Wind power generation efficiency optimization method and system based on complex meteorological conditions

Abstract

The invention relates to the technical field of wind power generation efficiency optimization, and relates to a wind power generation efficiency optimization method and system based on complex meteorological conditions, and the method comprises the steps: obtaining the real-time meteorological data of a wind power plant and the operation parameters of a fan, and carrying out the turbulence intensity analysis based on the real-time meteorological data to obtain the turbulence intensity data; fatigue load prediction analysis is carried out on the turbulence intensity data according to fan operation parameters to obtain fatigue load prediction data, then power generation efficiency loss association mining is carried out based on the prediction data to obtain power generation efficiency loss association data, and finally control strategy optimization matching is carried out according to the association data. And fan control parameters are adjusted through the control strategy optimization matching data to obtain fan control parameter optimization data, so that the problem that a current fan control system lacks accurate perception of real-time turbulence load under a high turbulence working condition is effectively solved, and the risk of structural fatigue damage caused by high turbulence is effectively avoided; and the probability of part abrasion and unplanned shutdown is reduced.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of wind power generation efficiency optimization, and relates to a wind power generation efficiency optimization method and system based on complex weather conditions. BACKGROUND

[0002] Plateau or mountain wind farms have the characteristics of dramatic terrain undulations, significant wind shear effects and complex and variable local climate, and are a renewable energy engineering with high technical intensity and extremely high safety requirements. The prevailing wind field in the high-altitude area where such wind farms are located is often accompanied by strong turbulent fluctuations and high-frequency oscillations of wind direction. Once the wind turbine blades are fatigued and broken, the tower structure is damaged or the transmission chain fails due to extreme turbulence events, major safety accidents such as unit collapse, fire and secondary geological disasters may be caused, thus highlighting the extreme importance of real-time turbulent load sensing and predictive control optimization.

[0003] The current wind turbine control strategy under turbulent conditions of plateau and mountain wind farms still needs to be systematically improved, which is embodied in the following key aspects: Firstly, the signal processing architecture of the existing control system does not conduct in-depth analysis from the perspective of multi-physical field coupling, lacks the ability to synchronously capture the cross-component related characteristics such as sudden change of blade root bending moment, acceleration response overrun at the top of the tower and generator torque oscillation caused by turbulence, is easy to cause the perception deviation between turbulence intensity evaluation and real structure load, increases the risk of rapid accumulation of fatigue life of key components caused by load misjudgment, cannot identify the limit load working condition induced by turbulence in the early stage, causes the unit to run in the non-design safety margin interval for a long time, and brings irreversible negative effects on the reliability of the whole farm power generation and the value preservation and appreciation of equipment assets.

[0004] Secondly, the current mainstream control logic does not construct a forward-looking prediction mechanism of the spatio-temporal evolution law of turbulence, cannot predict the time, intensity and duration of the arrival of turbulence group based on the advance measurement information of the upstream wind measurement tower or the cabin radar, is difficult to realize the coordinated adjustment of the active wind of the yaw system, the dynamic optimization of the variable pitch rate and the intelligent correction of the torque given under the premise of ensuring the safety of the structure, cannot complete the preventive power enhancement before the attack of turbulence, cannot restore the optimal power generation state in time during the turbulence decay period, finally misses a large number of power generation opportunities that can be captured, directly leads to the significant reduction of the annual equivalent full-load hours of the whole farm, cannot effectively reduce the direct economic loss of green energy caused by excessive conservative control strategy, and cannot guarantee the scientific balance between the maximization of power generation benefit and the safety of operation of the wind farm under complex terrain. SUMMARY

[0005] In view of the problems existing in the prior art, the present application provides a wind power generation efficiency optimization method and system based on complex weather conditions, which solves the above technical problems.

[0006] To achieve the above and other objectives, the technical solution adopted by the present invention is as follows: The first aspect of this invention provides a method for optimizing wind power generation efficiency based on complex meteorological conditions, the method comprising: Step S1: Obtain real-time meteorological data and wind turbine operating parameters of the wind farm. The real-time meteorological data of the wind farm includes wind speed, wind direction, temperature, air pressure and turbulence intensity measurements. The wind turbine operating parameters include wind turbine blade angle, generator speed, power output and controller settings. Perform turbulence intensity analysis based on the real-time meteorological data of the wind farm to obtain turbulence intensity data. Step S2: Perform fatigue load prediction analysis on the turbulence intensity data based on the wind turbine operating parameters to obtain fatigue load prediction data. The fatigue load prediction analysis includes using a wind turbine dynamics model to simulate the stress distribution and cumulative fatigue damage of the wind turbine blades, tower, and transmission chain under high turbulence conditions. Based on the fatigue load prediction data, perform power generation efficiency loss correlation mining to obtain power generation efficiency loss correlation data. The power generation efficiency loss correlation mining includes analyzing the power generation loss caused by the conservatism of the control strategy when the fatigue load exceeds the limit. Step S3: Based on the power generation efficiency loss correlation data, optimize and match the control strategy to generate control strategy optimization matching data. The control strategy optimization matching includes dynamically adjusting the wind turbine blade angle and generator speed to maximize power generation efficiency while ensuring that the fatigue load does not exceed the limit. Adjust the wind turbine control parameters through the control strategy optimization matching data to obtain wind turbine control parameter optimization data, thereby realizing the optimization of wind power generation efficiency under complex weather conditions.

[0007] A second aspect of the present invention provides a wind power generation efficiency optimization system based on complex meteorological conditions, the system comprising: Turbulence intensity analysis module: Acquires real-time meteorological data and wind turbine operating parameters of the wind farm. The real-time meteorological data of the wind farm includes wind speed, wind direction, temperature, air pressure and turbulence intensity measurements. The wind turbine operating parameters include wind turbine blade angle, generator speed, power output and controller setpoints. Based on the real-time meteorological data of the wind farm, turbulence intensity analysis is performed to obtain turbulence intensity data. Efficiency loss calculation module: Based on the wind turbine operating parameters, fatigue load prediction analysis is performed on the turbulence intensity data to obtain fatigue load prediction data; based on the fatigue load prediction data, power generation efficiency loss correlation mining is performed to obtain power generation efficiency loss correlation data, which includes analyzing the power generation loss caused by the conservatism of the control strategy when the fatigue load exceeds the limit. Wind turbine parameter optimization module: Based on the correlation data of power generation efficiency loss, the module optimizes and matches the control strategy to generate control strategy optimization matching data. The control strategy optimization matching includes dynamically adjusting the wind turbine blade angle and generator speed to maximize power generation efficiency while ensuring that fatigue load does not exceed the limit. The module adjusts the wind turbine control parameters through the control strategy optimization matching data to obtain the wind turbine control parameter optimization data.

