A compound load reduction control method for a two-blade wind turbine under extreme wind conditions

Abstract

The application discloses a two-blade wind turbine compound load reduction control method under extreme wind conditions, which comprises the following steps: collecting wind condition data through a wind direction and wind speed sensor, inputting the collected original wind condition data into a wind condition prediction model to predict the wind speed and wind direction, judging the wind condition in the prediction time period according to the wind condition prediction data, selecting a corresponding strategy according to the wind condition to perform yaw control on the two-blade wind turbine, and performing independent variable pitch control on each blade of the two-blade wind turbine to reduce the load of each blade. Through the prediction of the wind speed and the wind direction, the accuracy and timeliness of the yaw control are improved, and through the design of a new independent variable pitch control equation applied to the two-blade wind turbine, the accuracy of the independent variable pitch control is improved, and the stability of the system is improved.

Description

Technical Field

[0001] This invention relates to wind power technology, wind condition prediction technology, yaw control technology and independent pitch technology, and particularly to a composite load reduction method for a two-bladed wind turbine. Background Technology

[0002] Under the strategic goals of promoting "dual carbon" and building a new power system, new energy sources, primarily wind power, are experiencing rapid development. As a key component of renewable energy, wind energy is making significant contributions to a zero-carbon society. With the development of the wind power industry, the country is actively promoting the construction of large-scale wind power and photovoltaic bases, focusing on desert, Gobi, and arid regions. These regions possess abundant wind energy resources, allowing wind turbines to maintain high power generation duration and output. Furthermore, their remoteness from human settlements makes them ideal locations for large-scale wind power development. However, the harsh environment of these regions leads to high turbine failure rates and difficulties in equipment transportation, significantly increasing operation and maintenance costs. Therefore, cost reduction and efficiency improvement have become urgent issues for the wind power industry. Research on new large-capacity turbines with two blades and a downwind orientation has successfully broken through new boundaries in terms of low cost per kilowatt-hour, large capacity, high operational reliability, strong environmental adaptability, and low transportation and installation costs.

[0003] Two-bladed wind turbines are typically deployed in areas with abundant wind energy resources but complex terrain, especially in deserts or arid regions, where extreme wind conditions are common, such as strong winds, gusts, and drastic wind speed changes. Because two-bladed turbines have longer blades and a larger wind-catching area, strong winds place a greater load on the blades, easily leading to blade overload or structural damage. Therefore, yaw control and independent pitch control are particularly important for two-bladed turbines. Furthermore, since two-bladed turbines have fewer blades than three-bladed turbines, aerodynamic changes are more drastic during sudden wind changes, potentially causing greater loads on the turbine structure. Therefore, a more intelligent wind condition prediction mechanism is needed to optimize the yaw system.

[0004] Currently, numerous domestic and international literature and patent studies have applied wind condition prediction, yaw control, and independent pitch control to wind turbine load reduction, and have proposed some effective solutions.

[0005] 1. In the paper titled "Wind direction prediction for yaw control of wind turbines," Song D et al. proposed a hybrid model integrating autoregressive moving average (ARIMA) into Kalman filtering (KF) for predicting wind direction and optimizing yaw control. However, the yaw control strategy of this prediction model mainly focuses on wind direction prediction, achieving the goal of initiating yaw action in advance, but neglecting wind speed prediction and failing to incorporate optimization of yaw control parameters. Therefore, yaw performance in different wind speed ranges needs improvement.

[0006] 2. In patent application CN202311029012.2, entitled "Load-Dependent Autonomous Yaw Control for Wind Turbines," P. Gaucher et al. configured a control action signal for the yaw mechanism of a wind turbine to be determined based on multiple wind conditions and multiple load conditions. The control system determines the start and stop of the yaw system based on monitoring multiple changes related to the load conditions. However, for two-bladed wind turbines, this yaw system only adjusts yaw by signal changes, wasting the wind-following capability of the two-bladed wind turbine as a downwind turbine, reducing efficiency and increasing the burden on the yaw motor.

