Method of controlling a wind turbine based on wind conditions

Abstract

The application provides a wind turbine control method based on wind conditions, comprising: obtaining operation data and wind condition data of the wind turbine; preprocessing the operation data and the wind condition data; extracting features from the preprocessed operation data and wind condition data to obtain operation feature data and wind condition feature data; identifying the wind condition type represented by the wind condition feature data; when the wind condition feature data indicates that the current wind condition is a preset wind condition type, combining the preset wind condition type and the operation feature data to select a corresponding unit control rule; and controlling the wind turbine based on the unit control rule, which can effectively improve the adaptability of the wind turbine under complex wind conditions, through dynamic feature recognition and adaptive control strategy selection, ensuring the operation safety of the unit and maintaining the power generation efficiency, avoiding the technical defects of response lag and single strategy, and realizing the intelligentization and refinement of the wind turbine control strategy.

Description

Technical Field

[0001] This invention relates to the field of wind power generation control technology, and in particular to a control method for wind turbine generators based on wind conditions. Background Technology

[0002] As a type of renewable energy power generation equipment, wind turbines often face various complex and changeable wind conditions during operation, such as gusts, turbulence, wind shear, and sudden strong winds. These special wind conditions can have a significant impact on the operating performance, safety, and lifespan of wind turbines.

[0003] Traditional wind turbine control methods are typically designed based on steady-state wind conditions, making it difficult to effectively cope with these complex and ever-changing special wind conditions. Although there are some control methods in existing technologies for special wind conditions, most of these methods have shortcomings, such as single control strategies, slow response speed, and poor adaptability. They are unable to cope with multiple special wind conditions simultaneously, nor can they adjust the control strategy in real time according to dynamic changes in wind conditions, thus limiting the operating performance and safety of wind turbines under complex wind conditions. Summary of the Invention

[0004] This invention provides a wind turbine control method based on wind conditions, which solves the technical problems of single control strategies and poor adaptability of wind turbines under complex wind conditions in the prior art.

[0005] On one hand, the present invention provides a control method for wind turbine generators based on wind conditions, comprising: Acquire operating data and wind condition data of wind turbines; The operational data and the wind condition data are preprocessed; Feature extraction is performed on the preprocessed operational data and wind condition data to obtain operational feature data and wind condition feature data. Identify the wind condition type represented by the wind condition feature data; When the wind condition characteristic data indicates that the current wind condition is a preset wind condition type, the corresponding unit control rule is selected by combining the preset wind condition type and the operating characteristic data; The wind turbine is controlled based on the aforementioned unit control rules.

[0006] According to a wind condition-based control method for wind turbines provided by the present invention, when the preset wind condition type is a storm, the turbine control rules include: If the current wind speed is detected to be greater than the preset cut-out wind speed, the power limit of the wind turbine is restricted, and the power setting value is dynamically adjusted according to the pre-created wind speed power interpolation table. The wind speed-power interpolation table includes multiple sets of mapping relationships between wind speed values ​​and corresponding power limits, and the power limits decrease as the wind speed increases, so as to keep the unit in power generation operation.

[0007] According to the present invention, when the preset wind condition type is gust, the unit control rules include: Monitor generator speed and acceleration; If the acceleration continues to increase and exceeds a preset threshold, the pitch angle is limited to a preset minimum angle limit. The minimum angle limit is determined through load iterative optimization to accelerate the response speed of the pitch system.

[0008] According to a wind condition-based control method for wind turbines provided by the present invention, when the preset wind condition type is a sudden change in wind direction, the turbine control rules include: Detect the current yaw deviation; If the yaw deviation exceeds the preset limit, the pitch angle is raised to the corresponding value according to the pre-created minimum pitch angle interpolation table for wind deviation. The interpolation table includes multiple sets of mapping relationships between yaw deviation values ​​and minimum pitch angle, and the pitch angle increases with the increase of deviation, so as to reduce the unbalanced load of the unit.

[0009] According to a wind condition-based control method for wind turbines provided by the present invention, when the preset wind condition type is an emergency crosswind, the turbine control rules include: If the generator speed of the wind turbine exceeds the safety threshold and the pitch system reports a fault signal, the yaw drive of the wind turbine will be controlled to perform a 90° crosswind action and the yaw state will be locked until manual reset.

[0010] According to a wind condition-based control method for wind turbines provided by the present invention, when the preset wind condition type is turbulence, the turbine control rules include: turbulence intensity is calculated based on a wind speed sequence of preset duration. The power limit of the wind turbine is dynamically adjusted based on the turbulence intensity and a pre-created turbulence power limit interpolation table. The turbulence power limit interpolation table includes multiple sets of mapping relationships between turbulence intensity values ​​and power limits, and the power limit decreases as the turbulence intensity increases, so as to reduce the fatigue load of the unit.

[0011] According to a wind condition-based control method for wind turbines provided by the present invention, when the preset wind condition type is a composite wind condition, the turbine control rules include: Based on preset priority rules, multiple unit control rules are dynamically integrated; The composite wind condition refers to a wind condition in which at least two preset wind condition types exist simultaneously; The priority rules are ranked based on the degree of impact of wind conditions on unit safety and power generation efficiency, and the control rules with the greatest impact on safety are executed first.

[0012] According to a wind condition-based control method for wind turbines provided by the present invention, when the preset wind condition type is a composite wind condition, the turbine control rules include: When the current wind condition is identified as one of the preset wind condition types, but its matching degree is lower than the preset matching threshold, the system enters the fuzzy recognition mode. In fuzzy recognition mode, multiple unit control rules with similar wind conditions are invoked, and weights are assigned to each control rule based on the operating data to generate fused control commands.

