An automatic compensation control system for aerodynamic imbalance of a wind turbine rotor

Abstract

This invention discloses an automatic compensation control system for wind turbine rotor aerodynamic imbalance, comprising: a compensation control module, which calculates a set of blade aerodynamic compensation angles based on parameters transmitted from a parameter optimization module, and superimposes these aerodynamic compensation angles with pitch control outputs to obtain the final command; a data acquisition module, used to collect and store nacelle front-to-back and left-to-right vibration data; an index evaluation module, which constructs an index evaluation function to evaluate the quality of the rotor aerodynamic compensation angle based on the nacelle front-to-back and left-to-right vibration data; and a parameter optimization module, which sets an optimization space and uses a traversal optimization method within the optimization space to find the optimal parameters, and transmits the optimal parameters to the compensation control module. This invention reduces the fatigue load on the turbine caused by rotor aerodynamic imbalance by obtaining the optimal rotor aerodynamic compensation angle.

Description

Technical Field

[0001] This invention relates to the technical field of wind turbine control, and in particular to an automatic compensation control system for aerodynamic imbalance of wind turbine rotor. Background Technology

[0002] A wind turbine generator set includes a rotor, gearbox, generator, and tower. The rotor is the main component of the wind turbine generator set. Generally, wind turbine generator sets use a three-bladed rotor, with the blade roots mounted on the hub pitch bearings and equipped with a pitch control system. Due to manufacturing errors, installation deviations, and contamination and creep deviations during operation, the rotor becomes aerodynamically unbalanced. Therefore, a certain blade pitch angle deviation must be considered during the design process to reflect the deviations generated during manufacturing, installation, and operation. Aerodynamic imbalance of the rotor can cause fatigue loads on the entire generator set, tower vibration, and even component damage. To reduce the fatigue loads and vibrations of the entire generator set, a technical solution that can effectively compensate for the aerodynamic imbalance of the rotor needs to be proposed. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings and deficiencies of the prior art and provide an automatic compensation control system for the aerodynamic imbalance of wind turbine rotors, so as to obtain the optimal aerodynamic compensation angle of the rotor and reduce the fatigue load of the unit caused by the aerodynamic imbalance of the rotor.

[0004] The objective of this invention is achieved through the following technical solution: an automatic compensation control system for aerodynamic imbalance of wind turbine rotor, comprising a compensation control module, a data acquisition module, an index evaluation module, and a parameter optimization module;

[0005] The compensation control module calculates a set of blade aerodynamic compensation angles based on the parameters passed from the parameter optimization module. The aerodynamic compensation angles of the three blades are included. The aerodynamic compensation angles are then superimposed with the pitch control output pitch command to obtain the final command, which is then sent to the pitch system for execution.

[0006] The data acquisition module is used to collect and store vibration data during unit operation, including front and rear vibration data of the nacelle and left and right vibration data of the nacelle, and transmit the stored vibration data to the index evaluation module.

[0007] The index evaluation module constructs an index evaluation function to evaluate the quality of the wind turbine aerodynamic compensation angle based on the front and rear vibration data and the left and right vibration data of the nacelle, and passes the calculated evaluation function value to the parameter optimization module.

[0008] The parameter optimization module sets up an optimization space, and uses a traversal optimization method to find the optimal parameters within the optimization space, and then passes the optimal parameters to the compensation control module.

[0009] Furthermore, the compensation control module uses trigonometric functions to construct aerodynamic compensation angles. The amplitude and phase of these trigonometric functions uniquely determine a set of blade aerodynamic compensation angles. The specific formula is defined as follows:

[0010]

[0011] In the above formula, This indicates the aerodynamic compensation angle of blade 1; This indicates the aerodynamic compensation angle of blade 2; θ represents the aerodynamic compensation angle of blade 3; r represents the amplitude of the trigonometric function; θ represents the phase of the trigonometric function.

