Fan yaw regulation and control method and device considering wake flow influence, terminal equipment and storage medium

Abstract

The invention discloses a draught fan yaw regulation and control method and device considering wake flow influence, terminal equipment and a storage medium, and belongs to the technical field of electric power systems.The method comprises the steps that wind regime data in a to-be-optimized time window and draught fan data of a target wind field are obtained; according to the wind regime data, determining a yaw angle optimization value, and corresponding optimization total power and yaw angle adjustment cost of a fan in the target wind field under a plurality of candidate yaw angle adjustment durations, and constructing a corresponding Pareto front; based on the fan data, the Pareto frontier is solved by taking maximization of the optimized total power of the fans in the target wind field and minimization of the yaw angle adjustment cost in the target wind field as targets, and the target yaw angle adjustment duration of the fans in the target wind field is obtained; and regulating and controlling the fan in the target wind field according to the yaw angle optimization value of the fan in the target wind field and the target yaw angle regulation duration. By implementing the method, the problems of high cost and low efficiency of an existing yaw control strategy can be solved, and high-efficiency and low-cost yaw angle regulation and control can be realized.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular, to a yaw control method, device, terminal device, and storage medium for a wind turbine considering wake effects. Background Art

[0002] With the increasing global emphasis on clean energy and sustainable development, wind energy, as an important renewable energy source, plays a crucial role in the energy transition. However, effectively integrating wind energy into the power system, especially in large-scale wind farm operations, faces numerous challenges. Yaw control is a key link in optimizing wind farm operations and maximizing power output. It involves adjusting the direction of the wind turbine to ensure its optimal alignment with the main wind direction, thereby improving power generation efficiency.

[0003] However, in wind farm operations, the wake effect becomes an important factor affecting the yaw control effect. The wake effect refers to the reduction in wind speed and the increase in turbulence after the upstream wind turbine extracts kinetic energy from the wind, which causes aerodynamic interference to the downstream wind turbine, reducing its efficiency and resulting in significant power losses. This effect creates a complex non-linear relationship among the wind turbine position, yaw angle, and overall power output, making it particularly complex to optimize the yaw control strategy. To mitigate the impact of the wake effect, existing research has explored various methods, such as wake modeling and optimization techniques. However, these methods usually rely on computational fluid dynamics (CFD) models to predict the impact of the upstream wind turbine on the downstream wind turbine. The fluid dynamics process involves dynamic analyses such as wind speed changes, turbulence generation, and propagation. To accurately simulate this fluid dynamics process, high-precision physical models, such as the Navier-Stokes equations, are required. These equations have a huge computational cost when solved, ultimately resulting in a high computational cost and time cost for obtaining the yaw control strategy. Summary of the Invention

[0004] Embodiments of the present invention provide a yaw control method, device, terminal device, and storage medium for a wind turbine considering wake effects, which can solve the problems of high cost and low efficiency in obtaining the yaw control strategy caused by the need to construct a high-precision physical model to simulate the fluid dynamics process when regulating the yaw angle in the target wind farm, and achieve high-efficiency and low-cost regulation of the yaw angle of the wind turbines in the target wind farm.

[0005] An embodiment of the present invention provides a yaw control method for a wind turbine considering wake effects, including:

[0006] Obtaining wind condition data within a time window to be optimized and wind turbine data of a target wind farm;

[0007] Determining an optimized value of the yaw angle of the wind turbines in the target wind farm according to the wind condition data;

[0008] Based on the optimized yaw angle values of the wind turbines in the target wind farm, determine the optimized total power and yaw angle adjustment cost corresponding to the wind turbines in the target wind farm at several candidate yaw angle adjustment durations;

[0009] According to the optimized total power and yaw angle adjustment cost, determine the Pareto front of the optimized total power of the wind turbines in the target wind farm and the yaw angle adjustment cost of the wind turbines in the target wind farm;

[0010] Based on the wind turbine data, with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, solve the Pareto front to obtain the target yaw angle adjustment duration of the wind turbines in the target wind farm;

[0011] Regulate the yaw angle of the wind turbines in the target wind farm with the optimized yaw angle values and the target yaw angle adjustment duration of the wind turbines in the target wind farm.

[0012] Further, the determining the optimized yaw angle values of the wind turbines in the target wind farm according to the wind condition data includes:

[0013] Determine the optimization order of the yaw angles of the wind turbines in the target wind farm according to the wind condition data;

[0014] According to the optimization order of the yaw angles, use a preset yaw angle optimization algorithm to iteratively optimize the yaw angles of the wind turbines in the target wind farm to obtain the optimized yaw angle values of the wind turbines in the target wind farm.

[0015] Further, the determining the optimization order of the yaw angles of the wind turbines in the target wind farm according to the wind condition data includes:

[0016] Generate a wake interaction matrix of the target wind farm according to the wind condition data; wherein, the wake interaction matrix contains several elements, and each element represents the wake influence relationship between two wind turbines in the target wind farm;

[0017] Determine the hierarchical index vector within the time window to be optimized according to the wake interaction matrix; wherein, each hierarchical index vector corresponds to a hierarchical index value, and the hierarchical index value represents the wake influence degree of the wind turbines in the target wind farm corresponding to the current hierarchical index value on the rest of the wind turbines in the target wind farm;

[0018] Determine the optimization order of the yaw angles of the wind turbines in the target wind farm according to the hierarchical index vector.

[0019] Further, the determining the hierarchical index vector within the time window to be optimized according to the wake interaction matrix includes:

[0020] Perform an index update operation on the wake interaction matrix to obtain the hierarchical index vector within the time window to be optimized;

[0021] Among them, the index update operation includes:

[0022] According to the wake interaction matrix, determine a number of first-class wind turbines and a number of second-class wind turbines in the target wind farm at the current time step; among them, the first-class wind turbines are wind turbines that will not cause wake effects on the remaining wind turbines, and the second-class wind turbines are wind turbines that will cause wake effects on the remaining wind turbines;

[0023] Set the corresponding elements of the first-class wind turbines in the index update vector to the first bit value, and set the corresponding elements of the second-class wind turbines in the index update vector to the second bit value;

[0024] According to the wake interaction matrix, determine the hierarchical index vectors of each wind turbine in the target wind farm at the current time step;

[0025] Use the sum of the hierarchical index vectors of each wind turbine in the target wind farm at the current time step and the index update vector to determine the hierarchical index vector of the next time step, and set the elements corresponding to the first-class wind turbines in the wake interaction matrix to 0;

[0026] Obtain the current execution count, and determine whether the current execution count reaches the execution threshold, where the execution threshold is determined according to the total number of wind turbines in the target wind farm;

[0027] If so, use the hierarchical index vector of the next time step as the hierarchical index vector within the time window to be optimized;

[0028] If not, increment the current execution count by 1 and execute the index update operation.

