Yaw control methods, devices, electronic equipment and storage media

Abstract

This application discloses a yaw control method, device, electronic device, and storage medium, relating to the field of yaw control technology. The yaw control method includes: acquiring the composite wind speed and composite wind direction at the wheel hub; calculating the yaw error of the yaw system based on the composite wind direction and the nacelle position; performing fuzzification and defuzzification operations on the composite wind speed and yaw error sequentially to obtain weighting coefficients; obtaining the predicted wind direction based on the composite wind direction; constructing a multi-step prediction model based on a preset target yaw prediction sequence; using the predicted wind direction as input to the multi-step prediction model to output the yaw prediction error; constructing a multi-objective optimization function based on the weighting coefficients and the target yaw prediction sequence; calculating the predicted yaw action based on the multi-objective optimization function and the yaw prediction error; and executing the predicted yaw action through a yaw actuator. This application predicts yaw actions based on the dynamic changes in wind speed and wind direction, reducing wind energy loss.

Description

Technical Field

[0001] This application relates to the field of yaw control technology, and in particular to a yaw control method, device, electronic device and storage medium. Background Technology

[0002] As a core component of clean and renewable energy, wind power generation has become a research hotspot in the energy field, with its technological development and efficiency improvement being particularly important. The power generation efficiency of wind turbines is highly dependent on their wind capture performance, and the yaw system, as a key device for adjusting the nacelle's orientation to align with the wind direction, directly determines the wind energy capture efficiency, thus affecting power generation revenue. In existing yaw control technologies, traditional predefined logic control is widely used due to its simple structure. However, this method has poor adaptability to dynamic changes in wind speed and direction, and is prone to lag under complex wind conditions, leading to accumulated yaw errors and wind energy loss. Summary of the Invention

[0003] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a yaw control method, device, electronic equipment, and storage medium, which can predict yaw actions based on dynamic changes in wind speed and direction, thereby reducing wind energy loss.

[0004] The yaw control method of the first aspect of this application includes: Obtain the composite wind speed and composite wind direction at the hub; Based on the combined wind direction and the position of the cabin, the yaw error of the yaw system is calculated; The synthesized wind speed and the yaw error are sequentially subjected to fuzzification and defuzzification operations to obtain weighting coefficients; Based on the synthesized wind direction, the wind direction data within the preset prediction step size is predicted to obtain the predicted wind direction; A multi-step prediction model is constructed based on a pre-defined target yaw prediction sequence. The predicted wind direction is used as the input to the multi-step prediction model to output the yaw prediction error, which is configured as the yaw error of the yaw system predicted within the step size. Based on the weighting coefficients and the target yaw prediction sequence, a multi-objective optimization function is constructed; Based on the multi-objective optimization function and the yaw prediction error, the predicted yaw action is calculated; The predicted yaw action is performed by the yaw actuator.

[0005] The yaw control method according to the embodiments of this application has at least the following beneficial effects: Based on the acquired synthetic wind direction, the yaw error is obtained. The yaw error and the acquired synthetic wind speed are fuzzified and defuzzified, so that the final weight coefficients are adaptively changed according to the dynamic changes of synthetic wind speed and synthetic wind direction. This solves the problem of poor adaptability to dynamic changes of wind speed and wind direction in traditional yaw control methods. The wind direction data within the prediction step is predicted to obtain the predicted wind direction. A multi-step prediction model is constructed. Based on the predicted wind direction and the multi-step prediction model, the yaw error within the prediction step is predicted to obtain the yaw prediction error. This facilitates the construction of a multi-objective function through the yaw prediction error and the weight coefficients. Then, the optimal solution of the yaw action at the next moment is calculated through the multi-objective function and the yaw prediction error to obtain the predicted yaw action. The predicted yaw action is then output to the yaw actuator for execution. This avoids the situation where the yaw action of the traditional yaw control method is prone to lag, reduces the accumulation of yaw error, and reduces wind energy loss.

[0006] According to some embodiments of this application, the step of sequentially performing fuzzification and defuzzification operations on the synthetic wind speed and the yaw error to obtain weighting coefficients includes: A fuzzy controller is constructed based on a preset fuzzy rule table; The synthesized wind speed and the yaw error are used as inputs to the fuzzy controller to output the original weight coefficients; The original weight coefficients are processed using the centroid method to obtain the weight coefficients.

[0007] According to some embodiments of this application, obtaining the composite wind speed and composite wind direction at the hub includes: The initial wind speed and initial wind direction at the wheel hub were obtained using a laser wind-measuring radar. The initial wind speed and initial wind direction are reconstructed to obtain the composite wind speed and composite wind direction.

[0008] According to some embodiments of this application, the step of predicting the wind direction data within a preset prediction step size based on the synthesized wind direction to obtain the predicted wind direction includes: The synthesized wind direction is preprocessed to obtain mid-frequency and low-frequency data; The mid-frequency data and the low-frequency data are used as inputs to the trained wind direction prediction model to output the predicted wind direction.

[0009] According to some embodiments of this application, the preprocessing of the synthetic wind direction to obtain mid-frequency data and low-frequency data includes: The composite wind direction is decomposed using three-level variational mode decomposition to obtain the eigenmode functions of the mid-frequency component and the eigenmode functions of the low-frequency component; Constraint operations and standardization operations are performed sequentially on the intrinsic mode functions of the mid-frequency component and the intrinsic mode functions of the low-frequency component to obtain the mid-frequency data and the low-frequency data.