[0013] According to the wind condition-based control method for wind turbines provided by the present invention, multiple control rules for turbines with similar wind condition types are invoked, and weights are allocated to each control rule based on operating data to generate a fused control command, including: Calculate the fuzzy membership degree between the current wind condition feature data and the preset wind condition type, and select wind condition types with a fuzzy membership degree greater than or equal to the preset fuzzy threshold as candidate control rule sets; Based on the unit safety status, grid dispatching requirements, and historical success rate in the operational data, the applicable weight of each candidate control rule is determined; wherein, the applicable weight reflects the degree of adaptation of the control rule to the current operational data, and the higher the weight value, the higher the priority of its adoption; The output control parameters of each control rule are weighted and fused to generate the final fused control command.

[0014] According to a wind condition-based control method for wind turbines provided by the present invention, when the current wind condition is identified as a transitional wind condition, the turbine control rules include: Based on the trend of wind condition changes within a preset time window, the wind conditions are divided into multiple evolutionary stages. Based on the dynamic coupling relationship between different evolutionary stages and operational characteristic data, staged response rules are defined.

[0015] The wind condition-based wind turbine control method provided by this invention establishes a wind condition identification mechanism in a multi-dimensional feature space, which can accurately distinguish different wind condition types and their combinations. It adopts a dynamic rule selection mechanism to replace the static control strategy, and can match the optimal control parameters according to real-time operating characteristics. This can effectively improve the adaptability of wind turbines under complex wind conditions. Through dynamic feature identification and adaptive control strategy selection, it not only ensures the safety of turbine operation but also maintains power generation efficiency, solving the technical defects of response lag and single strategy, and realizing the intelligence and refinement of wind turbine control strategy. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a schematic flowchart of the wind turbine control method based on wind conditions provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of the wind turbine control device based on wind conditions provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0019] Figure 1 This is a flowchart illustrating the wind turbine control method based on wind conditions provided in an embodiment of the present invention.

[0020] See Figure 1 The wind turbine control method based on wind conditions may include the following steps 101 to 106.

[0021] Step 101: Obtain the operating data and wind condition data of the wind turbine.

[0022] In this step, operational data refers to the set of parameters reflecting the unit's operating status. Specifically, this can be achieved using real-time monitoring data such as generator speed, blade pitch angle, and power output, which characterizes the unit's current operating status. Wind condition data refers to the set of parameters reflecting ambient wind force. Specifically, this can be achieved using data collected by sensors such as anemometers, wind vanes, and three-dimensional ultrasonic anemometers, which characterizes the current wind field environmental characteristics.

[0023] Step 102: Preprocess the operational data and wind condition data.

[0024] In this step, preprocessing refers to the standardization of the raw data. This can be achieved through methods such as data cleaning, outlier removal, and time series alignment to ensure the validity and consistency of the data for subsequent analysis. Based on the preprocessed data, filters can be applied for further processing to remove noise and smooth the data.

[0025] Step 103: Extract features from the preprocessed operational data and wind condition data to obtain operational feature data and wind condition feature data.

[0026] In this step, feature extraction refers to the process of extracting key information from preprocessed data. Specifically, it can be achieved using algorithms such as time-domain statistical analysis, frequency-domain transformation, and wavelet decomposition to construct the correlation features between wind conditions and unit status.

[0027] Step 104: Identify the wind condition type represented by the wind condition feature data.

[0028] In this step, the preset wind condition type refers to a predefined wind condition classification system, which can be implemented by classifying wind conditions into categories such as storms, gusts, and turbulence according to meteorological standards. Each category corresponds to a specific set of control strategies.

[0029] Step 105: When the wind condition characteristic data indicates that the current wind condition is a preset wind condition type, select the corresponding unit control rule by combining the preset wind condition type and the operating characteristic data.

[0030] Step 106: Control the wind turbine based on the unit control rules.

[0031] In this embodiment, specifically, the unit's operating parameters and environmental wind condition parameters are collected in real time through a sensor network, and the raw data is processed by timestamp alignment and dimension normalization. A sliding window mechanism is used to extract features from continuous time-series data, such as calculating feature vectors for wind speed mean variance, wind direction change rate, and power fluctuation amplitude. A machine learning-based wind condition classification model is established, and real-time feature data is matched and identified against a feature library of preset wind condition types. When the current wind condition is detected to belong to a preset type, the corresponding control rule library is invoked, for example, a power limiting strategy is adopted for storm conditions, and a rapid pitch angle adjustment strategy is adopted for gust conditions. The control command generation module dynamically adjusts the control parameters according to the real-time operating characteristics to ensure that the unit maintains its optimal operating state within the safety threshold.

[0032] This embodiment establishes a wind condition recognition mechanism based on a multi-dimensional feature space, enabling accurate differentiation of different wind condition types and their combinations. A dynamic rule selection mechanism replaces the static control strategy, allowing for the matching of optimal control parameters based on real-time operational characteristics. This technical solution effectively enhances the adaptability of wind turbines under complex wind conditions. Through dynamic feature recognition and adaptive control strategy selection, it ensures both operational safety and power generation efficiency. This solution overcomes the technical shortcomings of traditional methods, such as lag and limited strategy simplification, achieving intelligent and refined wind turbine control strategies.

