Variable pitch system control method and device, variable pitch system, medium and program product

Abstract

The invention relates to a variable pitch system control method and device, a variable pitch system, a medium and a program product, relates to the technical field of wind power generation control, and can avoid energy efficiency and power stability imbalance of the variable pitch system. The method comprises the following steps: acquiring operation parameters of the generator set; according to the operation parameters, a first pitch angle adjusting instruction is generated, a variable pitch mechanism is controlled to operate according to the first pitch angle adjusting instruction, and the pitch angle of the variable pitch mechanism is obtained; the deviation between the pitch angle and a preset target pitch angle is determined, and a membership correction coefficient is generated according to the operation parameters; under the condition that the deviation is larger than a deviation threshold value, a second pitch angle adjusting instruction is generated according to the membership correction coefficient and the operation parameters; and according to the second pitch angle adjusting instruction, the variable pitch mechanism is controlled to operate until the output power of the generator set is larger than or equal to the power required by the power grid and the wind energy capture efficiency corresponding to the output power is optimal.

Description

Technical Field

[0001] This disclosure relates to the field of wind power generation control technology, specifically to a pitch system control method, device, pitch system, medium, and program product. Background Technology

[0002] In related technologies, pitch control often adopts fixed-threshold PID (Proportional-Integral-Derivative) or wind speed-power lookup table method, relying on preset rules to adjust the pitch angle. However, with the expansion of wind farm scale and the diversification of grid dispatch requirements, the pitch system needs to take into account both wind energy capture efficiency and grid power dynamic response under complex operating conditions. This results in the pitch system being unable to adjust the decoupling strategy according to real-time operating conditions, and there is a conflict between wind energy capture and grid power tracking. Summary of the Invention

[0003] To address the shortcomings of related technologies, this disclosure provides a pitch system control method, device, pitch system, medium, and program product.

[0004] To achieve the above objectives, in a first aspect, this disclosure provides a pitch system control method, the method comprising: Obtain the operating parameters of the generator set; Based on the operating parameters, a first pitch angle adjustment command is generated, the pitch mechanism is controlled to operate according to the first pitch angle adjustment command, and the pitch angle of the pitch mechanism is obtained. Determine the deviation between the pitch angle and the preset target pitch angle, and generate a membership correction coefficient based on the operating parameters; If the deviation is greater than the deviation threshold, a second pitch angle adjustment command is generated based on the membership correction coefficient and the operating parameters. The pitch mechanism is controlled to operate according to the second pitch angle adjustment command until the output power of the generator set is greater than or equal to the power demand of the power grid and the wind energy capture efficiency corresponding to the output power is optimal.

[0005] Optionally, the operating parameters include wind speed, generator speed, and grid demand power. Generating a first pitch angle adjustment command based on the operating parameters includes: Based on adaptive fuzzy rules, the wind speed, the generator speed, and the power demand of the power grid are dynamically decoupled to generate a first pitch angle adjustment command.

[0006] Optionally, the step of dynamically decoupling the wind speed, the generator speed, and the power grid demand based on adaptive fuzzy rules to generate a first pitch angle adjustment command includes: The wind speed, generator speed, and power grid demand are normalized and noise filtered to obtain the processed wind speed, generator speed, and power grid demand. Based on a preset rule base for the association between wind speed and pitch angle, a weighted average calculation is performed on the processed wind speed, generator speed, and power demand of the power grid to generate a first pitch angle adjustment command.

[0007] Optionally, generating the second pitch angle adjustment command based on the membership correction coefficient and the operating parameters includes: Based on the membership correction coefficient, the activation threshold and output gain of the association rule base of the preset wind speed and pitch angle are dynamically adjusted to obtain the intermediate association rule base, and the rule weight of the intermediate association rule base is updated to obtain the target association rule base. The association rule base of the preset wind speed and pitch angle adopts a hierarchical storage structure. Based on the sliding window, wavelet transform noise reduction processing is performed on the operating parameters to obtain an effective signal; Based on the target association rule base, the effective signals are weighted and averaged to generate a second pitch angle adjustment command.

[0008] Optionally, the operating parameters include grid demand power, and the step of updating the rule weights of the intermediate association rule base to obtain the target association rule base includes: Determine the power deviation between the power demand of the power grid and the theoretical maximum power of the power grid, and use the square of the power deviation as the determination of the update gradient; Based on the update gradient and the preset learning rate, the original rule weights of the intermediate association rule base are updated to obtain the target rule weights. Based on the target rule weights, the target association rule base is obtained.

[0009] Optionally, updating the original rule weights of the intermediate association rule base according to the update gradient and the preset learning rate to obtain the target rule weights includes: Substituting the update gradient, the original rule weights of the intermediate association rule base, and the preset learning rate into the following calculation formula, the target rule weights are obtained; , Among them, W i new The weights of the target rules, W i η represents the original rule weights, η represents the preset learning rate, and J represents the updated gradient.

