Variable-pitch dynamic adjustment method and related device

Abstract

The invention belongs to the technical field of wind power generation, and particularly relates to a variable pitch dynamic adjusting method and a related device. The method comprises the following steps: acquiring the rotating speed of a generator and the characteristic data of the wind generating set; the generator speed serves as a target variable; carrying out correlation calculation on the target variable and the feature data, and reserving the feature data with relatively high correlation with the target variable as input features; performing data standardization on the input features to obtain standard data; inputting the input features into a pre-constructed LSTM neural network model, and outputting a pitch angle adjustment amount based on a set generator rotation speed target value; and generating a variable pitch control signal by adopting a PID control algorithm, and sending the variable pitch control signal to a variable pitch system of the wind generating set to adjust the pitch angle. The problem that the variable pitch cannot be accurately adjusted in the prior art is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power generation, and particularly relates to a variable pitch dynamic adjustment method and related devices. Background Art

[0002] With the global energy structure transforming towards low-carbon, wind energy, as a core form of clean energy, faces numerous challenges in terms of power generation efficiency and stability. Wind turbines need to achieve maximum wind energy capture under complex meteorological conditions while balancing power output and mechanical loads. This poses higher requirements for the dynamic regulation of the variable pitch system, especially in the environment of low wind speed resources (3 - 8 m / s).

[0003] Although existing variable pitch technologies have evolved from active pitch, variable speed constant frequency technologies to incorporating algorithms such as deep learning and model predictive control (MPC), they still face the following problems: The monitoring of the wind speed profile is limited by the sensor resolution (e.g., the error of ultrasonic anemometry > 5%), which affects the rolling optimization effect of the MPC model; There is a conflict between the goal of maximizing power generation efficiency and minimizing mechanical loads, and it is necessary to develop multi-objective optimization algorithms to balance the two; The durability of the variable pitch motor and energy storage module under extreme working conditions still needs to be improved. For example, in a low-temperature environment of -30°C, the variable pitch mechanism may experience a response delay.

[0004] In summary, the existing technology has the problem that the variable pitch cannot be accurately adjusted. Summary of the Invention

[0005] The purpose of the present invention is to provide a variable pitch dynamic adjustment method and related devices, which solve the problem that the variable pitch in the existing technology cannot be accurately adjusted.

[0006] The present invention is realized through the following technical solutions: The present invention discloses a variable pitch dynamic adjustment method, including the following steps: S1. Collect the generator speed and characteristic data of the wind turbine; the generator speed is used as the target variable; S2. Calculate the correlation between the target variable and the characteristic data, and retain the characteristic data with a higher correlation with the target variable as the input features; S3. Normalize the input features to obtain the standard data; S4. Input the input features into a pre-constructed LSTM neural network model, and based on the set target value of the generator speed, output the pitch angle adjustment amount; S5. Use the PID control algorithm to generate a variable pitch control signal, and send the variable pitch control signal to the variable pitch system of the wind turbine to adjust the pitch angle.

[0007] Further, in S1, the characteristic data includes wind speed, wind direction, blade pitch angle, and generator output power.

[0008] Further, S2 is specifically as follows: Calculate the Pearson correlation coefficient between the target variable and the characteristic data. If the Pearson correlation coefficient is greater than the preset requirement, retain the characteristic data as the input feature.

[0009] Further, S3 is specifically as follows: Perform Z-score standardization on the input features to make the mean of the data 0 and the standard deviation 1; The standardization formula is: X' = (X - μ) / σ; Where, X is the original data of the input feature, μ is the data mean, σ is the data standard deviation; X' is the standardized data.

[0010] Further, in S4, the construction process of the LSTM neural network model is as follows: Collect historical data of the wind turbine generator set. The historical data includes generator speed and characteristic data of the wind turbine generator set; the characteristic data includes wind speed, wind direction, blade pitch angle, power; the generator speed is used as the target variable; Perform correlation calculation on the target variable and the characteristic data, and retain the characteristic data with high correlation with the target variable as the input feature; Perform data standardization on the input features to obtain standard data; Divide the standard data into a training set and a validation set; Pre-establish an LSTM neural network with 3 LSTM layers. The number of neurons in each layer is 128, 64, and 32 respectively; the dimension of the input layer of the network is the number of features, and the dimension of the output layer is 1; Input the training set into the established LSTM neural network to train the LSTM neural network. If the loss value of the validation set does not decrease for several consecutive times, stop training and save the current best model to obtain the pre-constructed LSTM neural network model.

