A method, system, device and medium for wind power variable pitch and rotational speed cooperative control

Abstract

The present application relates to the field of wind power research, and discloses a wind power variable pitch and rotating speed cooperative control method, system, device and medium, comprising: extracting the operation space-time working condition characteristics in the target time window according to the obtained operation state data and incoming flow environment data, obtaining the working condition space-time comprehensive characteristic index; calculating the unit working condition average power value corresponding to the target time window, forming the working condition level power characteristic set, and obtaining the working condition power difference index according to the discrete degree of data; performing feature matching on the obtained index, calculating the cooperative control fusion weight on the basis of the variable pitch channel basic weight and the rotating speed or torque channel basic weight; generating a cooperative control instruction according to the power grid dispatching information and the unit control constraint condition, and the wind turbine unit executes the cooperative control instruction to complete the cooperative control of variable pitch and rotating speed. The present application reduces power fluctuation and load, reduces variable pitch wear and improves robustness under the condition of meeting grid connection and safety constraints.

Description

Technical Field

[0001] This invention relates to the field of wind power research technology, and in particular to a method, system, equipment and medium for coordinated control of wind turbine pitch and rotational speed. Background Technology

[0002] Wind turbine generators typically employ a zoned control strategy: in low-wind-speed areas, energy capture is enhanced through constant optimal tip speed ratio or power torque control; in rated areas, pitch control is the primary method to maintain rated power and limit speed. Existing projects often design the "pitch loop" and "speed / torque loop" relatively independently: the pitch angle is used to adjust aerodynamic thrust and power, while the generator's electromagnetic torque or speed is used for fine-tuning power and balancing inertia.

[0003] However, under conditions such as strong turbulence, rapid wind direction shifts and yaw mismatch, wake interference, and grid-side power limiting / peak shaving, target conflicts and coupling competition arise between the two loops. On the one hand, frequent pitch control leads to actuator heat / wear and structural load fluctuations; on the other hand, relying solely on speed / torque introduces power oscillations and degrades grid connection quality. Meanwhile, measurement and modeling also present challenges: ① Power measured at the grid connection point includes both converter and pitch losses; ② Individual differences caused by air density, temperature, and station topology result in caliber drift in the application of the "reference power curve"; ③ Incoming flow sensors (nacelle wind speed / direction, airborne lidar) may be unavailable during salt spray, icing, or maintenance, causing operational condition assessment and feedforward control to fail. Therefore, existing technologies lack a collaborative control mechanism that can quantitatively balance spatiotemporal operational condition variations and power consistency / load, and maintain robustness under sensor anomalies and grid constraints. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method, system, device, and medium for coordinated control of wind turbine pitch and speed, which solves the problems that existing independent pitch and speed control of wind turbines cannot balance power stability and load suppression under complex operating conditions, and lacks robust handling of loss stripping and sensor failure.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] In a first aspect, the present invention provides a method for coordinated control of wind turbine pitch and rotational speed, comprising:

[0008] Based on the acquired operating status data and incoming flow environment data of the wind turbine generator set, the spatiotemporal operating condition characteristics of the wind turbine generator set within the target time window are extracted to obtain the spatiotemporal comprehensive characteristic index of the operating condition.

[0009] Based on the operating status data, incoming flow environment data, and reference power curve, the average power value per unit operating condition corresponding to the target time window is calculated to form an operating condition-level power feature set, and the operating condition power difference index is obtained based on the dispersion of the data in the set.

[0010] Feature matching is performed on the spatiotemporal comprehensive characteristic index of the operating condition and the power difference index of the operating condition. Based on the basic weight of the pitch channel and the basic weight of the speed or torque channel, the collaborative control fusion weight is calculated.

[0011] Based on the collaborative control fusion weights, grid dispatch information, and unit control constraints, collaborative control commands are generated, and the wind turbine generator sets execute the collaborative control commands to complete the collaborative control of pitch and speed.

[0012] As a preferred embodiment of the wind turbine pitch and speed coordinated control method described in this invention, the operating status data of the wind turbine generator set includes rotor speed, pitch angle, generator active power, electromagnetic torque, nacelle yaw angle, converter load rate, and pitch system duty cycle; the incoming flow environment data includes wind speed, wind direction, and turbulence intensity.

[0013] Different converter load rates and pitch system duty cycles correspond to different base loss power models, which are used to separate base losses from measured electrical power to obtain net output power.

[0014] As a preferred embodiment of the wind power pitch and speed coordinated control method described in this invention, the calculation of the spatiotemporal comprehensive characteristic index of the operating condition includes:

[0015] ;

[0016] in, Indicates the spatiotemporal comprehensive characteristics of the operating condition. Indicates the end time of the target time window. Indicates the start time of the target time window. This represents the cumulative change in wind direction within the target time window. This represents the average wind speed within the target time window. This represents the average absolute yaw mismatch angle within the target time window. This represents the average turbulence intensity within the target time window. , , , , All of these represent the spatiotemporal feature weighting coefficients.

