Yaw predictive control method, system, medium, and apparatus for a wind turbine generator system

Abstract

The application relates to a yaw prediction control method, system, medium and equipment of a wind turbine generator set, and comprises the following steps: constructing a Y-direction dynamic characteristic twin model of the wind turbine generator set; constructing an X-direction dynamic characteristic twin model of the wind turbine generator set; integrating the X-direction dynamic characteristic twin model and the Y-direction dynamic characteristic twin model to obtain a limit dynamic characteristic model; obtaining current state data and limit parameters of the wind turbine generator set; performing prediction according to the current state data, the limit parameters and the limit dynamic characteristic model to obtain a limit wind speed of the wind turbine generator set; and performing yaw control on the wind turbine generator set according to the limit wind speed and an actual wind speed of the wind turbine generator set. Through the application, the maximum wind condition limit that can be borne by the wind turbine generator set in a future time domain can be predicted, and yaw control action in the future time domain is determined, so that safe operation of the unit under the condition of minimized power loss is ensured.

Description

Technical Field

[0001] This invention relates to the technical field of wind power generation, and more specifically, to a yaw prediction control method, system, medium, and device for wind turbine generator sets. Background Technology

[0002] With the trend towards larger wind turbine generators, the working height of the nacelle is constantly increasing, making the safety of the nacelle load and yaw process a critical concern for the safe operation of wind turbine generators. In particular, the influence of factors such as turbulence, wake, and wind shear can cause drastic changes in the nacelle load, rapidly approaching and exceeding the ultimate load value, leading to damage to the unit or even tower fracture, resulting in significant losses.

[0003] Currently, conservative strategies are typically employed for controlling the yaw limit load of generator units. For example, yaw control strategies are designed considering the worst-case operating conditions, or control strategies are designed based on engineering experience. These conservative strategies can lead to untimely control action, causing the nacelle load to rapidly approach the load limit value, thereby triggering safety accidents and resulting in power generation losses. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a yaw prediction control method, system, medium and equipment for wind turbine generator sets, addressing the problems existing in the prior art.

[0005] The technical solution adopted by this invention to solve its technical problem is: to construct a yaw prediction and control method for wind turbine generators, comprising the following steps:

[0006] Step S10: Construct a twin model of the dynamic characteristics of the wind turbine generator in the Y direction;

[0007] Step S20: Construct a twin model of the dynamic characteristics of the wind turbine generator in the X direction;

[0008] Step S30: Integrate the X-direction dynamic characteristic twin model and the Y-direction dynamic characteristic twin model to obtain the ultimate dynamic characteristic model;

[0009] Step S40: Obtain the current state data and limit parameters of the wind turbine generator set;

[0010] Step S50: Based on the current state data, the limiting parameters, and the limiting dynamic characteristic model, predict the limiting wind speed of the wind turbine generator set;

[0011] Step S60: Perform yaw control on the wind turbine generator set based on the extreme wind speed and the actual wind speed of the wind turbine generator set.

[0012] In the yaw prediction control method for wind turbine generators described in this invention, step S10, which involves constructing a twin model of the dynamic characteristics of the wind turbine generator in the Y direction, includes:

[0013] A deep neural network model is used to model the dynamic characteristics of the wind turbine generator in the Y direction during yaw, and a twin model of the dynamic characteristics in the Y direction is obtained.

[0014] In the yaw prediction and control method for wind turbine generators described in this invention, the deep neural network model is a combination of a convolutional neural network model, a long short-term memory model, and an attention model.

[0015] The process of modeling the dynamic characteristics of the wind turbine generator in the Y direction during yaw using a deep neural network model to obtain a twin model of the dynamic characteristics in the Y direction includes:

[0016] Obtain data on factors influencing the yaw process of the wind turbine generator set;

[0017] The influencing factor data is preprocessed to obtain preprocessed data;

[0018] Based on the preprocessed data, construct combined input data for multiple influencing factors;

[0019] The convolutional neural network model is used to extract the local nonlinear coupling features of the combined input data;

[0020] The long short-term memory model is used to obtain the trend and mutation characteristics of the combined input data;

[0021] The attention model is used to process the local nonlinear coupling features, the trend features, and the mutation features to obtain the feature data of the influencing factors.

[0022] Modeling is performed based on the characteristic data of the influencing factors to obtain the twin model of the dynamic characteristics in the Y direction.

