Maximum wind energy capture control method and system for wind turbines

Abstract

The present invention provides a maximum wind energy capture control method and system for a wind turbine generator set, relating to the field of wind power generation. The method comprises: determining a plurality of key wind speed detection positions; establishing a multi-objective tip speed ratio optimization function; obtaining a real-time wind speed matrix based on the plurality of key wind speed detection positions, performing feature extraction, and generating a real-time wind speed feature matrix; generating a tip speed ratio value range and a reference tip speed ratio based on the real-time wind speed feature matrix through a tip speed ratio optimization model; generating a plurality of sampled tip speed ratios based on the tip speed ratio value range; calculating an optimized value of each sampled tip speed ratio through finite element analysis, a multi-objective tip speed ratio optimization function, and a reference tip speed ratio; determining an optimal tip speed ratio of the wind turbine generator set based on the optimized value of each sampled tip speed ratio through a genetic algorithm; and controlling the wind turbine generator set to capture wind energy according to the optimal tip speed ratio of the wind turbine generator set, thereby improving the wind energy capture efficiency of the wind turbine generator set.

Description

Technical Field

[0001] The present invention relates to the field of wind power generation, and in particular to a maximum wind energy capture control method and system applied to a wind turbine generator set. Background Art

[0002] The efficiency of a wind turbine in capturing energy from natural wind is essentially determined by three key factors: the rotor's rotational speed, the incoming wind speed, and the blade pitch angle. The core function of the pitch control mechanism is to dynamically adjust and limit the rotor speed to prevent mechanical failure or excessive aerodynamic noise pollution caused by excessive speed. When the wind speed does not reach the rated threshold, the pitch control system typically remains inactive. At this point, the wind turbine's power conversion efficiency is closely related only to the single variable, the tip speed ratio. Specifically, for any wind speed condition below the rated wind speed, there exists an optimal rotor speed value that matches it and maximizes the efficiency of converting wind energy into electrical energy. Therefore, the core goal of wind turbine maximum wind energy capture control is to precisely control the generator's electromagnetic torque, guiding the rotor speed to gradually approach this theoretically optimal speed value, thereby maximizing wind energy capture efficiency.

[0003] The core of the existing tip speed ratio control method is to calculate the reference speed of the wind rotor based on the real-time measured wind speed data. However, in practical applications, the layout of wind speed sensors is often limited to a single measuring point such as the top of the nacelle. This limitation makes it difficult to fully capture the complex characteristics of the wind field. Factors such as wind shear effect, wake interference, turbulent characteristics of wind speed, and the dynamic response delay of the anemometer itself will significantly affect the accuracy of wind speed measurement. In addition, due to the measurement deviation caused by different installation positions, the wind speed data of a single measuring point is often difficult to truly reflect the effective wind speed distribution within the swept surface of the wind rotor, resulting in low wind energy capture efficiency of the wind turbine.

[0004] Therefore, it is necessary to provide a maximum wind energy capture control method and system applied to a wind turbine generator set, so as to improve the wind energy capture efficiency of the wind turbine generator set. Summary of the Invention

[0005] The present invention provides a maximum wind energy capture control method for a wind turbine generator, comprising: determining a plurality of key wind speed detection positions; establishing a multi-objective tip speed ratio optimization function; obtaining a real-time wind speed matrix based on the plurality of key wind speed detection positions, wherein the real-time wind speed matrix includes a plurality of real-time wind speed vectors, one real-time wind speed vector corresponds to one key wind speed detection position, and the real-time wind speed vector includes the wind speed of the corresponding key wind speed detection position at a plurality of time points; performing feature extraction on the real-time wind speed matrix to generate a real-time wind speed feature matrix; generating a tip speed ratio value range and a reference tip speed ratio based on the real-time wind speed feature matrix through a tip speed ratio optimization model; generating a plurality of sampled tip speed ratios based on the tip speed ratio value range; calculating an optimized value of each sampled tip speed ratio through finite element analysis, the multi-objective tip speed ratio optimization function and the reference tip speed ratio; determining an optimal tip speed ratio of the wind turbine generator based on the optimized value of each sampled tip speed ratio through a genetic algorithm; and controlling the wind turbine generator to capture wind energy according to the optimal tip speed ratio of the wind turbine generator.

[0006] Furthermore, multiple key wind speed detection positions are determined, including: determining multiple wind speed detection positions to be screened; obtaining the wind speed of each wind speed detection position to be screened in multiple test time periods; obtaining the target tip speed ratio of the wind turbine in multiple test time periods based on a hill climbing search algorithm; for each wind speed detection position to be screened, calculating the influence coefficient of the wind speed detection position to be screened on the tip speed ratio of the wind turbine based on the wind speed of the wind speed detection position to be screened in multiple test time periods and the target tip speed ratio of the wind turbine in multiple test time periods; and determining multiple key wind speed detection positions based on the influence coefficient of each wind speed detection position to be screened on the tip speed ratio of the wind turbine.

[0007] Furthermore, based on the influence coefficient of each wind speed detection position to be screened on the tip speed ratio of the wind turbine generator set, multiple key wind speed detection positions are determined, including: based on the influence coefficient of each wind speed detection position to be screened on the tip speed ratio of the wind turbine generator set, multiple first target wind speed detection positions are determined from the multiple wind speed detection positions to be screened; for any two first target wind speed detection positions, the wind speed correlation coefficients of the two first target wind speed detection positions are calculated based on the wind speeds of the two first target wind speed detection positions in multiple test time periods; and through a clustering algorithm, multiple key wind speed detection positions are determined based on the wind speed correlation coefficients of any two first target wind speed detection positions.

[0008] Furthermore, through a clustering algorithm, multiple key wind speed detection positions are determined based on the wind speed correlation coefficients of any two first target wind speed detection positions, including: through a first clustering algorithm, multiple first target wind speed detection positions are clustered according to the wind speed correlation coefficients of any two first target wind speed detection positions to determine multiple first position clusters, and multiple second target wind speed detection positions are determined based on the cluster center of each first position cluster; through a second clustering algorithm, multiple second target wind speed detection positions are clustered according to the spatial distance between any two second target wind speed detection positions to determine multiple second position clusters; for each second position cluster, the key wind speed detection position corresponding to the second position cluster is determined based on the spatial coordinates of the second target wind speed detection position included in the second position cluster.

