Wind farm yaw control method, system, device and medium based on hierarchical optimization

Abstract

The present application provides a wind farm yaw control method, system, device and medium based on hierarchical optimization, which relates to the technical field of wind farms. The method includes: collecting real-time data of the wind farm; the real-time data includes wind speed, wind direction and wind turbine status information; preprocessing the real-time data to obtain initial wind farm data; combining the wake model, performing benchmark calculation and hierarchical optimization on the initial wind farm data to obtain the optimal yaw angle; the wake model is used to simulate the effect of mutual influence between wind turbines; hierarchical optimization includes a coarse search stage and a fine search stage; according to the optimal yaw angle, the wind turbine is controlled to perform yaw. The present application improves the computational efficiency and adaptability of yaw control by adopting hierarchical optimization, thereby improving the overall power generation efficiency of the wind farm.

Description

Technical Field

[0001] The present application relates to the technical field of wind farms, and in particular to a wind farm yaw control method, system, device and medium based on hierarchical optimization. Background Art

[0002] With the continuous expansion of wind farms, especially the centralized construction of wind farms, existing wind farms are facing a series of efficiency bottlenecks. Although the power of a single wind turbine continues to increase, the overall power generation efficiency of the wind farm has not been maximized due to the wake effect between wind turbines and the density of wind farm layout. The traditional yaw control method only performs wind-to-wind operations to maximize the power characteristics of a single wind turbine, but fails to fully consider the problem of reduced power generation efficiency caused by the wake effect, and cannot achieve global power maximization of the wind farm. To this end, the existing technology reduces the wake effect and improves power generation efficiency by optimizing the yaw angle. They mainly include yaw control methods based on computational fluid dynamics simulation, genetic algorithms, reinforcement learning and digital twins. However, they have problems such as low computational efficiency, reliance on training set accuracy, poor generalization ability and high computational cost, and have not been widely used in actual wind farms due to the large amount of calculation. These shortcomings limit their application effect and real-time performance in complex and changing environments. Summary of the invention

[0003] The present application provides a hierarchically optimized wind farm yaw control method, system, device and medium to solve one or more technical problems existing in the prior art and at least provide a beneficial option or create conditions.

[0004] On the one hand, the present application provides a wind farm yaw control method based on hierarchical optimization, comprising the following steps:

[0005] Collecting real-time data of the wind farm; the real-time data includes wind speed, wind direction and wind turbine status information;

[0006] Preprocessing the real-time data to obtain initial wind farm data;

[0007] Combined with the wake model, benchmark calculation and hierarchical optimization are performed on the initial wind farm data to obtain an optimal yaw angle;

[0008] The wake model is used to simulate the mutual influence between wind turbines; the hierarchical optimization includes a coarse search phase and a fine search phase; the coarse search phase refers to the process of quickly finding an approximate optimal solution with a larger step size within a larger parameter range; the fine search phase refers to the process of accurately finding an optimal solution with a smaller step size near the approximate optimal solution found in the coarse search phase;

[0009] The wind turbine is controlled to perform yaw according to the optimal yaw angle.

[0010] Furthermore, the preprocessing of the real-time data to obtain initial wind farm data includes:

[0011] Removing outliers and invalid data from the real-time data to filter out valid data;

[0012] Performing interpolation processing on the valid data to fill in the gaps in the valid data to obtain complete data;

[0013] The complete data is synchronized with the timestamp so that the timestamps of the complete data from different sources are consistent, thereby obtaining the initial wind farm data.

[0014] Furthermore, the combination of the wake model, performing benchmark calculation and hierarchical optimization on the initial wind farm data to obtain the optimal yaw angle includes:

[0015] In combination with the wake model, a benchmark calculation is performed on the initial wind farm data to obtain benchmark data;

[0016] Perform a rough search based on the benchmark data and the wake model to obtain a first yaw angle;

[0017] According to the first yaw angle and in combination with the wake model, a greedy strategy is adopted to perform a detailed search to obtain the optimal yaw angle.

[0018] Furthermore, the combining the wake model to perform a benchmark calculation on the initial wind farm data to obtain benchmark data includes:

[0019] Determining the initial operating state of the wind turbine according to the initial wind farm data; the initial operating state includes the yaw angle, pitch angle, rotation speed and power of the wind turbine when it does not perform active yaw;

[0020] According to the initial operating state and in combination with the wake model, a reference wind farm power and a reference wake velocity field when the wind turbine does not perform active yaw are calculated and obtained as the reference data.

[0021] Further, the performing a rough search based on the reference data and in combination with the wake model to obtain a first yaw angle includes:

[0022] Within a preset yaw angle coarse search range, traverse each yaw angle according to a preset coarse search step angle, and calculate the wind farm power corresponding to the yaw angle using the wake model to obtain a coarse search wind farm power set;

[0023] The yaw angle corresponding to the maximum value in the rough search wind farm power set is used as the first yaw angle.

[0024] Further, the step of performing a detailed search based on the first yaw angle and in combination with the wake model using a greedy strategy to obtain the optimal yaw angle includes:

[0025] According to the first yaw angle, setting a yaw angle detailed search range;

[0026] In the yaw angle fine search range, taking the first yaw angle as a starting point, traversing multiple yaw angles according to a preset fine search step angle;

[0027] Acquire a difference between the wind farm power at a current yaw angle and the wind farm power at a previous yaw angle as a power increment at the current yaw angle;

[0028] When the power increment of the current yaw angle is greater than a preset value, the next yaw angle is determined as the current yaw angle, and the process returns to the step of obtaining the difference between the wind farm power at the current yaw angle and the wind farm power at the previous yaw angle as the power increment of the current yaw angle, until the power increment of the current yaw angle is less than the preset value, and the previous yaw angle is determined as the optimal yaw angle.

[0029] Furthermore, the wake model includes a Jimenez wake model, a Gauss deflection model and a wake accumulation curling model; wherein the Jimenez wake model is used to simulate the velocity attenuation characteristics of the wake in the downstream; the Gauss deflection model is used to describe the directional deflection of the wake; and the wake accumulation curling model is used to consider the interaction between the wakes of multiple wind turbines.