[0011] Further, in S4, the PID control algorithm is used to generate the pitch control signal, specifically as follows: Take the pitch angle adjustment amount output by the LSTM neural network model as the set value of the PID controller, and the actually collected pitch angle as the feedback value; The PID controller calculates the error between the set value and the feedback value, and then processes the error according to the proportional link, integral link, and differential link to generate the pitch control signal.

[0012] The present invention also discloses a pitch dynamic adjustment system, including: A data acquisition module for acquiring the generator speed and the characteristic data of the wind turbine generator set; the characteristic data includes wind speed, wind direction, pitch angle, and power; the generator speed is used as the target variable; A correlation calculation module for performing correlation calculation on the target variable and the characteristic data, and retaining the characteristic data with high correlation with the target variable as the input features; A data preprocessing module for performing data standardization on the input features to obtain standard data; A pitch angle calculation module for inputting the input features into a pre-constructed LSTM neural network model, and outputting a pitch angle adjustment amount based on a set target value of the generator speed; An adjustment module for generating a pitch control signal using a PID control algorithm, sending the pitch control signal to the pitch system of the wind turbine generator set, and adjusting the pitch angle.

[0013] The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the pitch angle dynamic adjustment method are implemented.

[0014] The present invention also discloses a computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the steps of the pitch angle dynamic adjustment method are implemented.

[0015] The present invention also discloses a computer program product, including a computer program / instructions, and when the computer program / instructions are executed by a processor, the steps of the pitch angle dynamic adjustment method are implemented.

[0016] Compared with the prior art, the present invention has the following beneficial technical effects: The present invention discloses a method for dynamically adjusting the pitch angle of a variable pitch propeller. In steps S1 and S2, by collecting characteristic data and calculating its correlation with the target variable of the generator speed, the data that has a great influence on the generator speed is screened out as input features, which can reduce the interference of invalid information caused by large sensor errors, improve the quality of wind speed data input into the subsequent model, and further enhance the optimization effect of the LSTM neural network model; in step S3, the input features are standardized to eliminate the dimensional differences between different characteristic data, enabling the model to treat each feature more fairly. Taking the generator speed as the target variable, the pitch angle adjustment amount is output based on the set target value of the generator speed through the LSTM neural network model. In a low wind speed environment, by reasonably setting the target value of the generator speed, the power generation efficiency and mechanical load can be balanced to a certain extent. The LSTM neural network model has a certain prediction ability and can predict the pitch angle adjustment amount based on historical data and current input features; before the arrival of extreme working conditions (such as a low temperature environment of -30°C), the model can predict the change in wind conditions and adjust the pitch angle in advance, which can reduce the working intensity of the pitch motor and energy storage module under extreme conditions, reduce the risk of hardware damage caused by problems such as response delay, and extend the service life of the hardware. Finally, the PID control algorithm is used to generate a pitch control signal, which can correct the deviation in the process of pitch angle adjustment in real time, further optimize the pitch angle adjustment strategy, and find a more suitable balance point between power generation efficiency and mechanical load. The stable control of the PID control algorithm can make the pitch angle adjustment smoother and reduce the impact on the pitch motor and energy storage module. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flowchart of a method for dynamically adjusting the pitch angle of a variable pitch propeller according to the present invention; Figure 2 is a block diagram of a system for dynamically adjusting the pitch angle of a variable pitch propeller according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer, the following further detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments.

[0019] The detailed description of the embodiments of the present invention provided in the following drawings is not intended to limit the scope of the claimed present invention, but merely represents a selected embodiment of the present invention. Based on the drawings and embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present invention.

[0020] It should be noted that: The term "comprising", "including" or any other variant is intended to cover non-exclusive inclusion, such that a process, element, method, article or device comprising a series of elements does not include only those elements but also other elements not expressly listed, or elements inherent to the process, element, method, article or device.

[0021] The features and performance of the present invention will be further described in detail below in conjunction with embodiments.

[0022] Embodiment 1 The construction process of the LSTM neural network is described from a design perspective. It mainly includes the following processes: 1. Data collection and preprocessing 1) Data collection: Install sensors on the wind turbine generator set to collect data such as wind speed, wind direction, pitch angle, generator speed, and generator output power in real time. The data collection frequency should be determined according to actual needs, generally 10Hz - 100Hz.

[0023] 2) Data preprocessing: Feature selection: Calculate the Pearson correlation coefficient between each feature and the target variable (such as generator speed), remove features with low correlation, and retain features with high correlation with the target variable. For example, if the correlation coefficient between wind speed and generator speed is greater than 0.8, then retain wind speed as an input feature.

[0024] Data standardization: Perform Z-score standardization on the input features so that the mean of the data is 0 and the standard deviation is 1.