[0009] Furthermore, feature extraction is performed on the real-time wind speed matrix to generate a real-time wind speed feature matrix, including: for each real-time wind speed vector, performing variational mode decomposition on the real-time wind speed vector to generate multiple frequency-amplitude modulation signal components corresponding to the real-time wind speed vector, extracting component features of each frequency-amplitude modulation signal component, and generating a real-time wind speed feature submatrix corresponding to the real-time wind speed vector, wherein a row vector of the real-time wind speed feature submatrix represents a component feature of a frequency-amplitude modulation signal component, and the real-time wind speed feature matrix includes the real-time wind speed feature submatrix corresponding to each real-time wind speed vector.

[0010] Furthermore, the multi-objective tip speed ratio optimization function is at least related to the tip speed ratio difference, the mechanical vibration and power fluctuation of the generator.

[0011] Furthermore, the optimized value of each sampled tip speed ratio is calculated through finite element analysis, a multi-objective tip speed ratio optimization function and a benchmark tip speed ratio, including: calculating the impeller speed corresponding to the sampled tip speed ratio based on the sampled tip speed ratio; generating a mechanical vibration simulation result of the generator corresponding to the sampled tip speed ratio based on the impeller speed corresponding to the sampled tip speed ratio through finite element analysis; calculating the power fluctuation of the wind turbine corresponding to the sampled tip speed ratio based on the sampled tip speed ratio; and calculating the optimized value of the sampled tip speed ratio based on the multi-objective tip speed ratio optimization function, the sampled tip speed ratio, the benchmark tip speed ratio, the mechanical vibration simulation result of the generator corresponding to the sampled tip speed ratio, and the power fluctuation.

[0012] Furthermore, based on the impeller speed corresponding to the sampled tip speed ratio, a mechanical vibration simulation result of the generator corresponding to the sampled tip speed ratio is generated through finite element analysis, including: establishing a finite element model of the generator; applying boundary conditions and loads to the finite element model of the generator based on the impeller speed corresponding to the sampled tip speed ratio; and solving the finite element model of the generator to generate the mechanical vibration simulation result of the generator corresponding to the sampled tip speed ratio.

[0013] Furthermore, the optimal tip speed ratio of the wind turbine is determined based on the optimized value of each sampled tip speed ratio by a genetic algorithm, including: S11, based on the optimized value of each sampled tip speed ratio, the tip speed ratio value range, and the tip speed ratio difference between each sampled tip speed ratio and the reference tip speed ratio, multiple sampled tip speed ratios are selected, crossed, and mutated to generate multiple sampled tip speed ratios of the current iteration round; S12, by finite element analysis, multi-objective tip speed ratio optimization function, and reference tip speed ratio, the optimized value of each sampled tip speed ratio of the current iteration round is calculated; S13, based on the optimized value of each sampled tip speed ratio of the current iteration round, it is determined whether Satisfy the preset conditions. If not, execute S14; if so, execute S15; S14, based on the optimized value of the sampled tip speed ratio of each current iteration round, the tip speed ratio value range, and the tip speed ratio difference between the sampled tip speed ratio of each current iteration round and the benchmark tip speed ratio, select, cross and mutate the multiple sampled tip speed ratios of the current iteration round to generate multiple sampled tip speed ratios of the next iteration round, use the multiple sampled tip speed ratios of the next iteration round as the multiple sampled tip speed ratios of the current iteration round, and execute S12; S15, based on the optimized value of the sampled tip speed ratio of each current iteration round, determine the optimal tip speed ratio of the wind turbine.

[0014] The present invention provides a maximum wind energy capture control system for a wind turbine generator system, comprising: a position determination module for determining a plurality of key wind speed detection positions; a function establishment module for establishing a multi-objective blade tip speed ratio optimization function; a wind speed monitoring module for obtaining a real-time wind speed matrix based on a plurality of key wind speed detection positions, wherein the real-time wind speed matrix comprises a plurality of real-time wind speed vectors, one real-time wind speed vector corresponds to one key wind speed detection position, and the real-time wind speed vector comprises the wind speed of the corresponding key wind speed detection position at a plurality of time points, and feature extraction is performed on the real-time wind speed matrix. The invention relates to a method for generating a real-time wind speed characteristic matrix by taking a tip speed ratio as an example; running an optimization module to generate a tip speed ratio value range and a reference tip speed ratio based on the real-time wind speed characteristic matrix through a tip speed ratio optimization model, and generating multiple sampled tip speed ratios based on the tip speed ratio value range; calculating the optimized value of each sampled tip speed ratio through finite element analysis, a multi-objective tip speed ratio optimization function and a reference tip speed ratio, determining the optimal tip speed ratio of the wind turbine generator set based on the optimized value of each sampled tip speed ratio through a genetic algorithm, and controlling the wind turbine generator set to capture wind energy according to the optimal tip speed ratio of the wind turbine generator set.

[0015] Compared with the prior art, the maximum wind energy capture control method and system for wind turbines provided by the present invention have at least the following beneficial effects:

[0016] 1. Through a multi-objective optimization function, a balance is struck between multiple objectives, such as wind energy capture efficiency, mechanical stress, and power fluctuations, rather than simply pursuing a single objective. This avoids mechanical overload (such as gearbox damage) or grid shock (such as severe power fluctuations) caused by the pursuit of efficiency. For example, in strong wind conditions, some efficiency can be sacrificed, but by reducing the pitch angle or rotational speed, mechanical stress can be significantly reduced, extending equipment life. Wind speed changes can be detected in advance through multi-position wind speed detection (such as at different blade heights and at different distances in front of the rotor) and feature extraction. In non-uniform wind fields (such as wind shear and turbulence), traditional methods may fall into local optima, while genetic algorithms can explore a wider solution space, adapt to dynamic wind conditions, and smooth power output.

[0017] 2. By calculating the influence coefficients of the wind speed detection locations to be screened on the tip speed ratio of the wind turbine, the locations with the greatest impact on wind turbine performance can be screened. The wind speed data at these locations is more representative of the wind speed variations in the actual operating environment of the wind turbine. Avoid placing sensors in locations where wind speed variations are insignificant or have little impact on wind turbine performance, thereby reducing redundant data and improving data processing efficiency. By calculating the influence coefficients of the wind speed detection locations to be screened on the tip speed ratio of the wind turbine, the degree of influence of wind speed at different locations on the tip speed ratio can be quantified. This helps to more accurately adjust the control parameters of the wind turbine to achieve more precise tip speed ratio control. Control based on wind speed data at key wind speed detection locations can reduce control errors caused by inaccurate wind speed measurements or improper location selection, thereby improving the power generation efficiency and stability of the wind turbine.