[0007] 3. In the paper titled "Feedback-feedforward individual pitch control for windturbine load reduction," the authors, Selvam K et al., used a Coleman transformation to convert all signals to a non-rotating reference frame. In the Coleman domain, using an optimal multivariable LQG controller and a wind speed-based feedforward disturbance rejection controller, the LQG control minimizes rotor tilt and yaw moments, while the feedforward controller achieves further improvements by suppressing the influence of low-frequency wind components on rotor torque. However, if the basic multi-blade coordinate transformation used for independent pitch control in three-blade wind turbines is applied to two-blade wind turbines, this transformation becomes singular, potentially causing the control system to malfunction, leading to system instability, and even affecting turbine operation and power output. Summary of the Invention

[0008] In view of this, and considering the technical problems of insufficient adjustment capability of existing two-bladed wind turbines under extreme wind conditions, low adaptability of existing adjustment methods to two-bladed wind turbines, and slow response speed, the purpose of this invention is to provide a composite load reduction control method for two-bladed wind turbines under extreme wind conditions, so as to achieve stable operation of two-bladed wind turbines under extreme wind conditions and improve the power generation efficiency of two-bladed wind turbines.

[0009] The composite load reduction control method for two-blade wind turbines under extreme wind conditions of the present invention includes the following steps:

[0010] 1) Wind condition data is collected by wind direction and wind speed sensors, and the wind condition data is time series data including wind speed and wind direction;

[0011] 2) Input the raw wind data collected in step 1) into the wind prediction model to predict wind speed and direction;

[0012] 3) Determine the wind conditions within the forecast period based on wind forecast data, and select the corresponding strategy to control the yaw of the two-bladed wind turbine based on the wind conditions:

[0013] ① Set the maximum wind speed and the maximum wind direction change angle within the prediction period. If the predicted wind speed obtained within the prediction period is less than the set maximum wind speed and the predicted wind direction change angle is less than the set maximum wind direction change angle, then the wind conditions within the prediction period are normal wind conditions; otherwise, the wind conditions within the prediction period are extreme wind conditions.

[0014] ② If the wind conditions are normal during the forecast period, a passive yaw control strategy will be adopted during the forecast period to automatically align the yaw angle of the wind turbine with the wind direction.

[0015] If extreme wind conditions occur during the forecast period, an active yaw control strategy will be used to adjust the yaw angle of the wind turbine during the forecast period.

[0016] 4) Implement independent pitch control for each blade of the two-blade wind turbine to reduce the load on each blade, including the following steps:

[0017] A) Calculate the total load on the entire blade;

[0018] B) Transfer the calculated total blade load from the rotating reference frame to the fixed non-rotating reference frame to obtain the matrix expression of the blade load. Then rewrite the matrix expression into a vector expression and obtain the collective mode and difference mode of the blade load from the vector expression equation.

[0019] C) Design independent pitch control equations for differential mode and independent pitch control equations for collective mode, and apply the pitch control equations for differential mode and collective mode in parallel to reduce blade load.

[0020] Furthermore, the wind condition prediction model described in step 2) consists of a VMD model, an EEMD model, an LSTM model, and an LSSVM model. The training and wind condition prediction steps of the wind condition prediction model include:

[0021] ① The original wind condition data is decomposed using the VMD model to obtain K subsequences and residuals;

[0022] ② The residual data obtained in step ① are further decomposed using the EEMD model to obtain N subsequences and residuals;

[0023] ③ Input the K subsequences obtained from step ① and the N subsequences obtained from step ②, along with the residuals, into the LSTM model to train the LSTM model. During the training process, obtain the wind condition prediction data and wind condition prediction error data output by the LSTM model.

[0024] ④ Input the wind condition prediction error data obtained in step ③ into the LSSVM model to train the LSSVM model. During the training process, the wind condition prediction error correction data output by the LSSVM model is obtained.

[0025] ⑤ Add the wind condition prediction data output by the LSTM model in step ③ to the wind condition prediction error correction data output by the LSSVM model in step ④ to obtain the final wind condition prediction data, which includes predicted wind speed and predicted wind direction.