[0030] On the other hand, the present application provides a wind farm yaw control system based on hierarchical optimization, comprising: a data collection module, a data preprocessing module, a hierarchical optimization module and a yaw control module;

[0031] The data collection module is used to collect real-time data of the wind farm; the real-time data includes wind speed, wind direction and wind turbine status information;

[0032] The data preprocessing module is used to preprocess the real-time data to obtain initial wind farm data;

[0033] The hierarchical optimization module is used to combine the wake model to perform benchmark calculation and hierarchical optimization on the initial wind farm data to obtain the optimal yaw angle; the wake model is used to simulate the effect of mutual influence between wind turbines; the hierarchical optimization includes a coarse search stage and a fine search stage; the coarse search stage refers to the process of quickly finding an approximate optimal solution with a larger step size within a larger parameter range; the fine search stage refers to the process of accurately finding an optimal solution with a smaller step size near the approximate optimal solution found in the coarse search stage;

[0034] The yaw control module is used to control the wind turbine to perform yaw according to the optimal yaw angle.

[0035] On the other hand, the present application provides a wind farm yaw control device based on hierarchical optimization, comprising: a wind turbine, a wind tower, a laser radar, a wind turbine SCADA system device, an edge computing center device and a centralized data center device; the edge computing center device is arranged on the wind turbine;

[0036] The wind tower and the laser radar are used to collect wind speed and wind direction of the wind farm; the wind turbine SCADA system is used to collect wind turbine status information; the wind speed, wind direction and wind turbine status information are used as real-time data;

[0037] The edge computing center device is used to pre-process the real-time data to obtain initial wind farm data, and send the initial wind farm data to the centralized data center;

[0038] The centralized data center device is used to combine the wake model to perform benchmark calculation and hierarchical optimization on the initial wind farm data to obtain an optimal yaw angle; generate a yaw control instruction according to the optimal yaw angle, and send the yaw control instruction to the edge computing center;

[0039] The wake model is used to simulate the effect of mutual influence between wind turbines; the hierarchical optimization includes a coarse search stage and a fine search stage; the coarse search stage refers to the process of quickly finding an approximate optimal solution with a larger step size within a larger parameter range; the fine search stage refers to the process of accurately finding an optimal solution with a smaller step size near the approximate optimal solution found in the coarse search stage;

[0040] The edge computing center device is also used to detect the yaw state of the wind turbine and send the yaw control instruction to the wind turbine SCADA system device;

[0041] The wind turbine SCADA system device is also used to control the wind turbine to perform yaw according to the yaw control instruction.

[0042] On the other hand, the present application provides a computer medium storing a program executable by a processor, wherein the program executable by the processor is used to implement the aforementioned wind farm yaw control method based on hierarchical optimization when executed by the processor.

[0043] The beneficial effects of the present application are as follows: the present application provides a wind farm yaw control method based on hierarchical optimization, comprising the following steps: collecting real-time data of the wind farm; the real-time data includes wind speed, wind direction and wind turbine status information; preprocessing the real-time data to obtain initial wind farm data; combining the wake model, performing benchmark calculation and hierarchical optimization on the initial wind farm data to obtain the optimal yaw angle; the wake model is used to simulate the effect of mutual influence between wind turbines; hierarchical optimization includes a coarse search stage and a fine search stage; the coarse search stage refers to the process of quickly finding an approximate optimal solution with a large step size within a large parameter range; the fine search stage refers to the process of accurately finding the optimal solution with a small step size near the approximate optimal solution found in the coarse search stage; according to the optimal yaw angle, controlling the wind turbine to perform yaw. The present application improves the computational efficiency and adaptability of yaw control by adopting hierarchical optimization, thereby improving the overall power generation efficiency of the wind farm. The present application also provides corresponding systems, devices and media, and the beneficial effects of the systems, devices and media are similar to those of the method, which will not be repeated here.

[0044] Other features and advantages of the present application will be described in the following description, and partly become apparent from the description, or understood by practicing the present application. The purpose and other advantages of the present application can be realized and obtained by the structures specifically pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings are used to provide a further understanding of the technical solution of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the technical solution of the present invention and do not constitute a limitation on the technical solution of the present invention.

[0046] Figure 1 is a flow chart of a wind farm yaw control method based on hierarchical optimization provided in this application;

[0047] Figure 2 It is a structural diagram of a wind farm yaw control system based on hierarchical optimization provided by this application;

[0048] Figure 3 It is a structural diagram of a wind farm yaw control device based on hierarchical optimization provided by the present application;

[0049] Figure 4 It is the result of active yaw control performed by 30 wind turbines with a rated power of 2 MW in a daily cycle provided by this application;

[0050] Figure 5 It is a comparative schematic diagram of wind speed, wind direction, and wind farm power increase value within a daily cycle provided by the present application;

[0051] Figure 6It is a schematic diagram of the wake effect of a wind turbine when active yaw optimization is performed at a certain moment provided in this application. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0053] The present application is further described below in conjunction with the accompanying drawings and specific embodiments. The described embodiments should not be regarded as limiting the present application, and all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0054] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0055] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0056] As one of the most mature renewable energy technologies, wind power generation has been widely used in the construction of large-scale wind farms. However, despite the continuous advancement of wind power technology, existing wind farms still face many challenges, especially in terms of improving power generation efficiency and reducing costs, and innovative solutions are urgently needed.

[0057] Traditional yaw control usually aims at the maximum power output of a single wind turbine, mainly by adjusting the orientation of the wind turbine nacelle so that the wind rotor always faces the wind direction, in order to optimize the wind energy capture efficiency of a single wind turbine. This method does not fully consider the mutual influence between multiple wind turbines and the overall power generation efficiency of the wind farm. It cannot fully improve the overall power generation efficiency in large-scale wind farms and wind farms with limited site density. Moreover, traditional methods usually require a long calculation time when calculating the optimal yaw angle, and cannot quickly respond to changes in the wind farm environment, resulting in poor real-time performance of yaw control. In order to cope with the problem of reduced power generation efficiency caused by the wake effect, most existing technologies have adopted wake models and began to focus on active yaw wake control technology.