[0023] In the yaw prediction control method for wind turbine generators described in this invention, step S20, which involves constructing a twin model of the dynamic characteristics of the wind turbine generator in the X direction, includes:

[0024] A physical information neural network model is used to model the dynamic characteristics of the wind turbine generator in the X direction during yaw, and a twin model of the dynamic characteristics in the X direction is obtained.

[0025] In the yaw prediction and control method for wind turbine generator sets according to the present invention, the step of using a physical information neural network model to model the dynamic characteristics of the wind turbine generator set in the X direction during yaw, and obtaining a twin model of the dynamic characteristics in the X direction, includes:

[0026] Acquire the inlet wind speed of the wind turbine generator set and the status data of the wind turbine generator set in the X direction;

[0027] The inlet wind speed and the state data in the X direction are preprocessed to obtain preprocessed data;

[0028] The preprocessed data is input into the physical information neural network model for training and iterative analysis to obtain the X-direction dynamic characteristic twin model.

[0029] In the yaw prediction and control method for wind turbine generator sets according to the present invention, step S60, which involves yaw control of the wind turbine generator set based on the extreme wind speed and the actual wind speed of the wind turbine generator set, includes:

[0030] The wind speed deviation value is obtained by calculating the deviation between the extreme wind speed and the actual wind speed.

[0031] Based on the wind speed deviation value, determine whether the wind turbine generator is operating within a safe area;

[0032] If so, then maintain the current control logic;

[0033] If not, then yaw control is performed on the wind turbine generator set based on the wind speed deviation value.

[0034] In the yaw prediction control method for wind turbine generator sets according to the present invention, the yaw control of the wind turbine generator set based on the wind speed deviation value includes:

[0035] The avoidance signal is determined based on the wind speed deviation value;

[0036] The yaw angle of the wind turbine is adjusted based on the avoidance signal.

[0037] In the yaw prediction control method for wind turbine generator sets according to the present invention, the yaw control of the wind turbine generator set based on the wind speed deviation value further includes:

[0038] After adjusting the yaw angle of the wind turbine generator, the following steps are performed:

[0039] Determine the prediction time domain;

[0040] Obtain the rated power output of the wind turbine generator set;

[0041] Based on the rated power generation, the predicted time domain, and the optimization objective, constraints are constructed.

[0042] The avoidance signal is optimized according to the constraints to obtain the yaw control signal in the predicted time domain.

[0043] The present invention also provides a yaw prediction and control system for a wind turbine generator set, comprising:

[0044] Y-direction model building unit, used to build a twin model of the dynamic characteristics of wind turbine generators in the Y direction;

[0045] The X-direction model building unit is used to build a twin model of the dynamic characteristics of the wind turbine generator in the X direction.

[0046] The limit model construction unit is used to integrate the X-direction dynamic characteristic twin model and the Y-direction dynamic characteristic twin model to obtain the limit dynamic characteristic model.

[0047] The data acquisition unit is used to acquire the current status data and limit parameters of the wind turbine generator set;

[0048] The extreme wind speed prediction unit is used to predict the extreme wind speed of the wind turbine generator based on the current state data, the extreme parameters and the extreme dynamic characteristic model.

[0049] A yaw control unit is used to perform yaw control on the wind turbine generator set based on the extreme wind speed and the actual wind speed of the wind turbine generator set.

[0050] In the yaw prediction control system of the wind turbine generator described in this invention, the twin model of dynamic characteristics in the Y direction includes:

[0051] The multi-influencing factor data input module is used to input combined data from multiple influencing factors.

[0052] A coupling feature extraction module is used to extract local nonlinear coupling features of the combined input data using the convolutional neural network model.

[0053] The long short-term feature extraction module is used to obtain the trend features and abrupt change features of the combined input data using the long short-term memory model.

[0054] The key feature extraction module is used to process the local nonlinear coupling features, the trend features, and the mutation features through the attention model to obtain feature data of the influencing factors;

[0055] The model output module is used to perform modeling based on the feature data of the influencing factors to obtain the twin model of the dynamic characteristics in the Y direction.

[0056] In the yaw prediction control system of the wind turbine generator set according to the present invention, the yaw control unit includes:

[0057] The deviation calculation module is used to calculate the deviation based on the extreme wind speed and the actual wind speed to obtain the wind speed deviation value.

[0058] The safe operation judgment module is used to determine whether the wind turbine generator is operating within a safe area based on the wind speed deviation value.