[0012] According to the formula for the aerodynamic compensation angle, if the amplitude and phase of the trigonometric function are determined, the aerodynamic compensation angle of the three blades can be calculated. Thus, finding the optimal aerodynamic compensation angle for the three blades is transformed into finding the amplitude and phase of the trigonometric function, that is, reducing the three-variable optimization space to a two-variable optimization space. The amplitude and phase of the trigonometric function can be obtained from the parameter optimization module.

[0013] Finally, the aerodynamic compensation angles of the three blades are obtained. These compensation angles are then superimposed with the pitch control outputs to obtain the final commands for the three blades. The specific formulas are defined as follows:

[0014]

[0015] In the above formula, This indicates the final instruction for blade 1; This indicates the final instruction for blade 2; This indicates the final instruction for blade 3; This indicates the angle command for blade 1 output by the pitch control; This indicates the angle command for blade 2 output by the pitch control; This indicates the angle command for blade 3 output by the pitch control.

[0016] Furthermore, the data acquisition module includes two storage arrays A and B, which are used to store the front-to-back vibration data and the left-to-right vibration data of the cabin, respectively. The front-to-back vibration data of the cabin is stored in storage array A, and the left-to-right vibration data of the cabin is stored in storage array B. Storage arrays A and B have a fixed length. In order to meet the requirements of data analysis and processing, the length of storage arrays A and B should be sufficient to store vibration data of ten minutes or more.

[0017] The definitions of storage arrays A and B are as follows:

[0018]

[0019] In the above formula, This represents the measured value of the fore-and-aft vibration of the cabin at time 1; This represents the measured value of the fore-and-aft vibration of the cabin at time 2; This represents the measured value of the fore-and-aft vibration of the cabin at time N; N represents the length of the storage array. This represents the measured value of the left-right vibration of the cabin at time 1; This represents the measured value of the left-right vibration of the cabin at time 2; This represents the measured value of the left-right vibration of the cabin at time N;

[0020] Once the vibration data in storage arrays A and B are full, the data is transferred to the indicator evaluation module, and then storage arrays A and B are cleared.

[0021] Furthermore, the indicator evaluation module constructs the indicator evaluation function according to the following scheme:

[0022] First, the front-to-back vibration data and the left-to-right vibration data of the cabin are filtered by a filter to extract the 1P frequency component from the vibration data.

[0023] Secondly, the root mean square error (RMSE) was calculated for the filtered forward / backward and left / right vibration data of the cabin. The RMSE reflects the average amplitude of the vibration; the larger the RMSE, the larger the vibration amplitude. The formula for calculating the RMSE is as follows:

[0024]

[0025] In the above formula, σ fa This indicates the root mean square error of the fore-and-aft vibration of the cabin; This represents the filtered value for the fore-and-aft vibration of the cabin at time 1; This represents the filtered value for the fore-and-aft vibration of the cabin at time 2; This represents the filtered value of the fore-and-aft vibration of the cabin at time N; N represents the length of the storage array; σ ss This indicates the root mean square error of the lateral vibration of the cabin; This represents the filtered value for the left-right vibration of the cabin at time 1; This represents the filter value for the left-right vibration of the cabin at time 2; This represents the filter value for the left-right vibration of the cabin at time N;

[0026] Finally, based on the root mean square error of cabin forward and backward vibration and the root mean square error of cabin left and right vibration, the following index evaluation function is constructed:

[0027] f(r,θ)=C1·σ fa +C2·σ ss

[0028] In the above formula, f(r,θ) represents the index evaluation function, r represents the amplitude of the trigonometric function, θ represents the phase of the trigonometric function, and the magnitude of the evaluation function value reflects the quality of r and θ; C1 represents the weighted value of the fore-and-aft vibration of the cabin; σ fa C2 represents the root mean square error of the fore-and-aft vibration of the nacelle; C2 represents the weighted value of the lateral vibration of the nacelle; σ ss This indicates the root mean square error of the left-right vibration of the cabin.

[0029] Furthermore, the filter is selected as a first-order low-pass, second-order low-pass, or band-pass filter to filter out the influence of other frequency components.