[0058] The patent with publication number CN119195973A proposes a method for increasing wind farm power generation based on active wake yaw control optimization. The solution first obtains basic information about the wind farm, establishes a computational fluid dynamics simulation model, and obtains wind turbine operation data through lidar. The Gaussian wake model is corrected through computational fluid dynamics simulation results and operation data, and then the optimal yaw angle is searched through the particle swarm algorithm to perform the yaw action.

[0059] The patent with publication number CN118934447A proposes a method, device, equipment and medium for dynamic yaw control of the wake of a wind farm. The solution first obtains basic information of the wind farm, calculates the inflow wind speed of each wind turbine based on the model method (Taylor background field and steady-state speed loss method), and then sets different yaw angles through a genetic algorithm, thereby solving the optimal yaw angle and performing yaw action.

[0060] The patent with publication number CN118981935A proposes a method and device for constructing a wind farm yaw angle optimization control model, as well as a control strategy and device. The solution first obtains basic information about the wind farm, calculates the wake using the Gaussian wake model, and then solves the optimal yaw angle using a directed graph and reinforcement learning training method to perform the yaw action.

[0061] The patent with publication number CN118971195A proposes a wind farm multi-objective collaborative optimization scheduling method and device based on digital twins. The solution is built on the wind farm inflow wind speed digital twin model, the wind speed digital twin model in front of the machine, the single machine digital twin model, the wake characteristics digital twin model and the wind farm field-level energy management twin model through reinforcement learning and other methods, and then uses the distributed predictive control algorithm to optimize the control parameters and control the operation of wind turbines for different typical scenarios.

[0062] Although the existing technology has made certain progress, there are still some problems. The solution combined with the computational fluid dynamics simulation model is less efficient in the calculation of the wake velocity field and is difficult to meet the needs of real-time control. The yaw angle optimization solution using optimization methods such as particle swarm optimization, reinforcement learning, and digital twins relies on the accuracy of the training set, and has poor generalization ability. It is not suitable for the complex and changeable actual wind farm scenes and has a high computational cost. Some methods require complex models and algorithms, which increases the complexity and implementation difficulty of the system, and is not conducive to promotion and application in actual wind farms.

[0063] In response to the problems existing in the related technologies, the embodiments of the present application provide a wind farm yaw control method, system, device and medium based on hierarchical optimization. First, the embodiments of the present application adopt a hierarchical optimization strategy of coarse-fine search, which effectively improves the calculation efficiency by reducing the amount of calculation of global optimization when optimizing the yaw angle. At the same time, the greedy strategy is introduced in the fine search stage to further improve the accuracy and real-time performance of the search. Secondly, the embodiments of the present application set up an edge computing center to obtain and process data from wind towers, lidars and SCADA systems in real time, thereby reducing the burden on centralized data centers, improving real-time performance, and greatly improving the response speed of yaw control instructions. In addition, the embodiments of the present application can better adapt to the complex and changeable actual wind farm environment, and effectively alleviate the problems of high computing cost, strong dependence on training sets and poor generalization ability in the prior art. Through these improvements, the embodiments of the present application effectively improve the overall power generation efficiency of the wind farm, and have important application value and broad market prospects.

[0064] First, the wind farm yaw control method based on hierarchical optimization provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0065] The wind farm yaw control method based on hierarchical optimization proposed in the embodiment of the present application can be applied to a terminal, a server, or software running in a terminal or a server. The terminal can be a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks, and big data and artificial intelligence platforms.

[0066] Reference Figure 1 The implementation process of the wind farm yaw control method based on hierarchical optimization provided in the embodiment of the present application includes but is not limited to the following steps.

[0067] Step 101: collect real-time data of the wind farm.

[0068] The real-time data includes wind speed, wind direction and wind turbine status information. Wind speed and wind direction determine the efficiency and direction of wind turbines in capturing wind energy, while wind turbine status information reflects its current operating status.

[0069] In step 101, collecting real-time data of the wind farm is the basis of the entire yaw control optimization method. Real-time data is downloaded synchronously from multiple data sources (such as wind towers, lidars, and wind turbine SCADA systems). These data include but are not limited to wind speed, wind direction, and status information of wind turbines (such as yaw angle, pitch angle, speed, and power, etc.). Collecting these real-time data can fully understand the current operating conditions and environmental conditions of the wind farm. These data provide the original input for subsequent preprocessing and optimization calculations, ensuring that the optimization process can be based on the latest actual conditions, thereby improving the accuracy and effectiveness of yaw control. Only by accurately and timely obtaining these real-time data can the smooth progress of subsequent steps and the reliability of optimization results be ensured.

[0070] Step 102: pre-process the real-time data to obtain initial wind farm data.

[0071] In step 102, the original real-time data from different data sources obtained previously may have problems such as missing data and noise interference, and direct use will affect the accuracy of the calculation results. Next, these data need to be preprocessed by screening, interpolation, and timestamp synchronization to ensure the consistency and accuracy of all data. This step not only eliminates the impact of noise and outliers, but also fills in possible data missing points through interpolation technology, thereby providing high-quality input data for subsequent wake model calculations. In addition, the preprocessing process also includes integrating data from different sources into a time-synchronized format to facilitate subsequent analysis and calculations.

[0072] Step 103 , combining the wake model, performing benchmark calculation and hierarchical optimization on the initial wind farm data to obtain the optimal yaw angle.

[0073] It should be noted that the wake model is used to simulate the effect of mutual influence between wind turbines. Hierarchical optimization includes a coarse search phase and a fine search phase. The coarse search phase refers to the process of quickly finding an approximate optimal solution with a larger step size within a larger parameter range. The fine search phase refers to the process of accurately finding the optimal solution with a smaller step size near the approximate optimal solution found in the coarse search phase.

