Wind power plant site selection decision-making method

Abstract

The invention relates to a wind power plant site selection decision-making method, which comprises the following steps of: obtaining related influence factor data of a historical reference wind power plant site selection case and operation measurement index data after the case is implemented, and establishing a three-level evaluation network structure; based on a variable evaluation network structure reasoning condition probability table, through a three-level evaluation network structure, establishing a cost function for reflecting score corresponding measurement index data validity; a cost function minimum value is obtained based on sequence least square method optimization, and a corresponding conditional probability table is obtained; obtaining related influence factor data of the candidate wind power plant; and evaluating the candidate wind power plants based on the obtained conditional probability table, and making an optimal wind power plant site selection decision according to an evaluation score. Compared with the prior art, the method has the advantages that adverse effects of subjective factors suggested by experts on final selection can be avoided, objective measurement indexes are innovatively introduced to serve as optimization evaluation standards, and objective, reasonable and systematic wind power plant site selection scheme decision making is achieved.

Description

Technical Field

[0001] The present invention relates to the field of offshore wind farm site selection decision-making, and in particular to a wind farm site selection decision-making method based on a user-defined cost function and sequential least squares optimization. Background Art

[0002] With the rapid development and progress of the wind power industry, the site selection of offshore wind farms has become a key issue in the field of wind power. The factors affecting the site selection of offshore wind farms include production benefits, construction costs, operating costs, construction risks and other aspects.

[0003] The technical problems existing in the traditional offshore wind farm site selection methods include: excessive reliance on expert opinions brings uncertain subjective factors to the decision-making results, and the lack of objective evaluation and measurement may cause the site selection decision results to deviate from the expected effects. Summary of the invention

[0004] The purpose of the present invention is to provide a wind farm site selection decision method based on a custom cost function and sequential least squares optimization, so as to achieve a reasonable and feasible wind farm site selection under an objective measurement index target.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] A wind farm site selection decision method based on a custom cost function and sequential least squares optimization includes the following steps:

[0007] S1, obtain the relevant influencing factor data of historical reference wind farm site selection cases, as well as the operational measurement indicator data after the implementation of the cases, and establish a three-level evaluation network structure for comprehensive analysis based on the relevant influencing factor data;

[0008] S2, based on the variable evaluation network structure inference conditional probability table, through the three-level evaluation network structure, establishes a cost function that reflects the validity of the score corresponding to the measurement indicator data;

[0009] S3, based on the sequential least squares optimization, obtain the minimum value of the cost function and obtain the corresponding conditional probability table;

[0010] S4, obtaining relevant influencing factor data of the candidate wind farm;

[0011] S5, based on the conditional probability table obtained in step S3, the candidate wind farms are evaluated, and the optimal wind farm site selection decision is made according to the evaluation scores.

[0012] In S1, the relevant influencing factor data obtained include four categories of influencing factors, namely, production benefit category, construction cost category, operating cost category and construction risk category, wherein the influencing factors of the production benefit category include average wind speed and the sea area involved in the plan, the influencing factors of the construction cost category include water depth, offshore distance of the wind farm, and equipment transportation distance, the influencing factors of the operating cost category include salt spray concentration, the sum of the grid access distance and 0.5 times the cable length, and the wind direction consistency coefficient, and the influencing factors of the construction risk category include the annual typhoon frequency, the sum of the annual typhoon intensity average value and 0.7 times the annual typhoon intensity maximum value, and the annual average wave height;

[0013] Based on the operation measurement index data of n wind farms, an operation measurement index sequence (a1, a2, ... a i ,…,a n ), where a i represents the i-th wind farm.

[0014] The three-level evaluation network structure is specifically as follows: the first level is the influencing factor layer, which represents the specific influencing factors under each category; the second level is the decision-making standard layer, which represents the categories of influencing factors; and the third level is the target layer, which represents the final comprehensive evaluation.

[0015] The S2 comprises the following steps:

[0016] S21, perform data fuzzification on the influencing factor data of historical reference wind farm cases;

[0017] S22, based on the variable evaluation network structure reasoning conditional probability table, calculate the comprehensive score b of the i-th wind farm i , and establish a reference wind farm scoring sequence (b1, b2, …, b i …,b n ), wherein the evaluation network structure reasoning conditional probability table determines the parameters in the table according to the hierarchical structure of the evaluation network and the evaluation level of each unit predefined in the network structure, and the values ​​of the parameters are variable;

[0018] S23, based on the operational measurement indicator sequence (a1, a2, ..., a n ) and wind farm scoring sequence (b1, b2, …, b n ), and establish a cost function to measure the effectiveness of the rating sequence.

