Wind turbine power optimization method and system

Abstract

The application discloses a wind turbine power optimization method and system, and relates to the technical field of wind power generation control. The method comprises the following steps: correcting time sequence operation data during the operation of a wind turbine, eliminating abnormal data, and performing signal synchronization, wind speed space correlation correction and other processing to obtain a state characteristic data set accurately reflecting the relationship between the operation state and the power output of the wind turbine; based on the characteristic data set, the optimal target power value that can be reached by the wind turbine under the current wind condition is determined by matching similar historical operation segments and performing power stability evaluation; the difference between the target power and the actual power is calculated and trend prediction is performed to generate a power demand trend; adjustment instruction sets of the pitch angle and the generator torque are obtained according to a preset adjustment strategy mapping table; and the instructions are sent to the wind turbine controller to adjust the operation power. Through the above method, the dynamic and accurate optimization of the power of the wind turbine is realized, and the power generation efficiency and grid connection stability are improved.

Description

Technical Field

[0001] This application relates to the field of wind power generation control technology, and in particular to a method and system for optimizing the power of wind turbine generator sets. Background Technology

[0002] The power output of wind turbines is susceptible to significant fluctuations due to the randomness of wind speed and turbulence characteristics. This not only reduces power generation efficiency but also poses a challenge to the stability of the power grid. Traditional control strategies are usually based on standard power curves and are difficult to adapt to complex and ever-changing actual wind conditions. As a result, the turbines often operate at non-optimal power points, failing to fully utilize wind energy resources and ensure stable power output. Summary of the Invention

[0003] The main purpose of this application is to provide a method and system for optimizing the power output of wind turbine generator sets, which can solve the technical problem of large power output fluctuations in existing wind turbine generator sets.

[0004] To achieve the above objectives, this application provides a method for optimizing the power output of a wind turbine generator set, the method comprising: The time-series operation data collected during the operation of the wind turbine generator is corrected to obtain a set of state characteristic data reflecting the relationship between the unit's operating status and power output; Based on the state feature data set, the optimized target power value that the wind turbine generator can achieve under the current wind conditions is obtained; Based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, a trend prediction is performed to obtain the power demand trend, and a set of adjustment instructions for the control parameters of the wind turbine generator set is obtained based on the power demand trend. The set of adjustment instructions is sent to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set.

[0005] In one embodiment, the step of correcting the time-series operational data set collected during the operation of the wind turbine generator set to obtain a characteristic data set reflecting the relationship between the generator set's operating status and power output includes: Anomaly removal and signal synchronization are performed on the original time-series operation data collected during the operation of the wind turbine generator set to obtain a regular time-series data set. Spatial correlation correction and turbulence intensity calculation are performed on the wind speed data in the regularized time series dataset to obtain equivalent inflow wind speed data; The equivalent inflow wind speed data is correlated and combined with the generator speed and pitch angle in the regularized time series data set to obtain a state feature data set.

[0006] In one embodiment, the step of performing spatial correlation correction and turbulence intensity calculation on the wind speed data in the regularized time-series data set to obtain equivalent inflow wind speed data includes: Determine the wind speed data in the regularized time-series data set, wherein the wind speed data is wind speed measurement data at the hub of the wind turbine generator set and at preset spatial distribution points; Based on the computational fluid dynamics model, the wind speed measurement data is spatially correlated and corrected to obtain a corrected wind speed data sequence. Turbulence intensity is calculated on the corrected wind speed data sequence to obtain turbulence intensity values ​​that reflect the characteristics of wind speed fluctuations; The turbulence intensity value and the wind speed measurement data are weighted and fused to obtain the equivalent inflow wind speed data.

[0007] In one embodiment, the step of obtaining the optimized target power value that the wind turbine can achieve under the current wind conditions based on the state feature data set includes: The wind condition features in the state feature data set are compared with the historical operation segments in the historical database. The historical operation segments with similarity within the target similarity range are combined to obtain the historical power data set. A power stability assessment is performed on the historical power data set to obtain a stability assessment value; Power sequences whose stable evaluation values ​​are greater than preset stable evaluation values ​​are used as candidate optimized power data; Based on the current power grid dispatch instructions or power change rate limits, boundary constraints are applied to the candidate optimized power data to obtain the optimized target power value that the wind turbine can achieve under the current wind conditions.

[0008] In one embodiment, before the step of performing a power stability assessment on the historical power data set to obtain a stability assessment value, the method further includes: The power data with different sampling frequencies in the historical power data set are normalized to a time scale to generate a power data sequence with a unified time base. Power fluctuation features are extracted from the power data sequence of the unified time base to identify abnormal fluctuation segments where the power change rate exceeds a preset threshold. The abnormal fluctuation segments are removed from the historical power data set.

[0009] In one embodiment, the step of performing trend prediction based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set to obtain a power demand trend, and obtaining a set of adjustment instructions for the control parameters of the wind turbine generator set based on the power demand trend, includes: Determine the power difference between the optimized target power value and the actual power value of the wind turbine generator set; Based on the power difference, time-segmented accumulation and trend prediction are performed to obtain the power demand trend; The power regulation demand intensity index is obtained based on the power demand trend. Based on the power regulation demand intensity index, the preset regulation strategy mapping table is queried to determine the reference value of the pitch angle increment and the reference value of the generator torque setpoint increment. The pitch angle increment reference value and the generator torque setpoint increment reference value are coupled to obtain a set of adjustment instructions for the wind turbine generator control parameters.

[0010] In one embodiment, the step of determining the reference value for the pitch angle increment and the reference value for the generator torque setpoint increment based on the power regulation demand intensity index and a preset regulation strategy mapping table includes: Based on the power regulation demand intensity index, a preset regulation strategy mapping table is queried to obtain the query result. The preset regulation strategy mapping table defines the correspondence between different intensity index ranges and the priority configuration of regulation parameters. Based on the query results, the priority of the adjustment parameters corresponding to the current intensity index range is determined. The priority configuration of the adjustment parameters includes either the pitch angle priority mode or the generator torque priority mode. In the pitch angle priority mode, based on the power regulation demand intensity index and the current pitch angle, the pitch angle increment reference value is obtained through the first data mapping relationship. By combining the power difference and the pitch angle increment reference value, the generator torque setpoint increment reference value is obtained through the second data mapping relationship; In the generator torque priority mode, based on the power regulation demand intensity index and the current generator torque, the incremental reference value of the generator torque setpoint is obtained through a third data mapping relationship. By combining the power difference and the incremental reference value of the generator torque setpoint, the incremental reference value of the pitch angle is obtained through the fourth data mapping relationship.

[0011] In one embodiment, after the step of sending the set of adjustment instructions to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set, the method further includes: Collect feedback operating data of the wind turbine generator set after executing the set of adjustment commands; By comparing the feedback operation data with the expected operation data, a quantitative evaluation index for the optimization effect is determined; When the quantitative evaluation index is lower than the preset threshold, the optimized target power value is updated.

[0012] In one embodiment, the step of updating the optimized target power value includes: Collect the feedback operation data, status feature data set and optimization effect evaluation results recorded in each optimization process, and construct an optimization effect analysis database; The optimization effects under different wind conditions in the optimization effect analysis database are compared and analyzed to identify the correlation between target wind condition characteristics and poor optimization effects; When the optimization effect of the target wind condition feature is consistently poor, the correlation weights of wind speed and power in the generation rules of the state feature data set are adjusted based on the correlation relationship to obtain the optimization weights. Update the threshold setting regarding the power change rate limit in the calculation rules for the optimized target power value to obtain the optimized threshold; A dynamic optimization strategy is generated based on the optimization weights and the optimization thresholds, and the optimization target power value is updated based on the dynamic optimization strategy.

