CN113685314B - Pitch control method, system and readable storage medium - Google Patents

Abstract

The present application provides an independent pitch control method, system and readable storage medium for a wind turbine. The independent pitch control method includes determining an objective function and a data-driven model, wherein the independent variable of the objective function includes the load reduction control objective, the dependent variable of the objective function is used to represent the structural load of the wind turbine in the pitch control, and the data-driven model is used. The independent variables include the pitch angle speed of the wind turbine, and the data-driven model is used to determine the functional relationship between at least part of the load reduction control target and the pitch angle speed; according to the objective function and the data-driven model, the target propeller corresponding to the minimum value of the objective function is determined Pitch angle speed; according to the target pitch angle speed, the fan is controlled by pitch angle. The load drop effect is good.

Description

Pitch control method, system and readable storage medium

technical field

The present invention relates to the field of wind power, and in particular, to a pitch control method, system and readable storage medium.

Background technique

As one of the core parts of the unit control system, the pitch system plays an important role in the safe, stable and efficient operation of the wind turbine. On the one hand, the pitch controller changes the resistance to the incoming wind speed by adjusting the pitch angle of the fan blades, thereby controlling the aerodynamic torque and power captured by the wind rotor. On the other hand, in the process of fan rotation, due to the influence of horizontal and vertical wind shear, tower shadow effect, wind turbulence, yaw misalignment and other factors, the aerodynamic load on the fan impeller surface is in an unbalanced state. Unbalanced aerodynamic loads can cause component structural loads on the fan impeller surface, resulting in fan vibration. In some fan load reduction control strategies, the pitch angle of each fan blade is independently adjusted by the fan pitch system, and the aerodynamic load on each fan blade is changed to reduce the periodic load of the fan blade and the fan impeller. Asymmetric loading of the face.

However, with the increase of the single capacity of the fan and the lengthening of the fan blades, the load reduction effect of these fan load reduction control strategies is reduced.

SUMMARY OF THE INVENTION

The present application provides a pitch control method, system and readable storage medium, which have better load reduction effect.

The present application provides a pitch control method, and the pitch control method includes:

Determine the objective function and the data-driven model, wherein the independent variable of the objective function includes a load reduction control objective, the dependent variable of the objective function is used to represent the size of the structural load of the wind turbine in the pitch control, and the data-driven model is used. for determining the functional relationship between at least part of the derating control target and the pitch angle rotational speed;

According to the objective function and the data-driven model, determine the target pitch angle rotation speed corresponding to the minimum value of the objective function;

According to the target pitch angle rotation speed, pitch control is performed on the wind turbine.

The present application provides a pitch control system, comprising one or more processors for implementing the pitch control method described in any one of the above

The present application provides a readable storage medium on which a program is stored, and when the program is executed by a processor, implements the pitch control method described in any one of the above.

The pitch control method determines the objective function and the data-driven model. According to the objective function and the data-driven model, the target pitch angle speed corresponding to the minimum structural load value is determined, and then the fan is pitched based on the target pitch angle speed. By controlling the pitch angle speed, the wind turbine can run at the pitch angle speed when the structural load is minimized during the pitching process, which avoids the problem of decreasing the effectiveness of the pitch control strategy caused by the constant pitch angle speed, and improves the The load reduction effect of the fan in the pitch process is obtained.

Description of drawings

Fig. 1 is a structural schematic diagram of a fan;

Fig. 2 is a partial enlarged cross-sectional view of the fan in Fig. 1;

Fig. 3 is a schematic diagram of a pitch control flow diagram of the fan in Fig. 1 in a technology;

Figure 4a is a time series comparison diagram of the pitch angle and blade root moment of the fan under one working condition;

Fig. 4b is another time series comparison diagram of the pitch angle and blade root moment of the fan under the same working conditions as Fig. 4a;

5 is a flowchart of a pitch control method provided by an embodiment of the present application;

Fig. 6 is the sub-flow chart of step S52 in Fig. 5;

FIG. 7 is a block diagram of modules of a pitch control system provided by an embodiment of the present application.

Detailed ways

Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. Where the following description refers to the drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatus and methods consistent with some aspects of one or more embodiments of this specification, as recited in the appended claims.

It should be noted that: in other embodiments, the steps of the corresponding methods are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may also be combined into a single step in other embodiments. describe.

FIG. 1 is a schematic structural diagram of a fan 100 . Referring to FIG. 1, a wind turbine 100 is a wind generator, or referred to as a wind turbine. The wind turbine 100 includes a tower 11 extending from a support system 14 , a nacelle 12 mounted on the tower 11 , and a rotor 13 coupled to the nacelle 12 . The rotor 13 includes a rotatable hub 131 coupled to the nacelle 12 and at least one blade 132 coupled to and extending outwardly from the hub 131 .

In some embodiments, where rotor 13 includes a plurality of blades 132, the blades 132 are spaced around hub 131 to facilitate rotation of rotor 13 to enable kinetic energy to be converted from wind energy to usable mechanical energy, and subsequently to electrical energy.

In this embodiment, the rotor 13 includes three blades 132 .

In other embodiments, the rotor 13 may include more or less than three blades 132 .

FIG. 2 is a partially enlarged cross-sectional view of the fan 100 of FIG. 1 . Referring to FIGS. 1 and 2 , the nacelle 12 includes a rotor shaft 122 (also referred to as a main shaft or low speed shaft), a gearbox 123 , a high speed shaft 124 , a coupling 125 and a motor 126 . Hub 131 is rotatably coupled to generator 126 within nacelle 12 by rotor shaft 122 , gearbox 123 , high speed shaft 124 , and coupling 125 . The blades 132 drive the rotation of the hub 131 , which drives the rotor shaft 122 to rotate, and the rotation of the rotor shaft 122 drives the gearbox 123 , which then drives the high-speed shaft 124 . The high-speed shaft 124 drives the generator 126 to generate electricity through the coupling 125 . In this way, the conversion from wind energy to mechanical energy and then to electrical energy is realized.