[0008] As described above, the wind power generation efficiency optimization method and system based on complex meteorological conditions provided by this invention have at least the following beneficial effects: The wind power generation efficiency optimization method and system based on complex meteorological conditions provided by this invention acquires real-time meteorological data and wind turbine operating parameters from the wind farm. Based on the real-time meteorological data, turbulence intensity analysis is performed to obtain turbulence intensity data. Then, fatigue load prediction analysis is performed on the turbulence intensity data according to the wind turbine operating parameters to obtain fatigue load prediction data. Subsequently, based on the prediction data, power generation efficiency loss correlation mining is performed to obtain power generation efficiency loss correlation data. Finally, control strategy optimization and matching are performed based on the correlation data, and wind turbine control parameters are adjusted through the control strategy optimization and matching data to obtain wind turbine control parameter optimization data. This effectively solves the problem that current wind turbine control systems lack accurate perception and forward-looking prediction of real-time turbulence loads under high turbulence conditions, and are unable to make a dynamic optimal trade-off between "ensuring safety" and "maximizing power generation". On the one hand, by integrating multi-dimensional real-time data with dynamic model simulation, precise analysis of turbulent loads from perception to prediction was achieved, effectively avoiding the risk of structural fatigue damage caused by high turbulence, reducing component wear and the probability of unplanned downtime, further reducing maintenance costs, and ensuring the safe and stable operation of the wind turbine structure. On the other hand, through dynamic optimization and matching of control strategies, a dynamic balance between the fatigue load safety threshold and power generation efficiency was achieved, which not only avoids the waste of wind energy resources caused by conservative control and significantly increases the annual power generation under complex meteorological conditions, but also effectively reduces the economic losses caused by efficiency loss, ensuring the accuracy and economy of wind farm operation analysis. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram showing the connections between the steps of the method of the present invention.

[0011] Figure 2 This is a schematic diagram showing the connections of the various modules in the system of the present invention. Detailed Implementation

[0012] The following description, in conjunction with the implementation of this invention, is merely an example and illustration of the concept of this invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the inventive concept or exceed the scope defined in these claims, all of which should fall within the protection scope of this invention. Example

[0013] Please see Figure 1 As shown, a method for optimizing wind power generation efficiency under complex meteorological conditions is proposed. This method includes the following steps: Step S1: Obtain real-time meteorological data and wind turbine operating parameters of the wind farm. The real-time meteorological data of the wind farm includes wind speed, wind direction, temperature, air pressure and turbulence intensity measurements. The wind turbine operating parameters include wind turbine blade angle, generator speed, power output and controller settings. Perform turbulence intensity analysis based on the real-time meteorological data of the wind farm to obtain turbulence intensity data.

[0014] For example, step S1 includes: Step S11: Based on the wind speed and wind direction data in the real-time meteorological data of the wind farm, wind vector synthesis is performed to obtain wind vector data; Step S12: Perform wind vector fluctuation analysis based on wind vector data to obtain wind vector fluctuation data; Step S13: Calculate the environmental disturbance intensity based on wind vector fluctuation data and temperature and air pressure data to obtain environmental disturbance intensity data; Step S14: Perform turbulence intensity correction based on environmental disturbance intensity data and turbulence intensity measurements to obtain turbulence intensity data.

[0015] For example, step S11 includes the following steps: Step S111: Based on wind speed and wind direction data, perform wind direction coordinate system transformation to obtain the east-west wind speed component and the north-south wind speed component; Step S112: Perform vector synthesis calculation based on the east-west wind speed components and the north-south wind speed components to obtain the synthesized wind vector data; Step S113: Extract instantaneous fluctuation features from the synthetic wind vector data to obtain instantaneous wind vector fluctuation data; Step S114: Evaluate the stability of wind vector fluctuations based on instantaneous wind vector fluctuation data to obtain wind vector fluctuation stability data; Step S115: Perform wind vector data calibration based on synthetic wind vector data and wind vector fluctuation stability data to obtain wind vector data.

[0016] In one specific embodiment, when performing wind direction coordinate system transformation on the collected wind speed and wind direction data, the wind direction angle data is first converted from polar coordinates to standard meteorological angle measurement. Then, for the wind speed value at each observation time, according to the direction indicated by the wind direction angle, the wind speed is decomposed into east-west projected components using the sine function calculation principle, that is, the wind speed value multiplied by the sine of the wind direction angle to obtain the east-west component. At the same time, the wind speed is decomposed into north-south projected components using the cosine function calculation principle, that is, the wind speed value multiplied by the cosine of the wind direction angle to obtain the north-south component. This transformation process is repeated for each sampling point to generate the east-west wind speed component time series and the north-south wind speed component time series with the same time resolution as the original data.

[0017] When performing vector synthesis calculations, the east-west and north-south components at the same moment are treated as the two legs of a right triangle. The length of the hypotenuse is calculated using the Pythagorean theorem. Specifically, the squares of the east-west and north-south components are added together, and the square root of the sum is taken to obtain the magnitude of the composite wind speed at that moment. At the same time, the arctangent function is used to determine the composite wind direction angle. Specifically, the east-west component is divided by the north-south component to obtain the ratio, and then the arctangent operation is performed on the ratio. The quadrant is determined based on the sign of the north-south component, and then converted into a standard angle value. Finally, a vector data sequence containing composite wind speed and composite wind direction is formed. When extracting instantaneous fluctuation features from synthetic wind vector data, the average value of synthetic wind speed over the entire observation period is first calculated as the reference wind. The average value is then subtracted from the synthetic wind speed at each moment to obtain the wind speed fluctuation. Simultaneously, the average angle of synthetic wind direction over the entire observation period is calculated as the reference direction. The average angle is then subtracted from the synthetic wind direction angle at each moment and angle normalization is performed to obtain the wind direction fluctuation. Finally, the wind speed fluctuation and wind direction fluctuation are combined to form the instantaneous fluctuation data of the wind vector.