[0059] The yaw control module is used to maintain the current control logic when the wind turbine is operating within a safe area; and to perform yaw control on the wind turbine based on the wind speed deviation value when the wind turbine is operating within a safe area.

[0060] The yaw prediction control system for wind turbine generators described in this invention further includes:

[0061] The yaw optimization control unit is used to construct constraints based on the rated power generation of the wind turbine generator set, the predicted time domain, and the optimization objective, and to optimize the avoidance signal based on the constraints to obtain the yaw control signal in the predicted time domain.

[0062] The present invention also provides a storage medium storing a computer program adapted for loading by a processor to execute the steps of the yaw prediction control method for wind turbine generators as described above.

[0063] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps of the yaw prediction control method for a wind turbine generator as described above by calling the computer program stored in the memory.

[0064] The yaw prediction control method, system, medium, and equipment for wind turbine generators implemented according to the present invention have the following beneficial effects: They include: constructing a twin model of the wind turbine generator's dynamic characteristics in the Y direction; constructing a twin model of the wind turbine generator's dynamic characteristics in the X direction; integrating the X-direction and Y-direction dynamic characteristic twin models to obtain a limiting dynamic characteristic model; acquiring the current state data and limiting parameters of the wind turbine generator; predicting the limiting wind speed of the wind turbine generator based on the current state data, limiting parameters, and limiting dynamic characteristic model; and performing yaw control on the wind turbine generator based on the limiting wind speed and the actual wind speed of the wind turbine generator. Through this invention, the maximum tolerable wind condition limit of the wind turbine generator in the future time domain can be predicted, and the yaw control action in the future time domain can be determined, ensuring the safe operation of the unit with minimal power loss. Attached Figure Description

[0065] The present invention will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0066] Figure 1 This is a flowchart illustrating the yaw prediction control method for wind turbine generators provided in an embodiment of the present invention.

[0067] Figure 2 This is a schematic diagram illustrating the yaw limit dynamic characteristic analysis of a wind turbine generator system provided in an embodiment of the present invention;

[0068] Figure 3 This is a construction logic diagram of the Y-direction dynamic characteristic twin model provided in this embodiment of the invention;

[0069] Figure 4 This is a construction logic diagram of the X-direction dynamic characteristic twin model provided in the embodiments of the present invention;

[0070] Figure 5 This is a schematic diagram of wind speed limit analysis using a limiting dynamic characteristic model provided in an embodiment of the present invention;

[0071] Figure 6 This is a logic block diagram of the yaw prediction control system for a wind turbine generator provided in an embodiment of the present invention;

[0072] Figure 7 This is a schematic diagram of the wind turbine load safety prediction controller provided in an embodiment of the present invention. Detailed Implementation

[0073] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0074] To address the shortcomings of conservative strategies in controlling the yaw limit load of wind turbine generators, this invention, from a digital twin perspective, first establishes and integrates dynamic characteristic twin models of the wind turbine generator in the X and Y directions. Then, it predicts the limiting wind speed using the integrated model to obtain the maximum load limit that the wind turbine generator can withstand in the future time domain. Based on the predicted data, it determines the yaw control action in the future time domain to ensure safe operation of the unit with minimal power loss.

[0075] refer to Figure 1 , Figure 1 This invention illustrates a preferred embodiment of the navigation predictive control method for wind turbine generator sets provided by the present invention.

[0076] Specifically, such as Figure 1 As shown, the yaw predictive control method for this wind turbine generator includes the following steps:

[0077] Step S10: Construct a twin model of the dynamic characteristics of the wind turbine generator in the Y direction.

[0078] First, before constructing the digital twin models (i.e., the dynamic characteristic twin models in the Y and X directions), the limiting dynamic characteristics of the wind turbine system's yaw are analyzed. Specifically, the wind turbine system exhibits nonlinear dynamic characteristics, which can be expressed as follows:

[0079] (1);

[0080] (2);

[0081] In the formula It is a state equation; It is the output equation; This includes the yaw status of the wind turbine generator, such as yaw angle, yaw speed, and blade pitch angle. It is the inlet wind speed of the wind turbine generator set; It is the load of the wind turbine generator nacelle.

