A wind turbine arrangement optimization method, device and equipment for a wind farm

Abstract

The application relates to the field of wind farm layout optimization, and discloses a wind turbine arrangement optimization method, device and equipment for a wind farm, wherein the method comprises the following steps: acquiring wind farm resource data, arranging the positions of a wind turbine array according to the wind farm resource data to determine an initial arrangement scheme of the wind turbine array; determining a direction arrangement vector of the initial arrangement scheme, iteratively updating the direction arrangement vector based on the wind farm resource data, wherein the direction arrangement vector represents a first direction interval proportion and a second direction normal translation; in any iteration round, the direction arrangement vector of the current iteration round is optimized to generate a plurality of candidate arrangement vectors, the optimal arrangement vector of the current iteration round is determined from the plurality of candidate arrangement vectors, and the optimal arrangement vector of the last iteration round is position decoded to determine a target arrangement scheme of the wind farm. The technical scheme provided by the application can improve the optimization efficiency and optimization performance of the wind turbine arrangement of the wind farm.

Description

Technical Field

[0001] This application relates to the field of wind farm layout optimization, and in particular to a method, apparatus and equipment for optimizing the wind turbine layout of a wind farm. Background Technology

[0002] In offshore wind farm projects, micro-site selection (wind turbine layout) is a core factor determining the feasibility of the project. When wind turbines are running, the incoming airflow creates a wake region after passing the upstream turbines, causing a decrease in downstream wind speed and directly affecting the performance of the wind farm.

[0003] Currently, the layout of wind turbines in offshore wind farms mainly adopts two methods: First, manual layout based on expert experience. This method is affected by subjective human factors, resulting in low optimization efficiency and difficulty in obtaining a turbine layout scheme that maximizes wind farm power generation, leading to poor optimization performance. Second, optimization layout based on heuristic algorithms. Using power generation and cost as objective functions, this method can increase power generation and improve optimization efficiency to some extent. However, the optimization results often exhibit irregular layouts, making it difficult to meet the requirements of engineering feasibility and intensive sea use, resulting in poor optimization performance.

[0004] Therefore, improving the optimization efficiency and performance of wind turbine layout in wind farms has become a pressing technical challenge for the micro-site selection of offshore wind farms. Summary of the Invention

[0005] This application provides a method, apparatus, and equipment for optimizing the layout of wind turbines in a wind farm. By refining the arrangement of the spacing between wind turbine rows and columns, the optimization efficiency and performance of wind turbine layout in a wind farm can be improved.

[0006] This application provides a method for optimizing the layout of wind turbines in a wind farm. The method includes: acquiring wind farm resource data; arranging the wind turbine array according to the wind farm resource data to determine an initial layout scheme for the wind turbine array; determining a directional layout vector for the initial layout scheme; iteratively updating the directional layout vector based on the wind farm resource data, wherein the directional layout vector represents a first directional spacing ratio and a second directional normal translation; in any iteration of the iterative update, acquiring the directional layout vector for the current iteration; optimizing the directional layout vector in the first direction and the second direction to generate multiple candidate layout vectors; determining the optimal layout vector for the current iteration based on a first objective constraint and a second objective constraint among the multiple candidate layout vectors; and in the final iteration of the iterative update, decoding the optimal layout vector for the final iteration to determine a target layout scheme for the wind farm.

[0007] In one embodiment, the wind farm resource data includes at least site boundary data and the main wind direction; arranging the wind turbine array according to the wind farm resource data to determine the initial arrangement scheme of the wind turbine array includes: determining the arrangement direction of the wind turbine array based on the main wind direction, the arrangement direction including a first direction and a second direction; within the site range characterized by the site boundary data, arranging the wind turbine array according to the first direction and the second direction as references; obtaining the position arrangement information of the wind turbine array after the position arrangement; and determining the initial arrangement scheme based on the position arrangement information, wherein the position arrangement information includes at least a first direction spacing ratio and a second direction normal translation.

[0008] In one embodiment, the first directional spacing ratio includes the spacing ratio between adjacent fans in any exhaust fan of the fan array along the first direction, and the second directional normal translation includes the normal translation between each exhaust fan of the fan array along the second direction.

[0009] In one implementation, performing first-direction optimization and second-direction optimization on the directional arrangement vector of the current iteration to generate multiple candidate arrangement vectors includes: obtaining a first-direction spacing ratio and a second-direction normal translation represented by the directional arrangement vector; performing first-direction optimization on the first-direction spacing ratio and second-direction optimization on the second-direction normal translation; and updating the directional arrangement vector multiple times based on the first-direction optimization result and the second-direction optimization result to determine multiple candidate arrangement vectors.

[0010] In one implementation, determining the optimal arrangement vector for the current iteration based on a first objective constraint and a second objective constraint among the plurality of candidate arrangement vectors includes: for any candidate arrangement vector, determining whether the candidate arrangement vector satisfies the first objective constraint; if the determination result indicates that the first objective constraint is satisfied, determining the fitness metric of the candidate arrangement vector based on the second objective constraint and a specified wake model; and determining the optimal arrangement vector among the plurality of candidate arrangement vectors according to the fitness metrics of each candidate arrangement vector.

[0011] In one implementation, determining the fitness metric of the candidate layout vector based on the second objective constraint and the specified wake model includes: determining the penalty parameter of the candidate layout vector based on the second objective constraint, and determining the power generation parameter of the candidate layout vector based on the specified wake model and wind farm resource data; inputting the power generation parameter and the penalty parameter of the candidate layout vector into a pre-constructed fitness evaluation model to output the fitness metric of the candidate layout vector.

[0012] In one implementation, if the determination result does not satisfy the first target constraint, the candidate arrangement vector is eliminated.

[0013] In one embodiment, the first target constraint is that the candidate arrangement parameter in the first direction determined based on the candidate arrangement vector is greater than a preset first spacing, and the second target constraint includes a second spacing constraint and a second direction order constraint.