[0033] In one embodiment of this specification, when the preset wind condition type is storm, the unit control rules include: If the current wind speed is detected to be greater than the preset cut-out wind speed, the power limit of the wind turbine is restricted, and the power setting value is dynamically adjusted according to the pre-created wind speed-power interpolation table. The wind speed-power interpolation table includes multiple sets of mapping relationships between wind speed values ​​and corresponding power limits, and the power limits decrease as the wind speed increases, so as to keep the unit in power generation operation.

[0034] In this embodiment, the cut-out wind speed refers to the maximum safe operating wind speed threshold allowed by the wind turbine design. This threshold can be set using wind turbine model parameters or historical operating data, and is used to determine whether the power limiting mechanism is triggered. The wind speed-power interpolation table is a database storing the relationship between different wind speeds and their corresponding power limits. It can be generated using linear interpolation or polynomial fitting algorithms and is used to dynamically adjust power output when the wind speed exceeds the cut-out wind speed. The power limit decreasing with increasing wind speed means that the upper limit of power is gradually reduced as the wind speed increases. This can be implemented using a piecewise function or gradient descent model to balance turbine safety and power generation continuity.

[0035] Specifically, once a storm condition is identified, real-time wind speed data is compared with the preset cut-out wind speed. If the current wind speed exceeds the cut-out wind speed, dynamic power control mode is activated. In this mode, the power limit is restricted to a value lower than the rated power to prevent the unit from triggering shutdown protection due to overload. Simultaneously, based on the mapping relationship stored in the wind speed-power interpolation table, the corresponding power limit is matched according to the current wind speed value. For example, when the wind speed reaches 105% of the cut-out wind speed, the power limit can be adjusted to 80% of the rated power; when the wind speed reaches 110% of the cut-out wind speed, the power limit is further reduced to 60% of the rated power. Through this decreasing mechanism, the power generation operation is maintained while ensuring the safety of the unit's mechanical structure.

[0036] This embodiment achieves continuous and controllable power output under storm conditions by dynamically adjusting the power limit, while avoiding frequent start-ups and shutdowns caused by instantaneous wind speed fluctuations. Furthermore, the power adjustment method based on interpolation tables enables a smooth transition and is more adaptable to the non-linear characteristics of wind speed changes compared to fixed threshold control strategies. Through these technical solutions, the safe operation of the unit and the demand for continuous power generation are effectively balanced under storm conditions, avoiding power generation losses caused by direct shutdowns. The dynamic reduction mechanism of the power limit reduces the load impact on the transmission system under extreme wind speeds, extending the service life of critical components. Simultaneously, the application of interpolation tables makes the power adjustment process predictable and controllable, providing a stable power output reference for grid dispatch.

[0037] In one embodiment of this specification, when the preset wind condition type is gust, the unit control rules include: Monitor generator speed and acceleration; If the acceleration continues to rise and exceeds the preset threshold, the pitch angle will be limited to the preset minimum angle limit. The minimum angle limit is determined through load iterative optimization to accelerate the response speed of the pitch system.

[0038] In this embodiment, generator speed acceleration refers to the change in generator speed per unit time. Specifically, it can be achieved by collecting real-time data from a speed sensor and performing differential calculations, characterizing the dynamic response state of the unit under gust wind impact. The preset threshold is the critical acceleration value that triggers pitch angle adjustment. Specifically, it can be determined through historical operating data analysis combined with simulation testing, used to identify abnormal acceleration states caused by gust winds. The minimum angle limit is the minimum pitch angle that the pitch system is allowed to adjust. Specifically, it can be determined through multi-condition load simulation combined with iterative optimization using measured data, used to achieve rapid response while ensuring structural safety. Load iterative optimization refers to the calculation process of repeatedly adjusting parameters based on structural load data under different pitch angles. Specifically, it can be achieved using a closed-loop feedback between a finite element model and measured data, used to balance the contradiction between pitch speed and unit load-bearing capacity.

[0039] Specifically, when the generator speed acceleration is detected to continuously exceed a preset threshold, it indicates that gusts have caused the unit to enter unsteady-state operation. At this point, the pitch angle is limited to a minimum angle limit optimized through load iteration. By increasing the pitch angle, the wind energy capture efficiency is reduced, thereby suppressing sudden changes in speed. This minimum angle limit was determined through multiple rounds of load simulation and comparison with measured data, ensuring that the pitch system can respond quickly while avoiding structural overload caused by excessive adjustments. This control method can intervene at the initial stage of gusts, suppressing speed fluctuations earlier.

[0040] This embodiment achieves precise intervention while ensuring structural safety by dynamically monitoring acceleration trends and combining them with optimized and verified minimum pitch angle limits. The load iterative optimization process effectively resolves the contradiction between pitch speed and mechanical strength, avoiding the conservative or overly aggressive problems caused by empirical parameter settings in existing technologies. Through the above technical solution, the unit instability problem caused by sudden changes in generator speed under gust wind conditions is effectively solved. By identifying abnormal acceleration trends in advance and adopting optimized minimum pitch angle limits, the power generation state is maintained while reducing mechanical impact. The load iterative optimization method ensures dynamic matching between pitch action and structural load-bearing capacity, improving the adaptability and reliability of the control system.

[0041] In one embodiment of this specification, when the preset wind condition type is a sudden change in wind direction, the unit control rules include: Detect the current yaw deviation; If the yaw deviation exceeds the preset limit, the pitch angle will be raised to the corresponding value according to the pre-created minimum pitch angle interpolation table for wind deviation. The interpolation table includes multiple sets of mapping relationships between yaw deviation values ​​and minimum pitch angle, and the pitch angle increases with the increase of deviation, so as to reduce the unbalanced load of the unit.