[0082] It is the limiting parameter with the following nonlinear equation:

[0083] (3);

[0084] The limiting parameters are driven by the states of all wind turbine generators and can be divided into X-direction states. and Y-direction state The state equation of a wind turbine generator can be written as follows:

[0085] (4);

[0086] (5);

[0087] Limiting parameters For a given input It can be determined by the X-direction state. and Y-direction state This indicates that the input is:

[0088] (6);

[0089] In order to obtain an online estimate of the dynamics of the limiting parameters, The rate of change is treated as a constant input. Therefore, we can obtain:

[0090] (7);

[0091] The limiting parameter dynamics result is then obtained as follows:

[0092] (8);

[0093] The process flow of the limiting dynamic characteristic model of the yaw of the wind turbine generator system is as follows: Figure 2 As shown. Therefore, based on equation (8), the dynamic characteristic twin model in the Y direction and the dynamic characteristic twin model in the X direction can be constructed.

[0094] Optionally, in this embodiment of the invention, step S10, constructing a twin model of the dynamic characteristics of the wind turbine generator in the Y direction, includes: using a deep neural network model to model the dynamic characteristics of the wind turbine generator in the Y direction during yaw, thereby obtaining a twin model of the dynamic characteristics in the Y direction. Preferably, the deep neural network model is a combination of a convolutional neural network (CNN), a long short-term memory (LSTM), and an attention model, i.e., a CNN-LSTM-Attention combination model. The process of modeling the dynamic characteristics of wind turbine generators in the Y-direction during yaw using a deep neural network model to obtain a twin model of the dynamic characteristics in the Y-direction includes: acquiring data on influencing factors during the yaw process of wind turbine generators; preprocessing the influencing factor data to obtain preprocessed data; constructing combined input data of multiple influencing factors based on the preprocessed data; extracting local nonlinear coupling features of the combined input data using a convolutional neural network model; acquiring trend and abrupt change features of the combined input data using a long short-term memory model; processing the local nonlinear coupling features, trend features, and abrupt change features using an attention model to obtain feature data of the influencing factors; and modeling based on the feature data of the influencing factors to obtain a twin model of the dynamic characteristics in the Y-direction.

[0095] Specifically, in this embodiment of the invention, the specific logic for constructing a twin model of the dynamic characteristics of a wind turbine generator in the Y-direction (or Y-direction) during yaw based on a CNN-LSTM-Attention combined model is as follows: Figure 3 As shown, the specific steps include the following:

[0096] Step a: First, obtain data on influencing factors such as yaw angle, yaw speed, and blade pitch angle;

[0097] Step b: Preprocess, segment, and normalize these influencing factor data to obtain preprocessed data;

[0098] Step c: Construct a data combination input with multiple influencing factors;

[0099] Step d: Use a CNN model to extract the local nonlinear coupling features of influencing factors such as yaw angle, yaw speed, and blade pitch angle;

[0100] Step e: Capture the trend characteristics (also known as long-term characteristics, i.e., overall change trend) and short-term characteristics (i.e., abrupt change characteristics) of influencing factors such as yaw angle, yaw speed, and blade pitch angle through the LSTM model.

[0101] Step f: Use the attention mechanism of the Attention model to highlight the feature data of these influencing factors (i.e., nonlinear, coupling, and abrupt change features).

[0102] Step g: Finally, accurately model the integral effect of the dynamic characteristics in the Y direction of the yaw process, that is, output the twin model of the dynamic characteristics in the Y direction.

[0103] Step S20: Construct a twin model of the dynamic characteristics of the wind turbine generator in the X direction.

[0104] Optionally, in this embodiment of the invention, step S20, constructing a twin model of the dynamic characteristics of the wind turbine generator in the X direction, includes: using a Physical-Informed Neural Network (PINN) model to model the dynamic characteristics of the wind turbine generator in the X direction during yaw, thereby obtaining a twin model of the dynamic characteristics of the X direction. Specifically, using a PINN model to model the dynamic characteristics of the wind turbine generator in the X direction during yaw, and obtaining the twin model of the dynamic characteristics of the X direction, includes: acquiring the inlet wind speed and the state data of the wind turbine generator in the X direction; preprocessing the inlet wind speed and the state data in the X direction to obtain preprocessed data; and inputting the preprocessed data into the PINN model for training and iterative analysis to obtain the twin model of the dynamic characteristics of the X direction.