[0018] When assessing fluctuation stability, instantaneous fluctuation data is divided into several continuous time segments of fixed duration. The variance of wind speed fluctuations within each segment is calculated to reflect the magnitude of the fluctuation energy, and the standard deviation of wind direction fluctuations within each segment is calculated to reflect the amplitude of directional oscillation. Simultaneously, the rate of change of variance between adjacent segments is calculated. The fluctuation energy change index is obtained by dividing the difference between the variance of the current segment and the variance of the previous segment by the variance of the previous segment. The variance, standard deviation, and rate of change are weighted and combined to generate fluctuation stability data. During wind vector data calibration, the synthetic wind vector data is time-aligned with the fluctuation stability data. When the stability data shows that the fluctuation energy exceeds a preset threshold, it is determined that the moment is affected by gusty winds and requires smoothing correction. Specifically, the arithmetic mean of the synthetic wind speed at that moment and its adjacent moments is taken as the corrected wind speed value, and the vector average of the synthetic wind direction angle at that moment and its adjacent moments is calculated as the corrected wind direction value. Finally, the calibrated wind vector data is output.

[0019] For example, step S12 includes the following steps: Step S121: Perform wind vector change trend analysis on the wind vector data to obtain wind vector change trend data; Step S122: Calculate the fluctuation slope based on the wind vector change trend data to obtain the wind vector fluctuation slope data; Step S123: Identify extreme points of wind vector fluctuations based on the wind vector fluctuation slope data to obtain extreme point data of wind vector fluctuations; Step S124: Based on the extreme point data and slope data of wind vector fluctuation, perform fluctuation intensity analysis to obtain wind vector fluctuation data.

[0020] In one specific embodiment, when analyzing the wind vector change trend of the calibrated wind vector data, the wind vector data is arranged in chronological order to form a sequence. A moving window linear regression method is used, in which a window of a set width is slid across the sequence, and a straight line is fitted to the data points within each window. The least squares principle is used to minimize the sum of the squares of the vertical distances from the straight line to each data point. The slope value of the straight line is calculated as the trend strength at the center of the window. A positive slope value indicates an upward trend, and a negative slope value indicates a downward trend. The intercept value of the straight line is calculated as the trend benchmark. At the same time, the goodness-of-fit value is calculated to reflect the reliability of the trend. The slope value, intercept value, and goodness-of-fit value are arranged in chronological order to generate wind vector change trend data.

[0021] When calculating the fluctuation slope, a first-order difference operation is performed on the slope value sequence in the trend data. The slope value of the next moment is subtracted from the slope value of the previous moment to obtain the slope change. This change reflects the trend acceleration. Then, a second-order difference operation is performed on the slope change sequence. The slope change of the next moment is subtracted from the slope change of the previous moment to obtain the slope curvature. This curvature reflects the severity of the trend change. At the same time, the instantaneous fluctuation slope at each moment is calculated. The slope value of that moment is subtracted from the average slope value of the entire period to obtain the relative slope offset. The relative slope offset is divided by the average slope value to obtain the normalized fluctuation slope. The acceleration measure, curvature value and normalized slope are integrated to form the wind vector fluctuation slope data.

[0022] When identifying extreme points of fluctuations, a sliding window peak detection method is used for the fluctuation slope data. The detection window width is set to an odd number of points. The slope value of the center point of the window is compared with the slope values ​​of all other points in the window one by one. When the value of the center point is greater than the value of each other point, it is marked as a candidate point for a maximum value. When the value of the center point is less than the value of each other point, it is marked as a candidate point for a minimum value. To avoid misjudgment, a continuous condition test is introduced, requiring that the slope values ​​of the two points before and after each candidate point change monotonically. At the same time, the time distance between adjacent extreme points is calculated as the fluctuation half-cycle. The extreme point type, amplitude, and occurrence time that meet the conditions are recorded as wind vector fluctuation extreme point data.

[0023] When analyzing the fluctuation intensity, for each identified fluctuation cycle, the wind speed values ​​at its initial and final extreme points are extracted, and the difference between the two wind speed values ​​is calculated as the wind speed fluctuation amplitude. The maximum range of wind direction angle change within the cycle is extracted as the wind direction fluctuation amplitude. The wind speed sequence is converted to the frequency domain using the Fast Fourier Transform principle, and the frequency component with the largest amplitude is identified as the main fluctuation frequency. The power spectral density value corresponding to this frequency is calculated as the frequency domain energy intensity. After standardizing the wind speed amplitude, wind direction amplitude, main frequency value, and frequency domain energy, they are weighted and summed according to preset weight coefficients to generate a comprehensive fluctuation intensity index. At the same time, the spatial continuity of the fluctuation intensity is calculated, and the intensity value of the current monitoring point is spatially averaged with the intensity values ​​of the surrounding adjacent monitoring points to obtain the smoothed fluctuation intensity. Finally, all parameters are integrated to form wind vector fluctuation data.

[0024] For example, step S13 includes the following steps: Step S131: Perform temperature-pressure coupling effect analysis based on temperature and air pressure data to obtain temperature-pressure coupling effect data; Step S132: Evaluate the disturbance impact effect based on temperature-pressure coupling effect data and wind vector fluctuation data to obtain disturbance impact effect data; Step S133: Analyze the environmental disturbance intensity based on the disturbance impact effect data to obtain the environmental disturbance intensity data.

[0025] In one specific embodiment, when analyzing the temperature-pressure coupling effect based on temperature and pressure data, the temperature and pressure sequences are time-aligned. The product of the temperature and pressure values ​​at each moment is calculated to obtain the temperature-pressure co-variance. The average of the temperature-pressure co-variance over all moments is used to obtain the temperature-pressure co-variance baseline. The temperature-pressure co-variance at each moment is subtracted from the baseline to obtain the temperature-pressure co-variance anomaly. The Pearson correlation coefficient between the temperature and pressure sequences is calculated. First, the covariance between the temperature and pressure values ​​is calculated, which is the average of the products of the temperature anomaly and the pressure anomaly at each moment. Then, the temperature sequence... The standard deviations of the temperature and pressure sequences are calculated, and the covariance is divided by the product of the two standard deviations to obtain the correlation coefficient. A covariance matrix of the temperature and pressure data is constructed, and eigenvalue decomposition is performed on the matrix. The eigenvector corresponding to the largest eigenvalue is extracted as the main mode of temperature-pressure coupling. The instantaneous rate of change of the temperature and pressure sequences is calculated, and the difference is obtained by subtracting the previous time value from the value of the next time step. The difference is divided by the time interval to obtain the rate of change. The ratio of the rate of change of temperature to the rate of change of pressure is used as the temperature-pressure coupling sensitivity. The correlation coefficient, the eigenvector of the main mode, and the sensitivity parameter are integrated into the temperature-pressure coupling effect data.

