Offshore wind farm site selection auxiliary decision-making and dynamic control method, device and medium

Abstract

The application discloses a kind of near-shore wind farm site selection auxiliary decision and dynamic control method, equipment and medium, it is related to near-shore wind farm technical field, the method includes: initial population of quantum genetic algorithm is randomly generated, for each quantum chromosome in initial population, based on the electromagnetic interference thermodynamic diagram of candidate area, determine the electromagnetic interference intensity of each selected candidate point in quantum chromosome, calculate the radar interference index corresponding to quantum chromosome, further calculate the fitness value corresponding to quantum chromosome, based on the fitness value corresponding to each quantum chromosome in initial population, update initial population using quantum genetic algorithm, obtain updated population, constantly iterate until reaching iteration termination condition, based on the fitness value corresponding to each quantum chromosome in updated population, determine several candidate site selection schemes, the application can meet the demand of near-shore wind farm efficient, safe, sustainable construction and operation.

Description

Technical Field

[0001] This application relates to the field of nearshore wind farm technology, and in particular to a method, equipment and medium for nearshore wind farm site selection auxiliary decision and dynamic control. Background Technology

[0002] Nearshore wind farms refer to wind power generation systems located close to the coast (e.g., ≤15–20 km from the shore), in shallow waters (e.g., 5–50 m deep), with foundations directly anchored to the seabed. As the scale of nearshore wind farm construction continues to expand, site selection decisions are becoming increasingly critical. However, traditional site selection support methods for nearshore wind farms have revealed many limitations in practical applications, making it difficult to meet the needs of efficient, safe, and sustainable construction and operation of nearshore wind farms.

[0003] Traditional nearshore wind farm site selection auxiliary decision-making methods have revealed the following two limitations in practical applications: electromagnetic interference and multi-objective optimization.

[0004] Regarding electromagnetic interference, traditional site selection methods for nearshore wind farms often overlook dynamic electromagnetic environment interference factors. The electromagnetic environment in nearshore areas is complex and variable, with various artificial and natural sources of electromagnetic radiation. The wind turbines themselves generate electromagnetic fields, which interact with the surrounding electromagnetic environment to produce interference. This interference can not only affect the normal operation of control and communication systems within the nearshore wind farm, reducing power generation, but also adversely affect other important electromagnetic equipment such as nearby radar systems. However, previous site selection methods did not fully consider this dynamically changing electromagnetic environment interference factor, making it difficult to accurately assess and mitigate potential risks.

[0005] In terms of multi-objective optimization, traditional nearshore wind farm site selection auxiliary decision-making methods have significant shortcomings, focusing only on the single dimension of power generation. While power generation is an important consideration in nearshore wind farm construction, other factors need to be comprehensively considered during the actual site selection process. Using power generation as the sole optimization objective may lead to significant shortcomings in other key aspects, such as excessively high construction costs resulting in poor economic viability of nearshore wind farms, thus affecting their overall feasibility and sustainability. Summary of the Invention

[0006] The purpose of this application is to provide a method, equipment, and medium for site selection auxiliary decision-making and dynamic control of nearshore wind farms, which can meet the needs of efficient, safe, and sustainable construction and operation of nearshore wind farms.

[0007] To achieve the above objectives, this application provides the following solution.

[0008] Firstly, this application provides a method for assisting in the site selection decision of nearshore wind farms, the method comprising:

[0009] Based on the candidate locations in the candidate region, an initial population for a quantum genetic algorithm is randomly generated; the initial population includes multiple quantum chromosomes, each quantum chromosome represents a site selection scheme, and the quantum chromosome includes the selection probability of all the candidate locations. When a candidate location is selected for the construction of a wind turbine, the selection probability is 1.

[0010] For each quantum chromosome in the initial population, based on the electromagnetic interference heatmap of the candidate region, the electromagnetic interference intensity of each selected candidate point in the quantum chromosome is determined. Based on the electromagnetic interference intensity of each selected candidate point in the quantum chromosome, the radar interference index corresponding to the quantum chromosome is calculated. Based on the quantum chromosome and the radar interference index corresponding to the quantum chromosome, the fitness value corresponding to the quantum chromosome is calculated. The fitness value is a weighted sum of power generation, radar interference index, construction cost deviation, and maintenance cost.

[0011] Based on the fitness value corresponding to each quantum chromosome in the initial population, the initial population is updated using a quantum genetic algorithm to obtain the updated population;

[0012] Determine whether the iteration termination condition has been met; if yes, determine several candidate site selection schemes based on the fitness value corresponding to each quantum chromosome in the updated population; if no, use the updated population as the initial population for the next iteration and return to the step of "for each quantum chromosome in the initial population, determine the electromagnetic interference intensity of each selected candidate point in the quantum chromosome based on the electromagnetic interference heatmap of the candidate region".

[0013] Secondly, this application provides a dynamic control method for nearshore wind farms, the dynamic control method for nearshore wind farms comprising:

[0014] Several candidate site selection schemes were determined using the above-mentioned nearshore wind farm site selection auxiliary decision-making method, and the final site selection scheme was determined based on the several candidate site selection schemes;

[0015] Obtain the real-time interference intensity of each wind turbine in the nearshore wind farm constructed using the final site selection scheme;

[0016] For each wind turbine, it is determined whether the real-time interference intensity of the wind turbine is greater than a preset threshold. If so, the yaw control quantity of the wind turbine is calculated based on the real-time interference intensity and the preset threshold, and the yaw control quantity of the wind turbine is sent to the wind turbine controller. The wind turbine controller is used to perform yaw control on the wind turbine based on the yaw control quantity of the wind turbine.

[0017] Thirdly, this application provides a computer device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to implement the above-described nearshore wind farm site selection auxiliary decision-making method or the above-described nearshore wind farm dynamic control method.

[0018] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the aforementioned nearshore wind farm site selection auxiliary decision-making method or the aforementioned nearshore wind farm dynamic control method.

[0019] According to the specific embodiments provided in this application, this application has the following technical effects:

[0020] This application provides a method, device, and medium for site selection auxiliary decision-making and dynamic control of nearshore wind farms. It introduces a quantum genetic algorithm to determine several candidate site selection schemes, which improves computational efficiency compared to traditional genetic algorithms. It also incorporates an electromagnetic interference heatmap of the candidate region to calculate the radar interference index and further calculate the fitness value to determine several candidate site selection schemes. This approach considers electromagnetic interference in site selection auxiliary decision-making, avoiding the construction of wind turbines in locations with high electromagnetic interference, thus improving safety. The fitness value is a weighted sum of power generation, radar interference index, construction cost deviation, and maintenance cost, enabling the calculation of fitness values ​​based on multiple objectives to determine several candidate site selection schemes. This multi-objective optimization allows for construction and operation at a lower cost, enabling sustainable construction and operation, thereby meeting the needs of efficient, safe, and sustainable construction and operation of nearshore wind farms. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is an application environment diagram of a nearshore wind farm site selection auxiliary decision-making method provided in Embodiment 1 of this application.