[0017] As a preferred embodiment of the wind power pitch and speed coordinated control method described in this invention, the obtained operating condition power difference index includes:

[0018] The current base loss power is determined based on the converter load rate and the pitch system duty cycle, and the current base loss power is subtracted from the measured power to obtain the net output power.

[0019] The net output power is integrated within the target time window and divided by the window duration to obtain the average power value under unit operating conditions.

[0020] A power feature set is constructed by dividing the data into two-dimensional boxes based on wind speed and wind direction. The average power value of the unit operating condition within the same box is aggregated based on the reference power curve after density correction. The number of samples of the average power value of the unit operating condition within each box is counted.

[0021] If the number of samples exceeds a preset threshold, the standard deviation of the average power value of all units under operating conditions within the sub-bin is calculated, and the standard deviation is used as the power difference index of the current sub-bin to characterize the dispersion of power output under operating conditions; if the number of samples is less than the preset threshold, the power difference index is not calculated for the time being.

[0022] As a preferred embodiment of the wind turbine pitch and speed coordinated control method described in this invention, the unit control constraints include upper and lower speed limits, upper limits for pitch angle and pitch rate, generator torque / current limits, nacelle / blade root / tower base load limits, and grid connection specifications; the coordinated control commands include a coordinated control priority table, pitch angle reference, speed reference, generator torque reference, pitch rate threshold, and control entry and exit timing; the fusion weights are used to update the priority or control gain of the pitch and speed or torque channels in real time.

[0023] As a preferred embodiment of the wind power pitch and speed coordinated control method described in this invention, the modification of the coordinated control command based on grid-connected power constraints includes:

[0024] When there is a power grid power limiting order or an energy storage peak shaving order, the priority is redistributed according to the channel weight after gating. While meeting the power limiting constraints, priority is given to minimizing the weighted combined value of the operating power difference index and the structural load index. The structural load index includes at least one of the blade root bending moment, tower bottom bending moment and drive chain torque.

[0025] As a preferred embodiment of the wind power pitch and speed coordinated control method described in this invention, it further includes:

[0026] When the wind speed and direction sensors or wind-measuring lidar of the unit are unavailable, the incoming flow data of the neighboring units and the wind measurement tower are obtained within the communication range of the wind farm. The data are then weighted according to the attenuation relationship of the distance to the unit and the consistency of the relative wind direction to estimate the wind speed, wind direction and turbulence of the unit. This data is used as substitute incoming flow data to calculate the spatiotemporal comprehensive characteristic index of the operating condition.

[0027] Once the sensors in this unit are available again, a deviation compensation amount is established based on the difference between the restored incoming flow data and the alternative incoming flow data. The wind speed, wind direction, and the difference between the average power value and the power under the unit operating condition are then re-estimated based on the deviation compensation amount to achieve smooth switching and accuracy correction.

[0028] Secondly, the present invention provides a system for coordinated control of wind turbine pitch and rotational speed, comprising:

[0029] The sensing module is used to acquire the operating status data and incoming flow environment data of the wind turbine generator set;

[0030] The operating condition characteristic calculation module is used to extract the spatiotemporal operating condition characteristics of the wind turbine generator set within the target time window based on the operating status data and the incoming flow environment data, and obtain the spatiotemporal comprehensive characteristic index of the operating condition.

[0031] The power characteristic analysis module is used to calculate the average power value per unit operating condition corresponding to the target time window based on the operating status data, incoming flow environment data and reference power curve, form an operating condition-level power characteristic set, and obtain the operating condition power difference index based on the dispersion of the data in the set.

[0032] The weight fusion module is used to perform feature matching on the spatiotemporal comprehensive characteristic index of the operating condition and the power difference index of the operating condition, and to calculate the collaborative control fusion weight based on the preset basic weight of the pitch channel and the basic weight of the speed or torque channel.

[0033] The control command generation and execution module is used to generate collaborative control commands based on the collaborative control fusion weights, grid dispatch information and unit control constraints. The wind turbine generator sets execute the collaborative control commands to complete the collaborative control of pitch and speed.

[0034] Thirdly, the present invention provides an electronic device, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor executes the computer-executable instructions to implement the steps of a method for coordinated control of wind turbine pitch and rotational speed.

[0035] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, implement the steps of a method for coordinated control of wind turbine pitch and rotational speed.

[0036] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention provides a method, system, device, and medium for coordinated control of wind turbine pitch and speed. By constructing a coordinated control mechanism that integrates spatiotemporal operating conditions and power differences, it achieves dynamic weight allocation and priority adjustment of the pitch and speed loops, effectively reducing power fluctuations and structural loads under complex inflow conditions, and reducing frequent operation and wear of the pitch actuator. This invention establishes a basic loss power model with converter load rate, pitch duty cycle, and ambient temperature as independent variables, decomposing the measured active power into net output power, significantly reducing power caliber deviations caused by different operating conditions / machine types / temperatures, and providing a consistent benchmark for subsequent statistics and optimization. Furthermore, when sensors fail, the method of this invention can intelligently fuse data from neighboring machines and meteorological towers and perform deviation compensation, significantly enhancing the adaptability and robustness of the unit under abnormal operating conditions, while better meeting the coordinated control requirements under grid peak shaving and power limiting constraints. Attached Figure Description

[0037] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a schematic diagram of the overall flow logic of a wind power pitch and speed coordinated control method provided in an embodiment of the present invention. Detailed Implementation

[0039] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0040] Example 1, referring to Figure 1 As one embodiment of the present invention, a method for coordinated control of wind turbine pitch and rotational speed is provided, such as... Figure 1 The specific steps shown are as follows:

[0041] S100: Based on the acquired operating status data and incoming flow environment data of the wind turbine generator set, extract the spatiotemporal operating condition characteristics of the wind turbine generator set within the target time window to obtain the spatiotemporal comprehensive characteristic index of the operating condition.