[0013] Furthermore, to achieve the above objectives, this application also proposes a wind turbine generator power optimization system, which includes: The data correction module is used to correct the time-series operating data set collected during the operation of the wind turbine generator set, and obtain a set of state characteristic data reflecting the relationship between the unit's operating status and power output. The target calculation module is used to obtain the optimized target power value that the wind turbine can achieve under the current wind conditions based on the state feature data set. The instruction generation module is used to predict the power demand trend based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, and to obtain the power demand trend. Based on the power demand trend, it obtains a set of adjustment instructions for the control parameters of the wind turbine generator set. The instruction issuing module is used to issue the set of adjustment instructions to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set.

[0014] In addition, to achieve the above objectives, this application also proposes a wind turbine power optimization device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the wind turbine power optimization method described above.

[0015] In addition, to achieve the above objectives, the present invention also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the wind turbine power optimization method described above.

[0016] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the wind turbine generator power optimization method described above.

[0017] This application provides a method for optimizing the power output of wind turbine generator sets. The method corrects the time-series operating data of the wind turbine generator set during operation, removes abnormal data, and performs signal synchronization and wind speed spatial correlation correction to obtain a set of state characteristic data that accurately reflects the relationship between the generator set's operating status and power output. Based on this characteristic data set, by matching similar historical operating segments and conducting power stability assessment, the method determines the optimal target power value achievable by the generator set under the current wind conditions. The difference between the target power and the actual power is calculated, and trend prediction is performed to generate a power demand trend. According to a preset adjustment strategy mapping table, a set of adjustment commands for the pitch angle and generator torque is obtained, and these commands are sent to the generator set controller to adjust its operating power. Through this approach, the method achieves dynamic and precise optimization of the generator set's power output, improving power generation efficiency and grid connection stability. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating an embodiment of the wind turbine generator power optimization method of this application. Figure 2 This is a schematic diagram of the wind power generation difference trend curve of an embodiment of the wind turbine generator power optimization method of this application; Figure 3 This is a schematic diagram of the module structure of the wind turbine generator power optimization system according to an embodiment of this application; Figure 4 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the wind turbine generator power optimization method in this application embodiment.

[0021] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0023] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The main solution of this application embodiment is to correct the time-series operation data set collected during the operation of the wind turbine generator set to obtain a state feature data set reflecting the relationship between the unit's operating status and power output; Based on the state feature data set, the optimized target power value that the wind turbine generator can achieve under the current wind conditions is obtained; Based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, a trend prediction is performed to obtain the power demand trend, and a set of adjustment instructions for the control parameters of the wind turbine generator set is obtained based on the power demand trend. The set of adjustment instructions is sent to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set.

[0025] The power output of wind turbines is susceptible to significant fluctuations due to the randomness of wind speed and turbulence characteristics. This not only reduces power generation efficiency but also poses a challenge to the stability of the power grid. Traditional control strategies are usually based on standard power curves and are difficult to adapt to complex and ever-changing actual wind conditions. As a result, the turbines often operate at non-optimal power points, failing to fully utilize wind energy resources and ensure stable power output.

[0026] This application provides a solution that corrects the time-series operating data of wind turbine generators during operation, removes abnormal data, and performs signal synchronization and wind speed spatial correlation correction to obtain a set of state characteristic data that accurately reflects the relationship between the generator's operating status and power output. Based on this characteristic data set, by matching similar historical operating segments and conducting power stability assessment, the optimal target power value that the generator can achieve under the current wind conditions is determined. The difference between the target power and the actual power is calculated and trend prediction is performed to generate a power demand trend. According to a preset adjustment strategy mapping table, a set of adjustment commands for pitch angle and generator torque is obtained, and the commands are sent to the generator controller to adjust its operating power. Through the above methods, this approach achieves dynamic and precise optimization of generator power, improving power generation efficiency and grid connection stability.

[0027] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device capable of performing the above functions, such as a wind turbine power optimization device. This embodiment does not specifically limit it in this regard. The following uses a wind turbine power optimization device as an example to describe this embodiment and the following embodiments.

[0028] This application provides a method for optimizing the power output of a wind turbine generator set, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the wind turbine generator power optimization method of this application.

[0029] In this embodiment, the wind turbine generator power optimization method includes steps S10~S40: Step S10: Correct the time-series operation data set collected during the operation of the wind turbine generator set to obtain a state characteristic data set reflecting the relationship between the unit's operating status and power output.

[0030] It should be noted that the time-series operation data set refers to a multi-dimensional data sequence that is continuously collected in chronological order and reflects the operating status of each component of the wind turbine generator set. The state feature data set refers to a standardized data set that has been cleaned, corrected and feature extracted, and can more accurately characterize the actual inflow wind conditions of the wind turbine and the overall operating attitude of the unit, and clearly show the causal relationship between them and the final power output.

[0031] Understandably, the process begins by using data quality detection algorithms, such as thresholding based on standard deviation or interquartile range, to identify and remove outliers from the original time-series operational data (including wind speed, engine speed, power, and pitch angle). Timestamp alignment techniques are then employed to address data asynchrony issues caused by different sensor sampling frequencies or transmission delays, generating well-organized time-series data. Next, for wind speed data, a pre-generated computational fluid dynamics model is used to correct the spatial correlation of wind measurement data at multiple points in front of the wind turbine hub, eliminating the effects of nacelle tower shadows and flow field distortion. Turbulence intensity within this time period is then calculated, and finally, weighted fusion yields the effective inflow wind speed representing the entire wind turbine. Finally, the processed equivalent inflow wind speed is correlated and combined with synchronized key state parameters such as generator speed and pitch angle to form a high-quality state feature dataset for subsequent power optimization analysis.

[0032] In one feasible implementation, the step of correcting the time-series operational data set collected during the operation of the wind turbine generator set to obtain a characteristic data set reflecting the relationship between the generator set's operating status and power output includes: Anomaly removal and signal synchronization are performed on the original time-series operation data collected during the operation of the wind turbine generator set to obtain a regular time-series data set. Spatial correlation correction and turbulence intensity calculation are performed on the wind speed data in the regularized time series dataset to obtain equivalent inflow wind speed data; The equivalent inflow wind speed data is correlated and combined with the generator speed and pitch angle in the regularized time series data set to obtain a state feature data set.

[0033] It should be noted that outlier removal refers to using statistical or modeling methods to identify and remove invalid data points that significantly deviate from the normal fluctuation range due to sensor malfunction or transient interference. Signal synchronization processing refers to resolving the problem of misaligned data timestamps caused by inconsistent data acquisition times or transmission delays from different sensors, ensuring precise alignment of signals from all data channels on the time axis. Spatial correlation correction addresses the limitation that data measured by anemometers mounted on the nacelle cannot fully represent the wind conditions experienced by the entire swept surface of the wind turbine; it uses a flow field model to convert single-point measurements into effective wind speeds equivalent to those acting on the entire wind turbine. Turbulence intensity is the ratio of the standard deviation of wind speed to the average wind speed over a specific time period.

[0034] In the specific implementation, the original time-series data is first preprocessed. For outlier removal, methods based on physical constraints and statistical distribution can be used. For example, reasonable ranges for each parameter are set, and data points exceeding the threshold are considered outliers and removed. or hour, .

[0035] in, express The raw data points at any given time (such as power and wind speed). and These are the lower and upper threshold values ​​set according to the wind turbine's design operating range. This represents an invalid value, which will be filtered or imputed in subsequent processing.

