A real-time cooperative yaw control method for a wind farm

Abstract

The application discloses a kind of real-time cooperative yaw control methods of wind farm, including the following steps: 1) obtaining the measured wind direction wind speed data and fan arrangement layout data of wind farm;2) establish the wind farm model based on two-dimensional coordinate;3) establish wind turbine unit model and wind turbine wake model;4) establish wind farm cooperative yaw control model;5) according to the wind direction wind speed data of wind farm real-time acquisition, obtain optimal wind turbine cooperative yaw control strategy;6) through yaw control model simulation real wind farm's working condition, the power effect of wind farm under optimal wind turbine cooperative yaw control strategy is verified.The application can be optimized in real time according to the wind environment change in wind farm, improve power generation efficiency, so as to obtain greater wind farm power generation.

Description

Technical Field

[0001] This invention relates to the field of wind power generation, and in particular to a real-time coordinated yaw control method for wind farms. Background Technology

[0002] With societal development, the scale of wind energy development is expanding. During the operation of wind farms, the wind turbine power generation control strategy has a crucial impact on the total power generation of the entire wind farm. Conventional wind farm control methods mostly employ a real-time greedy wind control strategy to maximize the power generation of a single turbine. However, this control strategy does not consider the impact of the wake effect of upstream turbines on downstream turbines, thus failing to maximize the power generation of the entire wind farm.

[0003] Cooperative yaw control methods maximize the total power generation of a wind farm by coordinating yaw control among the turbines and reducing the impact of wake effects on downstream turbine power generation. However, in real-world wind farms, the incoming wind speed and direction change in real time, and no real-time cooperative control method considering time-varying wind conditions has yet been proposed. Since the wind environment in a wind farm is constantly changing and significantly affects the performance of cooperative yaw control, traditional artificial intelligence methods require numerous iterations to reach the optimal state and are prone to getting trapped in local optima. This leads to significant time delays and negatively impacts the efficiency of wind farm control optimization.

[0004] Therefore, there is an urgent need to propose an efficient real-time coordinated yaw control method for wind farms. Summary of the Invention

[0005] To address the aforementioned shortcomings of existing technologies, the present invention aims to solve the problems of poor optimization of wind farm power generation, high control strategy delay, resulting in low power generation efficiency and low power output. It provides a real-time coordinated yaw control method for wind farms that can optimize in real time based on changes in the wind environment within the wind farm, thereby improving power generation efficiency and achieving greater wind farm power generation.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a real-time coordinated yaw control method for wind farms, characterized by comprising the following steps:

[0007] 1) Obtain measured wind direction and speed data and wind turbine layout data of the wind farm;

[0008] 2) Based on the wind turbine layout data, establish a wind farm model based on two-dimensional coordinates to simulate the spatial distribution of the wind farm;

[0009] 3) Establish wind turbine unit model and wind turbine wake model;

[0010] 4) Combine the wind farm model with the wind turbine model and the wind turbine wake model to establish a wind farm cooperative yaw control model;

[0011] 5) Based on the wind direction and speed data collected in real time from the wind farm, the yaw angle parameters of each wind turbine at each time moment are optimized using the Bayesian optimization algorithm to obtain the yaw angle of each wind turbine when the wind farm generates the most electricity, which is the optimal wind turbine coordinated yaw control strategy.

[0012] 6) The working conditions of a real wind farm are simulated by a yaw control model to verify the power effect of the wind farm under the optimal wind turbine coordinated yaw control strategy.

[0013] Furthermore, the fan layout data includes the number of fans, the fan arrangement method, and the spacing between fans.

[0014] Furthermore, based on the wind farm model, the two-dimensional coordinates of each wind turbine within the wind farm are calculated to obtain the spatial coordinates of each wind turbine.

[0015] Further, in step 3), the wind turbine model includes setting the rotor surface radius, yaw speed, hub height and power curve of the selected wind turbine prototype, and setting the initial yaw angle according to the wind farm inflow conditions;

[0016] The wind turbine wake model includes a velocity loss model and a turbulence intensity model. The velocity loss model adopts a yaw wake model based on a two-dimensional Gaussian model to obtain the wake velocity distribution of the wind turbine under yaw conditions. The turbulence intensity model adopts the Crespo-Hernandez wake turbulence model to calculate the change in wind field turbulence intensity distribution caused by the wind turbine wake effect.

