Wind turbine cluster collaborative optimization control method, electronic equipment and program product

Abstract

The invention provides a wind turbine cluster collaborative optimization control method, electronic equipment, a readable storage medium and a computer program product. The method comprises the following steps: acquiring look-ahead information representing a wind condition in a future time period; based on the look-ahead information, constructing a model prediction control problem including a plurality of optimization targets; solving the model prediction control problem to obtain an optimal control instruction of each wind turbine; and the optimal control instruction is issued to the corresponding wind turbines to be executed.

Description

Technical Field

[0001] This disclosure relates to a collaborative optimization control method, control system, electronic equipment, storage medium, and program product for wind turbine clusters. Background Technology

[0002] In the operation of large-scale wind farms, the wake effect between wind turbine clusters not only causes significant power generation losses, but also leads to higher fatigue loads on key components of downstream units and causes drastic fluctuations in the total output power of the entire farm, which seriously restricts the economic efficiency, equipment lifespan and grid friendliness of wind farms.

[0003] To address these challenges, existing technologies attempt to incorporate sensing methods such as lidar for collaborative control. For example, Chinese patent CN108953060B proposes using a forward-mounted lidar to measure wind direction information and combine it with a wake model to calculate a yaw correction factor, thereby achieving field-level yaw control. Chinese patent CN119532106B, on the other hand, deploys Doppler lidar to collect real-time wind speed, wind direction, and turbulence data, establishes a wake influence model that includes attenuation and diffusion coefficients, and adjusts the operating status of upstream units accordingly to optimize power generation efficiency.

[0004] Existing technologies generally suffer from four limitations: First, they employ simplified or steady-state wake models, making it difficult to accurately reflect the complex coupling between dynamic wind fields and wakes; second, control relies on current measurement data, lacking forward-looking utilization of future wind conditions; third, optimization is limited to single units or local coordination, making it difficult to achieve global optimization across the entire wind farm; and fourth, they have a singular objective, failing to coordinate optimization across multiple performance dimensions such as power generation efficiency, load accumulation, and power fluctuations. These problems make it difficult for existing technologies to systematically improve the overall operational performance of wind farms. Summary of the Invention

[0005] This disclosure provides a collaborative optimization control method, control system, electronic equipment, storage medium, and program product for wind turbine clusters.

[0006] According to one aspect of this disclosure, a collaborative optimization control method for a wind turbine cluster is provided, comprising the following steps: acquiring forward-looking information characterizing wind conditions in future periods; constructing a model predictive control problem containing multiple optimization objectives based on the forward-looking information; solving the model predictive control problem to obtain the optimal control command for each wind turbine; and issuing the optimal control command to the corresponding wind turbine for execution.

[0007] According to one technical solution, by acquiring forward-looking information on future wind conditions and constructing a multi-objective model to predict and control the problem, it is possible to achieve advanced collaborative optimization at the wind farm level. By solving this problem and generating and issuing optimal control commands, it is possible to effectively suppress wind turbine load and power fluctuations while increasing total power generation.

[0008] According to at least one embodiment of the control method of this disclosure, the acquisition of forward-looking information characterizing wind conditions in future periods includes: acquiring three-dimensional wind field data for future periods through remote sensing measurement equipment deployed upstream of the wind farm.

[0009] According to one technical solution, by acquiring three-dimensional wind field data for future periods through remote sensing measurement equipment deployed upstream of the wind farm, it is possible to obtain forward-looking information to characterize future wind conditions.

[0010] According to at least one embodiment of the control method of this disclosure, the construction of a model predictive control problem including multiple optimization objectives includes: predicting the total power generation of the wind farm, the load of key components of each wind turbine, and the total output power fluctuation in a future period based on the prospective information; and taking the maximization of the total power generation, the minimization of the load of the key components, and the minimization of the total output power fluctuation as the multiple optimization objectives.

[0011] According to one technical solution, by predicting the total power generation, key component load, and total output power fluctuations based on forward-looking information, and taking these three as the targets for maximization or minimization, it is possible to achieve synergistic optimization of power generation efficiency, equipment lifespan, and grid friendliness.