[0010] According to some embodiments of this application, before using the mid-frequency data and the low-frequency data as input to the trained wind direction prediction model to output the predicted wind direction, the method further includes: Based on the mid-frequency data and the low-frequency data, a time series sample is constructed; The time series samples are processed using a sliding window to obtain the actual dataset and the real label dataset; Based on the actual dataset and the real label dataset, the preset initial wind direction prediction model is trained to obtain the trained wind direction prediction model.

[0011] According to some embodiments of this application, the multi-step prediction model is as follows: ; Where k represents time k, The yaw prediction error at time k+m is represented by . The predicted wind direction at time k+m is represented by [the given information]. This represents the predicted position of the cabin at time k+m-1. T represents the yaw prediction rate at time k+m. C represents the control period, and m represents the prediction step size.

[0012] According to a second aspect of this application, a yaw control device is applied to a yaw system, the yaw system including a yaw actuator and a hub. The device includes: The first parameter acquisition module is configured to acquire the composite wind speed and composite wind direction at the hub; and calculate the yaw error of the yaw system based on the composite wind direction and the position of the nacelle. Fuzzy inference module; the fuzzy inference module is configured to perform fuzzification and defuzzification operations on the synthetic wind speed and the yaw error in sequence to obtain weighting coefficients; The second parameter acquisition module is configured to predict the wind direction data within a preset prediction step size based on the synthesized wind direction, and obtain the predicted wind direction. Yaw Decision Module: The yaw decision module is configured to construct a multi-step prediction model based on a preset target yaw prediction sequence; the predicted wind direction is used as the input to the multi-step prediction model to output a yaw prediction error, which is configured as the yaw error of the yaw system predicted within the step size. A multi-objective optimization function is constructed based on the weighting coefficients and the target yaw prediction sequence; the predicted yaw action is calculated based on the multi-objective optimization function and the yaw prediction error; and the predicted yaw action is executed through the yaw actuator.

[0013] The yaw control device according to the second aspect of this application has at least the following beneficial effects: A first parameter acquisition module acquires the synthetic wind direction to obtain the yaw error; a fuzzy inference module performs fuzzification and defuzzification operations on the yaw error and the acquired synthetic wind speed, so that the final weight coefficients adapt to the dynamic changes of the synthetic wind speed and synthetic wind direction, solving the problem of poor adaptability to dynamic changes in wind speed and wind direction in traditional yaw control methods; a second parameter acquisition module predicts the wind direction data within the prediction step to obtain the predicted wind direction; a yaw decision module constructs a multi-step prediction model, and based on the predicted wind direction and the multi-step prediction model, predicts the yaw error within the prediction step to obtain the yaw prediction error. This facilitates the construction of a multi-objective function using the yaw prediction error and weight coefficients, and then calculates the optimal solution for the yaw action at the next moment using the multi-objective function and the yaw prediction error to obtain the predicted yaw action. The predicted yaw action is then output to the yaw execution mechanism for execution, avoiding the lag in yaw action that is common in traditional yaw control methods, reducing the accumulation of yaw error, and reducing wind energy loss.

[0014] An electronic device according to a third aspect of this application includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the voltage regulation method of the voltage converter according to a first aspect of this application.

[0015] According to a fourth aspect embodiment of the present application, a computer-readable storage medium stores a computer program that, when executed by a processor, implements the voltage regulation method of the voltage converter described in the first aspect embodiment of the present application.

[0016] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0017] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a flowchart illustrating the steps of the yaw control method according to an embodiment of this application. Figure 2 This is a schematic diagram of a specific process for step S101; Figure 3 This is a flowchart illustrating a specific step in step S103; Figure 4 This is a flowchart illustrating a specific step in step S104; Figure 5 This is a flowchart illustrating a specific step in step S401; Figure 6 This is a flowchart illustrating steps S601 to S603. Figure 7 This is a schematic diagram of the yaw control device according to an embodiment of this application; Figure 8 This is a schematic diagram of the wind speed membership function in an embodiment of this application; Figure 9 This is a schematic diagram of the yaw error membership function in an embodiment of this application; Figure 10 This is a schematic diagram of the original weight coefficient membership function in an embodiment of this application; Figure 11 This is a schematic diagram of a fuzzy rule table in an embodiment of this application; Figure 12 This is a schematic diagram of the laser wind measuring radar emitting a laser beam according to an embodiment of this application; Figure 13 This diagram illustrates a performance comparison of the yaw control method, the traditional yaw control method, and the fixed-coefficient model prediction yaw control method according to embodiments of this application. Figure 14 This is a schematic diagram of the absolute yaw error and yaw time of the yaw control method in the median wind speed range according to an embodiment of this application. Figure 15 A schematic diagram illustrating the absolute yaw error and yaw time of the yaw control method in the median wind speed range for a fixed coefficient model prediction. Figure 16 This is a schematic diagram of the absolute yaw error and yaw time of the traditional yaw control method in the median wind speed range; Figure 17 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0018] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0019] In the description of this application, it should be understood that the orientation descriptions, such as up, down, front, back, left, right, etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0020] In the description of this application, "several" means one or more, "more than" means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. The use of "first" and "second" in the description is merely for distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0021] In the description of this application, unless otherwise expressly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0022] Currently, driven by the global energy transition and the "dual carbon" goal, wind power, as a core component of clean and renewable energy, has become a research hotspot in the energy field, with its technological development and efficiency improvement becoming a key area of ​​focus. The power generation efficiency of wind turbines is highly dependent on their wind capture performance, and the yaw system, as a crucial device for adjusting the nacelle's orientation to align with the wind direction, directly determines the wind energy capture efficiency, thus affecting power generation revenue. Among existing yaw control technologies, traditional predefined logic control is widely used due to its simple structure. However, this method has poor adaptability to dynamic changes in wind speed and direction, and is prone to lag under complex wind conditions, leading to accumulated yaw errors and wind energy losses. Studies show that approximately 2.7% of wind power generation losses are caused by yaw errors, significantly impacting the economic benefits of wind farms over long-term operation. Furthermore, some existing yaw control technologies cannot even adapt to dynamic changes, relying solely on fixed model parameters to predict yaw actions.

