Multi-machine cooperative yaw control method and device considering dynamic wind direction and medium

Abstract

The invention relates to a multi-machine cooperative yaw control method and device considering a dynamic wind direction and a medium. The multi-machine cooperative yaw control method comprises the steps that S1, the dynamic relation between the wind direction and a wake flow area is established based on a Gaussian-curl hybrid model; s2, constructing a single-machine yaw optimization model based on model predictive control, optimizing single-machine yaw by taking maximum generating capacity of a wind power plant and reduction of fatigue load as optimization objectives, and obtaining a single-machine optimal yaw angle through rolling optimization and feedback correction; and S3, a serial optimization method is adopted to carry out multi-round iterative optimization on one fan by one fan, and an optimal multi-fan cooperative yaw control result is obtained. Compared with the prior art, the method has the advantages that higher total generating capacity and more balanced fatigue load can be realized, so that the whole wind power plant reaches the optimal state of power and load in a large range.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power yaw control, and in particular to a multi-machine coordinated yaw control method, equipment and medium taking into account dynamic wind directions. Background Art

[0002] When a wind turbine is running, the rotor extracts energy from the airflow, which causes the downstream airflow speed to decrease and form turbulence. This phenomenon is called the wake effect. In large wind farms, the wake generated by the upstream unit causes a power loss of 10% to 20%. In order to reduce the impact of the wake, active wake control is becoming increasingly important, and the main methods include yaw control and pitch control. Among them, yaw control has become the main control method because of its large offset range in the wake area and its significant effect in reducing the impact of the wake.

[0003] Traditional yaw control mainly follows the single-machine maximum power tracking strategy (MPPT), that is, each wind turbine is an independent individual, and the wind turbine direction is uniformly adjusted through yaw to achieve the design value on the power curve. Dipesh Kumar et al. compared the commonly used MPPT control algorithms: indirect power control, direct power control and hybrid control, and found that the adaptive MPPT control algorithm has higher efficiency.

[0004] However, as the scale of wind farms increases, the MPPT strategy leads to an intensification of the wake effect of upstream units. In order to solve this problem, Xu Yang et al. established a multi-machine equivalent model under the influence of wake and used the particle swarm optimization algorithm (PSO) for unified yaw control; Cai Wei et al. proposed a wind turbine yaw angle optimization method based on wake decoupling and clustering, and independently optimized multiple groups in parallel; DOU et al. used the covariance matrix adaptive evolutionary algorithm based on the three-dimensional yaw wake model to optimize the unit yaw angle and achieve an increase in wind farm power generation. The above analysis shows that yaw control considering wake can reduce wind farm power loss, but the feasibility of the proposed optimization method is verified under fixed wind direction, and the influence of wind direction volatility and randomness on the yaw effect is not considered.

[0005] Existing studies on wind direction modeling in yaw control can be divided into statistical methods and data-driven methods according to the mechanism. In terms of statistics, Erdem et al. predicted wind direction based on the autoregressive moving average model (ARMA), and Masseran et al. proposed a method based on a finite mixture of von-Mises distribution to achieve short-term prediction of wind direction. However, statistical methods have the problem of large data requirements and are not suitable for nonlinear wind directions. In terms of data-driven, Mohandes et al. and Zhou et al. respectively used support vector machines and least squares support vector machines (LSSVM) to achieve wind direction prediction. Since these algorithms are difficult to mine and learn the deep features of wind direction, the prediction accuracy does not meet the needs of real-time yaw optimization. Summary of the invention

[0006] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a multi-machine coordinated yaw control method, equipment and medium taking into account dynamic wind direction.

[0007] The purpose of the present invention can be achieved by the following technical solutions:

[0008] According to a first aspect of the present invention, a multi-machine coordinated yaw control method taking into account dynamic wind direction is provided, comprising:

[0009] S1. Establish the dynamic relationship between wind direction and wake area based on Gaussian-curl mixed model;

[0010] S2. Construct a single-machine yaw optimization model based on model predictive control, optimize the single-machine yaw with the optimization objectives of maximizing wind farm power generation and reducing fatigue load, and obtain the optimal yaw angle of the single machine through rolling optimization and feedback correction;

[0011] S3. Use the serial optimization method to perform multiple rounds of iterative optimization for each wind turbine to obtain the optimal multi-machine coordinated yaw control result.