[0014] A second aspect of this application provides a wind turbine layout optimization device for a wind farm. The device includes: an initial layout unit, configured to acquire wind farm resource data and arrange the wind turbine array according to the wind farm resource data to determine an initial layout scheme for the wind turbine array; a parameter determination unit, configured to determine the directional layout vector of the initial layout scheme and iteratively update the directional layout vector based on the wind farm resource data, wherein the directional layout vector represents a first directional spacing ratio and a second directional normal translation; a parameter optimization unit, configured to acquire the directional layout vector of the current iteration in any iteration of the iterative update, perform first directional optimization and second directional optimization on the directional layout vector to generate multiple candidate layout vectors, and determine the optimal layout vector of the current iteration among the multiple candidate layout vectors; and a scheme determination unit, configured to perform position decoding on the optimal layout vector of the last iteration in the last iteration of the iterative update based on a first target constraint and a second target constraint to determine the target layout scheme of the wind farm.

[0015] A third aspect of this application provides a computer device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the wind turbine layout optimization method for wind farms described in the first aspect.

[0016] The technical solution provided in one or more embodiments of this application significantly improves the power generation performance of wind turbines while enhancing the optimization efficiency and performance of wind turbine layout in wind farms through refined arrangement of the spacing between wind turbines in all directions. Specifically, by transforming the wind turbine layout optimization from a high-dimensional coordinate space to a low-dimensional parameter space, and while maintaining the overall regularity and order of the wind turbine array and meeting engineering constraints, the layout vectors representing row and column spacing are finely adjusted through iterative search, using the coordinates of the four corner wind turbines as constraints. After the iteration meets the termination condition, the final optimal layout vector is decoded and converted back into the actual wind turbine coordinate positions, thereby achieving automatic optimization of wind turbine micro-site selection and obtaining a regular layout scheme with better power generation performance that can be directly applied in engineering. This technical solution can significantly improve the optimization convergence efficiency and solution quality through local fine search while maintaining the overall layout form and engineering feasibility, effectively improving the optimization efficiency and performance of wind turbine layout. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the specific embodiments of this application or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0018] Figure 1 A schematic diagram illustrating the steps of a wind turbine layout optimization method for a wind farm provided in an embodiment of this application; Figure 2 This is a schematic diagram of the arrangement of a wind turbine array according to one embodiment of this application; Figure 3 A schematic diagram illustrating the optimized spacing of a fan arrangement according to one embodiment of this application; Figure 4 A schematic diagram illustrating the radius of the wake influence range provided in one embodiment of this application; Figure 5(a) is a schematic diagram of fan layout with only optimized row spacing provided in an embodiment of this application; Figure 5(b) is a schematic diagram of a fan arrangement with only optimized column spacing provided in an embodiment of this application; Figure 5(c) is a schematic diagram of a row-column joint optimization of the fan layout provided in an embodiment of this application; Figure 6 A schematic diagram of an optimized wind turbine layout for a wind farm, provided as one embodiment of this application; Figure 7 This is a schematic diagram of the structure of a computer device provided in one embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] Furthermore, the use of terms such as "first," "second," etc., in this application is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of embodiments in this application, unless otherwise stated, "multiple" means two or more. Additionally, the use of "based on" or "according to" implies openness and inclusiveness, because processes, steps, calculations, or other actions "based on" or "according to" one or more of the stated conditions or values ​​may in practice be based on additional conditions or beyond the stated values.

[0021] With the rapid development of the offshore wind power industry, the installed capacity and single-unit power of wind farms are constantly increasing, placing higher demands on the micro-site selection of wind farms (i.e., wind turbine layout optimization). The wind turbine layout directly determines the degree of influence of the wake effect, which in turn affects the power generation performance of the wind farm. Due to the low roughness of the marine environment and the slow attenuation of the wake, the wake interference between the front and rear rows of wind turbines is particularly significant, with wake losses reaching more than 20%, affecting the working performance of the wind farm.

[0022] In practice, wind turbine arrays are typically arranged according to expert experience. Designers, based on relevant technical specifications, position the arrays perpendicular to the main wind direction and repeatedly adjust row and column spacing to continuously verify power generation through calculations. This method is inefficient, time-consuming, and labor-intensive, and is limited by subjective factors, resulting in a limited pool of alternatives and difficulty in obtaining an array that approximates optimal power generation, leading to poor optimization performance. To improve optimization efficiency, heuristic optimization methods such as genetic algorithms and particle swarm optimization have been introduced. These methods aim to maximize power generation or minimize cost, performing continuous or discrete optimization of the wind turbine coordinates.

[0023] However, in related technologies, heuristic optimization methods for wind turbine location optimization often pursue the ultimate increase in theoretical power generation, resulting in irregular and disordered layouts. Furthermore, the feasible solutions obtained by this optimization method are limited, potentially leading to the global optimum not being within the solution space. Moreover, this irregular layout violates the principle of intensive use of the sea area for centralized submarine cable transmission channels, adversely affecting navigation safety, operation and maintenance, and subsequent marine development. It also makes it difficult to ensure that the outermost wind turbines are precisely located at the site boundary, easily leading to wasted marginal sites. Consequently, the optimized scheme is difficult to directly apply to actual engineering projects, resulting in poor optimization performance.

[0024] In view of the above, one or more embodiments of this application provide a method, apparatus and equipment for optimizing the layout of wind turbines in a wind farm, which can solve the above problems, significantly improve the power generation performance of wind turbines, improve the optimization efficiency and performance of wind turbine layout in the wind farm, maintain the overall regularity and order of the wind turbine array, meet the requirements of engineering feasibility and intensive use of sea area, and realize the automatic optimization of wind turbine micro-site selection.

[0025] Please see Figure 1 One embodiment of this application provides a method for optimizing the wind turbine layout in a wind farm, which may include the following steps: S1: Obtain wind farm resource data, and arrange the wind turbine array according to the wind farm resource data to determine the initial layout scheme of the wind turbine array.

[0026] The aforementioned wind farm resource data serves as the fundamental input information required for wind farm planning and design. This data may include site boundary data, wind resource data, turbine foundation parameters, and engineering constraints. For example, site boundary data includes site boundary coordinates; wind resource data includes hourly wind speed, hourly wind direction, and Weibull distribution parameters representing the year; turbine foundation parameters include rotor diameter, hub height, dynamic power curve, and thrust coefficient curve; and engineering constraints include minimum row and column spacing constraints (e.g., row spacing not less than 3 times the rotor diameter, column spacing not less than 7 times the rotor diameter) and site boundary constraints.