[0010] Optionally, obtaining the pitch angle of the pitch mechanism includes: Obtain the original signal of the blade angle in the pitch mechanism; The original signal is subjected to anti-interference filtering to obtain the filtered original signal, and the filtered original signal is subjected to analog-to-digital conversion to obtain the digital angle value. The environmental parameters of the pitch mechanism are obtained, including temperature and humidity. Based on the environmental parameters, the digital angle value is subjected to linear error compensation processing to obtain the propeller pitch angle.

[0011] Secondly, this disclosure provides a pitch system control device, the device comprising: The signal acquisition module is configured to acquire the operating parameters of the generator set; The first control module is configured to generate a first pitch angle adjustment command based on the operating parameters, control the pitch mechanism to operate based on the first pitch angle adjustment command, and obtain the pitch angle of the pitch mechanism. The first execution module is configured to determine the deviation between the pitch angle and the preset target pitch angle, and generate a membership correction coefficient based on the operating parameters. The second control module is configured to generate a second pitch angle adjustment command based on the membership correction coefficient and the operating parameters when the deviation is greater than the deviation threshold. The second execution module is configured to control the operation of the pitch mechanism according to the second pitch angle adjustment command until the output power of the generator set is greater than or equal to the power demand of the power grid and the wind energy capture efficiency corresponding to the output power is optimal.

[0012] Thirdly, this disclosure provides a pitch system including the pitch system control device described in the second aspect.

[0013] Fourthly, this disclosure provides a non-transitory computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method described in the first aspect.

[0014] Fifthly, this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] The above technical solution generates a first pitch angle adjustment command based on the generator set's operating parameters, controls the pitch mechanism to operate according to the first pitch angle adjustment command, and obtains the pitch angle of the pitch mechanism. Based on the deviation between the pitch angle of the pitch mechanism and the preset target pitch angle, a second pitch angle control command and torque compensation signal are generated based on the correction coefficient generated from the operating parameters and the operating parameters. The pitch mechanism is then controlled again based on the second pitch angle control command and torque compensation signal until the generator set's output power is greater than or equal to the grid's demand power, and the wind energy capture efficiency corresponding to the output power is optimal. This achieves coordinated optimization of wind energy capture efficiency and grid power demand, enabling the pitch system to adaptively adjust the blade angle when wind speed fluctuates or grid dispatch demand changes, thereby improving wind energy utilization efficiency and reducing power output deviation.

[0016] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0017] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings: Figure 1 This is a flowchart illustrating a pitch system control method according to an exemplary embodiment of the present disclosure.

[0018] Figure 2 This is another flowchart illustrating a pitch system control method according to an exemplary embodiment of the present disclosure.

[0019] Figure 3 This is a block diagram of a pitch system control device according to an exemplary embodiment of the present disclosure.

[0020] Figure 4 This is a system architecture diagram of a pitch system according to an exemplary embodiment of the present disclosure. Detailed Implementation

[0021] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0022] It should be noted that all actions involving the acquisition of signals, information, or data in this disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with authorization from the owner of the relevant device.

[0023] It is worth noting that the pitch control system is the core control device of the wind turbine generator. By adjusting the blade angle, it controls the wind energy captured by the wind turbine, which directly affects the power generation efficiency and grid connection stability. Traditional pitch control often uses fixed threshold PID or a lookup table method based on wind speed and power, relying on preset rules to adjust the pitch angle. However, with the expansion of wind farm scale and the diversification of grid dispatching needs, the pitch control system needs to balance wind energy capture efficiency and grid power dynamic response under complex operating conditions. This places higher demands on the adaptability and multi-objective coordination capabilities of the control strategy.

[0024] In related technologies, pitch control systems process data in isolation when dealing with sudden changes in wind speed, turbulence interference, and fluctuations in grid demand. Multi-source parameters such as wind speed, speed, and power demand lack dynamic coupling analysis, making it difficult to generate globally optimal pitch angle commands. Wind energy utilization efficiency is low, and decoupling strategies cannot be adjusted according to real-time operating conditions. There is a conflict between wind energy capture and grid power tracking.

[0025] In view of this, the present disclosure provides a pitch system control method, device, pitch system, medium and program product, which can improve the problem of difficulty in achieving dynamic coordinated optimization of wind energy capture and grid demand, resulting in an imbalance between energy efficiency and power stability.

[0026] Figure 1 This is a flowchart illustrating a pitch system control method according to an exemplary embodiment of this disclosure, such as... Figure 1 As shown, the pitch system control method may include the following steps: In step S11, the operating parameters of the generator set are obtained.

[0027] In step S12, a first pitch angle adjustment command is generated according to the operating parameters, the pitch mechanism is controlled to operate according to the first pitch angle adjustment command, and the pitch angle of the pitch mechanism is obtained.

[0028] The pitch mechanism converts motor torque into linear pitch displacement via a planetary gearbox and a ball screw.

[0029] In step S13, the deviation between the pitch angle and the preset target pitch angle is determined, and a membership correction coefficient is generated based on the operating parameters.

[0030] In step S14, if the deviation is greater than the deviation threshold, a second pitch angle adjustment command is generated based on the membership correction coefficient and the operating parameters.

[0031] In step S15, the pitch mechanism is controlled to operate according to the second pitch angle adjustment command until the output power of the generator set is greater than or equal to the power demand of the power grid and the wind energy capture efficiency corresponding to the output power is optimal.