[0012] Preferably, the dynamic relationship between wind direction and wake area is established based on the Gaussian-curl mixed model, specifically:

[0013] Assume that unit j is the upstream unit of unit i. When unit j is not completely aligned with the wind direction, the centerline of its wake is offset. At this time, the axial position x of unit i in the wake area of ​​unit j is:

[0014] x=Lcos(θ L -θ+0.3C T α i |) (1)

[0015] Where: L is the distance between unit i and unit j, θ L is the angle between unit i and unit j, θ is the wind direction angle, C T is the thrust coefficient, α i is the yaw angle;

[0016] The speed distribution of unit i follows:

[0017]

[0018] Where: v0 is the incoming wind speed, D is the wind wheel diameter, k a is the wake depth coefficient, k b is the wake diffusion coefficient, α i is the yaw angle, δ is the wake diffusion width;

[0019] Combining equations (1) and (2), we can obtain the velocity model of unit i in a single wake region;

[0020] For wind farms with multiple wind turbines, the wake superposition model is used to analyze the wake within the field based on the wake modeling of a single wind turbine. After the wake is superimposed, the dynamic relationship between the wind direction and the wake superposition area is obtained;

[0021] Among them, the actual wind speed of unit i is:

[0022]

[0023] Where: v j is the actual wind speed of unit j; v0 is the incoming wind speed.

[0024] Preferably, the single-machine yaw optimization model based on model predictive control specifically includes:

[0025] Control objective: Taking increasing power generation and reducing unit load as the optimization objectives, construct the multi-objective function F of unit i i :

[0026]

[0027] Where: ω is the weight factor; T C is the prediction time step; P i,t is the actual power of unit i under yaw condition; L i,t is the structural fatigue load of unit i under yaw condition;

[0028] Constraint function: fatigue load is not greater than the load in wind cut-out condition, and yaw angle is controlled within the set range.

[0029] Preferably, the actual power P of unit i under the yaw condition is i,t , the calculation expression is:

[0030]

[0031] Where: ρ is the air density, A is the swept area of ​​the fan impeller, C P is the power factor, f P is the attenuation coefficient, α i,t is the yaw angle of unit i at time t, v i,t is the actual wind speed of unit i at time t.

[0032] Preferably, the structural fatigue load L of unit i under the yaw condition is i,t , the calculation expression is:

[0033]

[0034] Where: ρ is the air density, A is the swept area of ​​the fan impeller, C P is the power coefficient, v i,t is the actual wind speed of unit i at time t.

[0035] Preferably, the constraint function: fatigue load is not greater than the load in the case of wind cut-out, and the yaw angle is controlled within a set range, specifically:

[0036]

[0037] Where: L i is the structural fatigue load of unit i under yaw condition, is the average wind speed in the current wind field, σ is the turbulence standard deviation, C T is the thrust coefficient, α i is the yaw angle of unit i under yaw condition, α min and α max are the minimum yaw angle and the maximum yaw angle respectively.

[0038] Preferably, the process of solving the optimal yaw angle of a single machine is specifically as follows:

[0039] Taking dynamic wind speed and wind direction as environmental input, and taking increasing wind farm power generation and reducing unit structural fatigue load as optimization goals, all wind turbines except unit i are fixed, and the model predictive controller is called to adjust the yaw angle α of unit i. i Control various state variables of the fan;

[0040] During the rolling optimization process, the quadratic programming solution is performed according to the optimization objective and constraint function. At each sampling moment, the optimal control sequence in the future limited time domain is obtained according to the optimization performance index at the current sampling moment.

[0041] Combined with feedback correction, the optimal yaw angle of a single machine is obtained.