[0030] Furthermore, the optimization space of the parameter optimization module is set to a two-dimensional space, that is, the optimization range of the magnitude and phase of the trigonometric function is set; since the phase range of the trigonometric function is 0 to 360 degrees, the optimization space is a circular region, and the optimization space is defined as follows:

[0031] S={(r,θ)|0≤r≤R,0°≤θ<360°}

[0032] In the above formula, S represents the optimization space, which contains two variables (r, θ); r represents the magnitude of the trigonometric function; R represents the upper limit of the magnitude of the trigonometric function; θ represents the phase of the trigonometric function; the optimization space S is a circular region with a radius less than or equal to R.

[0033] Furthermore, the parameter optimization module uses a traversal optimization method that divides the entire optimization space into several parts by setting an optimization step size, and evaluates the merits of each segmentation node. For the optimization space S, the optimization step size Δr is set in the radial direction and the optimization step size Δθ is set in the circumferential direction. The entire optimization space forms a spider web diagram, and each intersection point is the corresponding optimization point.

[0034] The traversal optimization process starts from the origin. First, it moves one step Δr radially, and then moves 0 degrees circumferentially with a step size Δθ until it completes one 360-degree traversal. Next, it moves one step Δr radially again, and then traverses one circumferentially. This process continues until the entire optimization space has been traversed.

[0035] Furthermore, during the traversal optimization process, the parameter optimization module first provides the current optimization position, i.e., the amplitude r and phase θ of the trigonometric function; secondly, the compensation control module calculates the aerodynamic compensation angle of the three blades based on the amplitude r and phase θ of the trigonometric function and performs compensation control; thirdly, the data acquisition module collects the vibration data of the engine room during unit operation; then, the index evaluation module calculates the evaluation function value based on the index evaluation function; finally, the parameter optimization module records the evaluation function value of the corresponding point and compares this point's evaluation function value with the historical optimal point's evaluation function value. If the evaluation function value of this point is less than the historical optimal point's evaluation function value, then the historical optimal point is updated to the current position point. Thus, the traversal optimization of a point is completed.

[0036] After completing the traversal and optimization process of all points, the parameter optimization module will output the optimal trigonometric function amplitude and phase corresponding to the historical optimal point. The compensation control module will then set the optimal aerodynamic compensation angle of the three blades based on the optimal trigonometric function amplitude and phase, thus completing the automatic optimization process.

[0037] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0038] 1. This invention proposes a trigonometric function method to construct the wind turbine aerodynamic compensation angle, which reduces the original three-variable optimization space to a two-variable optimization space, thereby reducing the difficulty and time of optimization.

[0039] 2. Based on nacelle vibration data, this invention proposes an evaluation function for the aerodynamic compensation angle of the wind turbine to assess its quality. This solution utilizes existing turbine nacelle vibration sensors, eliminating the need for additional sensors.

[0040] 3. The automatic compensation control system for wind turbine aerodynamic imbalance proposed in this invention can effectively eliminate wind turbine aerodynamic imbalance caused by manufacturing, installation and operation, thereby effectively reducing the fatigue load of the unit. Attached Figure Description

[0041] Figure 1 This is an architecture diagram of the system of the present invention. Detailed Implementation

[0042] The present invention will be further described in detail below with reference to the embodiments and accompanying drawings, but the embodiments of the present invention are not limited thereto.

[0043] This embodiment discloses an automatic compensation control system for aerodynamic imbalance of wind turbine rotor, such as... Figure 1 As shown, the system comprises four functional modules: a compensation control module, a data acquisition module, an index evaluation module, and a parameter optimization module. The specific details of each functional module are as follows:

[0044] The main function of the compensation control module is to eliminate the aerodynamic imbalance bending moment of the wind turbine. This module contains a set of blade aerodynamic compensation angles. Based on the parameters passed from the parameter optimization module, the aerodynamic compensation angles of the three blades are calculated. These compensation angles are then superimposed with the pitch control output of the pitch control to obtain the final command, which is then sent to the pitch system for execution.