[0023] Based on this, this application proposes a yaw control method, device, electronic device, and storage medium, which aims to predict yaw actions based on the dynamic changes of wind speed and wind direction, so as to reduce the accumulation of yaw error and thus reduce wind energy loss. At the same time, the weight coefficients of the multi-objective optimization function for predicting yaw actions can be adaptively adjusted with the dynamic changes of wind speed and wind direction to adapt to complex wind conditions and improve the accuracy of prediction.

[0024] The yaw control method of the first aspect of this application is applied to a yaw system, the yaw system including a wind turbine, the wind turbine including a yaw actuator, a hub and a nacelle.

[0025] The first aspect of this application is based on... Figure 1 A schematic diagram of yaw control method. (Refer to...) Figure 1 , Figure 1 This is a flowchart illustrating the steps of the yaw control method according to an embodiment of this application. Figure 1 The illustrated process steps include, but are not limited to, steps S101 to S109.

[0026] Step S101: Obtain the composite wind speed and composite wind direction at the hub.

[0027] Step S102: Based on the synthetic wind direction and the position of the cabin, the yaw error of the yaw system is calculated.

[0028] Step S103: Perform fuzzification and defuzzification operations on the synthesized wind speed and yaw error in sequence to obtain the weighting coefficients.

[0029] Step S104: Based on the synthetic wind direction, predict the wind direction data within the preset prediction step size to obtain the predicted wind direction.

[0030] Step S105: Construct a multi-step prediction model based on the preset target yaw prediction sequence.

[0031] Step S106: The predicted wind direction is used as the input to the multi-step prediction model to output the yaw prediction error, which is configured as the yaw error of the predicted yaw system within the step size.

[0032] Step S107: Construct a multi-objective optimization function based on the weighting coefficients and the target yaw prediction sequence.

[0033] Step S108: Based on the multi-objective optimization function and yaw prediction error, the predicted yaw action is calculated.

[0034] Step S109: Perform the predicted yaw action through the yaw actuator.

[0035] The yaw control method according to the embodiments of this application has at least the following beneficial effects: Based on the acquired synthetic wind direction, the yaw error is obtained. The yaw error and the acquired synthetic wind speed are fuzzified and defuzzified, so that the final weight coefficients are adaptively changed according to the dynamic changes of synthetic wind speed and synthetic wind direction. This solves the problem of poor adaptability to dynamic changes of wind speed and wind direction in traditional yaw control methods. The wind direction data within the prediction step is predicted to obtain the predicted wind direction. A multi-step prediction model is constructed. Based on the predicted wind direction and the multi-step prediction model, the yaw error within the prediction step is predicted to obtain the yaw prediction error. This facilitates the construction of a multi-objective function through the yaw prediction error and the weight coefficients. Then, the optimal solution of the yaw action at the next moment is calculated through the multi-objective function and the yaw prediction error to obtain the predicted yaw action. The predicted yaw action is then output to the yaw actuator for execution. This avoids the situation where the yaw action of the traditional yaw control method is prone to lag, reduces the accumulation of yaw error, and reduces wind energy loss.

[0036] In some embodiments, refer to Figure 2 Step S101 may include, but is not limited to, steps S201 to S202.

[0037] Step S201: Use a laser wind radar to obtain the initial wind speed and initial wind direction at the wheel hub.

[0038] Step S201: Reconstruct the initial wind speed and initial wind direction respectively to obtain the composite wind speed and composite wind direction.

[0039] In step S201 of some embodiments, referring to Figure 12 The laser wind measuring radar is a four-beam laser wind measuring radar. It acquires line-of-sight wind speed data at a predetermined distance (e.g., 52.2m) in front of the wind turbine to obtain the initial wind speed and direction. The specific process is as follows: Under the assumption of uniform incoming wind, the four-beam laser wind measuring radar is placed on top of the turbine hub. It emits four laser beams: OA, OB, OC, and OD. OA and OB form one pair of horizontal beams, and OC and OD form another pair of horizontal beams. The angle between OA and OB is 30°, and the angle between OC and OD is also 30°. OA and OD form one pair of vertical beams, and OB and OC form another pair of vertical beams. The angle between OA and OD is 25°, and the angle between OB and OC is also 25°. The radial wind speeds of OA, OB, OC, and OD are then measured respectively. , , , .

[0040] Radial wind speed is decomposed into a spatial rectangular coordinate system to obtain orthogonal vectors. , , Where each radial wind speed is about an orthogonal vector , , The orthogonal analytic expression is shown below: ; ; ; ; ; In this formula, , , These are orthogonal vectors on the X, Y, and Z axes, respectively. , , , Radial wind velocities for OA, OB, OC, and OD, respectively. The angle between plane BOC or plane AOD and plane XZ in the spatial rectangular coordinate system. The angle between the laser beam (OA, OB, OC, or OD) and the plane XY in the spatial rectangular coordinate system is given. Let X be the angle between the laser beam (OA, OB, OC, or OD) and the X-axis in the Cartesian coordinate system, where plane XZ is the plane formed by the X-axis and Z-axis, and plane XY is the plane formed by the X-axis and Y-axis.