[0017] Further, in step 4), the spatial distribution of wind turbines in the wind turbine wake model, wind turbine unit model and wind farm model are combined to obtain the wake velocity of each wind turbine at the downstream wind turbine in the wind farm model. The incoming flow velocity of each wind turbine under the influence of the wake is calculated by superimposing the model. By establishing a wind farm collaborative yaw control model, the inflow wind speed data at the hub of each wind turbine in the wind farm at any yaw angle are obtained.

[0018] Furthermore, step 5) specifically includes the following steps:

[0019] a) Determine the relevant parameters involved in the real-time coordinated yaw control of the wind farm, including: the time step of yaw target update, the yaw action speed of the yaw target and the yaw action range of the yaw target. Each wind turbine in the wind farm has any yaw angle within the yaw action range. Integrating the yaw angles of all wind turbines together is a coordinated yaw control scheme for a wind farm.

[0020] b) Randomly assign an initial yaw angle to each wind turbine;

[0021] c) Calculate the overall power generation of the wind farm based on the power curve function and establish the yaw control matrix of the wind farm;

[0022] d) Model the power output function f using Gaussian process regression:

[0023]

[0024] In the formula, K is the covariance matrix, K ij =k(x i x j ) corresponds to the (i, j)th input and k T =(k(x) 1 ,x),...,k(x) n ,x));Noise variance Used for quantizing power output y 1：n ={y 1 , ..., y n The noise level in};

[0025] e) Use the Gaussian process averaging function μ(x|D) n ) and variance function σ 2 (x|D n Select the wind turbine action x for the next iteration. n+1 :

[0026] x n+1 =arg max(μ(x|D n )+ρ n σ 2 (x|D n ));

[0027] In the formula, the parameter ρ n Used to balance search efficiency and search quality in selection functions.

[0028] f) Stop iterating when the number of calculations reaches the set value; otherwise, repeat steps c)-e) to obtain the optimal wind turbine coordinated yaw control strategy.

[0029] Furthermore, the process of calculating the overall power generation of a wind farm based on the power curve is as follows:

[0030] For a given wind speed U and wind direction θ W The yaw angle of the wind farm cluster is considered, and the wake interaction between the turbines is taken into account. The power generation P of each turbine is calculated using power curves. i Then the overall power generation of the wind farm can be represented by y:

[0031]

[0032] In the formula, x = (x1, ..., x i ,…,x N ), x i =γ i P represents the yaw action of the target wind turbine i; i The power of the i-th fan can be expressed as the local inflow v. i The function is determined by the selected wind turbine prototype.

[0033] Compared with existing technologies, this invention has the following advantages: Addressing the problem that conventional real-time wind chasing control methods for wind farms neglect the influence of turbine wake effects and fail to achieve optimal overall power generation, this invention proposes establishing a wind farm collaborative yaw control model. This model considers the impact of upstream turbine wakes on downstream turbines, thereby reducing wind power generation losses. Simultaneously, a Bayesian optimization algorithm is used to efficiently solve the yaw control optimization problem, meeting the high efficiency requirements of real-time collaborative control. Through this real-time collaborative yaw control method, better wind farm power generation can be obtained than with traditional control methods. It effectively provides the optimal turbine control scheme for real-time wind farm operation, enabling real-time optimization based on changes in the wind environment within the wind farm, improving power generation efficiency, and thus achieving greater wind farm power generation, maximizing the overall wind farm power generation. Attached Figure Description

[0034] Figure 1 This is a flowchart of the real-time coordinated yaw control method for wind farms disclosed in this invention.

[0035] Figure 2 Flowchart of Bayesian optimization of wind farm cooperative yaw control;

[0036] Figure 3 This is a simulation-based wind turbine layout diagram of a wind farm in the embodiment.

[0037] Figure 4 This is a diagram illustrating the effect of greedy control in a wind farm in the embodiment.

[0038] Figure 5 This is a diagram illustrating the effect of wind farm coordinated yaw control in the embodiment.