In some embodiments, hub 131 includes pitch assembly 130 . Pitch assembly 130 includes pitch drive system 135 , and pitch controller 134 operatively coupled to pitch drive system 135 . Each pitch drive system 135 is coupled to a corresponding blade 132 . The pitch controller 134 changes the pitch angle or blade pitch of the corresponding blade 132 along the pitch axis 133 by controlling the pitch drive system 135 (ie, determines the perspective angle of the blade 132 relative to the direction 17 of the wind) , to control the load and power generated by the fan 100 . For example, when the wind turbine 100 is working, the pitch controller 134 can control the pitch drive system 135 to control the blades 132 to rotate around the pitch axis 133 , so that the blades 132 rotate to the feathering position, thereby helping to reduce the rotational speed of the rotor 13 And/or it is beneficial to stop the rotor 13 . For another example, when the wind turbine 100 is working, the pitch controller 134 can control the pitch angle of the blades 132 by controlling the pitch drive system 135 to control the rotational speed of the rotor 13 , thereby controlling the power of the generator 126 . Specifically, pitch assembly 130 includes at least one pitch bearing 136 coupled to hub 131 and coupled to respective blades 132 . Pitch drive system 135 includes pitch drive motor 137 , pitch drive gearbox 138 , and pitch drive pinion 139 . Pitch drive motor 137 is coupled to pitch drive gearbox 138 such that pitch drive motor 137 applies mechanical force to pitch drive gearbox 138 . Pitch drive gearbox 138 is coupled to pitch drive pinion 139 such that pitch drive pinion 139 is rotated by pitch drive gearbox 138 . Pitch bearing 136 is coupled to pitch drive pinion 139 such that rotation of pitch drive pinion 139 causes pitch bearing 136 to rotate, which in turn rotates blade 132 about pitch axis 133 to change the pitch angle or blade of blade 132 pitch.

In some embodiments, pitch controller 134 may be a centralized pitch controller associated with multiple pitch drive systems 135 . Each pitch controller 134 simultaneously controls a plurality of pitch drive systems 135 to control the pitch angle or blade pitch of the blades 132 corresponding to the plurality of pitch drive systems 135 .

In some other embodiments, pitch controller 134 includes a distributed plurality of independent pitch controllers. Each pitch controller 134 controls a corresponding pitch drive system 135 to control the pitch angle or blade pitch of the blade 132 corresponding to the pitch drive system 135 .

In some embodiments, the nacelle 12 includes a main controller 121 . The pitch controller 134 is connected in communication with the main controller 121 . Pitch controller 134 receives one or more control signals from main controller 121 and/or sends signals to pitch controller 134 representing operating information of blades 132 to adjust the pitch angle or blade pitch of blades 132 Take control.

FIG. 3 is a schematic diagram of a pitch control process of the wind turbine 100 in FIG. 1 in one technology.

Referring to FIG. 3 , the pitch controller 134 includes a centralized pitch controller 1341 and an independent pitch controller 1342 . The centralized pitch controller 1341 obtains the theoretically required pitch angle β _c of the three blades 132 according to the current rotational speed w of the rotor 13 and the reference rotational speed w _ref obtained based on the predicted wind speed, which is used to control the aerodynamic torque captured by the fan 100 and power; the independent pitch controller 1342 obtains the pitch angles β ₁ , β ₂ , β ₃ that the three blades 132 theoretically need to reach according to the current torques M1 , M2 , M3 of the three blades 132 sensed by the sensor , which is used to control the load reduction of the fan 100 . The pitch controller 134 controls the pitch of the blades 132 by controlling the pitch drive system 135 according to the pitch angles β _{c ,} β ₁ , β ₂ , and β ₃ . When controlling the aerodynamic torque and power captured by the fan 100, The load reduction control is performed on the fan 100 .

In FIG. 3 , the pitch control principle of the wind turbine 100 is as follows: the pitch controller 134 controls the pitch drive system 135 according to the pitch angle that each blade 132 theoretically needs to achieve, so that the blades 132 are driven at a constant pitch angle. The speed is rotated to the corresponding angular position. However, with the development of fan technology, the single-machine capacity of the fan 100 increases, the blades 132 are lengthened, the volume of the fan 100 increases, and the inertia of the blades 132 during rotation increases. Based on the pitch control technology with constant pitch angle and rotational speed, there is a problem that the control effect of load reduction decreases. Specifically, on the one hand, the pitch angle of the blades 132 may not be accurately controlled. For example, if the pitch angle rotation speed of the blade 132 is set too large, after the blade 132 rotates to the theoretically required pitch angle position, it is likely that the inertia cannot be braked in time due to the excessive inertia, thereby causing the actual rotation of the blade 132 There is a deviation between the reached pitch angle position and the theoretically required pitch angle position, so that the effect of the load reduction control of the wind turbine 100 is reduced; on the other hand, the timeliness of the load reduction control is reduced. For example, when the pitch angle rotation speed of the blades 132 is small, the torque driving the blades 132 to rotate is small, and the time for the blades 132 to change from the stopped state to the rotating state is long, which reduces the timeliness of load reduction control and reduces the effect of load reduction control. It can be seen that, in order to improve the load reduction control effect of the wind turbine 100, it is necessary to perform pitch control on the wind turbine 100 based on a pitch control strategy with variable pitch angle and rotational speed. For example, in the early stage of the pitch control, the blades 132 are controlled at a relatively large pitch angular rotation speed to improve the timeliness of the load reduction control; in the later stage of the pitch control, the blades 132 are controlled at a relatively small pitch angular rotation speed. Pitching to reduce the pitch angle deviation caused by inertia.