[0019] The cost function is:

[0020]

[0021] in, is the auxiliary calculation sequence; x is the independent variable sequence composed of the variable evaluation network structure inference conditional probability table parameters; n is the number of reference wind farm selection cases.

[0022] In S3, based on the sequential least squares method, the solution to the minimum value of the cost function is transformed into the following problem:

[0023]

[0024] Among them, x is the independent variable sequence composed of the parameters of the variable evaluation network structure inference conditional probability table, h i represents an equality constraint, g j represents inequality constraints; m is the number of equality constraints; p is the number of inequality constraints.

[0025] In S3, the iterative calculation process based on the sequential least squares method includes the following steps:

[0026] S31, construct an approximate quadratic programming problem and solve the problem to obtain a search direction;

[0027] S32, determining the search step size based on the line search method, and updating the current solution in the search direction;

[0028] S33, convergence judgment.

[0029] In S31, based on the principle of sequential least squares method, in the current solution x l At , construct the following approximate quadratic programming problem:

[0030]

[0031] in

[0032]

[0033] is the Lagrangian function, λ,μ are the Lagrangian multipliers of equality and inequality constraints respectively, l is the number of iterations, represents the gradient, represents the Laplace operator, f(x) is the cost function;

[0034] After solving the quadratic programming problem, the corresponding search direction d after the lth iteration is obtained. l .

[0035] In S32, the current solution is updated according to the search step size and the search direction:

[0036] x l+1 =x l +α l d l

[0037] Among them, α lThe search step size obtained based on the line search method is:

[0038]

[0039] α is the step size.

[0040] In S33, when the search direction d l When the norm of the cost function or the change in the cost function is less than the set cutoff threshold, the algorithm terminates and obtains the optimal solution of the current cost function f(x l ) and the optimal solution x of the corresponding conditional probability table l .

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] (1) The present invention comprehensively considers various factors affecting the site selection decision of offshore wind farms. Each wind farm involves a variety of relevant original data, which improves the comprehensiveness and accuracy of the final decision;

[0043] (2) The present invention introduces objective measurement index data as optimization target data and performs optimization based on the sequential least squares method, thereby improving the objectivity, rationality and scientificity of the final decision;

[0044] (3) The present invention avoids reliance on expert opinions and innovatively solves the problem of decision-making errors caused by the uncertainty and subjective factors of expert opinions in evaluation decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 is a flow chart of the method of the present invention;

[0046] Figure 2 is a three-level evaluation network structure diagram in an embodiment of the present invention;

[0047] Figure 3 The following is a flowchart of the sequential least squares method iteration in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is implemented based on the technical solution of the present invention, and provides a detailed implementation method and specific operation process, but the protection scope of the present invention is not limited to the following embodiments.

[0049] This embodiment provides a wind farm site selection decision method based on a custom cost function and sequential least squares optimization, such as Figure 1 As shown, the following steps are included:

[0050] S1, obtain the relevant influencing factor data of historical reference wind farm site selection cases, as well as the operational measurement indicator data after the implementation of the cases, and establish a three-level evaluation network structure for comprehensive analysis based on the relevant influencing factor data.

[0051] In this embodiment, the relevant influencing factor data obtained include four categories of influencing factors, namely production benefit, construction cost, operating cost and construction risk. Among them, the influencing factors of the production benefit category include average wind speed and the sea area involved in the plan. The influencing factors of the construction cost category include water depth, offshore distance of the wind farm, and equipment transportation distance. The influencing factors of the operating cost category include salt spray concentration, the sum of the grid access distance and 0.5 times the cable length, and the wind direction consistency coefficient. The influencing factors of the construction risk category include typhoon annual frequency, the sum of the average annual typhoon intensity and 0.7 times the maximum annual typhoon intensity, and the annual average wave height. Based on the operation measurement indicator data of n wind farms, establish an operation measurement indicator sequence (a1, a2, ... a i ,…,a n ), where a i represents the i-th wind farm. Then, based on the current reference case data, a three-level evaluation network structure is established, such as Figure 2 As shown, the first level is the influencing factor layer, which represents the specific influencing factors under each category; the second level is the decision-making criterion layer, which represents the categories of influencing factors; and the third level is the target layer, which represents the final comprehensive evaluation.

[0052] The operation measurement indicator data selected in this embodiment is the average annual benefit of the wind farm, and its unit is: million US dollars.

[0053] S2, based on the variable evaluation network structure inference conditional probability table, through the three-level evaluation network structure, establishes a cost function that reflects the validity of the scoring corresponding measurement indicator data.

[0054] S2 includes the following steps:

[0055] S21, perform data fuzzification on the influencing factor data of historical reference wind farm cases.