[0029] Furthermore, the wind turbine data includes: the optimized output power of each wind turbine in the target wind farm, the total number of wind turbines in the target wind farm, the value of the yaw bearing, the rated life of the yaw bearing, the historical wind direction data, the data on the yaw angle limit range of the wind turbine, the speed limit of the yaw bearing, and the wake adjustment limit value;

[0030] Based on the wind turbine data, with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, solve the Pareto front to obtain the target yaw angle adjustment duration of the wind turbines in the target wind farm, including:

[0031] Based on the wind turbine data, with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, establish a multi-objective balance function; among them, the multi-objective balance function includes a first coefficient as the weight of the optimized total power and a second coefficient as the weight of the yaw angle adjustment cost;

[0032] Establish the yaw angle constraint, yaw rate constraint, and wake overlap constraint of the multi-objective balance function based on the fan data;

[0033] Solve the multi-objective balance function under the constraints of the yaw angle constraint, yaw rate constraint, and wake overlap constraint to obtain the target first coefficient and target second coefficient under maximizing the total optimized power of the fans in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm;

[0034] Obtain the target yaw angle adjustment duration of the fans in the target wind farm from the Pareto front according to the target first coefficient and target second coefficient.

[0035] Furthermore, the multi-objective balance function is specifically:

[0036]

[0037] where J represents the value of the multi-objective balance function; ω1 is the first coefficient; ω2 is the second coefficient; N is the total number of fans in the target wind farm; P i is the optimized output power of the i-th fan in the target wind farm; C is the value of the yaw bearing; L0 is the rated life of the yaw bearing; t c is the target yaw angle adjustment duration; θ i (t) is the yaw angle adjusted at each time step; σ ωd is the standard deviation of the wind direction historical data; μ ωd is the average value of the wind direction historical data; γ is a tuning parameter for controlling the sensitivity of weight adjustment; FI is the wind condition fluctuation index.

[0038] Furthermore, the yaw angle constraint is specifically:

[0039] θ min ≤θ i ≤θ max ;

[0040] where θ i is the yaw angle of fan i; θ min is the lower limit value of the yaw angle adjustment of the fan; θ max is the upper limit value of the yaw angle adjustment of the fan;

[0041] The yaw rate constraint is specifically:

[0042]

[0043] where ω max represents the yaw bearing speed limit;

[0044] The wake overlap constraint is specifically:

[0045] Δvi ≤v threshold ;

[0046] where Δv i is the wake effect from the upstream wind turbine; v threshold represents the wake adjustment limit value.

[0047] Based on the above method embodiment, the present invention correspondingly provides an apparatus embodiment;

[0048] An embodiment of the present invention correspondingly provides a wind turbine yaw control device considering wake influence, including: a data acquisition module, a yaw angle optimization value determination module, an optimization strategy determination module, a strategy solution module, and a control module;

[0049] The data acquisition module is configured to acquire wind condition data within a time window to be optimized and wind turbine data of a target wind farm;

[0050] The yaw angle optimization value determination module is configured to determine the yaw angle optimization value of the wind turbine within the target wind farm in the time window to be optimized according to the wind condition data;

[0051] The strategy determination module is configured to, based on the yaw angle optimization value of the wind turbine within the target wind farm, determine the corresponding total optimized power and yaw angle adjustment cost of the wind turbine within the target wind farm under several candidate yaw angle adjustment durations; and determine the Pareto frontier between the total optimized power of the wind turbine within the target wind farm and the yaw angle adjustment cost of the wind turbine within the target wind farm according to the total optimized power and the yaw angle adjustment cost;

[0052] The strategy solution module is configured to, based on the wind turbine data, solve the Pareto frontier with the goal of maximizing the total optimized power of the wind turbine within the target wind farm and minimizing the yaw angle adjustment cost within the target wind farm, and obtain the target yaw angle adjustment duration of the wind turbine within the target wind farm;

[0053] The control module is configured to control the yaw angle of the wind turbine within the target wind farm with the yaw angle optimization value and the target yaw angle adjustment duration of the wind turbine within the target wind farm.

[0054] Another embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method for controlling the yaw of a wind turbine considering wake influence described in the above embodiment of the present invention.

[0055] Another embodiment of the present invention provides a storage medium, which includes a stored computer program. When the computer program runs, it controls the device where the storage medium is located to execute the method for controlling the yaw of a wind turbine considering wake influence described in the above embodiment of the present invention.

[0056] The implementation of the present invention has the following beneficial effects:

[0057] The present invention provides a yaw control method, device, terminal device and storage medium for a wind turbine considering wake influence. The method includes obtaining wind condition data within a time window to be optimized and wind turbine data of a target wind farm; determining an optimized yaw angle value of the wind turbines in the target wind farm according to the wind condition data; based on the optimized yaw angle value of the wind turbines in the target wind farm, determining the optimized total power and yaw angle adjustment cost corresponding to the wind turbines in the target wind farm under several candidate yaw angle adjustment durations; and determining the Pareto frontier between the optimized total power of the wind turbines in the target wind farm and the yaw angle adjustment cost of the wind turbines in the target wind farm according to the optimized total power and the yaw angle adjustment cost. By using the optimized total power and yaw angle adjustment cost corresponding to the wind turbines in the target wind farm under several candidate yaw angle adjustment durations to construct the Pareto frontier between the optimized total power of the wind turbines in the target wind farm and the yaw angle adjustment cost of the wind turbines in the target wind farm, and then solving the Pareto frontier, the analysis of complex fluid dynamics processes is avoided, there is no need to construct a high-precision physical model for simulating fluid dynamics processes, the fluid dynamics analysis process is transformed into a problem of solving the Pareto frontier, and then by solving the Pareto frontier, the target yaw angle adjustment duration of the wind turbines in the target wind farm within the time window to be optimized can be obtained, realizing the control of the yaw angle of the wind turbines in the target wind farm with high efficiency and low cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 FIG. is a schematic flowchart of a yaw control method for a wind turbine considering wake influence provided by an embodiment of the present invention.

[0059] Figure 2 FIG. is a schematic diagram of a wind farm layout provided by an embodiment of the present invention.

[0060] Figure 3 FIG. is a schematic diagram of the actual wind conditions during a 60-minute period provided by an embodiment of the present invention.

[0061] Figure 4 FIG. is a schematic diagram of the result of yaw control of a wind farm by a yaw optimization method for a wind turbine considering wake influence provided by an embodiment of the present invention.

[0062] Figure 5 FIG. is a schematic diagram of a yaw bearing failure provided by an embodiment of the present invention.