[0026] When assessing the impact of disturbances based on temperature-pressure coupling effect data and wind vector fluctuation data, the characteristic vector of the temperature-pressure coupling principal mode is vector-synthesized with the wind vector fluctuation data. The east-west components of the two vectors are added to obtain the synthesized east-west component, and the north-south components of the two vectors are added to obtain the synthesized north-south component. The Pythagorean theorem is used to calculate the magnitude of the synthesized vector. When the synthesized north-south component is not zero, the ratio of the synthesized east-west component to the synthesized north-south component is calculated and then the arctangent is performed to obtain the synthesized direction. The synthesized magnitude is divided by the time window length to obtain the disturbance propagation rate. The distance attenuation model is used to assess the impact range. With the disturbance source as the center, the distance between the target point and the disturbance source is calculated. The disturbance intensity is multiplied by the negative exponential function of the distance value to obtain the attenuated intensity. The synthesized direction, propagation rate, and attenuated intensity are integrated into the disturbance impact effect data. When analyzing the intensity of environmental disturbances based on disturbance impact data, the time series of disturbance intensity is extracted, and its arithmetic mean is calculated as the baseline intensity. The ratio of its standard deviation to the mean is calculated to obtain the coefficient of variation, which reflects the stability of the intensity. A first threshold and a second threshold are set to classify the intensity into levels. Intensity values ​​below the first threshold are classified as weak disturbances, intensity values ​​between the first and second thresholds are classified as moderate disturbances, and intensity values ​​above the second threshold are classified as strong disturbances. The spatial distribution gradient of the disturbance intensity is calculated, the difference between the intensity values ​​of adjacent monitoring points is calculated, and the difference is divided by the distance between the two points to obtain the gradient value. The classification results, baseline intensity, coefficient of variation, and gradient value are integrated to generate environmental disturbance intensity data.

[0027] For example, step S14 includes the following steps: Step S141: Perform disturbance-measurement deviation analysis based on environmental disturbance intensity data and turbulence intensity measurements to obtain disturbance-measurement deviation data; Step S142: Analyze the correction factor based on the disturbance-measurement bias data to obtain the turbulence correction factor; Step S143: Correct the turbulence intensity measurement value based on the turbulence correction factor to obtain turbulence intensity data.

[0028] In one specific embodiment, when performing disturbance-measurement deviation analysis based on environmental disturbance intensity data and turbulence intensity measurements, the environmental disturbance intensity data and turbulence intensity measurements are aligned with the same time reference. The difference between the environmental disturbance intensity and the turbulence intensity measurements at each moment is calculated to obtain the absolute deviation. The absolute deviation is divided by the environmental disturbance intensity to obtain the relative deviation rate. For the deviation at each moment, the average deviation of the deviation at the adjacent moments before and after it is calculated as the local deviation reference. The deviation fluctuation is obtained by subtracting the local deviation reference from the current deviation. The deviation change rate is obtained by dividing the deviation fluctuation by the sampling time interval. The absolute deviation, relative deviation, deviation fluctuation, and change rate are integrated to form disturbance-measurement deviation data. When analyzing correction factors based on disturbance-measurement deviation data, the deviation data is used as the dependent variable and the environmental disturbance intensity data is used as the independent variable. A model of the relationship between deviation and disturbance intensity is established using multiple linear regression. The least squares principle is used to minimize the sum of squares of the differences between the model prediction value and the actual deviation value. The model coefficients are then solved and normalized to map their range to the interval between zero and one. The normalized coefficients are used as the basic correction factor. At the same time, the cumulative sum of deviations at each time step is calculated. When the cumulative sum exceeds a preset limit, a dynamic adjustment mechanism for the correction factor is triggered. The basic correction factor is multiplied by a dynamic gain coefficient to obtain the real-time correction factor. The basic factor and the real-time factor are then integrated into a turbulence correction factor. When calculating the correction of turbulence intensity measurements based on the turbulence correction factor, the turbulence intensity measurement at each moment is multiplied by the corresponding real-time correction factor to obtain a preliminary correction value. Each preliminary correction value is then judged to see if it exceeds the physically reasonable range. If the correction value is greater than the theoretical maximum value, it is restricted to the theoretical maximum value; if the correction value is less than the theoretical minimum value, it is restricted to the theoretical minimum value. The value after the range restriction is taken as the final correction value. The final correction values ​​at all moments are then smoothed over time. The moving average method is used to calculate the arithmetic mean of the correction values ​​at the current moment and the two moments before, resulting in the smoothed turbulence intensity data.

[0029] Step S2: Perform fatigue load prediction analysis on the turbulence intensity data based on the wind turbine operating parameters to obtain fatigue load prediction data. The fatigue load prediction analysis includes using a wind turbine dynamics model to simulate the stress distribution and cumulative fatigue damage of the wind turbine blades, tower, and transmission chain under high turbulence conditions. Based on the fatigue load prediction data, perform power generation efficiency loss correlation mining to obtain power generation efficiency loss correlation data. The power generation efficiency loss correlation mining includes analyzing the power generation loss caused by the conservatism of the control strategy when the fatigue load exceeds the limit.

[0030] For example, the operational logic of step S2 is as follows: Step S21: Perform load characteristic analysis on the turbulence intensity data based on the wind turbine operating parameters to obtain load characteristic data. The load characteristic analysis includes calculating the dynamic load change trend of wind turbine components under turbulence based on wind speed, wind direction, temperature, air pressure and turbulence intensity measurements, as well as wind turbine blade angle, generator speed, power output and controller settings. Step S22: Based on the load characteristic data, use the wind turbine dynamics model to simulate the stress distribution of the wind turbine blades, tower and transmission chain under high turbulence conditions to obtain stress distribution data. The stress distribution simulation includes solving the component stress response equation using the wind turbine dynamics model to obtain the spatiotemporal distribution of stress. Step S23: Calculate fatigue cumulative damage based on stress distribution data to obtain fatigue load prediction data. The fatigue cumulative damage calculation includes applying a fatigue damage accumulation model based on stress distribution data to evaluate the degree of damage to the wind turbine components within a preset time range. Step S24: Based on fatigue load prediction data, perform power generation loss analysis caused by the conservatism of the control strategy to obtain power generation efficiency loss correlation data. The power generation loss analysis includes comparing the difference between the adjustment of the controller setpoint and the ideal control strategy when the fatigue load exceeds the limit, and quantifying the power generation loss value.

