Fan yaw cooperative control method and system for offshore floating type wind power plant

Abstract

The invention discloses a fan yaw cooperative control method and system for an offshore floating type wind power plant, and relates to the technical field of offshore wind power, and the method comprises the steps: constructing a digital twinborn body of the offshore floating type wind power plant, and obtaining a twinborn wind power plant; the method comprises the following steps: locally calling historical wind regime data, solving a yaw coordination strategy in a twin wind power plant based on the historical wind regime data, and constructing a reference coordination control library; accessing a far-end weather forecast center, and acquiring predicted wind regime data falling into a prediction time window; traversing the reference cooperative control library based on the predicted wind regime data, and outputting a reference cooperative yaw angle; taking the reference cooperative yaw angle as an initial point, performing yaw cooperative control optimization in the twin wind power plant according to the predicted wind regime data, and outputting a predicted cooperative yaw angle; and in the prediction time window, performing power generation intervention on the offshore floating type wind power plant according to the predicted cooperative yaw angle. Therefore, the technical effects of prospective power generation intervention and improvement of power generation efficiency and stability are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of offshore wind power, in particular to a wind turbine yawing cooperative control method and system for offshore floating wind farm. BACKGROUND

[0002] In the operation process of offshore floating wind farm, wind turbine yawing control is crucial to improve power generation efficiency. The traditional wind turbine yawing control method of offshore floating wind farm mainly adjusts the yawing of individual wind turbine according to real-time wind condition data.

[0003] The traditional wind turbine yawing control of offshore floating wind farm has the following technical defects: first, it only relies on real-time wind condition data and cannot fully utilize the rules and information in historical wind condition data; second, it cannot effectively combine meteorological forecast data for forward-looking control, and cannot predict the impact of wind condition changes on the cooperative operation of the entire wind farm in advance, thereby affecting the overall efficiency and equipment safety of the wind farm. SUMMARY

[0004] The present application provides a wind turbine yawing cooperative control method and system for offshore floating wind farm to solve the technical problems of lack of wind condition change and historical rule perspective, single control dimension and insufficient prediction in the prior art, and achieves the technical effects of forward-looking power generation intervention and improved power generation efficiency and stability.

[0005] In a first aspect, the present application provides a wind turbine yawing cooperative control method for offshore floating wind farm, wherein the wind turbine yawing cooperative control method for offshore floating wind farm comprises:

[0006] Constructing a digital twin of offshore floating wind farm to obtain a twin wind farm.

[0007] After locally calling historical wind condition data, yawing cooperative strategy solving is performed in the twin wind farm according to the historical wind condition data to construct a baseline cooperative control library.

[0008] Accessing a remote meteorological forecast center to call predicted wind condition data falling within a prediction time window.

[0009] Traversing the baseline cooperative control library using the predicted wind condition data to output a baseline cooperative yawing angle.

[0010] Taking the baseline cooperative yawing angle as a starting point, yawing cooperative control optimization is performed in the twin wind farm according to the predicted wind condition data to output a predicted cooperative yawing angle.

[0011] In the prediction time window, power generation intervention is performed on the offshore floating wind farm according to the predicted cooperative yawing angle.

[0012] In a feasible implementation, a digital twin of an offshore floating wind farm is constructed to obtain a twin wind farm, comprising:

[0013] Design parameters of the offshore floating wind farm are collected to establish a wind farm control coupling model.

[0014] A CFD wake model is integrated into the wind farm control coupling model to quantify the wake superposition effect and the space-time difference effect of the wind turbine cluster, and the twin wind farm is output.

[0015] According to the historical wind condition data, the historical wind turbine yawing coordination data and the historical power generation data are called in time sequence synchronization.

[0016] The historical wind condition data, the historical wind turbine yawing coordination data and the historical power generation data are used to calibrate the accuracy of the wake field distribution of the twin wind farm.

[0017] In a feasible implementation, after calling the historical wind condition data locally, the yawing coordination strategy is solved in the twin wind farm according to the historical wind condition data to construct a benchmark coordination control library, comprising:

[0018] The historical wind condition data is processed based on interval discretization to obtain a plurality of sample discrete wind condition scenarios.

[0019] The plurality of sample coordination yawing angles of the plurality of sample discrete wind condition scenarios are solved in the twin wind farm with the goal of maximizing the overall power generation.

[0020] The plurality of sample discrete wind condition scenarios and the plurality of sample coordination yawing angles are stored in association to output the benchmark coordination control library.

[0021] In a feasible implementation, the historical wind condition data is processed based on interval discretization to obtain a plurality of sample discrete wind condition scenarios, comprising:

[0022] A discretization core parameter index is set, wherein the discretization core parameter index is composed of a wind speed index, a wind direction index and a sea wave height index.

[0023] The index interval division rule of the discretization core parameter index is set.

[0024] A plurality of historical wind condition records corresponding to a plurality of historical collection nodes are called from the historical wind condition data with the discretization core parameter index as a constraint.

[0025] The plurality of historical wind condition records are divided into a plurality of wind condition scenario discrete matrices according to the index interval division rule.

[0026] The plurality of wind condition scenario discrete matrices are aggregated to obtain the plurality of sample discrete wind condition scenarios.

[0027] In a feasible implementation, multiple sample cooperative yaw angles of the multiple sample discrete wind condition scenarios are solved in the twin wind farm, aiming at maximizing the total power generation, including:

[0028] The historical cooperative yaw angles are called by historical yaw characteristics using the first sample discrete wind condition scenario, and multiple historical cooperative power generations of multiple historical cooperative yaw angles are obtained.