[0026] Furthermore, the active yaw control strategy described in step 3) includes the following steps:

[0027] a) Calculate the average predicted wind speed over the prediction period, and set different yaw error angle thresholds T based on the average predicted wind speed. angle and yaw error duration threshold T time ;

[0028] b) Obtain the yaw error angle φ and duration T by using the predicted average wind direction every 5 seconds and the nacelle angle θ of the wind turbine. When the yaw error angle φ is greater than the preset yaw error angle threshold T... angle The duration T is greater than the preset yaw error duration threshold T. time At that time, the control cabin yaws at a speed of 0.6 (°) / s. After m seconds, the yaw error is φ. The yaw adjustment time is calculated to be t = φ / 0.6.

[0029] c) Initiate the yaw maneuver (mt) seconds in advance until θ = real-time wind direction angle;

[0030] d) Finally, if the time exceeds the current forecast period, the average forecast wind speed for the next forecast period is recalculated.

[0031] Furthermore, step A) calculating the total load on the entire blade includes:

[0032] Divide the blade into n segments and calculate the relative wind speed of each segment.

[0033] (1)

[0034] in, It predicts wind speed. It is the blade angular velocity. It is the radius of rotation of the midpoint of the leaf segment;

[0035] Calculate the lift and drag per unit area on the leaf segment.

[0036] (2)

[0037] Where L is lift. D is the lift coefficient, and D is the drag coefficient. It is the drag coefficient. It is the arc length of the leaf segment. It is air density;

[0038] The formula for calculating the torque of aerodynamic forces on each blade segment is as follows:

[0039] (3)

[0040] in, It is the torque of aerodynamic force. It is the windward angle of the blades;

[0041] Integrating the torque over each blade segment yields the total load on the entire blade:

[0042] (4)

[0043] in, It is the total load. It is the first The torque of a leaf segment.

[0044] Furthermore, in step B), the calculated total blade load is transferred from the rotating reference frame to a fixed non-rotating reference frame to obtain the matrix expression of the blade load, including:

[0045] Based on the following signal conversion relationship:

[0046] (5)

[0047] (6)

[0048] Where: B is the total number of blades, b is the blade number, and q is the number of blades. (b) For the total load, express equations (5) and (6) in matrix form, until:

[0049] (7)

[0050] In the formula, using , and These represent the loads on each blade;

[0051] Rewriting formula (7) in vector representation yields...

[0052] (8)

[0053] in, For collective mode, It is a differential mode.

[0054] Furthermore, in step C): the independent pitch control equations for the differential mode are designed as follows:

[0055] (9)

[0056] in: For proportional gain; It is the Laplace operator.

[0057] The calculation formula for an inverse notch filter set at a frequency of 1P is as follows:

[0058] (10)

[0059] In the formula The frequency of the inverse notch filter 1P is... and For the parameters of the inverse notch filter 1P respectively;

[0060] The calculation formula for a high-pass filter with a cutoff frequency of 0.4Hz is as follows:

[0061] (11)

[0062] In the formula These are the parameters of the high-pass filter. and These are the frequencies of the high-pass filter;

[0063] It is a low-pass filter with a cutoff frequency of 2Hz, and its calculation formula is as follows:

[0064] (12)

[0065] In the formula This is the frequency of the low-pass filter. These are the parameters of the low-pass filter;

[0066] The calculation formula for a standard notch filter set at the 3P frequency is as follows:

[0067] (13)

[0068] In the formula For the 3P frequency of the notch filter, and These are the 3P parameters of the notch filter;

[0069] The independent pitch control equations designed for collective mode are as follows:

[0070] (14)

[0071] in: For proportional gain,

[0072] It is an inverse notch filter set at the 2P frequency, and its calculation formula is as follows:

[0073] (15)

[0074] In the formula The frequency of the inverse notch filter 2P is... and These are the parameters of the inverse notch filter 2P;

[0075] It is a high-pass filter with a cutoff frequency of 3 Hz, and its calculation formula is as follows:

[0076] (16)

[0077] In the formula These are the parameters of the high-pass filter. and These are the frequencies of the high-pass filter;

[0078] It is a low-pass filter with a cutoff frequency of 1.3 Hz, and its calculation formula is as follows:

[0079] (17)

[0080] In the formula This is the frequency of the low-pass filter. These are the parameters for the low-pass filter.

