Real-time dynamic adjustment control method and system for fan operation based on environmental data

Abstract

The present invention discloses a real-time dynamic adjustment control method and system for the operation of a wind turbine based on environmental data, including: collecting multi-dimensional dynamic parameters of the environment where the wind turbine is located in real time based on a distributed sensor network; constructing a dynamic coupling prediction model, inputting the multi-dimensional dynamic parameters into a two-way verification module, and obtaining an environmental state prediction matrix within a future time window; obtaining a joint adjustment instruction set for the pitch angle, yaw angle, and generator torque of the wind turbine based on the environmental state prediction matrix, where the instruction set includes main strategy parameters and an emergency tolerance interval; based on an incremental learning mechanism, dynamically updating the model parameters according to the deviation between the actual environmental parameters and the predicted values, and adaptively adjusting the update period according to the change in turbulence intensity. The advantages of the present invention are as follows: realizing accurate prediction and optimal control of the wind turbine in a complex environment, breaking through the limitation of the lack of physical interpretability of traditional data-driven models, and improving the stability and efficiency of the wind turbine operation.

Description

Technical Field

[0001] The present invention relates to dynamic adjustment technology, and particularly to a real-time dynamic adjustment control method and system for the operation of a wind turbine based on environmental data. Background Art

[0002] With the rapid development of renewable energy, wind power generation, as a clean energy source, has been widely used globally. In a wind power generation system, the operating efficiency and stability of a wind turbine are often affected by environmental factors, such as changes in meteorological conditions like wind speed, wind direction, temperature, and air pressure. Therefore, how to dynamically adjust the wind turbine according to real-time environmental data to improve the operating efficiency of the wind turbine, extend the equipment life, and ensure the stability and safety of power generation has become an important research direction.

[0003] The current real-time dynamic adjustment control methods for the operation of wind turbines on the market mainly focus on using an intelligent control system to monitor and adjust the operating state of the wind turbine in real time to achieve efficient and stable operation. Common adjustment control methods include data collection based on sensors, and the operating conditions of the wind turbine are monitored in real time through key parameters such as temperature, pressure, vibration, and flow rate, so as to accurately adjust the rotational speed, load, air volume, etc. of the wind turbine. However, only focusing on the wind turbine itself still has limitations, lacking dynamic response to real-time environmental data and being unable to adapt to environmental changes such as wind speed, temperature, and humidity in real time. Under complex environmental conditions, traditional methods may lead to low efficiency of the wind turbine, energy consumption waste, or premature wear of the equipment. Summary of the Invention

[0004] In order to improve the existing real-time dynamic adjustment control method and system for the operation of a wind turbine, a real-time dynamic adjustment control method and system for the operation of a wind turbine based on environmental data are provided. This method combines a distributed sensor network with a dynamic coupling prediction model to achieve precise prediction and optimal control of the wind turbine in a complex environment. By dynamically adjusting the model parameters through an incremental learning mechanism, the stability and efficiency of the wind turbine operation are improved, especially suitable for variable environmental conditions.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A real-time dynamic adjustment control method for the operation of a wind turbine based on environmental data, comprising:

[0007] Real-time collecting multi-dimensional dynamic parameters of the environment where the wind turbine is located based on a distributed sensor network;

[0008] Constructing a dynamic coupling prediction model, and inputting the multi-dimensional dynamic parameters into a two-way verification module based on the multi-dimensional dynamic parameters to obtain an environmental state prediction matrix within a future time window;

[0009] Based on the environmental state prediction matrix, obtain a joint adjustment instruction set for the wind turbine pitch angle, yaw angle, and generator torque. The instruction set includes main strategy parameters and an emergency tolerance interval;

[0010] Based on the incremental learning mechanism, dynamically update the model parameters according to the deviation between the actual environmental parameters and the predicted values. The update period is adaptively adjusted according to the turbulence intensity.