[0018] 3. By calculating the influence coefficients of the selected wind speed detection locations on the tip speed ratio of the wind turbine, locations with influence coefficients greater than a first threshold (positive) or less than a second threshold (negative) are selected as the first target wind speed detection locations. Wind speed variations at these locations significantly impact wind turbine performance and more accurately reflect wind speed variations in the wind turbine's actual operating environment. The detection locations are clustered using the wind speed correlation coefficient, grouping locations with similar wind speed fluctuation patterns into the same cluster. This avoids selecting highly correlated, redundant locations. The most representative cluster center locations are selected from the large number of original locations to reduce subsequent computational effort. In a wind farm, if wind speeds of multiple wind turbines fluctuate synchronously due to terrain influences, retaining only the cluster center wind turbine can represent the wind speed characteristics of the area. The cluster center locations selected in the first stage are further clustered by spatial distance, merging geographically proximal clusters. Spatially proximal areas with similar wind speed patterns are merged to avoid over-concentration of key locations. Key locations are ensured to be evenly distributed across the wind farm, enhancing spatial representativeness. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further described in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, like numbers represent like structures, wherein:

[0020] Figure 1 is a flow chart of a maximum wind energy capture control method applied to a wind turbine generator system according to some embodiments of this specification;

[0021] Figure 2 is a schematic diagram of a process for determining multiple key wind speed detection positions according to some embodiments of this specification;

[0022] Figure 3 is a schematic diagram of a process for determining an optimal tip speed ratio of a wind turbine generator set according to some embodiments of this specification;

[0023] Figure 4 This is a module schematic diagram of a maximum wind energy capture control system applied to a wind turbine generator according to some embodiments of this specification. DETAILED DESCRIPTION

[0024] To more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly describes the drawings required for describing the embodiments. Obviously, the drawings described below are merely examples or embodiments of this specification. Those skilled in the art can apply this specification to other similar scenarios based on these drawings without inventive effort. Unless otherwise apparent from the context or otherwise noted, the same reference numerals in the figures represent the same structure or operation.

[0025] Figure 1 is a flow chart of a maximum wind energy capture control method applied to a wind turbine generator system according to some embodiments of this specification, such as Figure 1 As shown, the maximum wind energy capture control method applied to a wind turbine generator system may include the following process.

[0026] Step 110: Determine multiple key wind speed detection locations.

[0027] Figure 2 is a flow chart of determining multiple key wind speed detection positions according to some embodiments of this specification, such as Figure 2 As shown, step 110 specifically includes:

[0028] Determine multiple wind speed detection locations to be screened;

[0029] Obtaining the wind speed of each wind speed detection position to be screened in multiple test time periods, wherein the wind speed of the wind speed detection position to be screened in a certain test time period may include the wind speed at multiple test time points in the test time period;

[0030] Based on the hill climbing search algorithm, the target tip speed ratio of the wind turbine in multiple test time periods is obtained;

[0031] For each wind speed detection position to be screened, based on the wind speed at the wind speed detection position to be screened in multiple test time periods and the target tip speed ratio of the wind turbine in multiple test time periods, calculating the influence coefficient of the wind speed detection position to be screened on the tip speed ratio of the wind turbine;

[0032] According to the influence coefficient of each wind speed detection position to be screened on the tip speed ratio of the wind turbine generator set, a plurality of key wind speed detection positions are determined.

[0033] Specifically, the multiple wind speed detection locations to be screened need to cover the key spatial features of the wind field around the wind turbine, including:

[0034] 1. Height coverage: includes hub height (typical value: 80~120m) and a certain range above and below (such as ±20m) to capture vertical wind shear effects.

[0035] 2. Horizontal distribution: covers the upstream (such as 1D~5D, where D is the impeller diameter), lateral and downstream areas of the unit to reflect the effects of wind field turbulence, wake, etc.

[0036] 3. Terrain correlation: In complex terrain (such as mountains and hills), points need to be placed at typical locations such as ridges, valleys, and windward slopes.

[0037] Example:

[0038] Horizontal direction: a wind speed detection position to be screened is arranged at 2D, 3D and 4D upstream of the unit.

[0039] Vertical direction: wind speed detection positions to be screened are arranged at hub height, hub height + 10m, and hub height - 15m.

[0040] The locations to be screened must meet the requirements for wind turbine operation safety and data reliability:

[0041] 1. Unobstructed: Avoid obstructions such as buildings and trees to ensure accurate wind speed measurement.

[0042] 2. Stay away from interference sources: Stay at least 3D away from the wake area of ​​other units to avoid interference from wake effects.

[0043] 3. Mechanical accessibility: easy to install and maintain wind speed sensors (such as lidar, ultrasonic anemometer).

[0044] The principle of the Hill-Climbing Search method is to gradually approach the maximum power point by continuously disturbing the impeller speed and observing the changes in output power.

[0045] Implementation of Hill-Climbing Search:

[0046] Periodically apply small perturbations to the impeller speed (increase or decrease the speed).

[0047] Observe the changes in output power:

[0048] If the power increases, continue to perturb the speed in the same direction.

[0049] If the power is reduced, the speed is disturbed in the opposite direction.

[0050] Repeat the above process until the difference between the output power and the maximum power point is less than the preset power difference, and reversely calculate the target tip speed ratio based on the impeller speed at this time.

[0051] For example, the target tip speed ratio can be inversely calculated according to the following formula:

[0052] ,

[0053] in, is the target tip speed ratio, is the impeller speed, is the impeller radius, is the average wind speed at the t-th test time point, which may be the average of the wind speeds at the t-th test time point at multiple wind speed detection positions to be screened.

[0054] For each test time period, the standard deviation of the wind speed at the wind speed detection position to be screened in the test time period can be calculated based on the wind speed at the wind speed detection position to be screened at the test time point, and the standard deviation of the target tip speed ratio of the wind turbine generator set in the test time period can be calculated. Through the calculation formula of the nonlinear correlation coefficient (for example, the maximum information coefficient, the distance correlation coefficient, the Spearman rank correlation coefficient, etc.), the nonlinear correlation coefficient between the wind speed at the wind speed detection position to be screened and the target tip speed ratio of the wind turbine generator set is calculated based on the wind speed standard deviation and the target tip speed ratio standard deviation at the wind speed detection position to be screened in multiple test time periods as the influence coefficient of the wind speed detection position to be screened on the tip speed ratio of the wind turbine generator set.