[0039] Figure 6 The diagram shows the power generation efficiency of each exhaust fan under different control strategies in the embodiment. Detailed Implementation

[0040] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0041] Example: See Figures 1 to 6 A real-time coordinated yaw control method for wind farms includes the following steps:

[0042] 1) Obtain measured wind direction and speed data and wind turbine layout data for the wind farm. Specifically, the measured wind direction and speed data for the target wind farm includes historical wind turbine inflow data (i.e., historical wind direction and speed data) and real-time generated wind direction and speed data; during implementation, wind direction and speed are collected by an anemometer. The wind turbine layout data includes the number of wind turbines, the arrangement method of the wind turbines (such as tandem, staggered, or array arrangement), and the spacing between the wind turbines (including longitudinal and lateral spacing).

[0043] 2) Based on the wind turbine layout data, a two-dimensional coordinate-based wind farm model is established to simulate the spatial distribution of the wind farm. Specifically, using the wind turbine layout data as a reference, the geographical coordinates of the wind farm are transformed into a two-dimensional coordinate system with a specified wind turbine as the reference (origin), thus establishing a two-dimensional coordinate-based wind farm model. Then, based on this wind farm model and the wind turbine layout data, the two-dimensional coordinates of each wind turbine within the wind farm are calculated to obtain the spatial coordinates of each wind turbine in the two-dimensional planar wind farm model, thereby accurately simulating the spatial distribution of the wind farm.

[0044] 3) Establish wind turbine unit model and wind turbine wake model. The wind turbine unit model includes setting the rotor radius, yaw speed, hub height, and power curve based on the selected wind turbine prototype, and setting the initial yaw angle according to the wind field inflow conditions. The wind turbine wake model includes a velocity loss model and a turbulence intensity model. The velocity loss model adopts a yaw wake model based on a two-dimensional Gaussian model to obtain the wake velocity distribution of the wind turbine under yaw conditions. The turbulence intensity model adopts the Crespo-Hernandez wake turbulence model to calculate the changes in wind field turbulence intensity distribution caused by the wind turbine wake effect.

[0045] 4) Combine the wind farm model with the wind turbine model and the wind turbine wake model to establish a wind farm collaborative yaw control model. Specifically, combine the wind turbine wake model, the wind turbine model, and the spatial distribution of wind turbines in the wind farm model to obtain the wake velocity of each wind turbine at the downstream wind turbine in the wind farm model. Calculate the incoming flow velocity of each wind turbine under the influence of the wake using a (sum-squared) superposition model. By establishing the wind farm collaborative yaw control model, obtain the inflow wind speed data at the hub of each wind turbine in the wind farm at any yaw angle. The specific process is as follows:

[0046] Since the simulation effect of the wake model has a significant impact on the optimization effect of yaw control, a yaw wake model based on a two-dimensional Gaussian model is selected to establish the wind farm cooperative yaw control model.

[0047] The wake velocity distribution of the wind turbine under yaw conditions is as follows:

[0048]

[0049] In the formula, and are the wind speeds at the observation point and the incoming flow respectively, C T is the thrust coefficient of the wind turbine, d is the impeller diameter of the wind turbine, γ is the yaw angle of the wind turbine, δ is the deflection distance of the wake center at each downwind position, z h is the hub height of the wind turbine, y and z are the distances of the observation point from the wind turbine in the spanwise and vertical directions; σ y and σ z are the extension widths of the wake in the spanwise and vertical directions, and their calculation formulas are:

[0050]

[0051] In the formula, x is the distance of the observation point from the wind turbine in the downwind direction, k y and k z are the wake growth rates in the spanwise and vertical directions respectively.

[0052] According to the existence of the potential core region of the wind turbine wake, the wake region is divided into the near-wake region and the far-wake region, and different formulas are used for calculation. Specifically, the calculation formula for the normalized length x0 / d of the potential core is:

[0053]

[0054] In the formula, α * = 2.320, β * = 0.154; I is the turbulence intensity of the incoming flow.

[0055] In the near-wake region (x < x0), the normalized distance where The calculation formula of is:

[0056]

[0057] In the far-wake region (x > x0), the calculation formula for the normalized distance δ / d is:

[0058]

[0059] When x approaches infinity, the wake deflection gradually becomes:

[0060]

[0061] The sum of squares (SS) superposition model is used to calculate the influence of the wind turbine by the wakes of multiple wind turbines, which is expressed as:

[0062]

[0063] In the formula, vi It is the incoming flow velocity of target wind turbine i located within the wake superposition region of upstream wind turbine j, v j It is the incoming flow velocity of the upstream fan j, v ij It is the wake velocity of upstream wind turbine j at the target wind turbine i.