[0081] The beneficial effects of this invention are:

[0082] 1. This invention presents a composite load reduction control method for two-bladed wind turbines under extreme wind conditions. It proposes a wind condition prediction model based on a combination of Variational Mode Decomposition (VMD), Ensemble Empirical Mode Decomposition (EEMD), Long Short-Term Memory (LSTM), and Least Squares Support Vector Machine (LSSVM). This model achieves accurate prediction of short-term wind speed and direction, enabling the turbine to perform more precise and rapid active yaw control to cope with extreme wind conditions. This reduces the impact on the blades and yaw motor under extreme wind conditions, thus improving the stability of the turbine system. Furthermore, accurate yaw control lays the foundation for independent blade pitch control.

[0083] 2. The composite load reduction control method for two-bladed wind turbines under extreme wind conditions of the present invention combines the passive yaw and active yaw of the two-bladed wind turbine. Under normal circumstances, the two-bladed wind turbine is used as a downwind wind turbine to automatically yaw, which can reduce the energy consumption of yaw control and improve the overall performance of the wind turbine.

[0084] 3. The present invention provides a composite load reduction control method for two-bladed wind turbines under extreme wind conditions. It offers a linear non-singular coordinate transformation applicable to two-bladed wind turbines and designs a novel independent pitch control equation for two-bladed wind turbines based on this, thereby improving the accuracy of independent pitch control and enhancing the stability of the system. Attached Figure Description

[0085] Figure 1 This is a schematic diagram of the wind condition prediction model structure.

[0086] Figure 2 This is a flowchart of the yaw control strategy.

[0087] Figure 3 This is a flowchart of the combined load reduction process for a two-bladed fan. Detailed Implementation

[0088] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0089] The composite load reduction control method for two-blade wind turbines under extreme wind conditions in this embodiment includes the following steps:

[0090] 1) Wind condition data is collected by wind direction and wind speed sensors. The wind condition data is time series data including wind speed and wind direction.

[0091] 2) Input the raw wind data collected in step 1) into the wind prediction model to predict wind speed and direction. The wind prediction model mentioned in this step consists of a VMD model, an EEMD model, an LSTM model, and an LSSVM model. The training and wind prediction steps of the wind prediction model include:

[0092] ① The original wind condition data is decomposed using the VMD model to obtain K subsequences and residuals. This step is based on the signal decomposition method, which can effectively extract features from the wind condition data and improve the accuracy of the prediction model.

[0093] ② The residual data obtained in step ① is further decomposed using the EEMD model to obtain N subsequences and residuals. Although VMD performs excellently in signal decomposition and feature extraction, the residual information remaining after VMD decomposition is discarded in existing methods, resulting in the loss of some information. This embodiment further extracts the residual information using the EEMD model, which helps to further improve the model's prediction accuracy.

[0094] ③ Input the K subsequences obtained in step ① and the N subsequences obtained in step ②, along with the residuals, into the LSTM model to train it. During the training process, the wind condition prediction data and wind condition prediction error data output by the LSTM model are obtained. The LSTM network can extract complex feature relationships between long-term and short-term time series. Since wind condition changes are closely related to time, this embodiment uses the LSTM network to extract complex feature relationships from wind condition time series data, which can significantly improve the accuracy of wind condition prediction.

[0095] ④ Input the wind condition prediction error data obtained in step ③ into the LSSVM model to train the LSSVM model. During the training process, the wind condition prediction error correction data output by the LSSVM model is obtained. LSSVM is a mini-batch data modeling algorithm that transforms linearly inseparable problems in low-dimensional space into high-dimensional space for solution. It has a good fitting ability. By correcting the prediction error through the LSSVM model, the prediction accuracy of the model can be further improved.