[0036] Signal synchronization typically employs timestamp alignment algorithms. All sensor data is resampled to the same time base (e.g., 1Hz). For data points with missing timestamps, methods such as linear interpolation can be used to fill in the gaps, ensuring that all data sequences are strictly synchronized in time, forming a well-organized time-series data set.

[0037] After acquiring the regularized time-series data, the focus is on processing the wind speed data. Since the anemometer is installed at the stern of the nacelle, its measurements are affected by the tower shadow effect, blade disturbance, and flow field distortion, and therefore cannot directly represent the effective wind speed driving the turbine. Spatial correlation correction aims to establish the relationship between the single-point measured wind speed and the equivalent inflow wind speed. In this embodiment, a correction coefficient is obtained through experience or CFD simulation. (Usually less than 1), used to compensate for wake loss at the rear of the nacelle.

[0038]

[0039] in, yes The original wind speed value measured by the anemometer at any given time. It is the equivalent inflow velocity obtained after correction.

[0040] At the same time, calculate the turbulence intensity ( This is used to quantify the volatility of wind speed. The calculation formula is:

[0041] in, It is the equivalent inflow velocity within a time window (e.g., 10 minutes). standard deviation Within this time window The average value. Turbulence intensity is a key parameter for assessing wind stability and its impact on unit load.

[0042] Then, the processed key parameters are correlated and combined to determine the equivalent inflow wind speed corresponding to each timestamp. turbulence intensity (Can be added as a dynamic feature), generator speed and pitch angle The parameters are combined to form a multidimensional feature vector.

[0043]

[0044] The feature vector set is the final state feature data set. The state feature data set eliminates noise and bias in the original data and more realistically reflects the operating status of the wind turbine under specific wind conditions.

[0045] In one feasible implementation, the step of performing spatial correlation correction and turbulence intensity calculation on the wind speed data in the regularized time-series data set to obtain equivalent inflow wind speed data includes: Determine the wind speed data in the regularized time-series data set, wherein the wind speed data is wind speed measurement data at the hub of the wind turbine generator set and at preset spatial distribution points; Based on the computational fluid dynamics model, the wind speed measurement data is spatially correlated and corrected to obtain a corrected wind speed data sequence. Turbulence intensity is calculated on the corrected wind speed data sequence to obtain turbulence intensity values ​​that reflect the characteristics of wind speed fluctuations; The turbulence intensity value and the wind speed measurement data are weighted and fused to obtain the equivalent inflow wind speed data.

[0046] It should be noted that the wind speed measurement data at the wind turbine hub refers to the wind speed value directly obtained by a sensor installed on the central shaft (hub) of the wind turbine. The wind speed measurement data at the preset spatial distribution points refers to the wind speed values ​​collected at specific locations in front of or around the wind turbine (such as by lidar or multiple ultrasonic anemometers), which can reflect the spatial distribution information of the incoming wind field.

[0047] In practical implementation, after determining the wind speed data, spatial correlation correction is first performed. A computational fluid dynamics (CFD) model is used to establish a mapping relationship from multi-point measured wind speeds to the equivalent wind speed experienced across the entire swept surface of the wind turbine. Through numerical simulation, the CFD model can calculate the transfer function or correction coefficient matrix between the wind speed at each measuring point in front of the wind turbine and the equivalent wind speed under ideal uniform flow conditions. .

[0048]

[0049] in, It is a column vector, representing the... A set of wind speed values ​​measured at the wheel hub and various preset spatial distribution points at all times. . It is a coefficient matrix (which can be a row vector or a set of coefficients suitable for weighted summation) predetermined through CFD simulation, which contains weights for spatial correlation correction. It is a sequence of one or more wind speed values ​​that are better represented by the inflow wind conditions after correction.

[0050] Next, turbulence intensity is calculated based on the corrected wind speed data sequence. Turbulence intensity It is a key parameter characterizing wind speed fluctuations. For a given time window... (For example, 10 minutes), the turbulence intensity is usually calculated based on the ratio of the standard deviation to the mean of the wind speed within that window.

[0051]

[0052] in, It is a time window Internally corrected wind speed sequence The average value, Within the same time window The standard deviation.

[0053] Finally, the turbulence intensity information and wind speed measurement data are weighted and fused to generate the final equivalent inflow wind speed data. This fusion can take into account the impact of wind speed fluctuations on the wind turbine power output. One implementation method is to assign weights to the average wind speed and turbulence intensity to calculate a comprehensive wind speed index.

[0054]

[0055] Alternatively, turbulence intensity can be treated as a factor influencing the equivalent wind speed and combined nonlinearly.

[0056] in, yes The equivalent inflow velocity calculated within the time window of the given moment. and These are weighting coefficients (the sum of which is 1 or determined based on physical relationships). It is a fusion function. It is obtained through weighted fusion. It not only includes average wind speed information, but also incorporates the characteristics of turbulence intensity, thus providing a more comprehensive representation of the real and effective wind energy input acting on the wind turbine.

[0057] Step S20: Based on the state feature data set, obtain the optimized target power value that the wind turbine generator can achieve under the current wind conditions.

[0058] It should be noted that the current wind conditions refer to the wind resource conditions currently experienced by the wind turbine rotor, as described by parameters such as the equivalent inflow wind speed and turbulence intensity in the state characteristic data set. The optimized target power value refers to the maximum possible output power or optimal economic operating power that the wind turbine should achieve under an ideal control strategy, determined through theoretical calculation or model optimization under the current specific wind conditions and turbine operating state. It can serve as a reference benchmark for evaluating the potential of actual output power or guiding power point tracking control.

[0059] Understandably, firstly, based on massive amounts of historical, normally operating, and high-performance state characteristic data (such as equivalent inflow wind speed, generator speed, and pitch angle) and corresponding actual power output data, a power performance prediction model is trained using supervised learning through machine learning algorithms (such as Gradient Boosting Decision Tree (GBDT) or neural networks). This model can learn the power curve of the unit under healthy conditions and optimal control. In practical applications, the real-time acquired set of current state characteristic data is input into the trained prediction model. The model will directly output a corresponding power prediction value based on the current combination of wind speed, speed, and pitch angle. This value represents the theoretically achievable optimal target power value of the unit under specific wind conditions and states. Furthermore, the theoretical power curve of the wind turbine design can be combined, and the curve can be corrected in real time based on the current air density and turbulence intensity to calculate the theoretical maximum captured power as the optimization target.

[0060] In one feasible implementation, the step of obtaining the optimized target power value that the wind turbine can achieve under the current wind conditions based on the state feature data set includes: The wind condition features in the state feature data set are compared with the historical operation segments in the historical database. The historical operation segments with similarity within the target similarity range are combined to obtain the historical power data set. A power stability assessment is performed on the historical power data set to obtain a stability assessment value; Power sequences whose stable evaluation values ​​are greater than preset stable evaluation values ​​are used as candidate optimized power data; Based on the current power grid dispatch instructions or power change rate limits, boundary constraints are applied to the candidate optimized power data to obtain the optimized target power value that the wind turbine can achieve under the current wind conditions.

[0061] It should be noted that wind condition characteristics refer to the set of key parameters extracted from the state characteristic data set to describe the wind resource status, typically including equivalent inflow wind speed, turbulence intensity, and wind shear. Historical operating segments refer to continuous time-series operating data records of wind turbine generators stored in the historical database over a certain period. Stability assessment values ​​are numerical indicators that quantify the degree of power stability. Preset stability assessment values ​​are threshold values ​​artificially set to determine whether a power sequence is sufficiently stable. Candidate optimized power data are a set of historical power sequences that have undergone preliminary screening and are considered to have good stability. Dispatch instructions are the active power output targets or limits issued by the grid operation center to the wind farm. Power change rate limits are the maximum permissible increase or decrease in power of wind turbine generators per unit time, stipulated to ensure grid stability.