Further, the inventor of the present application found that under the same working condition (meaning that the operating environment of the fan 100 is the same, for example, the wind speed and wind direction at the location are the same), different settings of the pitch angle and rotation speed will affect the blades of the blades 132. Root moment magnitude. Referring to Figures 4a and 4b of the dynamic characteristics of the wind turbine under the same operating conditions, Figure 4a shows the difference of the pitch angle and blade root moment of the wind turbine 100 when the rotational speed of the blade 132 is set to [-3.5, 3.5] deg/s A time series comparison chart. Figure 4b is a time series comparison diagram of the pitch angle and blade root moment of the wind turbine 100 when the rotational speed of the blade 132 is set to [-5,5]deg/s. In FIGS. 4a and 4b, the curves 132_1A, 132_1A, and 132_1A respectively represent the variation characteristics of the pitch angles of the three blades 132 over time, and the pitch angle rotation speeds in different time periods can be determined based on the curves 132_1A, 132_1A, and 132_1A. The curves 132_1B, 132_1B, and 132_1B correspond to the curves 132_1A, 132_1A, and 132_1A one-to-one, and respectively represent the variation characteristics of the blade root moment of the three blades 132 with time. It can be seen from FIG. 4a and FIG. 4b that under different pitch angle rotation speed settings, the position of the pitch angle of the wind turbine 100 is affected, which further causes the blade root moments of the three blades 132 to be different. Different blade root moments will cause the impeller surface load moment of the fan 100 to be unbalanced, and the unbalanced impeller surface load moment will aggravate the vibration of the fan 100 and increase the structural load of the fan 100; on the contrary, if the impeller surface load moment is balanced, The vibration of the fan 100 is small, and the structural load of the fan 100 is small.

To sum up, when pitch control is performed on the wind turbine 100 based on the pitch control strategy with variable pitch angle speed, the influence of the pitch angle speed on the structural load needs to be considered. In order to achieve a better control effect of load reduction, it is necessary to select the pitch angle speed corresponding to the minimum structural load. Based on the above description, the present application proposes a pitch control method, the basic idea of which is: within a control period of pitch control, predict a future period of time after the control period (also referred to as the prediction period corresponding to the control period) Based on the predicted wind speed information, the pitch angle rotation speed in the prediction period is determined, so as to control the pitch of the wind turbine in the prediction period according to the predicted pitch angle rotation speed. Wherein, the predicted pitch angular rotation speed satisfies the following conditions: the structural load of the wind turbine 100 in the predicted period can be minimized.

FIG. 5 is a flowchart of a pitch control method provided by an embodiment of the present application. The pitch control method can be applied to the wind turbine 100 in FIG. 1 , for example, to the main controller 121 of the wind turbine 100 . The pitch control method includes steps S51 to S53.

Step S51, determine the objective function and the data-driven model, wherein the independent variable of the objective function includes the load reduction control objective, the dependent variable of the objective function is used to represent the size of the structural load of the wind turbine in the pitch control, and the data-driven model is used to determine at least The functional relationship between the partial derating control objective and the pitch angle speed.

In some embodiments, the load reduction control target refers to the influence factor of the structural load of the wind turbine 100 . Taking the force value of the front and rear directions of the top of the tower 11 of the wind turbine 100 as an example, the force value of the front and rear directions of the top of the tower 11 will affect the structural load of the wind turbine 100, and the change of the force value of the front and rear directions of the top of the tower 11 will affect the The size of the structural load of the fan 100 . By controlling the force value of the front and rear directions of the top of the tower 11 of the wind turbine 100 , the structural load of the wind turbine 100 can be controlled. Therefore, the force value of the front and rear direction of the top of the tower 11 can be called a load reduction control target. In this embodiment, the load reduction control targets include, but are not limited to, the pitch angular velocity change rate of the wind turbine 100 during the pitch control process, the force value of the front and rear directions of the top of the tower 11 , and the average value of the blade root moment of the blades 132 and each blade 132 The blade root moment difference.

The following is an example of the load reduction control target of the present application including only the pitch angular velocity change rate, the force value of the front and rear directions of the top of the tower 11, and the average value of the blade root moment of the blades 132 and the blade root moment of each blade 132. Be explained.

In some embodiments, the expression of the objective function can be expressed as expression (1):

表示三个转子叶片1、2、3在预测周期内的叶根力矩均值；M_i表示第i个转子叶片在预测周期的第z个预测步长内的叶根力矩，其中，z的取值为1、2……、n；

表示叶片132的叶根力矩均值与各叶片132的叶根力矩差异；F_fore-aft表示预测周期的第z个预测步长Δt内的塔架11顶部的前后方向受力值；

表示第i个转子叶片在预测周期的第z个预测步长Δt内的桨距角速度变化率；通过上述相关描述可知，

F_fore-aft、

分别表示降载控制目标；w₁(v_wind)、w₂(v_wind)、w₃(v_wind)分别表示对应降载控制目标的权重值。其中，

(F_fore-aft)²、

也可以为绝对值的形式，比如

用于保证降载控制目标的取值为正数。Among them, J represents the objective function value. The larger the value of J is, the greater the structural load of the wind turbine 100 in a future period of time (prediction period) after a control period; on the contrary, the greater the structural load of the wind turbine 100 in the future period of time (prediction period) after a control period Small. t ₀ represents the start time point of the forecast period, n represents the forecast step size of the forecast period; Δt represents the duration of each forecast step size;