[0074] In step 103, the core step of the whole method is to perform benchmark calculation and hierarchical optimization on the initial wind farm data in combination with the wake model. The wake model is used to simulate the mutual influence between wind turbines, and the benchmark calculation is used to determine the wind farm power when the active yaw strategy is not implemented, as a reference for subsequent optimization. The hierarchical optimization includes a coarse search stage and a fine search stage. By searching in different ranges with different step lengths, the approximate optimal solution is quickly found and the optimal solution is accurately determined. This process can effectively reduce the amount of calculation and improve the search efficiency, while ensuring that the optimal yaw angle found can maximize the overall power output of the wind farm. Through this step, the optimal yaw control strategy under the current working conditions can be obtained, providing a scientific basis for the subsequent yaw operation of the wind turbine.

[0075] Step 104: Control the wind turbine to yaw according to the optimal yaw angle.

[0076] In step 104, controlling the wind turbine to perform yaw according to the optimal yaw angle is the key to applying the optimization results to the actual operation of the wind farm. By sending the optimal yaw angle command to the control system of the wind turbine and adjusting the yaw angle of the wind turbine, the wind turbine can be operated according to the optimized strategy, thereby effectively reducing the wake impact of the upstream wind turbine on the downstream wind turbine and improving the power generation efficiency of the entire wind farm. This step realizes the transformation from theoretical optimization to practical application, ensures that the optimization results can play a role in actual operation, and ultimately achieves the purpose of increasing the overall power generation of the wind farm.

[0077] In some embodiments of the present application, in step 102, the implementation process of preprocessing the real-time data to obtain the initial wind farm data includes but is not limited to the following steps.

[0078] Step 201, remove outliers and invalid data in real-time data, and filter out valid data.

[0079] In step 201, outliers and invalid data in real-time data are removed, valid data is screened out, and the data used for subsequent analysis and calculation is ensured to be accurate and reliable. Outliers may be caused by sensor failure, data transmission errors or other external interference, while invalid data may be caused by incomplete data or data that does not conform to the expected format. By removing these outliers and invalid data, it is possible to avoid them from interfering with subsequent interpolation processing and timestamp synchronization, thereby improving the overall quality and credibility of the data and laying a solid foundation for subsequent steps.

[0080] Step 202, interpolation processing is performed on the valid data to fill in the gaps in the valid data to obtain complete data.

[0081] In step 202, interpolation processing is performed on the valid data to fill in the gaps in the data and ensure the integrity and continuity of the data so that subsequent analysis and calculation can proceed smoothly. In practical applications, gaps may appear in the data due to various reasons (such as sensor failure, data loss, etc.). Through interpolation processing, the value of the missing data can be estimated based on the surrounding data points to fill these gaps. This not only improves the availability of the data, but also avoids calculation errors or analysis deviations caused by missing data, ensuring that subsequent steps can be performed based on a complete data set.

[0082] Step 203 , performing time stamp synchronization processing on the complete data, so that the time stamps of the complete data from different sources are consistent, and obtaining initial wind farm data.

[0083] In step 203, the complete data is timestamped and synchronized to make the timestamps of data from different sources consistent, ensuring that all data are aligned in time so that subsequent analysis and calculation can accurately reflect the real-time operating status of the wind farm. In practical applications, data may come from multiple different sensors and systems, and the timestamps of these data may be inconsistent. Through timestamp synchronization, all data can be adjusted to the same time reference to ensure the synchronization of data in time. This not only improves the consistency and comparability of the data, but also provides an accurate time reference for subsequent benchmark calculations and hierarchical optimization, ensuring the accuracy and reliability of the entire yaw control optimization process.

[0084] In some embodiments of the present application, in step 103, the initial wind farm data is subjected to benchmark calculation and hierarchical optimization in combination with the wake model, and the process of obtaining the optimal yaw angle includes but is not limited to the following steps.

[0085] Step 301 , combining a wake model, performing a benchmark calculation on the initial wind farm data to obtain benchmark data.

[0086] In step 301, the initial wind farm data is benchmarked in combination with the wake model in order to determine the operating state of the wind farm when no active yaw strategy is implemented. This step simulates the wake interaction between wind turbines to calculate the wind farm power distribution and wake velocity field when yaw optimization is not performed. The benchmark data, as a reference point for subsequent optimization processes, can help evaluate the impact of active yaw strategies on the overall performance of the wind farm. Through the benchmark calculation, the actual situation of the power generation efficiency and wake effect of the current wind farm can be clarified, providing the necessary initial conditions for subsequent hierarchical optimization.

[0087] Step 302: Perform a rough search based on the reference data and the wake model to obtain a first yaw angle.

[0088] In step 302, a rough search is performed based on the benchmark data in combination with the wake model in order to quickly find an approximate optimal solution within a larger yaw angle range. This step calculates the wind farm power at each yaw angle by traversing the preset yaw angle rough search range with a larger step size. The purpose of the rough search is to narrow the range of the optimal yaw angle and reduce the amount of calculation for subsequent fine searches. Through the rough search, a near-optimal yaw angle can be quickly determined, providing a clear search interval for subsequent fine searches, thereby improving the efficiency of the optimization process.

[0089] Step 303: Based on the first yaw angle and in combination with the wake model, a greedy strategy is adopted to perform a detailed search to obtain an optimal yaw angle.

[0090] In step 303, a greedy strategy is used to perform a fine search based on the first yaw angle obtained by the coarse search in combination with the wake model in order to accurately find the optimal yaw angle within a smaller range. This step calculates the wind farm power at each yaw angle by searching with a smaller step size near the first yaw angle. The purpose of the fine search is to further optimize the yaw angle to ensure that the optimal yaw angle found can maximize the overall power output of the wind farm. The greedy strategy ensures the efficiency and accuracy of the search process by gradually adjusting the yaw angle and stopping the optimization when the power no longer increases. Through the fine search, the optimal yaw angle can be accurately determined, thereby maximizing the power generation efficiency of the wind farm.