[0062] In the specific implementation, the features representing wind conditions in the current state feature data set are organized into a feature vector. Then, the similarity between this current wind condition feature vector and the corresponding wind condition feature vector for each historical running segment in the historical database is calculated. For example, cosine similarity can be used:

[0063] in, It is the current wind condition feature vector. This is the average wind condition feature vector for a specific historical operating segment. The closer the calculated similarity value is to 1, the more similar the wind conditions. The system sets a target similarity interval (e.g., [0.85, 1.0]), retrieves all historical operating segments whose similarity falls within this interval, and combines their corresponding power output data into a historical power data set. .

[0064] For each power sequence in the historical power dataset Power stability is assessed using data spanning 10 minutes (e.g., data from a single data point). The stability assessment calculates the coefficient of variation (CV) of the power series, which is the ratio of the standard deviation to the mean, and uses this CV as the stability assessment value. :

[0065] in, It is a power sequence standard deviation It is a power sequence average value. The smaller the value, the smaller the fluctuation and the more stable the power sequence. A preset stability evaluation value is set. , will satisfy The power sequences were selected as candidate power data for optimization. These data represent the power levels that the units have consistently operated and output under similar wind conditions.

[0066] Finally, constraints are imposed on the candidate optimized power data based on the actual requirements of the current power grid. It is assumed that a representative power value is obtained from the candidate data. (For example, the average value of a candidate power sequence). This needs to be determined according to grid dispatch instructions. (If applicable) and power change rate limit (Unit: kW / min) This is corrected to obtain the final optimized target power value. This typically involves a constrained optimization problem or simple boundary pruning: First, ensure the target power meets the dispatch command: if a valid grid dispatch command exists, the optimized target power value should preferably be close to or equal to the grid dispatch command (within the unit's capacity). Second, ensure power changes are smooth and do not exceed limits: the final set... Compared with the current actual power The difference, divided by the adjustment time The resulting rate of change of power must satisfy Through this constraint process, the final optimized target power value not only references historical excellent operating data but also ensures compliance with the real-time operating specifications of the power grid.

[0067] In one feasible implementation, before the step of performing a power stability assessment on the historical power data set to obtain a stability assessment value, the method further includes: The power data with different sampling frequencies in the historical power data set are normalized to a time scale to generate a power data sequence with a unified time base. Power fluctuation features are extracted from the power data sequence of the unified time base to identify abnormal fluctuation segments where the power change rate exceeds a preset threshold. The abnormal fluctuation segments are removed from the historical power data set.

[0068] It should be noted that time-scale normalization refers to the process of converting time-series data with different sampling frequencies to a common time resolution reference through interpolation or aggregation. A power data series with a unified time reference refers to power time-series data where the time intervals of all data points are equal, facilitating subsequent consistency analysis. Power fluctuation feature extraction refers to the calculation of indicators reflecting the severity of power changes, such as the power change rate, from the power data series using mathematical methods. The power change rate is the absolute value of power change per unit time, used to measure the severity of power fluctuations. Abnormal fluctuation segments refer to power data intervals where the power change rate continuously or instantaneously exceeds a preset threshold, possibly caused by abnormal operating conditions such as faults or extreme turbulence.

[0069] In practical implementation, since the power data in the historical database may come from data acquisition systems of different periods and configurations, their sampling frequencies may differ (e.g., 1Hz, 0.1Hz, etc.). To conduct effective stability assessment, time scale normalization is necessary first. This is typically achieved using linear interpolation or resampling methods to standardize all data to a specific time interval. (For example, 1 second) on.

[0070] The linear interpolation formula is:

[0071] in, It is a point in time on the standard timeline. and It is the adjacent data in the original data Two time points, and This corresponds to the power value. Using this method, a power value can be generated for each point on the standard time axis, thus obtaining a power data sequence with a unified time base.

[0072] For the normalized power sequence Calculate its rate of change of power. In the discrete-time series, the rate of change of power... It can be approximated by first-order difference:

[0073] in, It is the normalized standard time interval. It quantifies the drasticness of power changes at each moment.

[0074] Next, a preset threshold for the rate of change of power is set. This threshold can be determined based on the rated power of the wind turbine, historical operating experience, or grid guidelines. (By comparison...) and Abnormal fluctuation segments were identified: Abnormal fragments

[0075] In practice, a period of time that continuously exceeds a threshold is usually defined as an abnormal fluctuation segment to avoid interference from single-point spikes.

[0076] All identified anomalous fluctuation segments (including the start and end timestamps of the segments) will be extracted from the power data sequence with a unified time base. The remaining data points, after being filtered out, form a cleaner set of historical power data, free from abnormally drastic fluctuations. This processed dataset is more representative of the power output characteristics of wind turbine generators under normal and stable operating conditions.

[0077] Step S30: Based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, a trend prediction is performed to obtain the power demand trend, and a set of adjustment instructions for the control parameters of the wind turbine generator set is obtained based on the power demand trend.

[0078] It should be noted that the actual power value refers to the active power output value of the wind turbine generator set as measured and fed back by sensors at the current moment. The power difference is the algebraic difference obtained by subtracting the actual power value from the optimized target power value; its sign and magnitude together indicate the direction and extent of the power demand gap. Trend forecasting refers to the process of predicting the direction and trend of future short-term changes based on current and historical power difference sequences using mathematical methods. The power demand trend is the output of trend forecasting; it is a qualitative judgment, such as "demand increasing," "demand decreasing," or "demand remaining stable," or a quantitative curve showing future power difference changes. Control parameters refer to adjustable variables that directly affect the power output of the wind turbine generator set, mainly including pitch angle and generator torque. The adjustment command set is a set of specific control commands that can be issued to the actuators in response to the power demand trend, such as the adjustment amount of the pitch angle setpoint and the change in the torque reference value.

[0079] Understandably, the process begins with continuously recording the power difference sequence at current and historical moments, forming a time series. Then, short-term prediction algorithms such as moving averages, exponential smoothing, or more advanced Kalman filtering are used to analyze this time series and predict the trend of the power difference over several control cycles, thus obtaining the power demand trend (e.g., "the power difference will remain positive and increase over the next 30 seconds"). Next, based on this trend and the current operating status of the wind turbine (e.g., rotational speed, wind speed), a pre-defined control strategy mapping table or a model-based predictive control (MPC) algorithm is used to calculate the adjustments needed for key control parameters of the wind turbine to achieve power point tracking. For example, if the trend indicates a need for a significant power increase, instructions to "reduce the pitch angle to capture more wind energy" and "appropriately increase the generator torque reference value" might be generated; if a smooth power reduction is needed, instructions to "gradually increase the pitch angle to limit power" might be generated. Finally, these adjustments for different control parameters are encapsulated into a complete set of adjustment instructions and sent to the underlying pitch and converter systems for execution.

[0080] In one feasible implementation, the step of predicting the power demand trend based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, and obtaining the set of adjustment instructions for the control parameters of the wind turbine generator set based on the power demand trend includes: Determine the power difference between the optimized target power value and the actual power value of the wind turbine generator set; Based on the power difference, time-segmented accumulation and trend prediction are performed to obtain the power demand trend; The power regulation demand intensity index is obtained based on the power demand trend. Based on the power regulation demand intensity index, the preset regulation strategy mapping table is queried to determine the reference value of the pitch angle increment and the reference value of the generator torque setpoint increment. The pitch angle increment reference value and the generator torque setpoint increment reference value are coupled to obtain a set of adjustment instructions for the wind turbine generator control parameters.