[0045] The aerodynamic compensation angle of the compensation control module includes the compensation angles of three blades: blade 1, blade 2, and blade 3. This scheme uses trigonometric functions to construct the aerodynamic compensation angles; the amplitude and phase of the trigonometric functions uniquely determine a set of aerodynamic compensation angles. The specific formulas are defined as follows:

[0046]

[0047] In the above formula, This indicates the aerodynamic compensation angle of blade 1; This indicates the aerodynamic compensation angle of blade 2; θ represents the aerodynamic compensation angle of blade 3; r represents the amplitude of the trigonometric function; θ represents the phase of the trigonometric function.

[0048] According to the formula for the aerodynamic compensation angle, if the amplitude and phase of the trigonometric functions are determined, the aerodynamic compensation angle of the three blades can be calculated. This transforms the search for the optimal aerodynamic compensation angle for the three blades into the optimization of the trigonometric function amplitude and phase, reducing the three-variable optimization space to a two-variable optimization space. The amplitude and phase of the trigonometric functions can be obtained from the parameter optimization module.

[0049] Finally, the aerodynamic compensation angles of the three blades are obtained. These compensation angles are then superimposed with the pitch control outputs to obtain the final commands for the three blades. The specific formula is as follows:

[0050]

[0051] In the above formula, This indicates the final instruction for blade 1; This indicates the final instruction for blade 2; This indicates the final instruction for blade 3; This indicates the angle command for blade 1 output by the pitch control; This indicates the angle command for blade 2 output by the pitch control; This indicates the angle command for blade 3 output by the pitch control.

[0052] The main function of the data acquisition module is to collect and store vibration data during the operation of the unit. After the aerodynamic imbalance compensation angle of the wind turbine is applied to the wind turbine, the vibration amplitude of the nacelle of the unit will inevitably change; therefore, the data acquisition module contains two storage arrays, A and B, which are used to store the front-to-back vibration data of the nacelle and the left-to-right vibration data of the nacelle, respectively.

[0053] The forward and aft vibration data of the nacelle is stored in storage array A, and the left and right vibration data of the nacelle is stored in storage array B. Storage arrays A and B have fixed lengths. To meet the requirements of data analysis and processing, the lengths of storage arrays A and B should be sufficient to store vibration data of ten minutes or more.

[0054] The definitions of storage arrays A and B are as follows:

[0055]

[0056] In the above formula, This represents the measured value of the fore-and-aft vibration of the cabin at time 1; This represents the measured value of the fore-and-aft vibration of the cabin at time 2; This represents the measured value of the fore-and-aft vibration of the cabin at time N; N represents the length of the storage array. This represents the measured value of the left-right vibration of the cabin at time 1; This represents the measured value of the left-right vibration of the cabin at time 2; This represents the measured value of the left-right vibration of the cabin at time N.

[0057] Once the vibration data in storage arrays A and B are full, the data is transferred to the indicator evaluation module, and then storage arrays A and B are cleared.

[0058] The main function of the index evaluation module is to construct an index evaluation function based on the nacelle's forward / backward and left / right vibration data to evaluate the quality of the rotor aerodynamic compensation angle. Aerodynamic imbalance of the rotor is directly reflected in the nacelle vibration data, including both forward / backward and left / right vibrations. When the rotor aerodynamic imbalance increases, the amplitude of the 1P frequency in the forward / backward and left / right vibration data will increase significantly; conversely, if the rotor aerodynamic imbalance is well compensated, the amplitude of the 1P frequency in the forward / backward and left / right vibration data will decrease significantly.

[0059] Therefore, the index evaluation function is constructed according to the following scheme:

[0060] First, the front-to-back vibration data and the left-to-right vibration data of the cabin are filtered to extract the 1P frequency component from the vibration data. The filter can be a first-order low-pass, second-order low-pass, or band-pass filter to filter out the influence of other frequency components.