[0063] Figure 6 FIG. is a schematic diagram of a node removal failure provided by an embodiment of the present invention.

[0064] Figure 7 FIG. is a schematic diagram of wind condition curves with two different fluctuation levels provided by an embodiment of the present invention.

[0065] Figure 8 It is a schematic structural diagram of a wind turbine yaw control device considering wake influence provided by an embodiment of the present invention. Detailed implementation manners

[0066] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0067] As Figure 1 shown, it is a method for controlling the yaw of a wind turbine considering wake influence provided by an embodiment of the present invention, including:

[0068] Step S1: Obtain wind condition data within a time window to be optimized and wind turbine data of a target wind farm;

[0069] Step S2: Determine the optimized yaw angle value of the wind turbines in the target wind farm according to the wind condition data;

[0070] Step S3: Based on the optimized yaw angle value of the wind turbines in the target wind farm, determine the optimized total power and yaw angle adjustment cost corresponding to the wind turbines in the target wind farm under several candidate yaw angle adjustment durations;

[0071] Step S4: According to the optimized total power and yaw angle adjustment cost, determine the Pareto frontier between the optimized total power of the wind turbines in the target wind farm and the yaw angle adjustment cost of the wind turbines in the target wind farm;

[0072] Step S5: Based on the wind turbine data, with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, solve the Pareto frontier to obtain the target yaw angle adjustment duration of the wind turbines in the target wind farm;

[0073] Step S6: Control the yaw angle of the wind turbines in the target wind farm with the optimized yaw angle value and the target yaw angle adjustment duration of the wind turbines in the target wind farm.

[0074] For step S1, in the present invention, the time window to be optimized refers to a certain future time period, and the yaw angle of the wind turbines in the target wind farm is adjusted according to the optimized yaw angle value.

[0075] The wind condition data within the time window to be optimized mainly includes wind speed data and wind direction data, and this wind condition data can be obtained through various means. For example, based on historical data and real-time data, through the Koopman method given in relevant literature, the wind speed data and wind direction data can be predicted within the time window to be optimized, thereby obtaining the wind condition data within the time window to be optimized.

[0076] The fan data of the target wind farm within the time window to be optimized mainly includes the optimized output power of each fan in the target wind farm, the total number of fans in the target wind farm, the value of the yaw bearing, the rated life of the yaw bearing, the historical wind direction data, the data of the fan yaw angle limit range, the speed limit of the yaw bearing, and the wake adjustment limit value. Among them, the fan yaw angle limit range data includes the lower limit value of the fan yaw angle adjustment and the upper limit value of the fan yaw angle adjustment.

[0077] In the present invention, the target wind farm is composed of multiple fans arranged according to a spatial layout, and the performance of each fan is affected by the wake effect generated by the upstream fan. By optimizing the yaw angle, the energy production of the entire wind farm can be optimized.

[0078] For step S2, the optimized yaw angle value of the fan in the target wind farm within the time window to be optimized can be obtained by combining the wind condition data with a preset yaw angle optimization algorithm.

[0079] In a preferred embodiment, determining the optimized yaw angle value of the fan in the target wind farm according to the wind condition data includes: determining the optimization order of the yaw angle of the fan in the target wind farm according to the wind condition data; according to the optimization order of the yaw angle, using a preset yaw angle optimization algorithm to iteratively optimize the yaw angle of the fan in the target wind farm to obtain the optimized yaw angle value of the fan in the target wind farm.

[0080] Specifically, the relative position of the fans in the target wind farm along the wind direction can be determined according to the wind condition data, and the optimization order of the yaw angle of the fans in the target wind farm can be determined based on this relative position to reduce the calculation amount required by the yaw angle optimization algorithm. Preferably, the optimization order of the yaw angle is: along the wind direction, processing from the downstream fan to the upstream fan all the time. The determination of this optimization order of the yaw angle is because the downstream fan is not affected by the wakes of other fans, and preferentially processing the downstream fan can effectively improve the calculation efficiency of the yaw angle optimization algorithm.

[0081] Exemplarily, the preset yaw angle optimization algorithm is as follows:

[0082] (1): Initialization: Assume that there are N fans in the target wind farm, set the initial yaw angles of the N fans as y o =[θ1, θ2, …, θ N , and calculate the reference power output P o .

[0083] During each optimization iteration, the adjustment of the yaw angle needs to comply with the physical constraints imposed by the mechanical properties of the wind turbine. Specifically, the yaw adjustment Δθ for each iteration is limited to:

[0084] Δθ = ω max ·t c ;(1)

[0085] where ω max is the yaw bearing speed limit; t c is the duration of the yaw angle adjustment.

[0086] By restricting the yaw adjustment during each iteration, it is ensured that the preset yaw angle optimization algorithm complies with the physical limitations of the wind turbine operation and prevents over-adjustment.

[0087] (2): Based on the optimization order of the yaw angle, determine at least one wind turbine to be processed in each optimization step, and perform iterative adjustments in sequence from the downstream wind turbine to the upstream wind turbine.

[0088] (3): Iterative adjustment: For each optimization step k:

[0089] For each wind turbine i in this adjustment, calculate the yaw angles of the forward adjustment strategy and the negative adjustment strategy:

[0090] y k,1 = y k +Δy;(2)

[0091] y k,2 = y k -Δy;(3)

[0092] where Δy k = [0,…,Δθ i ,…,0] is the adjustment degree of the yaw angle in the kth optimization step, indicating that only the wind turbine i has an adjustment degree of Δθ in this adjustment i ; y k is the yaw angle of N wind turbines in the kth optimization step; y k,1 is the yaw angle according to the forward adjustment strategy in the kth optimization step; y k,2 is the yaw angle according to the negative adjustment strategy in the kth optimization step.

[0093] Evaluate the power output of the entire wind farm before and after the adjustment under different adjustment strategies: P 调整前 = f(y k ), P 调整后,1 = f(y k,1 ), P 调整后,2 = f(y k,2) Among them, f(·) is the power prediction function, for example, the power prediction function given by the modified FLORIS model. P 调整前 is the wind farm power output before adjustment, P 调整后,1 is the wind farm power output after adjustment under the positive adjustment strategy; P 调整后,2 is the wind farm power output after adjustment under the negative adjustment strategy.

[0094] Apply optimal adjustment:

[0095]

[0096] Update the yaw angle:

[0097]

[0098] Among them, is the optimal adjustment degree of the yaw angle; y k+1 is the yaw angle after optimal adjustment at the k-th optimization step.

[0099] Based on the optimization order of the yaw angle, continue with the next optimization step k + 1. If the wind turbine being processed in the current iteration is at the most upstream level, return to the turbine at the most downstream level for processing.

[0100] (4): Convergence check: Repeat step (3) until all turbines have been optimized or the maximum number of iterations is reached.