[0029] According to the multiple historical cooperative power generations, a reference cooperative yaw angle is screened from the multiple historical cooperative yaw angles.

[0030] The first sample discrete wind condition scenario and the reference cooperative yaw angle are loaded in the twin wind farm to initialize the wind turbine layout and the floating platform dynamic parameters.

[0031] The gradient descent method is used to adjust the wind turbine yaw angle in the twin wind farm starting from the reference cooperative yaw angle, until the first sample cooperative yaw angle maximizing the total power generation is output.

[0032] In a feasible implementation, the reference cooperative yaw angle is output by traversing the reference cooperative control library using the predicted wind condition data, including:

[0033] According to the numerical inclusion relationship between the predicted wind condition data and the multiple sample discrete wind condition scenarios in the reference cooperative control library, the reference cooperative yaw angle is located in the multiple sample cooperative yaw angles.

[0034] The reference cooperative yaw angle is called and output to the twin wind farm for cooperative yaw control optimization.

[0035] In a feasible implementation, the power generation intervention of the offshore floating wind farm is performed according to the predicted cooperative yaw angle within the prediction time window, including:

[0036] The real-time cooperative yaw angle is called.

[0037] The cooperative difference quantitative value of the real-time cooperative yaw angle and the predicted cooperative yaw angle is calculated.

[0038] If the cooperative difference quantitative value is less than the preset difference threshold, the real-time cooperative yaw angle is continued to perform the power generation intervention of the offshore floating wind farm within the prediction time window.

[0039] If the cooperative difference quantitative value is greater than the preset difference threshold, the power generation intervention of the offshore floating wind farm is performed according to the predicted cooperative yaw angle within the prediction time window.

[0040] In a second aspect, the present application further provides a yawing cooperative control system for wind turbines of a floating offshore wind farm, wherein the yawing cooperative control system for wind turbines of the floating offshore wind farm comprises:

[0041] a twin body construction module for constructing a digital twin of the floating offshore wind farm to obtain a twin wind farm.

[0042] a benchmark library construction module for solving a yawing cooperative strategy in the twin wind farm according to historical wind condition data to construct a benchmark cooperative control library after locally calling the historical wind condition data.

[0043] a weather forecast access module for accessing a remote weather forecast center to call predicted wind condition data falling within a prediction time window.

[0044] a yaw angle matching module for traversing the benchmark cooperative control library using the predicted wind condition data to output a benchmark cooperative yaw angle.

[0045] a control optimization module for performing yawing cooperative control optimization in the twin wind farm according to the predicted wind condition data with the benchmark cooperative yaw angle as a starting point to output a predicted cooperative yaw angle.

[0046] a power generation intervention module for performing power generation intervention of the floating offshore wind farm according to the predicted cooperative yaw angle within the prediction time window.

[0047] The present application discloses a yawing cooperative control method and system for wind turbines of a floating offshore wind farm, comprising: constructing a digital twin model corresponding to a target floating offshore wind farm to form a twin wind farm for control simulation; performing yawing cooperative control strategy solving in the twin wind farm based on locally called historical wind condition data to construct and store a benchmark cooperative control strategy library; accessing a remote weather forecast service to obtain predicted wind condition data within a target prediction time window; inputting the predicted wind condition data into the benchmark cooperative control strategy library to perform control strategy matching and output a corresponding benchmark cooperative yaw angle; performing cooperative control optimization in the twin wind farm based on the predicted wind condition data with the benchmark cooperative yaw angle as an initial control condition to obtain an optimized predicted cooperative yaw angle; and implementing yawing control intervention of the floating offshore wind farm according to the predicted cooperative yaw angle within the prediction time window. The yawing cooperative control method and system for wind turbines of a floating offshore wind farm disclosed in the present application solve the technical problems of lacking wind condition change and historical rule perspective, single control dimension, and insufficient predictability, and achieve the technical effects of forward-looking power generation intervention and improved power generation efficiency and stability. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 It is a flowchart of the yawing cooperative control method for wind turbines of a floating offshore wind farm.

[0049] Figure 2 Fig. 1 is a structural schematic diagram of a yaw cooperative control system of a wind turbine of a floating offshore wind farm according to the present application.

[0050] Fig. 1 is a structural schematic diagram of a yaw cooperative control system of a wind turbine of a floating offshore wind farm according to the present application. DETAILED DESCRIPTION

[0051] The above technical solutions will be described in detail below in combination with the drawings and specific embodiments, so as to better understand the above technical solutions. Obviously, the described embodiments are only part of the embodiments of the present application, not all embodiments of the present application, and it should be understood that the present application is not limited to the example embodiments for explaining the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application. In addition, it should be noted that, for convenience of description, only the parts related to the present application are shown in the drawings, not all.

[0052] Embodiment one, as Figure 1 Fig. 1 is a structural schematic diagram of a yaw cooperative control system of a wind turbine of a floating offshore wind farm according to the present application.

[0053] S100: Construct a digital twin of a floating offshore wind farm to obtain a twin wind farm.

[0054] Specifically, the digital twin of a floating offshore wind farm refers to a virtual simulation model and real-time synchronization mechanism based on the structure, state and environmental data of an actual floating wind farm, which is used to simulate, predict and feedback the running state of various devices in the wind farm and environmental changes. The twin wind farm is the mapping result of the digital twin in operation to the real wind farm, which has the ability of synchronous data input, state evolution and behavior output with the real system.

[0055] The twin wind farm obtained through the above construction process can simulate highly consistent dynamic response and predict parameters without directly intervening in the actual wind turbine, providing real-time visual feedback and prediction evaluation basis for the coordinated adjustment of the yaw angle of the wind turbine.