Represents the mean value of the blade root moment of the three rotor blades 1, 2, and 3 in the prediction period; M _i represents the blade root moment of the ith rotor blade in the zth prediction step of the prediction period, where the value of z is 1, 2..., n;

Represents the difference between the mean value of the blade root moment of the blades 132 and the blade root moment of each blade 132; F _fore-aft represents the force value of the front and rear direction of the top of the tower 11 within the zth prediction step Δt of the prediction cycle;

Represents the pitch angular velocity change rate of the ith rotor blade within the zth prediction step Δt of the prediction cycle; from the above related descriptions, we can see that,

F _fore-aft ,

respectively represent the load shedding control targets; w ₁ (v _wind ), w ₂ (v _wind ), and w ₃ (v _wind ) respectively represent the weight values corresponding to the load shedding control targets. in,

(F _fore-aft ) ² ,

It can also be in the form of an absolute value, such as

The value used to ensure the load shedding control target is a positive number.

In some embodiments, considering that the PLC controller of the wind turbine does not support the integral algorithm in the execution process, the expression (1) can also be expressed as a discrete form in the expression (2):

Among them, k represents the kth control cycle, and j represents the jth prediction step size in the kth control cycle.

为例，数据驱动模型可以确定第i个叶片132的叶根力矩M_i和桨距角转速的函数关系，基于该函数关系，进而可以确定降载控制目标

与桨距角转速的函数关系；再以降载控制目标F_fore-aft为例，基于数据驱动模型，便可直接确定降载控制目标F_fore-aft与桨距角转速的函数关系。将根据数据驱动模型确定的降载控制目标与桨距角转速的函数关系代入到上述表达式(1)或(2)中，便可建立目标函数值与桨距角转速的函数关系。In some embodiments, the data-driven model may be built based on Kriging interpolation and operating data of the wind turbine 100 . The operation data of the wind turbine 100 may be the simulation data of the wind turbine 100, including the pitch angle rotational speed, wind speed, wind direction, pitch angle, blade root moment of the blade 132, and the tower 11 at multiple sampling time points of the wind turbine 100 under different operating conditions. Data such as the force value in the front and rear directions of the top. The data-driven model takes the fan state as output and the independent variable parameters related to the fan state as input. The state of the fan includes, but is not limited to, the blade root moment _Mi of the blade 132 and the force value F _fore-aft of the top of the tower 11 in the front and rear directions. The independent variable parameters related to the fan state include pitch angle speed. Different fan states correspond to different data-driven models. control target with load shedding

For example, the data-driven model can determine the functional relationship between the blade root moment M _i of the ith blade 132 and the pitch angle rotational speed, and based on the functional relationship, the load reduction control target can be determined.

The functional relationship with the pitch angle speed; taking the load reduction control target F _fore-aft as an example, based on the data-driven model, the functional relationship between the load reduction control target F _fore-aft and the pitch angle speed can be directly determined. Substituting the functional relationship between the load reduction control target and the pitch angle speed determined according to the data-driven model into the above expression (1) or (2), the functional relationship between the objective function value and the pitch angle speed can be established.

In some embodiments, the data-driven model established based on the kriging interpolation method can be expressed as expression (3):

According to the principle of kriging method, the meaning of each parameter in expression (3) is as follows:

表示对Y(x)全局近似的回归模型；f_p(x)为回归模型的基函数；c＝[c₁,c₂,…,c_p]^T，为回归模型的回归系数向量；m表示与风机状态相关的自变量参数的个数；Z(x)表示回归模型

近似Y(x)的系统偏差函数，其均值为0，其协方差矩阵可以表达为表达式(4)：Y(x) represents the fan state; x represents the independent variable parameter vector related to the fan state;

Represents a regression model that approximates Y(x) globally; f _p (x) is the basis function of the regression model; c=[c ₁ , c ₂ ,..., c _p ] ^T , is the regression coefficient vector of the regression model; m represents The number of independent variable parameters related to the fan state; Z(x) represents the regression model

Approximate the systematic deviation function of Y(x), its mean is 0, and its covariance matrix can be expressed as expression (4):

表示Z(x)的方差，

是以θ为超参数的相关函数，表示试验点x和x′之间的空间相关关系。在本实施例中，R(x_p,x_p′,θ_p)为一维相关函数，具体采用高斯函数，表示为表达式(5)：in,

represents the variance of Z(x),

is a correlation function with θ as a hyperparameter, which represents the spatial correlation between the test points x and x′. In this embodiment, R(x _p , x _p ′, θ _p ) is a one-dimensional correlation function, specifically a Gaussian function, which is expressed as Expression (5):

It can be understood that, for different fan states, x and Y(x) in expression (3) are different, that is, the data-driven models corresponding to different fan states are different. For example, when the fan state is the blade root moment of the blade 132, Y(x) represents the value of the blade root moment, x represents the independent variable parameter vector related to the blade root moment, and m represents the number of independent variable parameters related to the blade root moment. For another example, when the fan state is the force value in the front and rear directions of the top of the tower 11, Y(x) represents the value of the force value in the front and rear directions of the tower 11, and x represents the self-relevant force value in the front and rear directions of the top of the tower 11. Variable parameter vector, m represents the number of independent variable parameters related to the force value of the top of the tower 11 in the front and rear directions. By performing regression analysis on the operation data of the fan 100, the regression coefficient vector c of the data-driven model corresponding to each fan state can be determined respectively, and then the data-driven model corresponding to each fan state can be determined. Here, the data-driven model is described by taking the fan state as the blade root moment of the blade 132 and the force value of the front and rear directions of the top of the tower 11 as examples.