[0042] In this embodiment, yaw deviation refers to the angular difference between the actual wind direction and the orientation of the wind turbine nacelle. This can be achieved by synchronously measuring the wind direction sensor and yaw encoder. This deviation value is used to determine whether the turbine is in an abnormal wind-fighting state. The minimum pitch angle interpolation table for wind-fighting deviation is a two-dimensional data table storing yaw deviation values ​​and corresponding pitch angle settings. It can be generated through offline simulation or fitting historical operating data. Its function is to establish a quantitative relationship between the deviation value and the pitch angle adjustment, avoiding delays caused by real-time calculations. The pitch angle increasing with deviation means that when the yaw deviation exceeds a preset limit, the pitch angle setting is increased linearly or non-linearly according to the interpolation table. This can be achieved by combining a lookup table with a linear interpolation algorithm. This mechanism can quickly respond to load imbalance problems caused by sudden changes in wind direction.

[0043] Specifically, when a sudden change in wind direction causes the yaw deviation to exceed a preset limit, such as 5 degrees, the pre-generated minimum pitch angle interpolation table for wind deviation is immediately invoked. For example, when the deviation is 6 degrees, the corresponding pitch angle is raised to 3 degrees; when the deviation is 10 degrees, the corresponding pitch angle is raised to 8 degrees. After obtaining the target pitch angle by looking up the table, the pitch system performs a rapid pitch change, increasing the blade angle of attack and thus reducing the load area on the wind turbine, thereby reducing the aerodynamic unbalance torque caused by the yaw deviation. During this process, the mapping relationship of the interpolation table is verified through load optimization to ensure that the power generation is maintained within an acceptable range while reducing the load.

[0044] This embodiment employs a dynamic pitch angle adjustment mechanism to prioritize reducing aerodynamic loads through pitch control before the yaw system completes its wind-fighting maneuvers, forming a dual protection mechanism. Simultaneously, the data-driven approach using interpolation tables, compared to traditional proportional control strategies, can more accurately match the optimal pitch angle setpoint under different deviation levels. This technical solution effectively solves the problem of a surge in structural loads caused by yaw lag during sudden wind direction changes. By raising the pitch angle in advance, the stress on the blades is reduced, preventing torque fluctuations in the transmission chain components from exceeding design limits. Furthermore, the pre-defined mapping relationship of the interpolation tables ensures the real-time performance and accuracy of the control response, achieving dynamic load balance control while maintaining continuous power generation.

[0045] In one embodiment of this specification, when the preset wind condition type is emergency crosswind, the unit control rules include: If the generator speed of the wind turbine exceeds the safety threshold and the pitch system reports a fault signal, the yaw drive of the wind turbine will be controlled to perform a 90° crosswind action and the yaw state will be locked until manual reset.

[0046] In this embodiment, an emergency crosswind refers to a wind condition where a sudden change in wind direction causes the unit to experience abnormal lateral loads. This can be achieved by combining a wind speed sensor and a wind vane for monitoring. This feature is used to identify extreme wind conditions that may lead to structural instability. The safety threshold refers to a pre-set upper limit for the generator speed. This can be calculated by adding a dynamic margin coefficient to the unit's rated speed, or it can be set empirically. This threshold is used to determine whether to trigger an emergency protection mechanism to prevent mechanical overload. The pitch system feedback fault signal refers to the abnormal status code returned by the pitch driver or sensor. This can be achieved through parsing via the bus communication protocol. This signal is used to confirm that the pitch system cannot adjust the pitch angle normally. The 90° crosswind action refers to the yaw system driving the nacelle to rotate so that the blade plane forms a perpendicular angle with the wind direction. This can be achieved through encoder positioning and servo motor linkage. This action is used to quickly reduce the blade's frontal area to reduce wind load. Locking the yaw state means prohibiting the yaw system from autonomously adjusting the nacelle angle. This can be achieved through interlocking the electromagnetic brake with the controller. This measure is used to maintain mechanical stability in emergency situations. Manual reset refers to the system being unlocked only after on-site confirmation by an operator. This can be achieved through a safety switch and an access control module. This mechanism ensures that automatic operation is not restored until the fault is resolved.

[0047] Specifically, when the wind turbine is in an emergency crosswind condition, the system performs dual verification by monitoring the generator speed in real time to ensure it does not exceed a safe threshold, combined with fault signals from the pitch system. If both the speed exceeding the limit and the pitch system failure condition are met simultaneously, the yaw drive is immediately triggered to perform a 90° crosswind maneuver, aligning the blade plane perpendicular to the wind direction to minimize aerodynamic loads. During this process, the yaw system enters a locked state to prevent accidental rotation and simultaneously sends an alarm signal to the monitoring center. This state remains in effect until maintenance personnel complete the equipment inspection and manually unlock it via a physical switch. This control logic proactively cuts off wind power input in the event of a pitch system failure, preventing mechanical damage due to uncontrolled speed.

[0048] This embodiment utilizes active nacelle angle adjustment to provide aerodynamic protection, maintaining unit mechanical stability in a locked state. This avoids overload risks and reduces power generation losses caused by unnecessary shutdowns. Through this technical solution, a dual protection mechanism of active yaw and status locking is established in the event of a sudden pitch system failure and an emergency crosswind. This effectively prevents mechanical damage caused by generator overspeed while providing a safe operating window for manual intervention, significantly improving equipment reliability under extreme conditions.