[0064] According to formula (1-6), the downstream wake velocity distribution of each wind turbine in the wind farm at any yaw angle can be obtained; by introducing the established two-dimensional coordinate model of the wind farm, the incoming flow velocity v of each wind turbine under the influence of the wake can be calculated by formula (7). i , where v i Let be the inflow velocity of the i-th fan, which is the actual wind speed data considering the wake effect.

[0065] 5) Based on the real-time wind direction and speed data collected from the wind farm, the yaw angle parameters of each wind turbine at each time point are optimized using a Bayesian optimization algorithm to obtain the yaw angle of each wind turbine when the wind farm's power generation is maximized, i.e., the optimal wind turbine coordinated yaw control strategy. As an example, the initial wind turbine control strategy is a random yaw angle control strategy. The Bayesian optimization algorithm is used to iteratively optimize the random yaw angle strategy and update the real-time control dataset until the number of iterations reaches a set value (set number of iterations), at which point the optimal wind turbine coordinated yaw control strategy is output. In practice, as a special case, the initial wind turbine control strategy can also be a greedy control strategy.

[0066] 6) Simulate the actual operation of a wind farm using a yaw control model to verify the power output of the wind farm under the optimal wind turbine coordinated yaw control strategy. If the verification is successful, meaning that the power generation of each wind turbine reaches its maximum value under the optimal wind turbine coordinated yaw control strategy, or the total power generation under the optimal wind turbine coordinated yaw control strategy is greater than the total power generation under the greedy control strategy, then the verified optimal wind turbine coordinated yaw control strategy is used to control the wind turbines in the actual wind farm; otherwise, repeat steps 5) and 6) until the verification is successful.

[0067] Specifically, in step 5), in order to establish a regression model of the power output function f(x) and find control measures to achieve the maximum power output of the wind farm, each iteration of Bayesian optimization (BO) includes two stages: learning and optimization. Specifically, it includes the following steps:

[0068] a) Determine the relevant parameters involved in the real-time coordinated yaw control of the wind farm, including: the time step of yaw target update, the yaw action speed of the yaw target and the yaw action range of the yaw target. Each wind turbine in the wind farm has any yaw angle within the yaw action range. Integrating the yaw angles of all wind turbines together is a coordinated yaw control scheme for a wind farm.

[0069] b) Yaw action initialization: The initial yaw scheme is given randomly, and each wind turbine takes any value within the allowable range of yaw action:

[0070] x l ≤x≤x u (8)

[0071] In the formula, x l and x u These are the lower and upper limits of the fan control actions, respectively;

[0072] c) Calculate overall power generation based on power curve function: Calculate the overall power generation of the wind farm based on the power curve function and establish the wind farm yaw control matrix; specifically:

[0073] For a given wind speed U and wind direction θ W The yaw angle of the wind farm cluster is considered, and the power generation P of each wind turbine is calculated using the power curve function method by taking into account the wake interaction between the turbines. i Then the overall power generation of the wind farm can be represented by y:

[0074]

[0075] In the formula, x = (x1, ..., x i ,…,x N ), x i =γ i P represents the yaw action of the target wind turbine i; i The power of the i-th wind turbine can be expressed as the induction factor α. i yaw angle γ i and incoming flow v i The function, that is:

[0076]

[0077] In the formula, ρ is the air density, A is the rotor area, and C is the power coefficient. P The formula for calculating (α,γ) is:

[0078] C p (α,γ)=4α(1-α) 2 ×0.77×(cosγ) 1.88 (11)

[0079] Based on the calculated overall power generation y and the corresponding wind turbine yaw action x = (x1, ..., x i ,…,x N Establish the yaw control matrix for the wind farm:

[0080] [D] n={(x i ,y i )|i=1,...,n}.