[0011] Preferably, the real-time acquisition of multi-dimensional dynamic parameters of the environment where the wind turbine is located based on the distributed sensor network specifically includes:

[0012] The multi-dimensional dynamic parameters specifically include wind speed gradient distribution, turbulence intensity, vertical temperature gradient, humidity change rate, and air pressure fluctuation data;

[0013] The wind speed gradient distribution is obtained by synchronously scanning with three lidars whose vertical height difference exceeds 20% of the hub diameter. The second lidar captures the characteristics of turbulent vortices at a 45-degree elevation angle, and the sampling frequency is dynamically adjusted according to the standard deviation of the instantaneous wind speed. When the standard deviation exceeds 1.5 m / s, the high-frequency scanning mode is activated.

[0014] Preferably, for the construction of the dynamic coupling prediction model, based on the multi-dimensional dynamic parameters, input them into the two-way verification module, and combine the LSTM neural network with the blade aerodynamic elastic coupling equation and the tower resonance frequency constraint to obtain the environmental state prediction matrix within the future time window, specifically including:

[0015] Based on the obtained multi-dimensional dynamic parameters, construct a dynamic coupling prediction model through a neural network, and input the multi-dimensional dynamic parameters into the two-way verification module;

[0016] The two-way verification module specifically includes an LSTM neural network and a physical constraint equation;

[0017] The physical constraint equation includes the blade aerodynamic elastic coupling equation and the tower resonance frequency constraint condition. The formula is:

[0018]

[0019] Where, is the aerodynamic load, is the pitch angle, is the structural damping coefficient, is the lateral displacement of the tower, is the time-varying material fatigue threshold, and this constraint condition is embedded in the LSTM loss function through the Lagrange multiplier method;

[0020] Based on the environmental state prediction data within the future time window given by the dynamic coupling prediction model and the actual environmental state data , obtain an accurate environmental state prediction matrix within a future time window, and the formula is:

[0021]

[0022] Among them, C is a calibration coefficient, which is used to balance the influence of predicted values and actual data.

[0023] Preferably, based on the environmental state prediction matrix, obtain a joint adjustment instruction set for the wind turbine pitch angle, yaw angle, and generator torque. The instruction set includes main strategy parameters and emergency tolerance intervals, specifically including:

[0024] Obtain the environmental state prediction matrix The pitch angle of the wind turbine in , the yaw angle , the generator torque data;

[0025] The pitch angle of the wind turbine controls the angle of the wind turbine blades to maximize the capture of wind energy or reduce excessive load, and its adjustment function is:

[0026]

[0027] The yaw angle controls the direction of the wind turbine relative to the wind to align the wind turbine with the wind direction, and its adjustment formula is:

[0028]

[0029] The generator torque mainly controls the power output of the wind turbine, and its adjustment formula is:

[0030]

[0031] Among them, is the wind speed data with dimension w within the future time window , is the temperature data within the future time window , is the turbulence intensity data within the future time window , is the wind direction data with dimension w within the future time window ;

[0032] Generate control instructions based on each adjustment function, and combine the main strategy parameters to construct a joint adjustment instruction set ,

[0033]

[0034] Among them, , , are weight coefficients, which are adjusted based on the actual situation;

[0035] Based on the pitch angle of the wind turbine , yaw angle , and generator torque , an emergency tolerance interval is set.

[0036] Preferably, the upper and lower limit ranges of the pitch angle of the wind turbine , yaw angle , and generator torque are used to set the emergency tolerance interval, and the emergency tolerance interval generates predicted values through Monte Carlo simulation for extreme working conditions within the range, and setting the strain rate threshold specifically includes:

[0037] Generated through Monte Carlo simulation, simulating extreme working conditions with environmental parameters within the predicted value range;

[0038] When it is detected in real time that the strain rate at the root of the blade exceeds 0.15 με / s, it automatically switches to the load reduction priority mode and triggers the emergency preloading action of the hydraulic pitch system.

[0039] Preferably, based on the incremental learning mechanism, the model parameters are dynamically updated according to the deviation between the actual environmental parameters and the predicted values, and the update period is adaptively adjusted according to the change of turbulence intensity, specifically including:

[0040] Obtain the real-time environmental parameter values (t), predicted value (t), and calculate their error value:

[0041]

[0042] Update the model parameters based on the gradient descent method, and the formula is:

[0043]

[0044] where is the learning rate, is the gradient of the error;

[0045] Based on the value of the turbulence intensity, obtain its influence on the update period, and adjust the next update period, and the formula is:

[0046]