[0056] In some embodiments, constructing a digital twin of a floating offshore wind farm to obtain a twin wind farm includes:

[0057] The design parameters of the offshore floating wind farm are collected, and a wind farm control coupling model is established; the CFD wake model is integrated in the wind farm control coupling model to quantify the wake superposition effect and the time and space difference effect, and the twin wind farm is output; according to the historical wind condition data, the historical wind turbine yawing cooperation data and the historical power generation data are called in time sequence synchronization; the historical wind condition data, the historical wind turbine yawing cooperation data and the historical power generation data are used to calibrate the accuracy of the wake field distribution of the twin wind farm.

[0058] Specifically, a digital twin of an offshore floating wind farm is constructed, which is based on the physical structure and operating characteristics of the actual wind farm, and generates a virtual model system capable of dynamically simulating and feeding back the state of the wind farm through modeling and data synchronization. The digital twin includes a wind farm control coupling model, a wake influence modeling module, and a data calibration module. The control coupling model is used to simulate the operating control relationship between multiple floating wind turbines, the CFD wake model is used to analyze the spatial propagation and superposition effect of the wind turbine wake, and the wake field distribution accuracy calibration is used to improve the accuracy of the twin model using historical data.

[0059] Specifically, first, the design parameters of the offshore floating wind farm are collected, including but not limited to the hub height of the wind turbine, the blade length, the six-degree-of-freedom response characteristics of the platform, the mooring parameters of the floating system, the cable layout information, and the wind turbine control strategy parameters, and then a wind farm control coupling model considering attitude disturbance, load transfer and control interaction is established; then, a CFD (Computational Fluid Dynamics) wake model is integrated in the coupling model to simulate the wake generated by a single and multiple wind turbines under complex wind conditions, wherein the time and space attenuation, lateral drift and superposition coupling effects of the wake need to be considered, and then the wake interference intensity and mutual influence relationship between the wind turbines in the wind farm are quantified, and a twin wind farm with wake correlation state mapping capability is output.

[0060] In addition, to enhance the prediction accuracy of the model, the historical wind condition data (such as wind speed and direction time series curve), historical wind turbine yawing cooperation control data and historical power generation data are further called in time sequence synchronization, and the deviation analysis and optimization adjustment of the CFD wake simulation results are performed based on the data, to complete the accuracy calibration of the wake field distribution of the twin wind farm.

[0061] The digital twin wind farm constructed by the above steps not only can accurately reproduce the yawing response and power generation performance of the offshore floating wind farm under specific wind conditions, but also can significantly improve the quantification ability of the wake disturbance effect and the response prediction accuracy of the wind turbine cooperation control. Compared with the traditional static model, the twin system has stronger dynamic consistency and time and space fitting capability, and can support the simulation verification, optimization and rolling update of the wind turbine yawing cooperation control strategy before deployment, thereby improving the overall power generation efficiency and energy scheduling flexibility of the offshore wind farm.

[0062] S200: After calling the historical wind condition data locally, a yaw coordination strategy solving is performed in the digital twin wind farm according to the historical wind condition data to construct a benchmark coordination control library.

[0063] Specifically, the pre-stored wind condition information is extracted from the local storage system, including but not limited to historical data records such as wind speed, wind direction, wind frequency, etc. Then, through a specific algorithm or model, the control strategy that can optimize the yaw coordination of wind turbines is found in the digital twin wind farm according to the historical wind condition data, so as to realize the efficient coordinated operation between wind turbines.

[0064] In some embodiments, after calling the historical wind condition data locally, a yaw coordination strategy solving is performed in the digital twin wind farm according to the historical wind condition data to construct a benchmark coordination control library, including:

[0065] The historical wind condition data is processed based on interval discretization to obtain a plurality of sample discrete wind condition scenarios; a plurality of sample coordinated yaw angles of the plurality of sample discrete wind condition scenarios are solved in the digital twin wind farm with the goal of maximizing the overall power generation; the plurality of sample discrete wind condition scenarios and the plurality of sample coordinated yaw angles are stored in association, and the benchmark coordination control library is output.

[0066] Specifically, interval discretization processing is to divide the originally continuously changing historical wind condition data (such as wind speed, wind direction) into a plurality of representative discrete intervals according to a certain interval division method, so as to form a limited number of typical wind condition scenarios, thereby facilitating subsequent batch optimization and strategy library construction.

[0067] Specifically, the coordinated yaw angle refers to the optimal yaw angle combination of all wind turbines in the wind farm determined through coordinated optimization under a specific wind condition scenario, which aims to improve the overall power generation or reduce the load, rather than the independent optimum of a single wind turbine. The benchmark coordination control library is a control strategy database that can be quickly called for actual operation, which is formed by mapping different wind condition scenarios and corresponding optimal coordinated yaw angles through offline simulation and optimization.

[0068] Specifically, first, the collected historical wind speed, wind direction and other wind condition data of the wind farm in recent years are discretized based on a predetermined discretization processing rule (such as discretization based on wind speed level interval); for example, the wind speed is divided into intervals of 3-5 m / s, 5-7 m / s, 7-9 m / s, etc., the wind direction is divided into intervals of 0-30°, 30-60°, etc., and a plurality of typical wind condition scenario samples (such as wind speed 5-7 m / s and wind direction 30-60°) are formed.

[0069] Then, subsequently, the twin wind farm model is used to cooperatively optimize each discrete wind condition scenario, that is, in each discretized wind condition scenario, the intelligent optimization methods such as genetic algorithm, particle swarm, etc. are used to solve the optimal cooperative yaw angle combination of all wind turbines in the wind farm with the goal of maximizing the total power generation of the wind farm.

