Wind driven generator control method and system and server

Abstract

The invention provides a wind driven generator control method and system and a server, and relates to the field of wind driven generator safety control, and the method takes a clearance radar as a safety hard constraint, obviously reduces the tower sweeping risk, can maintain the control reliability under the conditions of rain, fog, clutter and the like, reduces the unnecessary power limitation and shutdown occurrence probability, and improves the safety of the wind driven generator. The misjudgment rate of the wind driven generator in the clearance control process is greatly reduced; in addition, according to the method, wind measurement load shedding and clearance safety can be coupled in the control process of the wind driven generator, feedforward control can be carried out under the working conditions of gust, shearing, yaw errors and the like, and the load of the wind driven generator is reduced.

Description

Technical Field

[0001] This invention relates to the field of wind turbine control, and in particular to a wind turbine control method, system and server. Background Technology

[0002] Current wind turbine blade sweep risk detection is mainly achieved through blade tip clearance monitoring devices. When the blade tip enters a threshold area, it triggers the wind turbine to execute protection strategies such as deceleration, pitch retraction, or shutdown. Specific solutions mainly involve: clearance detection and protection schemes based on lidar / camera / IMU, schemes based on lidar ranging to calculate clearance and transmit it to the main control unit for control, and methods for retrieving the true blade tip clearance value from three-line clearance radar ranging.

[0003] In the existing technology, wind-measuring lidar has been widely used for yaw / control optimization, thereby reducing additional load and increasing the power generation of wind turbines. However, its reliability decreases significantly under conditions such as rain and fog, resulting in a high misjudgment rate in the airspace control process of wind turbines under severe weather conditions. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a wind turbine control method, system, and server. This method uses airspace radar as a hard constraint for safety, significantly reducing the risk of tower sweeping, and can maintain control reliability under conditions such as rain, fog, and clutter, reducing unnecessary power limitations and the probability of downtime, and greatly reducing the misjudgment rate of wind turbines in the airspace control process. In addition, this method can couple wind measurement load reduction with airspace safety in the wind turbine control process, and can perform feedforward control under conditions such as gusts, shear, and yaw errors, reducing the load on the wind turbine.

[0005] In a first aspect, embodiments of the present invention provide a wind turbine control method, the method comprising: Acquire wind field data collected by wind-measuring lidar, clearance distance data collected by clearance radar, and operating status data of wind turbine generators; Based on wind farm data, clearance distance data, and operational status data, the clearance distance of wind turbine generators is predicted within a future time window to obtain the predicted clearance value. When an airspace risk is determined based on the airspace forecast, a control command is output to the wind turbine generator.

[0006] Optionally, the step of predicting the clearance distance of wind turbine generators within a future time window based on wind farm data, clearance distance data, and operational status data includes: The wind farm data, clearance distance data, and operational status data are aligned according to timestamps to generate a fused time-series dataset; The fused time-series dataset is input into a pre-built headroom prediction model, which learns the mapping relationship between wind field changes, unit response and headroom dynamic evolution, and outputs headroom prediction values.

[0007] Optionally, the net airspace prediction model is a deep learning model built based on a temporal neural network; the method further includes: Obtain wind field data, airspace data, and unit status data under historical operating conditions as training samples; Using the actual airspace distance at future moments as a supervision label, a temporal neural network is trained to learn a nonlinear mapping from historical time-series data to future airspace distance.

[0008] Optionally, the method further includes: Perform signal quality assessment on wind field data and / or clearance data, and generate corresponding quality assessment parameters; Based on the quality assessment parameters, the fusion weights of wind field data and / or airspace distance data in predicting airspace distance are dynamically determined.

[0009] Optionally, the method further includes: Obtain the prediction confidence level corresponding to the net airspace forecast value; When the predicted net airspace value is lower than the preset safe net airspace threshold and the prediction confidence level is higher than the preset confidence level threshold, it is determined that there is a net airspace risk.

[0010] Optionally, the method further includes: The airspace risk assessment strategy is dynamically adjusted based on the availability or reliability of data from wind-measuring lidar and airspace radar. When wind measurement lidar data is unavailable or its reliability is below the first threshold, switch to the first judgment mode that prioritizes airspace radar data. When the airspace radar data is unavailable or its reliability is below the second threshold, switch to the second judgment mode that prioritizes wind measurement lidar data. When data from both sensors is unavailable or the reliability of both is below the corresponding threshold, a preset power limit or shutdown command is output.