[0056] The specific operation of fuzzy processing of influencing factor data is as follows: calculate the average value of each influencing factor And the standard deviation σ, and the corresponding fuzzy standard is divided into five levels of evaluation, from bad to good: VB (Very Bad), B (Bad), A (Average), G (Good), VG (Very Good), where the original data value corresponding to 100% membership of VB and VG evaluation is The original data value corresponding to 100% membership of B and G evaluation is A evaluates the original data value corresponding to 100% membership Finally, according to the triangle fuzzification principle, the data is fuzzified to obtain the evaluation data of the evaluation type and its membership combination. The corresponding VB or VG depends on whether the current influencing factor value increases the wind farm evaluation to see whether it is a gain, and the evaluation of B and G is similar. Finally, the fuzzy standard shown in Table 1 is obtained in the specific data of the embodiment:

[0057] Table 1 Fuzzy standard table

[0058]

[0059]

[0060] S22, based on the variable evaluation network structure reasoning conditional probability table, calculate the comprehensive score b of the i-th wind farm i , and establish a reference wind farm scoring sequence (b1, b2, …, b i …,b n ), wherein the evaluation network structure inference conditional probability table determines the parameters in the table according to the hierarchical structure of the evaluation network and the evaluation level of each unit predefined in the network structure, and the parameter values ​​are variable.

[0061] S23, based on the operational measurement indicator sequence (a1, a2, ..., a n ) and wind farm scoring sequence (b1, b2, …, b n ), and establish a cost function to measure the effectiveness of the rating sequence.

[0062] For the final score (b1, b2, ..., b n ), if we can make its proportions consistent with the evaluation index sequence (a1, a2, …, a n ), the closer the corresponding ratios are, the higher the effectiveness of the score is. Consider the following cost function, where the sequence (b1k, b2k, …, b1k) is obtained by multiplying each number in the final score by the same proportional coefficient k. n k) can be as close as possible to the evaluation index sequence (a1, a2, ..., a n ) are equal, so the corresponding ratios of the two original sequences will be the most similar, and the corresponding score validity will be the highest:

[0063]

[0064] Obviously, only when a t =b t k, the part i=t may be 0. t ≠b tk will increase the cost function according to the difference in the ratio of the two, so the process of finding the optimal scoring network is transferred to the optimal scoring sequence (b1, b2, ..., b n ) is obtained, and then transferred to the process of minimizing the cost function W(k).

[0065] To facilitate further calculations, the cost function can be transformed into the following form to make it more conducive to mathematical calculations. At the same time, it is obvious that the goal of minimizing W(k) after the following transformation will not affect the determination of the optimal scoring sequence:

[0066]

[0067] For this function, first find its corresponding minimum value and consider taking the derivative of k on both sides of the equation:

[0068]

[0069] make Then, when When , the corresponding minimum value of the cost function W(k) can be obtained as:

[0070]

[0071] Then establish a custom cost function to measure the effectiveness of the scoring sequence:

[0072]

[0073] in, is the auxiliary calculation sequence; x is the independent variable sequence composed of the variable evaluation network structure inference conditional probability table parameters; n is the number of reference wind farm selection cases.

[0074] S3, based on the sequential least squares optimization, obtain the minimum value of the cost function and obtain the corresponding conditional probability table.

[0075] Based on the sequential least squares method, the solution to the minimum value of the cost function is transformed into the following problem:

[0076]

[0077] Among them, x is the independent variable sequence composed of the parameters of the variable evaluation network structure inference conditional probability table, h i represents an equality constraint, g j represents inequality constraints; m is the number of equality constraints; p is the number of inequality constraints.

[0078] Specifically, the factors that have a decisive influence on the accuracy of the model are the setting of the conditional probability table and the weighted distribution of the total score of the parameters in the final four categories for calculating the overall evaluation. Consider defining all the units involved in the setting of the conditional probability table and the weights of the final weighted distribution as function variables. As for the number of function variables, since both the influencing factors and the decision categories are divided into five levels, the first decision category (production efficiency category) involves 5 2 = 25 criteria, the second to fourth decision categories (construction cost, operation cost and construction risk) involve 5 3 = 125 criteria, each criterion involves 5 evaluation results, plus the final weight allocation involves 4 coefficients, so there are 25*5+125*5*3+4=2004 independent variables. Correspondingly, add constraints: for equal constraints, the sum of the 5 evaluation results in each criterion should be 1, and the sum of the 4 coefficients of the final weight allocation is also 1, so there are 25+125*3+1=401 equal constraints; for unequal constraints, all independent variables should be real numbers between 0 and 1, corresponding to a total of 2*2004=4008 unequal constraints. Then, the effectiveness of the final score is optimized under the constrained solution space to finally determine a more effective conditional probability table setting scheme.