[0096] ⑤ Add the wind condition prediction data output by the LSTM model in step ③ to the wind condition prediction error correction data output by the LSSVM model in step ④ to obtain the final wind condition prediction data, which includes predicted wind speed and predicted wind direction.

[0097] 3) Determine the wind conditions within the forecast period based on wind forecast data, and select the corresponding strategy to control the yaw of the two-bladed wind turbine based on the wind conditions:

[0098] ① Set the maximum wind speed and the maximum wind direction change angle within the prediction period. If the predicted wind speed obtained within the prediction period is less than the set maximum wind speed and the predicted wind direction change angle is less than the set maximum wind direction change angle, then the wind conditions within the prediction period are normal wind conditions; otherwise, the wind conditions within the prediction period are extreme wind conditions.

[0099] ② If the wind conditions are normal during the forecast period, a passive yaw control strategy will be adopted during the forecast period to automatically align the yaw angle of the wind turbine with the wind direction.

[0100] Under normal wind conditions, the aerodynamic torque of a wind turbine can be approximated by the windward area of ​​the turbine blades and the wind speed. Assuming the yaw angle of the wind turbine... It is the angle of deviation from the ideal direction of the wind, the aerodynamic torque. It can be represented as:

[0101]

[0102] in, It is an aerodynamic coefficient related to the design of wind turbine blades, reflecting the ability of wind turbine blades to generate aerodynamic torque. It is the effective windward area of ​​the wind turbine blades. This is the real-time wind speed.

[0103] Based on the generated aerodynamic torque, the wind turbine will adjust its yaw angle. The yaw angle is adjusted until it is as close as possible to the wind direction (ideal yaw angle). The yaw angle of the wind turbine changes with wind speed and direction, and the rate of change of the yaw angle can be described as follows:

[0104]

[0105] in, It is the rate of change of the yaw angle. It is the yaw moment of inertia of the wind turbine.

[0106] Combining the above two equations, we obtain the rate of change of yaw angle for a two-bladed wind turbine in passive yaw mode:

[0107]

[0108] If extreme wind conditions occur during the forecast period, the wind turbines require more precise adjustments to ensure optimal operating efficiency. Therefore, an active yaw control strategy is employed to adjust the yaw angle of the wind turbines during the forecast period. The active yaw control strategy includes the following steps:

[0109] a) Calculate the average predicted wind speed over the prediction period, and set different yaw error angle thresholds T based on the average predicted wind speed. angleand yaw error duration threshold T time The smaller the predicted average wind speed, the smaller the wind's impact on the system. The set yaw error angle threshold T... angle and yaw error duration threshold T time The larger the value of the predicted average wind speed, the more significant the wind's impact on the system. The set yaw error angle threshold T... angle and yaw error duration threshold T time The smaller the value, the better. Specifically, wind speeds can be divided into different ranges, such as: a low wind speed range (less than 5 m / s), a medium wind speed range (5 to 10 m / s), and a high wind speed range (greater than 10 m / s). Each wind speed range corresponds to a yaw error angle threshold T. angle and yaw error duration threshold T time Thus, based on the wind speed range described by the predicted average wind speed, the corresponding yaw error angle threshold T can be obtained. angle and yaw error duration threshold T time The finer the wind speed range is divided, the higher the yaw error angle threshold T should be. angle and yaw error duration threshold T time The more suitable.

[0110] In this embodiment The optimization range is (5°, 6°, 7°, ..., 30°). The optimization range is (5, 10, 15, ..., 60 s). The average predicted wind speed over 30 minutes is used as the basis for setting the yaw control parameters, and the average predicted wind direction every 5 seconds is used as the basis for judging whether the yaw action is premature. Selecting the average wind direction can smooth out small-range and irregular fluctuations in the second-level wind direction data, which can effectively improve the accuracy of the prediction model and the accuracy of the yaw action judgment.