[0061] Secondly, the root mean square error (RMSE) was calculated for both the forward / backward and left / right vibration data of the nacelle after filtering. The RMSE reflects the average amplitude of the vibration; a larger RMSE indicates a larger vibration amplitude. The formula for calculating the RMSE is as follows:

[0062]

[0063] In the above formula, σ fa This indicates the root mean square error of the fore-and-aft vibration of the cabin; This represents the filtered value for the fore-and-aft vibration of the cabin at time 1; This represents the filtered value for the fore-and-aft vibration of the cabin at time 2; This represents the filtered value of the fore-and-aft vibration of the cabin at time N; N represents the length of the storage array; σ ss This indicates the root mean square error of the lateral vibration of the cabin; This represents the filtered value for the left-right vibration of the cabin at time 1; This represents the filter value for the left-right vibration of the cabin at time 2; This represents the filter value for the left-right vibration of the cabin at time N.

[0064] Finally, based on the root mean square error of cabin forward and backward vibration and the root mean square error of cabin left and right vibration, the following index evaluation function is constructed:

[0065] f(r,θ)=C1·σ fa +C2·σ ss

[0066] In the above formula, f(r,θ) represents the index evaluation function, r represents the amplitude of the trigonometric function, θ represents the phase of the trigonometric function, and the magnitude of the evaluation function value reflects the quality of r and θ; C1 represents the weighted value of the fore-and-aft vibration of the cabin; σ fa C2 represents the root mean square error of the fore-and-aft vibration of the nacelle; C2 represents the weighted value of the lateral vibration of the nacelle; σ ss This indicates the root mean square error of the left-right vibration of the cabin.

[0067] The evaluation function value calculated by the indicator evaluation module is passed to the parameter optimization module.

[0068] The main function of the parameter optimization module is to set the optimization space, and then use the traversal optimization method to find the optimal parameters within the optimization space to obtain the optimal wind turbine aerodynamic compensation angle.

[0069] The optimization space is set as a two-dimensional space, that is, the optimization range for the amplitude and phase of the trigonometric functions is defined. Since the phase range of the trigonometric functions is from 0 to 360 degrees, the optimization space is a circular region. The optimization space is defined as follows:

[0070] S={(r,θ)|0≤r≤R,0°≤θ≤360°}

[0071] In the above formula, S represents the optimization space, which contains two variables (r, θ); r represents the magnitude of the trigonometric function; R represents the upper limit of the magnitude of the trigonometric function; θ represents the phase of the trigonometric function; the optimization space S is a circular region with a radius less than or equal to R.

[0072] The traversal optimization method divides the entire optimization space into several parts by setting an optimization step size, and evaluates the merits of each segment node. Specifically, for the optimization space S, the optimization step size Δr is set in the radial direction and the optimization step size Δθ is set in the circumferential direction. The entire optimization space forms a spider web diagram, and each intersection point is the corresponding optimization point.

[0073] The traversal optimization process starts from the origin. First, it moves one step Δr radially, and then moves 0 degrees circumferentially with a step size Δθ until it completes one 360-degree traversal. Next, it moves one step Δr radially again, and then traverses one circumferentially. This process continues until the entire optimization space has been traversed.

[0074] During the traversal optimization process, the parameter optimization module first provides the current optimization position, i.e., the trigonometric function amplitude *r* and phase *θ*. Second, the compensation control module calculates the aerodynamic compensation angles of the three blades based on the trigonometric function amplitude *r* and phase *θ*, and performs compensation control. Third, the data acquisition module collects vibration data from the generator nacelle. Then, the performance evaluation module calculates the evaluation function value based on the performance evaluation function. Finally, the parameter optimization module records the evaluation function value for this point and compares it with the historical optimal value. If the evaluation function value for this point is less than the historical optimal value, the historical optimal value is updated to the current position. This completes the traversal optimization for one point.

[0075] After completing the optimization process across all points, the parameter optimization module will output the optimal trigonometric function amplitude and phase corresponding to the historical optimal points. Based on these optimal trigonometric function amplitudes and phases, the compensation control module sets the optimal aerodynamic compensation angles for the three blades, completing the automatic optimization process.