[0031] For example, step S21 includes the following steps: The initial impact analysis of turbulence on wind turbine load is conducted based on measurements of wind speed, wind direction, temperature, air pressure, and turbulence intensity. The initial impact data of turbulence load are obtained. The initial impact analysis includes calculating the correlation between turbulence intensity and wind turbine foundation load based on meteorological data. Based on the initial turbulent load influence data and the wind turbine blade angle, generator speed, power output and controller setpoints, the dynamic load response of wind turbine components is calculated to obtain the dynamic load response data of the components. The dynamic load response calculation includes simulating the instantaneous load changes of the blades, tower and drive train under turbulent action by combining wind turbine operating parameters. Load change trend is extracted from the dynamic load response data of the component to obtain the dynamic load change trend as load feature data. The change trend extraction includes analyzing the change law of load response data over time to determine peak value and fluctuation characteristics.

[0032] In this embodiment of the invention, the initial impact of turbulence on wind turbine load is analyzed based on measurements of wind speed, wind direction, temperature, air pressure, and turbulence intensity, yielding initial turbulence load impact data. The initial impact analysis integrates meteorological parameters to calculate the correlation between turbulence intensity and wind turbine foundation load. Specifically, the process involves first calculating the ratio of the standard deviation of wind speed fluctuation to the average wind speed using wind speed measurements to determine turbulence intensity, and then correcting for air density using temperature and air pressure data. This establishes a quantitative relationship model between turbulence intensity and wind turbine tower foundation loads such as thrust and bending moment. A regression analysis method is used to fit the impact curve of turbulence intensity changes on foundation load, obtaining initial impact data characterizing load sensitivity. Next, based on the initial turbulence load impact data and wind turbine blade angle, generator speed, power output, and controller settings, dynamic load response calculations for wind turbine components are performed, yielding component dynamic load response data. The dynamic load response calculation employs a multibody dynamics simulation method, incorporating the initial turbulence load... Using the influencing data as input boundary conditions, and combining it with the wind turbine operating parameters, the instantaneous load changes of the blades, tower, and drive train under turbulent conditions are simulated. Specifically, an aeroelastic model of the wind turbine is constructed, the aerodynamic load distribution is calculated based on the blade angle and generator speed, and the wind turbine operating state is adjusted using controller setpoints. Simultaneously, finite element analysis is applied to solve the stress response of the tower and drive train in the dynamic wind field, thus outputting a time-varying sequence of component loads. Finally, the dynamic load response data of the components is analyzed to extract the load variation trend, which is then used as load characteristic data. The trend extraction employs time series analysis, using a sliding window technique to calculate the moving average and standard deviation of the load data, identifying load peaks and fluctuation characteristics. Specifically, the variation law of the load response data over time is analyzed, peak detection algorithms are used to locate load extreme points, and fluctuation frequency analysis is combined to determine the periodicity of the load, thereby extracting trend indicators characterizing the dynamic behavior of the load.

[0033] For example, step S22 includes the following steps: Load condition mapping is performed on wind turbine blades, tower and drive train based on load characteristic data to obtain model load input data. Load condition mapping includes determining the input load parameters required for wind turbine dynamic model according to dynamic load change trend. Based on the model load input data, the stress response equation is solved using the wind turbine dynamics model to obtain the component stress response data. The stress response equation solution includes calculating the instantaneous stress values ​​of each component of the wind turbine under high turbulence conditions. Spatiotemporal distribution features are extracted from the stress response data of the component to obtain stress distribution data. The spatiotemporal distribution feature extraction includes integrating the changes of instantaneous stress values ​​with time and plane to generate the spatiotemporal distribution of stress.

[0034] For example, step S23 includes the following steps: Stress time history features are extracted from stress distribution data to obtain stress cycle feature data. The stress time history feature extraction includes analyzing the amplitude and number of cycles of stress change of each component over time in the stress distribution data. Single-cycle damage calculation is performed based on stress cycle characteristic data and wind turbine component material properties to obtain single-cycle damage data. The single-cycle damage calculation includes determining the amount of damage caused by each stress cycle based on the stress amplitude and the material SN curve. Damage accumulation calculation is performed on the single-cycle damage data within a preset time range to obtain cumulative damage data. The damage accumulation calculation includes adding the single-cycle damage data within the preset time range using a linear accumulation rule. Fatigue life prediction of wind turbine components is performed based on cumulative damage data to obtain fatigue load prediction data. The fatigue life prediction includes assessing the degree of fatigue damage of each component within a preset time range based on the cumulative damage data.

[0035] For example, step S24 includes the following steps: Fatigue load over-limit event identification is performed based on fatigue load prediction data to obtain over-limit event data. The fatigue load over-limit event identification includes determining the operating time period during which fatigue damage to wind turbine components exceeds a preset safety threshold based on fatigue load prediction data. Based on the load over-limit event data, the actual controller setpoint adjustment is extracted to obtain the actual setpoint adjustment data. The actual controller setpoint adjustment extraction includes recording the actual modification parameters of the controller setpoint during the load over-limit event. The ideal controller setpoint is calculated based on the load over-limit event data to obtain the ideal setpoint data. The ideal controller setpoint calculation includes simulating the controller setpoint that maximizes power generation efficiency under the same load conditions without considering fatigue constraints. Based on the actual setpoint adjustment data and the ideal setpoint data, a setpoint difference analysis is performed to obtain setpoint difference data. The setpoint difference analysis includes calculating the numerical deviation between the actual setpoint adjustment data and the ideal setpoint data. The power generation loss value is quantified based on the setpoint difference data to obtain the power generation efficiency loss correlation data. The power generation loss value quantification includes evaluating the power generation loss value caused by the conservatism of the control strategy based on the setpoint difference data.