[0041] Furthermore, disregarding the wind speed vector along the Z-axis, the wind speed and direction on the upper plane can be derived: ; ; ; ; In this formula, and Radial wind velocities for OA and OB, respectively. and They are respectively and Orthogonal vectors on the X and Y axes The angle between plane BOC or plane AOD and plane XZ in the spatial rectangular coordinate system. Let OA or OB be the angle between OA or OB and the XY plane in the spatial rectangular coordinate system. Let OA or OB be the angle between OA or OB and the X-axis in the rectangular coordinate system. The wind speed at the top plane. The wind direction is indicated by the upper plane; where A and B are at the same horizontal height, and the upper plane is the horizontal plane at the height of A and B.

[0042] Similarly, the wind speed and direction on the lower plane can also be derived: ; ; ; ; In this formula, and Radial wind velocities at OC and OD, respectively. and They are respectively and Orthogonal vectors on the X and Y axes The angle between plane BOC or plane AOD and plane XZ in the spatial rectangular coordinate system. The angle between OC or OD and the XY plane in the spatial rectangular coordinate system. The angle between OC or OD and the X-axis in the Cartesian coordinate system. The wind speed on the lower plane. The lower plane represents the wind direction; where C and D are at the same horizontal height, and the lower plane is the horizontal plane at the height of C and D. The wind speeds and directions on the upper and lower planes are the initial wind speeds and directions, respectively.

[0043] It should be noted that ABCD forms a rectangle.

[0044] In step S201 of some embodiments, the desired resultant wind speed and resultant wind direction can be calculated from the initial wind speed and initial wind direction: ; ; ; ; In this formula, The height of the upper plane. The height of the lower plane. The wind speed at the top plane. The wind direction is on the upper plane. The wind speed on the lower plane. The wind direction is on the lower plane. This is vertical wind shear. This represents the rate of change of vertical wind direction. The height at the wheel hub. To synthesize wind speed, This represents the composite wind direction.

[0045] In steps S201 to S202 of this application, the wind speed along the laser can be obtained by the laser wind radar, but the wind direction is not necessarily along the direction of the laser. Therefore, converting the radial wind speed into an orthogonal vector can facilitate subsequent calculations.

[0046] In step S102 of some embodiments, the yaw error is obtained by subtracting the angle quantified by the synthetic wind direction from the angle of the cabin orientation.

[0047] In some embodiments, refer to Figure 3 Step S103 may include, but is not limited to, steps S301 to S303.

[0048] Step S301: Construct a fuzzy controller based on a preset fuzzy rule table.

[0049] In step S302, the synthesized wind speed and yaw error are used as inputs to the fuzzy controller to output the original weight coefficients.

[0050] Step S303: The original weight coefficients are processed using the centroid method to obtain the weight coefficients.

[0051] In step S301 of some embodiments, the constructed fuzzy rule table is as follows: Figure 11 As shown, refer to Figure 8 , Figure 9 and Figure 10 The fuzzy controller includes the input wind speed membership function and yaw error membership function, as well as the output original weight coefficient membership function.

[0052] In step S302 of some embodiments, the synthesized wind speed and yaw error are fuzzified using a fuzzy controller to output the original weighting coefficients. The specific logic flow of fuzzification is as follows: (Refer to...) Figure 8 , Figure 9 and Figure 10 Assuming the composite wind speed is 6, its membership degree in the wind speed membership function is 0.216 in S, 0.895 in M, and 0.002 in L. Assuming the yaw error is 10, its membership degree in the yaw error membership function is 0.111 in VS, 0.781 in S, 0.557 in M, and 0 in both L and VL. (Refer to...) Figure 11 The activation intensity is calculated based on the fuzzy rule table. Rule 1: If the composite wind speed is S and the yaw error is VS, then the original weighting coefficient of the output is VS.

[0053] Rule 15: If the composite wind speed is L and the yaw error is VL, then the output raw weighting coefficient is L, according to... Figure 11 There are a total of 15 rules, which will not be described one by one.

[0054] The activation intensity of Rule 1 is min (the membership degree of the composite wind speed S and the membership degree of the yaw error VS), which is equal to min(0.216, 0.111) = 0.111.

[0055] The activation intensity of rule 15 is min(the membership degree of the composite wind speed L and the membership degree of the yaw error VL), which is equal to min(0.002,0)=0. The activation intensity of the other rules can be obtained from this.

[0056] Next, calculate the maximum activation intensity corresponding to the original weight coefficients VS, S, M, L and VL respectively to obtain the original weight coefficients mapped by the original weight coefficient membership function. For example, the output original weight coefficients are VS including rules 1, 2, 6, 11 and 12. Assuming that the activation intensity of rule 1 is the largest, i.e. 0.111, substitute 0.111 as the x-axis into VS of the original weight coefficient membership function to obtain the corresponding original weight coefficient 0.716.

[0057] It should be noted that: yaw error reflects the wind turbine's alignment accuracy. To reduce the impact of uncertainty and improve generalization ability, the absolute value of the statistical average of yaw error is selected as the input of the fuzzy controller.

