Fan decoupling control method for coordinating active power tracking and fatigue load optimization

Abstract

The invention relates to the technical field of wind fields, and discloses a fan decoupling control method for coordinating active power tracking and fatigue load optimization so as to achieve the maximum power control requirement, meet a wind field dispatching instruction, achieve power tracking and restrain fatigue loads. According to the method, on the basis of a model prediction control framework, through decoupling control over the rotating speed and the pitch angle of a wind driven generator rotor, multi-target optimization of wind field reference power tracking and fatigue load suppression of components such as fan blades, a gearbox and a low-speed shaft is achieved.

Description

Technical Field

[0001] The invention relates to the technical field of wind farms, and in particular to a wind turbine decoupling control method for coordinating active power tracking and fatigue load optimization. Background Art

[0002] Wind power generation has become one of the fastest growing renewable energy sources, but the grid connection of a large number of wind farms has also affected the safety and stability of the power grid. Many countries with rapid development of wind power have emphasized that wind farms should improve the friendliness of the power grid and have the ability to control active power. Wind farms need to have reference power tracking capabilities and effective active power control capabilities, which requires wind farms to distribute power instructions to wind turbines in a reasonable manner. The research focus in this field is to improve the economic efficiency of wind farm operation while completing power distribution. Since wind power has no fuel cost, reducing the operating loss of the unit naturally becomes the focus of optimization. However, the previous control strategy still uses multiple controllers for separate control, and each controller only controls one parameter. Since the control strategy is relatively simple, the degree of freedom of control is sacrificed, and there is still a lot of room for optimization for fatigue loads. Therefore, there is an urgent need for a method to control the effective power and reduce the fatigue load of the entire wind farm while supporting the grid frequency. Summary of the invention

[0003] In view of the above problems, the present invention proposes a wind turbine decoupling control method that coordinates active power tracking and fatigue load optimization to achieve maximum power control requirements, meet wind farm scheduling instructions to achieve power tracking, and suppress fatigue loads. This method is based on a model predictive control framework and achieves wind farm reference power tracking and multi-objective optimization of fatigue load suppression of wind turbine blades, gearboxes, low-speed shafts and other components through decoupling control of wind turbine rotor speed and pitch angle.

[0004] The technical solution adopted by the present invention to solve its technical problem is:

[0005] The wind turbine decoupling control method for coordinating active power tracking and fatigue load optimization includes the following steps:

[0006] Step 1: Establish a nonlinear model of the wind turbine generator, including an aerodynamic model, a mechanical transmission system analysis model, a generator torque control model, a pitch angle servo system model, and a tower model;

[0007] Step 2: Establish a decoupled control framework for wind turbine rotor speed and pitch angle to achieve reference power tracking and reduce wind turbine fatigue;

[0008] Step 3: Linearize the nonlinear model of the wind turbine; obtain the linear description of the relevant model by constructing the linear increment of the nonlinear model;

[0009] Step 4: Design of active power controller;

[0010] Step 5: Feedback correction.

[0011] Furthermore, the step 1 is specifically as follows:

[0012] 1.1) Aerodynamic model

[0013] The following mathematical model is used to describe its aerodynamic characteristics:

[0014] P m =0.5πρR 2 C p (λ,β)v 3 #(1)

[0015] F t =0.5πρR 2 C t (λ,β)v 2 #(2)

[0016] P m and F t are the mechanical power and thrust of the wind wheel respectively, ρ is the air density, R is the radius of the wind wheel blade, λ is the tip speed ratio, C p is the wind turbine mechanical power coefficient, C t is the wind rotor thrust coefficient, v is the wind speed, and β is the wind turbine pitch angle;

[0017] 1.2) Mechanical transmission system analysis model

[0018] The mechanical transmission system of a wind turbine includes a wind wheel, a low-speed shaft, a gearbox, a high-speed shaft and a generator. The three-mass spring-damper system model is used to describe the mechanical transmission system:

[0019] Dynamic analysis and calculation of wind wheel:

[0020]

[0021] T s =T K1 +T D1 #(4)

[0022]

[0023]

[0024]