[0101] The pseudocode for executing the above steps (3) and (4) is as follows:

[0102] for iteration k = 1 to Kopt do

[0103] for wind turbine i = 1 to N in hierarchical order do

[0104] Select the adjustment and update the yaw angle according to formulas (1) to (5).

[0105] end for

[0106] Check whether convergence has occurred or the maximum number of iterations has been reached.

[0107] end for

[0108] (5): Power growth evaluation: Calculate the total optimized power P of each turbine in the wind farm at each yaw angle optimization step total and evaluate the improvement:

[0109]

[0110] (6): Select the yaw angle of the wind turbine corresponding to the yaw angle adjustment scheme with the maximum optimized total power as the optimized value of the yaw angle of the wind turbines in the target wind farm within the time window to be optimized.

[0111] The yaw angle optimization algorithm proposed by the present invention adopts a greedy optimization method, and its iterative nature allows continuous improvement of power output through small, incremental yaw adjustments. The algorithm dynamically evaluates and adjusts the yaw angle of the wind turbines within each time window to be optimized. In this way, it can respond in a timely manner to the changing wind conditions over time, ensuring that the yaw angle can be adjusted in a timely manner with the wind conditions to maximize the total power output. In addition, the algorithm also uses a dynamic model (such as the FLORIS model) specifically for evaluating the power output or performance of a wind farm for prediction and evaluation. By comparing the power outputs of different yaw configurations, the best direction for each wind turbine is selected, thereby increasing the overall power output of the target wind farm.

[0112] In a preferred embodiment, determining the optimization order of the yaw angles of the wind turbines in the target wind farm according to the wind condition data includes: generating a wake interaction matrix of the target wind farm according to the wind condition data; wherein, the wake interaction matrix contains a number of elements, and each element characterizes the wake influence relationship between two wind turbines in the target wind farm; determining a hierarchical index vector within the time window to be optimized according to the wake interaction matrix; wherein, each hierarchical index vector corresponds to a hierarchical index value, and the hierarchical index value characterizes the wake influence degree of the wind turbines in the target wind farm corresponding to the current hierarchical index value on the remaining wind turbines in the target wind farm; determining the optimization order of the yaw angles of the wind turbines in the target wind farm according to the hierarchical index vector.

[0113] Specifically, first generate a wake interaction matrix of the target wind farm according to the wind condition data, and this wake interaction matrix is used to describe the wake influence relationship between two wind turbines in the target wind farm.

[0114] The wake effect of the upstream wind turbines in the wind farm has a significant impact on the performance of the downstream wind turbines, reducing their effective wind speed and efficiency. Preferably, the wake influence of wind turbine i on wind turbine j can be described by the following formula:

[0115]

[0116] Wherein, represents the wake influence of wind turbine i on wind turbine j; T is a certain point within the wake area of wind turbine i; z T is the height of point T; z0 is the hub height of wind turbine i; r i,T is the wake attenuation at T due to the wake of wind turbine i itself; r j,T is the wake attenuation at T due to wind turbine j; α is a constant that adjusts the influence of height in wake calculation; E T is a parameter of the wind power generation efficiency at point T.

[0117] This formula calculates the total wake effect of wind turbine i on wind turbine j by estimating all the contributions to the influence of the observation point T in wind turbine i on wind turbine j. A higher value indicates a stronger wake effect of wind turbine i on wind turbine j.

[0118] After calculating the wake effect relationships between all wind turbines in the target wind farm using the above formula, the following wake interaction matrix Φ can be obtained:

[0119]

[0120] Determine the hierarchical index vector within the time window to be optimized according to the wake interaction matrix. Each hierarchical index value in the hierarchical index vector corresponds to a hierarchical index value, and the hierarchical index value characterizes the wake influence degree of the wind turbines in the target wind farm corresponding to the current hierarchical index value on the wind turbines in the target wind farm other than those corresponding to the current hierarchical index value.

[0121] In a preferred embodiment, the determining the hierarchical index vector within the time window to be optimized according to the wake interaction matrix includes: performing an index update operation on the wake interaction matrix to obtain the hierarchical index vector within the time window to be optimized.

[0122] Specifically, in the present invention, the execution threshold is the total number of wind turbines in the target wind farm minus 1. By repeatedly executing steps (a) to (e), the hierarchical index vector within the time window to be optimized is obtained.

[0123] Among them, the index update operation includes:

[0124] (a): According to the wake interaction matrix, determine a number of first-type wind turbines and a number of second-type wind turbines in the target wind farm at the current time step; wherein, the first-type wind turbines are wind turbines that do not cause wake effects on the remaining wind turbines, and the second-type wind turbines are wind turbines that cause wake effects on the remaining wind turbines;

[0125] (b): Set the corresponding elements of the first-type wind turbines in the index update vector to a first bit value, and set the corresponding elements of the second-type wind turbines in the index update vector to a second bit value; wherein, each element in the index update vector is used to represent the update value of the element in the corresponding hierarchical index vector.

[0126] (c): According to the wake interaction matrix, determine the hierarchical index vector of each wind turbine in the target wind farm at the current time step;

[0127] (d): Determine the hierarchical index vector for the next time step by summing the hierarchical index vector and the index update vector of each wind turbine in the target wind farm at the current time step, and set the elements in the wake interaction matrix corresponding to the first type of wind turbines to 0;

[0128] (e): Obtain the current execution count and determine whether the current execution count reaches the execution threshold. If so, use the hierarchical index vector for the next time step as the hierarchical index vector within the time window to be optimized. If not, increment the current execution count by 1, use the wake interaction matrix in (d) as the latest wake interaction matrix, and return to step (a) for execution.

[0129] Specifically, according to the specific algorithm used, 0 can be used as the first bit value and 1 as the second bit value, or 1 as the first bit value and 0 as the second bit value. The specific values of the bit values are not limited by the description in this specification.

[0130] Exemplarily, with the first bit value being 1 and the second bit value being 0, the above steps (a) and (b) can be implemented using the following formula:

[0131]

[0132] where, is the index update vector; R(t) is an intermediate variable, whose initial value is the wake interaction matrix; bit(A, b) is a binary calculation function indicating whether each instance of A is equal to b; 1 N is a column vector of all 1s with length N.

[0133] The following formula is used to implement step (d):

[0134]

[0135] where, H(t) is the hierarchical index vector at the current time step, and the value of H(0) is 1 N ; H(t + 1) is the hierarchical index vector for the next time step;

[0136] In case, the following operations are further included:

[0137] R(t + 1) ← r ij (t + 1) = 0; (11)

[0138] where, R(t + 1) is the wake interaction matrix for the next time step; r ij (t + 1) is the element in the i-th row and j-th (where j ranges from 1 to N) column of the wake interaction matrix for the next time step.