[0076] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. An automatic compensation control system for aerodynamic imbalance of a wind turbine rotor, characterized in that, It includes a compensation control module, a data acquisition module, an indicator evaluation module, and a parameter optimization module; The compensation control module calculates a set of blade aerodynamic compensation angles based on the parameters passed from the parameter optimization module. The aerodynamic compensation angles of the three blades are included. The aerodynamic compensation angles are then superimposed with the pitch control output pitch command to obtain the final command, which is then sent to the pitch system for execution. The data acquisition module is used to collect and store vibration data during unit operation, including front and rear vibration data of the nacelle and left and right vibration data of the nacelle, and transmit the stored vibration data to the index evaluation module. The index evaluation module constructs an index evaluation function to evaluate the quality of the wind turbine aerodynamic compensation angle based on the front and rear vibration data and the left and right vibration data of the nacelle, and passes the calculated evaluation function value to the parameter optimization module. The parameter optimization module sets up an optimization space, and uses a traversal optimization method to find the optimal parameters within the optimization space, and then passes the optimal parameters to the compensation control module. The compensation control module uses trigonometric functions to construct aerodynamic compensation angles. The amplitude and phase of the trigonometric functions can uniquely determine a set of blade aerodynamic compensation angles. The specific formula is defined as follows: ； In the above formula, This indicates the aerodynamic compensation angle of blade 1; This indicates the aerodynamic compensation angle of blade 2; This indicates the aerodynamic compensation angle of blade 3; Represents the magnitude of a trigonometric function; Indicates the phase of a trigonometric function; According to the formula for the aerodynamic compensation angle, if the amplitude and phase of the trigonometric function are determined, the aerodynamic compensation angle of the three blades can be calculated. Thus, finding the optimal aerodynamic compensation angle for the three blades is transformed into finding the amplitude and phase of the trigonometric function, that is, reducing the three-variable optimization space to a two-variable optimization space. The amplitude and phase of the trigonometric function can be obtained from the parameter optimization module. Finally, the aerodynamic compensation angles of the three blades are obtained. These compensation angles are then superimposed with the pitch control outputs to obtain the final commands for the three blades. The specific formulas are defined as follows: ； In the above formula, This indicates the final instruction for blade 1; This indicates the final instruction for blade 2; This indicates the final instruction for blade 3; This indicates the angle command for blade 1 output by the pitch control; This indicates the angle command for blade 2 output by the pitch control; This indicates the angle command for blade 3 output by the pitch control.

2. The automatic compensation control system for aerodynamic imbalance of wind turbine rotor according to claim 1, characterized in that: The data acquisition module includes two storage arrays, A and B, which are used to store the front-to-back vibration data and the left-to-right vibration data of the cabin, respectively. The front-to-back vibration data of the cabin is stored in storage array A, and the left-to-right vibration data of the cabin is stored in storage array B. Storage arrays A and B have a fixed length. In order to meet the requirements of data analysis and processing, the length of storage arrays A and B should be sufficient to store vibration data of more than ten minutes. The definitions of storage arrays A and B are as follows: ； In the above formula, This represents the measured value of the fore-and-aft vibration of the cabin at time 1; This represents the measured value of the fore-and-aft vibration of the cabin at time 2; This represents the measured value of the fore-and-aft vibration of the cabin at time N; N represents the length of the storage array. This represents the measured value of the left-right vibration of the cabin at time 1; This represents the measured value of the left-right vibration of the cabin at time 2; This represents the measured value of the left-right vibration of the cabin at time N; Once the vibration data in storage arrays A and B are full, the data is transferred to the indicator evaluation module, and then storage arrays A and B are cleared.