[0058] In step S303 of some embodiments, the specific process of defuzzification is as follows: using the centroid method, the centroid of the area enclosed by the membership function of the original weight coefficients and the original weight coefficients is taken as the weight coefficient, and the specific formula is: ; In this formula, These are the original weighting coefficients. The membership function of the original weight coefficients. Original weighting coefficients The scope of the discourse, These are the weighting coefficients.

[0059] Steps S301 to S303 of this application process the synthetic wind speed and yaw error through fuzzing and defuzzing to obtain weighting coefficients that are affected by the dynamic changes of the synthetic wind speed and yaw error.

[0060] In some embodiments, refer to Figure 4 Step S104 may include, but is not limited to, steps S401 to S402.

[0061] Step S401: Preprocess the composite wind direction to obtain mid-frequency and low-frequency data.

[0062] Step S402: Use the mid-frequency data and low-frequency data as input to the trained wind direction prediction model to output the predicted wind direction.

[0063] In some embodiments, refer to Figure 5 Step S401 may include, but is not limited to, steps S501 to S502.

[0064] Step S501: The composite wind direction is decomposed using three-layer variational mode decomposition to obtain the intrinsic mode functions of the mid-frequency component and the intrinsic mode functions of the low-frequency component.

[0065] Step S502: Constraint and standardization operations are performed on the intrinsic mode functions of the mid-frequency component and the intrinsic mode functions of the low-frequency component respectively to obtain mid-frequency data and low-frequency data.

[0066] In step S501 of some embodiments, the synthetic wind direction is decomposed into high-frequency component eigenmode functions, mid-frequency component eigenmode functions, and low-frequency component eigenmode functions using three-layer variational mode decomposition.

[0067] In step S502 of some embodiments, to ensure that the decomposed sequence is a mode function with a finite bandwidth and a center frequency, and that the sum of the bandwidths of each mode function is minimized, the constraint condition for the constraint operation is that the sum of all mode functions is equal to the synthetic wind direction sequence. The constraint variational model is as follows: ; In this formula, k can take values ​​from 1 to 3. It is the k-th mode function (i.e., the intrinsic mode function of the high-frequency component, the intrinsic mode function of the mid-frequency component, or the intrinsic mode function of the low-frequency component). For Dirac function, To synthesize the wind direction, Represents the partial derivative operator. Represented as the imaginary unit, where It is the Hilbert transform, used to obtain an analytical representation of a signal. t represents the time variable. middle It is the center frequency of the k-th modal function. Represented as the product of angular frequency and time, it is used to form a complex exponential function, frequency modulated on the signal, and modulated onto the baseband to evaluate its bandwidth. `st` is a standard abbreviation in optimization problems, meaning constrained by; here, the constraint is the function of all modes. The sum equals the original signal (i.e., the synthetic wind direction) ensures that the decomposed modes can completely reconstruct the original signal.

[0068] Find the optimal solution through continuous iteration and updates. , This method extracts and separates the synthetic wind direction frequency, obtaining the corresponding high-frequency, mid-frequency, and low-frequency terms. The residual frequency data after subtracting the extracted high-frequency, mid-frequency, and low-frequency terms from the synthetic wind direction is the residual term. Since the residual term has a small amplitude, it can be ignored. Therefore, the mid-frequency and low-frequency terms, which have more pronounced wind direction characteristics, are used as the wind direction reference for yaw control. The mid-frequency and low-frequency terms are then standardized using the following formula: ; ; ; In this formula, For mid-frequency or low-frequency terms, This is the mean of the mid-frequency or low-frequency term. The standard deviation of the mid-frequency or low-frequency term. This refers to standardized mid-frequency or low-frequency data.

[0069] Steps S501 to S502 of this application, through constraint operations, decompose the synthetic wind direction into high-frequency component intrinsic mode functions, mid-frequency component intrinsic mode functions, and low-frequency component intrinsic mode functions with finite bandwidth and center frequency, and minimize the sum of the bandwidths of each mode function; through standardization, the obtained mid-frequency and low-frequency data are scaled to a distribution range with a mean of 0 and a standard deviation of 1, so that the values ​​in the dataset are of similar magnitude, unify the scale of each component intrinsic mode function, and accelerate the convergence speed.

[0070] In some embodiments, refer to Figure 6 Before step S502, steps S601 to S603 may also be included, but are not limited to.

[0071] Step S601: Construct time series samples based on mid-frequency and low-frequency data.

[0072] Step S602: Use a sliding window to process the time series samples to obtain the actual dataset and the real label dataset.

[0073] Step S603: Based on the actual dataset and the real label dataset, train the preset initial wind direction prediction model to obtain the trained wind direction prediction model.

[0074] In step S601 of some embodiments, the intermediate frequency data and low frequency data are constructed into a time series sample.

[0075] In step S602 of some embodiments, a sliding window with a step size of one is used to process the time series samples to generate input-output pairs, specifically as follows: For the i-th sample, the input window covers the time step... Where T is 180, meaning the length of the input window is 180 time steps; the corresponding output window covers time steps. M is 60, meaning the length of the output window is 60 time steps (i.e., the prediction step size), to obtain the actual dataset X with an array shape of (N, T) and the true label dataset Y with an array shape of (N, M), where N is the total number of samples.

[0076] In step S603 of some embodiments, the wind direction prediction model is an LSTM (Long Short-Term Memory) model. The actual dataset and the real-label dataset are imported into the wind direction prediction model for training. The wind direction prediction model uses two LSTM layers, one dropout layer, and one output layer. Both the LSTM layer and the output layer contain 32 neurons. The dropout rate of the dropout layer is 0.2 to prevent overfitting. The trained wind direction prediction model can then be further evaluated using non-training samples to predict outputs, and its predictive performance can be assessed using various metrics. The coefficients, mean absolute error (MAE), or root mean square error (RMSE) are calculated using the following formulas: ; ; ; In this formula, The actual value; This is a predicted value; is the average of the true values; n is the total number of data points.