[0042] Preferably, the serial optimization method is used to perform multiple rounds of iterative optimization of each wind turbine to obtain the optimal multi-machine coordinated yaw control result, specifically:

[0043] 1) Adjust the yaw angle α of unit i through rolling optimization i , and update the wake impact area K i ;

[0044] 2) Traverse all wake impact areas and update the actual wind speed of the unit based on the speed model of the unit in a single wake area;

[0045] 3) Update the yaw angle through the model predictive controller if the multi-objective function value satisfies: F i ≤F baseline and ε is the preset threshold, then the current yaw angle is output, otherwise the multi-objective function benchmark value F is updated baseline is the optimal solution in the current optimization process, return to 1).

[0046] According to a second aspect of the present invention, there is provided an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the program, any one of the methods described above is implemented.

[0047] According to a third aspect of the present invention, there is provided a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, any one of the methods described above is implemented.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention establishes the relationship between the dynamic wind direction and the wake model under the condition of incomplete wind alignment, proposes a single-machine yaw optimization based on model predictive control under multiple objectives, and uses the output results of serial optimization and multi-round iterative model predictive control to obtain the optimal yaw angle of each wind turbine. By differentiating the yaw angles of the wind turbines, the influence of the wake on the yaw optimization effect is reduced, thereby increasing the power generation of the wind farm and reducing the fatigue load of the unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a flow chart of the method of the present invention;

[0051] Figure 2 It is an illustration of the wind direction angle in the scenarios of complete facing wind and incomplete facing wind;

[0052] Figure 3 It is a schematic diagram of the relationship between power coefficient and thrust coefficient and wind speed;

[0053] Figure 4 Schematic diagram of the model predictive control framework;

[0054] Figure 5 It is a prediction timing diagram in the embodiment. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0056] Example

[0057] like Figure 1As shown, this embodiment provides a multi-machine coordinated yaw control method taking into account dynamic wind direction, including:

[0058] S1. Establish the dynamic relationship between wind direction and wake area based on Gaussian-curl mixed model;

[0059] S2. Construct a single-machine yaw optimization model based on model predictive control, optimize the single-machine yaw with the optimization objectives of maximizing wind farm power generation and reducing fatigue load, and obtain the optimal yaw angle of the single machine through rolling optimization and feedback correction;

[0060] S3. Use the serial optimization method to perform multiple rounds of iterative optimization for each wind turbine to obtain the optimal multi-machine coordinated yaw control result.

[0061] Next, the method of this embodiment is introduced in detail.

[0062] S1. The dynamic relationship between wind direction and wake area is established based on the Gaussian-curl mixed model.

[0063] In order to reduce the influence of the wake effect in the wind farm on power generation and fatigue load in the subsequent yaw control optimization process, it is necessary to accurately characterize the wake model under dynamic wind speed and wind direction.

[0064] Therefore, this embodiment establishes a wake model that takes into account the dynamic wind direction under the condition of incomplete wind direction.

[0065] A wake model based on the combination of Gaussian distribution and curl effect (Gaussian-Curl Hybrid, GCH) is used. In the application of the GCH model, when the wind turbine is not completely aligned with the wind direction, the wake will bend and deviate to the leeward side of the wind turbine. The GCH model can calculate this curling effect and can more accurately calculate the impact of the wake on the downstream wind turbine.

[0066] To increase the power generation of downstream wind turbines, the upstream units may not be completely aligned with the wind direction. Figure 2 As shown in the figure, let unit j be the upstream unit of unit i. When unit j is not completely aligned with the wind direction, the centerline of its wake will be offset, which will affect the axial position x of unit i in the wake area of ​​unit j:

[0067] x=Lcos(|θ L -θ+0.3C T α i |) (1)

[0068] Where: L is the distance between unit i and unit j, θ L is the angle between unit i and unit j, θ is the wind direction angle, C T is the thrust coefficient.