[0032] In this embodiment, operating parameters include, but are not limited to, wind speed, generator speed, and grid power demand. Specifically, real-time wind speed and turbulence intensity can be measured by fusing lidar and ultrasonic sensors, increasing multi-sensor data fusion, and eliminating measurement bias of a single sensor in complex wind fields based on the Kalman filter algorithm, thus optimizing the frequency domain analysis capability of turbulence intensity; generator rotor speed can be synchronously acquired by Hall sensors and encoders, with redundant acquisition circuits designed to support real-time cross-verification of Hall sensor and encoder signals, and the time alignment accuracy of speed data can be improved through a timestamp synchronization mechanism; grid dispatch instructions can be received and target power thresholds can be parsed, and a grid dispatch protocol parser is integrated to support multi-protocol compatibility, achieving low-latency parsing and priority scheduling of target power thresholds.

[0033] It is worth noting that the state prediction formula of the Kalman filter algorithm is: , in, Let A represent the k-th fused wind speed estimate, and let A represent the state transition of the fused wind speed estimate. B represents the (k-1)th fused wind speed estimate, and B represents the control input matrix. Characterizes the control input at time k-1.

[0034] Measurement update formula: , , in, Characterizing the Kalman gain at time k, The covariance of the prediction error representing the fused wind speed estimate at time k is R is the transpose of the observation matrix H, where H represents the observation matrix and R represents the measurement noise covariance. The weight of z represents the influence of the control input on the fused wind speed estimate at time k. k Characterizes the raw data from the sensor; The formula for frequency domain analysis of turbulence intensity is: , , Where, σ v The standard deviation of wind speed is represented by N, which represents the total number of wind speed samples. i The instantaneous wind speed value representing the i-th wind speed sample. The arithmetic mean of the Nth wind speed sample value.

[0035] Redundancy check formula in speed sensing unit: The Hall sensor outputs ω H The encoder outputs ω E In this case, the rotational speed is: , Where ω represents the rotational speed, α represents the weighting coefficient of the Hall sensor and encoder output rotational speed, and σ H Characterizing the error of the sensor, σ E This represents the standard deviation of the sensor.

[0036] In the above technical solution, a first pitch angle adjustment command is generated based on the operating parameters of the generator set. The pitch mechanism is controlled to operate according to the first pitch angle adjustment command, and the pitch angle of the pitch mechanism is obtained. Based on the deviation between the pitch angle of the pitch mechanism and the preset target pitch angle, a second pitch angle control command and a torque compensation signal are generated based on the correction coefficient generated from the operating parameters and the operating parameters. The pitch mechanism is controlled to operate again based on the second pitch angle control command and the torque compensation signal until the output power of the generator set is greater than or equal to the power demand of the grid, and the wind energy capture efficiency corresponding to the output power is optimal. This achieves the coordinated optimization of wind energy capture efficiency and grid power demand, enabling the pitch system to adaptively adjust the blade angle when wind speed fluctuates or grid dispatch demand changes, thereby improving wind energy utilization efficiency and reducing power output deviation.

[0037] To facilitate a better understanding of the pitch system control method provided in this disclosure by those skilled in the art, the pitch system control method will be described in detail below.

[0038] In a feasible embodiment, when the operating parameters include wind speed, generator speed, and grid power demand, step S11, which involves generating a first pitch angle adjustment command based on the operating parameters, may include: Based on adaptive fuzzy rules, the wind speed, the generator speed, and the power demand of the power grid are dynamically decoupled to generate a first pitch angle adjustment command.

[0039] In this embodiment, fuzzy logic control is a control method based on fuzzy sets and fuzzy inference, capable of handling uncertainties and fuzziness in the system. Fuzzy logic control describes the input-output relationship of the system through fuzzy rules. Adaptive fuzzy rules refer to fuzzy rules that can automatically adjust according to the system's operating state and performance indicators. This adaptability can be achieved through learning algorithms, such as fuzzy clustering, genetic algorithms, and neural networks. Dynamic decoupling is a control strategy used to eliminate variable coupling in multivariable systems, allowing each variable to be controlled independently. In wind power generation systems, there is a coupling relationship between wind speed, generator speed, and grid power demand. Dynamic decoupling can decouple these variables, enabling independent control.

[0040] In the above technical solution, the coupling problem between wind speed, generator speed and grid power demand can be effectively handled by adaptive fuzzy rules and dynamic decoupling matrix, generating accurate pitch angle adjustment commands and significantly improving the performance and stability of wind power generation system.

[0041] In one feasible embodiment, the step of dynamically decoupling the wind speed, the generator speed, and the power grid demand based on adaptive fuzzy rules to generate a first pitch angle adjustment command may include: The wind speed, generator speed, and power grid demand are normalized and noise filtered to obtain the processed wind speed, generator speed, and power grid demand. Based on a preset rule base for the association between wind speed and pitch angle, a weighted average calculation is performed on the processed wind speed, generator speed, and power demand of the power grid to generate a first pitch angle adjustment command.