[0105] Specifically, as can be seen from the aforementioned equation (5), the wind turbine generator set possesses dynamic characteristics in the X direction (which can also be described as X-direction) and exhibits nonlinearity. Therefore, this invention can utilize the Physical Information Neural Network (PINN) model to construct a twin model of the dynamic characteristics of the wind turbine generator set in the X direction. In the process of constructing the twin model of the dynamic characteristics in the X direction, the main input is the incoming wind speed. and state data in the X direction The neural network is trained and iteratively analyzed to determine its deviation from the partial differential physical laws governing wind turbine operation, achieving convergence. The final output is a twin model of the dynamic characteristics of the wind turbine in the X-direction. Its specific construction logic is as follows: Figure 4 As shown.

[0106] Step S30: Integrate the X-direction dynamic characteristic twin model and the Y-direction dynamic characteristic twin model to obtain the ultimate dynamic characteristic model.

[0107] Specifically, after constructing the X-direction dynamic characteristic twin model and the Y-direction dynamic characteristic twin model of the wind turbine generator in steps S10 and S20 respectively, the models are integrated to obtain the limiting dynamic characteristic model. The integration of the limiting dynamic characteristic model is as follows: Figure 5 As shown.

[0108] Step S40: Obtain the current state data and limit parameters of the wind turbine generator set.

[0109] The current status data of the wind turbine generator includes: vibration velocity and acceleration in the X direction, vibration velocity and acceleration in the Y direction, and the limiting parameter is thrust. .

[0110] Step S50: Based on the current state data, limiting parameters, and limiting dynamic characteristic model, predict the limiting wind speed of the wind turbine generator.

[0111] Specifically, in this step, after obtaining the limiting dynamic characteristic model, the current state data and limiting parameters (i.e., thrust) of the wind turbine generator can be used as a basis. Predictions are made to forecast the limiting wind speed of wind turbines (i.e., the maximum load limit that wind turbines can withstand in the future time domain), defined as... Among them, the limiting parameter thrust It can be determined in advance.

[0112] Step S60: Perform yaw control on the wind turbine generator set based on the extreme wind speed and the actual wind speed of the wind turbine generator set.

[0113] Optionally, in this embodiment of the invention, step S60, performing yaw control on the wind turbine based on the extreme wind speed and the actual wind speed of the wind turbine, includes: calculating the deviation based on the extreme wind speed and the actual wind speed to obtain a wind speed deviation value; determining whether the wind turbine is operating within a safe area based on the wind speed deviation value; if yes, maintaining the current control logic; if no, performing yaw control on the wind turbine based on the wind speed deviation value. Specifically, performing yaw control on the wind turbine based on the wind speed deviation value includes: determining an avoidance signal based on the wind speed deviation value; and adjusting the yaw angle of the wind turbine based on the avoidance signal.

[0114] Specifically, this invention detects excessive loads, i.e., excessive thrust, by comparing envelope wind velocities. The wind speed deviation can be calculated using the following formula:

[0115] (9);

[0116] In the formula, For extreme wind speeds, Actual wind speed. Maximum wind speed. The actual wind speed can be predicted through step S50. It can be obtained through measurements using wind speed sensors or lidar equipment.

[0117] The wind speed deviation value calculated using equation (9) can be used to determine whether the wind turbine is operating within a safe zone. That is, when the actual wind speed is less than or equal to the limiting wind speed (…), the wind turbine is operating within a safe zone. When the actual wind speed exceeds the limit wind speed, the wind turbine operates within the gap constraints, allowing it to operate under acceptable loads within a safe area. Conversely, when the actual wind speed exceeds the limit wind speed (…), the wind turbine operates within the gap constraints, allowing it to operate under acceptable loads within a safe area. When the wind turbine generator starts to operate under excessive load in an unsafe area, yaw control is required to ensure its safe operation.

[0118] In this invention, yaw control is based on the wind speed deviation value. A suitable avoidance signal is determined, and this signal is then used to adjust the yaw angle of the wind turbine generator to reduce operating load and ensure safe operation during yaw. The avoidance signal can be determined using the following formula:

[0119] (10);

[0120] In the formula, It serves as a reference setpoint for yaw control, used to avoid collision signals. These are design parameters chosen for effective avoidance. Among them, when the actual wind speed equals the estimated envelope wind speed (i.e., the limiting wind speed), it will cause the thrust response to trigger a predefined safety limit.