[0079] like Figure 3 As shown, the iterative calculation process based on the sequential least squares method includes the following steps:

[0080] S31, construct an approximate quadratic programming problem and solve the problem to obtain a search direction;

[0081] S32, determining the search step size based on the line search method, and updating the current solution in the search direction;

[0082] S33, convergence judgment.

[0083] In S31, based on the principle of sequential least squares, in the current solution x l At , we can construct the following approximate quadratic programming problem:

[0084]

[0085] in

[0086]

[0087] is the Lagrangian function, λ,μ are the Lagrangian multipliers of equality and inequality constraints respectively, l is the number of iterations, represents the gradient, represents the Laplace operator, f(x) is the cost function;

[0088] After solving the quadratic programming problem, the corresponding search direction d after the lth iteration is obtained. l .

[0089] In S32, the current solution is updated according to the search step size and search direction:

[0090] x l+1 =x l +α l d l

[0091] Among them, α l The search step size obtained based on the line search method is:

[0092]

[0093] α is the step size. Note that in the actual process, in order to speed up the calculation of the search step size, the inexact line search method based on the Armijo condition or the Wolfe condition is generally used to optimize the search step size calculation.

[0094] In S33, when the search direction d l When the norm of the cost function or the change in the cost function is less than the set cutoff threshold, the algorithm terminates and obtains the optimal solution of the current cost function f(x l ) and the optimal solution x of the corresponding conditional probability table l .

[0095] S4, obtaining relevant influencing factor data of the candidate wind farm.

[0096] In this regard, actual wind farm data that is not in the reference wind farm case is selected as a substitute, so that after obtaining the score, the feasibility of the method can be verified based on its evaluation index. A total of 5 wind farm data are selected to simulate the candidate wind farm data, as shown in Table 2:

[0097] Table 2 Data of simulated candidate wind farms

[0098]

[0099] S5, based on the conditional probability table obtained in step S3, the candidate wind farms are evaluated, and the optimal wind farm site selection decision is made according to the evaluation scores.

[0100] Based on the existing data on influencing factors of candidate wind farms, data fuzzification is first performed based on Table 1:

[0101] Table 3 Fuzzy table of simulated candidate wind farm data

[0102]

[0103] Through the existing data on influencing factors of candidate wind farms and the optimal conditional probability table parameters obtained by the final optimization, comprehensive evaluation reasoning is carried out in the three-level evaluation network, and the utility values ​​of VB, B, A, G, and VG are given as 0, 0.35, 0.55, 0.85, and 1.00 respectively, and the weighted comprehensive score is calculated, as shown in Table 4:

[0104] Table 4 Final scores obtained based on the method of the present invention

[0105]

[0106] Based on the average method reasoning, the final score is shown in Table 5:

[0107] Table 5 Final score obtained based on average method reasoning

[0108] OWF-1 OWF-2 OWF-3 OWF-4 OWF-5 Overall rating 0.65 0.68 0.35 0.73 0.52 Ranking 3 2 5 1 4

[0109] The actual measurement indicators of the simulated candidate wind farm cases are shown in Table 6:

[0110] Table 6 Actual measurement index data of simulated candidate wind farm cases

[0111]

[0112] By comparing Table 4, Table 5 and Table 6, it is found that the final score obtained by this method is almost exactly the same as the target measurement index ratio, and the corresponding calculated cost function value is 0.12, which can effectively reflect the size of the target measurement index (annual income in this embodiment) based on the score size; while the final score obtained by reasoning based on the average method cannot provide a clear reference for the site selection decision, and the corresponding calculated cost function value is 11.35, which cannot reflect the actual site selection of the wind farm. This reflects the effectiveness and scientificity of the method of the present invention.

[0113] Finally, through the comprehensive scoring results, the candidate wind farm OWF-1 with the highest score was selected as the optimal decision.

[0114] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning, or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.

Claims

1. A wind farm site selection decision method based on a custom cost function and sequential least squares optimization, characterized in that: The following steps are involved: S1, obtain the relevant influencing factor data of historical reference wind farm site selection cases, as well as the operational measurement indicator data after the implementation of the cases, and establish a three-level evaluation network structure for comprehensive analysis based on the relevant influencing factor data; S2, based on the variable evaluation network structure inference conditional probability table, through the three-level evaluation network structure, establishes a cost function that reflects the validity of the score corresponding to the measurement indicator data; S3, based on the sequential least squares optimization, obtain the minimum value of the cost function and obtain the corresponding conditional probability table; S4, obtaining relevant influencing factor data of the candidate wind farm; S5, based on the conditional probability table obtained in step S3, the candidate wind farms are evaluated, and the optimal wind farm site selection decision is made according to the evaluation scores.