[0047] where is the adjustment coefficient, controlling the influence of the turbulence intensity on the update period, is time turbulence intensity data at this time

[0048] Furthermore, a real-time dynamic regulation and control system for the operation of a wind turbine based on environmental data is proposed, including:

[0049] Distributed sensors: The real-time distributed sensors are mainly used to obtain multi-dimensional dynamic parameters, including wind speed sensors, temperature sensors, humidity sensors, pressure sensors, lidar;

[0050] Prediction model module: The prediction model module is mainly used to obtain the environmental state prediction matrix within a future time window through the two-way verification module;

[0051] Two-way verification module: The two-way verification module is mainly used for the operation and processing of the LSTM neural network and physical constraint equations;

[0052] Adjustment function module: The adjustment function module is mainly used for the pitch angle of the wind turbine under different environmental state conditions , yaw angle , generator torque data adjustment;

[0053] Joint adjustment instruction set module: The joint adjustment instruction set is mainly used to coordinate and optimize multiple control variables to improve the performance and stability of the system, and at the same time set an emergency tolerance interval to ensure the operation of the wind turbine;

[0054] Parameter optimization module: The parameter optimization module is mainly used to dynamically update the model parameters according to the deviation between the actual environmental parameters and the predicted values;

[0055] Processor: The processor is mainly used for the calculation process of each formula and the construction and calculation process of each model.

[0056] Compared with the prior art, the advantages of the present invention are:

[0057] Based on the distributed sensor network to collect environmental data in real time, and through the dynamic coupling prediction model to predict the state of the wind turbine, it can effectively capture the real-time impact of environmental changes on the operation of the wind turbine and achieve a more accurate control strategy. Predict the future environmental state through the two-way verification module, and adjust the pitch angle, yaw angle and generator torque of the wind turbine based on the prediction matrix, which can optimize the operation efficiency of the wind turbine under different environmental conditions and ensure its safety at the same time. The incremental learning mechanism dynamically adjusts the model parameters according to the deviation between the actual environmental parameters and the predicted values, enabling the control system to adapt to environmental changes, especially being more flexible and accurate when dealing with sudden changes such as turbulence. This regulation and control method based on real-time dynamic data greatly improves the adaptability and operation stability of the wind turbine, reduces the losses caused by environmental fluctuations, improves the wind energy utilization efficiency, and reduces the failure risk. Description of the Drawings

[0058] Figure 1 Schematic diagram of real-time dynamic adjustment control of fan operation based on environmental data proposed by the present invention;

[0059] Figure 2 Schematic diagram of multi-dimensional dynamic parameter acquisition proposed by the present invention;

[0060] Figure 3 Schematic diagram of environmental state prediction matrix proposed by the present invention;

[0061] Figure 4 Schematic diagram of joint adjustment instruction set proposed by the present invention;

[0062] Figure 5 Schematic diagram of emergency tolerance interval proposed by the present invention;

[0063] Figure 6 Schematic diagram of dynamically updating model parameters proposed by the present invention;

[0064] Figure 7 Architecture diagram of the electronic device in this solution;

[0065] Figure 8 Schematic diagram of the structure of the computer-readable storage medium in this solution. Detailed implementation

[0066] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations.

[0067] The real-time dynamic adjustment control system for fan operation based on environmental data includes:

[0068] Distributed sensors: The real-time distributed sensors are mainly used to obtain multi-dimensional dynamic parameters, including wind speed sensors, temperature sensors, humidity sensors, barometric pressure sensors, lidar;

[0069] Prediction model module: The prediction model module is mainly used to obtain the environmental state prediction matrix within a future time window through the two-way verification module;

[0070] Two-way verification module: The two-way verification module is mainly used for the operation and processing of the LSTM neural network and physical constraint equations;

[0071] Adjustment function module: The adjustment function module is mainly used for the pitch angle of the fan under different environmental state conditions , yaw angle , generator torque Data adjustment;

[0072] Joint Regulation Instruction Set Module: The joint regulation instruction set is mainly used to coordinate and optimize multiple control variables to improve the performance and stability of the system, and at the same time set an emergency tolerance interval to ensure the operation of the fan;

[0073] Parameter Optimization Module: The parameter optimization module is mainly used to dynamically update the model parameters according to the deviation between the actual environmental parameters and the predicted values;

[0074] Processor: The processor is mainly used for the calculation process of each formula and the construction and calculation process of each model.