[0012] According to at least one embodiment of the control method of this disclosure, the prediction of the total power generation of the wind farm, the load of key components of each wind turbine, and the fluctuation of total output power in a future period includes: using a dynamic wake model to calculate the inflow wind speed of each wind turbine based on the three-dimensional wind farm data, and making predictions based on the inflow wind speed.

[0013] According to one technical solution, by using a dynamic wake model to calculate and predict the inflow wind speed based on three-dimensional wind field data, it is possible to accurately obtain the total power generation of the wind farm, the load of key components of each wind turbine, and the fluctuation of total output power in the future.

[0014] According to at least one embodiment of the control method of this disclosure, obtaining the optimal control command for each wind turbine includes: obtaining a yaw angle command sequence and a pitch angle command sequence for each wind turbine in a future time period; and taking the command corresponding to the current time in the command sequence as the optimal control command.

[0015] According to one technical solution, by generating a sequence of yaw angle and pitch angle commands for future time periods and taking the current command as the optimal control command, it is possible to achieve roll optimization execution under model predictive control.

[0016] According to at least one embodiment of the control method of this disclosure, after the optimal control command is issued to each corresponding wind turbine for execution, the method further includes: collecting actual operation feedback data of each wind turbine; and performing online correction on the prediction model used to construct the model predictive control problem based on the actual operation feedback data.

[0017] According to one technical solution, by collecting actual operational feedback data and performing online correction on the prediction model, the prediction accuracy and control robustness of the model can be continuously improved.

[0018] According to another aspect of this disclosure, a wind turbine cluster collaborative optimization control system is provided, comprising: an information acquisition unit for acquiring forward-looking information characterizing wind conditions in future periods; a problem construction unit for constructing a model predictive control problem containing multiple optimization objectives based on the forward-looking information; an instruction generation unit for solving the model predictive control problem to obtain the optimal control instruction for each wind turbine; and an instruction execution unit for issuing the optimal control instruction to the corresponding wind turbine for execution.

[0019] According to another aspect of this disclosure, an electronic device is provided, including a processor and a memory, the memory storing a computer program that, when executed by the processor, implements a control method according to any embodiment of this disclosure.

[0020] According to another aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the control method of any embodiment of this disclosure.

[0021] According to another aspect of this disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements a control method according to any embodiment of this disclosure. Attached Figure Description

[0022] The accompanying drawings illustrate exemplary embodiments of the present disclosure and, together with the description thereof, serve to explain the principles of the present disclosure. These drawings are included to provide a further understanding of the present disclosure and are incorporated in and constitute a part of this specification.

[0023] Figure 1 This is a schematic diagram illustrating a control method according to one embodiment of the present disclosure, showing an application scenario.

[0024] Figure 2 This is a schematic interactive flowchart of a control method according to one embodiment of the present disclosure.

[0025] Figure 3 This is a detailed flowchart of a specific embodiment of the method disclosed herein.

[0026] Figure 4 This is a comparison chart of wind farm power output under conventional independent control and collaborative control under a specific implementation method.

[0027] Figure 5 This is a comparison diagram of the fatigue load of the tower base bending moment under conventional control and system control of a specific implementation method. Detailed Implementation

[0028] The present disclosure will now be described in further detail with reference to the accompanying drawings and examples. It should be understood that the specific examples described herein are for illustrative purposes only and are not intended to limit the scope of the disclosure. Furthermore, it should be noted that, for ease of description, only the parts relevant to the present disclosure are shown in the accompanying drawings.

[0029] It should be noted that, where there is no conflict, the embodiments and features described in this disclosure can be combined with each other. The technical solutions of this disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0030] In the actual operation of large-scale wind farms, wind conditions exhibit strong randomness and rapid time-varying characteristics. Through careful research, the inventors discovered that existing collaborative control methods, lacking effective utilization of future wind conditions, result in control commands lagging significantly behind the dynamic changes in the wind farm under typical operating conditions such as gusts and sudden wind direction changes. This not only fails to effectively avoid wake impacts but may also exacerbate load impacts and power oscillations in downstream units due to erroneous scheduling. Furthermore, while pursuing high power generation, neglecting the active suppression of fatigue damage to key wind turbine components (such as blade roots and tower bases) will significantly shorten equipment lifespan; and drastic fluctuations in overall power will increase the grid's frequency regulation burden, limiting wind power grid-connected capacity. These interdependent and even conflicting objectives are difficult to resolve through traditional single-objective or reactive control strategies.