[0027] Under the premise of meeting the requirements of engineering feasibility and intensive use of marine resources, the wind turbine array is arranged based on expert experience. That is, based on engineering experience and constraints, a baseline arrangement of the wind turbine array is determined as the initial arrangement scheme. For example, during the arrangement process, the number of wind turbines, their starting and ending positions, and their orientation are determined for each direction of the wind turbine array (first direction wind turbine arrangement, second direction wind turbine arrangement). The direction perpendicular to the main wind energy direction is called the first direction, and the direction parallel to the main wind energy direction is called the second direction. Preferably, as many wind turbines as possible are arranged in the first direction wind turbine arrangement to facilitate centralized transmission of submarine cables and to a certain extent improve the power generation performance of the wind power array. Within the site boundary, the turbines are arranged according to engineering constraints using equidistant or quasi-equidistant rules to form an initial arrangement scheme with application feasibility. Furthermore, the initial arrangement scheme can serve as the starting point for subsequent iterative optimization, allowing subsequent optimization to perform local fine-grained searches near existing feasible solutions, significantly improving convergence efficiency and solution quality.

[0028] S3: Determine the directional arrangement vector of the initial arrangement scheme, and iteratively update the directional arrangement vector based on the wind farm resource data, wherein the directional arrangement vector represents the first directional spacing ratio and the second directional normal translation.

[0029] In this embodiment, the positional arrangement information of the wind turbine array is converted into a parameterized vector form. Specifically, the positional arrangement information is determined based on an initial arrangement scheme, including a first-direction spacing ratio and a second-direction normal translation. The first-direction spacing ratio and the second-direction normal translation are encapsulated into a directional arrangement vector representing the initial arrangement scheme. The first-direction spacing ratio corresponds to the arrangement characteristics of the wind turbine array along the first direction, describing the relative positional relationship of adjacent wind turbines in each row (each row of wind turbines in the first direction) using the form of a spacing ratio; that is, calculating the proportion of the spacing between adjacent wind turbines in each row to the total length of the row. The second-direction normal translation corresponds to the arrangement characteristics of the wind turbine array along the second direction, describing the degree of offset of each row of wind turbines (each row of wind turbines in the first direction) relative to a reference position in the second direction using the form of a normal translation; that is, using the first row as a reference, determining the relative position of each row of wind turbines along the second direction.

[0030] In this embodiment, by combining wind farm resource data, the parameters representing the directional arrangement vector are iteratively and dynamically adjusted to gradually search for a better arrangement scheme. Specifically, the first directional spacing ratio and the second directional normal translation represented by the directional arrangement vector are randomly perturbed to determine multiple possible candidate arrangement vectors. Based on the wind farm resource data, the optimal arrangement vector is determined in each iteration. The optimal arrangement vector is used as the optimization basis for the next iteration, i.e., as the directional arrangement vector for the next iteration. For example, guided by maximizing power generation, power generation is calculated using wind resource data and a wake model, and the optimal candidate arrangement vector is searched as the optimal arrangement vector. It should be noted that the initial arrangement scheme determines the fixed turbine coordinates of the four corner turbines of the wind turbine array. The fixed turbine coordinates remain unchanged during iterative optimization to ensure that the final scheme is consistent with the initial scheme in terms of overall layout. Power generation is improved only through local fine-tuning, balancing power generation improvement and engineering feasibility.

[0031] S5: In any iteration of the iterative update, obtain the directional arrangement vector of the current iteration, perform first direction optimization and second direction optimization on the directional arrangement vector to generate multiple candidate arrangement vectors, and determine the optimal arrangement vector of the current iteration based on the first target constraint and the second target constraint among the multiple candidate arrangement vectors.

[0032] Wherein, if the current iteration round is the first round, the directional arrangement vector of the initial arrangement scheme determined in step S3 is used as the directional arrangement vector of the current iteration round; if the current iteration round is not the first round, the optimal arrangement vector determined in the previous iteration round is used as the directional arrangement vector of the current iteration round.

[0033] In this embodiment, the first-direction optimization and the second-direction optimization can be understood as optimizing and adjusting the spacing parameters of the two directions represented by the directional arrangement vector. The spacing optimization in each direction can be achieved through operations such as random perturbation, proportional adjustment, and translational variation. For example, the first-direction spacing ratio is randomly perturbed to adjust the spacing ratio in the first direction, and the second-direction normal translation is randomly perturbed to adjust the normal translation in the second direction. By performing the first-direction optimization and the second-direction optimization on the directional arrangement vector, multiple candidate arrangement vectors can be generated, representing multiple possible combinations of spacing parameters. Each candidate arrangement vector, after decoding, corresponds to a specific wind turbine arrangement scheme.

[0034] In this embodiment, the optimal layout vector for the current iteration is determined based on the fitness evaluation results of each candidate layout vector. Specifically, a fitness evaluation is performed on each candidate layout vector to determine the optimal layout vector for each iteration. Specifically, based on the fixed wind turbine coordinates determined by the initial layout scheme, each candidate layout vector is mapped to the specific wind turbine coordinates of the wind turbine array. The fitness of the candidate layout vector is determined based on the specific wind turbine coordinates, the objective function, and constraint checks. The objective function can be to maximize power generation, and the constraint checks can be to check whether engineering constraints such as row spacing, column spacing, and column order are met. Based on the fitness of each candidate layout vector, the optimal layout vector is determined from multiple candidate layout vectors.

[0035] In this embodiment, when determining the optimal layout vector, it is also necessary to consider whether the candidate layout vectors satisfy the target constraints. Specifically, after evaluating the fitness of each candidate layout vector, it is necessary to determine whether each candidate layout vector satisfies the target constraints. If it does not, the candidate layout vector is eliminated, so that the optimal layout vector for each round is determined based on fitness among the candidate layout vectors that all satisfy the target constraints. Optionally, the determination of whether the target constraints are satisfied can also be performed before the fitness evaluation, so that the optimal layout vector for each round is determined based on fitness among the multiple candidate layout vectors after elimination. For example, the above-mentioned target constraints include a first target constraint and a second target constraint. The first target constraint represents the optimization target constraint in a first direction, and the second target constraint represents the optimization target constraint in a second direction. By applying optimization constraints in the first and second directions, candidate layout vectors with unreasonable spacing or non-compliant layout can be eliminated.