[0139] Through steps (a) to (e), a higher hierarchical index value is assigned to the wind turbines with less impact on the wakes of other wind turbines, and a lower hierarchical index value is assigned to the wind turbines with greater impact on the wakes of other wind turbines.

[0140] Exemplarily, it is assumed that there are 4 wind turbines arranged in sequence in the target wind farm. During the time window to be optimized, wind turbine 1 is located upstream in the wind direction, and wind turbine 4 is located downstream in the wind direction. The wake interaction matrix of the 4 wind turbines is as follows:

[0141]

[0142] At time t = 0, the first iterative calculation is performed. Substituting the above wake interaction matrix (i.e., R(0)) into formula (9), we can obtain:

[0143]

[0144] In the index update vector obtained above Substituting into formula (10), we can obtain:

[0145] H(2) ← (1 1 1 2) + (0 0 1 1) = 1 1 2 3; (15)

[0146] Since According to formula (11), set r in R(t) i3 (including r 13 , r 23 , r 33 , r 43 ) to 0, and we get:

[0147]

[0148] And so on. After 3 iterative calculations are completed, the hierarchical index vector can be obtained as:

[0149] H(3) = (1 2 3 4); (17)

[0150] In the above example, the downstream wind turbine 4 is assigned the largest hierarchical index value, and the upstream wind turbine 1 is assigned the smallest hierarchical index value.

[0151] In addition, when the first bit value is 0 and the second bit value is 1, formula (9) can be transformed to invert its result. In the finally obtained hierarchical index vector, the downstream wind turbines are assigned the smallest hierarchical index values, and the upstream wind turbines are assigned the largest hierarchical index values. This will not be elaborated here.

[0152] According to the hierarchical index vector, determine the optimization order of the yaw angles of the wind turbines in the target wind farm.

[0153] Specifically, based on the hierarchical index vector, the relative positions of the wind turbines in the target wind farm are determined, and the order from the downstream wind turbine to the upstream wind turbine is used as the optimization order of the yaw angles of the wind turbines in the target wind farm. For example, for the 4 wind turbines in the examples from (a) to (e), the yaw angles are optimized in the order of wind turbine 4, wind turbine 3, wind turbine 2, and wind turbine 1.

[0154] According to the wake interaction matrix, the hierarchical index vector within the time window to be optimized is determined; according to the hierarchical index vector, the optimization order of the yaw angles of the wind turbines in the target wind farm is determined. Since the hierarchical index vector is used to reflect the wake influence degree of each wind turbine on other wind turbines in the target wind farm, therefore, through the hierarchical index vector, the upstream wind turbines and downstream wind turbines that conform to the wind direction within the time window to be optimized can be efficiently and accurately determined from the numerous wind turbines in the wind farm; furthermore, the yaw angle optimization can start from the downstream wind turbines first. Since the wake influence of the downstream wind turbines on other wind turbines is weak, adjusting the yaw angles of the downstream wind turbines first reduces the chain reaction of the yaw adjustment of each wind turbine, and can concentrate the computational resources required for yaw angle optimization on the wind turbines that have the greatest impact on the output of the entire wind farm, ensuring a significant and measurable increase in power output within fewer iterations, thereby effectively improving the computational efficiency of the yaw angle optimization algorithm.

[0155] The hierarchical index vector is determined based on the wake interaction matrix, and the wake interaction matrix is calculated according to the wind conditions and the situation of the in-service wind turbines within each time window to be optimized. When the wake interaction matrix and the hierarchical index vector are updated, newly connected wind turbines will be incorporated into the updated hierarchical index vector, while wind turbines that are temporarily offline due to maintenance or failure will be excluded. Therefore, the embodiments of the present invention can seamlessly adapt to the connection and disconnection of wind turbines in the wind farm, ensuring that the wind turbines always maintain the best output state.

[0156] For step S3, based on the optimized yaw angle values of the wind turbines in the target wind farm within the time window to be optimized, the total optimized power and the yaw angle adjustment cost corresponding to the wind turbines in the target wind farm under several candidate yaw angle adjustment durations are determined. It should be noted that the yaw angle adjustment duration is the time length required to adjust the yaw angle of the wind turbine. Since the wind conditions change with time, when adjusting the same yaw angle, the shorter the yaw angle adjustment duration, the faster the wind turbine can adapt to the wind conditions, and the greater the output power that can be obtained, but at the same time, it will cause greater mechanical wear of the wind turbine, thereby increasing the yaw angle adjustment cost.

[0157] The total optimized power of the wind turbines in the target wind farm is the sum of the output powers of each wind turbine in the target wind farm.

[0158] Exemplarily, the calculation formula for the total optimized power of the wind turbines in the target wind farm is shown as follows:

[0159]

[0160] Among them, N is the total number of wind turbines in the target wind farm.

[0161] The wake effect is considered when calculating the power of each downstream wind turbine. The power generated by wind turbine i is shown in the following formula:

[0162]

[0163] Among them, ρ is the air density; A is the swept area of the wind turbine blades; C p is the power coefficient (a function of wind speed and wind turbine design); v i is the wind speed at wind turbine i; Δv i represents the wake effect from the upstream wind turbine; P i is the output power of wind turbine i.

[0164] Preferably, the Gaussian wake model can be used to model the wake effect to calculate the wake effect from the upstream wind turbine. This model approximately represents the wake influence caused by the wind speed deficit upstream propagating to the downstream with distance. For each wind turbine i, its wake effect on the downstream wind turbine j is given by the following formula:

[0165]

[0166] Among them, v i is the wind speed at wind turbine i; d ij is the distance between wind turbine i and wind turbine j; σ is the wake decay constant characterizing wake diffusion, and its value is related to the yaw angle of the wind turbine.

[0167] For wind turbine i, the wake effect Δv i from the upstream wind turbine can be determined by the sum of the wake effects Δv ij of each upstream wind turbine j on wind turbine i.

[0168] Exemplarily, the calculation formula for the yaw angle adjustment cost is as follows:

[0169]

[0170] Among them, L yaw is the yaw angle adjustment cost; C is the value of the yaw bearing; L0 is the rated life; t c is the yaw angle adjustment duration.

[0171] For step S4, according to the optimized total power and yaw angle adjustment cost corresponding to the wind turbines in the target wind farm under several candidate yaw angle adjustment durations, determine the Pareto frontier of the optimized total power of the wind turbines in the target wind farm and the yaw angle adjustment cost of the wind turbines in the target wind farm.

[0172] Specifically, the Pareto frontiers of the optimized total power of the wind turbines in the target wind farm and the yaw angle adjustment cost under different yaw angle adjustment durations can be plotted. The Pareto frontiers provide a visual representation of the trade-off between maximizing the output power and minimizing the control cost in the target wind farm.