[0025] The standardization formula is: X'=(X - μ) / σ; Where X is the original data, μ is the data mean, σ is the data standard deviation, and X' is the standardized data.

[0026] Data division: Divide the data into a training set and a validation set in a ratio of 7:3. The training set is used for model training, and the validation set is used for model performance evaluation.

[0027] 2. Construction and training of the LSTM neural network 1) Network structure design: Design a neural network containing 3 LSTM layers, with the number of neurons in each layer being 128, 64, and 32 respectively. The dimension of the input layer of the network is the number of features, and the dimension of the output layer is 1 (the predicted pitch angle adjustment amount).

[0028] 2) Parameter initialization: Use the He initialization method to initialize the weights of the network, and initialize the biases to 0. The He initialization method can effectively alleviate the problem of gradient disappearance in the neural network and accelerate the convergence of the model.

[0029] 3) Model training: Use the training set data to train the LSTM neural network. During the training process, the Adam optimization algorithm is adopted, with a learning rate of 0.001 and a loss function of mean squared error (MSE). The number of training epochs is set to 100, and the batch size is set to 32.

[0030] 4) Model optimization: During the training process, record the loss value of the validation set every 10 epochs. If the loss value of the validation set does not decrease for 5 consecutive times, stop the training and save the current best model. By this method, overfitting of the model is avoided and the generalization ability of the model is improved.

[0031] 3. Control strategy design Goal setting: Suppose the rated value of the generator speed is set to 1500 r / min, and the control goal is to maintain the generator speed near the rated value, with a fluctuation range not exceeding ±50 r / min. At the same time, it is required that the mechanical stress does not exceed 80% of the design value.

[0032] Control signal generation: According to the output of the LSTM neural network (pitch angle adjustment amount), use the PID control algorithm to generate the pitch control signal. The parameters of the PID controller (proportional coefficient Kp, integral coefficient Ki, derivative coefficient Kd) are adjusted by the trial and error method to achieve the best control effect.

[0033] 4. Simulation and testing Simulation platform construction: Build a model of the wind power generation system on the simulation platform, including a wind speed model, a pitch system model, etc. The wind speed model adopts a wind speed generation method based on time series, which can simulate the wind speed changes under low wind speed and complex wind conditions.

[0034] Simulation verification: Integrate the trained LSTM neural network control algorithm into the simulation model for simulation testing. The simulation time is 1000 seconds, and the sampling frequency is 10 Hz. Through the simulation testing, observe the changes of key indicators such as generator speed and mechanical stress to verify the effectiveness of the control algorithm.

[0035] Performance evaluation: Compare with the traditional PID control method and evaluate the performance of the LSTM neural network control algorithm in improving system stability, reducing mechanical stress, etc. The performance evaluation indicators include the generator speed volatility, the magnitude of mechanical stress, etc. For example, the calculation formula of the generator speed volatility is: volatility = (maximum speed - minimum speed) / rated speed × 100%.

[0036] Example 2 As Figure 1 shown, the present invention discloses a method for dynamically adjusting the pitch of a wind turbine blade, which includes the following steps: S1. Collect the generator speed and characteristic data of the wind turbine generator set; the characteristic data includes wind speed, wind direction, pitch angle, and power; the generator speed is used as the target variable. S2. Calculate the correlation between the target variable and the characteristic data, and retain the characteristic data with high correlation with the target variable as the input features. S3. Standardize the input features to obtain the standard data. S4. Input the input features into the pre-constructed LSTM neural network model, and based on the set target value of the generator speed, output the pitch angle adjustment amount. S5. Use the PID control algorithm to generate a pitch control signal, and send the pitch control signal to the pitch system of the wind turbine generator set to adjust the pitch angle.

[0037] Aiming at the limitations of the existing model predictive control pitch, deep learning pitch is adopted, and through the LSTM neural network control algorithm, the efficiency improvement amplitude and the load rate are optimized and reduced; it can adapt to low wind speed and complex wind speed scenarios, and improve the operation efficiency and stability of the wind turbine generator set under low wind speed conditions; through the multi-objective optimization algorithm, the contradiction between power generation efficiency and mechanical load is effectively balanced.

[0038] Embodiment 3 The present invention provides a pitch control method for a wind turbine generator set based on an LSTM neural network, including the following steps: Step 1. Data collection and preprocessing.

[0039] Install sensors on the wind turbine generator set to collect data such as wind speed, wind direction, pitch angle, generator speed, and generator output power in real time, and the data collection frequency is 50 Hz. Perform feature selection on the collected data, calculate the Pearson correlation coefficient between each feature and the target variable (generator speed), and retain the features with higher correlation as the input features.