[0069] Therefore, the speed distribution of unit i follows:

[0070]

[0071] Where: v0 is the incoming wind speed, D is the wind wheel diameter, k a is the wake depth coefficient, k b is the wake diffusion coefficient, α i is the yaw angle, δ is the wake diffusion width. According to design experience

[10] , k a The value is 0.38, k b The value is 0.004.

[0072] The combined equations (1) and (2) constitute the velocity model of unit i in a single wake region.

[0073] When a wind farm contains multiple wind turbines, a wake superposition model can be used to perform wake analysis within the farm based on the wake modeling of a single wind turbine. This embodiment uses an energy conservation superposition model to describe the nonlinear effect of multiple wake interactions. i is the set of upstream units of unit i, {j∈J i} is the upstream unit index of unit i. After the wake superposition, the actual wind speed of unit i is:

[0074]

[0075] In the formula, v j is the actual wind speed of unit j. Through the above formula, the dynamic relationship between wind direction and wake superposition area can be obtained.

[0076] S2. Construct a single-machine yaw optimization model based on model predictive control. Take maximizing wind farm power generation and reducing fatigue load as optimization goals, optimize the single-machine yaw, and obtain the optimal yaw angle of the single machine through rolling optimization and feedback correction.

[0077] It is required to solve the optimal yaw angle of a single wind turbine in the wind farm. Considering the large time margin of the yaw condition, MPC is more suitable for solving the yaw angle. In this section, a single-machine yaw optimization model considering dynamic wind direction changes is established based on MPC. First, the objective function and constraints of the optimization process are determined, and then the optimal yaw angle is obtained through rolling optimization and feedback correction.

[0078] 1) Objective function

[0079] Under yaw conditions, the actual power and structural fatigue load of the unit are:

[0080]

[0081] Where ρ is the air density, A is the swept area of ​​the fan impeller, C P is the power coefficient, C P and CT The value of satisfies Figure 3 The curve shown.

[0082] In order to achieve multi-objective optimization to increase power generation and reduce unit load during yaw, the weight factor ω is used to construct the objective function F of unit i. i :

[0083]

[0084] Among them, T C is the prediction time step. Considering the impact of the yaw start procedure on the yaw speed, the duration of each yaw movement should be longer than the start time. According to design experience, the start process takes about 1s. Therefore, the prediction time step T is set C 5s.

[0085] 2) Constraints

[0086] According to IEC standards, fatigue loads should not be greater than those in wind cut-out conditions, and the yaw angle should be controlled within a certain range:

[0087]

[0088] In the formula is the average wind speed in the current wind field, and σ is the turbulence standard deviation.

[0089] 3) Solution process

[0090] The single-machine yaw optimization uses dynamic wind speed and wind direction as environmental inputs, with the goal of increasing wind farm power generation and reducing unit structural fatigue loads. All wind turbines except unit i are fixed, such as Figure 4 As shown, the MPC controller is called to adjust the yaw angle α of unit i i Control the fan state variables.

[0091] In the process of rolling optimization, the optimization target J is established, and the quadratic programming solution is performed under constraints. At each sampling moment, the optimal control sequence in the future finite time domain is obtained according to the optimization performance index at that moment. Figure 5 As shown, starting from the T time period, based on the future time interval (T+T C ,T+kT C )’s objective function prediction, implements the instructions of the first time interval, and effectively overcomes the impact of system uncertainty.

[0092] S3. Use the serial optimization method to perform multiple rounds of iterative optimization for each wind turbine to obtain the optimal multi-machine coordinated yaw control result.

[0093] According to the solved optimal yaw angle of a single machine, the serial optimization method is used to perform multiple rounds of iterations for each wind turbine to complete the coordinated control of multiple units in the field, thereby achieving the multiple goals of increasing the power generation of the wind farm and reducing fatigue loads.

[0094] Due to the large time scale of yaw control, it is difficult for MPC to achieve coordinated optimization of multiple units simultaneously. To solve this problem, this embodiment adopts a serial optimization (SR) method to approach the optimal solution through multiple rounds of iterative optimization of each wind turbine. The optimal yaw angle of a single unit obtained based on MPC will be used as input to participate in the serial optimization SR process. After adjusting each unit one by one, the optimal yaw angle of each wind turbine in the field is obtained to achieve coordinated yaw control of multiple units.