[0075] Refer to Figure 1 As shown, the real-time dynamic regulation and control method for the operation of a wind turbine based on environmental data includes:

[0076] Step 1: Real-time collect multi-dimensional dynamic parameters of the environment where the wind turbine is located based on a distributed sensor network;

[0077] Step 2: Construct a dynamic coupling prediction model, input it into a two-way verification module based on the multi-dimensional dynamic parameters, and obtain the environmental state prediction matrix within the future time window;

[0078] Step 3: Based on the environmental state prediction matrix, obtain the joint regulation instruction set for the pitch angle, yaw angle and generator torque of the wind turbine, and the instruction set includes main strategy parameters and an emergency tolerance interval;

[0079] Step 4: Based on the incremental learning mechanism, dynamically update the model parameters according to the deviation between the actual environmental parameters and the predicted values, and the update period is adaptively adjusted according to the change of turbulence intensity.

[0080] Refer to Figure 2 As shown, the real-time collection of multi-dimensional dynamic parameters of the environment where the wind turbine is located based on a distributed sensor network specifically includes:

[0081] The multi-dimensional dynamic parameters specifically include wind speed gradient distribution, turbulence intensity, vertical temperature gradient, humidity change rate, and air pressure fluctuation data;

[0082] The wind speed gradient distribution is obtained by synchronous scanning of three groups of lidars with a vertical height difference exceeding 20% of the hub diameter. Among them, the second group of radars captures the characteristics of turbulent vortices at a 45-degree elevation angle, and the sampling frequency is dynamically adjusted according to the standard deviation of the instantaneous wind speed. When the standard deviation exceeds 1.5 m / s, the high-frequency scanning mode is started.

[0083] Refer to Figure 3 As shown, constructing a dynamic coupling prediction model, inputting it into a two-way verification module based on multi-dimensional dynamic parameters, and obtaining the environmental state prediction matrix within the future time window specifically includes:

[0084] Based on the acquired multi-dimensional dynamic parameters, a dynamic coupling prediction model is constructed through a neural network, and the multi-dimensional dynamic parameters are input into the bidirectional verification module;

[0085] The bidirectional verification module specifically includes an LSTM neural network and a physical constraint equation;

[0086] The physical constraint equation includes a blade aeroelastic coupling equation and a tower resonance frequency constraint condition, and the formula is:

[0087]

[0088] Among them, is the aerodynamic load, is the pitch angle, is the structural damping coefficient, is the lateral displacement of the tower, is the time-varying material fatigue threshold, and this constraint condition is embedded in the LSTM loss function through the Lagrange multiplier method;

[0089] Based on the environmental state prediction data within the future time window given by the dynamic coupling prediction model and the actual environmental state data an accurate environmental state prediction matrix within the future time window is obtained, and the formula is:

[0090]

[0091] Among them, C is the verification coefficient, which is used to balance the influence of predicted values and actual data.

[0092] Specifically, the physical constraint equation transforms the dynamic response characteristics of the flexible body described by the blade aeroelastic coupling equation and the inherent vibration mode law expressed by the tower resonance frequency constraint condition into a regularization term in the form of a differential equation and embeds it in the training process. This bidirectional verification mechanism synchronously optimizes the network weights and physical parameters by comparing the predicted data with the actual monitoring data in real time using error backpropagation, which not only ensures that the prediction results conform to the basic principles of fluid mechanics and structural dynamics but also can adaptively correct the model deviation. The finally output environmental state prediction matrix not only includes meteorological elements such as wind speed, wind direction, and turbulence within the future time window but also integrates key parameters such as the blade dynamic load spectrum and the tower vibration frequency spectrum to form a multi-dimensional spatio-temporal correlation feature tensor, providing high-precision prediction support for the active pitch control and resonance avoidance strategy of wind turbines.