[0111] b) Obtain the yaw error angle φ and duration T by using the predicted average wind direction every 5 seconds and the nacelle angle θ of the wind turbine. When the yaw error angle φ is greater than the preset yaw error angle threshold T... angle The duration T is greater than the preset yaw error duration threshold T. time At that time, the control cabin yaws at a speed of 0.6 (°) / s. After m seconds, the yaw error is φ. The yaw adjustment time is t=φ / 0.6.

[0112] c) Start the yaw action (mt) seconds in advance until θ = real-time wind direction angle.

[0113] d) Finally, if the time exceeds the current forecast period, the average forecast wind speed for the next forecast period is recalculated.

[0114] 4) Implement independent pitch control for each blade of the two-blade wind turbine to reduce the load on each blade, including the following steps:

[0115] A) Calculate the total load on the entire blade, including:

[0116] Divide the blade into n segments and calculate the relative wind speed of each segment.

[0117] (1)

[0118] in, It predicts wind speed. It is the blade angular velocity. It is the radius of rotation of the midpoint of the leaf segment.

[0119] Calculate the lift and drag per unit area on the leaf segment.

[0120] (2)

[0121] Where L is lift. D is the lift coefficient, and D is the drag coefficient. It is the drag coefficient. It is the arc length of the leaf segment. It refers to air density.

[0122] The formula for calculating the torque of aerodynamic forces on each blade segment is as follows:

[0123] (3)

[0124] in, It is the torque of aerodynamic force. It refers to the angle at which the blades face the wind.

[0125] Integrating the torque over each blade segment yields the total load on the entire blade:

[0126] (4)

[0127] in, It is the total load. It is the first The torque of a leaf segment.

[0128] B) Transfer the calculated total blade load from the rotating reference frame to a fixed non-rotating reference frame to obtain the matrix expression of the blade load. Then, rewrite the matrix expression as a vector representation, and derive the collective and difference modes of the blade load from the vector equations. This step specifically includes:

[0129] Based on the following signal conversion relationship:

[0130] (5)

[0131] (6)

[0132] Where: B is the total number of blades, b is the blade number, and q is the number of blades. (b) For the total load, express equations (5) and (6) in matrix form, until:

[0133] (7)

[0134] In the formula, using , and These represent the loads on each blade;

[0135] Rewriting formula (7) in vector representation yields...

[0136] (8)

[0137] in, For collective mode, It is a differential mode;

[0138] The corresponding inverse transformation is given by the following equation:

[0139] (9)

[0140] in, , These are the rotation angles of the two blade rotation domains, respectively. and These represent the rotation angles in the non-rotational domains of the two blades, respectively.

[0141] C) Design independent pitch control equations for differential mode and independent pitch control equations for collective mode, and apply the pitch control equations for both differential and collective modes in parallel to reduce blade loading. This step specifically includes:

[0142] The independent pitch control equations designed for differential mode are as follows:

[0143] (10)

[0144] in: For proportional gain; It is the Laplace operator.

[0145] The calculation formula for an inverse notch filter set at a frequency of 1P is as follows:

[0146] (11)

[0147] In the formula The frequency of the inverse notch filter 1P is... and To determine the parameters of the inverse notch filter 1P, in this embodiment, Set to 0.695 Hz. Set to 0.9, Set to 0.009.

[0148] The calculation formula for a high-pass filter with a cutoff frequency of 0.4Hz is as follows:

[0149] (12)

[0150] In the formula These are the parameters of the high-pass filter. and These are the frequencies of the high-pass filter; in this embodiment, Set to 0.7, Set to 0.4*2πrad / s; Set to 0.0004.2 rad / second.

[0151] It is a low-pass filter with a cutoff frequency of 2Hz, and its calculation formula is as follows:

[0152] (13)

[0153] In the formula This is the frequency of the low-pass filter. These are the parameters of the low-pass filter; in this embodiment, Set to 4πrad / s Set it to 0.7.