[0095] The pseudo code corresponding to the multi-round iterative optimization process of each wind turbine by the serial optimization method is as follows:

[0096] 1) Adjust α by rolling optimization i , and use FLORIS to update the wake impact area K i

[0097] 2)If k∈Ki then

[0098] 3) Update v according to formula (1) (2) k

[0099] 4)EndIf

[0100] 5) Update α based on MPC i , α i =argminF i ,

[0101] 6)If F i ≤F baseline and

[0102] 7) Output α i

[0103] 8) Else

[0104] 9) Update F baseline This is the optimal solution in the current optimization process. Return to 1)

[0105] 10)EndIf

[0106] This multi-stage iterative optimization search process can deeply explore the power improvement potential of each wind turbine under different wind speed and wind direction conditions, and carefully reduce the wake effect. Through SR iteration, this method can achieve global optimization, thereby achieving higher total power generation and more balanced fatigue load, so that the entire wind farm can achieve the optimal power and load over a large range.

[0107] The present invention establishes the relationship between the dynamic wind direction and the wake model under the condition of incomplete wind alignment, proposes a single-machine yaw optimization based on model predictive control under multiple objectives, and uses the output results of serial optimization and multi-round iterative model predictive control to obtain the optimal yaw angle of each wind turbine. By differentiating the yaw angles of the wind turbines, the influence of the wake on the yaw optimization effect is reduced, thereby increasing the power generation of the wind farm and reducing the fatigue load of the unit.

[0108] The electronic device of the present invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) or loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The CPU, ROM and RAM are connected to each other via a bus. An input / output (I / O) interface is also connected to the bus.

[0109] Multiple components in the device are connected to the I / O interface, including: input units, such as keyboards, mice, etc.; output units, such as various types of displays, speakers, etc.; storage units, such as disks, optical disks, etc.; and communication units, such as network cards, modems, wireless communication transceivers, etc. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunication networks.

[0110] The processing unit performs the various methods and processes described above, such as methods S1 to S3. For example, in some embodiments, methods S1 to S3 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the device via a ROM and / or a communication unit. When the computer program is loaded into the RAM and executed by the CPU, one or more steps of methods S1 to S3 described above may be performed. Alternatively, in other embodiments, the CPU may be configured to execute methods S1 to S3 in any other appropriate manner (e.g., by means of firmware).

[0111] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chip systems (SOCs), load programmable logic devices (CPLDs), and the like.

[0112] The program code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0113] In the context of the present invention, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0114] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A multi-machine coordinated yaw control method taking into account dynamic wind direction, characterized in that: include: S1. Establish the dynamic relationship between wind direction and wake area based on Gaussian-curl mixed model; S2. Construct a single-machine yaw optimization model based on model predictive control, optimize the single-machine yaw with the optimization objectives of maximizing wind farm power generation and reducing fatigue load, and obtain the optimal yaw angle of the single machine through rolling optimization and feedback correction; S3. Use the serial optimization method to perform multiple rounds of iterative optimization for each wind turbine to obtain the optimal multi-machine coordinated yaw control result.

2. The multi-machine coordinated yaw control method taking into account dynamic wind direction according to claim 1, characterized in that: The dynamic relationship between wind direction and wake area is established based on the Gaussian-curl mixed model, specifically: Assume that unit j is the upstream unit of unit i. When unit j is not completely aligned with the wind direction, the centerline of its wake is offset. At this time, the axial position x of unit i in the wake area of ​​unit j is: x=Lcos(θ L -θ+0.3C T a i |) (1) Where: L is the distance between unit i and unit j, θ L is the angle between unit i and unit j, θ is the wind direction angle, C T is the thrust coefficient, α i is the yaw angle; The speed distribution of unit i follows: Where: v0 is the incoming wind speed, D is the wind wheel diameter, k a is the wake depth coefficient, k b is the wake diffusion coefficient, α i is the yaw angle, δ is the wake diffusion width; Combining equations (1) and (2), we can obtain the velocity model of unit i in a single wake region; For wind farms with multiple wind turbines, the wake superposition model is used to analyze the wake within the field based on the wake modeling of a single wind turbine. After the wake is superimposed, the dynamic relationship between the wind direction and the wake superposition area is obtained; Among them, the actual wind speed of unit i is: Where: v j is the actual wind speed of unit j; v0 is the incoming wind speed.