[0042] In this embodiment, a vector control strategy can be used to dynamically decouple wind speed, generator speed, and the power demand of the power grid, achieving high-precision torque and speed decoupling. Furthermore, during the dynamic decoupling process, a dead-zone compensation algorithm is used to correct the output voltage of the inverter in the generator. The correction principle can be found in the following calculation formula: , Among them, V comp V represents the corrected inverter output voltage. ref The inverter output voltage before correction is represented by I, the motor phase current is represented by I, and the dead zone voltage loss is represented by ΔV.

[0043] The weighted average can be calculated using the following formula: , Among them, W i The weights representing the i-th signal. Representation of association rule base, x j Characterizes the operating parameters of the generator set.

[0044] The first pitch angle adjustment command is shown below: , in, The first pitch angle adjustment command is represented by m, which represents the number of association rules, and θ is the value of the first pitch angle adjustment command. i Represents the i-th data item involved in the calculation.

[0045] It is worth noting that the association rule base adopts a hierarchical storage structure, supports online rule addition and deletion, and can optimize the initial rule weight allocation through historical experience.

[0046] In a feasible embodiment, after determining the deviation between the pitch angle and the preset target pitch angle in step S13, the method may further include: PID correction is performed based on the deviation between the pitch angle and the preset target pitch angle to obtain the correction amount.

[0047] It is worth noting that during PID correction, the critical gain K can be calculated using the following formula. u and oscillation period T u Determine the parameters: , , , Among them, K p K represents the proportional coefficient of a PID controller. i K represents the integral coefficient of the PID controller. d The derivative coefficients characterize the PID controller.

[0048] In this embodiment, the correction amount can be used to generate a second pitch angle adjustment command, thereby dynamically matching pitch speed and load changes and reducing tracking error.

[0049] In a feasible embodiment, step S13, generating the second pitch angle adjustment command based on the membership correction coefficient and the operating parameters, may include: Based on the membership correction coefficient, the activation threshold and output gain of the association rule base of the preset wind speed and pitch angle are dynamically adjusted to obtain the intermediate association rule base, and the rule weight of the intermediate association rule base is updated to obtain the target association rule base. The association rule base of the preset wind speed and pitch angle adopts a hierarchical storage structure. Based on the sliding window, wavelet transform noise reduction processing is performed on the operating parameters to obtain an effective signal; Based on the target association rule base, the effective signals are weighted and averaged to generate a second pitch angle adjustment command.

[0050] In this embodiment, reinforcement learning Q-learning can be used to dynamically adjust the activation threshold and output gain of the association rule base based on historical control effects and membership correction coefficients. The specific update principle is shown in the following calculation formula: , Where s represents the system state, a represents the rule adjustment action, η represents the learning rate, r represents the immediate reward signal, and γ represents the discount factor.

[0051] Wavelet transform noise reduction can be achieved using the following formula: , Where x(t) represents the effective signal at time t, ψ represents the mother wavelet function, a represents the scaling factor, and b represents the translation factor.

[0052] In one feasible embodiment, when the operating parameters include grid demand power, updating the rule weights of the intermediate association rule base to obtain the target association rule base includes: Determine the power deviation between the power demand of the power grid and the theoretical maximum power of the power grid, and use the square of the power deviation as the determination of the update gradient; Based on the update gradient and the preset learning rate, the original rule weights of the intermediate association rule base are updated to obtain the target rule weights. Based on the target rule weights, the target association rule base is obtained.

[0053] In this embodiment, the theoretical maximum power of the power grid can be determined by the following formula: , Among them, P max ρ represents the theoretical maximum power of the power grid, A represents the swept area of ​​the wind turbine rotor, and v represents the real-time wind speed. The maximum value of the wind energy capture coefficient.

[0054] Therefore, the power deviation ΔP = |P real -P max | where ΔP characterizes the power deviation, P real Characterizes the power demand of the power grid.

[0055] Accordingly, update the gradient J=(P) real -P max ) 2 .

[0056] In a feasible embodiment, updating the original rule weights of the intermediate association rule base according to the update gradient and the preset learning rate to obtain the target rule weights may include: Substituting the update gradient, the original rule weights of the intermediate association rule base, and the preset learning rate into the following calculation formula, the target rule weights are obtained; , Among them, W i new The weights of the target rules, W i η represents the original rule weights, η represents the preset learning rate, and J represents the updated gradient.

[0057] It is worth noting that by iteratively adjusting the rule weights of the fuzzy control module using the gradient descent algorithm, the actual power output can be made to approximate the theoretical value. This can also be achieved based on the momentum term v. t To optimize rule weights, please refer to the following calculation formula for the specific principles: , , Among them, v t The momentum term at time t is represented by β, and the momentum factor by v. t-1 The momentum term at time t-1 is represented. J represents the gradient, W t The rule weights at time t are used to represent the rules.

[0058] Furthermore, it can monitor changes in grid demand based on historical optimization rule weights and quickly invoke preset control strategies when grid demand changes abruptly. Specifically, a time-series database can be used to store historical optimization rule weights, and similar operating conditions can be matched and strategies recommended based on time series data. The principle of similar operating conditions matching based on time series data can be found in the following calculation formula: , Where d represents the similarity distance, x k Characterizing time series data of operating conditions, y k It represents the time series data of operating conditions stored in the historical operating condition database.