[0070] Further, the sample discrete wind condition scenarios and the corresponding sample cooperative yaw angles are associated and stored, and finally a reference cooperative control library is formed. For example, the reference cooperative control library contains the following contents: the yaw angle set corresponding to scenario 1 (wind speed 5-10 m / s, wind direction 45-90°, sea wave height 1-2 m) is {wind turbine A: 15°, wind turbine B: 20°, wind turbine C: 10°};The yaw angle set corresponding to scenario 2 (wind speed 10-15 m / s, wind direction 90-135°, sea wave height 2-3 m) is {wind turbine A: 25°, wind turbine B: 30°, wind turbine C: 20°} and so on.

[0071] Through the above process, the standardization and efficiency of the wind farm operation strategy can be realized. First, the reference cooperative control library established based on historical data and simulation optimization can significantly improve the overall power generation efficiency of the wind farm and reduce the wake loss. Second, the pre-construction of the control library greatly reduces the online calculation burden, so that the wind farm can quickly respond to changes in wind conditions in actual operation, improving the self-adaptability of the wind farm. In addition, this method has good scalability and maintainability, and the control library can be updated regularly according to new historical data collected in the future to continuously optimize the operation strategy of the wind farm.

[0072] In some implementations, the historical wind condition data is discretized based on intervals to obtain a plurality of sample discrete wind condition scenarios, including:

[0073] The discretization core parameter index is set, wherein the discretization core parameter index is composed of a wind speed index, a wind direction index, and a sea wave height index. The index interval division rule of the discretization core parameter index is set. The plurality of historical wind condition records corresponding to the plurality of historical collection nodes are called from the historical wind condition data with the discretization core parameter index as a constraint. The plurality of historical wind condition records are segmented into a plurality of wind condition scene discrete matrices according to the index interval division rule. The plurality of wind condition scene discrete matrices are aggregated to obtain the plurality of sample discrete wind condition scenarios.

[0074] Specifically, the discretization core parameter index refers to the key physical quantity selected when the wind condition is characterized and classified. These parameters directly affect the operation state and power generation efficiency of the wind turbine, including wind speed, wind direction, and sea wave height.

[0075] Specifically, the index interval division rule is a specific method for interval division of the above-mentioned parameters, such as equal interval division, equal frequency division, or adaptive division based on statistical distribution. The historical acquisition node refers to various sensors or measuring points arranged in the wind farm for collecting wind condition data. The wind condition scene discrete matrix refers to the organization of historical wind condition data in the form of a matrix under a specific parameter interval, which is used for subsequent analysis and modeling.

[0076] Specifically, first, the discretization core parameter index is set, such as wind speed (m / s), wind direction (°), and sea wave height (m). Then, for each index, the interval division rule is set, for example, wind speed can be set to 0-4, 4-8, 8-12 m / s, wind direction can be set to 0-60°, 60-120°, 120-180°, etc., and sea wave height can be set to 0-1 m, 1-2 m, 2-3 m, etc. Then, from the historical wind condition data, the wind speed, wind direction, sea wave height, etc. of all historical acquisition nodes in the corresponding time period are extracted, and all historical wind condition records are divided into different wind condition scene discrete matrices according to the interval division rule. For example, at time A, the wind speed of node A is 6 m / s, the wind direction is 75°, and the sea wave height is 1.2 m. This data is classified into the scene of "wind speed 4-8 m / s, wind direction 60-120°, sea wave height 1-2 m".

[0077] Further, similar processing is performed for all acquisition nodes and times to obtain multiple groups of wind condition scene discrete matrices, and the multiple groups of matrices obtained are aggregated according to the interval combination method to obtain a series of representative sample discrete wind condition scenes. Each scene represents a certain range of wind condition characteristics. For example, a sample discrete wind condition scene may include all historical records with wind speed in the range of 5-10 m / s, wind direction in the range of 45-90°, and sea wave height in the range of 1-2 m.

[0078] Through the above process, the historical wind condition data is structured and standardized, and typical wind condition scenes of wind farm operation can be effectively extracted. The interval discretization method based on multiple core parameters not only considers the direct impact of wind speed and wind direction on wind turbine power generation, but also introduces environmental factors such as sea wave height, making the sample scene more representative and applicable. By aggregating a large amount of historical data, the interference of incidental extreme wind conditions on subsequent optimization can be reduced, and the statistical reliability of the sample can be improved.

[0079] In some implementations, a plurality of sample cooperative yaw angles of the plurality of sample discrete wind condition scenes are solved in the twin wind farm with the goal of maximizing the overall power generation, including:

[0080] The historical yawing feature is called under the first sample discrete wind condition scene to obtain multiple historical cooperative power generations of multiple historical cooperative yawing angles; the reference cooperative yawing angle is screened from the multiple historical cooperative yawing angles according to the multiple historical cooperative power generations; the first sample discrete wind condition scene and the reference cooperative yawing angle are loaded in the twin wind farm to initialize the wind turbine layout and the floating platform dynamic parameters; the gradient descent method is used to adjust the wind turbine yawing angle in the twin wind farm with the reference cooperative yawing angle as the starting point until the first sample cooperative yawing angle maximizing the whole field power generation is output.

[0081] Specifically, the historical cooperative yawing angle refers to the specific yawing setting of each wind turbine under the historical wind condition. The historical cooperative power generation refers to the actual power generation under these yawing settings. The reference cooperative yawing angle is the angle combination with the optimal power generation selected from the historical yawing angles by comparing the historical cooperative power generations.