3. The multi-machine coordinated yaw control method taking into account dynamic wind direction according to claim 2, characterized in that: The single-machine yaw optimization model based on model predictive control specifically includes: Control objective: To increase power generation and reduce unit load as the optimization objectives, construct a multi-objective function F for unit i i : Where: ω is the weight factor; T C is the prediction time step; P i,t is the actual power of unit i under yaw condition; L i,t is the structural fatigue load of unit i under yaw condition; Constraint function: fatigue load is not greater than the load in wind cut-out condition, and yaw angle is controlled within the set range.

4. The multi-machine coordinated yaw control method taking into account dynamic wind direction according to claim 3, characterized in that: The actual power P of unit i under the yaw condition i,t , the calculation expression is: Where: ρ is the air density, A is the swept area of ​​the fan impeller, C P is the power factor, f P is the attenuation coefficient, α i,t is the yaw angle of unit i at time t, v i,t is the actual wind speed of unit i at time t.

5. The multi-machine coordinated yaw control method taking into account dynamic wind direction according to claim 3, characterized in that: The structural fatigue load L of unit i under the yaw condition i,t , the calculation expression is: Where: ρ is the air density, A is the swept area of ​​the fan impeller, C P is the power coefficient, v i,t is the actual wind speed of unit i at time t.

6. The multi-machine coordinated yaw control method taking into account dynamic wind direction according to claim 3, characterized in that: The constraint function is: the fatigue load is not greater than the load in the wind cut-out condition, and the yaw angle is controlled within the set range, specifically: Where: L i is the structural fatigue load of unit i under yaw condition, is the average wind speed in the current wind field, σ is the turbulence standard deviation, C T is the thrust coefficient, α i is the yaw angle of unit i under yaw condition, α min and α max are the minimum yaw angle and the maximum yaw angle respectively.

7. The multi-machine coordinated yaw control method taking into account dynamic wind direction according to claim 3, characterized in that: The specific process of solving the optimal yaw angle of a single machine is as follows: Taking dynamic wind speed and wind direction as environmental input, and taking increasing wind farm power generation and reducing unit structural fatigue load as optimization goals, all wind turbines except unit i are fixed, and the model predictive controller is called to adjust the yaw angle α of unit i. i Control various state variables of the fan; During the rolling optimization process, the quadratic programming solution is performed according to the optimization objective and constraint function. At each sampling moment, the optimal control sequence in the future limited time domain is obtained according to the optimization performance index at the current sampling moment. Combined with feedback correction, the optimal yaw angle of a single machine is obtained.

8. The multi-machine coordinated yaw control method taking into account dynamic wind direction according to claim 3, characterized in that: The serial optimization method is used to perform multiple rounds of iterative optimization of each wind turbine to obtain the optimal multi-machine coordinated yaw control result, specifically: 1) Adjust the yaw angle α of unit i through rolling optimization i , and update the wake impact area K i ; 2) Traverse all wake impact areas and update the actual wind speed of the unit based on the speed model of the unit in a single wake area; 3) Update the yaw angle through the model predictive controller if the multi-objective function value satisfies: F i ≤F baseline and ε is the preset threshold, then the current yaw angle is output, otherwise the multi-objective function benchmark value F is updated baseline is the optimal solution in the current optimization process, return to 1).

9. An electronic device comprising a memory and a processor, wherein a computer program is stored in the memory, wherein: When the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images