[0036] S7: In the last iteration of the iterative update, the optimal layout vector of the last iteration is position-decoded to determine the target layout scheme of the wind farm.

[0037] In this embodiment, the iteration loop terminates when any iteration meets the termination condition. The termination condition can be that the iteration update reaches a preset iteration number, or that the fitness value corresponding to the candidate layout vector in the current iteration converges to a stable level. In the final iteration, all candidate layout vectors are compared to determine the optimal layout vector, which encapsulates the optimal layout information accumulated throughout the optimization process. Further, the optimal layout vector is position-decoded, i.e., the parameterized directional layout vector is converted inversely to the wind turbine coordinates in the actual physical space. This yields the final wind turbine arrangement result, forming the target layout scheme. The target layout scheme includes at least the two-dimensional coordinates, row and column assignments, and spacing configuration of each wind turbine.

[0038] In one embodiment, the first and second direction optimization processes can be understood as particle update processes in vector space. Each candidate arrangement vector is considered a learning particle, and its update speed and update position are determined based on the direction arrangement vector (reference particle) of the current iteration round. Here, the direction arrangement vector of the current iteration round is either the direction arrangement vector corresponding to the initial arrangement scheme or the optimal arrangement vector of the previous iteration round. Specifically, the update speed of the learning particle represents the step size and orientation of the perturbation, and the update position represents the perturbation position. Each learning particle optimizes towards the update speed and update direction determined by the reference particle. For example, for any learning particle, its update speed can be expressed as: Its update position can be represented as: ,in, Let i be the position of the i-th learning particle in the k-th iteration in the vector space. Let be the velocity of the i-th learning particle in the vector space during the k-th iteration. For the reference particle's position in vector space, This represents the current optimal position of the entire population in the vector space. For inertial weights, For individual learning factors, As a social learning factor, , , is a random number.

[0039] Based on the above ideas, the technical solution provided in this embodiment of the application significantly improves the power generation performance of wind turbines while enhancing the optimization efficiency and performance of wind turbine layout in wind farms through refined arrangement of the spacing between wind turbines in all directions. Specifically, the wind turbine layout optimization is transformed from a high-dimensional coordinate space to a low-dimensional parameter space. While maintaining the overall regularity and order of the wind turbine array and meeting engineering constraints, the coordinates of the four corner wind turbines are fixed as constraints. Iterative search is used to finely adjust the arrangement vectors representing row and column spacing. After the iteration meets the termination condition, the final optimal arrangement vector is decoded and converted back into the actual wind turbine coordinate positions. This achieves automatic optimization of wind turbine micro-site selection, obtaining a regular layout scheme with better power generation performance that can be directly applied in engineering. This technical solution can significantly improve the optimization convergence efficiency and solution quality through local fine-grained search while maintaining the overall layout form and engineering feasibility, effectively improving the optimization efficiency and performance of wind turbine layout.

[0040] In one implementation, the aforementioned wind farm resource data includes at least site boundary data and the prevailing wind direction. Based on the site boundary data and the prevailing wind direction, the wind turbine array is positioned to determine an initial layout scheme. Specifically, the layout direction of the wind turbine array is determined based on the prevailing wind direction, which includes a first direction and a second direction. The first direction is perpendicular to the prevailing wind direction, and the second direction is parallel to the prevailing wind direction. Generally, in the wind farm field, the direction perpendicular to the prevailing wind direction is called the column direction, and the direction parallel to the prevailing wind direction is called the row direction. Within the site area represented by the site boundary data, the wind turbine array is positioned using the first and second directions as references. The positional layout information of the wind turbine array after positioning is obtained. Based on the positional layout information, an initial layout scheme is determined, wherein the positional layout information includes at least the first direction spacing ratio and the second direction normal translation.

[0041] In one embodiment, the aforementioned first directional spacing ratio includes the spacing ratio between adjacent fans in any row of fans in the fan array along the first direction, and the aforementioned second directional normal translation includes the normal translation between each row of fans in the fan array along the second direction. Specifically, the first direction is called the column direction, and the second direction is called the row direction. The horizontal arrangement of the fan array in the first direction is a column, and the horizontal arrangement of the fan array in the second direction is a row. The first directional spacing ratio includes the spacing ratio between adjacent fans in each column, determined based on the row spacing, and the second directional normal translation is the normal translation between each column of fans, characterized by the column spacing. In the initial layout scheme, since no optimization has been performed in the first and second directions, the second directional normal translation of the initial layout scheme should be zero.

[0042] In one embodiment, see Figure 2 Taking a three-column, four-row wind turbine array as an example, the main wind energy direction is vertical. Therefore, the horizontal direction is the column direction, and the vertical direction is the row direction. The directional arrangement vector corresponding to the initial layout scheme can be expressed as: .in, Indicates the proportion of spacing. The normal translation amount is represented, where the spacing ratio for each column exists. , , ,in, Minimum line spacing (engineering constraint). The total length of the column. All spacing proportions constitute the first direction spacing proportion, and all normal translations constitute the second direction normal translation.

[0043] The technical solution provided in this embodiment realizes the standardization and parameterization of the initial wind turbine layout, providing a unified and quantifiable parameter basis for subsequent iterative optimization of wind turbine layout. Specifically, the complex layout information of the wind turbine array is transformed into a low-dimensional parameterized vector form. By utilizing the sum constraint of the spacing ratio (summing to 1), the row length conservation and minimum spacing constraints are automatically satisfied. At the same time, the relative position between columns is represented by the normal translation amount. Thus, while maintaining the overall regularity and order of the wind turbine array and making full use of the site boundaries, a clear and dimensionally simplified optimization basis is provided for subsequent iterative optimization, effectively improving the efficiency and performance of wind turbine layout optimization.