[0121] Furthermore, in this embodiment of the invention, yaw control of the wind turbine generator based on wind speed deviation further includes: after adjusting the yaw angle of the wind turbine generator, performing the following steps: determining the prediction time domain; obtaining the rated power output of the wind turbine generator; constructing constraints based on the rated power output, the prediction time domain, and the optimization objective; optimizing the avoidance signal according to the constraints to obtain the yaw control signal in the prediction time domain. The optimization objective is to minimize the deviation between the wind turbine generator's power output and the rated power output in the future time domain. The prediction time domain can be defined as T. The yaw control signal is defined as... .

[0122] To minimize power generation loss during the control process, this invention constructs a corresponding rolling time-domain optimization problem based on the aforementioned predictive control framework. Specifically, it involves: determining the prediction time domain T, defining the optimization objective (minimizing the deviation between the wind turbine's future power generation and its rated power), and constructing constraints based on the prediction time domain and the optimization objective. These constraints are discrete-time equations, as detailed below:

[0123] (11);

[0124] In the formula, It is air density. It refers to the diameter of the wind turbine rotor. It's the air intake speed. It is the power factor of the unit. It refers to the power generation capacity of the wind turbine generator set.

[0125] This invention optimizes the design parameters by using the discrete-time equation (11) with the goal of minimizing the deviation between the wind turbine generator's power output in the future time domain and its rated power output. This optimizes the yaw control action (i.e., the yaw control signal). This ensures safe yaw operation of the unit while minimizing power generation loss.

[0126] The yaw prediction control method for wind turbine generators of the present invention is a safety prediction control method driven by digital twins of the dynamic characteristics of wind turbine generators in the X-direction and Y-direction. It includes the construction of the X-direction dynamic characteristic twin model and the Y-dynamic characteristic twin model of the wind turbine generator, the online maximum wind condition calculation of the wind turbine generator, and the yaw safety prediction controller of the wind turbine generator.

[0127] The yaw prediction control method for wind turbine generator sets of the present invention ensures safe operation of the nacelle XY load while minimizing power loss under wind speed prediction conditions. This avoids safety accidents caused by untimely control action leading to the nacelle load rapidly approaching the load limit value. At the same time, it can also minimize power generation loss, significantly improving the economic efficiency of the generator set while ensuring its safety.

[0128] refer to Figure 6 The present invention also provides a yaw prediction and control system for wind turbine generator sets.

[0129] Specifically, such as Figure 6 As shown, the yaw prediction control system of this wind turbine generator includes:

[0130] Y-direction model building unit 601 is used to build a twin model of the dynamic characteristics of a wind turbine generator in the Y direction. Preferably, in this embodiment of the invention, the Y-direction dynamic characteristic twin model includes: a multi-influencing factor data input module (such as...) Figure 3 The input is used to input combined input data with multiple influencing factors; coupled feature extraction modules (such as...) Figure 3 CNN), used to extract local nonlinear coupling features of combined input data using a convolutional neural network model; long and short term feature extraction modules (such as... Figure 3LSTM), used to acquire trend and abrupt change features of combined input data using a long short-term memory model; key feature extraction module (such as... Figure 3 The attention mechanism is used to process local nonlinear coupling features, trend features, and abrupt change features through an attention model to obtain feature data of influencing factors; the model output module (such as...) Figure 3 The output is used to model based on the feature data of influencing factors to obtain a twin model of dynamic characteristics in the Y direction.

[0131] X-direction model building unit 602 is used to build a twin model of the dynamic characteristics of wind turbine generators in the X direction.

[0132] The limit model building unit 603 is used to integrate the X-direction dynamic characteristic twin model and the Y-direction dynamic characteristic twin model to obtain the limit dynamic characteristic model.

[0133] The data acquisition unit 604 is used to acquire the current status data and limit parameters of the wind turbine generator set.

[0134] The extreme wind speed prediction unit 605 is used to predict the extreme wind speed of the wind turbine generator based on the current state data, extreme parameters and extreme dynamic characteristic model.

[0135] The yaw control unit 606 is used to control the yaw of the wind turbine generator set based on the extreme wind speed and the actual wind speed of the wind turbine generator set.

[0136] Preferably, in this embodiment of the invention, the yaw control unit includes: a deviation calculation module, used to calculate the deviation based on the extreme wind speed and the actual wind speed to obtain a wind speed deviation value; a safe operation judgment module, used to determine whether the wind turbine is operating within a safe area based on the wind speed deviation value; and a yaw adjustment module, used to maintain the current control logic when the wind turbine is operating within a safe area, and to perform yaw control on the wind turbine based on the wind speed deviation value when the wind turbine is operating within a safe area.