[0031] To this end, this application proposes a collaborative optimization control method for wind turbine clusters. Its core is to: acquire forward-looking information representing wind conditions in future periods, construct and continuously solve a model predictive control problem involving multiple objectives, thereby generating and executing optimal control commands that can be globally coordinated and respond in advance, and continuously optimize control performance through an online feedback correction mechanism.

[0032] Figure 1 A schematic diagram of an application scenario of this disclosure is shown. In this application scenario, it may include a wind turbine 100, a pulsed Doppler lidar (LiDAR) 200, and a central controller 300.

[0033] The LiDAR 200 should have at least two scanning modes: Plane Position Indication (PPI) mode: a sector scan with a fixed elevation angle, used to measure wind speed and direction profiles at different azimuth angles; Distance and Height Indication (RHI) mode: a vertical scan with a fixed azimuth angle, used to measure wind shear and vertical wind direction changes.

[0034] The local controller of the separate generator 100 has a data acquisition unit. The local controller of each wind turbine acquires the unit's operating status data at a frequency of not less than 1Hz through the built-in data acquisition and monitoring control system (SCADA). This data includes, but is not limited to: nacelle anemometer and wind vane data, generator output power, pitch angle of the speed pitch system, yaw angle of the yaw system, and yaw error.

[0035] The central processing unit 300 can be, for example, an industrial-grade central controller (high-performance industrial PC or server), which communicates with the local controllers of all wind turbines and the LiDAR equipment via a high-speed local area network. This unit is responsible for running the core algorithm and calculating the optimal control commands.

[0036] The communication network is a high-speed, low-latency industrial Ethernet used to connect the LiDAR100, the local controllers of each wind turbine, and the central controller 300, ensuring real-time data transmission.

[0037] Figure 2 A schematic flowchart of a control method according to one embodiment of the present disclosure is shown. Figure 3 A detailed flowchart of a specific embodiment of the method of this disclosure is shown. For example... Figure 2 , Figure 3 The method shown includes steps S210 to S240. This method can be performed by an electronic device.

[0038] In step S210, forward-looking information representing wind conditions in the future period is obtained.

[0039] This step involves the acquisition and fusion of multi-source forward-looking data, which combines forward-looking wind field information from LiDAR with real-time status information of wind turbines to generate high-precision, high-confidence future wind field forecasts.

[0040] Data preprocessing: The raw LiDAR data undergoes coordinate transformation, converting it from a polar coordinate system (distance, azimuth, elevation) to a Cartesian coordinate system (x, y, z) with wind field reference. Outlier removal and filtering smoothing are performed on the SCADA data to reduce measurement noise. For spatiotemporal alignment and fusion algorithms, the Extended Kalman Filter (EKF) algorithm is used for fusion. State vector definition:

[0041] That is, the three-dimensional wind speed components at time k. Where X(k) represents the three-dimensional wind speed state vector at time k; U x (k), U y (k), U z (k) represents the components of the wind speed at time k in the x, y, and z directions in the Cartesian coordinate system; the superscript... This indicates the transpose of a vector, converting a row vector into a column vector.

[0042] State prediction equation: Predict the state at the next moment based on a simplified wind field physical model.

[0043]

[0044] Where X(k∣k) 1) Indicates based on k The one-step prediction of the state vector at time k based on information from time 1 and earlier; X(k 1) For k The estimated state vector at time 1; F is the state transition matrix, describing the dynamic evolution of the system; w(k) is the process noise vector at time k, usually assumed to be zero-mean Gaussian white noise.

[0045]

[0046] Where Z(k) represents the observation vector at time k; X(k) represents the state vector at time k; H is the observation matrix, used to map the state space to the observation space, and its construction depends on the beam pointing of the LiDAR and the position of the nacelle anemometer; v(k) is the observation noise vector at time k, which is usually assumed to be zero-mean Gaussian white noise.