[0173] For step S5, based on the wind turbine data, with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, solve the Pareto frontiers to obtain the target yaw angle adjustment durations of the wind turbines in the target wind farm within the time window to be optimized.

[0174] In a preferred embodiment, the step of solving the Pareto frontiers based on the wind turbine data with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm to obtain the target yaw angle adjustment durations of the wind turbines in the target wind farm includes: based on the wind turbine data, establish a multi-objective balance function with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm; wherein the multi-objective balance function includes a first coefficient as the weight of the optimized total power and a second coefficient as the weight of the yaw angle adjustment cost; based on the wind turbine data, establish the yaw angle constraint, yaw rate constraint, and wake overlap constraint of the multi-objective balance function; solve the multi-objective balance function under the constraints of the yaw angle constraint, yaw rate constraint, and wake overlap constraint to obtain the target first coefficient and target second coefficient under the condition of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm; obtain the target yaw angle adjustment durations of the wind turbines in the target wind farm from the Pareto frontiers according to the target first coefficient and target second coefficient.

[0175] In a preferred embodiment, the multi-objective balance function is specifically:

[0176]

[0177] where J represents the value of the multi-objective balance function; ω1 is the first coefficient; ω2 is the second coefficient; N is the total number of wind turbines in the target wind farm; P i is the optimized output power of the i-th wind turbine in the target wind farm; C is the value of the yaw bearing; L0 is the rated life of the yaw bearing; t c is the target yaw angle adjustment duration; θ i (t) is the yaw angle adjusted at each time step; FI is the wind condition fluctuation index.

[0178] A point on the Pareto frontiers can be selected according to the wind condition fluctuation index (FI) to balance the optimized total power of the wind turbines in the target wind farm and minimize the yaw angle adjustment cost. The wind condition fluctuation index is defined as:

[0179]

[0180] Among them, σ ωd is the standard deviation of the historical wind direction data; μ ωd is the average value of the historical wind direction data. For the condition of height variation of wind conditions, a smaller yaw angle adjustment duration is preferably selected to respond to wind condition changes in a timely manner; for the stable condition of wind conditions, a larger yaw angle adjustment duration can minimize the control cost while maintaining an appropriate power output.

[0181] Furthermore, according to the following formula, the first coefficient and the second coefficient can be determined by using the wind condition fluctuation index within the time window to be optimized:

[0182]

[0183] Among them, γ is a tuning parameter for controlling the sensitivity of weight adjustment. A larger value of γ means that the wind condition fluctuation index (FI) has a more significant impact on the weight.

[0184] In a preferred embodiment, during the process of solving the multi-objective balance function, the constraint conditions of the function also need to be constructed, including yaw angle constraint, yaw rate constraint, and wake overlap constraint.

[0185] The purpose of the yaw angle constraint is to limit the yaw angle of each wind turbine within a feasible range so that it does not exceed the yaw angle limit range of the wind turbine.

[0186] The yaw angle constraint is specifically:

[0187] θ min ≤θ i ≤θ max ; (26)

[0188] Among them, θ i is the yaw angle of wind turbine i; θ min is the lower limit value of the yaw angle adjustment of the wind turbine; θ max is the upper limit value of the yaw angle adjustment of the wind turbine;

[0189] The purpose of the yaw rate constraint is to prevent excessive wear of the yaw bearing, and it is necessary to limit the maximum rate of change of the yaw angle Δθ i .

[0190] The yaw rate constraint is specifically:

[0191]

[0192] Among them, ω max represents the yaw bearing speed limit;

[0193] The purpose of wake overlap constraint is to consider the wake interference between wind turbines and ensure that downstream wind turbines are not in extreme wake conditions that may reduce their efficiency.

[0194] The wake overlap constraint is specifically as follows:

[0195] Δv i ≤v threshold ;(28)

[0196] Where, Δv i is the wake effect from the upstream wind turbine; v threshold represents the wake adjustment limit value, which is a predefined limit value. When this limit value is exceeded, the efficiency of the downstream wind turbine will decrease significantly.

[0197] Under the constraints of formulas (26)-(28), solve the multi-objective balance function of formula (22) to obtain the target first coefficient and the target second coefficient of the multi-objective balance function under the conditions of maximizing the total optimized power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm. Then, use the target first coefficient and the target second coefficient to select a point from the Pareto front as the optimal solution of the multi-objective balance function, and use the candidate yaw angle adjustment duration corresponding to the optimal solution of the multi-objective balance function as the target yaw angle adjustment duration.

[0198] For step S6, use the optimized yaw angle value and the target yaw angle adjustment duration of the wind turbines in the target wind farm within the to-be-optimized time window to regulate the yaw angle of the wind turbines in the target wind farm.

[0199] It should be added that, as Figure 2 shown, it is a schematic diagram of a target wind farm including wind turbines 1, 2, 3, 4, 5, and 6. Here, taking a wind farm composed of six wind turbines as an example, the beneficial technical effects of the present invention are described.

[0200] Figure 3 (a) shows the actual wind direction recorded within one hour, Figure 3 (b) shows the actual wind speed recorded within one hour. These measured values represent the typical wind conditions used in the experiment. The control duration of the adjustment period is fixed at once per minute. According to the technical report of the National Renewable Energy Laboratory (NREL), considering the cost and expected life of the yaw bearing, the cost is about 6 million US dollars and the life is about 280,000 hours. The angular velocity of the yaw system is set at 0.357 degrees per second.

[0201] Figure 4 Shows the application of the wind turbine yaw control method provided by the present invention (hereinafter referred to as OCO (Online Cluster Optimization)) within the control time frame and the corresponding real-time power output of each wind turbine. Among them,Figure 4 (a) is a comparison graph of the power generation of a wind farm. The legend Output1 is the power generation curve of the wind farm obtained by the wind turbine yaw control method proposed by the present invention, and the legend Output2 is the power generation curve of the wind farm obtained by the existing wind turbine yaw control method; Figure 4 (b) is a graph showing the change of the wind turbine yaw angle over time. The legend Turbine1 is the curve of the yaw angle change of wind turbine 1 over time, Turbine2 is the curve of the yaw angle change of wind turbine 2 over time, Turbine3 is the curve of the yaw angle change of wind turbine 3 over time, Turbine4 is the curve of the yaw angle change of wind turbine 4 over time, Turbine5 is the curve of the yaw angle change of wind turbine 5 over time, and Turbine6 is the curve of the yaw angle change of wind turbine 6 over time. Under typical wind conditions, wind turbines 3 and 6 are always located downstream, and their wakes will not significantly affect other wind turbines. Therefore, these wind turbines adopt a direct windward strategy, resulting in overlapping curves. Generally speaking, compared with the basic greedy strategy where all wind turbines face the wind directly, the power production has increased by 15.24%. The dynamic modeling environment (Floridyn) demonstrates the predictive ability of the algorithm, and the time required for the wake to pass through the wind farm after the initial adjustment shows improved performance. To demonstrate the flexibility of the OCO method of the present invention in handling node failures, a series of tests were conducted. The results are as Figure 5 shown, with a focus on two different fault scenarios. Among them, Figure 5 (a) is a comparison graph of the power generation of a wind farm; Figure 5 (b) is a graph showing the change of the wind turbine yaw angle over time. Taking Figure 5 (b) the 30 - minute node as the test node, the first test scenario is introduced.