2. A wind farm site selection decision method based on a custom cost function and sequential least squares optimization according to claim 1, characterized in that: In S1, the relevant influencing factor data obtained include four categories of influencing factors, namely, production benefit category, construction cost category, operating cost category and construction risk category, wherein the influencing factors of the production benefit category include average wind speed and the sea area involved in the plan, the influencing factors of the construction cost category include water depth, offshore distance of the wind farm, and equipment transportation distance, the influencing factors of the operating cost category include salt spray concentration, the sum of the grid access distance and 0.5 times the cable length, and the wind direction consistency coefficient, and the influencing factors of the construction risk category include the annual typhoon frequency, the sum of the annual typhoon intensity average value and 0.7 times the annual typhoon intensity maximum value, and the annual average wave height; Based on the operation measurement index data of n wind farms, an operation measurement index sequence (a1, a2, ... a i ,…,a n ), where a i represents the i-th wind farm.

3. A method for wind farm site selection decision based on user-defined cost function and sequential least squares optimization according to claim 1, characterized in that: The three-level evaluation network structure is specifically as follows: the first level is the influencing factor layer, which represents the specific influencing factors under each category; the second level is the decision-making standard layer, which represents the categories of influencing factors; and the third level is the target layer, which represents the final comprehensive evaluation.

4. A method for wind farm site selection decision based on user-defined cost function and sequential least squares optimization according to claim 2, characterized in that: The S2 comprises the following steps: S21, perform data fuzzification processing on the influencing factor data of historical reference wind farm cases; S22, based on the variable evaluation network structure reasoning conditional probability table, calculate the comprehensive score b of the i-th wind farm i , and establish a reference wind farm scoring sequence (b1, b2, …, b i …,b n ), wherein the evaluation network structure reasoning conditional probability table determines the parameters in the table according to the hierarchical structure of the evaluation network and the evaluation level of each unit predefined in the network structure, and the values ​​of the parameters are variable; S23, based on the operational measurement indicator sequence (a1, a2, ..., a n ) and wind farm scoring sequence (b1, b2, …, b n ), and establish a cost function to measure the effectiveness of the rating sequence.

5. A method for wind farm site selection decision based on user-defined cost function and sequential least squares optimization according to claim 4, characterized in that: The cost function is: in, is the auxiliary calculation sequence; x is the independent variable sequence composed of the variable evaluation network structure inference conditional probability table parameters; n is the number of reference wind farm selection cases.

6. A method for wind farm site selection decision based on user-defined cost function and sequential least squares optimization according to claim 1, characterized in that: In S3, based on the sequential least squares method, the solution to the minimum value of the cost function is transformed into the following problem: Among them, x is the independent variable sequence composed of the parameters of the variable evaluation network structure inference conditional probability table, h i represents an equality constraint, g j represents inequality constraints; m is the number of equality constraints; p is the number of inequality constraints.

7. A method for wind farm site selection decision based on user-defined cost function and sequential least squares optimization according to claim 1, characterized in that: In S3, the iterative calculation process based on the sequential least squares method includes the following steps: S31, construct an approximate quadratic programming problem and solve the problem to obtain a search direction; S32, determining the search step size based on the line search method, and updating the current solution in the search direction; S33, convergence judgment.

8. A method for wind farm site selection decision based on user-defined cost function and sequential least squares optimization according to claim 7, characterized in that: In S31, based on the principle of sequential least squares method, in the current solution x l At , construct the following approximate quadratic programming problem: in is the Lagrangian function, λ,μ are the Lagrangian multipliers of equality and inequality constraints respectively, l is the number of iterations, represents the gradient, represents the Laplace operator, f(x) is the cost function; After solving the quadratic programming problem, the corresponding search direction d after the lth iteration is obtained. l .

9. A method for wind farm site selection decision based on user-defined cost function and sequential least squares optimization according to claim 8, characterized in that: In S32, the current solution is updated according to the search step size and the search direction: x l+1 =x l +a l d l Among them, α l The search step size obtained based on the line search method is: α is the step size.

10. A method for wind farm site selection decision based on user-defined cost function and sequential least squares optimization according to claim 9, characterized in that: In S33, when the search direction d l When the norm of the cost function or the change in the cost function is less than the set cutoff threshold, the algorithm terminates and obtains the optimal solution of the current cost function f(x l ) and the optimal solution x of the corresponding conditional probability table l .

Images