[0047] The optimal estimation of the three-dimensional wind speed field across the entire field is achieved through the prediction and update loop of the extended Kalman filter (EKF), and the forecast sequence {X(k+1|k),X(k+2|k), ...} for the next tens of seconds is output.

[0048] In step S220, a model predictive control problem containing multiple optimization objectives is constructed based on the prospective information.

[0049] Based on the fused high-precision wind field forecast, the future inflow conditions and performance of each wind turbine are dynamically predicted.

[0050] Dynamic wake prediction model: The Larsen wake model is adopted. The model inputs are: the real-time thrust coefficient Ct of the upstream wind turbine, the rotor diameter D, and the predicted wind speed U∞ and turbulence intensity TI from step S1.

[0051] The core of the model is to solve a relaxation equation for the wake velocity deficit ΔU, simulating its dynamic process of propagation downstream:

[0052] Wherein, ΔU represents the wake velocity deficit (i.e., the difference between the free-flow wind speed and the wake region wind speed). ΔU s τ represents the wake velocity deficit under steady state; τ is a time constant that reflects the speed of wake dynamic recovery and is usually related to atmospheric turbulence intensity and convection velocity; t is a time variable.

[0053] For the downstream j-th wind turbine, its inflow velocity U j The free-flow velocity is the sum of the wake velocity losses generated by all upstream wind turbines i in front of it.

[0054] Performance and load prediction: Power prediction: based on the predicted inflow wind speed U j The future power output P can be predicted by analyzing the wind power curve. j .

[0055] Load prediction: Establish key loads (such as the tower base bending moment M) y )and The relationship between inflow conditions and turbine operating conditions (such as pitch angle β, speed ω) Transfer function model. Its prediction equation is:

[0056] Among them, M y,j U represents the bending moment about the y-axis at the root of the j-th wind turbine tower or blade (usually referring to the critical fatigue load in the lateral or longitudinal direction); j Let be the inflow velocity of the j-th fan; TI be the turbulence intensity; β be the inflow velocity of the j-th fan. j Let ω be the yaw angle of the j-th wind turbine; j The rotor speed of the j-th wind turbine is denoted as 'Rotorspeed'; "..." indicates other operating or environmental parameters that may affect the bending moment, such as pitch angle, wind shear, and air density.

[0057] Model predictive control (MPC) optimization problem construction.

[0058] The optimization objective is to maximize the total power generation of the wind farm while minimizing total power fluctuations and total load accumulation. This is a multi-objective optimization problem, which can be weighed using weighting coefficients.

[0059] Determine the control variables: the set values ​​of the pitch angle (β) and generator torque (T) for each wind turbine.

[0060] Define constraints: including physical and safety constraints such as wind turbine speed range, power limit, pitch angle change rate, and torque change rate.

[0061] Constructing a rolling time-domain optimization problem: In each control cycle, based on the current state and the prediction model, solve for the optimal control sequence within a finite future time domain, but only implement the first control command, then roll over to the next cycle to re-optimize. The wind farm coordinated control problem is thus formulated as a standard rolling time-domain optimization problem.

[0062] Optimization objective (cost function): Design a multi-objective cost function J, aiming to achieve the following within a future prediction time domain Np (e.g., 30 seconds):

[0063] in: J represents the comprehensive performance index of multi-objective optimization; P total (k) represents the total power generation of the wind farm at time k in the prediction time domain; ΔP total (k)=P total (k), P total (k 1) Represents the change in total power between adjacent time points; the sum of its squares is used to measure the power fluctuation across the entire field; M y,total (k) represents a certain aggregate quantity (such as the sum of squares or weighted sum) of the bending moments about the y-axis of all key components of the wind turbine (such as the tower base) at time k, used to characterize the cumulative fatigue load; α, β, γ are non-negative weighting coefficients used to balance the relative importance of different optimization objectives; N is the prediction time domain length of the model predictive control.

[0064] Control variables and constraints: Control variables: The optimization variables are the sequence of pitch angle β and generator torque T of all wind turbines in the future control time domain Nc (e.g., 10 seconds).

[0065] Constraints: Inequality constraint: β_min ≤ β(k) ≤ β_max (pitch angle) Travel limit); T_min ≤ T(k) ≤ T_max (torque limit); | Δβ(k)| ≤ Δβ max (Pitch angle change rate limit); |ΔT(k)|≤ ΔT_max (torque change rate limit).