[0091] In some embodiments of the present application, in step 301, a baseline calculation is performed on the initial wind farm data in combination with a wake model, and the implementation process of obtaining the baseline data includes but is not limited to the following steps.

[0092] Step 401, determining the initial operating state of the wind turbine according to the initial wind farm data.

[0093] The initial operating state includes the yaw angle, pitch angle, rotation speed and power of the wind turbine when it does not perform active yaw.

[0094] In step 401, the initial operating state of the wind turbine is determined based on the initial wind farm data, which is the basis for the benchmark calculation. This step obtains the key operating parameters of the wind turbine when the active yaw strategy is not implemented by analyzing the initial data, including yaw angle, pitch angle, speed and power. These parameters reflect the operating status of the wind turbine under natural wind field conditions and provide necessary input for subsequent wake model calculations. By accurately determining the initial operating state, the accuracy and reliability of the benchmark calculation can be ensured, thereby providing a reliable starting point for the subsequent optimization process.

[0095] Step 402 : According to the initial operation state and in combination with the wake model, a reference wind farm power and a reference wake velocity field when the wind turbine does not perform active yaw are calculated as reference data.

[0096] In step 402, based on the initial operating state and in combination with the wake model, the benchmark wind farm power and benchmark wake velocity field when the wind turbine is not performing active yaw are calculated. This step calculates the overall power output and wake velocity field distribution of the wind farm under the current operating state by simulating the wake interaction between wind turbines. The benchmark wind farm power and benchmark wake velocity field serve as benchmark data, providing a reference point for subsequent hierarchical optimization. Through this step, the actual situation of the power generation efficiency and wake effect of the current wind farm can be clarified, providing the necessary basic data for the subsequent optimization process, ensuring that the optimization process can accurately evaluate the effect of the active yaw strategy.

[0097] In some embodiments of the present application, in step 302, a rough search is performed based on the reference data in combination with a wake model to obtain the first yaw angle, and the implementation process includes but is not limited to the following steps.

[0098] Step 501 , within a preset yaw angle coarse search range, traverse each yaw angle according to a preset coarse search step angle, and calculate the wind farm power corresponding to the yaw angle using a wake model to obtain a coarse search wind farm power set.

[0099] In step 501, within the preset yaw angle coarse search range, each yaw angle is traversed according to the preset coarse search step angle, and the corresponding wind farm power is calculated using the wake model. The purpose of this step is to obtain wind farm power data at different yaw angles by performing a rapid search within a larger yaw angle range. By traversing the yaw angles within the preset range, the impact of different yaw strategies on wind farm power can be fully understood, thereby providing data support for the subsequent determination of the first yaw angle. This step simulates the mutual influence between wind turbines through the wake model, calculates the wind farm power at each yaw angle, and provides basic data for the subsequent optimization process.

[0100] Step 502: taking the yaw angle corresponding to the maximum value in the rough search wind farm power set as the first yaw angle.

[0101] In step 502, the yaw angle corresponding to the maximum value in the rough search wind farm power set is used as the first yaw angle. The purpose of this step is to quickly determine an approximate optimal solution from the large amount of data obtained in the rough search stage. By comparing the wind farm power under different yaw angles, the yaw angle corresponding to the maximum power value is selected as the first yaw angle. This step provides a clear search starting point for the subsequent fine search stage, narrows the search range, and improves the efficiency of the optimization process. Through this step, a yaw angle close to the optimal can be quickly determined, laying the foundation for subsequent precise optimization.

[0102] In some embodiments of the present application, the implementation process of the rough search phase includes but is not limited to the following steps.

[0103] First, the yaw angle in the coarse search phase is set to , yaw angle coarse search range , coarse search step angle .

[0104] Then, within the preset yaw angle coarse search range, each yaw angle is traversed according to the preset coarse search step angle, and the corresponding wind farm power is calculated using the wake model to obtain the coarse search wind farm power set. Corresponding wind farm power Satisfies the following formula (1):

[0105] (1);

[0106] In formula (1), represents the wake model, Indicates wind speed, Indicates wind direction. Indicates wind turbine status information.

[0107] Finally, the yaw angle corresponding to the maximum value in the rough search wind farm power set is taken as the first yaw angle , Satisfies the following formula (2):

[0108] (2).

[0109] In some embodiments of the present application, in step 303, according to the first yaw angle, combined with the wake model, a greedy strategy is adopted to perform a detailed search to obtain the optimal wind farm power and its corresponding optimal yaw angle, and the implementation process includes but is not limited to the following steps.

[0110] Step 601: setting a yaw angle detailed search range according to a first yaw angle.

[0111] In step 601, a yaw angle fine search range is set according to the first yaw angle. The purpose of this step is to determine a smaller range near the first yaw angle found in the coarse search phase for a more detailed search. By setting a fine search range, computing resources and attention can be focused on the area that is most likely to contain the optimal solution, thereby improving the accuracy and efficiency of the search. This step provides a clear search interval for subsequent fine searches, ensuring that the optimization process can conduct in-depth exploration in key areas.

[0112] Step 602: within the yaw angle fine search range, taking the first yaw angle as a starting point, traverse multiple yaw angles according to a preset fine search step angle.

[0113] In step 602, within the yaw angle fine search range, starting from the first yaw angle, the yaw angle is gradually adjusted with a smaller step size to calculate the wind farm power corresponding to each yaw angle. By traversing multiple yaw angles, the impact of different yaw strategies on wind farm power can be more finely evaluated, so as to find a more optimal yaw angle. This step provides necessary data support for the subsequent greedy strategy by gradually adjusting the yaw angle and calculating the wind farm power in combination with the wake model.

[0114] Step 603: Obtain a difference between the wind farm power at the current yaw angle and the wind farm power at the previous yaw angle as a power increment at the current yaw angle.