[0082] Specifically, initializing the wind turbine layout and the floating platform dynamic parameters refers to setting the spatial arrangement of the wind turbines and the dynamics state of the floating foundation in the twin wind farm simulation environment.

[0083] Specifically, first, a scene is randomly selected as the first sample discrete wind condition scene, and all yawing angle combinations and their power generation records under similar wind conditions in the historical database are correspondingly called. For example, for the scene of "wind speed 4-8 m / s, wind direction 60-120°, sea wave height 1-2 m", the yawing angles (such as [8°, 10°, 9°], [6°, 11°, 7°], etc.) once used by the corresponding wind turbines in the historical operation and the corresponding whole field power generation are retrieved. Then, the combination with the highest cooperative power generation is selected from these historical cooperative yawing angles as the reference cooperative yawing angle of the scene.

[0084] Further, in the twin wind farm simulation model, the spatial position of the wind turbines, the dynamic parameters of the floating platform, and other simulation environments are initialized based on the sample wind condition scene and the reference cooperative yawing angle. Then, the gradient descent method is used to fine-tune the wind turbine yawing angle with the reference cooperative yawing angle as the initial point. The whole field power generation is calculated after each adjustment, and the optimal cooperative yawing angle under the scene is finally output until the power generation reaches the maximum value.

[0085] Through the above process, the historical operation experience and the simulation optimization capability are fully integrated to further improve the power generation performance. Among them, starting from the historical optimal yawing angle helps to accelerate the convergence speed of the gradient descent method, reduce invalid search, and improve the optimization efficiency; at the same time, the introduction of the twin wind farm ensures the physical feasibility and engineering applicability of the optimization results.

[0086] S300: Access a remote weather forecast center to call predicted wind condition data falling within a prediction time window.

[0087] Specifically, the remote weather forecast center refers to an external weather forecast service center for providing real-time weather forecast data. The predicted wind condition data includes, for example, wind speed, wind direction, air pressure and the like in a future period of time. The prediction time window is a preset time range, which can be determined according to the actual demand of the wind farm and the accuracy of the weather forecast, such as 24 hours, 48 hours or 72 hours, etc., to ensure that the obtained weather forecast data has high reference value. The predicted wind condition data is the prediction information about the future wind speed, wind direction and the like provided by the weather forecast center, which will be used for the forward-looking control of the wind farm.

[0088] S400: Traverse the reference cooperative control library using the predicted wind condition data, and output a reference cooperative yaw angle.

[0089] Specifically, the predicted wind condition data is a continuous data stream of wind speed, wind direction and the like in a future period of time obtained from the wind farm environment perception system or an external weather model. Through the traversal operation, the predicted wind condition can be matched and identified with the scene in the control library, and the yaw angle corresponding to the closest scene is output as the reference cooperative yaw angle, which is used to provide a better optimization starting point for subsequent control optimization, thereby improving the decision-making efficiency of the yaw control of the current wind farm.

[0090] In some embodiments, traversing the reference cooperative control library using the predicted wind condition data and outputting a reference cooperative yaw angle comprises:

[0091] According to the numerical inclusion relationship between the predicted wind condition data and the plurality of sample discrete wind condition scenes in the reference cooperative control library, a reference cooperative yaw angle is positioned among the plurality of sample cooperative yaw angles; the reference cooperative yaw angle is called and output to the twin wind farm for yaw cooperative control optimization.

[0092] Specifically, first, the predicted wind condition data is obtained, which includes wind speed, wind direction, height profile information and the like at each sub-time in the prediction period; then, the predicted wind condition data is compared with the plurality of sample discrete wind condition scenes in the reference cooperative control library in terms of numerical inclusion or similarity, to identify the nearest sample scene corresponding to the predicted wind condition data in the library. For example, the Euclidean distance, wind direction interval envelope, wind speed vector similarity and the like can be used for judgment.

[0093] Further, after identifying the most matching sample, the yaw angle in the corresponding cooperative yaw angle set of the sample is taken as the reference cooperative yaw angle, that is, the cooperative yaw control strategy of the fan group recommended in the current wind condition; finally, the reference cooperative yaw angle is input into the digital twin wind farm control model as the initial input or search guide item of the yaw control optimization, and further iteration and scheduling optimization of the control strategy are carried out.

[0094] The above sample matching and reference yaw angle output process based on predicted wind condition data enables the control system to generate a cooperative control initial value with global reference value in advance before the future wind condition changes, effectively shortens the optimization time delay and avoids entering the invalid solution area; through the high-quality sample strategy pre-embedded in the reference cooperative control library, the minimization control of the wake effect can be realized while ensuring the stability of the overall output power of the wind farm, thereby improving the group power generation efficiency and the utilization rate of the fan life.

[0095] S500: Taking the reference cooperative yaw angle as the starting point, performing yaw cooperative control optimization in the twin wind farm according to the predicted wind condition data, and outputting a predicted cooperative yaw angle.

[0096] Specifically, in actual operation, the yaw strategy of the wind farm needs to be dynamically adjusted according to the future wind condition change: first, the reference cooperative yaw angle obtained by the historical data optimization is taken as the initial solution, and the latest predicted wind condition data (such as wind speed of 7 m / s, wind direction of 90°, sea wave height of 1.5 m, etc.) is loaded into the twin wind farm simulation platform. Then, under the constraint condition of the predicted wind condition, a numerical optimization method (such as gradient descent method or genetic algorithm) is used to cooperatively optimize the yaw angle of each fan.