[0066] Equation constraints: Equations from the wind turbine dynamic model and prediction model.

[0067] In step S230, the model predictive control problem is solved to obtain the optimal control commands for each wind turbine.

[0068] To address the issue of high computational cost in centralized optimization of large-scale wind farms, a distributed optimization algorithm is adopted.

[0069] The Alternating Directional Multiplier Method (ADMM) is adopted: the global optimization problem is decomposed into several subproblems, and each subproblem corresponds to the local optimization of a wind turbine i.

[0070] Local subproblem: Wind turbine i solves a small-scale MPC problem locally, whose cost function is only related to its own state and control variables, and is coupled with other wind turbines by introducing consistency constraints.

[0071] Global coordination: Through multiple iterations, each wind turbine exchanges optimization results (i.e., estimates of coupled variables) with its neighboring turbines, and updates the Lagrange multipliers under the coordination of the central controller, ultimately bringing all local solutions to the global optimum.

[0072] Algorithm steps (in the l-th iteration): Local optimization: Each wind turbine solves its own local optimization problem in parallel.

[0073] Global variable update: The central controller collects all local solutions and updates the global consistency variable z.

[0074] Multiplier update: The central controller updates the Lagrange multiplier λ based on the difference between the updated global variables and local variables.

[0075] Convergence check: Check if the original residual and the dual residual are less than the set tolerance. If converged, exit; otherwise, return to S210.

[0076] This architecture greatly reduces computational complexity and enhances system reliability (a single point of failure does not affect the whole).

[0077] In step S240, the optimal control command is sent to each corresponding wind turbine for execution.

[0078] At the end of each control cycle (e.g., 1 second), the first value of the optimal control command sequence obtained by ADMM is calculated. The information is sent to the local controller of the corresponding fan i.

[0079] The local controller receives instructions and executes the new setpoints through the underlying pitch and torque control systems.

[0080] The system enters the next control cycle, repeating steps S210 to S240 to achieve rolling optimization and closed-loop feedback control.

[0081] Reference Figure 4 A comparison chart of wind farm power output under traditional independent control and collaborative control under a specific implementation method. Figure 5 As can be seen from the comparison diagram of fatigue load of tower base bending moment under traditional independent control and system control of a specific implementation method, the control method disclosed herein has significant and good technical effects.

[0082] This disclosure also provides a wind turbine cluster collaborative optimization control system, comprising: an information acquisition unit for acquiring forward-looking information characterizing wind conditions in future periods; a problem construction unit for constructing a model predictive control problem containing multiple optimization objectives based on the forward-looking information; an instruction generation unit for solving the model predictive control problem to obtain the optimal control instruction for each wind turbine; and an instruction execution unit for issuing the optimal control instruction to the corresponding wind turbine for execution.

[0083] According to another aspect of this disclosure, an electronic device is provided, including a processor and a memory, the memory storing a computer program that, when executed by the processor, implements a control method according to any embodiment of this disclosure.

[0084] The hardware architecture of electronic devices / devices can be implemented using a bus architecture. A bus architecture can include any number of interconnect buses and bridges, depending on the specific application and overall design constraints of the hardware. A bus connects various circuits, including one or more processors, memories, and / or hardware modules. A bus can also connect various other circuits such as peripherals, voltage regulators, power management circuits, external antennas, etc. Buses can be Industry Standard Architecture (ISA) buses, Peripheral Component Interconnect (PCI) buses, or Extended Industry Standard Component (EISA) buses, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, this diagram uses only one connecting line, but it does not represent a single bus or a single type of bus.

[0085] For ease of explanation, certain steps of the above method are described in relation to modules. It should be understood that the corresponding module performing one or more steps of the above method may be one or more hardware modules specifically configured to perform the corresponding step, or implemented by a processor configured to perform the corresponding step, or stored in a computer-readable medium for implementation by a processor, or implemented by some combination thereof.

[0086] The specific implementation of each module in the above-mentioned device can be referred to the implementation process of the corresponding steps in the above-mentioned method implementation method of this disclosure, and will not be repeated here.