[0049] In one embodiment of this specification, when the preset wind condition type is turbulence, the unit control rules include: Based on a wind speed sequence of a preset duration, the turbulence intensity is calculated as shown in the following formula (1): (1); in, Let be the turbulence intensity at time t. The average wind speed within the preset time window. This represents the standard deviation of the wind speed sequence within a preset time window. The preset time window can be set, for example, to 10 seconds or 10 minutes.

[0050] When the preset time window is 10 minutes, the standard deviation of the wind speed sequence can be obtained by formula (2): (2); If each second can be considered a point in time, then 10 minutes can have 600 points in time. Let i be the wind speed at the i-th time point, where i ranges from 1 to 600, representing 600 wind speed data points.

[0051] The power limit of the wind turbine is dynamically adjusted based on the turbulence intensity and a pre-created turbulence power limit interpolation table. The turbulence power limit interpolation table includes multiple sets of mapping relationships between turbulence intensity values ​​and power limits. The power limit decreases as the turbulence intensity increases, in order to reduce the fatigue load on the unit.

[0052] In this embodiment, turbulence intensity refers to a quantitative index of airflow fluctuation calculated by the ratio of the standard deviation of the wind speed sequence to the average wind speed within a preset time window. Specifically, it can be implemented using a sliding time window statistical method, and is used to characterize the impact level of the current turbulence on the unit structure. The turbulence-limited power interpolation table is a two-dimensional data table storing the correspondence between turbulence intensity and power limits. Specifically, it can be implemented using a piecewise linear interpolation algorithm, reducing the stress cycle number of unit components by decreasing the power output under high turbulence intensity. Dynamically adjusting the upper power limit refers to matching the power limit in the interpolation table according to the real-time calculated turbulence intensity and updating the generator torque or pitch angle control commands. Specifically, it can be implemented using a closed-loop feedback control strategy, enabling the unit to maintain stable operation under turbulent conditions.

[0053] Specifically, upon detecting turbulent wind conditions, a continuous sequence of wind speed data is first collected at a fixed sampling period, such as using wind speed samples per second within a 10-second time window. The current turbulence intensity is obtained by calculating the ratio of the standard deviation of this sequence to the average wind speed. Subsequently, a pre-established turbulence power limit interpolation table is consulted to match the power limit corresponding to the current turbulence intensity; for example, a turbulence intensity of 0.3 corresponds to 80% of the rated power. The control module adjusts the generator torque setpoint or pitch angle in real time based on this limit, ensuring that the unit's output power remains below the limit threshold. This step-wise attenuation of power output effectively reduces the dynamic loads on key components such as blade root bending moments and tower vibrations.

[0054] This embodiment achieves a balanced optimization of load control and power generation efficiency by establishing a quantitative correspondence between turbulence intensity and power limits. Through this technical solution, power output can be gradually reduced as turbulence intensity continuously increases, avoiding grid impact caused by sudden power limitations. Simultaneously, by reducing the stress amplitude of key components, the service life of blade bearings and gearboxes is effectively extended. This control strategy enables the unit to maintain the cumulative rate of structural fatigue damage within a safe threshold range while ensuring grid-connected operation.

[0055] In one embodiment of this specification, when the preset wind condition type is a composite wind condition, the unit control rules include: Based on preset priority rules, multiple unit control rules are dynamically integrated; Among them, a composite wind condition is a wind condition in which at least two preset wind condition types exist simultaneously; Priority rules are ranked based on the degree of impact of wind conditions on unit safety and power generation efficiency, with the control rules that have the greatest impact on safety being executed first.

[0056] In this embodiment, composite wind conditions refer to wind environments where at least two preset wind condition types coexist. This can be achieved through multi-dimensional wind condition feature data fusion analysis, such as jointly modeling parameters like sudden wind speed changes, turbulence intensity, and wind direction deviation to identify the superimposed state of multiple wind conditions. Priority rules refer to the execution order established based on the degree of impact of wind conditions on unit operation safety and power generation efficiency. This can be implemented using a decision tree algorithm based on risk level assessment, for example, setting the wind condition type that causes mechanical overload as the highest priority. Dynamic integration refers to the process of coordinating the execution of multiple control rules based on real-time wind condition changes. This can be achieved using a multi-objective optimization algorithm, such as weighted fusion of various control commands under safety constraints.

[0057] Specifically, when a complex wind condition is detected, the feature extraction module first identifies the existing wind condition combination, such as the simultaneous presence of gales and turbulence. Then, a preset priority rule library is invoked; for example, the power limiting rule corresponding to gusts is set as the highest priority, and the power limiting rule corresponding to turbulence is set as the second highest priority. The control command generation module coordinates the output of each rule according to the priority order; for example, under gales conditions, the power upper limit constraint is executed first, while the dynamic power limiting command corresponding to turbulence conditions is superimposed. During implementation, if a safety-related parameter is detected to reach a critical threshold, the execution of low-priority rules is immediately terminated; for example, when the generator speed exceeds a safety threshold, control commands related to power generation efficiency optimization are suspended.

[0058] This embodiment establishes a dynamic integration mechanism to coordinate the interactions between multiple rules in real time. For example, when sudden wind changes and emergency crosswinds overlap, the yaw lock command is executed first, followed by pitch angle compensation control, ensuring both structural safety and maintaining power generation continuity. This technical solution solves the problems of strategy uniformity and execution conflicts inherent in traditional control methods under complex wind conditions, achieving collaborative control in scenarios with overlapping wind conditions. The dynamic integration mechanism of priority rules ensures that safety hazards are eliminated first under complex operating conditions while also considering power generation efficiency optimization needs, effectively improving the unit's operational reliability and control response accuracy under extreme weather conditions.