[0036] In this embodiment of the invention, a threshold comparison method is used to compare the component damage degree in the fatigue load prediction data with a preset safety threshold in real time. Time series analysis is used to determine the specific operating time periods during which the fatigue damage of wind turbine components exceeds the safety limit, and the start and duration of these time periods are recorded to obtain load over-limit event data. Subsequently, based on the load over-limit event data, actual controller setpoint adjustments are extracted. By querying the historical operation logs of the wind turbine control system, the actual modification parameters of the controller to the wind turbine blade angle and generator speed during the identified over-limit events are extracted, including the adjustment magnitude and adjustment time, to obtain actual setpoint adjustment data. Simultaneously, based on the load over-limit event data, ideal controller setpoints are calculated using a power generation efficiency maximization model under the same load conditions. The simulation of wind turbine operation without considering fatigue constraints calculates the optimal values ​​of turbine blade angle and generator speed using optimization algorithms to maximize power generation efficiency, thus obtaining ideal setpoint data. Next, based on actual setpoint adjustment data and ideal setpoint data, a setpoint difference analysis is performed. A numerical deviation calculation method is applied to calculate the differences between the actual and ideal setpoint data in parameter values ​​and adjustment timing, including absolute and relative deviations, thus obtaining setpoint difference data. Finally, based on the setpoint difference data, power generation loss is quantified. By establishing a correlation model between setpoint deviation and power generation loss, combined with the wind turbine power output characteristic curve, the power generation loss due to the conservative control strategy is evaluated, and the total loss is calculated by integration, thus obtaining power generation efficiency loss correlation data.

[0037] Step S3: Based on the power generation efficiency loss correlation data, optimize and match the control strategy to generate control strategy optimization matching data. The control strategy optimization matching includes dynamically adjusting the wind turbine blade angle and generator speed to maximize power generation efficiency while ensuring that the fatigue load does not exceed the limit. Adjust the wind turbine control parameters through the control strategy optimization matching data to obtain wind turbine control parameter optimization data, thereby realizing the optimization of wind power generation efficiency under complex weather conditions.

[0038] For example, step S3 includes the following steps: Step S31: Based on the power generation efficiency loss correlation data and fatigue load prediction data, optimize and match the control strategy to generate control strategy optimization matching data. The control strategy optimization matching includes analyzing the power generation efficiency loss correlation data to determine the optimization adjustment scheme of the wind turbine blade angle and generator speed, and maximizing power generation efficiency while ensuring that the fatigue load does not exceed the limit under the constraint of the fatigue load prediction data. Step S32: Adjust the wind turbine control parameters by optimizing the matching data through the control strategy to obtain optimized wind turbine control parameter data. The wind turbine control parameter adjustment includes setting the control parameter values ​​of the wind turbine blade angle and generator speed according to the optimized matching data of the control strategy.

[0039] For example, step S31 includes the following steps: Step S311: Identify key parameters of efficiency loss based on power generation efficiency loss correlation data to obtain key parameter data of efficiency loss. The identification of key parameters of efficiency loss includes analyzing power generation efficiency loss correlation data to determine the main wind turbine operating parameters that cause power generation loss. Step S312: Based on the key parameter data of efficiency loss and fatigue load prediction data, perform control strategy constraint analysis to obtain control strategy constraint data. The control strategy constraint analysis includes evaluating the adjustment range of wind turbine blade angle and generator speed under the premise that the fatigue load does not exceed the limit, in combination with fatigue load prediction data. Step S313: Calculate the wind turbine parameter optimization scheme based on the control strategy constraint data to obtain wind turbine parameter optimization scheme data. The wind turbine parameter optimization scheme calculation includes determining the optimized values ​​of wind turbine blade angle and generator speed under the constraint conditions through an iterative optimization algorithm to maximize power generation efficiency. Step S314: Based on the wind turbine parameter optimization scheme data, perform control strategy matching and integration to generate control strategy optimization matching data, wherein the control strategy matching and integration includes integrating the optimized wind turbine parameters into an executable control strategy scheme.

[0040] In this embodiment of the invention, a multivariate correlation analysis method is used to analyze the correlation between operating parameters such as wind turbine blade angle and generator speed and power generation loss in the power generation efficiency loss correlation data. By calculating and ranking the contribution weight of each parameter to efficiency loss, the main wind turbine operating parameters causing power generation loss are screened out, thereby obtaining key parameter data for efficiency loss. Subsequently, based on the key parameter data for efficiency loss and fatigue load prediction data, control strategy constraint analysis is performed. Constraint optimization theory is used to evaluate the adjustment range of wind turbine blade angle and generator speed under the premise that fatigue load does not exceed the limit. By establishing a mapping relationship between load safety threshold and parameter adjustment, and combining the damage degree information in the fatigue load prediction data, the allowable fluctuation range of each operating parameter is calculated. The process involves several steps: First, control strategy constraint data is obtained. Second, based on this data, wind turbine parameter optimization schemes are calculated. An iterative optimization algorithm is applied to search for the optimal values ​​of wind turbine blade angle and generator speed under the constraints. By simulating the power generation efficiency under different parameter combinations and comparing the output results, the parameters are gradually adjusted until a solution that maximizes power generation efficiency is found, thus obtaining wind turbine parameter optimization scheme data. Finally, control strategy matching and integration are performed based on the wind turbine parameter optimization scheme data. A strategy fusion method is used to convert the optimized wind turbine parameters into actual executable control commands, including setting parameter adjustment sequences and switching logic, to ensure the smooth transition and effective execution of the control strategy under different operating conditions, thereby generating control strategy optimization matching data.