[0202] In the first test scenario, a yaw bearing failure of wind turbine 2 was introduced at the 30th minute. This failure fixed the direction of the wind turbine at a 30 - degree angle, preventing it from adjusting its yaw to face the wind directly. Despite this major constraint, the OCO algorithm remained flexible and optimized the yaw angles of the remaining wind turbines without significantly affecting the overall performance. As Figure 5 shown, the relative power production increased by 15.06% during the control period. Using the Floridyn dynamic model ensured accurate evaluation even under these fault conditions, reflecting that the algorithm is still effective under real - world operating conditions.

[0203] Figure 6 (a) is a comparison graph of the power generation of a wind farm; Figure 6 (b) is a graph showing the change of the wind turbine yaw angle over time. Taking Figure 6(b) The 30 - minute node is a test node, introducing a second test scenario. In the second test scenario, at the 30 - minute mark, a generator node removal fault of wind turbine 1 is introduced, effectively removing it from the operating power grid. The OCO algorithm continues to optimize the remaining wind turbines by updating the wind turbine stratification metrics. It is worth noting that wind turbine 2, which was originally within the wake region of wind turbine 1, only slightly adjusts its yaw angle to provide sufficient wake space for wind turbine 3 downstream. The relative power production increased by 7.64% during the control period. The Floridyn dynamic model again played a key role in maintaining the evaluation accuracy, demonstrating the robustness and adaptability of the optimization strategy.

[0204] To further study the impact of control cost on performance, a multi - objective optimization analysis was conducted. First, data of wind condition curves with two different fluctuation levels within one hour was used. Despite the variability, the filtered wind direction trends were consistent, and the wind direction for the next 120 minutes could be predicted. Using these predictions, the optimal power output and control cost were tested within different control - cycle intervals (1, 2, 3, 4, 5, 6, 8, 10, 12, 15, 20, 30, 40, 60, 120 minutes). The results are summarized in Figure 7 which shows the Pareto curve obtained from this analysis. The tangent slopes of the Pareto fronts for the two wind condition curves are - 960√FI. The respective tangents intersect the Pareto curve at points A and B, indicating that for the first wind condition, the next control cycle should be updated to Tc = 3 minutes, and for the second wind condition, it should be updated to Tc = 40 minutes. This analysis provides valuable insights into balancing performance and control cost, guiding the optimization strategy for future wind farms.

[0205] Based on the above - mentioned method item embodiments, the present invention correspondingly provides device item embodiments.

[0206] As Figure 8 shown, an embodiment of the present invention provides a wind turbine yaw control device considering wake influence, including: a data acquisition module, a yaw angle optimization value determination module, an optimization strategy determination module, a strategy solution module, and a control module;

[0207] The data acquisition module is used to acquire wind condition data within the time window to be optimized and wind turbine data of the target wind farm;

[0208] The yaw angle optimization value determination module is used to determine the yaw angle optimization value of the wind turbines in the target wind farm according to the wind condition data;

[0209] The strategy determination module is configured to determine the optimized total power and yaw angle adjustment cost corresponding to the wind turbines in the target wind farm at several candidate yaw angle adjustment durations based on the optimized yaw angle values of the wind turbines in the target wind farm; and determine the Pareto frontier between the optimized total power of the wind turbines in the target wind farm and the yaw angle adjustment cost of the wind turbines in the target wind farm according to the optimized total power and the yaw angle adjustment cost.

[0210] The strategy solution module is configured to solve the Pareto frontier with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm based on the wind turbine data, so as to obtain the target yaw angle adjustment duration of the wind turbines in the target wind farm.

[0211] The regulation module is configured to regulate the yaw angle of the wind turbines in the target wind farm with the optimized yaw angle value and the target yaw angle adjustment duration of the wind turbines in the target wind farm.

[0212] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. In addition, in the attached drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. Those of ordinary skill in the art can understand and implement without creative work.

[0213] Those skilled in the art can clearly understand that for the sake of convenience and brevity, the specific working process of the device described above can refer to the corresponding process in the foregoing method embodiment, which will not be elaborated here.

[0214] Based on the above method item embodiments, the present invention correspondingly provides terminal device item embodiments.

[0215] An embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a wind turbine yaw regulation method considering wake effect according to any one of the present invention.

[0216] The terminal device can be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The terminal device may include, but is not limited to, a processor and a memory.

[0217] The so-called processor may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The processor is the control center of the terminal device and connects all parts of the entire terminal device through various interfaces and circuits.

[0218] The memory can be used to store the computer program. The processor realizes various functions of the terminal device by running or executing the computer program stored in the memory and calling the data stored in the memory. The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include high-speed random access memory and can also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

[0219] Based on the above method item embodiments, the present invention correspondingly provides storage medium item embodiments.

[0220] An embodiment of the present invention provides a storage medium, which includes a stored computer program. Among them, when the computer program runs, it controls the device where the storage medium is located to execute a method for regulating the yaw of a wind turbine considering wake influence according to any one of the present invention.

[0221] The storage medium is a computer-readable storage medium, and the computer program is stored in the computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0222] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art of this technology, without departing from the principle of the present invention, several improvements and refinements can still be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A yaw control method for a wind turbine considering wake influence, characterized in that, Including: Obtain the wind condition data within the time window to be optimized and the fan data of the target wind farm; Determine the optimized yaw angle value of the fans in the target wind farm according to the wind condition data; Based on the optimized yaw angle value of the fans in the target wind farm, determine the optimized total power and yaw angle adjustment cost corresponding to the fans in the target wind farm under several candidate yaw angle adjustment durations; According to the optimized total power and yaw angle adjustment cost, determine the Pareto frontier of the optimized total power of the fans in the target wind farm and the yaw angle adjustment cost of the fans in the target wind farm; Based on the fan data, with the goal of maximizing the optimized total power of the fans in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, solve the Pareto frontier to obtain the target yaw angle adjustment duration of the fans in the target wind farm; Use the optimized yaw angle value and the target yaw angle adjustment duration of the fans in the target wind farm to control the yaw angle of the fans in the target wind farm.