[0059] This allows us to determine the preset parameter that minimizes d in the historical strategy.

[0060] In one feasible embodiment, obtaining the pitch angle of the pitch mechanism may include: Obtain the original signal of the blade angle in the pitch mechanism; The original signal is subjected to anti-interference filtering to obtain the filtered original signal, and the filtered original signal is subjected to analog-to-digital conversion to obtain the digital angle value. The environmental parameters of the pitch mechanism are obtained, including temperature and humidity. Based on the environmental parameters, the digital angle value is subjected to linear error compensation processing to obtain the propeller pitch angle.

[0061] In this embodiment, the original signal of the blade angle can be captured in real time by a high-resolution photoelectric encoder.

[0062] like Figure 2 As shown below, a complete example is used to illustrate the pitch system control method provided in this disclosure: I. Real-time wind speed is acquired through the wind speed detection unit in the signal acquisition module. and turbulence intensity The speed sensor unit collects the generator speed. The power grid interface unit analyzes the target power. Collect demand data and generate standardized sensor datasets; Standardized dataset Dnorm={v norm ω norm P norm The following formula is used to calculate: ; Among them, v rated ω rated P rated For the system's rated parameters, when D norm If the validity check passes, the data is complete, and the delay is <10ms, trigger II. II. Input the standardized sensor dataset into the fuzzy control module. The data preprocessing unit of the fuzzy control module processes the data... norm Filtering is performed to extract feature vectors and generate initial pitch angle adjustment commands; The fuzzy inference unit calculates θ based on the initial rule base. target Initial pitch angle adjustment command θ target Through the calculation formula: ; in, For membership function, If |θ target -θ real If the temperature exceeds 0.5°, trigger III; otherwise, enter static power fine-tuning mode. III. The permanent magnet synchronous pitch drive module receives and executes the initial pitch angle adjustment command θ. target It drives the pitch actuator and obtains the actual pitch angle θ through the photoelectric encoder via the pitch angle feedback module. real θ is generated after error compensation. feedback The feedback is then sent to the fuzzy control module; Compensated feedback signal θ feedback Through the calculation formula: ; Where ΔT and ΔH are the temperature and humidity drift, and k1 and k2 are the compensation gains. If the tracking error e θ =|θ feedback -θ target |>0.3°, triggering IV; IV. When the deviation between the actual pitch angle fed back by III and the target value exceeds the threshold, the state monitoring module triggers the fault tolerance strategy, and at the same time the power optimization module calculates the wind energy capture efficiency deviation and generates the membership correction coefficient. Condition monitoring and power optimization triggering, condition monitoring module analysis e θ Vibration data V rms Temperature rise ΔT, power optimization module calculates theoretical maximum power The power deviation is ΔP = |P real -P max |; V. The fuzzy control module updates the rule base according to the correction coefficient of IV, generates secondary adjustment instructions, and drives the pitching action. The real-time optimization unit updates the rule weights using gradient descent, according to the following formula: ; The fuzzy control module loads the new rule base R. new Generate secondary instruction θ optimized Updated control commands Membership function parameter μ new If ΔP decreases by less than 1% / second, convergence is determined, and VI is triggered; otherwise, proceed to II. VI. Repeat steps II to V until the actual power output meets the grid demand and the wind energy capture efficiency reaches its optimal level, and output a stable pitch angle and optimization report; Repeat II-V until the termination condition is met. The position closed-loop unit ensures tracking accuracy through PID correction, using the following formula: ; Stabilized pitch angle θ final Power optimization report; The termination condition is |P real -P grid |≤2%P grid And lasting for 10 seconds, e θ ≤0.1°, no system alarm.

[0063] Based on the same inventive concept, this disclosure also provides a pitch system control device, such as... Figure 3 As shown, the pitch system control device 300 includes: Signal acquisition module 301 is configured to acquire the operating parameters of the generator set; The first control module 302 is configured to generate a first pitch angle adjustment command based on the operating parameters, control the operation of the pitch mechanism based on the first pitch angle adjustment command, and obtain the pitch angle of the pitch mechanism. The first execution module 303 is configured to determine the deviation between the pitch angle and the preset target pitch angle, and generate a membership correction coefficient based on the operating parameters. The second control module 304 is configured to generate a second pitch angle adjustment command based on the membership correction coefficient and the operating parameters when the deviation is greater than the deviation threshold. The second execution module 305 is configured to control the operation of the pitch mechanism according to the second pitch angle adjustment command until the output power of the generator set is greater than or equal to the power demand of the power grid and the wind energy capture efficiency corresponding to the output power is optimal.

[0064] In the above technical solution, a first pitch angle adjustment command is generated based on the operating parameters of the generator set. The pitch mechanism is controlled to operate according to the first pitch angle adjustment command, and the pitch angle of the pitch mechanism is obtained. Based on the deviation between the pitch angle of the pitch mechanism and the preset target pitch angle, a second pitch angle control command and a torque compensation signal are generated based on the correction coefficient generated from the operating parameters and the operating parameters. The pitch mechanism is controlled to operate again based on the second pitch angle control command and the torque compensation signal until the output power of the generator set is greater than or equal to the power demand of the grid, and the wind energy capture efficiency corresponding to the output power is optimal. This achieves the coordinated optimization of wind energy capture efficiency and grid power demand, enabling the pitch system to adaptively adjust the blade angle when wind speed fluctuates or grid dispatch demand changes, thereby improving wind energy utilization efficiency and reducing power output deviation.