[0041] For example, step S32 includes the following steps: Step S321: Based on the control strategy optimization matching data, analyze the wind turbine control parameters to obtain control parameter analysis data, wherein the wind turbine control parameter analysis includes extracting the optimized setpoints of wind turbine blade angle and generator speed from the control strategy optimization matching data; Step S322: Perform control parameter feasibility verification based on control parameter parsing data to obtain control parameter feasibility data. The control parameter feasibility verification includes evaluating whether the optimized setpoint is within the safe range of the wind turbine operating parameters. Step S323: Based on the feasibility data of the control parameters, optimize and integrate the control parameters to obtain optimized control parameter data for the wind turbine. The optimization and integration of control parameters includes integrating the optimized setpoints that have passed the feasibility verification into executable wind turbine control parameters. Example

[0042] like Figure 2 As shown, a wind power generation efficiency optimization system based on complex meteorological conditions is implemented. The system includes: Turbulence intensity analysis module: Acquires real-time meteorological data and wind turbine operating parameters of the wind farm. The real-time meteorological data of the wind farm includes wind speed, wind direction, temperature, air pressure and turbulence intensity measurements. The wind turbine operating parameters include wind turbine blade angle, generator speed, power output and controller setpoints. Based on the real-time meteorological data of the wind farm, turbulence intensity analysis is performed to obtain turbulence intensity data. Efficiency loss calculation module: Based on the wind turbine operating parameters, fatigue load prediction analysis is performed on the turbulence intensity data to obtain fatigue load prediction data; based on the fatigue load prediction data, power generation efficiency loss correlation mining is performed to obtain power generation efficiency loss correlation data, which includes analyzing the power generation loss caused by the conservatism of the control strategy when the fatigue load exceeds the limit. Wind turbine parameter optimization module: Based on the correlation data of power generation efficiency loss, the module optimizes and matches the control strategy to generate control strategy optimization matching data. The control strategy optimization matching includes dynamically adjusting the wind turbine blade angle and generator speed to maximize power generation efficiency while ensuring that fatigue load does not exceed the limit. The module adjusts the wind turbine control parameters through the control strategy optimization matching data to obtain the wind turbine control parameter optimization data.

[0043] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0044] It should be understood that determining B based on A does not mean determining B solely based on A; it also means determining B based on A and / or other information.

[0045] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0046] In conclusion, the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing wind power generation efficiency based on complex weather conditions, characterized in that, The method comprises: Step S1: obtaining wind farm real-time meteorological data and wind turbine operating parameters, wherein the wind farm real-time meteorological data comprises wind speed, wind direction, temperature, air pressure and turbulence intensity measurement values, and the wind turbine operating parameters comprise wind turbine blade angle, generator speed, power output and controller set value; turbulence intensity analysis is performed based on the wind farm real-time meteorological data, so as to obtain turbulence intensity data; Step S2: fatigue load prediction analysis is performed on the turbulence intensity data according to the wind turbine operating parameters, so as to obtain fatigue load prediction data; power generation efficiency loss correlation mining is performed based on the fatigue load prediction data, so as to obtain power generation efficiency loss correlation data, wherein the power generation efficiency loss correlation mining comprises analyzing power generation loss caused by the conservativeness of the control strategy when the fatigue load exceeds the limit; Step S3: control strategy optimization matching is performed according to the power generation efficiency loss correlation data, so as to generate control strategy optimization matching data; wind turbine control parameter adjustment is performed through the control strategy optimization matching data, so as to obtain wind turbine control parameter optimization data.

2. The method for optimizing wind power generation efficiency based on complex meteorological conditions according to claim 1, characterized in that, Step S1 comprises: Step S11: wind vector synthesis is performed based on wind speed data and wind direction data in the wind farm real-time meteorological data, so as to obtain wind vector data; Step S12: wind vector fluctuation analysis is performed based on the wind vector data, so as to obtain wind vector fluctuation data; Step S13: environmental disturbance intensity calculation is performed based on the wind vector fluctuation data and temperature and air pressure data, so as to obtain environmental disturbance intensity data; Step S14: turbulence intensity correction is performed based on the environmental disturbance intensity data and the turbulence intensity measurement values, so as to obtain turbulence intensity data.

3. The method for optimizing wind power generation efficiency based on complex meteorological conditions according to claim 2, characterized in that, Step S13 comprises the following steps: Step S131: temperature-pressure coupling effect analysis is performed based on the temperature data and the air pressure data, so as to obtain temperature-pressure coupling effect data; Step S132: disturbance influence effect evaluation is performed based on the temperature-pressure coupling effect data and the wind vector fluctuation data, so as to obtain disturbance influence effect data; Step S133: environmental disturbance intensity analysis is performed based on the disturbance influence effect data, so as to obtain environmental disturbance intensity data.

4. The method for optimizing wind power generation efficiency based on complex meteorological conditions according to claim 1, characterized in that, The operation logic of step S2 is as follows: Step S21: load characteristic analysis is performed on the turbulence intensity data according to the wind turbine operating parameters, so as to obtain load characteristic data, wherein the load characteristic analysis comprises calculating the dynamic load change trend of the wind turbine components under the action of turbulence based on the wind speed, wind direction, temperature, air pressure and turbulence intensity measurement values and the wind turbine blade angle, generator speed, power output and controller set value; Step S22: stress distribution simulation of the wind turbine blades, tower and transmission chain under high turbulence conditions is performed based on the load characteristic data using the wind turbine dynamics model, so as to obtain stress distribution data, wherein the stress distribution simulation comprises solving the component stress response equation using the wind turbine dynamics model to obtain the stress space-time distribution; Step S23: fatigue cumulative damage calculation is performed based on the stress distribution data, so as to obtain fatigue load prediction data, wherein the fatigue cumulative damage calculation comprises evaluating the damage degree of the wind turbine components within a preset time range by applying a fatigue damage accumulation model according to the stress distribution data; Step S24: Analyzing the loss of power generation caused by the conservativeness of the control strategy based on the fatigue load prediction data, obtaining the loss of power generation efficiency correlation data, wherein the loss of power generation analysis includes comparing the difference between the adjustment of the controller set value when the fatigue load is exceeded and the ideal control strategy, and quantifying the loss of power generation.

5. The method for optimizing wind power generation efficiency based on complex meteorological conditions according to claim 4, characterized in that, Step S21 includes the following steps: Based on the wind speed, wind direction, temperature, air pressure and turbulence intensity measurement values, the initial influence of turbulence on the fan load is analyzed to obtain the initial load influence data of turbulence, wherein the initial influence analysis includes calculating the correlation between turbulence intensity and fan base load according to meteorological data; Based on the initial load influence data of turbulence and the fan blade angle, generator speed, power output and controller set value, the dynamic load response calculation of the fan components is carried out to obtain the dynamic load response data of the components, wherein the dynamic load response calculation includes simulating the instantaneous load change of the blade, tower and transmission chain under the action of turbulence combined with the fan operation parameters; The load change trend extraction is performed on the component dynamic load response data to obtain the dynamic load change trend as the load characteristic data, wherein the change trend extraction includes analyzing the change law of the load response data with time to determine the peak value and fluctuation characteristics.