[0023] Figure 2 This is a flowchart illustrating a nearshore wind farm site selection auxiliary decision-making method provided in Embodiment 1 of this application.

[0024] Figure 3 This is a flowchart illustrating a dynamic control method for a nearshore wind farm provided in Embodiment 2 of this application.

[0025] Figure 4This is a schematic diagram of a nearshore wind farm site selection auxiliary decision and dynamic control system provided in Embodiment 3 of this application.

[0026] Figure 5 This is a schematic diagram of the structure of a computer device provided in Embodiment 4 of this application. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and 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.

[0028] Example 1.

[0029] The nearshore wind farm site selection auxiliary decision-making method provided in this application embodiment can be applied to, for example... Figure 1 The application environment shown is as follows. The terminal communicates with the server via a network. The data storage system stores the data that the server needs to process. The data storage system can be set up independently, integrated into the server, or placed in the cloud or on another server. The terminal can send a pending address selection auxiliary decision request to the server. After receiving the pending address selection auxiliary decision request, the server, based on the candidate points in the candidate region, randomly generates an initial population for a quantum genetic algorithm. For each quantum chromosome in the initial population, based on the electromagnetic interference heatmap of the candidate region, the server determines the electromagnetic interference intensity of each selected candidate point in the quantum chromosome. Based on the electromagnetic interference intensity of each selected candidate point in the quantum chromosome, the server calculates the radar interference index corresponding to the quantum chromosome. Based on the quantum chromosome and its corresponding radar interference index, the server calculates the radar interference index. The process involves calculating the interference index and the fitness value of the corresponding quantum chromosome. Based on the fitness value of each quantum chromosome in the initial population, a quantum genetic algorithm is used to update the initial population, resulting in an updated population. The process then determines whether the iteration termination condition has been met. If so, several candidate site selection schemes are determined based on the fitness value of each quantum chromosome in the updated population. If not, the updated population is used as the initial population for the next iteration, and the process returns to the step of "determining the electromagnetic interference intensity of each selected candidate point in the quantum chromosome based on the electromagnetic interference heatmap of the candidate region for each quantum chromosome in the initial population." The server can then feed back the obtained site selection auxiliary decision result—several candidate site selection schemes for the site selection auxiliary decision request—to the terminal.

[0030] In addition, in some embodiments, the nearshore wind farm site selection auxiliary decision method can also be implemented by a server or a terminal. For example, the terminal can directly process the site selection auxiliary decision request to be processed, or the server can obtain the site selection auxiliary decision request to be processed from the data storage system and process it.

[0031] In one exemplary embodiment, such as Figure 2 As shown, a nearshore wind farm site selection auxiliary decision-making method is provided. This method is executed by computer equipment, specifically by a terminal or server alone, or by both a terminal and a server. In this embodiment, the method is applied to... Figure 1 The following steps are used as an example of a server in the example.

[0032] Step S1: Based on the candidate locations in the candidate region, an initial population for the quantum genetic algorithm is randomly generated; the initial population includes multiple quantum chromosomes, each quantum chromosome represents a site selection scheme, and the quantum chromosome includes the selection probability of all the candidate locations. When a candidate location is selected for the construction of a wind turbine, the selection probability is 1.

[0033] Step S2: For each quantum chromosome in the initial population, based on the electromagnetic interference heatmap of the candidate region, determine the electromagnetic interference intensity of each selected candidate point in the quantum chromosome; based on the electromagnetic interference intensity of each selected candidate point in the quantum chromosome, calculate the radar interference index corresponding to the quantum chromosome; based on the quantum chromosome and the radar interference index corresponding to the quantum chromosome, calculate the fitness value corresponding to the quantum chromosome; the fitness value is a weighted sum of power generation, radar interference index, construction cost deviation, and maintenance cost.

[0034] Step S3: Based on the fitness value corresponding to each quantum chromosome in the initial population, the initial population is updated using a quantum genetic algorithm to obtain the updated population.

[0035] Step S4: Determine whether the iteration termination condition has been met; if yes, determine several candidate site selection schemes based on the fitness value corresponding to each quantum chromosome in the updated population; if no, use the updated population as the initial population for the next iteration and return to the step of "for each quantum chromosome in the initial population, determine the electromagnetic interference intensity of each selected candidate point in the quantum chromosome based on the electromagnetic interference heatmap of the candidate region".

[0036] By implementing steps S1 to S4 above, this embodiment introduces a quantum genetic algorithm to determine several candidate site selection schemes. Compared with traditional genetic algorithms, it can improve computational efficiency. It introduces an electromagnetic interference heat map of the candidate region to calculate the radar interference index and further calculate the fitness value to determine several candidate site selection schemes. It can consider electromagnetic interference in site selection auxiliary decision-making, avoid building wind turbines in locations with high electromagnetic interference, and improve safety. The fitness value is introduced as a weighted sum of power generation, radar interference index, construction cost deviation, and maintenance cost. It can comprehensively calculate the fitness value based on multiple objectives and determine several candidate site selection schemes, completing multi-objective optimization. It not only considers power generation but also costs (construction cost and maintenance cost), enabling operation with high power generation and completion of construction and operation at low cost. It can carry out sustainable construction and operation, and can meet the needs of efficient, safe, and sustainable construction and operation of nearshore wind farms.

[0037] Given the limitations of traditional nearshore wind farm site selection auxiliary decision-making methods, there is an urgent need for an intelligent site selection auxiliary decision-making method that integrates electromagnetic environment simulation, machine learning, and multi-objective optimization. This method would overcome the shortcomings of related technologies, provide an intelligent, efficient, and dynamic solution to electromagnetic interference problems, and offer a scientific, accurate, efficient, and comprehensive solution for nearshore wind farm site selection, thereby promoting the healthy development of the nearshore wind farm industry.

[0038] The following is a detailed description of a nearshore wind farm site selection auxiliary decision-making method used in this embodiment, including the following steps.

[0039] (a) Generate the initial population.

[0040] In this embodiment, candidate areas for constructing nearshore wind farms are first determined, and candidate sites for constructing wind turbines within these areas are determined based on experience or randomly. Then, based on the candidate sites within the candidate areas, an initial population for a quantum genetic algorithm is randomly generated. The initial population includes multiple quantum chromosomes, each representing a site selection scheme. Each quantum chromosome contains the selection probability of all candidate sites. When a candidate site is selected for wind turbine construction, the selection probability is 1; when a candidate site is not selected, the selection probability is 0.

[0041] This embodiment introduces a quantum-inspired genetic algorithm. First, quantum chromosome encoding is performed. The quantum chromosome encoding mechanism is as follows: the selected state of each candidate point is represented by a single qubit. The quantum chromosome, composed of all qubits, represents a possible addressing scheme. That is, the quantum chromosome represents an addressing scheme, which includes multiple qubits. The number of qubits is the same as the number of candidate points, and there is a one-to-one correspondence between qubits and candidate points. The selected state of a candidate point is represented by qubits. For example, if the number of candidate points is 30, then the number of qubits is 30. At this time, the length of the quantum chromosome is... =30, and the qubit is represented as: , For the first The probability of selecting the nth candidate point, which represents the probability of selecting the nth candidate point. The probability that a candidate site is "selected for wind turbine construction". For the first The first probability amplitude of each candidate point. For the first The probability that a candidate point is not selected. The representative was not selected. For the first The second probability amplitude of each candidate point. For the first The probability that a candidate point is selected. The representative was selected, and the constraints were: .