[0081] d) Modeling the power output function based on Gaussian process regression: In the learning phase of the nth iteration of Bayesian optimization, the yaw control matrix D of the wind farm from previous iterations is used. n ={(x i ,y i )|i=1,…,n}, using a Gaussian distribution d(x)~N(μ(x|D n ),σ 2 (x|D n Determine the posterior distribution of the power output function f(x) for an unknown input x, and model the power output function f:

[0082]

[0083] In the formula, K is the covariance matrix, K ij =k(x i ,x j ) corresponds to the (i,j)th input and k T =(k(x) 1 ,x),…,k(x n ,x));Noise variance Used for quantizing power output y 1:n ={y 1 ,…,y n The noise level in}; its mean and variance functions are expressed as follows:

[0084]

[0085]

[0086] e) Selecting a new yaw action based on the UCB function: In the nth iteration of Bayesian optimization, the Gaussian process average function μ(x|D) is used. n ) and variance function σ 2 (x|D n Select the wind turbine action x for the next iteration. n+1 The function is obtained using the maximum confidence upper bound (UCB), as shown in the following formula:

[0087] x n+1 =arg max(μ(x|D n )+ρ n σ 2 (x|D n (15)

[0088] In the formula, the parameter ρ nUsed to balance search efficiency and search quality in selection functions.

[0089] f) Convergence determination: Stop iteration when the number of calculations reaches the set value; otherwise, repeat steps c)-e) to obtain the optimal wind turbine coordinated yaw control strategy.

[0090] g) Output the optimal wind turbine coordinated yaw control strategy.

[0091] The optimal wind turbine coordinated yaw control strategy obtained through this scheme is used to calculate relevant parameters of the wind farm, including the real-time overall power generation curve of the wind farm and the power generation of each wind turbine.

[0092] As a specific embodiment:

[0093] This study uses a simulated wind farm turbine layout as the research object and employs Vestas V-80 2MW wind turbines to investigate the wind farm control performance. The turbine rotor diameter is 80m and the hub height is 70m. The turbine's cut-in, cut-out, and rated wind speeds are 4, 25, and 15 m / s, respectively. The wind farm layout is as follows: Figure 3 As shown, 25 wind turbines were arranged in series, with the front-to-back spacing Sx maintained at 4D and the lateral spacing Sy maintained at 2D, to test the effectiveness of the yaw control strategy.

[0094] This invention employs a real-time cooperative yaw control method for wind farms, combining the spatial distribution of the wind farm with the wind turbine wake model to establish a cooperative yaw control model. Then, based on measured wind speed and direction data from the wind farm's inflow, both a traditional greedy control strategy and a Bayesian optimized cooperative yaw strategy are used to obtain different wind turbine yaw control effect diagrams, such as… Figure 4 , 5 As shown, where Figure 4 This is a diagram illustrating the effect of a traditional greedy control strategy. Figure 5 The diagram shows the effect of the optimal cooperative yaw strategy after Bayesian optimization.

[0095] The power curve function was used to verify and compare the wind farm power generation efficiency (current power generation / rated power) of the yaw control scheme, as shown in Table 1. Table 1 shows that, compared with the traditional greedy control strategy, the proposed Bayesian optimized cooperative yaw strategy can increase the overall power generation of the wind farm by approximately 0.98%, effectively mitigating the loss of wind power generation due to the wake effect and significantly improving the economic efficiency of the wind farm.

[0096] Table 1 Comparison of wind farm power generation efficiency

[0097]

[0098] The power generation statistics of each wind turbine in the wind farm are as follows: Figure 6As shown, due to the large number of wind turbines, the power generation of each row of turbines was combined during the statistical analysis, and the changes in power generation across the five rows of turbines were compared. Figure 6 It can be seen that although the Bayesian optimization cooperative yaw strategy reduces the power generation of the front wind turbines, it reduces the wind speed loss caused by the wake effect in the wind farm as a whole, and significantly increases the power generation of the rear wind turbines, thereby improving the overall power generation efficiency of the wind farm.

[0099] Traditional wind farm control strategies neglect the power generation reduction caused by the wake effect of wind turbines. This invention proposes to establish a cooperative yaw control model that considers the wake model. By coordinating yaw control among the wind turbines in the wind farm, the impact of the wake effect on the power generation of downstream wind turbines is reduced. At the same time, a Bayesian optimization algorithm is used to achieve efficient optimization calculation when solving for the optimal solution of yaw control, thus meeting the high efficiency requirements of real-time cooperative yaw control in wind farms.