[0065] Furthermore, given that the operating parameters include wind speed, generator speed, and grid power demand, the first control module 302 is configured to dynamically decouple the wind speed, generator speed, and grid power demand based on adaptive fuzzy rules to generate a first pitch angle adjustment command.

[0066] Furthermore, the first control module 302 is configured to normalize and filter the wind speed, the generator speed, and the power grid demand to obtain the processed wind speed, generator speed, and power grid demand. Based on a preset rule base for the association between wind speed and pitch angle, a weighted average calculation is performed on the processed wind speed, generator speed, and power demand of the power grid to generate a first pitch angle adjustment command.

[0067] Furthermore, the second control module 304 is configured to dynamically adjust the activation threshold and output gain of the preset wind speed and pitch angle association rule library according to the membership correction coefficient, to obtain an intermediate association rule library, and update the rule weight of the intermediate association rule library to obtain a target association rule library. The preset wind speed and pitch angle association rule library adopts a hierarchical storage structure. Based on the sliding window, wavelet transform noise reduction processing is performed on the operating parameters to obtain an effective signal; Based on the target association rule base, the effective signals are weighted and averaged to generate a second pitch angle adjustment command.

[0068] Furthermore, when the operating parameters include the grid demand power, the second control module 304 is configured to determine the power deviation between the grid demand power and the grid theoretical maximum power, and use the square of the power deviation as the determination of the update gradient. Based on the update gradient and the preset learning rate, the original rule weights of the intermediate association rule base are updated to obtain the target rule weights. Based on the target rule weights, the target association rule base is obtained.

[0069] Furthermore, the second control module 304 is configured to update the original rule weights of the intermediate association rule base according to the update gradient and the preset learning rate to obtain the target rule weights, including: Substituting the update gradient, the original rule weights of the intermediate association rule base, and the preset learning rate into the following calculation formula, the target rule weights are obtained; , Among them, W i new The weights of the target rules, W i η represents the original rule weights, η represents the preset learning rate, and J represents the updated gradient.

[0070] Furthermore, the first control module 302 is configured to acquire the original signal of the blade angle in the pitch mechanism; The original signal is subjected to anti-interference filtering to obtain the filtered original signal, and the filtered original signal is subjected to analog-to-digital conversion to obtain the digital angle value. The environmental parameters of the pitch mechanism are obtained, including temperature and humidity. Based on the environmental parameters, the digital angle value is subjected to linear error compensation processing to obtain the propeller pitch angle.

[0071] Regarding the pitch system control device 300 in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated here.

[0072] Based on the same inventive concept, this disclosure also provides a pitch system, including the above-mentioned pitch system control device.

[0073] like Figure 4 As shown below, the composition of the pitch system and the operating principle of each component are explained using a specific implementation method: I. The pitch system may include a signal acquisition module, a fuzzy control module, a permanent magnet synchronous pitch drive module, a pitch angle feedback module, a power optimization module, and a safety protection module; Among them, the signal acquisition module is used to collect wind speed, generator speed, blade pitch angle and grid power demand parameters in real time and generate multi-source sensor data streams; The fuzzy control module is connected to the signal acquisition module and is used to dynamically decouple multi-source sensor data based on adaptive fuzzy rules, and output pitch angle adjustment commands and torque compensation signals for permanent magnet synchronous motors. The permanent magnet synchronous pitch drive module is connected to the fuzzy control module and is used to drive the permanent magnet synchronous motor to perform pitch control actions according to the pitch angle adjustment command, and to feed back the actual pitch angle to the status monitoring module. The pitch angle feedback module communicates bidirectionally with the permanent magnet synchronous pitch drive module and the fuzzy control module to acquire actual pitch angle data in real time and perform nonlinear error compensation to generate a high-precision feedback signal. The power optimization module is used to integrate the power demand of the power grid with the wind energy capture efficiency model, and dynamically correct the membership function and rule weights of the fuzzy control module to achieve maximum power point tracking. The safety protection module is used for hardware-level fault isolation.

[0074] II. The signal acquisition module may include: The wind speed detection unit is used to measure real-time wind speed and turbulence intensity by fusing lidar and ultrasonic sensors, increasing multi-sensor data fusion, Kalman filtering algorithm to eliminate measurement bias of single sensor in complex wind fields, and optimizing the frequency domain analysis capability of turbulence intensity. The speed sensing unit is used to synchronously acquire the generator rotor speed through a Hall sensor and an encoder. It features a redundant acquisition circuit, supports real-time cross-verification of Hall sensor and encoder signals, and improves the time alignment accuracy of speed data through a timestamp synchronization mechanism. The blade pitch angle feedback unit is used to obtain the current blade angle through a high-precision photoelectric encoder. It introduces a photoelectric encoder temperature drift compensation model to dynamically correct the angle measurement value based on the ambient temperature, thereby reducing nonlinear errors caused by temperature changes. The power grid interface unit is used to receive power grid dispatch instructions and parse target power thresholds. It integrates a power grid dispatch protocol parser, supports multi-protocol compatibility, and achieves low-latency parsing and priority scheduling of target power thresholds.