[0055] It can be understood that by calculating the influence coefficient of the wind speed detection position to be screened on the tip speed ratio of the wind turbine, the position that has the greatest impact on the performance of the wind turbine can be screened. The wind speed data at these positions can better represent the wind speed changes in the actual operating environment of the wind turbine. Avoid placing sensors at positions where the wind speed changes are not significant or have little impact on the performance of the wind turbine, thereby reducing redundant data and improving data processing efficiency. By calculating the influence coefficient of the wind speed detection position to be screened on the tip speed ratio of the wind turbine, the degree of influence of wind speed at different positions on the tip speed ratio can be quantified. This helps to more accurately adjust the control parameters of the wind turbine to achieve more precise tip speed ratio control. Control based on wind speed data at key wind speed detection positions can reduce control errors caused by inaccurate wind speed measurement or improper position selection, and improve the power generation efficiency and stability of the wind turbine.

[0056] Preferably, a plurality of key wind speed detection positions are determined according to the influence coefficient of each wind speed detection position to be screened on the tip speed ratio of the wind turbine generator set, including:

[0057] Determining a plurality of first target wind speed detection positions from the plurality of wind speed detection positions to be screened according to the influence coefficient of each wind speed detection position to be screened on the tip speed ratio of the wind turbine generator set, for example, taking a wind speed detection position to be screened whose tip speed ratio influence coefficient is greater than a first threshold or less than a second threshold as the first target wind speed detection position, wherein the first threshold is a positive number and the second threshold is a negative number;

[0058] For any two first target wind speed detection positions, calculating the wind speed correlation coefficients of the two first target wind speed detection positions according to the wind speeds of the two first target wind speed detection positions in multiple test time periods;

[0059] A clustering algorithm is used to determine a plurality of key wind speed detection positions according to wind speed correlation coefficients of any two first target wind speed detection positions.

[0060] Specifically, for any two first target wind speed detection positions, the nonlinear correlation coefficient of the wind speeds at the two first target wind speed detection positions is calculated according to the calculation formula of the nonlinear correlation coefficient (for example, the maximum information coefficient, the distance correlation coefficient, the Spearman rank correlation coefficient, etc.) based on the standard deviation of the wind speeds at the two first target wind speed detection positions in multiple test time periods, and is used as the wind speed correlation coefficient of the two first target wind speed detection positions.

[0061] Preferably, a clustering algorithm is used to determine a plurality of key wind speed detection positions according to the wind speed correlation coefficients of any two first target wind speed detection positions, including:

[0062] Clustering the plurality of first target wind speed detection positions according to the wind speed correlation coefficients of any two first target wind speed detection positions using a first clustering algorithm (e.g., K-Means, agglomerative hierarchical clustering algorithm, etc.) to determine a plurality of first position clusters; and determining a plurality of second target wind speed detection positions based on the cluster center of each first position cluster. For example, the first target wind speed detection position corresponding to the cluster center of the first position cluster may be used as the second target wind speed detection position.

[0063] Clustering the plurality of second target wind speed detection positions according to the spatial distance between any two second target wind speed detection positions using a second clustering algorithm (e.g., K-Means, agglomerative hierarchical clustering algorithm, etc.) to determine a plurality of second position clusters;

[0064] For each second position cluster, the key wind speed detection position corresponding to the second position cluster is determined based on the spatial coordinates of the second target wind speed detection positions included in the second position cluster. For example, the spatial coordinates of the second target wind speed detection positions included in the second position cluster can be averaged as the key wind speed detection position corresponding to the second position cluster.

[0065] As can be understood, by calculating the influence coefficients of the selected wind speed detection locations on the tip speed ratio of the wind turbine, locations with influence coefficients greater than a first threshold (positive) or less than a second threshold (negative) are selected as the first target wind speed detection locations. Wind speed variations at these locations significantly impact wind turbine performance and more accurately reflect wind speed variations in the wind turbine's actual operating environment. The detection locations are clustered using the wind speed correlation coefficient, grouping locations with similar wind speed fluctuation patterns into the same cluster. This avoids selecting highly correlated, redundant locations. The most representative cluster center locations are selected from the large number of original locations to reduce subsequent computational effort. In a wind farm, if the wind speeds of multiple wind turbines fluctuate synchronously due to terrain influences, retaining only the cluster center wind turbine can represent the wind speed characteristics of the area. The cluster center locations selected in the first stage are further clustered by spatial distance, merging geographically proximal clusters. Spatially proximal areas with similar wind speed patterns are merged to avoid over-concentration of key locations. This ensures that key locations are evenly distributed across the wind farm, enhancing spatial representativeness.

[0066] Step 120: Establish a multi-objective tip speed ratio optimization function.

[0067] Preferably, the multi-objective tip speed ratio optimization function is at least related to the tip speed ratio difference, the mechanical vibration and power fluctuation of the generator.

[0068] Step 130: Acquire a real-time wind speed matrix based on multiple key wind speed detection positions.

[0069] The real-time wind speed matrix includes multiple real-time wind speed vectors, one real-time wind speed vector corresponds to one key wind speed detection position, and the real-time wind speed vector includes the wind speed of the corresponding key wind speed detection position at multiple time points.

[0070] Step 140 : extracting features from the real-time wind speed matrix to generate a real-time wind speed feature matrix.

[0071] Specifically include:

[0072] For each real-time wind speed vector, variational modal decomposition is performed on the real-time wind speed vector to generate multiple frequency-amplitude modulation signal components corresponding to the real-time wind speed vector, and the component features of each frequency-amplitude modulation signal component are extracted to generate a real-time wind speed feature submatrix corresponding to the real-time wind speed vector, wherein a row vector of the real-time wind speed feature submatrix represents the component features of a frequency-amplitude modulation signal component, and the real-time wind speed feature matrix includes the real-time wind speed feature submatrix corresponding to each real-time wind speed vector.

[0073] Specifically, variational mode decomposition (VMD) is an adaptive signal decomposition method that can decompose complex signals into multiple frequency- and amplitude-modulated intrinsic mode functions (FM / AM signal components). Its core concept is to iteratively search for the optimal solution of the variational model to determine the center frequency and bandwidth of each FM / AM signal component, thereby achieving frequency-domain segmentation and effective separation of the signal.

[0074] Non-recursive and adaptive: Variational mode decomposition determines the center frequency and bandwidth of the FM and AM signal components through iterative optimization, avoiding the modal aliasing problem in empirical mode decomposition.