其中，V_wind表示风机100所在环境的风速；θ_wind表示风机100所在环境的风向；σ_wind表示风剪切系数；θ_azimuth表示叶片132在转动过程中的方位角，用于描述塔影效应；β_i表示第i个叶片132的桨距角；

表示第i个叶片132的桨距角转速；ω_r表示第i个叶片132的转子13的转速。第i个叶片132的叶根力矩可以表达为表达式(6)：In some embodiments, when the fan state is the blade root moment of the blade 132, the independent variable parameter vector related to the blade root moment

Wherein, V _wind represents the wind speed of the environment where the fan 100 is located; _{θ wind} represents the wind direction of the environment where the fan 100 is located; σ _wind represents the wind shear coefficient; β _i represents the pitch angle of the ith blade 132;

represents the pitch angle rotation speed of the ith blade 132 ; ω _r represents the rotation speed of the rotor 13 of the ith blade 132 . The blade root moment of the i-th blade 132 can be expressed as Expression (6):

表示与第i个叶片132的叶根力矩对应的自变量参数。In the expression (6), M _i represents the blade root moment of the i-th blade 132 of the fan 100,

represents the independent variable parameter corresponding to the blade root moment of the i-th blade 132 .

其中，V_wind表示风速；θ_wind表示风向；σ_wind表示风剪切系数；θ_azimuth表示叶片132在转动过程中的方位角，用于描述塔影效应；β₁,β₂,β₃分别表示三个叶片132的桨距角；

分别表示三个叶片132的桨距角转速。塔架11顶部的前后方向受力值可以表达为表达式(7)：In the same way, when the fan state is the force value of the front and rear directions of the top of the tower 11, the independent variable parameter vector related to the force value of the front and rear directions of the top of the tower 11

Wherein, V _wind represents the wind speed; θ _wind represents the wind direction; σ _wind represents the wind shear coefficient; θ _azimuth represents the _azimuth angle _of the blade 132 during the rotation process, which is used to describe the tower shadow effect _; the pitch angles of the three blades 132;

respectively represent the pitch angle rotational speeds of the three blades 132 . The force value of the front and rear direction of the top of the tower 11 can be expressed as Expression (7):

In the expression (7), F _fore-aft represents the force value in the front-rear direction of the top of the tower 11 .

来表示，以建立降载控制目标和桨距角转速

的函数关系。以桨距角β和转子13的转速ω_r为例：In expressions (6) and (7), the wind speed V _wind , the wind direction θ _wind , and the wind shear coefficient σ _wind represent the predicted incoming wind information, which are predicted by lidar or the like in any control period of the pitch control. Wind speed V _wind , wind direction θ _wind , and wind shear coefficient σ _wind in a future period of time (ie, forecast period). The predicted incoming wind information can be directly substituted into the above expression (6) or (7), and at the same time, the other parameters are passed through the pitch angle speed.

to represent to establish the derating control target and the pitch angle speed

functional relationship. Take the pitch angle β and the rotational speed ω _r of the rotor 13 as an example:

来表示桨距角β的表达式可以表达为表达式(8)：In one embodiment, the rotational speed through the pitch angle

The expression to express the pitch angle β can be expressed as expression (8):

where t represents time.

来表示转子13的转速ω_r的表达式可以基于如下方法确定：In one embodiment, the rotational speed through the pitch angle

The expression to express the rotational speed ω _r of the rotor 13 can be determined based on the following method:

According to the relevant knowledge of the fan, the power calculation formula of the fan 100 can be expressed as expression (9):

Further, λ can be expressed as expression (10):

Among them, λ represents the reference tip speed ratio; ρ is the air density; R is the radius of the rotor 13 , and C _P (λ, β) is the wind energy capture coefficient. Further, C _P (λ, β) can be expressed as expression (11):

Further, λ _i can be determined based on expression (12):

In some embodiments, a ₁ in Expression (11) and Expression (12) is 0.5176, a ₂ is 116, a ₃ is 0.4, a ₄ is 5, a ₅ is 21, a ₆ is 0.0068, and b ₁ is 0.08 and _b2 is 0.035.

来表示的转子13转速ω_r的表达式。According to Expressions (9) to (12), it is possible to determine the rotational speed by the pitch angle

to express the expression of the rotor 13 rotational speed ω _r .

转子13的转速ω_r、以及风机100的有效功率输出的约束条件如下：In some embodiments, since the operation data of the wind turbine 100 is obtained by simulation, a wind turbine model needs to be established for the wind turbine 100, and the value range of each parameter in the data-driven model is constrained based on the wind turbine model. In one of the embodiments, the pitch angle speed

The constraints on the rotational speed ω _r of the rotor 13 and the effective power output of the fan 100 are as follows:

ω _{r_cut} ≤ω _r ≤ω _{r_rate}

95%·P _ref ≤P _g ≤105%·P _ref

表示风机100的最小桨距角转速；

表示风机100的最大桨距角转速；in,

Indicates the minimum pitch angle speed of the wind turbine 100;

Indicates the maximum pitch angle speed of the fan 100;

ω _{r_cut} represents the cut-in rotational speed of the rotor 13 ; ω _{r_rate} represents the rated rotational speed of the rotor 13 ;

P _ref represents the effective power reference value of the fan 100 .

之差，因此，针对降载控制目标

可以无需建立对应的数据驱动模型来确定降载控制目标

和桨距角转速的函数关系。It should be noted that in expressions (1) and (2), since the rate of change of the pitch angular velocity can be directly expressed as the pitch angular rotation speed of the two prediction steps before and after

The difference, therefore, for the shedding control objective

It is possible to determine load shedding control targets without establishing a corresponding data-driven model

as a function of the pitch angle speed.