[0042] Population initialization specifically includes: setting the population size. =200, meaning a total of 200 quantum chromosomes are generated. Initially, each qubit of each quantum chromosome is set to... ,Right now = =0.5, meaning that all candidate points have the same initial probability of being selected, and each candidate point has a 50% probability of being selected or not selected. The quantum chromosome is generated through observation, as shown in the following formula:

[0043] ;

[0044] in, For the first One quantum chromosome; For the first The first quantum chromosome The probability of selecting each candidate site is as follows: a selection probability of 1 means that a wind turbine will be built at this candidate site, and a selection probability of 0 means that a wind turbine will not be built at this candidate site. These are random numbers that follow a uniform distribution of 0-1.

[0045] In this way, a quantum chromosome can probabilistically represent multiple potential addressing schemes.

[0046] (ii) Fitness value calculation.

[0047] In this embodiment, for each quantum chromosome in the initial population, the electromagnetic interference intensity of each selected candidate point in the quantum chromosome is determined based on the electromagnetic interference heatmap of the candidate region. Based on the electromagnetic interference intensity of each selected candidate point in the quantum chromosome, the radar interference index corresponding to the quantum chromosome is calculated. Based on the quantum chromosome and the radar interference index corresponding to the quantum chromosome, the fitness value corresponding to the quantum chromosome is calculated. The fitness value is a weighted sum of power generation, radar interference index, construction cost deviation, and maintenance cost.

[0048] The process of determining the electromagnetic interference intensity of each selected candidate point in the quantum chromosome based on the electromagnetic interference heatmap of the candidate region includes the following steps.

[0049] (1) Normalize the electromagnetic interference heat map of the candidate region to obtain the normalized electromagnetic interference heat map. The electromagnetic interference heat map includes the predicted electromagnetic interference intensity of each grid in the candidate region. The normalized electromagnetic interference heat map includes the predicted normalized electromagnetic interference intensity of each grid in the candidate region.

[0050] This embodiment first obtains an electromagnetic interference (EMI) heatmap. Specifically, it receives an EMI heatmap from an EMI prediction system. The EMI heatmap can be determined by any existing method. The EMI heatmap represents the EMI intensity distribution of a gridded candidate region (the candidate region is divided into grids so that the candidate region includes multiple grids). The EMI heatmap covers the candidate region, and the latitude and longitude range of the candidate region is: , Minimum longitude, For the maximum longitude, Minimum latitude For maximum latitude, the WGS84 latitude and longitude coordinate system can be used, with a network resolution of [missing information]. ,like: , Longitude interval The data format of the electromagnetic interference heatmap is a two-dimensional matrix, with latitudinal intervals. Each element Indicates position (i.e., the candidate region) Line number Column grid, =1, 2, ..., , The number of rows in the grid. =1, 2, ..., , The electromagnetic interference intensity prediction value (in dBm) is the number of columns in the grid. This electromagnetic interference intensity prediction value has been spatiotemporally fused to reflect the average electromagnetic interference intensity level of all locations within the geographical area covered by the grid in the future time period (e.g., the next hour).

[0051] The electromagnetic interference (EMI) heatmap is normalized to obtain a normalized EMI heatmap, which provides the required EMI intensity for calculating the fitness value in the quantum genetic algorithm. The normalization calculation formula is as follows:

[0052] ;

[0053] in, For the first Line number Normalized electromagnetic interference intensity prediction values ​​for the grid of columns; For the first Line number Predicted electromagnetic interference intensity values ​​for the grid of columns; Two-dimensional matrix The average value of all elements in the electromagnetic interference heatmap, that is, the average value of the predicted electromagnetic interference intensity for each grid in the electromagnetic interference heatmap. Two-dimensional matrix The standard deviation of all elements in the electromagnetic interference heatmap is the standard deviation of the predicted electromagnetic interference intensity for each grid in the heatmap.

[0054] (2) For each selected candidate point in the quantum chromosome, determine the grid in which the selected candidate point is located in the candidate region to obtain the target grid, and calculate the electromagnetic interference intensity of the selected candidate point based on the normalized electromagnetic interference intensity prediction value of the neighboring grid corresponding to the target grid.

[0055] A candidate point is a location point within the candidate region, possessing latitude and longitude coordinates suitable for wind turbine installation. Therefore, a candidate point will be located within a specific grid within the candidate region. The grid containing the selected candidate point is designated as the target grid. Using the target grid as the center, the neighborhood of the target grid is determined, further defining the corresponding neighboring grids. The neighboring grids are all grids within the target grid's neighborhood, excluding the target grid itself. The size of the neighborhood can be customized based on requirements; for example, it could be... Neighborhood.

[0056] After determining the neighboring grids corresponding to the target grid, this embodiment performs neighborhood aggregation and... Scaling is applied to calculate the electromagnetic interference intensity at selected candidate points based on the normalized predicted electromagnetic interference intensity values ​​of the neighboring grids corresponding to the target grid. .

[0057] The formula for calculating the electromagnetic interference intensity of the selected candidate points is:

[0058] ;

[0059] in, The selected candidate points The intensity of electromagnetic interference; for function, , As an input variable, its function is to compress the value to... between; The selected candidate points The number of neighboring grids corresponding to the target grid; The selected candidate points The set of neighboring grids corresponding to the target grid; For the position located at the Line number The predicted electromagnetic interference intensity after normalization of the neighborhood grid of the column.

[0060] The radar interference index is the summation or weighted summation of the electromagnetic interference intensities of all selected candidate sites in the quantum chromosome, representing the total interference level of the entire site selection scheme. Specifically, the radar interference index is calculated based on the electromagnetic interference intensity of each selected candidate site in the quantum chromosome. This calculation includes either summing the electromagnetic interference intensities of each selected candidate site to obtain the radar interference index, or performing a weighted summation of the electromagnetic interference intensities of each selected candidate site to obtain the radar interference index.

[0061] Therefore, after obtaining the electromagnetic interference intensity of each selected candidate point in the quantum chromosome, the radar interference index can be calculated by summation. Specifically, the electromagnetic interference intensity of all selected candidate points in the quantum chromosome is summed to obtain the radar interference index. , The radar interference index can be calculated using a weighted summation method, which involves summing the electromagnetic interference intensities of all selected candidate sites in the quantum chromosome. , For the first The weight of the selected candidate point, where the weight can be the weight of the first selected candidate point. The power of the wind turbines installed at each of the selected candidate sites is used to calculate the radar interference index, which represents the total interference risk of the entire nearshore wind farm layout.

