model between collision damage factors and damage evaluation indicators includes the following steps: First, interpolate the collision damage data, and use the binary polynomial equation shown in Equation (1) to fit the damage indicators under different collision speeds and ship masses: G(x) = g1x 3 + g2y 3 + g3x 2 y + g4y 2 x + g5x 2 + g6y 2 + g7xy + g g x + g9y + 9 10 (1) where x represents the mass of the ship, y represents the collision speed, and g1, g2... g 10 represent fitting coefficients; secondly, use the tangent function shown in Equation (2) to fit the damage index under different oblique collision angles: f(x) = a·arctan(bx 2 + cx + d)+ e (2) In the formula, x represents the offset collision angle; a, b, c, d, and e are all fitting coefficients.

7. The method for evaluating the collision damage of an offshore wind turbine by using the improved empowerment method according to any one of claims 1-6, characterized in that, In step S3, the membership function is a Z-type membership function: When the index value is positively correlated with the damage degree, use M1(x); when the index value is negatively correlated with the damage degree, use M2(x).

8. The method for evaluating the collision damage of an offshore wind turbine by the improved empowerment method according to claim 7, characterized in that, In step S3, use the Monte Carlo simulation method to process the output results of the membership function of a single damage evaluation indicator, generate the damage membership vector of the index value, combine all the damage membership vectors of the index values to obtain the first-order fuzzy comprehensive evaluation matrix R1, the weight of the first-order fuzzy evaluation matrix is A1, and perform a fuzzy transformation on the first-order fuzzy comprehensive evaluation matrix R1 and the weight A1 to obtain the first-order fuzzy comprehensive evaluation result B1: Take the B1 calculated for each factor in the collision damage factor set as the row vector of the second-order fuzzy comprehensive evaluation matrix to obtain the second-order fuzzy comprehensive evaluation matrix R2. The weight vector of the collision damage factor set is A2, and perform a fuzzy transformation on the second-order fuzzy comprehensive evaluation matrix R2 and the weight vector A2 of the collision damage factor set to obtain the second-order comprehensive evaluation result B2:

9. The method for evaluating the collision damage of an offshore wind turbine using the improved empowerment method according to claim 8, characterized in that, In step S3, A1 = [K1, K2, K3, K4], where K1, K2, K3, and K4 respectively represent the membership degrees of the four indicators of area damage rate, natural frequency of the fan, maximum displacement at the top of the tower barrel, and maximum displacement of the pile foundation mud surface in the comprehensive fuzzy evaluation; A2 = [P1, P2], where P1 and P2 respectively represent the membership degrees of the coupling factor of collision speed & ship mass and the offset collision angle in the comprehensive evaluation process.

Images