[0093] Refer to Figure 4 As shown, based on the environmental state prediction matrix, a joint adjustment instruction set for the pitch angle, yaw angle, and generator torque of the wind turbine is obtained. The instruction set includes main strategy parameters and an emergency tolerance interval, specifically including:

[0094] Obtain the environmental status prediction matrix for the pitch angle of the wind turbine , yaw angle , generator torque data;

[0095] The pitch angle of the wind turbine controls the angle of the wind turbine blades to maximize the capture of wind energy or reduce excessive loads, and its adjustment function is:

[0096]

[0097] The yaw angle controls the direction of the wind turbine relative to the wind direction to align the wind turbine with the wind direction, and its adjustment formula is:

[0098]

[0099] The generator torque mainly controls the power output of the wind turbine, and its adjustment formula is:

[0100]

[0101] where, is the wind speed data with dimension w within the future time window , is the temperature data within the future time window , is the turbulence intensity data within the future time window , is the wind direction data with dimension w within the future time window ;

[0102] Generate control instructions based on each adjustment function, and combine with the main strategy parameters to construct a joint adjustment instruction set ,

[0103]

[0104] where, , , are weight coefficients, which are adjusted based on the actual situation;

[0105] Based on the upper and lower limit ranges of the pitch angle of the wind turbine, yaw angle , and generator torque , set the emergency tolerance interval.

[0106] It is understandable that when constructing the joint regulation instruction set, the main policy parameters may not be consistent with the requirements of each regulation function, resulting in conflicts or inability to coordinate among the control instruction sets. For example, the regulation of generator torque may conflict with the regulation of the wind turbine pitch angle, thus affecting the effect of the overall control strategy. Therefore, when constructing the joint regulation instruction set, an optimization algorithm should be used for parameter matching to ensure the coordinated operation among each regulation function. At the same time, by introducing constraint conditions, it can be ensured that each control instruction does not interfere with each other, achieving the optimal overall regulation effect.

[0107] Refer to Figure 5 As shown, based on the upper and lower limit ranges of the wind turbine pitch angle , yaw angle , and generator torque , setting the emergency tolerance interval specifically includes:

[0108] Generated through Monte Carlo simulation, simulating extreme working conditions within the predicted value range of environmental parameters;

[0109] When the strain rate at the root of the blade is monitored in real time and exceeds 0.15 με / s, automatically switch to the load reduction priority mode and trigger the emergency preloading action of the hydraulic pitch system.

[0110] Specifically, in the generation of Monte Carlo simulation, a certain range of variables are set, such as the upper and lower limits of wind speed, and the fluctuation range of other environmental parameters. Run the simulation model, generate different environmental conditions through random sampling, so as to simulate various possible extreme working conditions. Evaluate the performance of the wind turbine under these extreme working conditions based on the simulation results, and identify potential risk factors. When the system monitors in real time that the strain rate at the root of the blade exceeds 0.15 με / s, it indicates that the wind turbine may be in an overloaded or unstable operating state. At this time, the system will automatically switch to the load reduction priority mode, reduce the wind turbine load, reduce the pressure on mechanical components, and prevent equipment damage. Immediately afterwards, the hydraulic pitch system will be activated to perform an emergency preloading operation, adjust the windward angle of the blade, reduce the wind pressure on the blade, thereby reducing the stress on the blade and maintaining the stable operation of the wind turbine.

[0111] Refer to Figure 6 As shown, based on the incremental learning mechanism, dynamically update the model parameters according to the deviation between the actual environmental parameters and the predicted values, and the update period is adaptively adjusted according to the turbulence intensity. Specifically includes:

[0112] Obtain the real-time environmental parameter value (t), predicted value (t), and calculate their error value:

[0113]

[0114] Update the model parameters based on the gradient descent method The formula is as follows:

[0115]

[0116] where is the learning rate, is the gradient of the error;

[0117] Based on the value of the turbulence intensity, obtain its influence on the update period, and adjust the next update period. The formula is:

[0118]

[0119] where is the adjustment coefficient, which controls the influence of the turbulence intensity on the update period, is the time and the turbulence intensity data at this time.