[0062] To determine the quality of quantum chromosomes, this embodiment introduces a quantum genetic algorithm for multi-objective optimization decision-making. A multi-objective function is designed to calculate the fitness value, and the formula for calculating the fitness value is as follows:

[0063] ;

[0064] in, This is the fitness value; It is the first weight; The power generation is calculated based on the historical wind speed time series of each selected candidate point in the quantum chromosome; As the second weight; Radar interference index; It is the third weight; Construction cost deviation The reciprocal of the construction cost deviation is the deviation between the construction cost and the budget of the site selection scheme represented by the quantum chromosome. It is the fourth weight; To maintain costs The reciprocal of the value, the maintenance cost is calculated based on the position of each selected candidate point in the quantum chromosome.

[0065] The first, second, third, and fourth weights are dynamically adjusted by a fuzzy controller. When the iteration termination condition is met, the first, second, third, and fourth weights are recorded as the final dynamic weight coefficients determined by the optimization.

[0066] The higher the fitness value, the better the overall performance. The optimization objectives are to maximize power generation, minimize radar interference (i.e., minimize radar interference index) and minimize cost (i.e. minimize construction cost deviation and maintenance cost). Ultimately, the goal is to maximize power generation, minimize radar interference and minimize cost.

[0067] The power generation, construction cost deviation, and maintenance cost are calculated based on the site selection scheme (i.e., the location of the wind turbine) represented by the quantum chromosome, by calling external models of the engineering project (such as wind resource model, wind resource database, cost model (construction cost model and maintenance cost model), and cost database).

[0068] The power generation calculation is achieved by calling the external model of the engineering project to obtain the historical wind speed time series of the candidate area. Based on the location of all wind turbines in the site selection scheme represented by the quantum chromosome (the location of the selected candidate point is the location of the wind turbine), the average value of the historical wind speed time series of all wind turbine locations is calculated to form an equivalent wind speed time series representing the wind speed of the entire nearshore wind farm. The ratio of the standard deviation to the mean of this equivalent wind speed time series is calculated to obtain the wind speed volatility. Power generation can then be further calculated based on the wind speed volatility.

[0069] Construction cost deviation is the difference between the construction cost of the site selection scheme represented by the quantum chromosome and the budget. It is calculated by calling the construction cost model. The construction cost includes: foundation costs: wind turbine equipment and tower itself; location-related costs: water depth (affecting the difficulty and cost of foundation construction), distance from the shore (affecting cable length and laying cost), geological conditions (affecting foundation type), waterway avoidance costs, etc. Then, the deviation between the construction cost and the budget is calculated to obtain the construction cost deviation, which directly reflects the economic feasibility of the site selection scheme. The larger the construction cost deviation, the more the construction cost of the site selection scheme exceeds the budget.

[0070] Maintenance cost is the estimated operation and maintenance cost of a nearshore wind farm throughout its entire life cycle. The average annual maintenance cost is calculated using a maintenance cost model. The average annual maintenance cost is the maintenance cost. The maintenance cost mainly includes: accessibility cost: distance from shore and weather window (which determines the number of days a ship can sail for maintenance). The farther the distance and the worse the weather, the higher the transportation and time costs for a single maintenance trip; economies of scale: whether the wind turbine layout is concentrated affects the number of wind turbines that can be maintained in a single voyage.

[0071] (iii) Population renewal.

[0072] In this embodiment, the initial population is updated using a quantum genetic algorithm based on the fitness value of each quantum chromosome in the initial population, resulting in an updated population.

[0073] The process involves updating the initial population using a quantum genetic algorithm based on the fitness value of each quantum chromosome in the initial population, resulting in an updated population. This process includes the following steps.

[0074] (1) Based on the fitness value of each quantum chromosome in the initial population, Pareto front screening is used to determine each quantum chromosome in the initial population as either a dominated solution or a non-dominated solution.

[0075] The fitness value of each quantum chromosome in the initial population is calculated using the fitness value calculation formula. The Pareto front screening mechanism is introduced to distinguish between non-dominated solutions (retained as elites) and dominated solutions (which need optimization) through Pareto front screening, and each quantum chromosome in the initial population is determined to be either a dominated solution or a non-dominated solution.

[0076] (2) For each quantum chromosome that is determined to be the dominant solution, the quantum chromosome that is determined to be the dominant solution is rotated using a quantum rotation gate to obtain the rotated quantum chromosome, and the rotated quantum chromosome is mutated using a quantum NOT gate to obtain the updated quantum chromosome.

[0077] This embodiment uses a quantum rotation gate to adaptively adjust the quantum state of the quantum chromosome determined as the dominant solution. The core is the adaptive adjustment of the rotation angle, where the rotation angle is adaptively determined based on the fitness value. It can be +0.05π / -0.03π / ±0.01π. Specifically, the rotation angle is selected based on the comparison results of the fitness values ​​(the fitness value corresponding to the quantum chromosome determined as the dominant solution is compared with the optimal fitness value and the average fitness value, respectively. The optimal fitness value is the maximum fitness value corresponding to each quantum chromosome in the initial population, and the average fitness value is the average fitness value corresponding to each quantum chromosome in the initial population, reflecting the overall convergence of the population. The optimal and average fitness values ​​provide a relative and dynamic reference for adjusting the rotation angle). (+0.05π / -0.03π / ±0.01π), if the fitness value corresponding to the quantum chromosome that is determined to be the dominant solution is equal to the optimal fitness value, then the rotation angle is determined to be +0.05. If the fitness value of the quantum chromosome identified as the dominant solution is less than the average fitness value, then the rotation angle is determined to be -0.03. Otherwise, set the rotation angle to +0.01. or -0.01 .

[0078] The adaptive adjustment rule for the rotation angle is a mechanism based on performance-based probability allocation. The advantages of this setting are: (1) reward mechanism ( (1) If a candidate point appears in an excellent (fitness value equals the optimal fitness value) location scheme, then the probability of this candidate point being selected in the next generation is increased; (2) Penalty mechanism ( If a candidate site appears in a poor (fitness value less than the average fitness value) site selection scheme, then the probability of this candidate site being selected in the next generation is reduced; (3) Exploration mechanism ( If a candidate location appears in a site selection scheme with moderate performance (i.e., fitness value greater than or equal to average fitness value, but less than optimal fitness value), then random fine-tuning is performed. This helps maintain population diversity and avoids prematurely falling into local optima.

[0079] The adaptive adjustment rules for the rotation angle are shown in Table 1 below.

[0080] Table 1 Adaptive Adjustment Rules for Rotation Angle

[0081]

[0082] In Table 1, To determine the fitness value corresponding to the quantum chromosome that is the dominant solution; This represents the optimal fitness value. This represents the average fitness value.