[0100] The yaw control method proposed in this invention can achieve a greater wind farm power generation than traditional methods, and is very suitable for coordinated yaw control during wind farm operation.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit the technical solutions. Those skilled in the art should understand that any modifications or equivalent substitutions to the technical solutions of the present invention without departing from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A real-time coordinated yaw control method for wind farms, characterized in that, Includes the following steps: 1) Obtain measured wind direction and speed data and wind turbine layout data for the wind farm; 2) Based on the wind turbine layout data, establish a wind farm model based on two-dimensional coordinates to simulate the spatial distribution of the wind farm; 3) Establish wind turbine unit model and wind turbine wake model; the wind turbine unit model includes setting the rotor surface radius, yaw speed, hub height and power curve according to the selected wind turbine prototype, and setting the initial yaw angle according to the wind field inflow conditions; The wind turbine wake model includes a velocity loss model and a turbulence intensity model. The velocity loss model adopts a yaw wake model based on a two-dimensional Gaussian model to obtain the wake velocity distribution of the wind turbine under yaw conditions. The turbulence intensity model adopts the Crespo-Hernandez wake turbulence model to calculate the change in wind field turbulence intensity distribution caused by the wind turbine wake effect. 4) Combine the wind farm model with the wind turbine model and the wind turbine wake model to establish a wind farm collaborative yaw control model; combine the wind turbine wake model, the wind turbine model and the spatial distribution of wind turbines in the wind farm model to obtain the wake velocity of each wind turbine at the downstream wind turbine in the wind farm model, and calculate the incoming flow velocity of each wind turbine under the influence of the wake by superimposing the model; by establishing a wind farm collaborative yaw control model, obtain the inflow wind speed data at the hub of each wind turbine in the wind farm at any yaw angle; 5) Based on the real-time wind direction and speed data collected from the wind farm, the yaw angle parameters of each wind turbine at each time point are optimized using a Bayesian optimization algorithm to obtain the yaw angle of each wind turbine when the wind farm generates the maximum power, i.e., the optimal wind turbine coordinated yaw control strategy; specifically including the following steps: a) Determine the relevant parameters involved in the real-time coordinated yaw control of the wind farm, including: the time step of yaw target update, the yaw action speed of the yaw target and the yaw action range of the yaw target. Each wind turbine in the wind farm has any yaw angle within the yaw action range. Integrating the yaw angles of all wind turbines together is a coordinated yaw control scheme for a wind farm. b) Randomly assign an initial yaw angle to each wind turbine; c) Calculate the overall power generation of the wind farm based on the power curve function and establish the wind farm yaw control matrix; the process of calculating the overall power generation of the wind farm based on the power curve is as follows: Given a wind speed U and wind direction The yaw angle of the wind farm cluster is considered, and the power generation of each wind turbine is calculated using power curves by taking into account the wake interaction between the turbines. The overall power generation capacity of the wind farm is used To indicate: ； In the formula, , This indicates the yaw action of the target wind turbine i; The power of the i-th wind turbine is expressed as the induction factor. Yaw angle and incoming flow The function, that is: ； In the formula, air density, Rotor area, power coefficient The calculation formula is: ； d) Apply Gaussian process regression to the power output function Modeling: ； In the formula, It is the covariance matrix. Corresponding to the input and noise variance Used for quantizing power output The noise level in; e) Use the Gaussian process averaging function Sum of variance functions Select the wind turbine action for the next iteration. : ； In the formula, the parameter Used to balance search efficiency and search quality in selection functions; f) Stop iterating when the number of calculations reaches the set value; otherwise, repeat steps c)-e) to obtain the optimal wind turbine coordinated yaw control strategy. 6) Simulate the actual working conditions of a wind farm using a yaw control model to verify the power effect of the wind farm under the optimal wind turbine coordinated yaw control strategy. If the verification is successful, that is, the power generation of each wind turbine reaches the maximum value under the optimal wind turbine coordinated yaw control strategy, or the total power generation under the optimal wind turbine coordinated yaw control strategy is greater than the total power generation under the greedy control strategy, then the verified optimal wind turbine coordinated yaw control strategy is used to control the wind turbines of the actual wind farm; otherwise, repeat steps 5) and 6) until the verification is successful.

2. The real-time coordinated yaw control method for wind farms according to claim 1, characterized in that, The fan layout data includes the number of fans, the arrangement of fans, and the spacing between fans.

3. A real-time coordinated yaw control method for wind farms according to claim 1 or 2, characterized in that, Based on the wind farm model, the two-dimensional coordinates of each wind turbine in the wind farm are calculated to obtain the spatial coordinates of each wind turbine.

Images