[0120] Furthermore, the method according to the embodiment of the present application can also be implemented with the aid of Figure 7 the architecture of the electronic device shown. As Figure 7 shown, the electronic device 500 may include a bus 501, one or more CPUs 502, a read-only memory (ROM) 503, a random access memory (RAM) 504, a communication port 505 connected to the network, an input / output component 506, a hard disk 507, etc. The storage device in the electronic device 500, such as the ROM 503 or the hard disk 507, may store the method and system for real-time dynamic adjustment and control of the fan operation based on environmental data provided by the present application. The electronic device 500 may also include a terminal interface 508. Of course, Figure 7 the architecture shown is only exemplary. When implementing different devices, one or more components shown in the electronic device may be omitted according to actual needs. Figure 7

[0121] Figure 8 is a schematic diagram of the structure of a computer-readable storage medium provided by an embodiment of the present application. As Figure 8 shown, it is a computer-readable storage medium 600 according to an embodiment of the present application. Computer-readable instructions are stored on the computer-readable storage medium 600. When the computer-readable instructions are run by a processor, the method and system for real-time dynamic adjustment and control of the fan operation based on environmental data according to the embodiment of the present application described with reference to the above drawings can be executed. The storage medium 600 includes but is not limited to, for example, volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and cache memory, etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. ​

[0122] It should be noted that: the above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. Also, the above description of specific embodiments of this specification has been made. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0123] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

[0124] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A real-time dynamic adjustment control method for the operation of a wind turbine based on environmental data, characterized in that, including: Real-time collecting multi-dimensional dynamic parameters of the environment where the wind turbine is located based on a distributed sensor network; Constructing a dynamic coupling prediction model, inputting the multi-dimensional dynamic parameters into a two-way verification module based on the multi-dimensional dynamic parameters, combining the LSTM neural network with the blade aeroelastic coupling equation and the tower resonance frequency constraint to obtain an environmental state prediction matrix within a future time window; Based on the environmental state prediction matrix, obtain a joint adjustment instruction set for the wind turbine pitch angle, yaw angle, and generator torque. The instruction set includes main policy parameters and an emergency tolerance interval, and the emergency tolerance interval generates predicted values through Monte Carlo simulation extreme operating conditions within the range, and set the strain rate threshold; Based on an incremental learning mechanism, dynamically updating the model parameters according to the deviation between the actual environmental parameters and the predicted values, and adaptively adjusting the update period according to the change of the turbulence intensity.

2. The real-time dynamic adjustment control method for fan operation based on environmental data according to claim 1, wherein The real-time collection of multi-dimensional dynamic parameters of the environment where the wind turbine is located based on the distributed sensor network specifically includes: The multi-dimensional dynamic parameters specifically include wind speed gradient distribution, turbulence intensity, temperature vertical gradient, humidity change rate, and air pressure fluctuation data; The wind speed gradient distribution is obtained by synchronous scanning of three groups of lidars with a vertical height difference exceeding 20% of the hub diameter. Among them, the second group of radars captures the characteristics of turbulent vortices at a 45-degree elevation angle, and the sampling frequency is dynamically adjusted according to the standard deviation of the instantaneous wind speed. When the standard deviation exceeds 1.5 m / s, a high-frequency scanning mode is started.

3. The real-time dynamic adjustment control method for fan operation based on environmental data according to claim 1, characterized in that, The construction of the dynamic coupling prediction model, inputting the multi-dimensional dynamic parameters into the two-way verification module, combining the LSTM neural network with the blade aeroelastic coupling equation and the tower resonance frequency constraint to obtain the environmental state prediction matrix within a future time window specifically includes: Based on the obtained multi-dimensional dynamic parameters, constructing a dynamic coupling prediction model through a neural network and inputting the multi-dimensional dynamic parameters into the two-way verification module; The two-way verification module specifically includes an LSTM neural network and a physical constraint equation; The physical constraint equation includes the blade aeroelastic coupling equation and the tower resonance frequency constraint condition, and the formula is: ； Among them, is the pneumatic load, is the pitch angle, is the structural damping coefficient, is the lateral displacement of the tower barrel, is the time-varying material fatigue threshold, and this constraint condition is embedded in the LSTM loss function through the Lagrange multiplier method, where t is the time within the time window; Environmental state prediction data within the future time window given by the dynamic coupling prediction model and the actual environmental state data to obtain an accurate environmental state prediction matrix within the future time window. The formula is as follows: ​ ； where C is a verification coefficient used to balance the influence of the predicted value and the actual data.