[0011] Optionally, when an airspace risk is determined to exist based on the airspace forecast value, the step of outputting control commands to the wind turbine generator includes: Obtain the airspace safety threshold corresponding to the airspace risk, and calculate the difference between the airspace safety threshold and the predicted airspace value; Based on the difference, determine the constraints corresponding to the control command, and obtain the control command when the structural load of the wind turbine generator is at its minimum value under the constraints. Control commands are sent to the wind turbine generator set.

[0012] Optionally, the control commands include at least one of yaw correction commands, pitch adjustment commands, torque adjustment commands, and power adjustment commands; Structural loads include at least one of the following: tower load, blade load, yaw system load, and drive train load.

[0013] In a second aspect, the present invention provides a wind turbine generator control system, the system comprising: Data acquisition module: used to acquire wind field data collected by wind lidar, clearance distance data collected by clearance radar, and operating status data of wind turbine generators; The clearance prediction module is used to predict the clearance distance of wind turbine generators within a future time window based on wind farm data, clearance distance data, and operating status data, and obtain the clearance prediction value. Control execution module: When it is determined that there is a risk to the airspace based on the airspace forecast value, the module outputs control commands to the wind turbine generator set.

[0014] Thirdly, embodiments of the present invention also provide a server, the server including a processor and a memory, the memory storing computer-executable instructions that can be executed by the processor, the processor executing the computer-executable instructions to implement the steps of the wind turbine control method provided in the first aspect.

[0015] This invention provides a wind turbine control method, system, and server. During the airspace control process for a wind turbine, the method first acquires wind field data collected by a wind-measuring lidar, airspace distance data collected by an airspace radar, and the operating status data of the wind turbine. Then, based on the wind field data, airspace distance data, and operating status data, it predicts the airspace distance of the wind turbine within a future time window, obtaining a predicted airspace value. When an airspace risk is determined based on the predicted airspace value, a control command is output to the wind turbine. This method uses the airspace radar as a safety hard constraint, significantly reducing the risk of tower sweeping and maintaining control reliability under conditions such as rain, fog, and clutter. It reduces unnecessary power limitations and the probability of downtime, significantly lowering the misjudgment rate of the wind turbine during airspace control. Furthermore, this method can couple wind-measuring load reduction with airspace safety during the wind turbine control process, enabling feedforward control under conditions such as gusts, shear, and yaw errors, reducing the load on the wind turbine.

[0016] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0017] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 A flowchart of a wind turbine control method provided in an embodiment of the present invention; Figure 2 A flowchart of another wind turbine control method provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a wind turbine control system provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a server provided in an embodiment of the present invention.

[0020] icon: 100 - Data Acquisition Module; 200 - Clearance Prediction Determination Module; 300 - Control Execution Module; 101 - Processor; 102 - Memory; 103 - Bus; 104 - Communication interface. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, 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.

[0022] To facilitate understanding of this embodiment, a wind turbine control method disclosed in this invention will first be introduced, such as... Figure 1 As shown, the method includes: Step S101: Acquire wind field data collected by wind-measuring lidar, clearance distance data collected by clearance radar, and operating status data of wind turbine generators.

[0023] First, we obtain three types of core data, as follows: Wind field data collected by wind-measuring lidar includes impeller equivalent incoming wind speed, wind shear parameters, gust amplitude and arrival time, yaw angle, turbulence intensity, etc., which are used to characterize the dynamic changes of the wind field and the future short-term incoming flow trend. Clearance distance data collected by clearance radar: The minimum clearance value between the blade tip and the tower is obtained through multi-beam ranging inversion, and the data reliability information is output simultaneously. Wind turbine generator operating status data: Key parameters such as rotor speed, blade pitch angle, nacelle yaw angle, power command, and torque command are obtained from the main control system of the unit to quantify the current operating conditions of the unit.

[0024] Step S102: Based on wind farm data, clearance distance data, and operating status data, predict the clearance distance of the wind turbine generator set within a future time window to obtain the clearance prediction value.