[0137] Furthermore, in this embodiment of the invention, the yaw prediction control system of the wind turbine generator set further includes: a yaw optimization control unit 607, used to construct constraints based on the rated power output of the wind turbine generator set, the prediction time domain, and the optimization objective, and to optimize the avoidance signal according to the constraints to obtain a yaw control signal in the prediction time domain. The yaw control unit 606 and the yaw optimization control unit 607 can be configured as a yaw safety prediction controller for the wind turbine generator set, specifically as follows... Figure 7 As shown.

[0138] Specifically, the specific coordination and operation process between the various units in the yaw prediction control system of the wind turbine generator set can be referred to the yaw prediction control method of the wind turbine generator set mentioned above, and will not be repeated here.

[0139] Furthermore, an electronic device of the present invention includes a memory and a processor; the memory is used to store a computer program; the processor is used to execute the computer program to implement the yaw prediction control method for a wind turbine generator as described in any of the above embodiments. Specifically, according to embodiments of the present invention, the processes described above with reference to the flowchart can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowchart. In such embodiments, when the computer program is downloaded, installed, and executed by an electronic device, it performs the functions defined in the methods of the embodiments of the present invention. The electronic device in the present invention can be a terminal such as a laptop, desktop computer, tablet computer, or smartphone, or it can be a server.

[0140] Furthermore, one type of storage medium of the present invention stores a computer program thereon, which, when executed by a processor, implements the yaw prediction control method for a wind turbine generator set as described above. Specifically, it should be noted that the storage medium described above in the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein computer-readable program code is carried. The transmitted data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0141] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0142] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0143] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0144] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0145] The above embodiments are only for illustrating the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They do not limit the scope of protection of the present invention. All equivalent changes and modifications made within the scope of the claims of the present invention should fall within the scope of the claims of the present invention.

Claims

1. A yaw prediction and control method for a wind turbine generator set, characterized in that, Includes the following steps: step S10: Construct a twin model of the dynamic characteristics of the wind turbine generator in the Y direction; Step S20: Construct a twin model of the dynamic characteristics of the wind turbine generator in the X direction; Step S30: Integrate the X-direction dynamic characteristic twin model and the Y-direction dynamic characteristic twin model to obtain the ultimate dynamic characteristic model; Step S40: Obtain the current state data and limit parameters of the wind turbine generator set; Step S50: Based on the current state data, the limiting parameters, and the limiting dynamic characteristic model, predict the limiting wind speed of the wind turbine generator set; Step S60: Perform yaw control on the wind turbine generator set based on the extreme wind speed and the actual wind speed of the wind turbine generator set.

2. The yaw prediction control method for wind turbine generator sets according to claim 1, characterized in that, In step S10, constructing the Y-direction dynamic characteristic twin model of the wind turbine generator includes: A deep neural network model is used to model the dynamic characteristics of the wind turbine generator in the Y direction during yaw, and a twin model of the dynamic characteristics in the Y direction is obtained.

3. The yaw prediction control method for wind turbine generators according to claim 2, characterized in that, The deep neural network model is a combination of a convolutional neural network model, a long short-term memory model, and an attention model. The process of modeling the dynamic characteristics of the wind turbine generator in the Y direction during yaw using a deep neural network model to obtain a twin model of the dynamic characteristics in the Y direction includes: Obtain data on factors influencing the yaw process of the wind turbine generator set; The influencing factor data is preprocessed to obtain preprocessed data; Based on the preprocessed data, construct combined input data for multiple influencing factors; The convolutional neural network model is used to extract the local nonlinear coupling features of the combined input data; The long short-term memory model is used to obtain the trend and mutation characteristics of the combined input data; The attention model is used to process the local nonlinear coupling features, the trend features, and the mutation features to obtain the feature data of the influencing factors. Modeling is performed based on the characteristic data of the influencing factors to obtain the twin model of the dynamic characteristics in the Y direction.

4. The yaw prediction control method for wind turbine generator sets according to claim 1, characterized in that, In step S20, constructing the X-direction dynamic characteristic twin model of the wind turbine generator includes: A physical information neural network model is used to model the dynamic characteristics of the wind turbine generator in the X direction during yaw, and a twin model of the dynamic characteristics in the X direction is obtained.