[0075] III. The fuzzy control module may include: The data preprocessing unit is used to normalize and filter noise from multi-source sensor data. It uses a combination of sliding window filtering and wavelet transform to separate high-frequency noise from effective signals and improve the data signal-to-noise ratio. The fuzzy inference unit has a built-in rule base that associates wind speed and blade pitch angle. It generates initial control commands through a weighted average method. The rule base adopts a hierarchical storage structure, supports online rule addition and deletion, and optimizes the initial rule weight allocation through an expert system. The adaptive correction unit is used to dynamically adjust the activation threshold and output gain of the rule base according to the membership correction coefficient output by the power optimization module. It embeds an online learning mechanism to dynamically optimize the rule activation threshold based on historical control effects, thereby avoiding local optima.

[0076] IV. The permanent magnet synchronous pitch drive module may include: The permanent magnet synchronous motor adopts a vector control strategy to achieve high-precision torque and speed decoupling. The dead-zone compensation algorithm is embedded in the field-oriented control (FOC) to suppress the influence of inverter nonlinear distortion on torque fluctuation. The pitch actuator converts the motor torque into linear pitch displacement through a planetary gearbox and a ball screw. The planetary gearbox is made of carburized steel to improve fatigue resistance. The ball screw preload is adjustable to compensate for transmission backlash under different loads. The position closed-loop unit is used to perform PID correction based on the deviation between the actual pitch angle and the target value, and feeds the correction amount back to the fuzzy control module. The PID parameters are dynamically adjusted through a self-tuning algorithm to match the pitch speed and load changes.

[0077] V. The pitch angle feedback module may include: The encoder acquisition unit is used to capture the original blade angle signal in real time using a high-resolution photoelectric encoder. The signal conditioning unit is used to perform anti-interference filtering and AD conversion on the original signal to generate a digital angle value; The error compensation unit has a built-in blade deformation compensation algorithm that combines ambient temperature and humidity data to correct nonlinear measurement errors. The closed-loop feedback unit is used to feed back the compensated actual pitch angle to the fuzzy control module to correct subsequent commands.

[0078] VI. The power optimization module may include: The wind energy model unit is used to calculate the theoretical maximum wind energy capture efficiency based on the Bates theory, and introduces a real-time air density correction factor and blade pollution coefficient to improve the environmental adaptability of theoretical power calculation. The real-time optimization unit iteratively adjusts the rule weights of the fuzzy control module through the gradient descent algorithm, making the actual power output closer to the theoretical value. The gradient descent algorithm incorporates the momentum term to accelerate convergence and avoid oscillations; it supports multi-threaded parallel computing to reduce optimization latency. The strategy storage unit is used to record historical optimization parameters and quickly invoke preset control strategies when there are sudden changes in grid demand. It uses a time-series database to store historical parameters and supports time-series-based matching of similar operating conditions and strategy recommendation.

[0079] VII. The safety protection module may include: The overload detection unit is used to identify motor stall and phase loss faults; The emergency heave unit is used to perform safe feathering maneuvers; The status self-check unit is used to assess the health of the system.

[0080] The above technical solution has the following beneficial effects: 1. By fusing multi-source sensor data and dynamically decoupling adaptive fuzzy rules, the system achieves coordinated optimization of wind energy capture efficiency and grid power demand. The signal acquisition module collects wind speed, rotational speed and grid demand data in real time. The fuzzy control module generates blade pitch angle adjustment commands based on a dynamic rule base. The power optimization module combines the wind energy model to correct the fuzzy membership degree, forming a closed-loop feedback. This enables the system to adaptively adjust the blade angle when wind speed fluctuates or grid dispatch demand changes, thereby improving wind energy utilization efficiency and reducing power output deviation.

[0081] 2. By linking nonlinear error compensation and fault-tolerant strategies, the robustness and operational stability of the pitch system are enhanced. The pitch angle feedback module compensates for temperature and humidity drift of the actual pitch angle and generates a high-precision feedback signal. The status monitoring module analyzes vibration and temperature rise data in real time. When the tracking error exceeds the limit, the redundant control strategy is triggered. The safety protection module performs hardware-level fault isolation to reduce the impact of environmental interference and mechanical wear on control accuracy, extend the life of key components, and reduce the risk of unplanned downtime.

[0082] 3. Through gradient descent optimization and historical strategy preloading, rapid response and long-term energy efficiency balance under complex operating conditions are achieved. The power optimization module calculates the maximum wind energy capture efficiency based on Bates theory and uses the gradient descent algorithm to iteratively adjust the weights of fuzzy rules so that the actual power approaches the theoretical value. The strategy storage unit records historical optimization parameters and calls the preset strategy when the grid demand changes suddenly, shortening the control convergence time, adapting to different wind conditions and grid dispatch scenarios, and maintaining the efficient and economical operation of the system.