[0042] S200: Based on operating status data, incoming flow environment data and reference power curve, calculate the average power value per unit operating condition corresponding to the target time window, form a set of operating condition-level power characteristics, and obtain the operating condition power difference index based on the dispersion of the data in the set.

[0043] S300: Perform feature matching on the spatiotemporal comprehensive characteristic index of operating conditions and the power difference index of operating conditions, and calculate the collaborative control fusion weight based on the preset basic weight of pitch channel and basic weight of speed or torque channel.

[0044] S400: Based on the collaborative control fusion weights, grid dispatch information and unit control constraints, a collaborative control command is generated. The wind turbine generator executes the collaborative control command to complete the collaborative control of pitch and speed.

[0045] It should be noted that in existing wind power control technologies, pitch control and speed / torque control are usually independent of each other. Under complex operating conditions such as turbulence, rapid changes in wind direction, wake interference, or grid power limitation, the pitch actuator is prone to frequent operation, increased structural load fluctuations, power output oscillations, and degraded grid connection quality due to conflicting objectives. At the same time, it is difficult to accurately separate the converter and pitch system losses included in the measured power, resulting in power control reference drift. Furthermore, there is a lack of reliable alternative data sources when the incoming flow sensor fails, resulting in insufficient robustness of the unit under abnormal operating conditions.

[0046] It should be noted that steps S100-S400 above, by constructing a collaborative control mechanism that integrates spatiotemporal operating condition characteristics and power differences, achieve dynamic weight allocation and priority adjustment of the pitch and speed loops, effectively reducing power fluctuations and structural loads under complex inflow conditions, and reducing frequent operation and wear of the pitch actuator. This invention establishes a basic loss power model with converter load rate, pitch duty cycle, and ambient temperature as independent variables, decomposing the measured active power into net output power, significantly reducing power caliber deviations caused by different operating conditions / machine types / temperatures, and providing a consistent benchmark for subsequent statistics and optimization. Furthermore, when sensors fail, the method of this invention can intelligently fuse data from neighboring units and meteorological towers and perform deviation compensation, significantly enhancing the adaptability and robustness of the unit under abnormal operating conditions, while better meeting the coordinated control requirements under grid peak shaving and power limiting constraints.

[0047] In this embodiment of the invention, step S100 extracts the spatiotemporal operating condition characteristics of the wind turbine generator set within the target time window based on the acquired operating status data and incoming flow environment data, and obtains the spatiotemporal comprehensive characteristic index of the operating condition, including the following sub-steps A1 and A2:

[0048] In A1: Obtain the operating status data and incoming flow environment data of the wind turbine generator set;

[0049] Specifically, the operating status data of wind turbine generator sets include rotor speed, pitch angle, generator active power, electromagnetic torque, nacelle yaw angle, converter load rate, and pitch system duty cycle; the incoming flow environment data includes wind speed, wind direction, and turbulence intensity.

[0050] Specifically, the data sampling frequency is no less than 10Hz, and the timestamps are unified to the same base and aligned by interpolation;

[0051] In this embodiment of the invention, different converter load rates and pitch system duty cycles correspond to different basic loss power models. The converter load rate is the ratio of the current output active power to the rated active power, the pitch system duty cycle is the ratio of the actuator drive time within the window to the window duration, and the basic loss power model is used to separate the basic loss from the measured power to obtain the net output power.

[0052] Specifically, the basic loss power model is obtained through a regression model, which takes at least the converter load rate, pitch duty cycle and nacelle ambient temperature as inputs; the update cycle of the basic loss power model is no more than 24 hours, and a sliding window is used for retraining to suppress drift.

[0053] In A2: Extract the spatiotemporal operating condition characteristics of the wind turbine generator set within the target time window to obtain the spatiotemporal comprehensive characteristic index of the operating condition;

[0054] Specifically, the operating conditions of the wind turbine generator set during the target time window are extracted, including the end time, start time, cumulative change in wind direction, average wind speed, average absolute yaw mismatch angle, and average turbulence intensity of the target time window.

[0055] Specifically, the calculation of the spatiotemporal comprehensive characteristic index of the operating condition includes:

[0056] ;

[0057] in, Indicates the spatiotemporal comprehensive characteristics of the operating condition. Indicates the end time of the target time window. Indicates the start time of the target time window. This represents the cumulative change in wind direction within the target time window. This represents the average wind speed within the target time window. This represents the average absolute yaw mismatch angle within the target time window. This represents the average turbulence intensity within the target time window. , , , , All of these represent the spatiotemporal feature weighting coefficients.