[0081] It should be noted that time-segmented cumulative sum trend prediction is a method that divides time into continuous intervals, integrates (cumulatively sums) the power difference within each interval, and then analyzes the changing patterns of these cumulative sums to infer future trends. The power regulation demand intensity index measures the urgency or magnitude of adjustments to control parameters to achieve power point tracking (PPT), and is typically determined by the slope of the power demand trend and the predicted magnitude of the power difference. The preset regulation strategy mapping table is a predefined data table or functional relationship that takes the power regulation demand intensity index, etc., as input and outputs recommended adjustment reference values ​​for pitch angle and generator torque. The pitch angle increment reference value is the change in the suggested pitch angle setpoint relative to the current value; a positive value usually indicates an increase in the pitch angle to reduce power capture. The generator torque setpoint increment reference value is the change in the suggested generator torque command relative to the current value; a positive value usually indicates an increase in torque to increase power generation or limit speed.

[0082] In the specific implementation, firstly, the power difference is continuously calculated and recorded. ,in This is for discrete-time indexing. Next, the time axis is divided into sections of length [length missing]. The sampling period is divided into time intervals. The power difference within each time interval is accumulated and summed to obtain the cumulative sum for that time interval. :

[0083] in, This indicates a time period index. This reflects the overall power shortage situation during that period (positive values ​​indicate insufficient power, negative values ​​indicate excess power). Trend forecasting analyzes several recent periods (e.g., , , Cumulative and sequence of ) To achieve this. (Refer to...) Figure 2 The trend line of the sequence can be fitted using methods such as linear regression or weighted moving average, and its slope can be determined. This represents the trend of power demand (e.g.) This indicates that the power demand gap is widening.

[0084] Power regulation demand intensity index Incorporating trend slope The cumulative sum of the power difference predicted for a future period (This can be obtained through time series forecasting models such as ARIMA or simple extrapolation). The weighted calculation formula is:

[0085] in, and It is a weighting coefficient used to adjust the proportion of the influence of the trend change rate and the predicted cumulative sum on the adjustment strength. The larger the value, the stronger the required adjustment action.

[0086] The preset adjustment strategy mapping table is a two-dimensional table or function. ,in This is the generator speed. Based on the calculated... and current speed By querying this mapping table, you can obtain the corresponding pitch angle increment reference value. and generator torque setpoint increment reference value :

[0087] This mapping table is pre-set based on wind turbine characteristics, load limitations, and controller design experience to ensure that it provides safe and effective parameter adjustment suggestions under different demand intensities.

[0088] Finally, the query results and Coupling is implemented. Coupling primarily considers the mutual influence between the two elements during dynamic processes. For example, when rapidly reducing power at high wind speeds, it's necessary to coordinately increase the pitch angle and reduce torque to prevent engine overspeed. The coupling logic can be a set of rules or a simple coupling function. Ultimately, this generates a coordinated set of adjustment instructions that can be directly issued. This ensures the overall optimality of the control actions.

[0089] In one feasible implementation, the step of determining the reference value for the pitch angle increment and the reference value for the generator torque setpoint increment based on the power regulation demand intensity index and a preset regulation strategy mapping table includes: Based on the power regulation demand intensity index, a preset regulation strategy mapping table is queried to obtain the query result. The preset regulation strategy mapping table defines the correspondence between different intensity index ranges and the priority configuration of regulation parameters. Based on the query results, the priority of the adjustment parameters corresponding to the current intensity index range is determined. The priority configuration of the adjustment parameters includes either the pitch angle priority mode or the generator torque priority mode. In the pitch angle priority mode, based on the power regulation demand intensity index and the current pitch angle, the pitch angle increment reference value is obtained through the first data mapping relationship. By combining the power difference and the pitch angle increment reference value, the generator torque setpoint increment reference value is obtained through the second data mapping relationship; In the generator torque priority mode, based on the power regulation demand intensity index and the current generator torque, the incremental reference value of the generator torque setpoint is obtained through a third data mapping relationship. By combining the power difference and the incremental reference value of the generator torque setpoint, the incremental reference value of the pitch angle is obtained through the fourth data mapping relationship.

[0090] It should be noted that the query results refer to the data entries or rule identifiers returned after matching and searching the preset regulation strategy mapping table using the power regulation demand intensity index as the input key. The regulation parameter priority configuration is defined in the mapping table, representing the control parameter adjustment order strategy set for different regulation demand intensity ranges. Its core is to clarify which parameter (pitch angle or generator torque) is the primary regulation means and is calculated first during joint regulation. The first data mapping relationship is a function or lookup table specifically used in pitch angle priority mode to directly calculate the pitch angle increment reference value based on the power regulation demand intensity index and the current pitch angle. The second data mapping relationship is a function or lookup table used to calculate the generator torque setpoint increment reference value after obtaining the pitch angle increment reference value and combining it with the current power difference. The third data mapping relationship is a function or lookup table specifically used in generator torque priority mode to directly calculate the generator torque setpoint increment reference value based on the power regulation demand intensity index and the current generator torque. The fourth data mapping relationship is a function or lookup table used to calculate the pitch angle increment reference value after obtaining the generator torque increment reference value and combining it with the current power difference.

[0091] In its implementation, the system predefines a regulation strategy mapping table. The core of this table is to map continuous power regulation demand intensity indicators. The system is divided into several discrete intervals, and each interval is assigned a priority configuration for adjustment parameters (i.e., "pitch angle priority mode" or "generator torque priority mode"). The query process is an interval matching logic: model

[0092] in, It is a preset mapping function. For example, when When the value is low, the generator torque priority mode with a faster response may be used, while when the value is high (usually corresponding to high wind speed conditions that require rapid power limiting), the pitch angle priority mode with a more direct effect is used to ensure safety.

[0093] In pitch angle priority mode, the system first determines the pitch angle adjustment amount, and then determines the generator torque compensation adjustment amount based on the effect of the pitch angle adjustment or the remaining power deviation. This is achieved by first utilizing the initial data mapping relationship. Calculate the reference value for pitch angle increment This mapping relationship is usually considered and current pitch angle (Because the effect of pitch angle adjustment is non-linearly related to its current position).

[0094]

[0095] It can be a two-dimensional lookup table or an empirical function, ensuring that the given... This is reasonable and effective under the current operating conditions. Subsequently, the system estimates and adjusts the pitch angle. The impact on power is then calculated, and the remaining power deviation that needs to be compensated by torque adjustment is determined. Second data mapping relationship. Based on the current power difference And the calculation just made Determine the reference value for the generator torque setpoint increment. .

[0096]

[0097] The design needs to take into account the power change caused by the pitch angle change, so as to calculate the torque compensation amount that is coordinated with it.

[0098] In generator torque-priority mode, the system first determines the adjustment amount of the generator torque, and then calculates the auxiliary adjustment amount of the pitch angle based on the expected effect of the torque adjustment. This is first achieved through a third-party data mapping relationship. Calculate the incremental reference value of the generator torque setpoint. Third, consider the data mapping relationship. and current generator torque .

[0099]

[0100] Then, the system predicts torque adjustment. The impact on power is calculated, and the remaining power deviation is determined. Fourth, the data mapping relationship is established. Based on the current power difference and Determine the auxiliary pitch angle increment reference value .

[0101]

[0102] Its function is to perform precise power tracking or speed stabilization compensation.

[0103] Step S40: The set of adjustment instructions is sent to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set.

[0104] It should be noted that the controller refers to the core hardware control unit in a wind turbine generator set that is responsible for receiving instructions from the upper level and driving the actuators to operate. It typically includes the main controller, pitch controller, and converter controller.