[0097] Specifically, in the optimization process, the simulation platform calculates the influence of different yaw combinations on the overall power generation in real time, adjusts the yaw angle gradually, until the power generation converges to the maximum value or reaches the preset optimization step number, selects the yaw angle combination that makes the overall power generation reach the maximum value, and finally outputs the optimal cooperative yaw angle under the predicted wind condition as the actual yaw setting of each unit in the next control period, which is real-time issued to the wind farm control system.

[0098] Through the above process, the forward-looking adaptive control of the wind farm on the future wind condition is realized, and the yaw strategy of each unit can be dynamically adjusted according to the predicted environment. Among them, taking the historical optimization result as the starting point, the convergence time of the optimization algorithm is significantly shortened, and the response speed and accuracy of the control strategy are improved. Through the high-precision simulation of the twin wind farm, the physical feasibility and engineering practicability of the optimization result are guaranteed.

[0099] S600: In the predicted time window, the power generation intervention of the offshore floating wind farm is performed according to the predicted cooperative yaw angle.

[0100] In some embodiments, in the predicted time window, the power generation intervention of the offshore floating wind farm is performed according to the predicted cooperative yaw angle, including:

[0101] The real-time cooperative yaw angle is called, the cooperative difference quantization value of the real-time cooperative yaw angle and the predicted cooperative yaw angle is calculated, if the cooperative difference quantization value is less than a preset difference threshold, the real-time cooperative yaw angle is continued to be used to perform the power generation intervention of the offshore floating wind farm in the predicted time window, and if the cooperative difference quantization value is greater than the preset difference threshold, the power generation intervention of the offshore floating wind farm is performed according to the predicted cooperative yaw angle in the predicted time window.

[0102] Specifically, the real-time cooperative yaw angle refers to the combination of the yaw angles actually applied to the wind turbines of the wind farm at present. The cooperative difference quantization value refers to the overall deviation degree between the real-time cooperative yaw angle and the predicted cooperative yaw angle quantitatively measured by a mathematical method such as Euclidean distance, weighted absolute value, etc. The preset difference threshold is a numerical limit preset for judging whether the current control strategy needs to be switched.

[0103] Specifically, in the predicted time window (for example, the next 10 minutes), the real-time cooperative yaw angle of the current wind turbines (for example, [10.0°, 11.0°, 9.0°]) is first obtained and compared with the predicted cooperative yaw angle (for example, [9.8°, 11.2°, 8.4°]) obtained by the foregoing optimization. A cooperative difference quantization method is used, for example, the Euclidean distance of the two groups of yaw angles is calculated.

[0104] Then, the cooperative difference quantization value is compared with the preset difference threshold. If the calculated cooperative difference quantization value is less than the preset difference threshold, it can be considered that the current real-time cooperative yaw angle is highly consistent with the predicted cooperative yaw angle, and no adjustment is needed. The current real-time cooperative yaw angle is continued to be used to perform the power generation intervention of the wind farm, so as to avoid mechanical loss and control system burden caused by frequent adjustment. If the cooperative difference quantization value is greater than the preset difference threshold, it indicates that there is a significant difference between the current actual yaw state and the optimal predicted yaw state, and the yaw angles of the wind turbines will be adjusted according to the predicted cooperative yaw angle, and a new power generation intervention strategy is implemented to improve the overall power generation performance of the wind farm.

[0105] Through the above process, the dynamic adaptive adjustment of the yaw control of the wind farm is realized, which can not only ensure the sensitive response of the system to the change of wind conditions, but also avoid frequent switching of control instructions due to slight deviations, prolong the service life of the equipment and improve the operation stability. In other words, the above steps effectively reduce the energy consumption and maintenance cost of the control system while maximizing the power generation benefit, thereby improving the economy and reliability of the wind farm.

[0106] In summary, the yaw cooperative control method of the offshore floating wind farm provided by the present application has the following technical effects:

[0107] By constructing a digital twin model corresponding to the target offshore floating wind farm, a twin wind farm for control simulation is formed; based on the locally called historical wind condition data, a yaw cooperative control strategy solution is executed in the twin wind farm to construct and store a reference cooperative control strategy library; a remote weather forecast service is accessed to obtain predicted wind condition data within a target prediction time window; the predicted wind condition data is input into the reference cooperative control strategy library for control strategy matching, and the corresponding reference cooperative yaw angle is output; the reference cooperative yaw angle is taken as the initial control condition, and the cooperative control optimization is executed in the twin wind farm based on the predicted wind condition data to obtain an optimized predicted cooperative yaw angle; within the prediction time window, the yaw control intervention of the offshore floating wind farm is implemented according to the predicted cooperative yaw angle, thereby realizing the technical effects of proactive power generation intervention and improving the power generation efficiency and stability.

[0108] Embodiment two, as Figure 2 is a structural schematic diagram of the yaw cooperative control system of the offshore floating wind farm of the present application. For example, Figure 1 The flowchart of the yaw cooperative control method of the offshore floating wind farm of the present application can be realized by the structure as Figure 2 indicated.

[0109] Based on the same idea as the yaw cooperative control method of the offshore floating wind farm in the above embodiment, the yaw cooperative control system of the offshore floating wind farm of the present application also includes:

[0110] The twin body construction module 11 is used to construct the digital twin body of the offshore floating wind farm to obtain the twin wind farm.

[0111] The reference library construction module 12 is used to perform yaw cooperative strategy solution in the twin wind farm according to the historical wind condition data after locally calling the historical wind condition data, so as to construct the reference cooperative control library.

[0112] The weather forecast access module 13 is used to access a remote weather forecast center to call predicted wind condition data falling within a prediction time window.

[0113] A yaw angle matching module 14 is configured to traverse the reference cooperative control library using the predicted wind condition data, and output a reference cooperative yaw angle.