[0083] After determining the rotation angle, the quantum rotation gate matrix of the quantum rotation gate is then determined based on the rotation angle. Using quantum rotation gate matrices to update quantum states , For the first The probability amplitude of the first update for each candidate point For the first The second updated probability amplitude of each candidate point can be used to update each qubit in the quantum chromosome based on the quantum rotation gate matrix. The quantum chromosome determined as the dominant solution can be rotated using a quantum rotation gate with a rotation angle to obtain the rotated quantum chromosome.

[0084] Introducing quantum NOT gate mutation (mutation probability) The quantum NOT gate matrix is ​​transformed (i.e., mutated) through a quantum NOT gate (X gate). Using quantum NOT gate matrices to update quantum states In order to maintain population diversity.

[0085] The triggering of mutation in a quantum NOT gate is a completely randomized process. Its core purpose is to maintain population diversity and escape local optima, introducing randomness. Generally, the mutation probability in genetic algorithms is between 0.01 and 0.1. Choosing 0.08 represents a trade-off between maintaining stability and introducing diversity. The mutation operation is probabilistic. Each qubit in the quantum chromosome is acted upon independently to effectively maintain population diversity and avoid premature convergence. For each qubit in the quantum chromosome, a random number uniformly distributed in the range [0, 1) is generated independently. If the result is positive, then a quantum NOT gate is applied to the qubit to perform a mutation operation; otherwise, the result remains unchanged.

[0086] In this embodiment, the quantum chromosome determined to be the dominant solution is rotated using a quantum rotation gate to obtain a rotated quantum chromosome. Then, a quantum NOT gate is used to mutate the rotated quantum chromosome to obtain an updated quantum chromosome. Specifically, this involves: comparing the fitness value corresponding to the quantum chromosome determined to be the dominant solution with the optimal fitness value and the average fitness value, obtaining the comparison result; determining the rotation angle based on the comparison result; and rotating the quantum chromosome determined to be the dominant solution using a quantum rotation gate with the rotation angle to obtain a rotated quantum chromosome. For each qubit in the rotated quantum chromosome, it is determined whether the randomly generated random number is less than the mutation probability. If so, the qubit is mutated using a quantum NOT gate to obtain an updated qubit. If not, the qubit is used as the updated qubit, and all the updated qubits are combined to form an updated quantum chromosome. The rotated quantum chromosome is then mutated using a quantum NOT gate to obtain an updated quantum chromosome.

[0087] Specifically, the fitness value corresponding to the quantum chromosome identified as the dominant solution is compared with both the optimal fitness value and the average fitness value. The rotation angle is determined based on the comparison results, specifically: if the fitness value corresponding to the quantum chromosome identified as the dominant solution is equal to the optimal fitness value, then the rotation angle is determined to be +0.05. If the fitness value of the quantum chromosome identified as the dominant solution is less than the average fitness value, then the rotation angle is determined to be -0.03. Otherwise, set the rotation angle to +0.01. or -0.01 .

[0088] (3) All quantum chromosomes that are determined to be non-dominated solutions and all updated quantum chromosomes are combined into an updated population, and the initial population is updated using a quantum genetic algorithm to obtain the updated population.

[0089] (iv) Generation of candidate site selection schemes.

[0090] In this embodiment, it determines whether the iteration termination condition has been met. If so, it determines several candidate site selection schemes based on the fitness value corresponding to each quantum chromosome in the updated population. If not, it uses the updated population as the initial population for the next iteration and returns to the step of "for each quantum chromosome in the initial population, determining the electromagnetic interference intensity of each selected candidate point in the quantum chromosome based on the electromagnetic interference heatmap of the candidate region".

[0091] The iteration termination condition can be reaching the maximum number of iterations. After iterating to the point where the termination condition is met, based on the fitness value corresponding to each quantum chromosome in the updated population, the Pareto optimal solution is output. The Pareto optimal solution is a non-dominated solution, and the location scheme represented by a non-dominated solution is considered as a candidate location scheme, thus determining several candidate location schemes. Through Pareto front screening and preservation of the elite solution set, a balance is achieved between multiple objectives such as power generation, radar interference, and cost, realizing multi-objective optimization.

[0092] This application also provides an application scenario in which the aforementioned nearshore wind farm site selection auxiliary decision-making method is applied. Specifically, the nearshore wind farm site selection auxiliary decision-making method provided in this embodiment can be applied to nearshore wind farm construction scenarios. The nearshore wind farm construction scenario includes an auxiliary decision-making stage, a decision-making stage, and a construction stage. The auxiliary decision-making stage is used to determine several candidate site selection schemes. The decision-making stage is used to determine the final site selection scheme based on these candidate schemes, such as selecting the most suitable candidate site selection scheme from several candidate schemes as the final site selection scheme based on requirements. The construction stage is used to construct the nearshore wind farm based on the final site selection scheme. The nearshore wind farm site selection auxiliary decision-making method provided in this embodiment belongs to the auxiliary decision-making stage.

[0093] Example 2.

[0094] Considering the lack of a real-time electromagnetic environment monitoring and feedback mechanism during the operation of nearshore wind farms after site selection and construction, and the fact that the electromagnetic environment is dynamic and affected by various factors such as solar activity, weather changes, and the usage of surrounding electromagnetic equipment, the electromagnetic environment may change at any time during the operation of nearshore wind farms. Traditional methods cannot obtain this information in real time and feed it back to the operation of nearshore wind farms, making it difficult to adjust and optimize the operation strategy of nearshore wind farms in a timely manner to adapt to the new electromagnetic environment, thus failing to ensure the long-term stable and efficient operation of nearshore wind farms.

[0095] To address this issue, this embodiment provides a dynamic control method for nearshore wind farms, such as... Figure 3 As shown, the nearshore wind farm dynamic control method includes the following steps.

[0096] Step T1: Use the nearshore wind farm site selection auxiliary decision-making method described in Example 1 to determine several candidate site selection schemes, and determine the final site selection scheme based on the several candidate site selection schemes.

[0097] Step T2: Obtain the real-time interference intensity of each wind turbine in the nearshore wind farm constructed using the final site selection scheme.

[0098] Step T3: For each wind turbine, determine whether the real-time interference intensity of the wind turbine is greater than a preset threshold. If so, calculate the yaw control quantity of the wind turbine based on the real-time interference intensity and the preset threshold, and send the yaw control quantity of the wind turbine to the wind turbine controller. The wind turbine controller is used to perform yaw control on the wind turbine based on the yaw control quantity of the wind turbine.

[0099] This embodiment implements dynamic edge computing control: deploying a dynamic threshold (i.e., a preset threshold) and a wind turbine yaw control linkage mechanism. A dynamic threshold is set, and when the real-time interference intensity is greater than the dynamic threshold, the yaw control quantity of the wind turbine is calculated through 5G-MEC (5G Multi-access Edge Computing), triggering the wind turbine yaw control command. The wind turbine controller performs yaw control on the wind turbine based on the wind turbine yaw control command, causing the wind turbine to perform yaw action.

[0100] The dynamic control mechanism for edge computing combined with 5G-MEC includes the following steps.

[0101] (1) Set dynamic threshold ,in, The historical average disturbance value (dBm) is given. As a dynamic adjustment factor, The standard deviation of historical interference is denoted as dBm.