[0105] Understandably, the adjustment command set includes reference values ​​for pitch angle increments and generator torque setpoint increments, which are sent to the wind turbine's main controller via a deterministic industrial real-time communication network (such as CANopen, Profinet, or EtherCAT). Upon receiving the command, the main controller first performs safety checks and limiting to ensure the command value remains within the safe range allowed by the turbine's current operating conditions, preventing overshoot or oscillation. Next, the main controller distributes the processed pitch angle command to the independent pitch controllers of each blade, which drive servo motors to adjust the blade angle. Simultaneously, it sends torque commands to the converter control system, which precisely controls the generator's electromagnetic torque by adjusting the switching states of the power electronics in the generator-side and grid-side converters. The entire process requires low latency and high reliability to ensure the real-time performance and stability of power adjustment, ultimately enabling the turbine's actual output power to quickly and smoothly track the optimized target power value.

[0106] In one feasible implementation, after the step of sending the set of adjustment instructions to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set, the method further includes: Collect feedback operating data of the wind turbine generator set after executing the set of adjustment commands; By comparing the feedback operation data with the expected operation data, a quantitative evaluation index for the optimization effect is determined; When the quantitative evaluation index is lower than the preset threshold, the optimized target power value is updated.

[0107] It should be noted that feedback operating data refers to the set of key state parameters of the wind turbine generator set measured and uploaded in real time by sensors after the controller executes adjustment commands. These mainly include actual power, generator speed, actual pitch angle, and wind speed. Expected operating data is the ideal operating state data that the wind turbine should achieve after adjustment, pre-calculated based on the optimized target power value and control commands, such as expected power and expected speed. Quantitative evaluation indicators are one or more numerical indicators used to objectively measure the degree of agreement between actual operating performance and expected targets, such as statistics on power point tracking error or loss function values ​​based on the error.

[0108] In practical implementation, after the adjustment command is issued and sufficient dynamic response time has elapsed (e.g., several control cycles to ensure the system enters a quasi-steady state), the data acquisition system reads feedback operation data within a time window from the SCADA system or controller memory. Let the acquired actual power sequence be... The expected power sequence is ,in, This represents the index of the sampling points within that time window. Additionally, data such as generator speed and pitch angle may also be collected for auxiliary analysis.

[0109] When quantitatively evaluating the optimization effect, power performance can be tracked. Therefore, the normalized root mean square error (NRMSE) can be used as a quantitative metric. The calculation formula is:

[0110] in, This is the rated power of the fan, used to normalize errors and improve the performance of the indicator. It is dimensionless and easy to set universal thresholds. The smaller the value, the better the actual power tracks the expected power. The system calculates this index to quantify the optimization effect of this control.

[0111] The calculated quantitative evaluation indicators With the preset threshold Compare. If This indicates that the control effect has not met expectations, the current target power value may be too aggressive or unrealistic, or the wind turbine environment may have changed. In this case, the system will trigger the logic to update the target power value. The update strategy is to adjust it downwards based on error feedback.

[0112] in, It is the previously used optimized target power value. It is the average value of the actual power within the data collection time window. It is a learning rate or damping coefficient between 0 and 1, used to control the magnitude of adjustment and avoid over-correction. (Updated) It will be used in the next round of power optimization control loop.

[0113] In one feasible implementation, the step of updating the optimized target power value includes: Collect the feedback operation data, status feature data set and optimization effect evaluation results recorded in each optimization process, and construct an optimization effect analysis database; The optimization effects under different wind conditions in the optimization effect analysis database are compared and analyzed to identify the correlation between target wind condition characteristics and poor optimization effects; When the optimization effect of the target wind condition feature is consistently poor, the correlation weights of wind speed and power in the generation rules of the state feature data set are adjusted based on the correlation relationship to obtain the optimization weights. Update the threshold setting regarding the power change rate limit in the calculation rules for the optimized target power value to obtain the optimized threshold; A dynamic optimization strategy is generated based on the optimization weights and the optimization thresholds, and the optimization target power value is updated based on the dynamic optimization strategy.

[0114] It should be noted that the optimization effect analysis database is a structured dataset used to store historical optimization process data. Its records include timestamps for each optimization, feedback operation data, state characteristic data sets, and quantitative evaluation indicators. Target wind characteristics are extracted from the state characteristic data set and are used to characterize typical wind parameter patterns of the wind turbine's operating environment, such as average wind speed, turbulence intensity, and wind shear coefficient within a specific range. Association weights are coefficients assigned in the generation rules to the degree of influence of different input variables (such as wind speeds at different altitudes) on output characteristics (such as equivalent wind speed at the hub). Optimization weights are a new set of weight values ​​obtained by correcting the original association weights based on historical data analysis. Power change rate constraints are upper and lower limits set to constrain the rate of change of the wind turbine's operating stability and grid security when calculating the optimization target power value.

[0115] In its implementation, after each optimization control loop, the system records the key data of that loop into the optimization effect analysis database. The recorded data structure can be formally represented as a single record. :

[0116] in, It's a timestamp. It is a set of state characteristic data (such as 10-minute average wind speed, turbulence intensity, wind direction, etc.). It provides feedback on operational data (such as actual power sequence, speed, etc.). These are quantitative evaluation indicators. Database. It continues to expand over time.

[0117] The system periodically checks the database. Clustering and association analysis are performed. First, based on the state feature data set... Clustering historical records to identify representative target wind condition feature patterns (For example, High wind speed and high turbulence (Medium wind speed, low turbulence). Then, the characteristic pattern for each wind condition is calculated. Corresponding average optimization performance index :

[0118] If a certain wind condition characteristic pattern If the average optimization level is consistently significantly higher than that of other models or the overall average level, then it is determined that this wind condition characteristic is related to poor optimization effect.

[0119] When the problem wind condition is identified Then, the system will adjust the rules accordingly. First, it analyzes whether the original rules for generating the state feature data set under this wind condition fail to accurately reflect the real wind field (e.g., the rules for calculating wind speed at wheel hub height). ,in These are wind speeds at different altitudes. It is the association weight). Through methods such as regression analysis, based on... The weights are recalculated based on the data to obtain the optimized weights. Simultaneously, the dynamic process of power point tracking under this wind condition is analyzed. If the original power change rate limit threshold is found... If the threshold is too stringent, leading to frequent limiting, or too lenient, causing power fluctuations, then the threshold should be adjusted based on the actual achievable smooth rate of change to obtain an optimized threshold. The adjustment strategy is as follows:

[0120] in, The adjustment factor is calculated based on statistics of successful tracking instances in historical data (such as a quantile of the rate of power change).

[0121] The obtained optimization weights and optimized threshold Encapsulate it into a condition-triggered dynamic optimization strategy Its trigger condition is the detection of current state characteristics and Matching. Use when the condition is met. New rules are used to calculate the optimized target power value. This allows for the coverage and optimization of existing fixed rules.

[0122] This embodiment provides a method for optimizing the power output of wind turbine generator sets. It corrects the time-series operating data of the wind turbine generator set during operation, removes abnormal data, and performs signal synchronization and wind speed spatial correlation correction to obtain a set of state characteristic data that accurately reflects the relationship between the generator set's operating status and power output. Based on this characteristic data set, by matching similar historical operating segments and conducting power stability assessment, it determines the optimized target power value achievable by the generator set under the current wind conditions. It calculates the difference between the target power and the actual power and performs trend prediction to generate a power demand trend. According to a preset adjustment strategy mapping table, it obtains a set of adjustment commands for the pitch angle and generator torque, and sends these commands to the generator set controller to adjust its operating power. Through this method, the method achieves dynamic and precise optimization of the generator set's power output, improving power generation efficiency and grid connection stability.