[0040] If the correlation coefficients between the wind speed, wind direction and the generator speed are both greater than 0.7, then retain the wind speed and wind direction as the input features. Standardize the data so that the mean of the data is 0 and the standard deviation is 1. The standardization formula is: X'=(X - μ) / σ, where X is the original data, μ is the data mean, and σ is the data standard deviation. Divide the standardized data into a training set and a validation set according to a ratio of 7:3.

[0041] Step 2. Construct and train the LSTM neural network model.

[0042] Design a neural network with 3 LSTM layers, where the number of neurons in each layer is 256, 128, and 64 respectively. The input layer dimension is the selected number of features, and the output layer dimension is 1 (the predicted pitch angle adjustment). Initialize the network weights using the He initialization method and the biases to 0. Train the LSTM neural network using the training set data, with the Adam optimization algorithm, a learning rate of 0.0005, the mean squared error as the loss function, 150 training epochs, and a batch size of 64. Record the loss value of the validation set every 20 epochs. If the validation set loss value does not decrease for 10 consecutive times, stop training and save the current best model.

[0043] Step 3: Determine whether the model performance meets the requirements. If so, go to Step 4; if not, return to Step 2 for model optimization.

[0044] Step 4: Design a control strategy based on the trained LSTM neural network model and generate a pitch control command.

[0045] Set the rated value of the generator speed to 1500 r / min. The control objective is to maintain the generator speed within the range of 1475 - 1525 r / min, while requiring that the mechanical stress does not exceed 75% of the design value. Based on the output of the LSTM neural network (pitch angle adjustment), use the PID control algorithm to generate a pitch control signal. The proportional coefficient Kp of the PID controller is 1.0, the integral coefficient Ki is 0.3, and the derivative coefficient Kd is 0.2. Send the pitch control command to the pitch system of the wind turbine generator set to achieve the adjustment of the pitch angle.

[0046] The LSTM neural network can process data with sequential properties. In wind power generation, as time goes by, factors such as wind speed and wind direction are dynamically changing, with obvious time series characteristics. Through its special gating mechanism, LSTM can effectively capture the information of these input features at different time steps and their complex relationships with the generator speed.

[0047] In the pre - constructed LSTM neural network model, when the filtered feature data is input, the model will learn the mapping relationship between the input features and the generator speed based on the existing training data. Based on the set target value of the generator speed, the model continuously adjusts the network parameters through the backpropagation algorithm, so that the output pitch angle adjustment can make the generator speed as close as possible to the target value.

[0048] The PID control algorithm is a classic feedback control algorithm. It adjusts the control signal according to the error between the actual output and the target output of the system, enabling the system to stably reach the target state. In pitch control, the pitch angle adjustment amount output by the LSTM neural network model is used as the set value of the PID controller, while the actual pitch angle is the feedback value.

[0049] The PID controller calculates the error between the set value and the feedback value, and then processes the error through three links: proportional (P), integral (I), and derivative (D) to generate a pitch control signal. The proportional link generates a corresponding control action according to the magnitude of the current error. The integral link is used to eliminate the steady-state error of the system, and the derivative link can predict the change trend of the error and perform control in advance to improve the response speed and stability of the system. Finally, the generated pitch control signal is sent to the pitch system of the wind turbine generator to achieve precise adjustment of the pitch angle, so that the generator speed is stabilized near the target value.

[0050] Example 4 As Figure 2 shown, the present invention discloses a pitch dynamic adjustment system, including: A data acquisition module for collecting the generator speed and the characteristic data of the wind turbine generator; the characteristic data includes wind speed, wind direction, pitch angle, and power; the generator speed is used as the target variable; A correlation calculation module for performing correlation calculation on the target variable and the characteristic data, and retaining the characteristic data with high correlation with the target variable as the input features; A data preprocessing module for normalizing the input features to obtain standard data; A pitch angle calculation module for inputting the input features into a pre-constructed LSTM neural network model and outputting a pitch angle adjustment amount based on the set target value of the generator speed; An adjustment module for generating a pitch control signal by using the PID control algorithm, sending the pitch control signal to the pitch system of the wind turbine generator, and realizing the adjustment of the pitch angle.

[0051] Example 5 The present invention also discloses a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the variable pitch dynamic adjustment method are implemented. Among them, the memory may include a memory, such as a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk memory, etc.; the processor, network interface, and memory are interconnected through an internal bus, and this internal bus can be an Industry Standard Architecture bus, a Peripheral Component Interconnect standard bus, an Extended Industry Standard Architecture bus, etc., and the bus can be divided into an address bus, a data bus, a control bus, etc. The memory is used to store programs. Specifically, the program can include program code, and the program code includes computer operation instructions. The memory can include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0052] Embodiment 6 The present invention also discloses a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps of the variable pitch dynamic adjustment method are implemented. Specifically, the computer-readable storage medium includes, but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory can include a random access memory and / or a cache memory, etc. The non-volatile memory can include a read-only memory, a hard disk, a flash memory, an optical disc, a magnetic disk, etc.