[0102] The formula for calculating the dynamic adjustment factor is:

[0103] ;

[0104] It should be noted that the dynamic adjustment factor is determined based on the long-term statistical characteristics of historical interference data (i.e., historical values ​​of interference intensity), and its update frequency is lower than that of the dynamic threshold. The update frequency, for example, the dynamic adjustment factor is based on the data within 24 hours (or other time intervals). The calculations are performed and updated every 24 hours. This update strategy ensures both a rapid response to transient disturbances and avoids instability in the control strategy caused by environmental noise.

[0105] In this embodiment, the dynamic threshold is updated in real time, with an update cycle of 1 minute.

[0106] (2) Linked with wind turbine yaw control, when the real-time interference intensity is greater than the dynamic threshold, the yaw control quantity of the wind turbine is calculated through 5G-MEC, and the wind turbine yaw control command is triggered. The wind turbine controller performs yaw control on the wind turbine based on the wind turbine yaw control command, so that the wind turbine performs yaw action.

[0107] The formula for calculating the yaw control value of the wind turbine is:

[0108] ;

[0109] in, This is the yaw control value for the wind turbine; The real-time interference intensity of the wind turbine is obtained by the spectrum monitoring sensor deployed on the wind turbine in real time and uploaded to the MEC platform through the 5G network, and its unit is dBm; This is a preset threshold.

[0110] Yaw control quantity of wind turbine The calculation adopts a proportional control strategy and is equipped with a 15° limit protection to ensure the stability of control and equipment safety. The reason for setting the limit protection is: (1) to avoid over-adjustment: to prevent the wind turbine from violent and unstable operation due to a single abnormal interference; (2) to ensure equipment safety: the wind turbine's yaw system has a maximum speed and angle limit to avoid mechanical impact and wear; (3) to ensure power generation stability: excessive sudden yaw may cause violent fluctuations in power output and affect the stability of the power grid.

[0111] Yaw control needs to satisfy yaw angle constraints, which are: , The yaw angle of the wind turbine after being controlled by the yaw control quantity of the wind turbine. This represents the maximum yaw angle of the wind turbine.

[0112] During the planning phase, the candidate site selection scheme with the least overall interference and the highest comprehensive benefits is selected from the perspective of the whole, based on the predicted electromagnetic interference heat map and quantum genetic algorithm. This is a "static" and one-time optimal decision-making behavior. During the operation phase, edge computing dynamic control is based on real-time measured interference intensity and local fine-tuning (yaw) of individual wind turbines to cope with short-term and sudden interference events (such as temporary changes in ship course or sudden high-intensity radar signals) that are not covered by static planning. This is a "dynamic" and continuous adaptive behavior.

[0113] This embodiment uses a blockchain evidence storage system (the evidence storage adopts a Merkle tree structure) to record decision parameters (through Hyperledger Fabric smart contracts). Specifically, the blockchain evidence storage system uses a Merkle tree structure to store key decision data, including: (1) Optimized input records: including the hash value of the electromagnetic interference heatmap used (hash). (1) and timestamp; (2) Optimal solution record: including candidate site selection schemes and their corresponding fitness values, final site selection schemes and their corresponding fitness values; (3) Weight record: including the dynamic weight coefficients finally determined by optimization (i.e., the first weight, the second weight, the third weight and the fourth weight); (4) Control command record: including the real-time interference intensity of the wind turbine, the dynamic threshold and the yaw control amount of the wind turbine. The above key decision data are stored in the leaf data blocks. The root hash value of the Merkle tree is recursively calculated from the hash values ​​of each leaf data block. SHA3-256 is a cryptographic hash function. For the first The root hash value is the hash value of each leaf data block, with || representing the concatenation operation. This root hash value, along with the timestamp, is written to the blockchain to ensure the integrity and immutability of all critical decision data, ensuring that every output decision is traceable, and that any tampering with critical decision data will cause the Merkle root hash verification to fail, providing a complete and trustworthy data chain for regulatory audits.

[0114] At this point, the nearshore wind farm dynamic control method of this embodiment further includes: using a blockchain evidence storage system to record site selection auxiliary decision data and dynamic control data. The site selection auxiliary decision data includes electromagnetic interference heat maps of candidate areas and several candidate site selection schemes. The dynamic control data includes the real-time interference intensity of the wind turbine, preset thresholds, and yaw control quantities of the wind turbine.

[0115] This embodiment belongs to the field of electromagnetic compatibility and auxiliary intelligent control optimization decision-making technology of wind power systems. Specifically, it relates to a nearshore wind farm site selection auxiliary decision-making and dynamic control method that integrates machine learning and multi-objective optimization. It is a wind turbine site selection auxiliary decision-making and dynamic control method based on quantum genetic algorithm, which is particularly suitable for solving the problem of how to effectively resist electromagnetic interference in the nearshore and breakwater wind power construction planning and site selection stage and during operation in areas with dense ship radar, so as to achieve refined management and efficient operation of wind turbines in terms of electromagnetic interference resistance.

[0116] This embodiment takes a wind farm site as an example. The geographical location of the wind farm site is: 120.5°E~121.2°E, 32.3°N~32.8°N (15km from the coastline); the ship traffic is: an average of 285 ships per day (including 37 large radar ships); the wind farm is planned to have an installed capacity of 300MW, with 30 10MW wind turbines to be installed; the interference sources are: ship navigation radar (S / X band) and shore-based radar stations (2 locations).

[0117] (1) Receive electromagnetic interference heat map from electromagnetic interference prediction system.

[0118] (2) Perform quantum chromosome encoding and optimization to obtain several candidate site selection schemes. The first weight is initially set to 0.6, the second weight is initially set to 0.3, the third weight is initially set to 0.08, the fourth weight is initially set to 0.02, the population size is 200, the maximum number of iterations is 150, the number of non-dominated solutions in the Pareto solution set is 37, and the crowding distance threshold is 0.7 times the average crowding distance.

[0119] (3) Dynamic control of edge computing: 1) Threshold setting: =-62dBm, =3.2dBm, Threshold=-52.4dBm; 2) Dynamic adjustment factor =3 (because) =0.052<0.2); 3) The blockchain evidence storage block size is 2MB (containing approximately 1200 transactions), and the evidence storage content includes electromagnetic interference heat map, candidate site selection scheme, wind turbine yaw control quantity, etc.

[0120] Compared to traditional methods, the site selection method in this embodiment can exclude wind turbines located in high interference areas. This is because the radar interference index is used as the core optimization target, which can actively avoid high interference areas and greatly expand the safe operating range. With the optimization of ship traffic flow, sufficient safety margin is provided for the operation and future expansion of nearshore wind farms.

[0121] This embodiment has the following advantages: (1) Improved interference distribution: The maximum interference value is reduced from -45dBm to -58dBm, a reduction of 62%; (2) Expanded safe area; (3) Optimized fan layout: After multiple iterations of optimization, the fans located in the interference area are removed; (4) Provides an intuitive electromagnetic compatibility assessment tool.