[0154] The calculation formula for a standard notch filter set at the 3P frequency is as follows:

[0155] (14)

[0156] In the formula For the 3P frequency of the notch filter, and These are the 3P parameters of the notch filter; in this embodiment, Set to 2.085 Hz. Set to 0.3, Set to 0.006.

[0157] The independent pitch control equations designed for collective mode are as follows:

[0158] (15)

[0159] in: For proportional gain,

[0160] It is an inverse notch filter set at the 2P frequency, and its calculation formula is as follows:

[0161] (16)

[0162] In the formula The frequency of the inverse notch filter 2P is... and These are the parameters of the inverse notch filter 2P; in this embodiment, Set to 1.39Hz. Set to 0.5. Set to 0.005.

[0163] It is a high-pass filter with a cutoff frequency of 3 Hz, and its calculation formula is as follows:

[0164] (17)

[0165] In the formula These are the parameters of the high-pass filter. and These are the frequencies of the high-pass filter; in this embodiment, Set to 0.7, Set to 2.6 rad / s, Set to 0.0008π rad / s.

[0166] It is a low-pass filter with a cutoff frequency of 1.3 Hz, and its calculation formula is as follows:

[0167] (18)

[0168] In the formula This is the frequency of the low-pass filter. These are the parameters of the low-pass filter; in this embodiment... Set to 1.39Hz. Set it to 0.7.

[0169] During the control of wind turbine pitch, by jointly executing the control equations (10) and (15) in the embodiment, the 1P and 2P loads in the blade root bending moment can be removed. Executing control equation (10) also removes the 1P rotating hub torque, reducing the load transfer from the blade to the shaft. Executing control equation (15) further reduces the 2P load in the tower base moment.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A composite load reduction control method for a two-bladed wind turbine under extreme wind conditions, characterized in that: Includes the following steps: 1) Wind condition data is collected by wind direction and wind speed sensors, and the wind condition data is time series data including wind speed and wind direction; 2) Input the raw wind data collected in step 1) into the wind prediction model to predict wind speed and direction; 3) Determine the wind conditions within the forecast period based on wind forecast data, and select the corresponding strategy to control the yaw of the two-bladed wind turbine based on the wind conditions: ① Set the maximum wind speed and the maximum wind direction change angle within the prediction period. If the predicted wind speed obtained within the prediction period is less than the set maximum wind speed and the predicted wind direction change angle is less than the set maximum wind direction change angle, then the wind conditions within the prediction period are normal wind conditions; otherwise, the wind conditions within the prediction period are extreme wind conditions. ② If the wind conditions are normal during the forecast period, a passive yaw control strategy will be adopted during the forecast period to automatically align the yaw angle of the wind turbine with the wind direction. If extreme wind conditions occur during the forecast period, an active yaw control strategy will be used to adjust the yaw angle of the wind turbine during the forecast period. 4) Implement independent pitch control for each blade of the two-blade wind turbine to reduce the load on each blade, including the following steps: A) Calculate the total load on the entire blade, including: Divide the blade into n segments and calculate the relative wind speed of each segment. （1） in, It predicts wind speed. It is the blade angular velocity. It is the radius of rotation of the midpoint of the leaf segment; Calculate the lift and drag per unit area on the leaf segment. （2） Where L is lift. D is the lift coefficient, and D is the drag coefficient. It is the drag coefficient. It is the arc length of the leaf segment. It is air density; The formula for calculating the torque of aerodynamic forces on each blade segment is as follows: （3） in, It is the torque of aerodynamic force. It is the windward angle of the blades; Integrating the torque over each blade segment yields the total load on the entire blade: （4） in, It is the total load. It is the first The torque of a leaf segment; B) The calculated total blade load is transferred from the rotating reference frame to a fixed non-rotating reference frame to obtain the matrix expression of the blade load. This matrix expression is then rewritten as a vector representation. From the vector equations, the collective and difference modes of the blade load are obtained, including: Based on the following signal conversion relationship: （5） （6） Where: B is the total number of blades, b is the blade number, and q is the number of blades. (b) For the total load, formulas (5) and (6) are expressed in matrix form, yielding: （7） In the formula, using , and These represent the loads on each blade; Rewriting formula (7) in vector representation yields: （8） in, For collective mode, It is a differential mode; C) Design independent pitch control equations for differential mode and independent pitch control equations for collective mode, and apply the pitch control equations for differential mode and collective mode in parallel to reduce blade load.