[0115] In step 603, the power increment is calculated by comparing the wind farm power under adjacent yaw angles, thereby evaluating the impact of yaw angle adjustment on wind farm power. The calculation of power increment provides a decision basis for the subsequent greedy strategy, helping to determine whether to continue adjusting the yaw angle. Through this step, the effect of yaw angle adjustment can be monitored in real time to ensure that the optimization process is carried out in the direction of increasing wind farm power.

[0116] Step 604, when the power increment of the current yaw angle is greater than the preset value, the next yaw angle is determined as the current yaw angle, and the process returns to the step of obtaining the difference between the wind farm power at the current yaw angle and the wind farm power at the previous yaw angle as the power increment of the current yaw angle, until the power increment of the current yaw angle is less than the preset value, and the previous yaw angle is determined as the optimal yaw angle.

[0117] In step 604, when the power increment of the current yaw angle is greater than the preset value, the next yaw angle is determined as the current yaw angle, and the process returns to the step of obtaining the difference between the wind farm power at the current yaw angle and the wind farm power at the previous yaw angle as the power increment of the current yaw angle, until the power increment of the current yaw angle is less than the preset value, and the previous yaw angle is determined as the optimal yaw angle. This step is the termination condition of the fine search phase. The yaw angle is gradually adjusted through a greedy strategy until the power increment no longer increases effectively, thereby determining the optimal yaw angle. This step ensures that the optimization process can stop at the local optimal solution, avoids unnecessary calculations, and improves the optimization efficiency. Through this step, the optimal yaw angle can be accurately determined to maximize the power of the wind farm.

[0118] In some embodiments of the present application, the implementation process of the detailed search phase includes but is not limited to the following steps.

[0119] First, set the yaw angle search range to the first yaw angle , fine search step angle .

[0120] Secondly, within the yaw angle fine search range, take the first yaw angle as the starting point, and initialize the yaw angle in the fine search phase to , according to the preset fine search step angle, traverse multiple yaw angles.

[0121] Then, the difference between the wind farm power at the current yaw angle and the wind farm power at the previous yaw angle is obtained as the power increment of the current yaw angle. .when When, update . Keep updating until , the previous yaw angle is determined as the optimal yaw angle. At this time, the optimal yaw angle .

[0122] In some embodiments of the present application, the wake model includes a Jimenez wake model, a Gauss deflection model, and a wake accumulation curl model.

[0123] The Jimenez wake model is used to simulate the velocity attenuation characteristics of the wake in the downstream. The model can accurately predict the velocity distribution of the wake in the downstream area by considering factors such as turbulent diffusion and velocity recovery in the wake. This is crucial for evaluating the impact of the wake on downstream wind turbines, as it can help determine the extent to which the wake weakens the wind speed of the downstream wind turbine, thereby providing a basis for optimizing the yaw angle of the wind turbine. By using the Jimenez wake model, the propagation and attenuation process of the wake in the wind farm can be simulated more accurately, thereby improving the accuracy of yaw control optimization.

[0124] The Gauss deflection model is used to describe the directional deflection of the wake. Based on the Gaussian distribution assumption, the model takes into account the diffusion characteristics of the wake in the lateral and vertical directions, and can accurately predict the directional changes of the wake in the downstream area. This is crucial for evaluating the yaw effect of the wake on the downstream wind turbine, because it can help determine the deflection angle of the wake in the downstream area, thereby providing a basis for optimizing the yaw angle of the wind turbine. By using the Gauss deflection model, the deflection and diffusion process of the wake in the wind farm can be simulated more accurately, thereby improving the accuracy of yaw control optimization.

[0125] The wake accumulation curling model is used to consider the interaction between multiple wind turbine wakes. The model can accurately predict the combined impact of multiple wind turbine wakes in the downstream area by considering the superposition and interference effects of multiple wakes in space. This is crucial for evaluating the mutual impact between multiple wind turbines in a wind farm, as it can help determine the cumulative effect of multiple wakes in the downstream area, thereby providing a basis for optimizing the yaw angle of the wind turbine. By using the wake accumulation curling model, the interaction process of multiple wakes in a wind farm can be more accurately simulated, thereby improving the accuracy of yaw control optimization.

[0126] Secondly, refer to Figure 2 The embodiment of the present application provides a wind farm yaw control system based on hierarchical optimization, including: a data collection module, a data preprocessing module, a hierarchical optimization module and a yaw control module.

[0127] The data collection module is used to collect real-time data of the wind farm, including wind speed, wind direction and wind turbine status information.

[0128] The data preprocessing module is used to preprocess the real-time data to obtain initial wind farm data.

[0129] The layered optimization module is used to combine the wake model to perform benchmark calculations and layered optimization on the initial wind farm data to obtain the optimal yaw angle.

[0130] The yaw control module is used to control the wind turbine to perform yaw according to the optimal yaw angle.

[0131] Furthermore, refer to Figure 3 The embodiment of the present application provides a wind farm yaw control device based on hierarchical optimization, including: a wind turbine, a wind tower, a laser radar, a wind turbine SCADA system device, an edge computing center device and a centralized data center device. The edge computing center device is on the wind turbine. The wind tower and the laser radar are used to collect the wind speed and wind direction of the wind farm. The wind turbine SCADA system is used to collect wind turbine status information. The wind speed, wind direction and wind turbine status information are used as real-time data.

[0132] The function of a wind tower is to collect basic data. It monitors wind speed and direction information in a wind farm in real time through sensors such as anemometers and wind vanes. These data are crucial for evaluating the operating environment of wind turbines and adjusting the yaw angle, because real-time changes in wind speed and direction directly affect the wind turbine's wind energy capture efficiency. The accurate data provided by the wind tower can ensure that the yaw control system can respond to wind farm changes in a timely manner, thereby optimizing the operating status of the wind turbine and improving the overall power generation efficiency of the wind farm.