[0044] In one implementation, performing first-direction optimization and second-direction optimization on the directional arrangement vector of the current iteration to generate multiple candidate arrangement vectors includes: obtaining a first-direction spacing ratio and a second-direction normal translation represented by the directional arrangement vector; performing first-direction optimization on the first-direction spacing ratio and second-direction optimization on the second-direction normal translation; and updating the directional arrangement vector multiple times based on the first-direction optimization results and the second-direction optimization results to determine multiple candidate arrangement vectors.

[0045] Specifically, the first direction spacing ratio represents the spacing ratio between adjacent wind turbines within the same column. First direction optimization involves adjusting the first direction spacing ratio to change the distribution density and relative position of wind turbines in the column direction, thereby reducing the wake impact of upstream wind turbines on downstream wind turbines within the same column. The second direction normal translation represents the translation amount between adjacent columns of wind turbines along the normal direction (perpendicular to the main wind energy direction, equivalent to the second direction). Second direction optimization involves adjusting the translation amount of each column along the normal direction to change the spacing width between columns, thereby offsetting the wake superposition areas between different columns.

[0046] In one embodiment, see Figure 3 Taking a three-column, four-row wind turbine array as an example, the spacing ratio in the first direction is... The normal translation in the second direction is The first direction optimization is performed on the determined directional arrangement vector: for each column of wind turbines in the first direction, the spacing ratio is used. This indicates that it satisfies: During optimization, only optimization is needed. and , Automatically generated. Specifically, for exist Random perturbations are generated within a certain range. ,in, Minimum line spacing (engineering constraint). Let be the total length of the column. Further, the directional arrangement vector after the first spacing perturbation is optimized in the second direction: for each column of fans in the first direction, except for the first and last columns, it is translated in the second direction, the translation amount being . This indicates that the positions of the first and last columns of fans remain unchanged to ensure that the coordinates of the fans at the four corners of the array remain constant. Specifically, since the fan array has only three columns in the first direction, only the middle column needs to be shifted, i.e., [the following text is incomplete and requires further context: "to adjust the position of the fans in the first and last columns, ensuring the coordinates of the fans at the four corners of the array remain unchanged."] or exist Random perturbations are generated within the range, where, The value can be 0.5 to 1 times the impeller diameter. It is required that the coordinates of each column strictly increase after translation. If the coordinates of each column in the second direction before translation are expressed as... Then the coordinates of each column after translation follow Furthermore, the spacing between each column in the second direction is greater than the preset column spacing.

[0047] The technical solution provided in this embodiment effectively weakens the wake interference and wake superposition effect of wind turbines by decoupling row spacing optimization (first direction optimization) and column spacing optimization (second direction optimization). Specifically, row spacing mainly affects the wake superposition of wind turbines in the row direction, while column spacing mainly affects the wake interference between columns. By jointly and randomly perturbing the spacing parameters in both directions, a two-way fine-tuning of the wind turbine layout is achieved, thereby enabling a systematic exploration of better spacing combinations, effectively reducing wake losses and improving the power generation performance of wind farms. Furthermore, compared to directly perturbing the coordinates of each wind turbine, a vast solution space is covered with less computational input, reducing the number of invalid evaluations and thus effectively improving the optimization efficiency and performance of wind turbine layout.

[0048] In one implementation, among multiple candidate arrangement vectors, the optimal arrangement vector for the current iteration is determined based on an fitness metric and target constraints. Specifically, for any candidate arrangement vector, it is determined whether the candidate arrangement vector satisfies a first target constraint. If the determination result indicates that the first target constraint is satisfied, the fitness metric of the candidate arrangement vector is determined based on a second target constraint and a specified wake model. Based on the fitness metrics of each candidate arrangement vector, the optimal arrangement vector is determined among the multiple candidate arrangement vectors. Furthermore, if the determination result indicates that the first target constraint is not satisfied, the candidate arrangement vector is eliminated.

[0049] The first objective constraint is a hard engineering constraint that the wind turbine layout must meet, serving as the foundation for ensuring the structural safety and operational space of the turbines. The second objective constraint is a soft optimization objective or auxiliary constraint pursued further, provided that the first objective constraint is met. The aforementioned wake model is used to determine the wake effect between wind turbines. By calculating the wind speed loss caused by upstream turbines to downstream turbines, it evaluates the power generation of the wind turbines represented by the candidate layout vectors. For example, the wake model specified above can employ Jensen's model, Park's model, Frandsen's model, RANS model, etc.

[0050] In this embodiment, the first objective constraint, based on the candidate arrangement vector, determines the candidate arrangement parameters in the first direction (i.e., the spacing between adjacent fans in each column of the first direction, i.e., row spacing), which are greater than a preset first spacing. Specifically, the actual distance between any adjacent fans in the same column must not be less than the preset first spacing (e.g., 3 times the impeller diameter). The second objective constraint includes a second spacing constraint and a second direction sequence constraint. The second spacing constraint is that the candidate arrangement parameters in the second direction, based on the candidate arrangement vector, are greater than the preset second spacing (i.e., the spacing between fans in each column of the first direction in the second direction, i.e., column spacing), which are greater than the preset second spacing. Specifically, the normal projection distance between adjacent columns is not less than the preset second spacing. The second direction sequence constraint indicates that the projection relationship of each row of fans in the fan array in the first direction, determined by the candidate arrangement vector, satisfies the initial order in the second direction. Specifically, the normal projection positions of each column maintain an increasing order to prevent column intersections.

[0051] In this embodiment, determining the fitness metric for candidate arrangement vectors based on the second objective constraint and the specified wake model includes the following steps: S61: Based on the second objective constraint, determine the penalty parameter of the candidate arrangement vector, and based on the specified wake model and wind farm resource data, determine the power generation parameter of the candidate arrangement vector. The above power generation parameter is usually the annual power generation. S63: Input the power generation parameters and penalty parameters of the candidate arrangement vectors into the pre-built fitness evaluation model to output the fitness metric of the candidate arrangement vectors.