[0025] J r is the moment of inertia of the wind wheel, T m is the mechanical torque of the wind wheel, T s is the low speed shaft torque, K rsis the spring constant between the wind wheel and the low-speed shaft, D rs is the damping coefficient between the wind wheel and the low-speed shaft, ω r is the wind wheel speed, ω s is the low speed shaft speed, is the difference between the wind wheel speed and the low-speed shaft speed; The difference between the rotation angles of the main shaft and the low-speed shaft of the wind wheel

[0026] The dynamic modeling of the low-speed shaft is as follows:

[0027]

[0028]

[0029]

[0030]

[0031] J s is the moment of inertia of the low-speed shaft, K sg is the spring constant between the low-speed shaft and the generator, D sg is the damping coefficient between the low-speed shaft and the generator, ω g is the generator speed, N gear is the gearbox transmission ratio, is the first-order derivative of the low-speed shaft speed with respect to time, is the difference between the low speed shaft speed and the generator speed, is the difference in rotational angle between the low-speed shaft and the generator shaft.

[0032] Dynamic analysis of generator:

[0033]

[0034] J g is the moment of inertia of the generator, T e is the generator torque, is the first-order derivative of the generator speed with respect to time;

[0035] 1.3) Generator torque control model

[0036] The generator torque control model adopts the first-order inertia model, and its expression is:

[0037]

[0038] P e =μω g T e #(14)

[0039] τ e is the inertia time of the generator, Te is the torque setting value of the generator, μ is the generator efficiency, is the torque reference value of the generator, The first derivative of the generator torque setting with respect to time;

[0040] 1.4) Pitch angle servo system model

[0041] The pitch angle servo system is simplified to a first-order inertial model as shown below:

[0042]

[0043] τ β is the inertia time of the pitch angle servo system, β is the pitch angle setting value, β ref is the pitch angle reference value;

[0044] 1.5) Tower model

[0045] M t =hF t #(16)

[0046] M t is the tower bending moment, h is the tower height, F t is the tower thrust.

[0047] Furthermore, the step 2 is specifically as follows:

[0048] 2.1) Power reference tracking

[0049] When the wind turbine is in the maximum power point tracking mode, the generator rotor speed and pitch angle are controlled to make the wind turbine operate at the optimal power coefficient; when the wind turbine needs to operate in the power reference tracking state, the wind rotor mechanical power coefficient C is adjusted. p It can be deviated from the optimal value; C in the power reference tracking state p Calculated by the following formula:

[0050]

[0051] is the reference power coefficient under power reference tracking, is the optimal power factor in the maximum power point tracking mode, is the reference power value, is the maximum power at the current wind speed;

[0052] 2.2) Decoupling control strategy of wind turbine rotor speed and pitch angle

[0053] Set the generator power reference value The wind speed v is input into the rotor speed and pitch angle control system, and the generator speed reference value is output. and pitch angle reference value β ref , and then the generator speed reference value Pitch angle reference value β ref The inputs are respectively into the reference torque control system and the pitch angle servo system, and finally the generator torque value and the pitch angle are output; the linear modeling and controller based on this strategy can realize power reference tracking and reduce the fatigue of the wind turbine.

[0054] in, is the first-order inertial system of the generator; is the first-order inertial system of pitch angle; Denotes the ratio divided by the gearbox ratio.

[0055] Furthermore, the step 3 is specifically as follows:

[0056] 3.1) Small signal analysis method

[0057] In order to obtain the incremental expression of the fan control model, the small signal analysis method is used;

[0058]

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065] The analysis is as follows:

[0066] a) Obtain the steady-state operating point of the system through power flow calculation;

[0067] b) Perform a first-order Taylor expansion at the steady-state operating point to obtain its incremental expression:

[0068]

[0069] Where Δx and Δu are the increment vectors of state variables and algebraic variables respectively;

[0070]

[0071] Simplify equation (26) into the system small signal state space model at the steady-state operating point:

[0072]

[0073] Where A is the state matrix of the system model,

[0074]

[0075] 3.2) The incremental linear model of the wind turbine system is as follows:

[0076] Aerodynamic model:

[0077]

[0078] in and Respectively represent T m Yes r , partial derivatives of v and β, where Δ is the incremental expression;

[0079] Mechanical transmission system:

[0080]