[0058] It should be noted that the above step S100 extracts spatiotemporal operating condition characteristic indicators that can comprehensively reflect wind direction changes, turbulence intensity, yaw mismatch and time span based on the incoming flow environment data within the target time window, thereby quantitatively assessing the unsteadiness and complexity of the current operating condition.

[0059] In this embodiment of the invention, step S200 calculates the average power value per unit operating condition corresponding to the target time window based on operating status data, incoming flow environment data, and reference power curve, forming a set of operating condition-level power characteristics, and obtains the operating condition power difference index based on the dispersion of the data within the set, including the following sub-steps B1~B4:

[0060] In B1: The current base loss power is determined based on the converter load rate, nacelle ambient temperature and pitch system duty cycle, and the current base loss power is subtracted from the measured power to obtain the net output power;

[0061] In B2: Integrate the net output power within the target time window and divide by the window duration to obtain the unit operating condition average power value. The formula is expressed as:

[0062] ;

[0063] in, Indicates the length of the target window. This indicates the net output power.

[0064] In B3: A power feature set is constructed by dividing the data into two-dimensional boxes based on wind speed and wind direction. The average power value of the unit operating condition within the same box is collected based on the corrected reference power curve. The number of samples of the average power value of the unit operating condition within each box is counted. The width of the wind speed box is 1 m / s and the width of the wind direction box is 10°.

[0065] Specifically, the correction of the reference power curve based on air density and temperature includes: calculating the current air density based on the real-time measurements of cabin temperature and air pressure sensors, and correcting the reference power curve under the standard air density according to the scaling relationship, thereby obtaining a power reference curve with consistent caliber that is applicable to the current actual atmospheric conditions, providing an accurate basis for the subsequent collection and comparison of average power under unit operating conditions.

[0066] In B4: If the number of samples exceeds the preset threshold, the standard deviation of the average power value of all units under the operating conditions in the sub-bin is calculated, and the standard deviation is used as the power difference index of the current sub-bin to characterize the dispersion of power output under the operating conditions; if the number of samples is less than the preset threshold, the power difference index is not calculated for the time being.

[0067] In this embodiment of the invention, the preset threshold is 20. This number is the minimum requirement to ensure that the power data in the sub-bin has enough samples to calculate the reliable standard deviation. This can avoid the distortion or excessive fluctuation of the difference index due to too few samples, and can also take into account the data accumulation speed of common operating conditions in the actual operation of wind farms, so as to ensure that the control system can update the power difference index in a timely and stable manner.

[0068] In this embodiment of the invention, the standard deviation of the average power value of all unit operating conditions within the sub-compartment is calculated using the following formula:

[0069] ;

[0070] in, This represents the power difference index under the current compartment conditions, where n represents the number of samples within the same compartment. This represents the average value of the current bins.

[0071] It should be noted that step S200 above calculates the average power under unit operating conditions and constructs a power feature set. It further combines the net power output analysis to obtain the power difference index, which effectively reveals the degree of dispersion of the unit's power output under the same incoming flow conditions. This provides a quantifiable basis for identifying performance degradation, wake effects, or control mismatch.

[0072] In this embodiment of the invention, step S300 performs feature matching on the spatiotemporal comprehensive characteristic index of operating conditions and the power difference index of operating conditions. Based on the preset basic weights of the pitch channel and the basic weights of the speed or torque channel, the calculation of the collaborative control fusion weights includes:

[0073] Specifically, the steps for feature matching of the spatiotemporal comprehensive characteristic index of operating conditions and the power difference index of operating conditions include:

[0074] The two types of indicators were standardized according to historical statistical ranges and divided into three levels: low, medium, and high (corresponding to stable / normal / strong operating conditions; small / medium / large power differences).

[0075] Establish a matching matrix: When the operating conditions are strong and the power difference is large, increase the priority of the pitch channel to quickly stabilize the angle and suppress the load; when the operating conditions are stable and the power difference is large, increase the priority of the speed / torque channel to reduce pitch action and wear; the remaining combinations should gradually transition from pitch-based to speed / torque-based.

[0076] To avoid frequent jitter, a hysteresis loop is set for gear shifting; shifting is only allowed when any indicator continuously crosses the threshold for a certain number of sampling cycles.

[0077] If the structural load is close to the limit or there is a power reduction command on the grid side, a penalty or bonus is added to the matching matrix result to tilt the priority towards the channel that is more conducive to load limiting and power stabilization.

[0078] The channel priority and control gain ratio at the current moment are obtained; the underlying control loop allocates the adjustment range of pitch, speed / torque accordingly.

[0079] Specifically, the normalized spatiotemporal comprehensive characteristic index of the operating condition and the power difference index of the operating condition are fused in multiple dimensions. The weighted linear combination is then input into the activation function for nonlinear mapping. Based on the preset basic weights of the pitch channel and the basic weights of the speed or torque channel, a collaborative control fusion weight between 0 and 1 is generated. This collaborative control fusion weight can dynamically reflect the comprehensive impact of the spatiotemporal unsteady characteristics and the power output dispersion on the control channel under the current operating condition, thereby achieving adaptive matching based on real-time operating condition characteristics.