[0075] Variational problem construction: Assuming that each FM / AM signal component is a finite bandwidth signal with a different center frequency, the variational problem is constructed by minimizing the sum of the bandwidths of all FM / AM signal components and the error between them and the original signal.

[0076] Optimization solution: Use optimization algorithms such as the alternating direction multiplier method to solve the variational problem, continuously update the parameters of the FM signal components (such as center frequency, amplitude, and phase), and gradually approach the optimal solution.

[0077] The real-time wind speed vector can be subjected to variational mode decomposition according to the following steps to generate multiple FM signal components corresponding to the real-time wind speed vector, and the component features of each FM signal component can be extracted:

[0078] S21. Data preparation:

[0079] Obtain real-time wind speed vector data to ensure data integrity and continuity.

[0080] S22. Variational mode decomposition parameter settings:

[0081] Number of modes: determines the number of decomposed FM / AM signal components and needs to be reasonably set according to the complexity of the wind speed signal.

[0082] Penalty parameter: controls the bandwidth of the FM / AM signal components and balances the degree of separation between modes.

[0083] Convergence criterion tolerance: Set the conditions for stopping the iteration, such as reaching the maximum number of iterations or the residual signal energy being lower than the threshold.

[0084] S23, variational mode decomposition:

[0085] The real-time wind speed vector is input into the variational mode decomposition algorithm and decomposed into K frequency-amplitude-modulation signal components through iterative optimization. Each frequency-amplitude-modulation signal component represents a specific frequency component in the wind speed signal and has frequency-amplitude-modulation characteristics.

[0086] S24. Extraction of component features

[0087] For each FM / AM signal component obtained by decomposition, the following features are extracted to describe its time-frequency characteristics:

[0088] 1. Center frequency:

[0089] Definition: The main frequency component of the FM / AM signal in the frequency domain.

[0090] Significance: Reflects the distribution of different frequency components in the wind speed signal.

[0091] 2. Bandwidth:

[0092] Definition: The frequency range of the FM and AM signal components in the frequency domain.

[0093] Significance: Describes the fluctuation range of different frequency components in the wind speed signal.

[0094] 3. Instantaneous amplitude:

[0095] Definition: The amplitude variation of the FM / AM signal components in the time domain.

[0096] Significance: Reflects the intensity changes of different frequency components in the wind speed signal.

[0097] 4. Instantaneous frequency:

[0098] Definition: Frequency variation of the FM / AM signal components in the time domain.

[0099] Significance: Describes the time-varying characteristics of different frequency components in wind speed signals.

[0100] 5. Energy

[0101] Definition: The energy of the FM / AM signal component in the time domain or frequency domain.

[0102] Significance: Quantify the energy contribution of different frequency components in the wind speed signal.

[0103] As you can understand, variational modal decomposition (VMD) can effectively separate the different frequency components (such as turbulence and gusts) in wind speed signals, facilitating subsequent analysis. By adaptively determining the center frequency and bandwidth of the modal components, VMD can provide higher time-frequency resolution and more accurately describe the dynamic characteristics of wind speed signals. VMD can accurately identify signal characteristics even in the presence of noise, making it suitable for wind speed monitoring in complex environments.

[0104] Step 150 : Generate a tip speed ratio value range and a reference tip speed ratio based on the real-time wind speed characteristic matrix through the tip speed ratio optimization model.

[0105] The tip speed ratio value range may be a possible value range of the tip speed ratio for the real-time wind speed characteristic matrix, and the reference tip speed ratio may be the tip speed ratio for maximum wind energy capture corresponding to the real-time wind speed characteristic matrix. The tip speed ratio optimization model may be a convolutional neural network (CNN) model.

[0106] The model architecture of the tip speed ratio optimization model may include:

[0107] Input layer:

[0108] A real-time wind speed characteristic matrix (such as a 32×32 two-dimensional matrix) and the influence coefficient of each key wind speed detection position on the tip speed ratio of the wind turbine.

[0109] Convolutional layer:

[0110] Multi-layer convolution (such as 3x3 convolution kernel) extracts the spatial features of wind speed (such as wind speed gradient, turbulence pattern).

[0111] Pooling layer:

[0112] Max pooling or average pooling reduces feature dimensions and enhances the robustness of the model.

[0113] Fully connected layer:

[0114] The convolution features are mapped to the tip speed ratio range and the benchmark tip speed ratio.

[0115] Output layer:

[0116] Two output branches:

[0117] Tip speed ratio value range: lower and upper limits of the output range.

[0118] Base tip speed ratio: Outputs the tip speed ratio for maximum wind energy capture corresponding to the current wind conditions.

[0119] The tip speed ratio optimization model has the following constraints:

[0120] 1. Maximum speed constraints: The mechanical strength of wind turbine blades and drive systems (such as gearboxes and generators) is limited. When the speed is too high, blades may deform or break due to excessive centrifugal force, and the drive system may be damaged due to overload. The maximum speed is determined by the wind turbine's design parameters (such as material strength, blade length, and mass distribution) and is a hard constraint on the wind turbine's safe operation.

[0121] 2. Grid power constraints: Wind turbine output power is closely related to wind speed, tip speed ratio, and pitch angle (β). When wind speeds are too high, the wind turbine may reach its rated power or even exceed the maximum power allowed by the grid. Exceeding the maximum power allowed by the grid can cause grid voltage fluctuations, frequency instability, and even damage grid equipment.

[0122] 3. Vibration Constraints: Vibration of the blades and tower can cause fatigue damage, reducing equipment life. Excessive vibration can trigger protective mechanisms and even cause downtime.

[0123] Step 160: Generate a plurality of sampled tip speed ratios based on the tip speed ratio value range.

[0124] For example, a plurality of sample tip speed ratios are randomly selected from the tip speed ratio value range.

[0125] Step 170 : Calculate the optimized value of each sampled tip speed ratio through finite element analysis, a multi-objective tip speed ratio optimization function, and a reference tip speed ratio.

[0126] Specifically include:

[0127] Based on the sampled tip speed ratio, calculating the impeller speed corresponding to the sampled tip speed ratio;

[0128] Generate mechanical vibration simulation results of the generator corresponding to the sampled tip speed ratios based on the impeller speed corresponding to the sampled tip speed ratios through finite element analysis;

[0129] Based on the sampled blade tip speed ratio, the power fluctuation of the wind turbine corresponding to the sampled blade tip speed ratio is calculated;

[0130] The optimized value of the sampled tip speed ratio is calculated according to the multi-objective tip speed ratio optimization function, the sampled tip speed ratio, the benchmark tip speed ratio, the mechanical vibration simulation results of the generator corresponding to the sampled tip speed ratio, and the power fluctuation.