[0025] Based on the wind field data, airspace distance data, and operating status data obtained in step S101, and combined with the unit dynamics characteristics, blade flexure deformation patterns, and wind field interference factors, an airspace risk assessment model is constructed.

[0026] Specifically, for a preset future time window (adaptable to blade rotation period and wind field change rate), the model can predict the trend of the change in the clearance distance between the blade tip and the tower within the window through calculation, obtain the clearance prediction value (including the predicted minimum clearance value), and simultaneously output the clearance safety margin to predict tower sweep risk in advance.

[0027] If the reliability of any sensor data falls below a preset threshold, a multi-source mutual verification control degradation strategy will be triggered: dynamically adjusting data weights during data fusion and prediction, or switching to a conservative prediction mode based on historical data to ensure the reliability of prediction results under harsh conditions.

[0028] Step S103: When it is determined that there is a risk to the airspace based on the airspace forecast value, output control commands to the wind turbine generator set.

[0029] The predicted airspace value obtained in step S102 is compared with the preset safe airspace threshold. If the predicted airspace value is lower than the threshold or the safety margin is insufficient, it is determined that there is an airspace risk.

[0030] At this point, with airspace safety as a hard constraint and reducing the load on key components of the unit (tower, blades, transmission chain, yaw system) as the optimization goal, load-airspace coordinated control commands are generated, including yaw correction commands, pitch adjustment commands, torque adjustment commands, and power adjustment commands.

[0031] If the reliability of sensor data is abnormal, the command will be updated according to the control degradation strategy. For example, when the reliability of wind lidar is insufficient, the control of airspace data constraint will be used first; when the reliability of airspace radar is insufficient, the command will be generated based on wind field feedforward and safety margin verification; finally, the control command will be sent to the unit actuator to control the unit to enter the power limiting and load reduction state or the safe shutdown state under critical risk, so as to ensure airspace safety while taking into account power generation efficiency.

[0032] Optionally, the process of predicting the airspace clearance of wind turbine generators within a future time window based on wind farm data, clearance distance data, and operational status data is achieved through the following steps: First, wind field data, clearance distance data, and operational status data are aligned according to timestamps to generate a fused time-series dataset; The fused time-series dataset is then input into a pre-built headroom prediction model, which learns the mapping relationship between wind field changes, unit response and headroom dynamics, and outputs headroom prediction values.

[0033] After acquiring wind field data and clearance distance, the first step is to precisely align the timestamps of the wind field data collected by the wind-measuring lidar, the clearance distance data collected by the clearance radar, and the operating status data of the wind turbine generators. The synchronization error must satisfy the following relationship: ; in, To account for the timestamp discrepancies in data from different sensors; This is the maximum allowed synchronization error threshold for the system.

[0034] By unifying the time dimension of the three types of data and eliminating time-series deviations caused by inconsistencies in sensor sampling frequencies and data transmission delays, a fused time-series dataset is formed, encompassing wind field characteristics, real-time tip clearance, and dynamic operating parameters of the generating unit. This dataset fully maps the correlation between the unit's current operating status, external wind field disturbances, and real-time changes in clearance, providing a highly timely and consistent data foundation for subsequent forecasting.

[0035] This dataset contains the impeller equivalent incoming air velocity. Specifically, it is calculated by performing multi-point inversion on the wind field data in the space in front of the rotor obtained by the wind-measuring lidar. The formula used is as follows: ; in, The swept area of ​​the fan impeller; It is a three-dimensional spatial point wind speed function, which is the instantaneous / average wind speed at spatial location (x,y,z) obtained by three-dimensional scanning wind measurement lidar. It includes three-dimensional distribution information of the wind field within the impeller sweeping surface (such as wind speed shear, turbulence intensity, spatial non-uniformity, etc.).

[0036] This dataset also includes minimum clearance values; for multi-beam clearance radar, the minimum clearance value at the blade tip is... It can be obtained through geometric inversion, using the following formula: ; in, Let be the measured distance of the i-th ranging beam; t represents the angle between the blade tip and the tower corresponding to the i-th ranging beam; t represents the current time.