[0077] In step S402 of some embodiments, after constructing the wind direction prediction model, a set of synthetic wind directions with a time scale of 180 is input to obtain the predicted wind direction with a time scale of 60.

[0078] In step S105 of some embodiments, a multi-step prediction model is constructed using the target yaw prediction sequence.

[0079] It is understandable that the target yaw prediction sequence includes multiple yaw prediction rates, and the multi-step prediction model is shown below: ; Where k represents time k, This represents the yaw prediction error at time k+m. This represents the predicted wind direction at time k+m. This represents the predicted position of the cabin at time k+m-1. T represents the target yaw prediction rate at time k+m. C represents the control period, and m represents the prediction step size.

[0080] For example, in some embodiments, the cabin position at the next moment can be obtained by the current cabin position, the target yaw prediction rate at the next moment, and the control period, as specified in the formula: ; In this formula, Let k be the cabin position at time k. Let be the cabin position at time k+1. The target yaw prediction rate at time k+1. To control the period, k represents the k-th time.

[0081] In step S106 of some embodiments, the predicted wind direction is used as the input to the multi-step prediction model, and then the yaw prediction error is output.

[0082] In step S107 of some embodiments, a multi-objective optimization function is constructed, as shown below: ; ; ; ; In this formula, Let N be the dimensionless power reduction factor, N be the total number of samples, m be the prediction step size, and k be the time step at time k. For yaw prediction error, , The longest duration of the yaw system's operation. To predict the actual action time within the step size, This is the maximum distance the yaw system can rotate. To predict the actual rotation distance within the step size, To control the cycle, QF(1), QF(2), and QF(3) are the first, second, and third parts of the multi-objective optimization function, respectively. QF is the value of the multi-objective optimization function, i.e., the loss value. The smaller QF is, the better the final result.

[0083] It should be noted that the first part of the multi-objective optimization function, QF(1), considers power loss, while the second part, QF(2), and the third part, QF(3), consider the loss of the yaw actuator.

[0084] In step S108 of some embodiments, the yaw prediction rate includes three possible values: ; In this formula, where, For reverse yaw , For positive yaw 0 indicates no yaw, and corresponding forward and reverse yaw cannot be triggered simultaneously, meaning that in a set of yaw actions... In this process, both positive and negative yaw cannot exist simultaneously to avoid frequent yaw maneuvers. It can be seen that there is a mapping relationship between the predicted yaw maneuver and the predicted yaw rate, and it includes three types of maneuvers: positive yaw, negative yaw, and no yaw.

[0085] The target yaw prediction sequence is obtained through an exhaustive method. For example, to predict the situation 4 seconds into the future, with both the sampling period and the control period being 1 second, it is necessary to predict the yaw actions for the next 4 steps. Initially, the first step is predicted. In the first control period, all possible yaw prediction rates mapped to the predicted yaw actions for the next 4 steps are listed to form the yaw prediction sequence, such as: S1: 0, 0, 0, 0 (corresponding to no yaw, no yaw, no yaw, no yaw), S2: 0 S1: 0, 0, 0, 0; S2: 0.5, 0, 0, 0; S3: 0, 0.5, 0, 0, etc. Then, sequences with continuous yaw prediction rates are filtered out from the yaw prediction sequences, resulting in: S1: 0, 0, 0, 0; S2: 0.5, 0, 0, 0; S3: 0, 0.5, 0, 0; S4: 0, 0, 0.5, 0; S5: 0, 0, 0, 0.5; S6: 0.5, 0.5, 0, 0; S7: 0, 0, 0.5, 0.5; S8: 0, 0, 0.5, 0.5; S9: ... 0.5, 0.5, 0.5, 0, S10: 0, 0.5, 0.5, 0.5, S11: 0.5, 0.5, 0.5, 0.5, S12: -0.5, 0, 0, 0, S13: 0, -0.5, 0, 0, S 14: 0, 0, -0.5, 0, S15: 0, 0, 0, -0.5, S16: -0.5, -0.5, 0, 0, S17: 0, 0, -0.5, -0.5, S18: 0, 0, -0.5, -0.5, S S19: -0.5, -0.5, -0.5, 0; S20: 0, -0.5, -0.5, -0.5; S21: -0.5, -0.5, -0.5, -0.5. These 21 sequences are all used as target yaw prediction sequences (in this example, one target yaw prediction sequence includes four yaw prediction rates). It can be seen that a large number of invalid solutions are avoided, the solution efficiency is accelerated, and frequent yaw start-stop control is avoided, protecting the safe operation of subsequent yaw actuators. Multiple target yaw prediction sequences are successively substituted into the multi-step prediction model to obtain yaw prediction errors. These errors are then substituted into the multi-objective optimization function, which is repeated to obtain multiple loss values. The action mapped by the first yaw prediction rate in the target yaw prediction sequence with the smallest loss value is selected as the first predicted yaw action. For example, if S1 corresponds to the smallest calculated loss value, the first yaw prediction rate of S1 is 0, corresponding to no yaw. Therefore, the first predicted yaw action is no yaw, and this no yaw action is output to the yaw actuator. After the next control cycle, the above steps are repeated to predict the second step. In the next control cycle, the initial wind speed, initial wind direction, and cabin position are remeasured and updated before the second step prediction is performed, and so on, until the prediction of all four yaw actions is completed.