[0052] In one embodiment, the fitness evaluation model is simply represented as: ,in, Annual power generation For penalty parameters, This represents the coordinate parameters. If the candidate arrangement vector satisfies all constraints, then... If the fitness value is zero, and a constraint is violated, the fitness value is increased by penalizing the parameter, thereby reducing the competitiveness of the candidate layout vector. Furthermore, among multiple candidate layout vectors, the candidate layout vector with the smallest fitness metric is selected as the optimal layout vector for the current iteration.

[0053] In one embodiment, the fitness evaluation model can be specifically represented as: This can be understood as the existence of a certain line spacing. The distance between fan i and fan j in column r is less than the preset first distance. At that time, the fitness metric equals Add the square of the violation amount to ensure that solutions that violate the constraints are eliminated. This parameter distinguishes between different degrees of violation; the more severe the violation, the larger the penalty increment. When all row spacings satisfy the first objective constraint, the adaptation metric considers annual power generation, the second objective constraint, and additional penalty constraints. Specifically, Annual power generation The penalty term characterizing the second spacing constraint, The penalty term characterizing the second-direction order constraint, The penalty term represents the additional penalty constraint, which is used to handle severe violations of the second objective constraint, such as an overall violation of column spacing greater than 1 times the fan diameter.

[0054] In this embodiment, the penalty term of the additional penalty constraint can be expressed as: .in, , Indicates two adjacent columns, A second spacing is preset. Specifically, the minimum spacing violation of all column spacings is calculated, and the square root of the summation over all column pairs is taken to obtain a measure of the overall violation degree of the second spacing constraint. If this value is greater than the wind turbine diameter D, an additional penalty for the second spacing constraint is triggered, assigning a maximum penalty value. Otherwise, the additional penalty corresponding to the second spacing constraint is zero. Specifically, calculate the order violation amount for all adjacent columns, and take... To ensure a positive violation only occurs when the projected coordinates of the subsequent column are less than those of the preceding column, the square root of the summation of all reverse-order cases is taken to obtain a measure of the overall violation severity of the second-direction sequence constraint. If this value is greater than the wind turbine diameter D, an additional penalty is triggered for the second spacing constraint, assigning a maximum penalty value. Otherwise, the additional penalty corresponding to the second spacing constraint is zero.

[0055] In this embodiment, the penalty term of the first objective constraint can be expressed as: There exists a certain line spacing. The distance between fan i and fan j in column r is less than the preset first distance. At that time, the fitness metric equals Add the square of the violation amount to ensure that solutions that violate the constraints are eliminated. This item can distinguish between different degrees of violation; the more serious the violation, the greater the increase in penalty.

[0056] In this embodiment, the penalty term for the second spacing constraint can be expressed as: This is used to calculate the minimum spacing violation penalty between all column spacings. Indicates the column number. The spacing between any two columns (the first column) Series of fans and the first The spacing between the fans is less than the preset second spacing. At that time, the violation of quantity The sum of the squares of the values ​​multiplied by the weight coefficient 2 forms the penalty value corresponding to the second spacing constraint. The penalty term for the second directional sequence constraint can be expressed as: This can be understood as iterating through all adjacent columns, checking whether the projected coordinates along the uniform normal maintain an increasing order, summing the squares of the violations, and multiplying them by a weighting coefficient of 5 to form the penalty value corresponding to the second direction order constraint.

[0057] The technical solution provided in this embodiment can accurately quantify the wake effect and power generation performance, and effectively screen solutions through a reasonable penalty mechanism, thereby improving the optimization efficiency and reliability of wind turbine layout in wind farms. Specifically, the fitness of each candidate layout vector is determined. The fitness function sets graded penalties for violations of row spacing, column spacing, and order constraints. Severe violations are given a very large penalty value to ensure the elimination of invalid solutions. Minor violations are weighted and accumulated according to the degree of violation. When the constraints are fully satisfied, the penalty is zero. Finally, the candidate layout vector with the smallest fitness metric is selected as the optimal layout vector for the current iteration. Thus, while ensuring that the candidate layout scheme meets the requirements of engineering safety and structural order, the optimization search direction is guided by maximizing power generation, effectively improving the screening accuracy of the optimal layout vector and the convergence stability of the optimization process, and improving the optimization efficiency and performance of wind turbine layout in wind farms.

[0058] In one embodiment, the aforementioned power generation parameters refer to the power generation of the wind turbine array. The power generation is calculated as follows: The wind turbine array coordinates and wind farm resource data are obtained. The wind farm resource data includes the power curve of each turbine (describing the turbine's output power at different wind speeds), inference coefficient curves, hourly wind speed data for at least one full year, and hourly wind direction data. Since wake calculation depends on wind direction and speed, to improve the accuracy of wind energy resource assessment, the wind speed needs to be corrected for Weibull distribution by wind direction sector. Then, the wind direction is discretized into fixed-width sectors (e.g., 4.5°), so that each sector has a sector frequency and Weibull distribution parameters within the sector. Under each sector, the turbine coordinates are rotated to the incoming flow coordinate system and sorted from upstream to downstream. The probability density function of the Weibull distribution is: ,in, For the incoming wind speed, For scale parameters, For shape parameters. Further, calculations are performed sequentially from upstream to downstream: It is determined which upstream wind turbines i affect the current wind turbine j via their wakes. Based on the Jensen wake model and the partial wake coefficient, the wind speed deficit of each upstream wind turbine i on the downstream wind turbine j is calculated. All deficits are then synthesized using a sum-of-squares wake superposition model to obtain the effective wind speed of the downstream wind turbine j. Finally, the effective wind speed is linearly interpolated using the corresponding power curve to obtain the corresponding output power. The power generation of the wind power array is determined based on the output power of each wind turbine.

[0059] In this embodiment, the wake radius of the upstream wind turbine is determined to assess whether it has a wake effect on the downstream wind turbine. Please refer to [link to relevant documentation]. Figure 4 The wake radius is determined as follows: ,in, The impeller radius of the upstream wind turbine. The horizontal distance downstream from the wind turbine The radius of the wake effect at a distance of meters. The wake attenuation coefficient is typically taken as 0.04 to 0.05 at sea. Furthermore, given the incoming wind speed for each wind turbine, determining the wake velocity of the upstream turbine allows us to determine the wind speed deficit of the downstream turbine, and thus, the wind speed of the downstream turbine. ,in, Normal incoming air velocity (m / s) The horizontal distance downstream from the wind turbine Downstream wind speed at a distance of meters. The thrust coefficient (obtained from the thrust coefficient curve based on the wind speed at the upstream turbine). Optionally, the wake attenuation coefficient can also be determined based on... calculate, The height of the wind turbine hub. This represents the ground roughness height.