[0037] The constructed fusion time series dataset is input into the pre-trained airspace prediction model. This model has been trained with a large amount of historical operating data and has the ability to learn the complex nonlinear mapping relationship between wind field dynamic changes, unit dynamic response (such as blade deflection and speed fluctuation) and airspace distance evolution.

[0038] After the model inputs fused time series data, it combines it with a preset future time window. (Adapting to blade rotation period and wind field change rate), predict and output the predicted clearance between the blade tip and the tower within this time window (including the minimum predicted clearance and prediction confidence level); the prediction process is based on the following formula: ; in, This is the predicted change in net air volume based on the trend of incoming flow changes and the unit's operating status.

[0039] The core of this implementation method lies in "time alignment + model prediction": first, it solves the problem of spatiotemporal consistency of multi-source data, and then uses a model with generalization ability to extrapolate the future. Compared with traditional threshold judgment, it is more forward-looking and can effectively predict tower sweep risks in advance, providing accurate decision-making basis for subsequent safety control.

[0040] The aforementioned airspace prediction model for predicting clearance distance can employ a deep learning model based on a temporal neural network. This type of model can accurately capture the temporal correlation between wind field changes, unit operation, and airspace evolution, significantly improving the accuracy and robustness of airspace prediction under complex operating conditions. In this implementation, the control method also includes a model training stage, where the airspace prediction model is a deep learning model based on a temporal neural network. Furthermore, the method further includes: Obtain wind field data, airspace data, and unit status data under historical operating conditions as training samples; Using the actual airspace distance at future moments as a supervision label, a temporal neural network is trained to learn a nonlinear mapping from historical time-series data to future airspace distance.

[0041] First, a training sample set is constructed. Specifically, wind field data, airspace data, and unit status data aligned with timestamps are collected during the historical full-condition operation of wind turbine generators. These data are used as training samples for the temporal neural network, covering various scenarios such as normal wind conditions, gusts, wind shear, yaw error, and rain and fog interference.

[0042] Then, supervised training of the model is performed. Specifically, the actual airspace distance corresponding to future moments in historical time series data is used as the supervision label. The time series neural network is trained through supervised learning, allowing the model to autonomously learn the nonlinear mapping relationship between historical multi-source time series data and future airspace distance. At the same time, the sample weights are optimized by combining the credibility of multi-source data to enhance the prediction stability of the model in harsh environments and provide reliable model support for real-time airspace prediction.

[0043] Optionally, the control method can also include a multi-source data quality assessment and dynamic fusion weight allocation stage to improve the accuracy of airspace prediction and adaptability to harsh operating conditions; in this case, the method further includes: Perform signal quality assessment on wind field data and / or clearance data, and generate corresponding quality assessment parameters; Based on the quality assessment parameters, the fusion weights of wind field data and / or airspace distance data in predicting airspace distance are dynamically determined.

[0044] First, real-time signal quality testing is conducted on wind field data collected by wind-measuring lidar and clearance distance data collected by blade clearance radar. Based on indicators such as data integrity, measurement stability, signal attenuation under rain / fog / clutter interference, and data reliability, quality assessment parameters corresponding to wind field data and clearance distance data are generated to quantitatively characterize the usable accuracy and reliability of the two types of data.

[0045] Subsequently, based on the quality assessment parameters, the fusion weights of wind field data and clearance distance data are dynamically adjusted during the data fusion stage of the clearance prediction model. Data sources with high-quality and reliable data are given higher fusion weights, while data sources with poor quality and high interference are given lower weights. This weakens the interference of low-quality data on the clearance prediction results, adapts to multi-source mutual verification and control degradation strategies, and ensures the stability and accuracy of clearance prediction under complex operating conditions by the time-series neural network.

[0046] Optionally, to improve the accuracy of airspace risk assessment and avoid misjudgments and unnecessary downtime due to prediction errors or low-confidence data, this method also adds a dual verification step for airspace risk based on prediction confidence; in this case, the method further includes: Obtain the prediction confidence level corresponding to the net airspace forecast value; When the predicted net airspace value is lower than the preset safe net airspace threshold and the prediction confidence level is higher than the preset confidence level threshold, it is determined that there is a net airspace risk.