2. The composite load reduction control method for two-bladed wind turbines under extreme wind conditions according to claim 1, characterized in that: Step 2) describes a wind condition prediction model composed of VMD, EEMD, LSTM, and LSSVM models. The training and prediction steps for this model include: ① The original wind condition data is decomposed using the VMD model to obtain K subsequences and residuals; ② The residual data obtained in step ① are further decomposed using the EEMD model to obtain N subsequences and residuals; ③ Input the K subsequences obtained from step ① and the N subsequences obtained from step ②, along with the residuals, into the LSTM model to train the LSTM model. During the training process, obtain the wind condition prediction data and wind condition prediction error data output by the LSTM model. ④ Input the wind condition prediction error data obtained in step ③ into the LSSVM model to train the LSSVM model. During the training process, the wind condition prediction error correction data output by the LSSVM model is obtained. ⑤ Add the wind condition prediction data output by the LSTM model in step ③ to the wind condition prediction error correction data output by the LSSVM model in step ④ to obtain the final wind condition prediction data, which includes predicted wind speed and predicted wind direction.

3. The composite load reduction control method for two-bladed wind turbines under extreme wind conditions according to claim 2, characterized in that: The active yaw control strategy described in step 3) includes the following steps: a) Calculate the average predicted wind speed over the prediction period, and set different yaw error angle thresholds T based on the average predicted wind speed. angle and yaw error duration threshold T time ; b) Obtain the yaw error angle φ and duration T by using the predicted average wind direction every 5 seconds and the nacelle angle θ of the wind turbine. When the yaw error angle φ is greater than the preset yaw error angle threshold T... angle The duration T is greater than the preset yaw error duration threshold T. time At that time, the control cabin yaws at a speed of 0.6 (°) / s. After m seconds, the yaw error is φ. The yaw adjustment time is calculated to be t = φ / 0.

6. c) Start the yaw maneuver (mt) seconds in advance until θ = real-time wind direction angle; d) Finally, if the time exceeds the current forecast period, the average forecast wind speed for the next forecast period is recalculated.

4. The composite load reduction control method for two-bladed wind turbines under extreme wind conditions according to claim 1, characterized in that: In step C): The independent pitch control equations for the differential mode are designed as follows: （9） in: For proportional gain; It is the Laplace operator; The calculation formula for an inverse notch filter set at a frequency of 1P is as follows: （10） In the formula The frequency of the inverse notch filter 1P is... and For the parameters of the inverse notch filter 1P respectively; The calculation formula for a high-pass filter with a cutoff frequency of 0.4Hz is as follows: （11） In the formula These are the parameters of the high-pass filter. and These are the frequencies of the high-pass filter; It is a low-pass filter with a cutoff frequency of 2Hz, and its calculation formula is as follows: （12） In the formula This is the frequency of the low-pass filter. These are the parameters of the low-pass filter; The calculation formula for a standard notch filter set at the 3P frequency is as follows: （13） In the formula For the 3P frequency of the notch filter, and These are the 3P parameters of the notch filter; The independent pitch control equations designed for collective mode are as follows: （14） in: For proportional gain, It is an inverse notch filter set at the 2P frequency, and its calculation formula is as follows: （15） In the formula The frequency of the inverse notch filter 2P is... and These are the parameters of the inverse notch filter 2P; It is a high-pass filter with a cutoff frequency of 3 Hz, and its calculation formula is as follows: （16） In the formula These are the parameters of the high-pass filter. and These are the frequencies of the high-pass filter; It is a low-pass filter with a cutoff frequency of 1.3 Hz, and its calculation formula is as follows: （17） In the formula This is the frequency of the low-pass filter. These are the parameters for the low-pass filter.

Images