[0123] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the wind turbine power optimization method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0124] This application also provides a wind turbine generator power optimization system; please refer to [reference needed]. Figure 3 The wind turbine power optimization system includes: Data correction module 10 is used to correct the time-series operation data set collected during the operation of the wind turbine generator set to obtain a set of state characteristic data reflecting the relationship between the unit's operating status and power output. The target calculation module 20 is used to obtain the optimized target power value that the wind turbine generator can achieve under the current wind conditions based on the state feature data set. The instruction generation module 30 is used to perform trend prediction based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, obtain the power demand trend, and obtain a set of adjustment instructions for the control parameters of the wind turbine generator set based on the power demand trend. The instruction issuing module 40 is used to issue the set of adjustment instructions to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set.

[0125] In one feasible implementation, the data correction module 10 is further used to perform abnormal data removal and signal synchronization processing on the original time-series operation data in the time-series operation data set collected during the operation of the wind turbine generator set, so as to obtain a regular time-series data set. Spatial correlation correction and turbulence intensity calculation are performed on the wind speed data in the regularized time series dataset to obtain equivalent inflow wind speed data; The equivalent inflow wind speed data is correlated and combined with the generator speed and pitch angle in the regularized time series data set to obtain a state feature data set.

[0126] In one feasible implementation, the data correction module 10 is further used to determine the wind speed data in the regularized time series data set, wherein the wind speed data is the wind speed measurement data at the hub of the wind turbine generator set and the preset spatial distribution points. Based on the computational fluid dynamics model, the wind speed measurement data is spatially correlated and corrected to obtain a corrected wind speed data sequence. Turbulence intensity is calculated on the corrected wind speed data sequence to obtain turbulence intensity values ​​that reflect the characteristics of wind speed fluctuations; The turbulence intensity value and the wind speed measurement data are weighted and fused to obtain the equivalent inflow wind speed data.

[0127] In one feasible implementation, the target calculation module 20 is further configured to perform similarity calculation between the wind condition features in the state feature data set and the historical operation segments in the historical database, and combine the historical operation segments whose similarity is within the target similarity interval to obtain a historical power data set. A power stability assessment is performed on the historical power data set to obtain a stability assessment value; Power sequences whose stable evaluation values ​​are greater than preset stable evaluation values ​​are used as candidate optimized power data; Based on the current power grid dispatch instructions or power change rate limits, boundary constraints are applied to the candidate optimized power data to obtain the optimized target power value that the wind turbine can achieve under the current wind conditions.

[0128] In one feasible implementation, the target calculation module 20 is further configured to perform time-scale normalization processing on power data with different sampling frequencies in the historical power data set to generate a power data sequence with a unified time reference. Power fluctuation features are extracted from the power data sequence of the unified time base to identify abnormal fluctuation segments where the power change rate exceeds a preset threshold. The abnormal fluctuation segments are removed from the historical power data set.

[0129] In one feasible implementation, the instruction generation module 30 is further configured to determine the power difference between the optimized target power value and the actual power value of the wind turbine generator set; Based on the power difference, time-segmented accumulation and trend prediction are performed to obtain the power demand trend; The power regulation demand intensity index is obtained based on the power demand trend. Based on the power regulation demand intensity index, the preset regulation strategy mapping table is queried to determine the reference value of the pitch angle increment and the reference value of the generator torque setpoint increment. The pitch angle increment reference value and the generator torque setpoint increment reference value are coupled to obtain a set of adjustment instructions for the wind turbine generator control parameters.

[0130] In one feasible implementation, the instruction generation module 30 is further configured to query a preset adjustment strategy mapping table based on the power adjustment demand intensity index to obtain the query result. The preset adjustment strategy mapping table defines the correspondence between different intensity index ranges and adjustment parameter priority configurations. Based on the query results, the priority of the adjustment parameters corresponding to the current intensity index range is determined. The priority configuration of the adjustment parameters includes either the pitch angle priority mode or the generator torque priority mode. In the pitch angle priority mode, based on the power regulation demand intensity index and the current pitch angle, the pitch angle increment reference value is obtained through the first data mapping relationship. By combining the power difference and the pitch angle increment reference value, the generator torque setpoint increment reference value is obtained through the second data mapping relationship; In the generator torque priority mode, based on the power regulation demand intensity index and the current generator torque, the incremental reference value of the generator torque setpoint is obtained through a third data mapping relationship. By combining the power difference and the incremental reference value of the generator torque setpoint, the incremental reference value of the pitch angle is obtained through the fourth data mapping relationship.

[0131] In one feasible implementation, the instruction issuing module 40 is also used to collect feedback operating data of the wind turbine generator set after the set of adjustment instructions is executed; By comparing the feedback operation data with the expected operation data, a quantitative evaluation index for the optimization effect is determined; When the quantitative evaluation index is lower than the preset threshold, the optimized target power value is updated.

[0132] In one feasible implementation, the instruction issuing module 40 is also used to collect the feedback running data, status feature data set and optimization effect evaluation results recorded in each optimization process, and to build an optimization effect analysis database; The optimization effects under different wind conditions in the optimization effect analysis database are compared and analyzed to identify the correlation between target wind condition characteristics and poor optimization effects; When the optimization effect of the target wind condition feature is consistently poor, the correlation weights of wind speed and power in the generation rules of the state feature data set are adjusted based on the correlation relationship to obtain the optimization weights. Update the threshold setting regarding the power change rate limit in the calculation rules for the optimized target power value to obtain the optimized threshold; A dynamic optimization strategy is generated based on the optimization weights and the optimization thresholds, and the optimization target power value is updated based on the dynamic optimization strategy.

[0133] The wind turbine power optimization system provided in this application, employing the wind turbine power optimization method described in the above embodiments, can solve the technical problem of large fluctuations in wind turbine power output. Compared with the prior art, the beneficial effects of the wind turbine power optimization system provided in this application are the same as those of the wind turbine power optimization method provided in the above embodiments, and other technical features of the wind turbine power optimization system are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0134] This application provides a wind turbine power optimization device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the wind turbine power optimization method in the above embodiment 1.

[0135] The following is for reference. Figure 4 The diagram illustrates a structural schematic suitable for implementing the wind turbine power optimization device in the embodiments of this application. The wind turbine power optimization device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, tablets, digital radio receivers, PDAs (Personal Digital Assistants), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The wind turbine power optimization device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0136] like Figure 4As shown, the wind turbine power optimization device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the wind turbine power optimization device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, LCDs (Liquid Crystal Displays), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the wind turbine power optimization equipment to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a wind turbine power optimization equipment with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0137] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0138] The wind turbine power optimization device provided in this application, employing the wind turbine power optimization method described in the above embodiments, can solve the technical problem of wind turbine power optimization. Compared with the prior art, the beneficial effects of the wind turbine power optimization device provided in this application are the same as those of the wind turbine power optimization method provided in the above embodiments, and other technical features of the wind turbine power optimization device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0139] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0140] 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.

[0141] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the wind turbine generator power optimization method in the above embodiments.

[0142] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory or Flash Memory), optical fibers, CD-ROM (CD-Read Only Memory), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0143] The aforementioned computer-readable storage medium may be included in the wind turbine power optimization device; or it may exist independently and not be assembled into the wind turbine power optimization device.

[0144] The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by the wind turbine power optimization device, the wind turbine power optimization device: corrects the time-series operation data set collected during the operation of the wind turbine to obtain a set of state characteristic data reflecting the relationship between the unit's operating status and power output. Based on the state feature data set, the optimized target power value that the wind turbine generator can achieve under the current wind conditions is obtained; Based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, a trend prediction is performed to obtain the power demand trend, and a set of adjustment instructions for the control parameters of the wind turbine generator set is obtained based on the power demand trend. The set of adjustment instructions is sent to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set.