[0131] Preferably, generating a mechanical vibration simulation result of the generator corresponding to the sampled tip speed ratio based on the impeller speed corresponding to the sampled tip speed ratio by finite element analysis includes:

[0132] Establish a finite element model of the generator;

[0133] Apply boundary conditions and loads to the finite element model of the generator based on the impeller speed corresponding to the sampled tip speed ratio;

[0134] The finite element model of the generator is solved to generate the mechanical vibration simulation results of the generator corresponding to the sampled tip speed ratios.

[0135] Specifically, the following processes may be included:

[0136] S31. Establish a finite element model of the generator

[0137] Modeling objects: Key components of the generator (such as rotor, stator, bearings, gearbox, main shaft, etc.), with special attention paid to structures that may cause vibration (such as rotor imbalance, gear mesh excitation, etc.).

[0138] Modeling method: Use CAD software to build a three-dimensional geometric model of the generator and simplify non-critical structures (such as bolts, small holes, etc.).

[0139] Meshing: Use finite element pre-processing software to perform meshing. Select the appropriate element type: solid elements (such as tetrahedrons and hexahedrons) are used for solid structures such as rotors and stators. Shell elements are used for thin-walled structures (such as casings). Beam elements are used for slender structures such as spindles.

[0140] Mesh density control: Increase mesh density near vibration excitation sources (such as gear meshing surfaces and bearing contact surfaces). Appropriately reduce mesh density in non-critical areas.

[0141] Material property definition: Define the material properties of each component (such as elastic modulus, Poisson's ratio, density, damping ratio, etc.).

[0142] S32. Apply boundary conditions and loads

[0143] Boundary conditions:

[0144] Fixed constraint: Apply full constraints (fix all degrees of freedom) to the fixed end of the generator (such as the flange connected to the tower).

[0145] Contact Definition: Define the contact relationship between the bearing and the shaft (such as friction contact or frictionless contact). Define the contact relationship between the gear meshing surfaces.

[0146] Load application:

[0147] Speed ​​load: Calculate the impeller speed based on the sampled tip speed ratio and convert it into the angular velocity of the rotor. In the finite element model, speed load is applied through rotating boundary conditions or inertial force loading:

[0148] Rotating boundary condition: The angular velocity is applied directly to the rotor surface.

[0149] Inertial force loading: Calculates the centrifugal force generated by the rotor during rotation and applies it to the rotor element as a body load.

[0150] Unbalance Excitation: Simulate rotor mass imbalance or apply an equivalent periodic force directly at the rotor mass center.

[0151] Gear mesh excitation: Simulates dynamic loads during gear meshing (such as time-varying mesh stiffness and meshing impact) by applying periodic force or displacement excitation.

[0152] Gravity load: Applies gravity acceleration in the vertical direction.

[0153] S33, solution and post-processing

[0154] Solution settings:

[0155] Analysis Type: Modal Analysis: Calculates the generator's natural frequencies and vibration modes, identifying possible resonant frequencies. Harmonic Response Analysis: Analyzes the generator's steady-state vibration response (e.g., displacement, acceleration) under simple harmonic excitation. Transient Dynamics Analysis: Analyzes the generator's vibration response under unsteady-state excitation (e.g., startup, shutdown, sudden load changes).

[0156] Solver selection: Mode Superposition Method.

[0157] Post-processing:

[0158] Vibration Response Extraction: Extract the vibration displacement, velocity, and acceleration time histories of multiple key nodes of the generator. Extract the frequency response function and analyze how the vibration amplitude changes with frequency.

[0159] The power prediction model can be used to predict the output power of the wind turbine corresponding to the sampled tip speed ratio based on the sampled tip speed ratio. The standard deviation of the predicted output power of the wind turbine corresponding to the sampled tip speed ratio and the output power of the wind turbine at the previous m consecutive time points is calculated as the power fluctuation of the wind turbine corresponding to the sampled tip speed ratio. Specifically, the power prediction model is used to predict the output power of the wind turbine corresponding to the sampled tip speed ratio based on the sampled tip speed ratio, which is recorded as P sample , before obtaining m The actual output power of the wind turbine at each consecutive time point is recorded as P t−1 , P t−2 ,…, P t−m , and the predicted output power of the wind turbine P sample Composition power sequence P t−1 ,P t−2 ,…, P t−m , P sample , calculate the power series according to the standard deviation calculation formula P t−1 , P t−2 ,…, P t−m , P sample The standard deviation of the sampled tip speed ratio is used as the power fluctuation of the wind turbine corresponding to the sampled tip speed ratio. m The output power of wind turbines at consecutive time points together constitutes a power sequence. The greater the difference between the power values ​​in the power sequence, that is, the higher the degree of dispersion, the larger the calculated standard deviation and the greater the power fluctuation.

[0160] As an example only, the multi-objective tip speed ratio optimization function is:

[0161] ,

[0162] in, is the multi-objective tip speed ratio optimization function, 、 and is the weight, 、 and greater than 0, , is the sampled tip speed ratio, is the base tip speed ratio, is the maximum amplitude of the i-th key node of the generator corresponding to the sampled tip speed ratio, is the total number of key nodes, is the power fluctuation of the wind turbine corresponding to the sampled tip speed ratio.

[0163] Understandably, optimizing only the power coefficient can lead to excessive mechanical stress or excessive power fluctuations. Controlling only power fluctuations can compromise power generation efficiency. A multi-objective tip speed ratio optimization function is used to find a balance between multiple objectives (i.e., efficiency, safety, and stability).

[0164] Step 180 : Determine the optimal tip speed ratio of the wind turbine generator system based on the optimized value of each sampled tip speed ratio by using a genetic algorithm.