[0086] In step S109 of some embodiments, the yaw actuator begins execution immediately upon receiving the predicted yaw action.

[0087] Refer to Figure 13. Figure 13 This paper evaluates the performance of three control methods under the same wind speed and direction scenario through simulation examples: the yaw control method proposed in this application, the traditional yaw control method, and the fixed-coefficient model prediction yaw control method (hereinafter referred to as the fixed-coefficient yaw control method). The traditional yaw control method uses predefined logic control; when the wind direction change exceeds a certain threshold, the nacelle rotation is triggered after extending the corresponding time threshold, and forward and reverse yaws are interlocked and cannot be triggered simultaneously. The fixed-coefficient yaw control method does not consider wind speed as an auxiliary parameter in yaw decision-making; that is, the influence of wind speed is not considered during yaw control. MAE reflects the average level of yaw error. The percentage of power loss reflecting the yaw error. This reflects the total activation time of the yaw actuator. From... Figure 13 It can be seen that the yaw control method of this application is superior to the other two yaw control methods in three aspects. And through... Figure 14 , Figure 15 and Figure 16 Furthermore, it can be seen that the absolute yaw error and yaw time of the yaw control method of this application are better than the other two yaw control methods. That is, the absolute value of the yaw error is smaller and the required yaw time is shorter, thereby improving the efficiency and accuracy of yaw control.

[0088] It should be noted that wind direction determines the yaw alignment of the wind turbine, and wind speed affects the turbine's power capture potential (greater wind captures more power). Higher wind alignment accuracy usually means increased yaw maneuvering; and the higher the wind speed, the greater the potential power loss due to yaw error. Therefore, this application introduces fuzzy inference using wind speed to select weighting coefficients. However, using wind speed alone without considering yaw error would lead to an overemphasis on wind speed in yaw decisions—no yaw at low wind speeds, yaw at high wind speeds. Therefore, the range of yaw error must also be considered; excessively large yaw errors would increase the load on the wind turbine. Thus, wind speed and yaw error are chosen as the core parameters for wind turbine yaw (i.e., as inputs to the fuzzy controller), resulting in a yaw control method that performs better than traditional yaw control methods.

[0089] Reference Figure 7 , Figure 7 This is a schematic diagram of the structure of a yaw control device according to a second aspect embodiment of this application; the yaw control device of this application is applied to a yaw system, which includes a yaw actuator and a wheel hub; the yaw control device of this application embodiment includes: The first parameter acquisition module 701 is configured to acquire the composite wind speed and composite wind direction at the hub; and calculate the yaw error of the yaw system based on the composite wind direction and the position of the nacelle. Fuzzy inference module 702; Fuzzy inference module 702 is configured to perform fuzzification and defuzzification operations on the synthetic wind speed and yaw error in sequence to obtain weight coefficients; Second parameter acquisition module 703; The second parameter acquisition module 703 is configured to predict the wind direction data within a preset prediction step size based on the synthetic wind direction, and obtain the predicted wind direction; Yaw Decision Module 704: The yaw decision module 704 is configured to construct a multi-step prediction model based on a preset target yaw prediction sequence; the predicted wind direction is used as the input to the multi-step prediction model to output the yaw prediction error, which is configured as the yaw error of the predicted yaw system within the step size. A multi-objective optimization function is constructed based on the weighting coefficients and the target yaw prediction sequence; the predicted yaw action is calculated based on the multi-objective optimization function and the yaw prediction error; and the predicted yaw action is executed through the yaw actuator.

[0090] The yaw control device according to the second aspect of this application has at least the following beneficial effects: The first parameter acquisition module 701 acquires the synthetic wind direction to obtain the yaw error; the fuzzy inference module 702 performs fuzzification and defuzzification operations on the yaw error and the acquired synthetic wind speed, so that the final weight coefficients adaptively change according to the dynamic changes of the synthetic wind speed and synthetic wind direction, solving the problem of poor adaptability to dynamic changes in wind speed and wind direction in traditional yaw control methods; and the second parameter acquisition module 703 predicts the wind direction data within the prediction step to obtain the predicted wind direction. A multi-step prediction model is constructed through the yaw decision module 704. Based on the predicted wind direction and the multi-step prediction model, the yaw error within the prediction step is predicted to obtain the yaw prediction error. This facilitates the construction of a multi-objective function using the yaw prediction error and weighting coefficients. The optimal solution for the yaw action at the next moment is then calculated using the multi-objective function and the yaw prediction error to obtain the predicted yaw action. The predicted yaw action is then output to the yaw actuator for execution. This avoids the lag in yaw action that is common in traditional yaw control methods, reduces the accumulation of yaw error, and reduces wind energy loss.

[0091] An embodiment of the third aspect of this application also provides an electronic device, which includes a memory 1702 and a processor 1701. The memory 1702 stores a computer program, and the processor 1701 executes the computer program to implement the voltage regulation method of the voltage converter described in the first aspect embodiment. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.

[0092] Reference Figure 17, Figure 17 This is a schematic diagram of the structure of an electronic device according to one embodiment. The electronic device includes: The processor 1701 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 1702 can be implemented as a read-only memory, static storage device, dynamic storage device, or random access memory (RAM). The memory 1702 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1702 and is called and executed by the processor 1701 to execute the television bezel laser etching method of the embodiments of this application. The input / output interface 1703 is used to implement information input and output; The communication interface 1704 is used to enable communication and interaction between this device and other devices. Communication can be achieved via wired or wireless means. Bus 1705 transmits information between the various components of the device; The processor 1701, memory 1702, input / output interface 1703 and communication interface 1704 are connected to each other within the device via bus 1705.