[0059] In one embodiment of this specification, when the preset wind condition type is a composite wind condition, the unit control rules include: When the current wind condition is identified as one of the preset wind condition types, but its matching degree is lower than the preset matching threshold, the system enters the fuzzy recognition mode. In fuzzy recognition mode, multiple unit control rules with similar wind conditions are invoked, and weights are assigned to each control rule based on the operating data to generate fused control commands.

[0060] In this embodiment, the fuzzy recognition mode refers to a multi-rule collaborative decision-making mechanism activated when the matching degree between wind condition features and preset types does not reach a deterministic threshold. Specifically, fuzzy logic algorithms can be used to calculate the membership degree of wind condition features, solving the problem of rigid control strategies caused by traditional binary judgment. Matching degree refers to a quantitative index of similarity between wind condition feature data and preset wind condition types, which can be calculated using Euclidean distance or cosine similarity algorithms, used to assess the degree of conformity between the current wind condition and the preset type. Weight allocation refers to the decision-making process of prioritizing multiple control rules based on the real-time operating status of the unit. Specifically, the analytic hierarchy process (AHP) combined with sensor data can be used to dynamically adjust the weight coefficients, achieving precise adaptation between the control strategy and the operating conditions.

[0061] Specifically, when the wind condition identification module detects that the matching degree between the current wind condition and the preset type is lower than a set threshold, the fuzzy logic operation unit is activated. This unit first calculates the membership values ​​of characteristic parameters such as current wind speed, wind direction, and turbulence intensity with various preset wind conditions, and filters out a set of candidate control rules with membership degrees higher than the critical value. Then, based on real-time operating parameters such as generator speed, pitch angle position, and grid dispatch instructions, a multi-objective optimization algorithm is used to determine the applicable weight of each candidate rule. Finally, the control parameters such as the power setpoint and pitch angle adjustment corresponding to each rule are linearly superimposed according to their weights to generate a composite control instruction that balances safety and power generation efficiency.

[0062] This embodiment effectively addresses the complex and compound wind conditions commonly encountered in wind farms by establishing a multi-rule collaborative decision-making mechanism. Existing technologies using fixed threshold-based control strategy switching methods are prone to control command oscillations, while the dynamic weight allocation mechanism employed in this solution smoothly transitions between different control rules, significantly improving the stability of unit operation. This technical solution effectively solves the technical problem of poor control strategy adaptability under complex wind conditions, automatically selecting the optimal combination of control rules when wind characteristics are unclear. Through the synergistic effect of fuzzy recognition and dynamic weight allocation, it avoids the limitations of a single control rule under complex operating conditions and prevents control conflicts caused by the parallel execution of multiple rules, thereby maximizing power generation efficiency while ensuring safe unit operation.

[0063] In one embodiment of this specification, multiple unit control rules with similar wind conditions are invoked, and weights are allocated to each control rule based on operating data to generate a fused control command, including: Calculate the fuzzy membership degree between the current wind condition feature data and the preset wind condition type, and select wind condition types with a fuzzy membership degree greater than or equal to the preset fuzzy threshold as candidate control rule sets; Based on the unit safety status, grid dispatching requirements, and historical success rate in the operational data, the applicable weight of each candidate control rule is determined; whereby the applicable weight reflects the degree of adaptation of the control rule to the current operational data, and the higher the weight value, the higher the priority of its adoption; The output control parameters of each control rule are weighted and fused to generate the final fused control command.

[0064] In this embodiment, fuzzy membership degree refers to a quantitative indicator of the degree of matching between current wind condition feature data and preset wind condition types. It can be implemented using fuzzy logic algorithms or similarity calculation models to filter candidate control rules with high correlation to the current wind condition. The candidate control rule set refers to a collection of multiple potentially applicable control rules filtered through fuzzy membership degree. It can be generated using threshold filtering or sorting selection methods to cover various control requirements that may exist under the current wind condition. Applicability weight refers to a parameter reflecting the priority of each candidate control rule in the current operating environment. It can be determined using a multi-factor weighted evaluation model, with the weight value dynamically adjusted based on the unit's safety status, grid dispatch requirements, and historical success rate to ensure the adaptability of the control strategy. Weighted fusion refers to the process of integrating the output parameters of different control rules according to their weights. It can be implemented using linear weighting or nonlinear interpolation algorithms to generate comprehensive control commands to balance multiple objective requirements.

[0065] Specifically, when the system detects that the matching degree between the current wind condition and the preset type is lower than a set threshold, the fuzzy recognition mode is activated. At this time, by calculating the fuzzy membership degree between wind condition feature data and each preset wind condition type, a set of candidate rules with matching degrees exceeding the minimum requirement is selected. Subsequently, based on real-time monitored unit safety status parameters, grid dispatch power demand, and historical control success rate data, each candidate rule is evaluated in multiple dimensions to determine its applicable weight. Finally, the control parameters corresponding to each rule (such as pitch angle setting, power limit, etc.) are fused and calculated according to their weights to generate a comprehensive control command that balances safety and power generation efficiency.