[0083] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of the pitch system control method described above. For example, the computer-readable storage medium may be a memory including program instructions, which may be executed by a processor of an electronic device to perform the pitch system control method described above.

[0084] In another exemplary embodiment, a computer program product is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a code portion for performing the above-described pitch system control method when executed by the programmable device.

[0085] The preferred embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. However, this disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of this disclosure, various simple modifications can be made to the technical solutions of this disclosure, and these simple modifications all fall within the protection scope of this disclosure.

[0086] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0087] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A pitch control system method, characterized in that, The method includes: Obtain the operating parameters of the generator set; Based on the operating parameters, a first pitch angle adjustment command is generated, the pitch mechanism is controlled to operate according to the first pitch angle adjustment command, and the pitch angle of the pitch mechanism is obtained. Determine the deviation between the pitch angle and the preset target pitch angle, and generate a membership correction coefficient based on the operating parameters; If the deviation is greater than the deviation threshold, a second pitch angle adjustment command is generated based on the membership correction coefficient and the operating parameters. The pitch mechanism is controlled to operate according to the second pitch angle adjustment command until the output power of the generator set is greater than or equal to the power demand of the grid and the wind energy capture efficiency corresponding to the output power is optimal.

2. The pitch control method according to claim 1, characterized in that, The operating parameters include wind speed, generator speed, and grid power demand. Generating a first pitch angle adjustment command based on these operating parameters includes: Based on adaptive fuzzy rules, the wind speed, the generator speed, and the power demand of the power grid are dynamically decoupled to generate a first pitch angle adjustment command.

3. The pitch control method according to claim 2, characterized in that, The process of dynamically decoupling the wind speed, generator speed, and grid power demand based on adaptive fuzzy rules to generate a first pitch angle adjustment command includes: The wind speed, generator speed, and power grid demand are normalized and noise filtered to obtain the processed wind speed, generator speed, and power grid demand. Based on a preset rule base for the association between wind speed and pitch angle, a weighted average calculation is performed on the processed wind speed, generator speed, and power demand of the power grid to generate a first pitch angle adjustment command.

4. The pitch control method according to claim 1, characterized in that, The step of generating a second pitch angle adjustment command based on the membership correction coefficient and the operating parameters includes: Based on the membership correction coefficient, the activation threshold and output gain of the association rule base of the preset wind speed and pitch angle are dynamically adjusted to obtain the intermediate association rule base, and the rule weight of the intermediate association rule base is updated to obtain the target association rule base. The association rule base of the preset wind speed and pitch angle adopts a hierarchical storage structure. Based on the sliding window, wavelet transform noise reduction processing is performed on the operating parameters to obtain an effective signal; Based on the target association rule base, the effective signals are weighted and averaged to generate a second pitch angle adjustment command.

5. The pitch system control method according to claim 4, characterized in that, The operating parameters include the power demand of the power grid. The step of updating the rule weights of the intermediate association rule base to obtain the target association rule base includes: Determine the power deviation between the power demand of the power grid and the theoretical maximum power of the power grid, and use the square of the power deviation as the determination of the update gradient; Based on the update gradient and the preset learning rate, the original rule weights of the intermediate association rule base are updated to obtain the target rule weights. Based on the target rule weights, the target association rule base is obtained.

6. The pitch system control method according to claim 5, characterized in that, The step of updating the original rule weights of the intermediate association rule base according to the update gradient and the preset learning rate to obtain the target rule weights includes: Substituting the update gradient, the original rule weights of the intermediate association rule base, and the preset learning rate into the following calculation formula, the target rule weights are obtained; ， Among them, W i new The weights of the target rules, W i η represents the original rule weights, η represents the preset learning rate, and J represents the updated gradient.

7. The pitch system control method according to claim 1, characterized in that, The step of obtaining the pitch angle of the pitch mechanism includes: Obtain the original signal of the blade angle in the pitch mechanism; The original signal is subjected to anti-interference filtering to obtain the filtered original signal, and the filtered original signal is subjected to analog-to-digital conversion to obtain the digital angle value. The environmental parameters of the pitch mechanism are obtained, including temperature and humidity. Based on the environmental parameters, the digital angle value is subjected to linear error compensation processing to obtain the propeller pitch angle.

8. A pitch control system, characterized in that, The device includes: The signal acquisition module is configured to acquire the operating parameters of the generator set; The first control module is configured to generate a first pitch angle adjustment command based on the operating parameters, control the pitch mechanism to operate based on the first pitch angle adjustment command, and obtain the pitch angle of the pitch mechanism. The first execution module is configured to determine the deviation between the pitch angle and the preset target pitch angle, and generate a membership correction coefficient based on the operating parameters. The second control module is configured to generate a second pitch angle adjustment command based on the membership correction coefficient and the operating parameters when the deviation is greater than the deviation threshold. The second execution module is configured to control the operation of the pitch mechanism according to the second pitch angle adjustment command until the output power of the generator set is greater than or equal to the power demand of the grid and the wind energy capture efficiency corresponding to the output power is optimal.

9. A pitch control system, characterized in that, Includes the pitch system control device as described in claim 8.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1-7.

11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1-7.

Images