[0053] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, optical memories, etc.) containing computer-usable program code.

[0054] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the specified functions in one Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0055] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to operate in a particular manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one or more processes and / or blocks Figure 1 of the process or processes and / or boxes Figure 1 or boxes specified.

[0056] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one or more processes and / or blocks Figure 1 of the process or processes and / or boxes Figure 1 or boxes specified.

[0057] 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 them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent substitutions can still be made to the specific embodiments of the present invention. Any modification or equivalent substitution that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A variable pitch dynamic adjustment method, characterized in that It includes the following steps: S1. Collect the generator speed and characteristic data of the wind turbine generator set; the generator speed is used as the target variable; S2. Calculate the correlation between the target variable and the characteristic data, and retain the characteristic data with a relatively high correlation with the target variable as the input features; S3. Standardize the input features to obtain the standard data; S4. Input the input features into the pre-constructed LSTM neural network model, and based on the set target value of the generator speed, output the pitch angle adjustment amount; S5. Use the PID control algorithm to generate a pitch control signal, and send the pitch control signal to the pitch system of the wind turbine generator set to adjust the pitch angle.

2. The variable pitch dynamic adjustment method according to claim 1, characterized in that, In S1, the characteristic data includes wind speed, wind direction, pitch angle, and generator output power.

3. The variable pitch dynamic adjustment method according to claim 1, wherein S2 specifically is: Calculate the Pearson correlation coefficient between the target variable and the characteristic data. If the Pearson correlation coefficient is greater than the preset requirement, retain the characteristic data as the input features.

4. A variable pitch dynamic adjustment method according to claim 1, characterized in that S3 specifically is: Perform Z-score standardization processing on the input features to make the mean of the data 0 and the standard deviation 1; The standardization formula is: X'=(X - μ) / σ; where X is the original data of the input features, μ is the data mean, σ is the data standard deviation; X' is the standardized data.

5. A variable pitch dynamic adjustment method according to claim 1, characterized in that In S4, the construction process of the LSTM neural network model is: Collect the historical data of the wind turbine generator set. The historical data includes the generator speed and the characteristic data of the wind turbine generator set; the characteristic data includes wind speed, wind direction, pitch angle, power; the generator speed is used as the target variable; Calculate the correlation between the target variable and the characteristic data, and retain the characteristic data with a high correlation with the target variable as the input features; Standardize the input features to obtain the standard data; Divide the standard data into a training set and a validation set; Pre-establish an LSTM neural network containing 3 LSTM layers, and the number of neurons in each layer is 128, 64, and 32 respectively; the dimension of the input layer of the network is the number of features, and the dimension of the output layer is 1; Input the training set into the established LSTM neural network to train the LSTM neural network. If the loss value of the validation set does not decrease continuously for multiple times, stop training and save the current best model to obtain the pre-constructed LSTM neural network model.

6. The variable pitch dynamic adjustment method according to claim 1, wherein, In S4, using the PID control algorithm to generate a pitch control signal specifically is: Use the pitch angle adjustment amount output by the LSTM neural network model as the set value of the PID controller, and the actually collected pitch angle as the feedback value; The PID controller calculates the error between the set value and the feedback value, and then processes the error according to the proportional link, integral link, and differential link to generate a pitch control signal.

7. A variable pitch dynamic adjustment system, characterized in that It includes: A data acquisition module for collecting the generator speed and characteristic data of the wind turbine generator set; the characteristic data includes wind speed, wind direction, pitch angle, power; the generator speed is used as the target variable; A correlation calculation module for calculating the correlation between the target variable and the characteristic data, and retaining the characteristic data with a high correlation with the target variable as the input features; A data preprocessing module for standardizing the input features to obtain the standard data; The pitch angle calculation module is used to input the input features into a pre-constructed LSTM neural network model, and based on the set target value of the generator speed, output the pitch angle adjustment amount; The adjustment module is used to generate a pitch control signal by using a PID control algorithm, and send the pitch control signal to the pitch system of the wind turbine generator set to adjust the pitch angle.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the pitch angle dynamic adjustment method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the pitch angle dynamic adjustment method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, it implements the steps of the pitch angle dynamic adjustment method according to any one of claims 1 to 6.

Images