[0133] The role of LiDAR is to measure wind speed and direction with high precision. It uses laser technology to measure wind speed and direction in wind farms at long distances and with high precision, providing more accurate wind farm data for yaw control systems. Compared with wind towers, LiDAR has higher measurement accuracy and longer measurement distance, and can more accurately capture subtle changes in wind farms. This is indispensable for control systems that need to precisely adjust the yaw angle to optimize wind turbine performance. The precise data provided by LiDAR allows for more effective yaw control, reduces wake effects, and improves wind farm power generation efficiency.

[0134] The wind turbine SCADA system device is the core data acquisition and execution unit in the yaw control device of the wind farm. It is responsible for collecting various status information of the wind turbine, such as yaw angle, pitch angle, speed and power, and transmitting this information to the edge computing center. At the same time, it also receives yaw control instructions from the edge computing center and controls the wind turbine to perform corresponding yaw operations according to the instructions. The SCADA system device monitors the operating status of the wind turbine in real time to ensure that the yaw control system can obtain the necessary data in a timely manner, thereby realizing precise control of the wind turbine. This is of great significance to improving the operating efficiency and reliability of wind farms.

[0135] The edge computing center device is used to pre-process the real-time data, obtain the initial wind farm data, and send the initial wind farm data to the centralized data center. The centralized data center device is used to combine the wake model to perform benchmark calculations and hierarchical optimization on the initial wind farm data to obtain the optimal wind farm power and the optimal yaw angle. According to the optimal wind farm power and the optimal yaw angle, a yaw control instruction is generated, and the yaw control instruction is sent to the edge computing center. The edge computing center device is also used to calibrate the yaw state of the wind turbine and send the yaw control instruction to the wind turbine SCADA system device. The wind turbine SCADA system device is also used to control the wind turbine to perform yaw according to the yaw control instruction.

[0136] The edge computing center device plays a role in data preprocessing and preliminary analysis in the yaw control device of the wind farm. It is installed on the wind turbine and can receive data from the wind tower, lidar and wind turbine SCADA system in real time, and preprocess these data to obtain initial wind farm data. The edge computing center ensures the accuracy and consistency of the data through operations such as data screening, cleaning and synchronization, providing a reliable data basis for subsequent optimization calculations. In addition, it is also responsible for sending the preprocessed data to the centralized data center and receiving the yaw control instructions from the centralized data center, further checking the yaw status of the wind turbine, and ensuring the accurate execution of the control instructions. This helps to improve the real-time and response speed of yaw control, thereby improving the overall performance of the wind farm.

[0137] The centralized data center device is the core computing and decision-making unit in the wind farm yaw control device. It is responsible for receiving the initial wind farm data from the edge computing center, and performing benchmark calculations and hierarchical optimization in combination with the wake model. Through complex calculations and analysis, the centralized data center can determine the optimal wind farm power and optimal yaw angle, and generate corresponding yaw control instructions. These instructions are then sent to the edge computing center to guide the yaw operation of the wind turbine. The centralized data center ensures the scientificity and effectiveness of the yaw control strategy through efficient data processing and optimization algorithms, thereby maximizing the power generation efficiency of the wind farm.

[0138] In some embodiments of the present application, reference Figures 4 to 6 , taking a wind farm consisting of 30 wind turbines with a rated power of 2MW as an example, Figure 4 Part a) shows the active yaw angles of 30 wind turbines at different times. In the current example, the maximum active yaw angle of the wind turbine is 25°, and a large number of wind turbines perform active yaw between 2pm and 7pm. Figure 4 Part b) shows the percentage of wind turbines in the wind farm that implement active yaw. At most, about 30% of the wind turbines implement the active yaw strategy, which shows that the power output of the wind farm can be increased by implementing active yaw in fewer wind turbines. Figure 4 Part c) in the middle shows the wind farm power output during the daily cycle. Figure 5 This is a comparison chart of wind speed, wind direction, and power increase of the wind farm within a daily cycle. It can be seen that between 2 pm and 7 pm, the wind speed and direction change greatly, and the power increase effect is more prominent at this time. This indicates that under conditions with high turbulence, the implementation of the active yaw control strategy has higher benefits. The entire wind farm has increased its power output by approximately 8 MW within a daily cycle, which is equivalent to adding approximately 4 wind turbines. Figure 6 Schematic diagram of the wake effect of the wind turbine when active yaw optimization is performed at a certain moment. It can be seen that the wake of the upstream wind turbine actively yaws, which makes its wake bypass the downstream wind turbine.

[0139] In addition, an embodiment of the present application further provides a computer medium, in which a program executable by a processor is stored. When the program executable by the processor is executed by the processor, it is used to implement the aforementioned wind farm yaw control method based on hierarchical optimization.

[0140] In summary, the wind farm yaw control method, system, device and medium based on hierarchical optimization provided in the embodiments of the present application have the following technical effects.

[0141] The embodiment of the present application uses a hierarchical optimization strategy of coarse-fine search to quickly determine the approximate optimal solution in the coarse search stage, narrow the search range, and then accurately find the optimal solution in the fine search stage, effectively improving the search efficiency and optimization accuracy. This method effectively reduces the amount of calculation, can quickly adapt to the dynamic changes of the wind farm, and adjust the yaw angle of the wind turbine in time, thereby minimizing the wake impact of the upstream wind turbine on the downstream wind turbine and improving the power generation efficiency of the entire wind farm. In addition, this method combines a variety of wake models, which can more accurately simulate the mutual influence between wind turbines, further improving the accuracy and reliability of the optimization results.

[0142] At the same time, the embodiment of the present application realizes efficient data processing and rapid generation and execution of optimization instructions through the collaborative work of the edge computing center and the centralized data center. The edge computing center preprocesses real-time data, reduces the delay of data transmission and processing, and ensures the real-time and accuracy of the optimization process. The centralized data center is responsible for complex optimization calculations, generates optimal yaw control instructions, and quickly transmits them to the wind turbine through the edge computing center, realizing the timely application of optimization results. This architecture not only improves the response speed of the system, but also enhances the stability and reliability of the system, ensuring that the wind farm can maintain the best operating state under different working conditions, and ultimately achieves an effective increase in the overall power generation of the wind farm.