[0060] In this embodiment, when the downstream wind turbine is affected by the wake, only the influence of the overlapping portion of the wake needs to be considered, and a wake coefficient is introduced. At this point, the effective wind speed of the downstream fan should be determined as follows: ,in, The wind speed (m / s) in front of the upstream wind turbine rotor. The radius of the upstream wind turbine rotor. The wind speed swept in front of the downstream wind turbine rotor, The horizontal distance between upstream wind turbine i and downstream wind turbine j. The wake factor is defined as the ratio of the area where the wake region intersects with the downstream wind turbine to the area of ​​the downstream wind turbine rotor. It is used when the entire wake is involved. Among them, when the downstream units j by n When considering the wake effect of the upstream unit, a sum-of-squares wake superposition model is used to synthesize all losses, and the effective wind speed of the downstream turbine j can be expressed as: .

[0061] In an optional embodiment of this example, the hourly incoming air velocity at each machine position is calculated using a time-series accumulation method. Then, by combining the dynamic power curves of the corresponding wind turbines, the hourly output power is obtained, and the power generation of the wind power array is determined based on the hourly output power of each wind turbine. Specifically, this can be represented as follows: ,in, It is the total number of wind turbines. This represents the effective wind speed of the j-th fan in hour T. This is the number of hours per year (8760 hours). The time step is used. The annual power generation calculated using the time-series accumulation method has high accuracy.

[0062] In another optional implementation of this embodiment, the power generation of the wind power array is determined based on the output power of each sector using the sector-wind speed interval integration method. Specifically, the output power of all wind turbines within the sector is summed, multiplied by the frequency and time coefficient of that sector, and the contributions of all sectors are summed to obtain the power generation of the entire wind power array within the wind farm. This can be represented as follows: ,in, This is the number of hours per year (8760 hours). This represents the total number of wind sector areas. Let represent the wind direction probability distribution of the i-th wind direction sector. The output power of the fan. Let be the wind speed probability distribution of the j-th wind turbine in the i-th wind direction sector.

[0063] This application provides an example of a micro-site selection scenario for a wind farm. The wind farm is a regular rectangle with a site capacity of 500MW, a single turbine capacity of 16MW, and a total of 32 turbines. The prevailing wind direction is confirmed to be north-south by a wind rose diagram. After expert experience in layout, the initial layout of the turbine array generates 2651.905GWh of electricity.

[0064] In this scenario example, when only row spacing is optimized, the wind turbine array changes as shown in Figure 5(a), with a power generation of 2656.074 GWh and an increase in operating hours of 8.142 h. When only column spacing is optimized, the wind turbine array changes as shown in Figure 5(b), with a power generation of 2653.561 GWh and an increase in operating hours of 3.234 h. When both row and column spacing are optimized, the wind turbine array changes as shown in Figure 5(c), with a power generation of 2656.496 GWh and an increase in operating hours of 8.967 h. The increase in operating hours represents the equivalent number of full-load operating hours converted from the increase. The combined optimization allows the wind farm to operate for an equivalent additional 8.967 hours, representing a significant efficiency improvement given the scarcity of offshore wind power resources. Therefore, optimizing either column or row spacing can increase power generation, while combining the optimization of both row and column spacing simultaneously avoids the influence of upstream and downstream wakes and extends the wake recovery zone, resulting in even greater power generation. The larger the offshore wind farm site, the greater the power generation benefit from jointly optimizing row and column spacing. Furthermore, compared to randomly optimizing the coordinates of individual wind turbines, jointly optimizing row and column spacing can improve the power generation performance of the wind farm and significantly enhance the optimization efficiency and performance of wind turbine layout.

[0065] The above description is merely a scenario example provided in the specification and is not intended to limit the present invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0066] Please see Figure 6 This application also provides a wind turbine layout optimization device for a wind farm, the device comprising: The initial layout unit 100 is used to acquire wind farm resource data and arrange the wind turbine array according to the wind farm resource data to determine the initial layout scheme of the wind turbine array. The parameter determination unit 200 is used to determine the directional arrangement vector of the initial arrangement scheme, and iteratively update the directional arrangement vector based on the wind farm resource data, wherein the directional arrangement vector represents the first directional spacing ratio and the second directional normal translation. The parameter optimization unit 300 is used to obtain the directional arrangement vector of the current iteration in any iteration of the iterative update, perform first directional optimization and second directional optimization on the directional arrangement vector to generate multiple candidate arrangement vectors, and determine the optimal arrangement vector of the current iteration based on the first objective constraint and the second objective constraint among the multiple candidate arrangement vectors. The scheme determination unit 400 is used to perform position decoding on the optimal arrangement vector of the last iteration in the last iteration of the iterative update, so as to determine the target arrangement scheme of the wind farm. in, In one embodiment, the initial layout unit 100 is specifically used to acquire wind farm resource data, which includes at least site boundary data and the main wind direction; determine the layout direction of the wind turbine array based on the main wind direction, which includes a first direction and a second direction; within the site range represented by the site boundary data, arrange the wind turbine array in position using the first direction and the second direction as references; acquire the position layout information of the wind turbine array after position layout; and determine the initial layout scheme based on the position layout information, wherein the position layout information includes at least a first direction spacing ratio and a second direction normal translation.

[0067] In one embodiment, the parameter optimization unit 300 is specifically configured to, in any iteration of the iterative update, obtain a first directional spacing ratio and a second directional normal translation represented by the directional arrangement vector, perform first directional optimization on the first directional spacing ratio, and perform second directional optimization on the second directional normal translation, update the directional arrangement vector multiple times based on the first directional optimization result and the second directional optimization result to determine multiple candidate arrangement vectors, and determine the optimal arrangement vector for the current iteration among the multiple candidate arrangement vectors.