[0114] A control optimization module 15 is configured to perform cooperative yaw control optimization in the twin wind farm based on the reference cooperative yaw angle as a starting point and the predicted wind condition data, and output a predicted cooperative yaw angle.

[0115] A power generation intervention module 16 is configured to perform power generation intervention in the offshore floating wind farm based on the predicted cooperative yaw angle within the predicted time window.

[0116] In some embodiments, the twin body construction module 11 includes:

[0117] A design parameter acquisition and coupling model establishment unit is configured to acquire design parameters of the offshore floating wind farm, and establish a wind farm control coupling model.

[0118] A CFD wake model integration and twin wind farm output unit is configured to integrate a CFD wake model in the wind farm control coupling model to quantify the wake superposition effect and the time-space difference effect of the wind turbine cluster, and output the twin wind farm.

[0119] A historical data time series synchronization calling unit is configured to time series synchronization call historical wind turbine yaw cooperative data and historical power generation data based on the historical wind condition data.

[0120] A wake field distribution precision calibration unit is configured to use the historical wind condition data, historical wind turbine yaw cooperative data, and historical power generation data to calibrate the wake field distribution precision of the twin wind farm.

[0121] In some embodiments, the reference library construction module 12 includes:

[0122] A historical wind condition data discretization processing unit is configured to discretize the historical wind condition data based on interval discretization to obtain a plurality of sample discrete wind condition scenarios.

[0123] A sample cooperative yaw angle solving unit is configured to solve a plurality of sample cooperative yaw angles of the plurality of sample discrete wind condition scenarios in the twin wind farm with the goal of maximizing the total power generation.

[0124] A reference cooperative control library output unit is configured to store the plurality of sample discrete wind condition scenarios and the plurality of sample cooperative yaw angles in association, and output the reference cooperative control library.

[0125] In some implementations, the historical wind condition data discretization processing unit in the reference library construction module 12 includes the following execution steps:

[0126] The discrete core parameter index is set, wherein the discrete core parameter index is composed of a wind speed index, a wind direction index, and a sea wave height index.

[0127] An index interval division rule of the discrete core parameter index is set.

[0128] A plurality of historical wind condition records corresponding to a plurality of historical collection nodes are called from the historical wind condition data, with the discrete core parameter index as a constraint.

[0129] The plurality of historical wind condition records are segmented into a plurality of wind condition scene discrete matrices according to the index interval division rule.

[0130] The plurality of wind condition scene discrete matrices are aggregated to obtain the plurality of sample discrete wind condition scenes.

[0131] In some implementations, the execution steps of the sample cooperative yaw angle solving unit in the benchmark library construction module 12 include:

[0132] A plurality of historical cooperative power generations of a plurality of historical cooperative yaw angles are obtained by performing historical yaw feature calling with the first sample discrete wind condition scene.

[0133] A benchmark cooperative yaw angle is screened from the plurality of historical cooperative yaw angles according to the plurality of historical cooperative power generations.

[0134] The first sample discrete wind condition scene and the benchmark cooperative yaw angle are loaded in the twin wind farm to initialize wind turbine layout and floating platform dynamic parameters.

[0135] The gradient descent method is used to adjust the wind turbine yaw angle in the twin wind farm with the benchmark cooperative yaw angle as a starting point until a first sample cooperative yaw angle that maximizes the overall power generation is output.

[0136] In some embodiments, the yaw angle matching module 14 includes:

[0137] The benchmark cooperative yaw angle positioning unit is configured to position a benchmark cooperative yaw angle from the plurality of sample cooperative yaw angles according to a numerical inclusion relationship between the predicted wind condition data and the plurality of sample discrete wind condition scenes in the benchmark cooperative control library.

[0138] The yaw cooperative control optimization calling unit is configured to call the benchmark cooperative yaw angle output to the twin wind farm for yaw cooperative control optimization.

[0139] In some embodiments, the power generation intervention module 16 includes:

[0140] The real-time cooperative yaw angle calling unit is configured to call a real-time cooperative yaw angle.

[0141] A cooperative difference quantization value calculation unit is configured to calculate a cooperative difference quantization value of the real-time cooperative yaw angle and the predicted cooperative yaw angle.

[0142] A power generation intervention decision unit is configured to, if the cooperative difference quantization value is less than a preset difference threshold, continue the real-time cooperative yaw angle to perform power generation intervention of the offshore floating wind farm in the predicted time window; and if the cooperative difference quantization value is greater than the preset difference threshold, perform power generation intervention of the offshore floating wind farm according to the predicted cooperative yaw angle in the predicted time window.

[0143] It should be understood that the embodiments mentioned in the specification focus on the differences between them and the specific embodiments in the first embodiment, and the wind turbine yaw cooperative control system of the offshore floating wind farm described in the second embodiment is also applicable. For the sake of brevity of the specification, it will not be further expanded here.

[0144] It should be understood that the embodiments disclosed in the present application and the above description can enable those skilled in the art to implement the present application. At the same time, the present application is not limited to the above-mentioned part of the embodiments, it should be understood that the ordinary skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A wind turbine yaw coordinated control method for an offshore floating wind farm, characterized in that: include: Build a digital twin of the offshore floating wind farm to obtain a twin wind farm; After locally calling historical wind condition data, solving a yaw coordination strategy in the twin wind farms based on the historical wind condition data to build a benchmark coordination control library; Access to a remote weather forecast center to retrieve forecast wind data that falls within the forecast time window; Using the predicted wind condition data to traverse the reference collaborative control library, and outputting a reference collaborative yaw angle; Taking the reference coordinated yaw angle as a starting point, performing yaw coordinated control optimization in the twin wind farm according to the predicted wind condition data, and outputting a predicted coordinated yaw angle; Within the prediction time window, power generation intervention of the offshore floating wind farm is performed according to the predicted coordinated yaw angle.