[0143] In some optional embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation schematic diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the application is provided by way of example, for the purpose of providing a more comprehensive understanding of the technology. The disclosed method is not limited to the operation and logic flow presented herein. The optional embodiment is expected, wherein the order of various operations is changed and the sub-operation of a part of the larger operation is wherein described is performed independently.

[0144] In addition, although the present application is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in a separate physical device or software module. It can also be understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present application. More specifically, in view of the properties, functions and internal relationships of the various functional modules in the devices disclosed herein, the actual implementation of the module will be understood within the conventional techniques of engineers. Therefore, those skilled in the art can implement the present application as set forth in the claims using ordinary techniques. It can also be understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present application, which is determined by the full scope of the attached claims and their equivalents.

[0145] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several programs to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0146] The logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable programs for implementing logical functions, and may be embodied in any computer-readable medium for use by a program execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch and execute a program from a program execution system, device or apparatus), or in conjunction with such program execution system, device or apparatus. For purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by a program execution system, device or apparatus, or in conjunction with such program execution system, device or apparatus.

[0147] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, for example, the program may be obtained electronically by optically scanning the paper or other medium, then editing, interpreting or processing it in a suitable manner as necessary, and then storing it in a computer memory.

[0148] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable program execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0149] In the above description of this specification, the description with reference to the terms "one embodiment / implementation", "another embodiment / implementation" or "certain embodiments / implementations" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiments or examples are included in the embodiments or examples of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0150] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0151] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present invention.

Claims

1. A wind farm yaw control method based on hierarchical optimization, characterized in that: The steps include: Collecting real-time data of the wind farm; the real-time data includes wind speed, wind direction and wind turbine status information; Preprocessing the real-time data to obtain initial wind farm data; Combined with the wake model, benchmark calculation and hierarchical optimization are performed on the initial wind farm data to obtain an optimal yaw angle; The wake model is used to simulate the mutual influence between wind turbines; the hierarchical optimization includes a coarse search phase and a fine search phase; the coarse search phase refers to the process of quickly finding an approximate optimal solution with a larger step size within a larger parameter range; the fine search phase refers to the process of accurately finding an optimal solution with a smaller step size near the approximate optimal solution found in the coarse search phase; According to the optimal yaw angle, controlling the wind turbine to perform yaw; The preprocessing of the real-time data to obtain initial wind farm data includes: Removing outliers and invalid data from the real-time data to filter out valid data; Performing interpolation processing on the valid data to fill in the gaps in the valid data to obtain complete data; Performing time stamp synchronization processing on the complete data so that the time stamps of the complete data from different sources are consistent, thereby obtaining the initial wind farm data; The combining of the wake model, performing benchmark calculation and hierarchical optimization on the initial wind farm data to obtain the optimal yaw angle, includes: In combination with the wake model, a benchmark calculation is performed on the initial wind farm data to obtain benchmark data; Perform a rough search based on the benchmark data and the wake model to obtain a first yaw angle; According to the first yaw angle, in combination with the wake model, a greedy strategy is adopted to perform a detailed search to obtain the optimal yaw angle; The wake model includes a Jimenez wake model, a Gauss deflection model and a wake accumulation curling model; wherein the Jimenez wake model is used to simulate the velocity attenuation characteristics of the wake in the downstream; the Gauss deflection model is used to describe the directional deflection of the wake; and the wake accumulation curling model is used to consider the interaction between the wakes of multiple wind turbines.

2. The wind farm yaw control method based on hierarchical optimization according to claim 1 is characterized in that: The step of combining the wake model with the initial wind farm data to perform a benchmark calculation to obtain benchmark data includes: Determining the initial operating state of the wind turbine according to the initial wind farm data; the initial operating state includes yaw angle, pitch angle, rotation speed and power; According to the initial operating state and in combination with the wake model, a reference wind farm power and a reference wake velocity field when the wind turbine does not perform active yaw are calculated and obtained as the reference data.

3. The wind farm yaw control method based on hierarchical optimization according to claim 1, characterized in that: The step of performing a rough search based on the reference data and in combination with the wake model to obtain a first yaw angle includes: Within a preset yaw angle coarse search range, traverse each yaw angle according to a preset coarse search step angle, and calculate the wind farm power corresponding to the yaw angle using the wake model to obtain a coarse search wind farm power set; The yaw angle corresponding to the maximum value in the rough search wind farm power set is used as the first yaw angle.

4. The wind farm yaw control method based on hierarchical optimization according to claim 1, characterized in that: The step of performing a detailed search based on the first yaw angle and the wake model using a greedy strategy to obtain the optimal yaw angle includes: According to the first yaw angle, setting a yaw angle detailed search range; In the yaw angle fine search range, taking the first yaw angle as a starting point, traversing multiple yaw angles according to a preset fine search step angle; Acquire a difference between the wind farm power at a current yaw angle and the wind farm power at a previous yaw angle as a power increment at the current yaw angle; When the power increment of the current yaw angle is greater than a preset value, the next yaw angle is determined as the current yaw angle, and the process returns to the step of obtaining the difference between the wind farm power at the current yaw angle and the wind farm power at the previous yaw angle as the power increment of the current yaw angle, until the power increment of the current yaw angle is less than the preset value, and the previous yaw angle is determined as the optimal yaw angle.

5. A wind farm yaw control system based on hierarchical optimization, characterized in that: The wind farm yaw control system is used to implement the wind farm yaw control method based on hierarchical optimization as described in any one of claims 1 to 4.

6. A wind farm yaw control device based on hierarchical optimization, characterized in that: The wind farm yaw control device is used to implement the wind farm yaw control system based on hierarchical optimization as claimed in claim 5.

7. A computer medium storing a program executable by a processor, characterized in that: The program executable by the processor is used to implement the wind farm yaw control method based on hierarchical optimization as claimed in any one of claims 1 to 4 when executed by the processor.

Images