2. The yaw control method for a wind turbine considering wake influence according to claim 1, wherein, The determining the optimized yaw angle value of the fans in the target wind farm according to the wind condition data includes: Determine the optimization order of the yaw angles of the fans in the target wind farm according to the wind condition data; According to the optimization order of the yaw angles, use a preset yaw angle optimization algorithm to iteratively optimize the yaw angles of the fans in the target wind farm to obtain the optimized yaw angle value of the fans in the target wind farm.

3. The yaw control method for a wind turbine considering wake influence according to claim 2, characterized in that The determining the optimization order of the yaw angles of the fans in the target wind farm according to the wind condition data includes: Generate a wake interaction matrix of the target wind farm according to the wind condition data; wherein, the wake interaction matrix contains several elements, and each element represents the wake influence relationship between two fans in the target wind farm; Determine the hierarchical index vector within the time window to be optimized according to the wake interaction matrix; wherein, each hierarchical index vector corresponds to a hierarchical index value, and the hierarchical index value represents the wake influence degree of the fans in the target wind farm corresponding to the current hierarchical index value on the remaining fans in the target wind farm; Determine the optimization order of the yaw angles of the fans in the target wind farm according to the hierarchical index vector.

4. The yaw control method of a wind turbine considering wake influence according to claim 3, wherein The determining the hierarchical index vector within the time window to be optimized according to the wake interaction matrix includes: Perform an index update operation on the wake interaction matrix to obtain the hierarchical index vector within the time window to be optimized; Wherein, the index update operation includes: According to the wake interaction matrix, determine several first-type fans and several second-type fans in the target wind farm at the current time step; wherein, the first-type fans are the fans that will not cause wake influence on the remaining fans, and the second-type fans are the fans that will cause wake influence on the remaining fans; Set the corresponding elements of the first-type fans in the index update vector to a first bit value, and set the corresponding elements of the second-type fans in the index update vector to a second bit value; According to the wake interaction matrix, determine the hierarchical index vector of each fan in the target wind farm at the current time step; Use the sum of the hierarchical index vectors of each fan in the target wind farm at the current time step and the index update vector to determine the hierarchical index vector of the next time step, and set the elements corresponding to the first-type fans in the wake interaction matrix to 0; Obtain the current execution count, and determine whether the current execution count reaches the execution threshold, where the execution threshold is determined according to the total number of wind turbines in the target wind farm; If so, use the hierarchical index vector of the next time step as the hierarchical index vector within the time window to be optimized; If not, increment the current execution count by 1 and perform the index update operation.

5. The yaw control method of a wind turbine considering wake influence according to claim 1, characterized in that, The wind turbine data includes: the optimized output power of each wind turbine in the target wind farm, the total number of wind turbines in the target wind farm, the value of the yaw bearing, the rated life of the yaw bearing, the historical wind direction data, the yaw angle limit range data of the wind turbine, the yaw bearing speed limit, and the wake adjustment limit value; Based on the wind turbine data, with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, solve the Pareto front to obtain the target yaw angle adjustment duration of the wind turbines in the target wind farm, including: Based on the wind turbine data, with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, establish a multi-objective balance function; where the multi-objective balance function includes a first coefficient as the weight of the optimized total power and a second coefficient as the weight of the yaw angle adjustment cost; Establish the yaw angle constraint, yaw rate constraint, and wake overlap constraint of the multi-objective balance function based on the wind turbine data; Solve the multi-objective balance function under the constraints of the yaw angle constraint, yaw rate constraint, and wake overlap constraint to obtain the target first coefficient and target second coefficient under the condition of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm; Obtain the target yaw angle adjustment duration of the wind turbines in the target wind farm from the Pareto front according to the target first coefficient and target second coefficient.

6. The yaw control method for a wind turbine considering wake influence according to claim 5, characterized in that, The multi-objective balance function is specifically: Among them, J represents the value of the multi-objective balance function; ω1 is the first coefficient; ω2 is the second coefficient; N is the total number of wind turbines in the target wind farm; P i is the optimized output power of the i-th wind turbine in the target wind farm; C is the value of the yaw bearing; L0 is the rated life of the yaw bearing; t c is the target yaw angle adjustment duration; θ i (t) is the yaw angle adjusted at each time step; σ ωd is the standard deviation of the wind direction historical data; μ ωd is the average value of the wind direction historical data; γ is a tuning parameter that controls the sensitivity of weight adjustment; FI is the wind condition fluctuation index.

7. The yaw control method for a wind turbine considering wake influence according to claim 5, characterized in that, The yaw angle constraint is specifically: θ min ≤ θ i ≤ θ max ; Among them, θ i is the yaw angle of wind turbine i; θ min is the lower limit value of the yaw angle adjustment of the wind turbine; θ max is the upper limit value of the yaw angle adjustment of the wind turbine; The yaw rate constraint is specifically: Among them, ω max represents the yaw bearing speed limit; The wake overlap constraint is specifically: Δv i ≤v threshold ; Among them, Δv i is the wake effect from the upstream fan; v threshold represents the wake adjustment limit value.

8. A yaw control device for a wind turbine considering wake influence, characterized in that, It includes: A data acquisition module, a yaw angle optimization value determination module, an optimization strategy determination module, a strategy solution module, and a regulation module; The data acquisition module is used to acquire the wind condition data within the time window to be optimized and the wind turbine data of the target wind farm; The yaw angle optimization value determination module is used to determine the yaw angle optimization value of the wind turbines in the target wind farm according to the wind condition data; The strategy determination module is used to determine the optimized total power and yaw angle adjustment cost corresponding to the wind turbines in the target wind farm under several candidate yaw angle adjustment durations based on the yaw angle optimization value of the wind turbines in the target wind farm; according to the optimized total power and yaw angle adjustment cost, determine the Pareto front of the optimized total power of the wind turbines in the target wind farm and the yaw angle adjustment cost of the wind turbines in the target wind farm; The strategy solution module is used to solve the Pareto front based on the wind turbine data, with the goal of maximizing the optimized total power of the wind turbines in the target wind farm and minimizing the yaw angle adjustment cost in the target wind farm, to obtain the target yaw angle adjustment duration of the wind turbines in the target wind farm; The control module is used to adjust the yaw angle of the wind turbines in the target wind farm with the optimized yaw angle value and the target yaw angle adjustment duration of the wind turbines in the target wind farm.

9. A terminal device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements a wind turbine yaw control method considering wake influence as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium includes a stored computer program. When the computer program runs, it controls the device where the storage medium is located to execute a wind turbine yaw control method considering wake influence as described in any one of claims 1 to 7.

Images