[0080] Specifically, collaborative control fusion weights W The calculation is expressed as:

[0081] ;

[0082] in, This represents the activation function mapping operation. This represents the normalized base weights of the pitch channel. This represents the base weight of the speed or torque channel after normalization. This represents the power difference index after normalization. This represents the spatiotemporal comprehensive characteristic index of the operating conditions after normalization. , , , All represent adjustment coefficients. Preset base weights for the pitch channel. Basic weights of speed or torque channels The values ​​of both are between 0 and 1, and the sum of the two is 1;

[0083] Specifically, the collaborative control fusion weights are used to generate or update the priorities or control gains of the pitch and speed or torque channels in real time; among them, the update of priorities or control gains adopts upper and lower hysteresis to avoid frequent switching.

[0084] It should be noted that step S300 above matches and fuses spatiotemporal operating condition characteristics with power difference index, and dynamically generates collaborative control fusion weights on the basis of preset pitch and speed channel base weights, thereby enabling adaptive adjustment of the priority and gain allocation of pitch and speed control loops according to different operating condition characteristics, thereby enhancing the system's coordination capability under complex conditions.

[0085] In this embodiment of the invention, step S400 generates a collaborative control command based on the collaborative control fusion weight, grid dispatch information, and unit control constraints. The wind turbine generator executes the collaborative control command to complete the collaborative control of pitch and speed, including the following sub-steps D1 to D4:

[0086] In D1: Based on the collaborative control fusion weights, grid dispatch information, and unit control constraints, collaborative control commands are generated;

[0087] Specifically, the unit control constraints include upper and lower speed limits, upper limits for pitch angle and pitch rate, generator torque / current limits, nacelle / blade root / tower base load limits, and grid connection specifications; the generated collaborative control commands include a collaborative control priority table, pitch angle reference, speed reference, generator torque reference, pitch rate threshold, and control entry and exit timing.

[0088] Specifically, when control objectives conflict, the priority table and gating gain are used, and saturation protection and anti-integral wind strategies are adopted. When any hard constraint is triggered, the grid connection specifications are satisfied first, and then the remaining degrees of freedom are allocated according to the updated collaborative control fusion weights.

[0089] In D2: Modifying the coordinated control command based on grid-connected power constraints includes: when there is a grid power limiting command or energy storage peak shaving command, the priority is reallocated according to the channel weight after gating. While satisfying the power limiting constraint, the weighted combined value of the operating condition power difference index and the structural load index is minimized. The structural load index includes at least one of the blade root bending moment, tower bottom bending moment and drive chain torque.

[0090] Specifically, the weighted composite value of the power difference index under operating conditions and the structural load index. J The calculation is as follows:

[0091] ;

[0092] in, This represents the weighted value of the normalized structural load indices: leaf root moment, tower base moment, and drive chain torque. , All represent weighting coefficients. and The weighted composite value J of the power difference index and the structural load index is solved approximately every 1–5 seconds using heuristics or quadratic programming to update the priority and reference values.

[0093] In D3: When the wind speed and direction sensors or wind measurement lidar of the unit are unavailable, the incoming flow data of the neighboring units and the wind measurement tower are acquired within the communication range of the wind farm. The data is then weighted according to the attenuation relationship of the distance to the unit and the consistency of the relative wind direction to estimate the wind speed, wind direction and turbulence of the unit. This data is used as alternative incoming flow data to calculate the spatiotemporal comprehensive characteristic index of the operating condition.

[0094] In D4: When the local unit's sensor becomes available again, a deviation compensation amount is established based on the difference between the recovered local incoming flow data and the replacement incoming flow data. The wind speed, wind direction, and the difference between the average power value and the power under the unit operating condition are re-estimated based on the deviation compensation amount to achieve smooth switching and accuracy correction.

[0095] Specifically, once the local sensor becomes available again, the system first calculates the difference between the recovered local incoming flow data and the replacement incoming flow data. Then, it uses an exponentially weighted moving average algorithm to recursively filter the difference sequence to generate a smooth and gradual deviation compensation amount. This compensation amount is applied in real time to continuously correct the currently estimated wind speed, wind direction, average power value per unit operating condition, and power difference index. This achieves a smooth transition from replacement data to local measured data in the early stage of sensor data recovery and effectively suppresses the jump in estimated values ​​at the moment of switching, ultimately achieving the purpose of accuracy correction and smooth transition of system state.

[0096] It should be noted that, specifically, the collaborative control commands are generated cyclically by the supervisory layer and then sent down to the underlying controllers for execution. The complete process is as follows:

[0097] a. Collect incoming flow, unit status, power grid dispatch and load monitoring data to determine whether there are power limiting, peak shaving or safety constraints triggered.