[0093] A fourth aspect of this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the voltage regulation method of the voltage converter described in the first aspect of the embodiment.

[0094] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0095] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0096] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0097] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0098] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0099] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0100] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0101] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0102] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0103] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0104] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0105] Furthermore, it should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent will be obtained first. Moreover, the collection, use, and processing of this data will comply with relevant laws, regulations, and standards. Additionally, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user will be obtained through pop-ups or redirects to confirmation pages. Only after obtaining the user's separate permission or consent will the necessary user-related data for the proper functioning of these embodiments be acquired.

[0106] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A yaw control method, characterized in that, Applied to a yaw system, the yaw system including a yaw actuator, a hub, and a nacelle; The method includes: Obtain the composite wind speed and composite wind direction at the hub; Based on the combined wind direction and the position of the cabin, the yaw error of the yaw system is calculated; The synthesized wind speed and the yaw error are sequentially subjected to fuzzification and defuzzification operations to obtain weighting coefficients; Based on the synthesized wind direction, the wind direction data within the preset prediction step size is predicted to obtain the predicted wind direction; A multi-step prediction model is constructed based on a pre-defined target yaw prediction sequence. The predicted wind direction is used as the input to the multi-step prediction model to output the yaw prediction error, which is configured as the yaw error of the yaw system predicted within the step size. Based on the weighting coefficients and the target yaw prediction sequence, a multi-objective optimization function is constructed; Based on the multi-objective optimization function and the yaw prediction error, the predicted yaw action is calculated; The predicted yaw action is performed by the yaw actuator.

2. The yaw control method according to claim 1, characterized in that, The process of sequentially performing fuzzification and defuzzification operations on the synthesized wind speed and the yaw error to obtain weighting coefficients includes: A fuzzy controller is constructed based on a preset fuzzy rule table; The synthesized wind speed and the yaw error are used as inputs to the fuzzy controller to output the original weight coefficients; The original weight coefficients are processed using the centroid method to obtain the weight coefficients.

3. The yaw control method according to claim 1, characterized in that, The process of obtaining the composite wind speed and composite wind direction at the wheel hub includes: The initial wind speed and initial wind direction at the wheel hub were obtained using a laser wind-measuring radar. The initial wind speed and initial wind direction are reconstructed to obtain the composite wind speed and composite wind direction.

4. The yaw control method according to claim 1, characterized in that, The step of predicting the wind direction based on the synthesized wind direction within a preset prediction step size to obtain the predicted wind direction includes: The synthesized wind direction is preprocessed to obtain mid-frequency and low-frequency data; The mid-frequency data and the low-frequency data are used as inputs to the trained wind direction prediction model to output the predicted wind direction.

5. The yaw control method according to claim 4, characterized in that, The preprocessing of the synthesized wind direction to obtain mid-frequency and low-frequency data includes: The composite wind direction is decomposed using three-level variational mode decomposition to obtain the eigenmode functions of the mid-frequency component and the eigenmode functions of the low-frequency component; Constraint operations and standardization operations are performed sequentially on the intrinsic mode functions of the mid-frequency component and the intrinsic mode functions of the low-frequency component to obtain the mid-frequency data and the low-frequency data.

6. The yaw control method according to claim 4, characterized in that, Before using the mid-frequency data and the low-frequency data as input to the trained wind direction prediction model to output the predicted wind direction, the method further includes: Based on the mid-frequency data and the low-frequency data, a time series sample is constructed; The time series samples are processed using a sliding window to obtain the actual dataset and the real label dataset; Based on the actual dataset and the real label dataset, the preset initial wind direction prediction model is trained to obtain the trained wind direction prediction model.

7. The yaw control method according to claim 1, characterized in that, The multi-step prediction model is shown below: ； Where k represents time k, The yaw prediction error at time k+m is represented by . The predicted wind direction at time k+m is represented by [symbol missing]. This represents the predicted position of the cabin at time k+m-1. T represents the yaw prediction rate at time k+m. C represents the control period, and m represents the prediction step size.

8. A yaw control device, characterized in that, Applied to a yaw system, the yaw system including a yaw actuator and a hub; The device includes: The first parameter acquisition module is configured to acquire the composite wind speed and composite wind direction at the hub; and calculate the yaw error of the yaw system based on the composite wind direction and the position of the nacelle. Fuzzy inference module; the fuzzy inference module is configured to perform fuzzification and defuzzification operations on the synthetic wind speed and the yaw error in sequence to obtain weighting coefficients; The second parameter acquisition module is configured to predict the wind direction data within a preset prediction step size based on the synthesized wind direction, and obtain the predicted wind direction. Yaw Decision Module: The yaw decision module is configured to construct a multi-step prediction model based on a preset target yaw prediction sequence; the predicted wind direction is used as the input to the multi-step prediction model to output a yaw prediction error, which is configured as the yaw error of the yaw system predicted within the step size. A multi-objective optimization function is constructed based on the weighting coefficients and the target yaw prediction sequence; the predicted yaw action is calculated based on the multi-objective optimization function and the yaw prediction error; and the predicted yaw action is executed through the yaw actuator.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the yaw control method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the yaw control method according to any one of claims 1 to 7.

Images