5. The yaw prediction control method for wind turbine generator sets according to claim 4, characterized in that, The method of using a physical information neural network model to model the dynamic characteristics of the wind turbine generator in the X direction during yaw, and obtaining a twin model of the dynamic characteristics in the X direction, includes: Acquire the inlet wind speed of the wind turbine generator set and the status data of the wind turbine generator set in the X direction; The inlet wind speed and the state data in the X direction are preprocessed to obtain preprocessed data; The preprocessed data is input into the physical information neural network model for training and iterative analysis to obtain the X-direction dynamic characteristic twin model.

6. The yaw prediction control method for wind turbine generator sets according to claim 1, characterized in that, In step S60, the yaw control of the wind turbine generator set based on the extreme wind speed and the actual wind speed of the wind turbine generator set includes: The wind speed deviation value is obtained by calculating the deviation between the extreme wind speed and the actual wind speed. Based on the wind speed deviation value, determine whether the wind turbine generator is operating within a safe area; If so, then maintain the current control logic; If not, then yaw control is performed on the wind turbine generator set based on the wind speed deviation value.

7. The yaw prediction control method for wind turbine generators according to claim 6, characterized in that, The yaw control of the wind turbine generator based on the wind speed deviation value includes: The avoidance signal is determined based on the wind speed deviation value; The yaw angle of the wind turbine is adjusted based on the avoidance signal.

8. The yaw prediction control method for wind turbine generators according to claim 7, characterized in that, The yaw control of the wind turbine generator based on the wind speed deviation value also includes: After adjusting the yaw angle of the wind turbine generator, the following steps are performed: Determine the prediction time domain; Obtain the rated power output of the wind turbine generator set; Based on the rated power generation, the predicted time domain, and the optimization objective, constraints are constructed. The avoidance signal is optimized according to the constraints to obtain the yaw control signal in the predicted time domain.

9. A yaw prediction and control system for a wind turbine generator set, characterized in that, include: Y-direction model building unit, used to build a twin model of the dynamic characteristics of wind turbine generators in the Y direction; The X-direction model building unit is used to build a twin model of the dynamic characteristics of the wind turbine generator in the X direction. The limit model construction unit is used to integrate the X-direction dynamic characteristic twin model and the Y-direction dynamic characteristic twin model to obtain the limit dynamic characteristic model. The data acquisition unit is used to acquire the current status data and limit parameters of the wind turbine generator set; The extreme wind speed prediction unit is used to predict the extreme wind speed of the wind turbine generator based on the current state data, the extreme parameters and the extreme dynamic characteristic model. A yaw control unit is used to perform yaw control on the wind turbine generator set based on the extreme wind speed and the actual wind speed of the wind turbine generator set.

10. The yaw prediction control system for a wind turbine generator according to claim 9, characterized in that, The Y-direction dynamic characteristic twin model includes: The multi-influencing factor data input module is used to input combined data from multiple influencing factors. The coupling feature extraction module is used to extract local nonlinear coupling features of the combined input data using a convolutional neural network model. The long-short-term feature extraction module is used to obtain the trend features and abrupt change features of the combined input data using a long short-term memory model. The key feature extraction module is used to process the local nonlinear coupling features, the trend features, and the mutation features through an attention model to obtain feature data of influencing factors; The model output module is used to perform modeling based on the feature data of the influencing factors to obtain the twin model of the dynamic characteristics in the Y direction.

11. The yaw prediction control system for a wind turbine generator set according to claim 9, characterized in that, The yaw control unit includes: The deviation calculation module is used to calculate the deviation based on the extreme wind speed and the actual wind speed to obtain the wind speed deviation value. The safe operation judgment module is used to determine whether the wind turbine generator is operating within a safe area based on the wind speed deviation value. The yaw control module is used to maintain the current control logic when the wind turbine is operating within a safe area; and to perform yaw control on the wind turbine based on the wind speed deviation value when the wind turbine is operating within a safe area.

12. The yaw prediction control system for a wind turbine generator set according to claim 9, characterized in that, Also includes: The yaw optimization control unit is used to construct constraints based on the rated power generation of the wind turbine generator set, the predicted time domain, and the optimization objective, and to optimize the avoidance signal based on the constraints to obtain the yaw control signal in the predicted time domain.

13. A storage medium, characterized in that, The storage medium stores a computer program adapted for loading by a processor to perform the steps of the yaw prediction control method for a wind turbine generator as described in any one of claims 1 to 8.

14. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the steps of the yaw prediction control method for a wind turbine generator as described in any one of claims 1 to 8 by calling the computer program stored in the memory.

Images