[0066] This embodiment effectively addresses control decision-making under boundary wind conditions by establishing a fuzzy recognition mechanism, enhancing control accuracy through multi-rule synergy. By incorporating multi-dimensional assessments such as unit safety status, grid demand, and historical success rate, the control strategy gains stronger environmental adaptability. This technical solution enables intelligent optimization of control strategies even in situations with fuzzy or mixed wind characteristics. This method effectively solves the problem of poor adaptability of traditional control methods under boundary conditions, reducing the risk of misjudgment through multi-rule fusion while ensuring the operational stability of the unit and grid dispatch response capabilities under complex operating conditions. It is particularly suitable for atypical wind conditions frequently encountered in wind farms, helping to extend equipment lifespan and improve power generation efficiency.

[0067] In one embodiment of this specification, when the current wind condition is identified as a transitional wind condition, the unit control rules include: Based on the trend of wind condition changes within a preset time window, the wind conditions are divided into multiple evolutionary stages. Based on the dynamic coupling relationship between different evolutionary stages and operational characteristic data, staged response rules are defined.

[0068] In this embodiment, transitional wind conditions refer to wind conditions where wind speed, direction, or turbulence intensity are continuously changing. This can be achieved by using time series analysis algorithms to identify trends in wind data, addressing the problem that traditional methods cannot effectively handle dynamically changing wind conditions. The evolutionary process stage refers to decomposing the continuously changing wind conditions into time periods with distinct characteristics. Specifically, a sliding window statistical method combined with clustering algorithms can be used to divide the wind condition change pattern into stages, enabling precise matching between control strategies and the wind condition evolution process. Dynamic coupling refers to the correlation characteristics between unit operating parameters and wind condition evolution stages. This can be achieved by establishing a correlation model using historical operating data, used to determine the adjustment priority of control strategies at different stages.

[0069] Specifically, when a transitional wind condition is detected, the system first collects a sequence of wind speed, wind direction angle, and turbulence intensity data within a preset time period. The direction of wind evolution is then determined by calculating the rate of change index. For example, if the standard deviation of the wind direction angle continues to increase and the rate of change of wind speed exceeds a threshold within a 10-second time window, it is determined that the system has entered a sudden wind direction evolution phase. Subsequently, based on the correlation model between each stage and the generator speed and pitch angle state, corresponding pitch angle adjustment rate limiting rules are generated. In the initial stage, gradual power regulation is used; when entering the mid-acceleration stage, it switches to rapid pitch control; and finally, in the stable stage, it reverts to the conventional control mode.

[0070] This embodiment dynamically divides the evolution phases and establishes phase response rules, enabling control parameters to adaptively adjust as wind conditions evolve, thus eliminating control lag. This technical solution provides a phased control strategy for the continuously changing characteristics of transitional wind conditions, avoiding mechanical shocks caused by sudden changes in control commands. Simultaneously, by dynamically matching phase characteristics with operating parameters, it reduces load fluctuations during wind condition evolution and improves operational stability.

[0071] In some other embodiments of this specification, feature extraction is performed on the preprocessed operational data and wind condition data to obtain operational feature data and wind condition feature data, including: Cluster analysis was performed on the preprocessed operational data and wind condition data to determine the centroid of each cluster. Operational features and wind condition features are extracted from the center points of each cluster to form operational feature data and wind condition feature data.

[0072] In this embodiment, the operational data includes parameters such as generator speed, blade pitch angle, and power output. Through cluster analysis, this data can be divided into different clusters, each representing a specific operational state mode. For example, one cluster might represent normal operation, while another might represent high-load operation. Wind condition data includes parameters such as wind speed, wind direction, and turbulence intensity. Through cluster analysis, this data can also be divided into different clusters, each representing a specific wind condition mode. For example, one cluster might represent steady wind conditions, while another might represent gusty wind conditions.

[0073] The center point of a cluster is a representative feature point of the data within that cluster, typically determined by calculating the mean of all data points within the cluster. This center point reflects the typical characteristics of the data within that cluster. Operational and wind condition features are extracted from each cluster center point to form operational and wind condition feature data.

[0074] Features are extracted from the central points of the operational data cluster, such as the mean generator speed, variance of pitch angle, and extreme values ​​of power output. These features characterize the typical features of this operational state mode. Similarly, features are extracted from the central points of the wind condition data cluster, such as the mean wind speed, rate of change of wind direction, and standard deviation of turbulence intensity. These features characterize the typical features of this wind condition mode. The extracted features are combined into feature vectors; the feature vectors of the operational data are called operational feature data, and the feature vectors of the wind condition data are called wind condition feature data.

[0075] This embodiment introduces cluster analysis to preprocess operational and wind data before extracting feature data, which improves the representativeness of feature extraction, enhances the accuracy of wind condition identification, improves the adaptability of control strategies, optimizes computational efficiency, and enhances the robustness of the system.

[0076] Based on the same general inventive concept, this invention also protects a wind turbine control device based on wind conditions, such as... Figure 2 As shown, Figure 2 This is a schematic diagram of the wind turbine control device based on wind conditions provided in an embodiment of the present invention. The wind turbine control device based on wind conditions provided by the present invention will be described below. The wind turbine control device described below can be referred to in correspondence with the wind turbine control method described above.

[0077] The wind condition-based wind turbine control device includes a data acquisition module 201, a preprocessing module 202, a feature extraction module 203, a wind condition recognition module 204, a rule generation module 205, and a turbine control module 206.