4. The real-time dynamic adjustment control method for the operation of a wind turbine based on environmental data according to claim 1, characterized in that, Based on the environmental state prediction matrix, obtaining a joint adjustment instruction set for the pitch angle, yaw angle, and generator torque of the wind turbine, and the instruction set includes main strategy parameters and an emergency tolerance interval, specifically including: Obtain the environmental state prediction matrix of the fan pitch angle , yaw angle , generator torque data; The pitch angle of the wind turbine controls the angle of the wind turbine blades to maximize the capture of wind energy or reduce excessive loads, and its adjustment function is as follows: ； The yaw angle controls the direction of the wind turbine relative to the wind to align the wind turbine with the wind direction, and its adjustment formula is: ； The generator torque is used to control the power output of the wind turbine, and its adjustment formula is: ； wherein, is the wind speed data with dimension w within the future time window , is the temperature data within the future time window , is the turbulence intensity data within the future time window , is the wind direction data with dimension w within the future time window . Generate control instructions based on each adjustment function, and combine with the main policy parameters to construct a joint adjustment instruction set , ； Among them, , , are weight coefficients, which are adjusted based on the actual situation; Based on the pitch angle of the fan , the yaw angle , and the upper and lower limit ranges of the generator torque , an emergency tolerance interval is set.

5. The real-time dynamic adjustment control method for fan operation based on environmental data according to claim 4, characterized in that Based on the pitch angle of the fan , yaw angle , generator torque The upper and lower limit ranges are used to set an emergency tolerance interval, and the emergency tolerance interval generates predicted values through Monte Carlo simulation The extreme working conditions within the range, and setting the strain rate threshold specifically includes: Generated by Monte Carlo simulation, simulating extreme working conditions within the predicted values of the simulation environment parameters range; When it is detected in real time that the strain rate at the root of the blade exceeds 0.15 με / s, automatically switch to the load reduction priority mode and trigger the emergency preloading action of the hydraulic pitch system.

6. The real-time dynamic adjustment and control method for the operation of a wind turbine based on environmental data according to claim 1, characterized in that The dynamic update of the model parameters according to the deviation between the actual environmental parameters and the predicted values based on the incremental learning mechanism, and the adaptive adjustment of the update period according to the change of the turbulence intensity specifically includes: Obtain real-time environmental parameter values (t), predicted value (t), and calculate its error value: ； Update the model parameters based on the gradient descent method The formula is as follows: ； Among them, is the learning rate, is the gradient of the error; Based on the value of the turbulence intensity, obtaining its influence on the update period and adjusting the next update period, and the formula is: ； Among them, is the adjustment coefficient, which controls the influence of the turbulence intensity on the update period, is time and is the turbulence intensity data at that time.

7. A real-time dynamic adjustment control method for fan operation in combination with environmental data is used to implement the real-time dynamic adjustment control system for fan operation based on environmental data as described in any one of claims 1-6, characterized in that, including: Distributed sensors: The real-time distributed sensors are mainly used to obtain multi-dimensional dynamic parameters, including wind speed sensors, temperature sensors, humidity sensors, air pressure sensors, and lidars; Prediction model module: The prediction model module is mainly used to obtain an environmental state prediction matrix within a future time window through the two-way verification module; Two-way verification module: The two-way verification module is mainly used for the operation and processing of the LSTM neural network and the physical constraint equation; Adjustment function module: The adjustment function module is mainly used to adjust the pitch angle of the fan , yaw angle , generator torque data; Joint adjustment instruction set module: The joint adjustment instruction set is mainly used to coordinate and optimize multiple control variables to improve the performance and stability of the system, and at the same time set an emergency tolerance interval to ensure the operation of the wind turbine; Parameter optimization module: The parameter optimization module is mainly used to dynamically update the model parameters according to the deviation between the actual environmental parameters and the predicted values; Processor: The processor is mainly used for the calculation process of each formula and the construction calculation process of each model.

8. An electronic device, characterized in that, Comprising: At least one processor; And a memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the real-time dynamic adjustment control method for the operation of the wind turbine based on environmental data as described in any one of claims 1-6.

9. A computer-readable storage medium storing computer-readable instructions, characterized in that, When the computer-readable instructions are executed by the processor, the real-time dynamic adjustment control method for the operation of the wind turbine based on environmental data as described in any one of claims 1-6 is implemented.

Images