[0098] b. Based on the matching results of S300, a channel priority table is formed, and constraints are superimposed in the following order: grid connection specification / current limit → structural load limit → speed safety margin → pitch rate limit;

[0099] c. Reference value generation:

[0100] Within the permissible pitch range, determine the adjustment direction and magnitude according to the priority table; when there is unsteady enhancement or load increase, moderately increase the pitch opening to stabilize the angle and limit the load; when the operating conditions are stable, reduce unnecessary pitch changes.

[0101] Within the rated range, power stability is prioritized while limiting speed deviation; in low wind areas, the target speed for energy capture is maintained first; if a power limiting command is issued, the torque reference is used to converge to the target first, and then small pitch adjustments are made to suppress oscillations.

[0102] The upper limit of the speed is set based on the actuator temperature rise and the frequency of operation, and a minimum hold time is set for rapid repetitive actions; when the speed or frequency exceeds the threshold, the pitch weight is automatically reduced.

[0103] After the incoming flow and power conditions have been stable for several cycles, the strategy is slowly transitioned to one that is primarily based on speed / torque. When rapid wind direction oscillation, increased turbulence, or load approaching the limit are detected, the strategy is quickly switched to one that is primarily based on pitch.

[0104] d. If a power limiting / peak shaving command is received, the priorities will be rearranged without breaking the hard constraints to minimize the combined performance of power fluctuations and critical loads; if necessary, the target speed will be reduced and the pitch retrace speed will be limited to prevent re-oscillation.

[0105] e. When wind speed / wind direction / lidar is unavailable, use alternative data from neighboring units and wind measurement towers and weight them according to distance and wind direction consistency; after the sensors recover, use a gradual deviation compensation method to smoothly switch back to avoid sudden changes in reference values.

[0106] f. Apply limiting and ramping to the generated pitch angle, speed, and torque reference values, package them into a coordinated control command and issue it; the bottom loop executes at a fixed frequency and transmits the actual execution and deviation back to the upper layer for correction in the next cycle.

[0107] It should be noted that, based on the real-time calculated fusion weights, grid dispatch requirements and unit safety constraints, the above step S400 dynamically generates collaborative control commands, and coordinates the pitch angle, speed and torque reference values, so that the wind turbine generator can achieve the collaborative control goal of stable power output and structural load optimization under the premise of meeting grid connection specifications and load limitations.

[0108] Example 2, based on the previous example, provides an application example of a method for coordinated control of wind turbine pitch and rotational speed, to verify and illustrate the technical effects of the method.

[0109] This embodiment uses a 3MW onshore wind turbine generator as an example. The generator can be configured for direct drive or semi-direct drive, with a rated wind speed of 12m / s. The system is equipped with sensors including a nacelle anemometer and wind vane (cup type + weather vane), tower top temperature and pressure sensors, blade root and tower bottom strain gauges, a main shaft torque meter, generator current and voltage sampling units, and a pitch servo current sensor. Data acquisition is performed at a 50Hz frequency, and time synchronization is achieved through a precise time protocol or a network time protocol. Finally, data frames with a unified time base are sent to the monitoring and control layer at a 1-second interval.

[0110] In this embodiment, a sliding window length of 60 seconds is used, with a 30-second overlap, meaning the control command update cycle is 30 seconds. The raw data undergoes preprocessing before feature extraction: outliers are removed using the median ± median absolute deviation (MAD) method; linear interpolation is performed on data segments with consecutive missing values ​​not exceeding 3 seconds; data exceeding this range are marked as invalid. To eliminate the influence of atmospheric condition variations, air density is calculated using measured temperature and pressure, and the standard reference power curve is scaled and corrected to establish a power benchmark adapted to the current environment.

[0111] Furthermore, the basic loss power model is estimated by a quadratic polynomial regression model characterized by converter load rate, pitch duty cycle and nacelle temperature. The model is trained based on 7-day sliding historical data and ridge regression regularization, and is retrained every 6 hours. At least two months of data are required for the initialization phase.

[0112] Furthermore, to evaluate power output characteristics, the system performs two-dimensional segmented statistics on the average power per unit operating condition, with a segment width of 1 m / s wind speed and 10° wind direction. At least 20 samples must be accumulated in each segment before the operating condition power difference index δop can be calculated. This index is defined as the standard deviation of the average power per unit operating condition in that segment. Before calculation, extreme values ​​must be removed again using the median ± 3MAD method.

[0113] Furthermore, the spatiotemporal characteristics of the operating conditions are determined by the window duration ΔT, the cumulative change in wind direction Δφ, the average wind speed vavg, the average absolute yaw mismatch angle γ, and the average turbulence intensity T. Iavg It consists of five parameters. Each parameter is normalized to its historical 5%–95th percentile range, and then linearly weighted and summed using a weighting coefficient α = [0.25, 0.20, 0.20, 0.20, 0.15] to obtain the spatiotemporal comprehensive characteristic index of the operating condition. The collaborative control fusion weight W is obtained by using the normalized prior channel weights (…). =0.6, =0.4), δop and F windAn activation function is input, with the adjustment coefficient β taking values ​​of [1.2, 0.8, 1.6, 1.0]. To suppress frequent weight fluctuations, a hysteresis interval of [0.45, 0.55] is set. The final channel weight allocation is: Pitch channel w'θ = W· Speed / torque channel w'ω=(1-W)· .