[0122] The improvement in interference distribution is mainly due to multi-objective optimization and dynamic control via edge computing. The fitness value calculation formula directly uses the radar interference index as the optimization target. The quantum genetic algorithm, through an adaptive adjustment mechanism of the rotation angle, continuously reduces the selection probability of candidate sites in high-interference areas. This allows candidate site selection schemes to actively exclude candidate sites located in the original hotspot area (-45dBm), thereby avoiding the construction of wind turbines in high-interference areas at the planning level and fundamentally reducing the maximum interference value. Simultaneously, for interference exceeding the standard due to unexpected situations during operation (such as the temporary approach of ships), yaw control is used to temporarily deviate the wind turbine from its maximum power point, instantaneously reducing its radar cross section (RCS), thus suppressing the real-time interference peak. This provides dynamic protection at the operational level, ensuring that the interference level is not excessive at any time.

[0123] The expansion of the safe zone is mainly due to the effects of neighborhood aggregation and sigmoid scaling in multi-objective optimization and data preprocessing. In multi-objective optimization, the Pareto optimal solution set pursued by the quantum genetic algorithm is essentially finding a well-distributed non-dominated solution set. This means that the quantum genetic algorithm not only seeks an optimal solution, but also a set of solutions that can well balance multiple objectives (power generation, radar interference, cost). The ultimately adopted candidate site selection schemes lead to a more dispersed spatial layout of wind turbines, actively avoiding large areas of low-interference safe zones, thereby expanding the overall safe zone. Simultaneously, in the preprocessing stage... Neighborhood aggregation and sigmoid scaling provide a smoother and more accurate assessment of the interference risk of each candidate point, avoiding assessment errors caused by grid partitioning. This enables the quantum genetic algorithm to more accurately identify the true safe areas and high-risk areas, thereby making more reasonable layout decisions and effectively utilizing the safe space.

[0124] Compared to traditional methods, this embodiment achieves the following beneficial effects: Significantly improved electromagnetic compatibility: Due to the adoption of multi-objective optimization, using the radar interference index as the core optimization objective, it can actively avoid high-interference areas, reducing the maximum interference intensity within the nearshore wind farm planning area from -45dBm to -58dBm, a reduction of 62%. Significantly expanded safe operating range: A Pareto solution set with better interference performance was generated, and the interference risk was accurately assessed by combining neighborhood aggregation and sigmoid scaling in data preprocessing. This resulted in an increased proportion of low-interference safe areas in the final wind turbine layout scheme, providing sufficient safety margin for the operation and future expansion of nearshore wind farms. Simultaneously, edge computing dynamic control provides a runtime guarantee mechanism for the above optimization effects, enabling real-time yaw control to respond to sudden interference events and ensure that the interference level remains under control.

[0125] This application also provides an application scenario in which the aforementioned nearshore wind farm dynamic control method is applied. Specifically, the nearshore wind farm dynamic control method provided in this embodiment can be applied in a dynamic control scenario. The dynamic control scenario includes a generation stage and a control stage. The generation stage is used to determine the yaw control amount of the wind turbine, and the control stage is used to control the wind turbine based on the yaw control amount. The nearshore wind farm dynamic control method provided in this embodiment belongs to the generation stage.

[0126] Example 3.

[0127] This embodiment provides a nearshore wind farm site selection auxiliary decision-making and dynamic control system, such as Figure 4 As shown, it includes the following modules.

[0128] Data receiving module: Used to receive electromagnetic interference heatmaps from the electromagnetic interference prediction system. The electromagnetic interference heatmap is a gridded distribution of electromagnetic interference intensity in candidate regions.

[0129] The data preprocessing module includes: a data normalization submodule, which normalizes the received electromagnetic interference heatmap; and an interference vector extraction submodule, which extracts the electromagnetic interference intensity of candidate points.

[0130] Encoding module: Uses qubit encoding to generate quantum chromosomes, which represent addressing schemes.

[0131] Optimization Decision Module: Responsible for defining and optimizing multi-objective functions.

[0132] Regulation and evidence storage module: performs dynamic regulation and evidence storage operations for edge computing.

[0133] This embodiment, based on electromagnetic simulation data, introduces a quantum genetic algorithm for multi-objective optimization and combines it with 5G-MEC edge computing for dynamic control. Its advantages include: breakthrough in algorithm efficiency: the quantum genetic algorithm is 40 times faster than the traditional genetic algorithm (GPU accelerated); improved engineering applicability: supports real-time control over a 50km² area (latency <200ms); compatibility with existing wind turbine control system protocols; and enhanced data security: blockchain-based evidence storage is certified by DNV-GL Maritime.

[0134] The core value of this embodiment lies in providing an electromagnetic interference-resistant nearshore wind farm site selection auxiliary decision-making and dynamic control system, which is particularly suitable for breakwaters and nearshore wind power construction planning in areas with dense ship radar. It integrates a variety of advanced technologies to form a closed-loop system that can dynamically optimize site selection, ensure long-term stability, and realize full-link electromagnetic interference-resistant management from static site selection to dynamic operation, taking into account both real-time performance and reliability.

[0135] Example 4.

[0136] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 5As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a nearshore wind farm site selection auxiliary decision-making method or a nearshore wind farm dynamic control method.

[0137] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0138] In one exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the nearshore wind farm site selection auxiliary decision-making method in Embodiment 1 or the nearshore wind farm dynamic control method in Embodiment 2.

[0139] Example 5.

[0140] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the nearshore wind farm site selection auxiliary decision-making method of Embodiment 1 or the nearshore wind farm dynamic control method of Embodiment 2.

[0141] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0142] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0143] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. An offshore wind farm siting aid decision method, characterized in that, The offshore wind farm site selection auxiliary decision-making method comprises the following steps: Based on the candidate points in the candidate area, an initial population of a quantum genetic algorithm is randomly generated; the initial population comprises a plurality of quantum chromosomes, and each quantum chromosome represents a site selection scheme; the quantum chromosome comprises selection probabilities of all the candidate points; when a candidate point is selected and used to build a wind turbine, the selection probability is 1; For each quantum chromosome in the initial population, the electromagnetic interference intensity of each selected candidate point in the quantum chromosome is determined based on the electromagnetic interference thermal map of the candidate area; the radar interference index corresponding to the quantum chromosome is calculated based on the electromagnetic interference intensity of each selected candidate point in the quantum chromosome; and the fitness value corresponding to the quantum chromosome is calculated based on the quantum chromosome and the radar interference index corresponding to the quantum chromosome; the fitness value is the weighted sum of the power generation, the radar interference index, the construction cost deviation and the maintenance cost. Based on the fitness value corresponding to each quantum chromosome in the initial population, the initial population is updated by using the quantum genetic algorithm to obtain an updated population; It is determined whether an iteration termination condition is reached; if yes, a plurality of candidate site selection schemes are determined based on the fitness value corresponding to each quantum chromosome in the updated population; and if no, the updated population is taken as the initial population of the next iteration, and the step of "for each quantum chromosome in the initial population, the electromagnetic interference intensity of each selected candidate point in the quantum chromosome is determined based on the electromagnetic interference thermal map of the candidate area" is returned.