[0047] First, obtain the airspace prediction confidence level. While outputting the airspace prediction value through the time-series neural network-type airspace prediction model, combine information such as model prediction error, multi-source data quality assessment parameters, and real-time data fusion weights to simultaneously generate the prediction confidence level corresponding to the airspace prediction value, thereby quantitatively characterizing the reliability of the current airspace prediction result.

[0048] Subsequently, the airspace risk is accurately assessed by using a dual-condition approach of "airspace prediction value + prediction confidence level". Only when the airspace prediction value within a future time window is lower than the preset safe airspace threshold and the prediction confidence level corresponding to the prediction value is higher than the preset confidence level threshold, is it determined that the wind turbine has an airspace risk of blade sweeping. This effectively avoids misjudgments caused by low confidence prediction results, adapts to multi-source mutual verification and control degradation strategies, and reduces unnecessary power restrictions and downtime probability under severe operating conditions.

[0049] To further improve the reliability and robustness of airspace risk assessment under complex conditions such as rain, fog, and clutter, this method also introduces a dynamic adaptive adjustment mechanism for the airspace risk assessment strategy based on dual-sensor status. Optionally, the method further includes: The airspace risk assessment strategy is dynamically adjusted based on the availability or reliability of data from wind-measuring lidar and airspace radar. When wind measurement lidar data is unavailable or its reliability is below the first threshold, switch to the first judgment mode that prioritizes airspace radar data. When the airspace radar data is unavailable or its reliability is below the second threshold, switch to the second judgment mode that prioritizes wind measurement lidar data. When data from both sensors is unavailable or the reliability of both is below the corresponding threshold, a preset power limit or shutdown command is output.

[0050] Specifically, the system first monitors the availability and reliability of data from wind-measuring lidar and airspace radar in real time. Based on the real-time status of the two types of sensors, it dynamically switches the appropriate airspace risk assessment strategy and simultaneously matches multi-source mutual verification and control degradation logic.

[0051] When the wind-measuring lidar data fails or its data reliability is lower than the preset first threshold, it automatically switches to the first judgment mode that prioritizes the airspace radar data, using the actual blade tip clearance distance measured by the airspace radar as the core judgment basis, and strictly follows the hard constraints of airspace safety to carry out risk judgment.

[0052] When the airspace radar data is abnormal or its reliability is lower than the preset second threshold, it automatically switches to the second judgment mode that prioritizes wind lidar data. It relies on wind field feedforward data, unit operating status and airspace prediction model to complete the airspace risk assessment and ensure the continuity of the control process.

[0053] When the data from both the wind-measuring lidar and the air-clearance radar are unavailable, or when the reliability of both is below the corresponding threshold, the conventional risk assessment process is immediately skipped, and a preset power-limited operation or emergency shutdown safety command is directly output to execute the ultimate protection strategy, thereby eliminating the risk of blade sweeping from the source and ensuring the safe operation of the unit.

[0054] When a wind turbine is determined to have a risk of blade sweeping to the tower based on the airspace forecast, the specific process for outputting and executing control commands to the turbine is as follows: Optionally, when an airspace risk is determined to exist based on the airspace forecast, the steps for outputting control commands to the wind turbine include: Obtain the airspace safety threshold corresponding to the airspace risk, and calculate the difference between the airspace safety threshold and the predicted airspace value; Based on the difference, determine the constraints corresponding to the control command, and obtain the control command when the structural load of the wind turbine generator is at its minimum value under the constraints. Control commands are sent to the wind turbine generator set.

[0055] Specifically, the first step is to calculate the difference in the safety margin of the airspace. By retrieving the preset safety threshold of the airspace of the wind turbine generator set (i.e., the core safety hard constraint of the anti-sweeping tower), the safety threshold is compared with the current predicted airspace value, and the difference between the two is calculated. This quantifies the current safety margin of the airspace and the risk level, providing a quantitative basis for setting subsequent control constraints.

[0056] Then, constraints are set and the optimal load control command is solved. Based on the above-mentioned clearance difference, the clearance safety constraints that the control command must meet are determined to ensure that the clearance distance is always not lower than the safety threshold during the unit adjustment process. At the same time, with the optimization objective of minimizing the structural load of key components such as towers, blades, transmission chains, and yaw equipment, the optimal control command that balances safety and load reduction is solved by the control algorithm under the premise of clearance safety constraints.