[0114] Control commands are updated every 1 to 2 seconds by the supervisory layer. The pitch angle reference increment and generator torque reference increment are calculated using a limited linear or piecewise linear mapping function based on information such as active power setpoint deviation, yaw mismatch angle, turbulence intensity, and speed safety margin, and then multiplied by the corresponding channel weights and gain coefficients. The bottom-level pitch, torque, and speed control loops operate at a frequency of 100Hz and employ saturation limits and anti-integral saturation measures. The system follows strict constraint priorities: grid connection specifications and current limits are highest, followed by structural loads and speed, and finally pitch rate limits. Typical operating constraints include: speed range 0.7–1.2 pu, absolute pitch rate not exceeding 8° / s, and generator torque not exceeding 1.05 pu.

[0115] Furthermore, under conditions of rapid wind direction changes (Δφ>30° / 5min) and enhanced turbulence, the system increases the priority of the pitch channel by increasing the fusion weight W (approaching 1) to quickly stabilize the blade tip angle of attack and suppress load fluctuations. Under stable downwind conditions, W decreases to 0.3–0.4, and the system relies more on the speed / torque channel for fine-tuning, which helps reduce generation noise and pitch system wear. The hysteresis gating mechanism effectively avoids frequent switching of weights near critical values, while the power difference index δop provides stable performance evaluation feedback after sufficient sample size.

[0116] As the above analysis shows, this invention, by constructing a collaborative control mechanism that integrates spatiotemporal operating condition characteristics and power differences, achieves dynamic weight allocation and priority adjustment of the pitch and speed loops, effectively reducing power fluctuations and structural loads under complex inflow conditions, and minimizing frequent operation and wear of the pitch actuator. This invention establishes a basic loss power model with converter load rate, pitch duty cycle, and ambient temperature as independent variables, decomposing the measured active power into net output power, significantly reducing power caliber deviations caused by different operating conditions / machine types / temperatures, and providing a consistent benchmark for subsequent statistics and optimization. Furthermore, when sensors fail, the method of this invention can intelligently fuse data from neighboring units and meteorological towers and perform deviation compensation, significantly enhancing the adaptability and robustness of the unit under abnormal operating conditions, while better meeting the coordinated control requirements under grid peak shaving and power limiting constraints.

[0117] Example 3: This example provides a system for coordinated control of wind turbine pitch and rotational speed, including:

[0118] The sensing module is used to acquire the operating status data and incoming flow environment data of the wind turbine generator set;

[0119] The operating condition characteristic calculation module is used to extract the spatiotemporal operating condition characteristics of the wind turbine generator set within the target time window based on the operating status data and the incoming flow environment data, and obtain the spatiotemporal comprehensive characteristic index of the operating condition.

[0120] The power characteristic analysis module is used to calculate the average power value per unit operating condition corresponding to the target time window based on operating status data, incoming flow environment data and reference power curves, form a set of operating condition-level power characteristics, and obtain the operating condition power difference index based on the dispersion of the data in the set.

[0121] The weight fusion module is used to perform feature matching on the spatiotemporal comprehensive characteristic index of operating conditions and the power difference index of operating conditions. Based on the preset basic weights of the pitch channel and the basic weights of the speed or torque channel, it calculates the collaborative control fusion weights.

[0122] The control command generation and execution module is used to generate collaborative control commands based on the collaborative control fusion weights, grid dispatch information and unit control constraints. The wind turbine generator sets execute the collaborative control commands to complete the collaborative control of pitch and speed.

[0123] It should be noted that the technical solution of the wind power pitch and speed coordinated control system is based on the same concept as the technical solution of the wind power pitch and speed coordinated control method described above. For details not described in detail in the technical solution of the wind power pitch and speed coordinated control system in this embodiment, please refer to the description of the technical solution of the wind power pitch and speed coordinated control method described above.

[0124] The above-mentioned unit modules can be embedded in the processor of the electronic device in hardware form or independent of it, or they can be stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0125] This embodiment also provides an electronic device, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for coordinated control of wind turbine pitch and speed. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0126] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method proposed in the above embodiments.

[0127] The storage medium proposed in this embodiment belongs to the same inventive concept as the method proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0128] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory, random access memory, flash memory, hard disk, or optical disk, and includes several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute the method of the embodiments of the present invention.