6. The method for optimizing wind power generation efficiency based on complex meteorological conditions according to claim 5, characterized in that, Step S22 includes the following steps: Based on the load characteristic data, the load condition mapping of the fan blade, tower and transmission chain is carried out to obtain the model load input data, wherein the load condition mapping includes determining the input load parameters required by the fan dynamics model according to the dynamic load change trend; Based on the model load input data, the stress response equation is solved using the fan dynamics model to obtain the component stress response data, wherein the stress response equation solving includes calculating the instantaneous stress value of each component of the fan under high turbulence conditions; The spatiotemporal distribution feature extraction is performed on the component stress response data to obtain the stress distribution data, wherein the spatiotemporal distribution feature extraction includes integrating the change of the instantaneous stress value with time and plane to generate the stress spatiotemporal distribution.

7. The method for optimizing wind power generation efficiency based on complex meteorological conditions according to claim 6, characterized in that, Step S23 includes the following steps: The stress time history feature extraction is performed on the stress distribution data to obtain the stress cycle characteristic data, wherein the stress time history feature extraction includes analyzing the amplitude and cycle number of the stress change with time of each component in the stress distribution data; Based on the stress cycle characteristic data and the material characteristics of the fan components, the single stress cycle damage calculation is carried out to obtain the single cycle damage data, wherein the single stress cycle damage calculation includes determining the damage caused by each stress cycle according to the stress amplitude and the material S-N curve; The damage accumulation calculation in the preset time range is performed on the single cycle damage data to obtain the cumulative damage data, wherein the damage accumulation calculation includes adding each single cycle damage data in the preset time range by linear accumulation rule; Based on the cumulative damage data, the fatigue life prediction of the fan components is carried out to obtain the fatigue load prediction data, wherein the fatigue life prediction includes evaluating the fatigue damage degree of each component in the preset time range according to the cumulative damage data.

8. The method for optimizing wind power generation efficiency based on complex meteorological conditions according to claim 4, characterized in that, Step S24 includes the following steps: The fatigue load over-limit event data is obtained by identifying fatigue load over-limit events based on the fatigue load prediction data, wherein the fatigue load over-limit event identification comprises determining the operation time period in which the fatigue damage of the wind turbine component exceeds the pre-set safety threshold according to the fatigue load prediction data; The actual controller set value adjustment data is obtained by extracting actual controller set value adjustments based on the load over-limit event data, wherein the actual controller set value adjustment extraction comprises recording the actual modification parameters of the controller set value during the load over-limit event; The ideal set value data is obtained by calculating ideal controller set values based on the load over-limit event data, wherein the ideal controller set value calculation comprises simulating the controller set value that maximizes the power generation efficiency without considering fatigue constraints under the same load condition; The set value difference data is obtained by performing set value difference analysis based on the actual set value adjustment data and the ideal set value data, wherein the set value difference analysis comprises calculating the numerical deviation between the actual set value adjustment data and the ideal set value data; The power generation efficiency loss correlation data is obtained by quantifying the power generation loss value based on the set value difference data, wherein the power generation loss value quantification comprises evaluating the power generation loss value caused by the conservatism of the control strategy according to the set value difference data.

9. The method for optimization of wind power generation efficiency based on complex meteorological conditions according to claim 1, characterized in that, Step S3 comprises the following steps: Step S31: generating control strategy optimization matching data by performing control strategy optimization matching based on the power generation efficiency loss correlation data and the fatigue load prediction data, wherein the control strategy optimization matching comprises analyzing the power generation efficiency loss correlation data to determine the optimization adjustment scheme of the wind turbine blade angle and the generator speed, and maximizing the power generation efficiency while ensuring that the fatigue load does not exceed the limit under the constraint of the fatigue load prediction data; Step S32: obtaining wind turbine control parameter optimization data by adjusting the wind turbine control parameters according to the control strategy optimization matching data, wherein the wind turbine control parameter adjustment comprises setting the control parameter values of the wind turbine blade angle and the generator speed according to the control strategy optimization matching data.

10. The method for optimizing wind power generation efficiency based on complex meteorological conditions according to claim 9, characterized in that, Step S31 comprises the following steps: Step S311: obtaining efficiency loss key parameter data by identifying efficiency loss key parameters based on the power generation efficiency loss correlation data; Step S312: obtaining control strategy constraint condition data by performing control strategy constraint condition analysis based on the efficiency loss key parameter data and the fatigue load prediction data; Step S313: obtaining wind turbine parameter optimization scheme data by performing wind turbine parameter optimization scheme calculation based on the control strategy constraint condition data, wherein the wind turbine parameter optimization scheme calculation comprises determining the optimization values of the wind turbine blade angle and the generator speed by an iterative optimization algorithm under the constraint condition to maximize the power generation efficiency; Step S314: generating control strategy optimization matching data by performing control strategy matching integration based on the wind turbine parameter optimization scheme data.

11. A system for optimizing wind power generation efficiency based on complex weather conditions, characterized in that, The system comprises: The turbulence intensity analysis module: obtain real-time meteorological data and wind turbine operating parameters of the wind farm, wherein the real-time meteorological data of the wind farm includes wind speed, wind direction, temperature, air pressure and turbulence intensity measurement values, and the wind turbine operating parameters include wind turbine blade angle, generator speed, power output and controller set value; based on the real-time meteorological data of the wind farm, turbulence intensity analysis is carried out to obtain turbulence intensity data; The efficiency loss calculation module: fatigue load prediction analysis is carried out on the turbulence intensity data according to the wind turbine operating parameters to obtain fatigue load prediction data; based on the fatigue load prediction data, power generation efficiency loss correlation mining is carried out to obtain power generation efficiency loss correlation data, wherein the power generation efficiency loss correlation mining includes analyzing the power generation loss caused by the conservativeness of the control strategy when the fatigue load is over limit; The wind turbine parameter optimization module: control strategy optimization matching is carried out according to the power generation efficiency loss correlation data to generate control strategy optimization matching data, wherein the control strategy optimization matching includes dynamically adjusting the wind turbine blade angle and the generator speed to maximize the power generation efficiency under the premise of ensuring that the fatigue load is not over limit; the wind turbine control parameter adjustment is carried out through the control strategy optimization matching data to obtain wind turbine control parameter optimization data.

Images