[0057] Finally, the command is issued to execute the unit adjustment. Specifically, the generated optimal control command is sent to the actuator of the wind turbine in real time to coordinate the adjustment of parameters such as yaw, pitch, torque, and power. While strictly avoiding tower sweep risk and meeting the hard constraints of airspace safety, the structural load of the unit is reduced to the minimum. It is adapted to multi-source mutual verification and control degradation strategies to ensure the safe and stable operation of the unit.

[0058] Optionally, the control commands include at least one of yaw correction commands, pitch adjustment commands, torque adjustment commands, and power adjustment commands; the structural loads include at least one of tower loads, blade loads, yaw system loads, and drive train loads.

[0059] like Figure 2 The flowchart of another wind turbine control method shown includes the following steps: S201: Use a wind-measuring lidar to collect the incoming wind field.

[0060] The deployment of a 3D scanning wind measurement lidar and a blade clearance radar on the wind turbine generator was completed, with time synchronization and coordinate system unification achieved simultaneously. Preset sampling periods were configured for both types of sensors. And perform data timestamp alignment calibration to ensure that the timestamp deviation of data from different sensors meets the system requirements (i.e. ,in For data timestamp deviation, This is the maximum allowed synchronization error threshold for the system, laying the foundation for subsequent data acquisition and fusion.

[0061] S202: Use airspace radar to collect and calculate the minimum airspace value.

[0062] A three-dimensional scanning lidar for wind measurement scans and measures the wind field in front of the impeller. After collecting raw wind field data, the equivalent incoming wind speed to the impeller is calculated through multi-point inversion. The yaw angle (characterizing the deviation between the incoming flow direction and the unit's orientation) is derived by combining measured wind direction and nacelle orientation. Finally, core wind field characteristic parameters such as impeller equivalent incoming wind speed, wind shear parameters, gust amplitude and arrival time, and turbulence intensity are extracted to construct a feedforward feature vector characterizing the future short-term incoming flow change trend, providing wind field input for load optimization and clearance prediction.

[0063] S203: Obtain the unit's operating status.

[0064] This step mainly involves using airspace radar to collect and calculate the minimum airspace value and obtain the unit's operating status.

[0065] Specifically, the blade clearance radar (which can use a multi-beam ranging radar) transmits radar signals in real time to measure the distance between the blade tip and the tower; for multi-beam radar, the distance is calculated through geometric inversion (which can be done using...). (to calculate), where Let be the measured distance of the i-th ranging beam. (The angle between the blade tip and the tower) is used to filter out the minimum clearance value between the blade tip and the tower, and the corresponding credibility information is output simultaneously.

[0066] The unit status acquisition module collects real-time operating status data from the wind turbine main control system, including key parameters such as rotor speed, blade pitch angle, nacelle yaw angle, power command, and torque command, to construct a unit operating status vector and comprehensively quantify the current operating conditions of the unit.

[0067] S204: Data fusion is performed through methods such as time alignment.

[0068] Time alignment optimization is performed on the wind field characteristic data collected by S202, the minimum airspace data obtained by S203, and the unit operating status data to ensure that the three types of data match in the same time dimension. Then, the multi-source data is integrated, denoised, and redundantly removed through a preset fusion algorithm to eliminate single data errors and form high-quality fused data, which serves as the core input for subsequent airspace risk prediction and control decisions.

[0069] S205: Predict net airspace risk for future time windows.

[0070] Based on the fused data from S204, and combining the blade deflection characteristics with the incoming flow variation patterns, a headspace risk assessment model is constructed. This model is designed for a pre-defined future time window. Calculate the predicted change in net airspace Specifically, it can be done through... The predicted minimum air clearance value is calculated. Then, this predicted value is compared with the preset safe air clearance threshold to generate an air clearance risk assessment result, clarify the air clearance safety margin, and predict the risk of tower sweeping.

[0071] S206: Multi-source mutual verification control downgrade strategy.