[0078] The data acquisition module 201 is used to acquire the operating data and wind condition data of the wind turbine. The preprocessing module 202 is used to preprocess the operating data and wind condition data; The feature extraction module 203 is used to extract features from the preprocessed operational data and wind condition data to obtain operational feature data and wind condition feature data. The wind condition recognition module 204 is used to identify the wind condition type represented by the wind condition feature data; The rule generation module 205 is used to select the corresponding unit control rule by combining the preset wind condition type and the operating characteristic data when the wind condition characteristic data indicates that the current wind condition is a preset wind condition type. The unit control module 206 is used to control the wind turbine based on the unit control rules.

[0079] Figure 3 This is a schematic diagram of the structure of the electronic device provided in an embodiment of the present invention.

[0080] like Figure 3As shown, the electronic device may include a processor 310, a communication interface 320, a memory 330, and a communication bus 340. The processor 310, communication interface 320, and memory 330 communicate with each other via the communication bus 340. The processor 310 can call logic instructions from the memory 330 to execute a wind turbine control method based on wind conditions.

[0081] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes 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 the present 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.

[0082] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the wind turbine control method based on wind conditions provided by the above methods.

[0083] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the wind condition-based wind turbine control methods provided by the methods described above.

[0084] The device embodiments described above are merely illustrative. 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 modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0085] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications 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.

Claims

1. A method of controlling a wind turbine based on wind conditions, characterized by, include: Acquire operating data and wind condition data of wind turbines; The operational data and the wind condition data are preprocessed; Feature extraction is performed on the preprocessed operational data and wind condition data to obtain operational feature data and wind condition feature data. Identify the wind condition type represented by the wind condition feature data; When the wind condition characteristic data indicates that the current wind condition is a preset wind condition type, the corresponding unit control rule is selected by combining the preset wind condition type and the operating characteristic data; wherein, the preset wind condition type includes single wind condition and compound wind condition; the single wind condition is any one of storm, gust, sudden wind direction change, turbulence, and transitional wind condition; Based on the aforementioned unit control rules, control the wind turbine unit; When the preset wind condition type is gust, the unit control rules include: Monitor generator speed and acceleration; If the acceleration continues to increase and exceeds a preset threshold, the pitch angle is limited to a preset minimum angle limit. The minimum angle limit is determined through load iterative optimization to accelerate the response speed of the pitch system. Wherein, when the preset wind condition type is a complex wind condition, the unit control rules include: Based on preset priority rules, multiple unit control rules are dynamically integrated; The composite wind condition refers to a wind condition in which at least two preset wind condition types exist simultaneously; The priority rules are ranked based on the degree of impact of wind conditions on unit safety and power generation efficiency, and the control rules with the greatest impact on safety are executed first. When the current wind condition is identified as one of the preset wind condition types, but its matching degree is lower than the preset matching threshold, the system enters the fuzzy recognition mode. In fuzzy recognition mode, the fuzzy membership degree between the current wind condition feature data and the preset wind condition type is calculated, and wind condition types with a fuzzy membership degree greater than or equal to the preset fuzzy threshold are selected as candidate control rule sets. Based on the unit safety status, grid dispatching requirements, and historical success rate in the operational data, the applicable weight of each candidate control rule is determined; wherein, the applicable weight reflects the degree of adaptation of the control rule to the current operational data, and the higher the weight value, the higher the priority of its adoption; The output control parameters of each control rule are weighted and fused to generate the final fused control command.

2. The wind turbine control method based on wind conditions according to claim 1, wherein, When the preset wind condition type is a storm, the unit control rules include: If the current wind speed is detected to be greater than the preset cut-out wind speed, the power limit of the wind turbine is restricted, and the power setting value is dynamically adjusted according to the pre-created wind speed power interpolation table. The wind speed-power interpolation table includes multiple sets of mapping relationships between wind speed values ​​and corresponding power limits, and the power limits decrease as the wind speed increases, so as to keep the unit in power generation operation.

3. The wind turbine control method based on wind conditions of claim 1, wherein, When the preset wind condition type is a sudden change in wind direction, the unit control rules include: Detect the current yaw deviation; If the yaw deviation exceeds the preset limit, the pitch angle is raised to the corresponding value according to the pre-created minimum pitch angle interpolation table for wind deviation. The interpolation table includes multiple sets of mapping relationships between yaw deviation values ​​and minimum pitch angle, and the pitch angle increases with the increase of deviation, so as to reduce the unbalanced load of the unit.

4. The wind turbine control method based on wind conditions according to claim 1, wherein, When the preset wind condition type is an emergency crosswind, the unit control rules include: If the generator speed of the wind turbine exceeds the safety threshold and the pitch system reports a fault signal, the yaw drive of the wind turbine will be controlled to perform a 90° crosswind action and the yaw state will be locked until manual reset.

5. The wind turbine control method based on wind conditions according to claim 1, wherein, When the preset wind condition type is turbulence, the unit control rules include: turbulence intensity is calculated based on a wind speed sequence of preset duration. The power limit of the wind turbine is dynamically adjusted based on the turbulence intensity and a pre-created turbulence power limit interpolation table. The turbulence power limit interpolation table includes multiple sets of mapping relationships between turbulence intensity values ​​and power limits, and the power limit decreases as the turbulence intensity increases, so as to reduce the fatigue load of the unit.

6. The wind turbine control method based on wind conditions according to claim 1, wherein, When the current wind condition is identified as a transitional wind condition, the unit control rules include: Based on the trend of wind condition changes within a preset time window, the wind conditions are divided into multiple evolutionary stages. Based on the dynamic coupling relationship between different evolutionary stages and operational characteristic data, staged response rules are defined.

Images