2. The wind turbine yaw coordinated control method for an offshore floating wind farm according to claim 1, characterized in that: Build a digital twin of the offshore floating wind farm to obtain a twin wind farm, including: collecting design parameters of the offshore floating wind farm and establishing a wind farm control coupling model; By integrating the CFD wake model into the wind farm control coupling model to quantify the wind farm wake superposition effect and time-space difference effect, the twin wind farm is output; Based on the historical wind condition data, historical wind turbine yaw coordination data and historical power generation data are synchronously called in a time series; The historical wind condition data, the historical wind turbine yaw coordination data and the historical power generation data are used to calibrate the accuracy of the wake field distribution of the twin wind farm.

3. The wind turbine yaw coordinated control method for an offshore floating wind farm according to claim 2, characterized in that: After locally calling the historical wind condition data, the yaw coordination strategy is solved in the twin wind farms based on the historical wind condition data to build a benchmark coordinated control library, including: Processing the historical wind condition data based on interval discretization to obtain a plurality of sample discrete wind condition scenarios; With the goal of maximizing the power generation of the entire field, solving multiple sample coordinated yaw angles of the multiple sample discrete wind condition scenarios in the twin wind farm; The plurality of sample discrete wind condition scenarios and the plurality of sample coordinated yaw angles are stored in association with each other, and the benchmark coordinated control library is output.

4. The wind turbine yaw coordinated control method for an offshore floating wind farm according to claim 3, characterized in that: The historical wind condition data is processed based on interval discretization to obtain multiple sample discrete wind condition scenarios, including: Setting a discretized core parameter index, wherein the discretized core parameter index is composed of a wind speed index, a wind direction index, and a wave height index; Setting the indicator interval division rule of the discretized core parameter indicator; Using the discretized core parameter index as a constraint, calling a plurality of historical wind condition records corresponding to a plurality of historical collection nodes from the historical wind condition data; dividing the plurality of historical wind condition records into a plurality of wind condition scene discrete matrices according to the index interval division rule; Aggregate the multiple wind scenario discrete matrices to obtain the multiple sample discrete wind scenario.

5. The wind turbine yaw coordinated control method for an offshore floating wind farm according to claim 4, characterized in that: With the goal of maximizing the power generation of the entire field, solving multiple sample coordinated yaw angles of the multiple sample discrete wind condition scenarios in the twin wind farm includes: The first sample discrete wind condition scenario is used to call historical yaw characteristics to obtain multiple historical coordinated power generation at multiple historical coordinated yaw angles; selecting a reference coordinated yaw angle from the plurality of historical coordinated yaw angles according to the plurality of historical coordinated power generation amounts; Loading a first sample discrete wind scenario and the reference coordinated yaw angle in the twin wind farm to initialize wind turbine layout and floating platform dynamic parameters; Taking the reference coordinated yaw angle as a starting point, a gradient descent method is used to adjust the yaw angles of wind turbines in the twin wind farm until a first sample coordinated yaw angle that maximizes the power generation of the entire farm is iteratively output.

6. The wind turbine yaw coordinated control method for an offshore floating wind farm according to claim 3, characterized in that: The predicted wind condition data is used to traverse the reference collaborative control library to output a reference collaborative yaw angle, including: Positioning a reference collaborative yaw angle among the multiple sample collaborative yaw angles according to a numerical inclusion relationship between the predicted wind condition data and the multiple sample discrete wind condition scenarios in the reference collaborative control library; The reference coordinated yaw angle is called and output to the twin wind farm to perform yaw coordinated control optimization.

7. The wind turbine yaw coordinated control method for an offshore floating wind farm according to claim 1, characterized in that: Performing power generation intervention of the offshore floating wind farm according to the predicted coordinated yaw angle within the prediction time window includes: Call real-time collaborative yaw angle; Calculating a quantified value of a cooperative difference between the real-time cooperative yaw angle and the predicted cooperative yaw angle; If the quantized value of the coordinated difference is less than a preset difference threshold, then within the prediction time window, continuing the real-time coordinated yaw angle to perform power generation intervention on the offshore floating wind farm; If the quantized value of the coordination difference is greater than a preset difference threshold, power generation intervention of the offshore floating wind farm is performed according to the predicted coordinated yaw angle within the prediction time window.

8. A wind turbine yaw coordinated control system for an offshore floating wind farm, characterized in that: A method for implementing wind turbine yaw coordinated control of an offshore floating wind farm according to any one of claims 1 to 7, comprising: A twin construction module is used to construct a digital twin of an offshore floating wind farm to obtain a twin wind farm; A benchmark library construction module is used to locally call historical wind condition data and solve the yaw coordination strategy in the twin wind farm based on the historical wind condition data to build a benchmark coordinated control library; The weather forecast access module is used to access the remote weather forecast center to call the forecast wind data falling within the forecast time window; A yaw angle matching module, configured to traverse the reference collaborative control library using the predicted wind condition data and output a reference collaborative yaw angle; a control optimization module, configured to perform yaw coordinated control optimization in the twin wind farm based on the predicted wind condition data, taking the reference coordinated yaw angle as a starting point, and output a predicted coordinated yaw angle; A power generation intervention module is used to perform power generation intervention of the offshore floating wind farm according to the predicted coordinated yaw angle within the predicted time window.

Images