[0145] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0146] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0147] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0148] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described wind turbine power optimization method, and is capable of solving the technical problem of wind turbine power optimization. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the wind turbine power optimization method provided in the above embodiments, and will not be repeated here.

[0149] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the wind turbine generator power optimization method described above.

[0150] The computer program product provided in this application can solve the technical problem of large power output fluctuations in wind turbine generators. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the wind turbine generator power optimization method provided in the above embodiments, and will not be repeated here.

[0151] The above are only some embodiments of this application and do not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for optimizing the power output of a wind turbine generator set, characterized in that, The wind turbine generator power optimization method includes: The time-series operation data collected during the operation of the wind turbine generator is corrected to obtain a set of state characteristic data reflecting the relationship between the unit's operating status and power output; Based on the state feature data set, the optimized target power value that the wind turbine generator can achieve under the current wind conditions is obtained; Based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, a trend prediction is performed to obtain the power demand trend, and a set of adjustment instructions for the control parameters of the wind turbine generator set is obtained based on the power demand trend. The set of adjustment instructions is sent to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set.

2. The method as described in claim 1, characterized in that, The step of correcting the time-series operational data set collected during the operation of the wind turbine generator set to obtain a characteristic data set reflecting the relationship between the unit's operating status and power output includes: Anomaly removal and signal synchronization are performed on the original time-series operation data collected during the operation of the wind turbine generator set to obtain a regular time-series data set. Spatial correlation correction and turbulence intensity calculation are performed on the wind speed data in the regularized time series dataset to obtain equivalent inflow wind speed data; The equivalent inflow wind speed data is correlated and combined with the generator speed and pitch angle in the regularized time series data set to obtain a state feature data set.

3. The method as described in claim 2, characterized in that, The step of performing spatial correlation correction and turbulence intensity calculation on the wind speed data in the regularized time-series dataset to obtain equivalent inflow wind speed data includes: Determine the wind speed data in the regularized time-series data set, wherein the wind speed data is wind speed measurement data at the hub of the wind turbine generator set and at preset spatial distribution points; Based on the computational fluid dynamics model, the wind speed measurement data is spatially correlated and corrected to obtain a corrected wind speed data sequence. Turbulence intensity is calculated on the corrected wind speed data sequence to obtain turbulence intensity values ​​that reflect the characteristics of wind speed fluctuations; The turbulence intensity value and the wind speed measurement data are weighted and fused to obtain the equivalent inflow wind speed data.

4. The method as described in claim 1, characterized in that, The step of obtaining the optimized target power value that the wind turbine can achieve under the current wind conditions based on the state feature data set includes: The wind condition features in the state feature data set are compared with the historical operation segments in the historical database. The historical operation segments with similarity within the target similarity range are combined to obtain the historical power data set. A power stability assessment is performed on the historical power data set to obtain a stability assessment value; Power sequences whose stable evaluation values ​​are greater than preset stable evaluation values ​​are used as candidate optimized power data; Based on the current power grid dispatch instructions or power change rate limits, boundary constraints are applied to the candidate optimized power data to obtain the optimized target power value that the wind turbine can achieve under the current wind conditions.

5. The method as described in claim 4, characterized in that, Before the step of performing a power stability assessment on the historical power data set to obtain a stability assessment value, the method further includes: The power data with different sampling frequencies in the historical power data set are normalized to a time scale to generate a power data sequence with a unified time base. Power fluctuation features are extracted from the power data sequence of the unified time base to identify abnormal fluctuation segments where the power change rate exceeds a preset threshold. The abnormal fluctuation segments are removed from the historical power data set.

6. The method as described in claim 1, characterized in that, The step of predicting the power demand trend based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, and obtaining the set of adjustment instructions for the control parameters of the wind turbine generator set based on the power demand trend includes: Determine the power difference between the optimized target power value and the actual power value of the wind turbine generator set; Based on the power difference, time-segmented cumulative summation and trend prediction are performed to obtain the power demand trend; The power regulation demand intensity index is obtained based on the power demand trend. Based on the power regulation demand intensity index, the preset regulation strategy mapping table is queried to determine the reference value of the pitch angle increment and the reference value of the generator torque setpoint increment. The pitch angle increment reference value and the generator torque setpoint increment reference value are coupled to obtain a set of adjustment instructions for the wind turbine generator control parameters.

7. The method as described in claim 6, characterized in that, The step of determining the reference values ​​for the pitch angle increment and the generator torque setpoint increment based on the power regulation demand intensity index and querying the preset regulation strategy mapping table includes: Based on the power regulation demand intensity index, a preset regulation strategy mapping table is queried to obtain the query result. The preset regulation strategy mapping table defines the correspondence between different intensity index ranges and the priority configuration of regulation parameters. Based on the query results, the priority of the adjustment parameters corresponding to the current intensity index range is determined. The priority configuration of the adjustment parameters includes either the pitch angle priority mode or the generator torque priority mode. In the pitch angle priority mode, based on the power regulation demand intensity index and the current pitch angle, the pitch angle increment reference value is obtained through the first data mapping relationship. By combining the power difference and the pitch angle increment reference value, the generator torque setpoint increment reference value is obtained through the second data mapping relationship; In the generator torque priority mode, based on the power regulation demand intensity index and the current generator torque, the incremental reference value of the generator torque setpoint is obtained through a third data mapping relationship. By combining the power difference and the incremental reference value of the generator torque setpoint, the incremental reference value of the pitch angle is obtained through the fourth data mapping relationship.

8. The method as described in claim 1, characterized in that, After the step of sending the set of adjustment instructions to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set, the method further includes: Collect feedback operating data of the wind turbine generator set after executing the set of adjustment commands; By comparing the feedback operation data with the expected operation data, a quantitative evaluation index for the optimization effect is determined; When the quantitative evaluation index is lower than the preset threshold, the optimized target power value is updated.

9. The method as described in claim 8, characterized in that, The steps for updating the optimized target power value include: Collect the feedback operation data, status feature data set and optimization effect evaluation results recorded in each optimization process, and construct an optimization effect analysis database; The optimization effects under different wind conditions in the optimization effect analysis database are compared and analyzed to identify the correlation between target wind condition characteristics and poor optimization effects; When the optimization effect of the target wind condition feature is consistently poor, the correlation weights of wind speed and power in the generation rules of the state feature data set are adjusted based on the correlation relationship to obtain the optimization weights. Update the threshold setting regarding the power change rate limit in the calculation rules for the optimized target power value to obtain the optimized threshold; A dynamic optimization strategy is generated based on the optimization weights and the optimization thresholds, and the optimization target power value is updated based on the dynamic optimization strategy.

10. A wind turbine generator power optimization system, characterized in that, The wind turbine power optimization system includes: The data correction module is used to correct the time-series operating data set collected during the operation of the wind turbine generator set, and obtain a set of state characteristic data reflecting the relationship between the unit's operating status and power output. The target calculation module is used to obtain the optimized target power value that the wind turbine can achieve under the current wind conditions based on the state feature data set. The instruction generation module is used to predict the power demand trend based on the power difference between the optimized target power value and the actual power value of the wind turbine generator set, and to obtain the power demand trend. Based on the power demand trend, it obtains a set of adjustment instructions for the control parameters of the wind turbine generator set. The instruction issuing module is used to issue the set of adjustment instructions to the controller of the wind turbine generator set to adjust the operating power of the wind turbine generator set.

Images