的权重w₃(v_wind)可能为0.2，但在风速18m/s时，预测周期内的桨距角速度变化率

的权重w₃(v_wind)可能需要调整到0.5。另一方面，在不同的风况下，每个控制周期获取的预测来流风信息可能不同，为使目标函数和数据驱动模型具有自适应性，提高预测准确性，在本申请一些实施例中，对于任一控制周期，根据该控制周期对应的预测来流风信息，确定对应该控制周期的目标函数中的多个降载控制目标分别对应的权重值，同时，数据驱动模型用于根据多个控制周期中每个控制周期对应的预测来流风信息，分别确定相应控制周期的至少部分降载控制目标与桨距角转速的函数关系。In some embodiments, considering that the wind condition of the environment where the wind turbine 100 is located is in a process of constant change, under different operating conditions, on the one hand, the load reduction control targets in the above expressions (1) and (2) weights may vary. For example, when the wind speed is 10m/s, the rate of change of the pitch angular velocity in the prediction period

The weight w ₃ (v _wind ) of the

The weight w ₃ (v _wind ) may need to be adjusted to 0.5. On the other hand, under different wind conditions, the predicted incoming wind information obtained in each control cycle may be different. In order to make the objective function and the data-driven model adaptive and improve the prediction accuracy, in some embodiments of the present application, For any control period, according to the predicted flow wind information corresponding to the control period, the weight values corresponding to the multiple load reduction control targets in the objective function corresponding to the control period are determined. At the same time, the data-driven model is used to control the The predicted incoming wind information corresponding to each control period in the period respectively determines the functional relationship between at least part of the load reduction control target of the corresponding control period and the pitch angle rotation speed.

Further, the predicted incoming wind information for each control period may include the predicted incoming wind information within a plurality of prediction steps after the corresponding control period. For any control period, the relationship between at least part of the derating control target and the pitch angle speed corresponding to the predicted step of the control period can be determined according to the predicted flow wind information in any predicted step after the control period. function relationship, and the weight value of each load-shedding control objective of the objective function corresponding to the prediction step size of the control period. Based on this, the above expression (6) can be expressed as expression (15):

Wherein, M _i (k+j) represents the blade root moment of the i-th blade 132 at the j-th prediction step of the k-th control cycle.

Similarly, the above expression (7) can be expressed as the expression (16):

表示塔架11在第k个控制周期的第j个预测步长的前后受力值。in,

Represents the force value of the tower 11 before and after the jth prediction step of the kth control cycle.

Continuing to refer to FIG. 5 , after determining the objective function and the data-driven model, step S52 may be performed.

Step S52, according to the objective function and the data-driven model, determine the target pitch angle rotation speed corresponding to the minimum value of the objective function.

It can be seen from the above related descriptions that the data-driven model is mainly used to determine the functional relationship between at least part of the load reduction control target and the pitch angle speed of each control cycle based on the prediction of each control cycle to flow wind information. The functional relationship between at least part of the load reduction control target determined in each control cycle and the pitch angle rotational speed is determined, and the corresponding target pitch angle rotational speed is respectively determined. Referring to FIG. 6 in conjunction, FIG. 6 is a sub-flow chart of step S52 in FIG. 5 , including step S61 and step S62.

Step S61, for any control period, determine the objective function to be solved according to the objective function and the functional relationship between at least part of the load reduction control target and the pitch angle rotational speed determined by the control period, wherein the independent variable of the objective function to be solved includes the propeller. The pitch angle speed, the dependent variable of the objective function to be solved is used to represent the structural load of the wind turbine in the pitch control.

In some embodiments, according to the above related descriptions, the functional relationship between at least part of the load reduction control target determined in the control period and the pitch angle speed can be substituted into the objective function to obtain the objective function to be solved for the control period.

Further, for a control cycle including a plurality of prediction steps, the target function can be determined according to the functional relationship between the objective function and any prediction step after the control cycle corresponding to at least part of the derating control target and the pitch angle speed. The objective function to be solved corresponding to the prediction step size of the control period. That is, the functional relationship between at least part of the load reduction control target corresponding to each prediction step and the pitch angular rotation speed is substituted into the objective function, so as to obtain the objective function to be solved corresponding to the prediction step.

Step S62, based on the intelligent optimization solution algorithm, the objective function to be solved is solved to determine the corresponding target pitch angle rotation speed.

For a control cycle including multiple prediction steps, the objective function to be solved corresponding to each prediction step of the control cycle can be solved based on an intelligent optimization solution algorithm to determine the corresponding target pitch angle speed. In some embodiments, M particles may be searched in space to determine a target pitch angle rotation speed sequence corresponding to a plurality of predicted step sizes for the control period. Specifically, it is possible to iterate N times in the control cycle, and for each iteration, for all prediction steps corresponding to the control cycle, that is, each iteration, determine a pitch angle rotation speed corresponding to each prediction step in the control cycle. . For example, if one control cycle corresponds to 10 prediction steps, and the fan 100 includes 3 blades 132, 10×3 pitch angular rotation speeds can be determined for each iteration, and the 10×3 pitch angular rotation speeds are respectively related to the 10 The prediction step corresponds to the optimal point that minimizes the value of the objective function shown in Expression (2) in the current iteration.

It is assumed here that in the i-th iteration of the k-th control cycle, the pitch angular rotation speeds of the three blades 132 corresponding to the j-th prediction step represented by the l-th particle can be expressed as Expression (17):

Wherein, l=1,2,...,M, i=1,2,...,N, i_1, i_2, i_3 are used to distinguish the three blades 132 .