2. The offshore wind farm site selection aid decision method according to claim 1, characterized in that, Based on the electromagnetic interference thermal map of the candidate area, the electromagnetic interference intensity of each selected candidate point in the quantum chromosome is determined, specifically comprising the following steps: The electromagnetic interference thermal map of the candidate area is normalized to obtain a normalized electromagnetic interference thermal map; the normalized electromagnetic interference thermal map comprises normalized electromagnetic interference intensity prediction values of each grid in the candidate area; For each selected candidate point in the quantum chromosome, a grid in which the selected candidate point is located in the candidate area is determined to obtain a target grid, and the electromagnetic interference intensity of the selected candidate point is calculated based on the normalized electromagnetic interference intensity prediction values of the neighbor grids corresponding to the target grid. The calculation formula of the electromagnetic interference intensity of the selected candidate point is as follows: ； wherein, is the selected candidate point ; is the electromagnetic interference intensity of the selected candidate point ; is a function; is the number of neighborhood grids corresponding to the target grid where the selected candidate point ; is a set composed of neighborhood grids corresponding to the target grid where the selected candidate point ; is the normalized electromagnetic interference intensity prediction value of the neighborhood grid located at the th row and the th column.

3. The offshore wind farm site selection aid decision method of claim 1, wherein, Based on the electromagnetic interference intensity of each selected candidate point in the quantum chromosome, the radar interference index corresponding to the quantum chromosome is calculated, specifically comprising the following steps: the electromagnetic interference intensities of each selected candidate point in the quantum chromosome are summed to obtain the radar interference index corresponding to the quantum chromosome; or the electromagnetic interference intensities of each selected candidate point in the quantum chromosome are weighted and summed to obtain the radar interference index corresponding to the quantum chromosome.

4. The offshore wind farm site selection aid decision method of claim 1, wherein, The calculation formula of the fitness value is as follows: ； in, This is the fitness value; It is the first weight; The power generation is calculated based on the historical wind speed time series of each selected candidate point in the quantum chromosome. As the second weight; Radar interference index; It is the third weight; Construction cost deviation The reciprocal of the construction cost deviation is the deviation between the construction cost and the budget of the site selection scheme represented by the quantum chromosome. It is the fourth weight; To maintain costs The reciprocal of the value, the maintenance cost is calculated based on the position of each selected candidate point in the quantum chromosome.

5. The offshore wind farm site selection aid decision method according to claim 1, characterized in that, Based on the fitness value corresponding to each quantum chromosome in the initial population, the initial population is updated by using a quantum genetic algorithm to obtain an updated population, specifically comprising: Based on the fitness value corresponding to each quantum chromosome in the initial population, each quantum chromosome in the initial population is determined as a dominated solution or a non-dominated solution by using a Pareto front screening. For each quantum chromosome determined as a dominated solution, the quantum chromosome determined as a dominated solution is rotated by using a quantum rotation gate to obtain a rotated quantum chromosome, and the rotated quantum chromosome is mutated by using a quantum NOT gate to obtain an updated quantum chromosome. All quantum chromosomes determined as non-dominated solutions and all updated quantum chromosomes are combined to form an updated population, so as to update the initial population by using a quantum genetic algorithm to obtain an updated population.

6. The offshore wind farm site selection aid decision method according to claim 5, characterized in that, For each quantum chromosome determined as a dominated solution, the quantum chromosome determined as a dominated solution is rotated by using a quantum rotation gate to obtain a rotated quantum chromosome, and the rotated quantum chromosome is mutated by using a quantum NOT gate to obtain an updated quantum chromosome, specifically comprising: The fitness value corresponding to the quantum chromosome determined as a dominated solution is compared with an optimal fitness value and an average fitness value respectively to obtain a comparison result, a rotation angle is determined based on the comparison result, and the quantum chromosome determined as a dominated solution is rotated by using a quantum rotation gate with the rotation angle to obtain a rotated quantum chromosome. For each quantum bit in the rotated quantum chromosome, it is judged whether a randomly generated random number is less than a mutation probability, if yes, the quantum bit is mutated by using a quantum NOT gate to obtain an updated quantum bit, if not, the quantum bit is taken as an updated quantum bit, and all updated quantum bits are combined to form an updated quantum chromosome, so as to mutate the rotated quantum chromosome by using a quantum NOT gate to obtain an updated quantum chromosome. The fitness value corresponding to the quantum chromosome determined as the dominated solution is compared with the optimal fitness value and the average fitness value respectively to obtain a comparison result, and a rotation angle is determined based on the comparison result, specifically including: if the fitness value corresponding to the quantum chromosome determined as the dominated solution is equal to the optimal fitness value, the rotation angle is determined as +0.05 , if the fitness value corresponding to the quantum chromosome determined as the dominated solution is less than the average fitness value, the rotation angle is determined as -0.03 , otherwise, the rotation angle is determined as +0.01 or -0.01 .

7. A method of dynamic control of an offshore wind farm, characterized in that, The offshore wind farm dynamic control method comprises: A number of candidate site selection schemes are determined by using the offshore wind farm site selection decision-making assistance method of any one of claims 1-6, and a final site selection scheme is determined based on the number of candidate site selection schemes; Real-time interference intensities of each wind turbine in the offshore wind farm constructed by using the final site selection scheme are obtained; For each wind turbine, it is judged whether the real-time interference intensity of the wind turbine is greater than a preset threshold, if yes, a yaw control amount of the wind turbine is calculated based on the real-time interference intensity of the wind turbine and the preset threshold, and the yaw control amount of the wind turbine is sent to a controller of the wind turbine; the controller of the wind turbine is used to perform yaw control on the wind turbine based on the yaw control amount of the wind turbine.

8. The offshore wind farm dynamic control method according to claim 7, characterized in that, The calculation formula of the yaw control amount of the wind turbine is: ； wherein, is a yaw control variable of the wind turbine; is a real-time disturbance intensity of the wind turbine; is a preset threshold value; ； wherein, is the historical interference mean; is the dynamic adjustment factor; is the historical interference standard deviation; ； The offshore wind farm dynamic control method further comprises recording site selection decision-making assistance data and dynamic control data by using a blockchain storage system, the site selection decision-making assistance data comprises an electromagnetic interference thermal map of a candidate area and a number of candidate site selection schemes, and the dynamic control data comprises a real-time interference intensity of a wind turbine, a preset threshold, and a yaw control amount of the wind turbine.

9. A computer device comprising: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, characterized in that the processor executes the computer program to implement the offshore wind farm site selection decision-making method of any one of claims 1-6 or the offshore wind farm dynamic control method of any one of claims 7-8.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the offshore wind farm site selection decision-making method of any one of claims 1-6 or the offshore wind farm dynamic control method of any one of claims 7-8.

Images