[0129] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for coordinated control of wind turbine pitch and rotational speed, characterized in that, include: Based on the acquired operating status data and incoming flow environment data of the wind turbine generator set, the spatiotemporal operating condition characteristics of the wind turbine generator set within the target time window are extracted to obtain the spatiotemporal comprehensive characteristic index of the operating condition. Based on the operating status data, incoming flow environment data, and reference power curve, the average power value per unit operating condition corresponding to the target time window is calculated to form an operating condition-level power feature set, and the operating condition power difference index is obtained based on the dispersion of the data in the set. Feature matching is performed on the spatiotemporal comprehensive characteristic index of the operating condition and the power difference index of the operating condition. Based on the basic weight of the pitch channel and the basic weight of the speed or torque channel, the collaborative control fusion weight is calculated. Based on the collaborative control fusion weights, grid dispatch information, and unit control constraints, a collaborative control command is generated, and the wind turbine generator executes the collaborative control command to complete the collaborative control of pitch and speed. The operating status data of the wind turbine generator set includes rotor speed, pitch angle, generator active power, electromagnetic torque, nacelle yaw angle, converter load rate, and pitch system duty cycle; the incoming flow environment data includes wind speed, wind direction, and turbulence intensity. Different converter load rates and pitch system duty cycles correspond to different base loss power models. The base loss power model is used to separate base losses from measured electrical power to obtain net output power. The calculation of the spatiotemporal comprehensive characteristic index of the working condition includes: in, Indicates the spatiotemporal comprehensive characteristics of the operating condition. Indicates the end time of the target time window. Indicates the start time of the target time window. This represents the cumulative change in wind direction within the target time window. This represents the average wind speed within the target time window. This represents the average absolute yaw mismatch angle within the target time window. This represents the average turbulence intensity within the target time window. All represent spatiotemporal feature weight coefficients; The obtained operating condition power difference index includes: The current base loss power is determined based on the converter load rate and the pitch system duty cycle, and the current base loss power is subtracted from the measured power to obtain the net output power. The net output power is integrated within the target time window and divided by the window duration to obtain the average power value under unit operating conditions. A power feature set is constructed by dividing the data into two-dimensional boxes based on wind speed and wind direction. The average power value of the unit operating condition within the same box is aggregated based on the reference power curve after density correction. The number of samples of the average power value of the unit operating condition within each box is counted. If the number of samples exceeds a preset threshold, the standard deviation of the average power value of all units under operating conditions within the sub-bin is calculated, and the standard deviation is used as the power difference index of the current sub-bin to characterize the dispersion of power output under operating conditions; if the number of samples is less than the preset threshold, the power difference index is not calculated for the time being.

2. The method for coordinated control of wind turbine pitch and rotational speed as described in claim 1, characterized in that, The unit control constraints include upper and lower speed limits, upper limits for pitch angle and pitch rate, generator torque / current limits, nacelle / blade root / tower base load limits, and grid connection specifications; the cooperative control commands include a cooperative control priority table, pitch angle reference, speed reference, generator torque reference, pitch rate threshold, and control entry and exit timing. The fusion weights are used to update the priority or control gain of the pitch and speed or torque channels in real time.

3. The method for coordinated control of wind turbine pitch and rotational speed as described in claim 1, characterized in that, Also includes: When the wind speed and direction sensors or wind-measuring lidar of the unit are unavailable, the incoming flow data of the neighboring units and the wind measurement tower are obtained within the communication range of the wind farm. The data are then weighted according to the attenuation relationship of the distance to the unit and the consistency of the relative wind direction to estimate the wind speed, wind direction and turbulence of the unit. This data is used as substitute incoming flow data to calculate the spatiotemporal comprehensive characteristic index of the operating condition. Once the sensors in this unit are available again, a deviation compensation amount is established based on the difference between the restored incoming flow data and the alternative incoming flow data. The wind speed, wind direction, and the difference between the average power value and the power under the unit operating condition are then re-estimated based on the deviation compensation amount to achieve smooth switching and accuracy correction.

4. A system for coordinated control of wind turbine pitch and rotational speed, employing the method for coordinated control of wind turbine pitch and rotational speed as described in any one of claims 1 to 3, characterized in that, include: The sensing module is used to acquire the operating status data and incoming flow environment data of the wind turbine generator set; The operating condition characteristic calculation module is used to extract the spatiotemporal operating condition characteristics of the wind turbine generator set within the target time window based on the operating status data and the incoming flow environment data, and obtain the spatiotemporal comprehensive characteristic index of the operating condition. The power characteristic analysis module is used to calculate the average power value per unit operating condition corresponding to the target time window based on the operating status data, incoming flow environment data and reference power curve, form an operating condition-level power characteristic set, and obtain the operating condition power difference index based on the dispersion of the data in the set. The weight fusion module is used to perform feature matching on the spatiotemporal comprehensive characteristic index of the operating condition and the power difference index of the operating condition, and to calculate the collaborative control fusion weight based on the preset basic weight of the pitch channel and the basic weight of the speed or torque channel. The control command generation and execution module is used to generate collaborative control commands based on the collaborative control fusion weights, grid dispatch information and unit control constraints. The wind turbine generator sets execute the collaborative control commands to complete the collaborative control of pitch and speed.

5. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store computer-executable instructions, and when the processor executes the computer-executable instructions, it implements the steps of the method for coordinated control of wind power pitch and rotation speed as described in any one of claims 1 to 3.

6. A computer-readable storage medium storing computer-executable instructions thereon, characterized in that: When the computer-executable instructions are executed by the processor, they implement the steps of the method for coordinated control of wind turbine pitch and rotation speed as described in any one of claims 1 to 3.

Images