[0087] This disclosure also provides a readable storage medium storing a computer program that, when executed by a processor, is used to implement the methods described above. A "readable storage medium" can be any means capable of containing, storing, communicating, propagating, or transmitting a program for use by or in conjunction with an instruction execution system, apparatus, or device. More specific examples of a readable storage medium include: an electrical connection with one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and portable read-only memory (CDROM), etc.

[0088] This disclosure also provides a computer program product, the methods of which can be implemented wholly or partially through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented wholly or partially as a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed, all or part of the processes or functions of this disclosure are performed.

[0089] Computer programs or instructions can be stored in a readable storage medium or transferred from one readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The readable storage medium can be any available medium capable of access, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; an optical medium, such as a digital video optical disc; or a semiconductor medium, such as a solid-state drive. The computer-readable storage medium can be a volatile or non-volatile storage medium, or it can include both volatile and non-volatile types of storage media.

[0090] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, systems, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0091] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0092] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0093] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0094] In the description of this specification, the references to terms such as "one embodiment / mode," "some embodiments / modes," "example," "specific example," or "some examples," etc., refer to specific features, structures, or characteristics described in connection with that embodiment / mode or example, which are included in at least one embodiment / mode or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment / mode or example. Moreover, the specific features, structures, or characteristics described may be combined in any suitable manner in one or more embodiments / modes or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments / modes or examples described in this specification, as well as the features of different embodiments / modes or examples.

[0095] Those skilled in the art should understand that the above embodiments are merely for illustrating the present disclosure and are not intended to limit the scope of the disclosure. Those skilled in the art can make other changes or modifications based on the above disclosure, and these changes or modifications still fall within the scope of the present disclosure.

Claims

1. A method for collaborative optimization control of a wind turbine cluster, characterized in that, Includes the following steps: Obtain forward-looking information that characterizes wind conditions in future periods; Based on the aforementioned prospective information, a model predictive control problem with multiple optimization objectives is constructed. Solve the model predictive control problem to obtain the optimal control commands for each wind turbine; The optimal control command is then sent to each wind turbine for execution.

2. The control method according to claim 1, characterized in that, The acquisition of forward-looking information representing future wind conditions includes: Three-dimensional wind field data for future periods can be obtained by deploying remote sensing measurement equipment upstream of wind farms.

3. The control method according to claim 2, characterized in that, The construction of the model predictive control problem, which includes multiple optimization objectives, includes: Based on the aforementioned forward-looking information, the total power generation of the wind farm, the load on key components of each wind turbine, and the fluctuation of total output power are predicted in the future period. The optimization objectives are to maximize the total power generation, minimize the load on the key components, and minimize the fluctuation of the total output power.

4. The control method according to claim 3, characterized in that, The predicted total power generation of the wind farm, the load on key components of each wind turbine, and the fluctuation of total output power over the future period include: Using a dynamic wake model, the inflow wind speed of each wind turbine is calculated based on the three-dimensional wind field data, and predictions are made based on the inflow wind speed.

5. The control method according to claim 1, characterized in that, The process of obtaining the optimal control commands for each wind turbine includes: The yaw angle command sequence and pitch angle command sequence for each wind turbine in the future time period are obtained; The instruction in the instruction sequence corresponding to the current moment is taken as the optimal control instruction.

6. The control method according to claim 1, characterized in that, After the optimal control command is issued to the corresponding wind turbines for execution, the method further includes: Collect actual operational feedback data from each wind turbine; Based on the actual operational feedback data, the predictive model used to construct the model predictive control problem is calibrated online.

7. A collaborative optimization control system for a wind turbine cluster, characterized in that, include: The information acquisition unit is used to acquire forward-looking information that characterizes wind conditions in future periods. The problem construction unit is used to construct a model predictive control problem containing multiple optimization objectives based on the prospective information. The instruction generation unit is used to solve the model predictive control problem and obtain the optimal control instructions for each wind turbine. The instruction execution unit is used to issue the optimal control instruction to the corresponding wind turbines for execution.

8. An electronic device, characterized in that, It includes a processor and a memory, the memory storing a computer program that, when executed by the processor, implements the control method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the control method as described in any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the control method according to any one of claims 1 to 6.

Images