Then in the i-th iteration of the k-th control cycle, the pitch angular rotation speeds of the three blades 132 corresponding to all the prediction steps represented by the l-th particle can be expressed as Expression (18):

where n represents the number of prediction steps.

When looking for the target pitch angle rotation speed corresponding to the minimum value of the objective function, the lth particle can define the trajectory in the parameter space according to expressions (19) and (20):

表示第i+1次迭代桨距角转速

相对于第i次迭代的桨距角转速

的变化率；

Represents the i+1th iteration pitch angle speed

The pitch angle speed relative to the ith iteration

rate of change;

表示第i+1次的桨距角转速；

Indicates the i+1th pitch angle speed;

为第i次迭代中使表达式(2)所示的目标函数取值最小时所对应的局部最优点；

is the local optimum point corresponding to the minimum value of the objective function shown in expression (2) in the ith iteration;

为所有1至i次迭代中使表达式(2)所示的目标函数取值最小时所对应的全局最优点。

is the global optimal point corresponding to the minimum value of the objective function shown in expression (2) in all 1 to i iterations.

c ₁ and c ₂ are learning factors;

r ₁ and r ₂ are two random numbers between uniform distribution and [0,1];

的速度变化率的继承。ω is the inertia weight, representing the pair

The rate of change of velocity is inherited.

则

可以表达为表达式(20)：Based on the above calculation, the pitch angle rotation speed of each prediction step corresponding to the minimum value of the objective function in the objective function can be finally determined after N iterations in the kth control cycle, and the pitch angle The rotational speed is taken as the target pitch angle rotational speed. Here, it is assumed that the target pitch angle speed is

but

can be expressed as expression (20):

表示第k个控制周期第1个预测步长对应的桨距角转速；in,

represents the pitch angle speed corresponding to the first prediction step of the kth control cycle;

表示第k个控制周期第2个预测步长对应的桨距角转速；akin,

represents the pitch angle speed corresponding to the second prediction step of the kth control cycle;

表示第k个控制周期第n个预测步长对应的桨距角转速。akin,

Indicates the pitch angle speed corresponding to the nth prediction step of the kth control cycle.

Based on the particle swarm algorithm to solve the objective function, the solution speed is fast and the convergence is better.

Continue to refer to FIG. 5 , after determining the target pitch angle rotation speed, step S53 may be continued.

Step S53, performing pitch control on the wind turbine 100 according to the target pitch angle rotation speed.

在该控制周期之后下一个控制周期内对风机100进行变桨控制。从而使得在下一个控制周期对风机100进行变桨控制时，桨距角转速使结构载荷最小。In some embodiments, for any control period, according to the target pitch angle speed determined in the control period

The pitch control of the wind turbine 100 is performed in the next control period after the control period. Therefore, when the pitch control is performed on the wind turbine 100 in the next control cycle, the rotational speed of the pitch angle minimizes the structural load.

Further, in some embodiments, the relationship between the control period, the prediction period and the prediction step size may be as follows: for example, assuming that every second is a control period, the prediction period corresponding to each control period is a period of 10 seconds after the corresponding control period. , in the prediction period, every 1 second is a prediction step.

Based on the above assumptions, it can be clearly seen that the first control period (for example: the first second) will determine the target pitch angle of each prediction step in the first prediction period (for example: the second to the 11th second). speed, the target pitch angle speed at the 2nd, 3rd, ... and 11th seconds will be determined respectively;

Similarly, in the second control cycle (second second), the target pitch angle speed of each prediction step in the second prediction cycle (second to 12th second) will be determined, that is, the third second, The target pitch angle rotation speed at the 4th second, ..., the 12th second.

在该控制周期之后下一个控制周期内对风机100进行变桨控制。如此，可以在变桨过程中，进一步降低风机100的结构载荷。According to the above example, it can be seen that in the first control period (the 1st second), it is only necessary to output the target pitch angle speed of the first prediction step (the 2nd second) in the first prediction period, which is used in the first prediction period. Two control cycles (the second second) to control the pitch of the wind turbine 100; when the wind turbine 100 runs to the third second, at this time, the second control cycle has output based on the latest predicted wind flow information for the third second. For the target pitch angle rotation speed, pitch control of the wind turbine 100 may be performed based on the target pitch angle rotation speed in the first prediction period (3rd second) output in the second control period (2nd second). That is, for any control period, the target pitch angle rotation speed corresponding to the first prediction step size of the control period can be used, that is, the target pitch angle rotation speed in the first column of Expression 20

The pitch control of the wind turbine 100 is performed in the next control period after the control period. In this way, the structural load of the wind turbine 100 can be further reduced during the pitching process.

Based on the above description, in some embodiments of the present application, the pitch control method determines an objective function and a data-driven model, and according to the objective function and the data-driven model, determines the target pitch angle rotation speed corresponding to the minimum value of the objective function, and then based on the target The pitch angle rotational speed controls the pitch of the wind turbine 100 . By controlling the pitch angle speed, the wind turbine 100 runs at the pitch angle speed when the structural load is minimized during the pitch process, avoiding the problem of decreasing the effectiveness of the pitch control strategy caused by the constant pitch angle speed. The load reduction effect of the fan 100 during the pitch pitch process is improved.

FIG. 7 is a module block diagram of a pitch control system 800 provided by an embodiment of the present application.

The pitch control system 800 includes one or more processors 801 for implementing the pitch control method described above. In some embodiments, the pitch control system 800 may include a readable storage medium 809, which may store programs that can be called by the processor 801, and may include a non-volatile storage medium.

In some embodiments, pitch control system 800 may include memory 808 and interface 807 .

In some embodiments, the pitch control system 800 may also include other hardware according to practical applications.