[0165] Figure 3 is a flow chart of determining the optimal tip speed ratio of a wind turbine generator set according to some embodiments of this specification, such as Figure 3 As shown, as an example, step 180 specifically includes:

[0166] S11. Based on the optimized value of each sampled tip speed ratio, the range of tip speed ratios, and the difference between each sampled tip speed ratio and the baseline tip speed ratio, multiple sampled tip speed ratios are selected, crossover, and mutated to generate multiple sampled tip speed ratios for the current iteration. Specifically, the best performing sampled tip speed ratios are retained based on the optimized value (e.g., roulette wheel selection or tournament selection). Two parent sampled tip speed ratios are combined (e.g., arithmetic crossover or single-point crossover) to generate child sampled tip speed ratios. Example: Parent sampled tip speed ratio A = 6, parent sampled tip speed ratio B = 10, child tip speed ratio = 8 (arithmetic crossover). Mutation: Randomly perturb some child tip speed ratios (e.g., add or subtract a small amount) to increase population diversity. Example: TSR = 8 mutates to TSR = 8.2 (random perturbation ±0.2). For each tip speed ratio difference range, the mean of the optimized values ​​of the sampled tip speed ratios within that tip speed ratio difference range can be calculated. The larger the mean, the greater the probability that the sampled tip speed ratios within the tip speed ratio difference range will be retained as the parent. It can be understood that by calculating the mean of the optimized values ​​within the tip speed ratio difference range, the overall performance of the sampled tip speed ratios within different tip speed ratio difference ranges can be quantified. The selection strategy based on statistical information is more objective and accurate, and helps to find the global optimal solution.

[0167] S12. Calculating an optimized value of the sampled tip speed ratio of each current iteration round through finite element analysis, a multi-objective tip speed ratio optimization function, and a benchmark tip speed ratio;

[0168] S13, based on the optimized value of the sampled tip speed ratio of each current iteration round, determine whether a preset condition is met, for example, whether the change of the optimization value for multiple generations is less than a threshold, whether the maximum number of iterations is reached, etc. If not, execute S14; if so, execute S15;

[0169] S14, based on the optimized value of the sampled tip speed ratio of each current iteration round, the value range of the tip speed ratio, and the tip speed ratio difference between the sampled tip speed ratio of each current iteration round and the reference tip speed ratio, performing selection, crossover, and mutation operations on the multiple sampled tip speed ratios of the current iteration round to generate multiple sampled tip speed ratios of the next iteration round, using the multiple sampled tip speed ratios of the next iteration round as the multiple sampled tip speed ratios of the current iteration round, and executing S12;

[0170] S15. Determine the optimal tip speed ratio of the wind turbine generator based on the optimized value of the sampled tip speed ratio in each current iteration round.

[0171] Step 190: Control the wind turbine generator set to capture wind energy according to the optimal tip speed ratio of the wind turbine generator set.

[0172] Specifically, after determining the optimal tip speed ratio of the wind turbine, it is necessary to coordinately adjust the operating state of the wind turbine through variable pitch control and variable speed control to make its actual tip speed ratio as close to the optimal tip speed ratio as possible, thereby achieving multi-objective optimization of wind energy capture efficiency.

[0173] Figure 4 is a module diagram of a maximum wind energy capture control system applied to a wind turbine generator system according to some embodiments of this specification, such as Figure 4 As shown, the maximum wind energy capture control system applied to a wind turbine generator system may include a position determination module, a function establishment module, a wind speed monitoring module and an operation optimization module.

[0174] A position determination module, used to determine multiple key wind speed detection positions;

[0175] Function building module, used to build multi-objective tip speed ratio optimization function;

[0176] A wind speed monitoring module is configured to obtain a real-time wind speed matrix based on multiple key wind speed detection locations, wherein the real-time wind speed matrix includes multiple real-time wind speed vectors, one real-time wind speed vector corresponds to one key wind speed detection location, and the real-time wind speed vector includes the wind speed of the corresponding key wind speed detection location at multiple time points, and perform feature extraction on the real-time wind speed matrix to generate a real-time wind speed feature matrix;

[0177] The operation optimization module is used to generate a tip speed ratio value range and a benchmark tip speed ratio based on the real-time wind speed characteristic matrix through a tip speed ratio optimization model, and generate multiple sampled tip speed ratios based on the tip speed ratio value range; calculate the optimized value of each sampled tip speed ratio through finite element analysis, a multi-objective tip speed ratio optimization function and a benchmark tip speed ratio; determine the optimal tip speed ratio of the wind turbine based on the optimized value of each sampled tip speed ratio through a genetic algorithm, and control the wind turbine to capture wind energy according to the optimal tip speed ratio of the wind turbine.

[0178] The maximum wind energy capture control system applied to a wind turbine generator set may be used to execute the maximum wind energy capture control method applied to a wind turbine generator set, which will not be described in detail here.

[0179] Finally, it should be understood that the embodiments described in this specification are intended only to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. A maximum wind energy capture control method applied to a wind turbine generator system, characterized in that: include: Identify multiple key wind speed detection locations; Establish a multi-objective tip speed ratio optimization function; Based on multiple key wind speed detection positions, a real-time wind speed matrix is ​​obtained, wherein the real-time wind speed matrix includes multiple real-time wind speed vectors, one real-time wind speed vector corresponds to one key wind speed detection position, and the real-time wind speed vector includes the wind speed of the corresponding key wind speed detection position at multiple time points; Extract features from the real-time wind speed matrix to generate a real-time wind speed feature matrix; Through the tip speed ratio optimization model, based on the real-time wind speed characteristic matrix, the tip speed ratio value range and the benchmark tip speed ratio are generated; Based on the tip speed ratio value range, a plurality of sample tip speed ratios are generated; The optimized value of each sampled tip speed ratio is calculated by finite element analysis, multi-objective tip speed ratio optimization function and benchmark tip speed ratio; Determine the optimal tip speed ratio of the wind turbine based on the optimized value of each sampled tip speed ratio through a genetic algorithm; Controlling the wind turbine to capture wind energy based on the optimal tip speed ratio of the wind turbine; Among them, several key wind speed detection locations are determined, including: Determine multiple wind speed detection locations to be screened; Obtain the wind speed of each wind speed detection location to be screened during multiple test time periods; Based on the hill climbing search algorithm, the target tip speed ratio of the wind turbine in multiple test time periods is obtained; For each wind speed detection position to be screened, based on the wind speed at the wind speed detection position to be screened in multiple test time periods and the target tip speed ratio of the wind turbine in multiple test time periods, calculating the influence coefficient of the wind speed detection position to be screened on the tip speed ratio of the wind turbine; According to the influence coefficient of each wind speed detection position to be screened on the tip speed ratio of the wind turbine generator set, a plurality of key wind speed detection positions are determined.