[0072] Based on the availability and reliability of data from wind-measuring lidar and airspace radar, the control strategy is dynamically adjusted: If the data quality of the wind-measuring lidar deteriorates or its reliability falls below the threshold due to operating conditions such as rain, fog, or obstruction, the control weight of the airspace radar data will be automatically increased, and the system will switch to a conservative control mode that prioritizes airspace safety. If the airspace radar output is abnormal or its reliability is insufficient, switch to a control mode based on predicted airspace and safety margin to ensure core safety constraints. If both types of radar data are unavailable, immediately implement the preset safety power limiting or shutdown strategy to avoid operational risks.

[0073] S207: Obtain load-headroom coordinated control decision.

[0074] The optimization objective is to reduce structural loads on the tower, blades, drivetrain, and yaw system, while maintaining a minimum predicted clearance that is not less than a preset safe clearance threshold as an insurmountable hard constraint. Under the premise of meeting these safety constraints, the algorithmic calculations of the fusion and control modules collaboratively generate yaw correction commands, pitch adjustment commands, and torque or power adjustment commands, achieving a dual balance between load optimization and clearance safety.

[0075] S208: Outputs control commands such as yaw, pitch, and power.

[0076] Through the actuator interface module, the coordinated control commands generated by S207 are sent in real time to the corresponding actuators of the unit, coordinating the adjustment of at least two types of mechanisms, including yaw, pitch, torque, and power. During the execution of the commands, the unit's operational feedback is continuously monitored, and parameters are dynamically corrected to ensure that the unit operates stably while meeting airspace safety and load requirements, effectively suppressing the risk of blade scorching, and simultaneously ensuring power generation efficiency.

[0077] As can be seen from the wind turbine control method mentioned in the above embodiments, this method uses the airspace radar as a hard safety constraint, which significantly reduces the risk of tower sweeping and can maintain control reliability under conditions such as rain, fog, and clutter, reducing unnecessary power limitations and the probability of downtime, and greatly reducing the misjudgment rate of wind turbines in the airspace control process. In addition, this method can couple wind measurement load reduction with airspace safety in the wind turbine control process, and can perform feedforward control under conditions such as gusts, shear, and yaw errors, thereby reducing the load on the wind turbine.

[0078] Corresponding to the above embodiments of the wind turbine control method, this invention also provides a wind turbine control system, such as... Figure 3 As shown, the system includes: Data acquisition module 100: used to acquire wind field data collected by wind lidar, clearance distance data collected by clearance radar, and operating status data of wind turbine generator set; Clearance prediction value determination module 200: Based on wind farm data, clearance distance data and operating status data, it predicts the clearance distance of wind turbine generators within a future time window and obtains the clearance prediction value. Control execution module 300: When it is determined that there is a risk to the airspace based on the airspace forecast value, it outputs control commands to the wind turbine generator set.

[0079] As can be seen from the above wind turbine control system, the system uses the airspace radar as a hard constraint for safety, which significantly reduces the risk of tower sweeping and can maintain control reliability in rain, fog, clutter and other conditions, reducing unnecessary power limitations and the probability of downtime, and greatly reducing the misjudgment rate of wind turbines in the airspace control process. In addition, the system can couple wind measurement and load reduction with airspace safety in the wind turbine control process, and can perform feedforward control under conditions such as gusts, shear and yaw errors to reduce the load on the wind turbine.

[0080] The wind turbine control system provided in this embodiment of the invention has the same implementation principle and technical effects as the aforementioned wind turbine control method embodiment. For the sake of brevity, any parts not mentioned in the system embodiment can be referred to the corresponding content in the aforementioned wind turbine control method embodiment.

[0081] This embodiment also provides a server, the structural diagram of which is shown below. Figure 4 As shown, the device includes a processor 101 and a memory 102; wherein, the memory 102 is used to store one or more computer instructions, which are executed by the processor to implement the steps of the wind turbine control method described above.

[0082] Figure 4 The server shown also includes a bus 103 and a communication interface 104. The processor 101, the communication interface 104, and the memory 102 are connected via the bus 103.

[0083] The memory 102 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. The bus 103 may be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0084] The communication interface 104 is used to connect to at least one user terminal and other network units through a network interface, and to send encapsulated IPv4 packets or IPv4 packets to the user terminal through the network interface.

[0085] Processor 101 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 101 or by instructions in software form. The processor 101 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this disclosure. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this disclosure can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 102. The processor 101 reads the information in memory 102 and, in conjunction with its hardware, completes the steps of the method described in the foregoing embodiments.