The readable storage medium 809 in this embodiment of the present application stores a program thereon, and when the program is executed by the processor 801 , is used to implement the pitch control method described above.

The present application may take the form of a computer program product embodied on one or more readable storage media 809 (including but not limited to disk storage, CD-ROM, optical storage, etc.) having program code embodied therein. Readable storage media 809 includes persistent and non-permanent, removable and non-removable media, and storage of information may be implemented by any method or technology. Information may be computer readable instructions, data structures, modules of programs, or other data. Examples of readable storage media 809 include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), Flash Memory or other memory technology, Compact Disc Read Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage , magnetic tape cartridges, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

The above descriptions are only preferred embodiments of this specification, and are not intended to limit this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this specification shall be included in this specification. within the scope of protection.

The above descriptions are only preferred embodiments of this specification, and are not intended to limit this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of this specification shall be included in this specification. within the scope of protection.

It should also be noted that the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device comprising a series of elements includes not only those elements, but also Other elements not expressly listed, or which are inherent to such a process, method, article of manufacture, or apparatus are also included. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, article of manufacture, or device that includes the element.

Claims (11)

1. a pitch control method, is characterized in that, described pitch control method comprises: Determine the objective function, wherein the independent variable of the objective function includes a load reduction control target, the load reduction control target is the influence factor of the structural load of the wind turbine in the pitch control, and the dependent variable of the objective function is used to represent the wind turbine Structural load magnitude in pitch control; According to the operating data of the wind turbine, a data-driven model is determined, wherein the data-driven model takes the state of the wind turbine as an output, and uses an independent variable parameter related to the state of the wind turbine as an input, and is used to determine at least part of the load reduction control target and the Function relationship of pitch angle speed; According to the objective function and the data-driven model, determine the target pitch angle rotation speed corresponding to the minimum value of the objective function; According to the target pitch angle rotation speed, pitch control is performed on the wind turbine. 2 . The pitch control method according to claim 1 , wherein the data-driven model is used to determine the corresponding control according to the predicted flow wind information corresponding to each of the multiple control cycles. 3 . a function of the derating control target and the pitch angle rotational speed for at least part of the cycle; The determining of the target pitch angle rotation speed corresponding to the minimum value of the objective function includes: According to the objective function and the functional relationship between at least part of the load reduction control target and the pitch angle rotational speed determined in each control cycle, the corresponding target pitch angle rotational speed is respectively determined; The pitch control of the fan includes: For any one of the control periods, pitch control is performed on the wind turbine in the next control period after the control period according to the target pitch angle rotational speed determined in the control period. 3. The pitch control method according to claim 2, wherein the independent variable of the objective function comprises a plurality of the load reduction control targets, and the pitch control method further comprises: For any one of the control periods, according to the predicted flow wind information corresponding to the control period, the weight values corresponding to the multiple load reduction control targets in the objective function corresponding to the control period are determined. 4 . The pitch control method according to claim 2 , wherein the functional relationship between at least part of the load reduction control target determined according to the objective function and each of the control cycles and the pitch angle rotation speed. 5 . , respectively determine the corresponding target pitch angle rotation speed, including: For any one of the control periods, the objective function to be solved is determined according to the objective function and the functional relationship between at least part of the load reduction control objective and the pitch angle rotation speed determined in the control period, wherein the objective function to be solved is The independent variable includes the pitch angular rotation speed, and the dependent variable of the objective function to be solved is used to represent the structural load of the wind turbine in the pitch control; Based on an intelligent optimization solution algorithm, the objective function to be solved is solved to determine the corresponding target pitch angle rotation speed. 5 . The pitch control method according to claim 4 , wherein the intelligent optimization solution algorithm comprises a particle swarm algorithm. 6 . 6. The pitch control method according to claim 4, wherein the predicted incoming wind information of each control period comprises the predicted incoming wind information within a plurality of prediction steps after the corresponding control period; For any one of the control periods, the determining the functional relationship between at least part of the load-shedding control target and the pitch angle rotational speed corresponding to the control period includes: According to the predicted incoming wind information in any of the predicted steps after the control period, determine the functional relationship between at least part of the load reduction control target and the pitch angle rotation speed corresponding to the predicted step in the control period; For any of the control periods, determining the objective function to be solved according to the objective function and the functional relationship between at least part of the load reduction control objective and the pitch angle rotational speed determined in the control period, including: According to the objective function and the functional relationship between at least part of the load reduction control target corresponding to any predicted step after the control cycle and the pitch angle rotation speed, determine the said control cycle corresponding to the predicted step. The objective function to be solved; For any one of the control periods, the solving of the objective function to be solved to determine the corresponding target pitch angle rotation speed includes: Based on an intelligent optimization solution algorithm, the objective function to be solved corresponding to each predicted step size of the control period is solved to determine the corresponding target pitch angle rotation speed. 7 . The pitch control method according to claim 6 , wherein, for any one of the control periods, according to the target pitch angle rotation speed determined in the control period, after the control period The pitch control is performed on the wind turbine in one control cycle, including: According to the target pitch angle rotation speed corresponding to the first predicted step size in the control period, pitch control is performed on the wind turbine in the next control period after the control period. 8 . The pitch control method according to claim 1 , wherein the load reduction control target comprises a pitch angular velocity change rate of the wind turbine during the pitch control process. 9 . 9 . The pitch control method of claim 1 , wherein the data-driven model is established based on a kriging interpolation method and operating data of the wind turbine. 10 . 10. A pitch control system, comprising one or more processors for implementing the pitch control method according to any one of claims 1-9. 11. A readable storage medium, wherein a program is stored thereon, and when the program is executed by a processor, the pitch control method according to any one of claims 1-9 is implemented.

Images