2. The maximum wind energy capture control method for a wind turbine generator system according to claim 1, characterized in that: According to the influence coefficient of each wind speed detection position to be screened on the tip speed ratio of the wind turbine, multiple key wind speed detection positions are determined, including: Determining a plurality of first target wind speed detection positions from a plurality of wind speed detection positions to be screened according to an influence coefficient of each wind speed detection position to be screened on a tip speed ratio of the wind turbine generator set; For any two first target wind speed detection positions, calculating the wind speed correlation coefficients of the two first target wind speed detection positions according to the wind speeds of the two first target wind speed detection positions in multiple test time periods; A clustering algorithm is used to determine a plurality of key wind speed detection positions according to wind speed correlation coefficients of any two first target wind speed detection positions.

3. The maximum wind energy capture control method for a wind turbine generator system according to claim 2, characterized in that: Through the clustering algorithm, multiple key wind speed detection locations are determined according to the wind speed correlation coefficients of any two first target wind speed detection locations, including: Clustering the plurality of first target wind speed detection positions according to the wind speed correlation coefficients of any two first target wind speed detection positions using a first clustering algorithm to determine a plurality of first position clusters, and determining a plurality of second target wind speed detection positions based on the cluster center of each first position cluster; Clustering the plurality of second target wind speed detection positions according to the spatial distance between any two second target wind speed detection positions using a second clustering algorithm to determine a plurality of second position clusters; For each second position cluster, the key wind speed detection position corresponding to the second position cluster is determined according to the spatial coordinates of the second target wind speed detection positions included in the second position cluster.

4. The maximum wind energy capture control method for a wind turbine generator system according to any one of claims 1 to 3, characterized in that: Perform feature extraction on the real-time wind speed matrix to generate a real-time wind speed feature matrix, including: For each real-time wind speed vector, variational modal decomposition is performed on the real-time wind speed vector to generate multiple frequency-amplitude modulation signal components corresponding to the real-time wind speed vector, and component features of each frequency-amplitude modulation signal component are extracted to generate a real-time wind speed feature submatrix corresponding to the real-time wind speed vector, wherein a row vector of the real-time wind speed feature submatrix represents a component feature of a frequency-amplitude modulation signal component, and the real-time wind speed feature matrix includes the real-time wind speed feature submatrix corresponding to each real-time wind speed vector.

5. The maximum wind energy capture control method for a wind turbine according to any one of claims 1 to 3, characterized in that: The multi-objective tip speed ratio optimization function is at least related to the tip speed ratio difference, the mechanical vibration and power fluctuation of the generator.

6. The maximum wind energy capture control method for a wind turbine generator system according to claim 5, characterized in that: The optimized value of each sampled tip speed ratio is calculated using finite element analysis, a multi-objective tip speed ratio optimization function, and a benchmark tip speed ratio, including: Based on the sampled tip speed ratio, calculating the impeller speed corresponding to the sampled tip speed ratio; Generate mechanical vibration simulation results of the generator corresponding to the sampled tip speed ratios based on the impeller speed corresponding to the sampled tip speed ratios through finite element analysis; Based on the sampled blade tip speed ratio, the power fluctuation of the wind turbine corresponding to the sampled blade tip speed ratio is calculated; The optimized value of the sampled tip speed ratio is calculated according to the multi-objective tip speed ratio optimization function, the sampled tip speed ratio, the benchmark tip speed ratio, the mechanical vibration simulation results of the generator corresponding to the sampled tip speed ratio, and the power fluctuation.

7. The maximum wind energy capture control method for a wind turbine generator system according to claim 6, characterized in that: Based on the impeller speed corresponding to the sampled tip speed ratio, the finite element analysis generates the mechanical vibration simulation results of the generator corresponding to the sampled tip speed ratio, including: Establish a finite element model of the generator; Apply boundary conditions and loads to the finite element model of the generator based on the impeller speed corresponding to the sampled tip speed ratio; The finite element model of the generator is solved to generate the mechanical vibration simulation results of the generator corresponding to the sampled tip speed ratios.

8. The maximum wind energy capture control method for a wind turbine generator system according to claim 5, characterized in that: The optimal tip speed ratio of the wind turbine is determined based on the optimized value of each sampled tip speed ratio through a genetic algorithm, including: S11, based on the optimized value of each sampled tip speed ratio, the value range of the tip speed ratio, and the tip speed ratio difference between each sampled tip speed ratio and the reference tip speed ratio, performing selection, crossover, and mutation operations on multiple sampled tip speed ratios to generate multiple sampled tip speed ratios for the current iteration round; S12. Calculating an optimized value of the sampled tip speed ratio of each current iteration round through finite element analysis, a multi-objective tip speed ratio optimization function, and a benchmark tip speed ratio; S13, based on the optimized value of the sampled tip speed ratio of each current iteration round, determine whether the preset conditions are met, if not, execute S14, if so, execute S15; S14, based on the optimized value of the sampled tip speed ratio of each current iteration round, the value range of the tip speed ratio, and the tip speed ratio difference between the sampled tip speed ratio of each current iteration round and the reference tip speed ratio, performing selection, crossover, and mutation operations on the multiple sampled tip speed ratios of the current iteration round to generate multiple sampled tip speed ratios of the next iteration round, using the multiple sampled tip speed ratios of the next iteration round as the multiple sampled tip speed ratios of the current iteration round, and executing S12; S15. Determine the optimal tip speed ratio of the wind turbine generator based on the optimized value of the sampled tip speed ratio in each current iteration round.

9. A maximum wind energy capture control system applied to a wind turbine generator system, characterized in that: The maximum wind energy capture control method for a wind turbine generator system according to any one of claims 1 to 8 comprises: A position determination module, used to determine multiple key wind speed detection positions; Function building module, used to build multi-objective tip speed ratio optimization function; A wind speed monitoring module is configured to obtain a real-time wind speed matrix based on multiple key wind speed detection positions, wherein the real-time wind speed matrix includes multiple real-time wind speed vectors, one real-time wind speed vector corresponds to one key wind speed detection position, and the real-time wind speed vector includes the wind speed of the corresponding key wind speed detection position at multiple time points, and perform feature extraction on the real-time wind speed matrix to generate a real-time wind speed feature matrix; The operation optimization module is used to generate a tip speed ratio value range and a benchmark tip speed ratio based on the real-time wind speed characteristic matrix through a tip speed ratio optimization model, and generate multiple sampled tip speed ratios based on the tip speed ratio value range; calculate the optimized value of each sampled tip speed ratio through finite element analysis, a multi-objective tip speed ratio optimization function and a benchmark tip speed ratio; determine the optimal tip speed ratio of the wind turbine based on the optimized value of each sampled tip speed ratio through a genetic algorithm, and control the wind turbine to capture wind energy according to the optimal tip speed ratio of the wind turbine.

Images