3. The automatic compensation control system for aerodynamic imbalance of wind turbine rotor according to claim 2, characterized in that: The indicator evaluation module constructs the indicator evaluation function according to the following scheme: First, the front-to-back vibration data and the left-to-right vibration data of the cabin are filtered by a filter to extract the 1P frequency component from the vibration data. Secondly, the root mean square error (RMSE) was calculated for the filtered forward / backward and left / right vibration data of the cabin. The RMSE reflects the average amplitude of the vibration; the larger the RMSE, the larger the vibration amplitude. The formula for calculating the RMSE is as follows: ； In the above formula, This indicates the root mean square error of the fore-and-aft vibration of the cabin; This represents the filtered value for the fore-and-aft vibration of the cabin at time 1; This represents the filtered value for the fore-and-aft vibration of the cabin at time 2; This represents the filtered value for the fore-and-aft vibration of the cabin at time N; N represents the length of the storage array; This indicates the root mean square error of the lateral vibration of the cabin; This represents the filtered value for the left-right vibration of the cabin at time 1; This represents the filter value for the left-right vibration of the cabin at time 2; This represents the filter value for the left-right vibration of the cabin at time N; Finally, based on the root mean square error of cabin forward and backward vibration and the root mean square error of cabin left and right vibration, the following index evaluation function is constructed: ； In the above formula, This represents the indicator evaluation function. This represents the magnitude of the trigonometric function. Representing the phase of a trigonometric function, evaluating the magnitude of the function value reflects... and The advantages and disadvantages; This represents the weighted value of the fore-and-aft vibration of the cabin; This indicates the root mean square error of the fore-and-aft vibration of the cabin; This represents the weighted value of the left-right vibration of the cabin; This indicates the root mean square error of the left-right vibration of the cabin.

4. An automatic compensation control system for aerodynamic imbalance of wind turbine rotor according to claim 3, characterized in that: The filter is selected as a first-order low-pass, second-order low-pass, or band-pass filter to filter out the influence of other frequency components.

5. An automatic compensation control system for aerodynamic imbalance of wind turbine rotor according to claim 1, characterized in that: The optimization space of the parameter optimization module is set to a two-dimensional space, that is, the optimization range of the magnitude and phase of the trigonometric function is set; since the phase range of the trigonometric function is 0 to 360 degrees, the optimization space is a circular region, and the optimization space is defined as follows: ； In the above formula, This represents the search space, containing two variables. ; Represents the magnitude of a trigonometric function; This represents the upper limit of the amplitude of a trigonometric function; Phase of trigonometric functions; search space It is a circular region with a radius less than or equal to .

6. An automatic compensation control system for aerodynamic imbalance of wind turbine rotor according to claim 5, characterized in that: The parameter optimization module employs a traversal optimization method that divides the entire optimization space into several parts by setting an optimization step size, and evaluates the merits of each segment node; for the optimization space... Set the optimization step size in the radius direction. Optimization step size is set in the circumferential direction. The entire optimization space forms a spider web diagram, with each intersection being the corresponding optimization point; The traversal optimization process starts from the origin and first moves one step radially. Then, along the circumference according to the step length Starting from 0 degrees, it completes one cycle by turning 360 degrees. Next, move one more step radially. Then traverse the circumferential direction once more; continue in this manner until the entire search space has been traversed.

7. An automatic compensation control system for aerodynamic imbalance of wind turbine rotor according to claim 6, characterized in that: During the traversal optimization process, the parameter optimization module first provides the current optimization position, which is the magnitude of the trigonometric function. and phase ; Secondly, the compensation control module is based on the amplitude of the trigonometric function. and phase The aerodynamic compensation angles of the three blades are calculated and compensation control is performed; next, the data acquisition module collects vibration data of the unit's nacelle during operation. Then, the indicator evaluation module calculates the evaluation function value based on the indicator evaluation function; finally, the parameter optimization module records the evaluation function value of the corresponding point and compares the evaluation function value of this point with the evaluation function value of the historical best point. If the evaluation function value of this point is less than the evaluation function value of the historical best point, the historical best point is updated to the current position point. Thus, the traversal optimization of a point is completed. After completing the traversal and optimization process of all points, the parameter optimization module will output the optimal trigonometric function amplitude and phase corresponding to the historical optimal point. The compensation control module will then set the optimal aerodynamic compensation angle of the three blades based on the optimal trigonometric function amplitude and phase, thus completing the automatic optimization process.

Images