[0068] Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0069] In this application embodiment, a wind turbine layout optimization device for a wind farm is presented in the form of a functional unit. Here, a unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that execute one or more software or fixed programs, or other devices that can provide the above functions.

[0070] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application, such as... Figure 7 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system). Figure 7 Take a processor 10 as an example.

[0071] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0072] The memory 20 stores instructions executable by at least one processor 10 to cause the at least one processor 10 to perform the method shown in the above embodiments.

[0073] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0074] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0075] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0076] This application also provides a computer-readable storage medium. The methods described in this application can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the methods shown in the above embodiments are implemented.

[0077] The apparatus or unit described in the above embodiments can be implemented by a computer chip or entity, or by a product having a certain function. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0078] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0079] Those skilled in the art will understand that embodiments of this application can be provided as methods, apparatus, or computer devices. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0080] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus, or computer devices according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0081] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0083] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0084] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0085] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0086] Although embodiments of this application have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of this application, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for optimizing the layout of wind turbines in a wind farm, characterized in that, The method includes: Acquire wind farm resource data, and arrange the wind turbine array according to the wind farm resource data to determine the initial layout scheme of the wind turbine array; The directional arrangement vector of the initial arrangement scheme is determined, and the directional arrangement vector is iteratively updated based on the wind farm resource data, wherein the directional arrangement vector represents the first directional spacing ratio and the second directional normal translation. In any iteration of the iterative update, the directional arrangement vector of the current iteration is obtained, and the directional arrangement vector is optimized in the first direction and the second direction to generate multiple candidate arrangement vectors. Among the multiple candidate arrangement vectors, the optimal arrangement vector of the current iteration is determined based on the first objective constraint and the second objective constraint. In the last iteration of the iterative update, the optimal layout vector of the last iteration is position-decoded to determine the target layout scheme of the wind farm. The first directional spacing ratio includes the spacing ratio between adjacent fans in any exhaust fan of the fan array along the first direction, and the second directional normal translation includes the normal translation of each exhaust fan in the fan array along the second direction. The first direction is perpendicular to the main wind energy direction, and the second direction is parallel to the main wind energy direction.

2. The method according to claim 1, characterized in that, The wind farm resource data includes at least site boundary data and the prevailing wind direction; the wind turbine array is positioned according to the wind farm resource data to determine the initial layout scheme of the wind turbine array, including: The arrangement direction of the wind turbine array is determined based on the main wind energy direction, and the arrangement direction includes a first direction and a second direction; Within the site area characterized by the site boundary data, the wind turbine array is arranged with the first direction and the second direction as references; The position arrangement information of the wind turbine array after the position arrangement is obtained, and the initial arrangement scheme is determined based on the position arrangement information, wherein the position arrangement information includes at least a first direction spacing ratio and a second direction normal translation.

3. The method according to claim 1, characterized in that, Performing first-direction optimization and second-direction optimization on the directional arrangement vectors to generate multiple candidate arrangement vectors includes: Obtain the first directional spacing ratio and the second directional normal translation represented by the directional arrangement vector; The first direction spacing ratio is optimized in the first direction, and the second direction normal translation is optimized in the second direction. Based on the optimization results of the first direction and the optimization results of the second direction, the directional arrangement vector is updated multiple times to determine multiple candidate arrangement vectors.

4. The method according to claim 1, characterized in that, Among the multiple candidate arrangement vectors, the optimal arrangement vector for the current iteration is determined based on the first objective constraint and the second objective constraint, including: For any of the candidate arrangement vectors, determine whether the candidate arrangement vector satisfies the first target constraint; If the judgment result indicates that the first objective constraint is met, the fitness metric of the candidate arrangement vector is determined based on the second objective constraint and the specified wake model. The optimal arrangement vector is determined from among the candidate arrangement vectors based on the fitness metric of each candidate arrangement vector.

5. The method according to claim 4, characterized in that, Based on the second objective constraint and the specified wake model, the fitness metric for determining the candidate arrangement vector includes: Based on the second objective constraint, the penalty parameter of the candidate arrangement vector is determined, and the power generation parameter of the candidate arrangement vector is determined based on the specified wake model and wind farm resource data. The power generation parameters and penalty parameters of the candidate arrangement vectors are input into a pre-built fitness evaluation model to output the fitness metric of the candidate arrangement vectors.

6. The method according to claim 4, characterized in that, If the judgment result does not meet the first objective constraint, the candidate arrangement vector is eliminated.

7. The method according to claim 1 or 4, characterized in that, The first target constraint is that the candidate arrangement parameter in the first direction determined based on the candidate arrangement vector is greater than a preset first spacing, and the second target constraint includes a second spacing constraint and a second direction order constraint.

8. A wind turbine layout optimization device for a wind farm, characterized in that, The device includes: An initial layout unit is used to acquire wind farm resource data and arrange the wind turbine array according to the wind farm resource data to determine the initial layout scheme of the wind turbine array. The parameter determination unit is used to determine the directional arrangement vector of the initial arrangement scheme, and iteratively update the directional arrangement vector based on the wind farm resource data, wherein the directional arrangement vector represents the first directional spacing ratio and the second directional normal translation. The parameter optimization unit is used to obtain the directional arrangement vector of the current iteration in any iteration of the iterative update, perform first directional optimization and second directional optimization on the directional arrangement vector to generate multiple candidate arrangement vectors, and determine the optimal arrangement vector of the current iteration among the multiple candidate arrangement vectors. The scheme determination unit is used to perform position decoding on the optimal arrangement vector of the last iteration round in the iterative update based on the first objective constraint and the second objective constraint, so as to determine the target arrangement scheme of the wind farm. The first directional spacing ratio includes the spacing ratio between adjacent fans in any exhaust fan of the fan array along the first direction, and the second directional normal translation includes the normal translation of each exhaust fan in the fan array along the second direction. The first direction is perpendicular to the main wind energy direction, and the second direction is parallel to the main wind energy direction.

9. A computer device, characterized in that, include: The system includes a memory and a processor, which are interconnected. The memory stores computer instructions, and the processor executes the computer instructions to perform the wind turbine layout optimization method for a wind farm as described in any one of claims 1 to 7.

Images