[0086] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, devices, and methods can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0087] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0088] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0089] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0090] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A wind turbine control method, characterized in that, The method includes: Acquire wind field data collected by wind-measuring lidar, clearance distance data collected by clearance radar, and operating status data of wind turbine generators; Based on the wind field data, the clearance distance data, and the operating status data, the clearance distance of the wind turbine generator set in a future time window is predicted to obtain the clearance prediction value. When it is determined that there is a risk to airspace based on the predicted airspace value, a control command is output to the wind turbine generator set.

2. The wind turbine control method according to claim 1, characterized in that, The step of predicting the airspace clearance of the wind turbine generator within a future time window based on the wind farm data, the clearance distance data, and the operating status data includes: The wind farm data, clearance distance data, and unit operating status data are aligned according to timestamps to generate a fused time-series dataset; The fused time-series dataset is input into a pre-built airspace prediction model, which is used to learn the mapping relationship between wind field changes, unit response and airspace dynamic evolution, and outputs the predicted airspace value.

3. The wind turbine control method according to claim 2, characterized in that, The airspace prediction model is a deep learning model built on a temporal neural network; the method further includes: Obtain wind field data, airspace data, and unit status data under historical operating conditions as training samples; Using the actual airspace distance at future moments as a supervision label, the temporal neural network is trained to learn a nonlinear mapping from historical time-series data to future airspace distance.

4. The wind turbine control method according to claim 1, characterized in that, The method further includes: Perform signal quality assessment on the wind field data and / or clearance distance data, and generate corresponding quality assessment parameters; Based on the quality assessment parameters, the fusion weights of the wind field data and / or airspace distance data in predicting airspace distance are dynamically determined.

5. The wind turbine control method according to claim 1, characterized in that, The method further includes: Obtain the prediction confidence level corresponding to the predicted net airspace value; When the predicted airspace value is lower than the preset safe airspace threshold and the prediction confidence level is higher than the preset confidence level threshold, it is determined that there is an airspace risk.

6. The wind turbine control method according to claim 1, characterized in that, The method further includes: The airspace risk assessment strategy is dynamically adjusted based on the availability or reliability of the data from the wind-measuring lidar and the airspace clearance radar. When wind measurement lidar data is unavailable or its reliability is below the first threshold, switch to the first judgment mode that prioritizes airspace radar data. When the airspace radar data is unavailable or its reliability is below the second threshold, switch to the second judgment mode that prioritizes wind measurement lidar data. When data from both sensors is unavailable or the reliability of both is below the corresponding threshold, a preset power limit or shutdown command is output.

7. The wind turbine generator control method according to claim 1, characterized in that, When an airspace risk is determined to exist based on the airspace forecast value, the step of outputting a control command to the wind turbine generator set includes: Obtain the airspace safety threshold corresponding to the airspace risk, and calculate the difference between the airspace safety threshold and the predicted airspace value; Based on the difference, the constraint conditions corresponding to the control command are determined, and the control command when the structural load of the wind turbine generator is at its minimum value is obtained under the constraint conditions. The control command is sent to the wind turbine generator set.

8. The wind turbine generator control method according to claim 7, characterized in that, The control commands include at least one of yaw correction commands, pitch adjustment commands, torque adjustment commands, and power adjustment commands; The structural loads include at least one of the following: tower load, blade load, yaw system load, and drive train load.

9. A wind turbine control system, characterized in that, The system includes: Data acquisition module: used to acquire wind field data collected by wind lidar, clearance distance data collected by clearance radar, and operating status data of wind turbine generators; Clearance prediction value determination module: used to predict the clearance distance of the wind turbine generator in a future time window based on the wind field data, the clearance distance data and the operating status data, and obtain the clearance prediction value; Control execution module: When it is determined that there is a risk to the airspace based on the airspace prediction value, the module outputs control commands to the wind turbine generator set.

10. A server, characterized in that, The server includes a processor and a memory, the memory storing computer-executable instructions that can be executed by the processor, the processor executing the computer-executable instructions to implement the steps of the wind turbine control method according to any one of claims 1 to 8.

Images