[0081]

[0082]

[0083]

[0084]

[0085] Generator torque control model:

[0086]

[0087] Pitch angle servo system model:

[0088]

[0089] Tower:

[0090]

[0091] in and Respectively represent M t Yes r , partial derivatives of v and β;

[0092] 3.3) Constructing state space expressions

[0093]

[0094] in:

[0095]

[0096]

[0097] d=Δv #(41)

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105] Furthermore, the step 4 is specifically as follows: 4.1) Prediction model

[0106] Formula (38) is discretized, and the discrete state space model is as follows:

[0107]

[0108] Among them A d , B d , C d , D d 、E d and W d Respectively expressed as:

[0109]

[0110]

[0111]

[0112] C d =C #(53)

[0113] D d =D #(54)

[0114] W d =W #(55)

[0115] t s is the sampling period of the discrete system, in seconds;

[0116] 4.2) Optimization goals

[0117] The cost function of the controller is expressed as:

[0118]

[0119] in:

[0120]

[0121] N p Expressed as the control interval, Q o is the optimization weight for penalizing output deviation, Q c is the optimized weight for penalizing the input variable, y ref is the reference value of the output variable;

[0122] The constraints include power reference tracking, power output limit, mechanical transmission system torque limit, tower thrust limit, generator torque limit and pitch angle limit. The specific constraints are as follows:

[0123]

[0124] in is the constrained minimum value of the input power variation, is the maximum value of the input power variation constraint, and They are the maximum constraint values ​​of the input bending moment variation and the maximum constraint values ​​of the input spindle torque variation respectively; and Δβ min are the constrained minimum value of the output generator torque change and the constrained minimum value of the output pitch angle change, respectively. and Δβ max They are respectively the maximum value constrained by the output generator torque variation and the maximum value constrained by the output pitch angle variation;

[0125] Preferably, the control interval N p is 10.

[0126] Furthermore, the step 5 is specifically as follows:

[0127] The correction is done by tracking the power command value as follows:

[0128] ΔP fc (k+1|k)=ΔP(k+1|k)+ζ[ΔP(k)-ΔP(k|k-1)] #(57)

[0129] ζ is the correction factor, P fcTo track the power command value in feedback correction, ΔP(k+1|k) is the power change at time k predicted at time k+1, ΔP(k) is the power change at time k, and ΔP(k|k-1) is the power change at time k predicted at time k-1.

[0130] The beneficial effects of the present invention are as follows: the present invention realizes wind farm reference power tracking, multi-objective optimization of fatigue load suppression of wind turbine blades, gearboxes, low-speed shafts and other components. In addition, the prior art is a single-degree-of-freedom control strategy, and the setting values ​​of the rotor speed and the pitch angle need to be completed by two modules respectively, and the setting value of the pitch angle requires the calculation of the rotor speed, while the present invention is a two-degree-of-freedom control strategy, and the active power controller generates reference values ​​of the rotor speed and the pitch angle at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0131] Figure 1 A simplified diagram of a three-mass spring damping system of the present invention;

[0132] Figure 2 is a block diagram of a wind turbine generator of the present invention;

[0133] Figure 3 This is a decoupling control block diagram of the wind turbine rotor speed and pitch angle of the present invention. DETAILED DESCRIPTION

[0134] The present invention is further described below in conjunction with the accompanying drawings.

[0135] The wind turbine decoupling control method for coordinating active power tracking and fatigue load optimization includes the following steps:

[0136] Step 1: Build a nonlinear model of the wind turbine

[0137] (1) Aerodynamic model

[0138] The power output of a wind turbine depends on the wind energy absorbed by the blades, and the following mathematical model is used to describe its aerodynamic characteristics:

[0139] P m =0.5πρR 2 C p (λ,β)v 3 #(1)

[0140] F t =0.5πρR 2 C t (λ,β)v 2 #(2)

[0141] P m and F tare the mechanical power and thrust of the wind wheel respectively, ρ is the air density, R is the radius of the wind wheel blade, λ is the blade speed ratio, C p is the wind turbine mechanical power coefficient, C t is the wind rotor thrust coefficient, and v is the wind speed.

[0142] (2) Mechanical transmission system

[0143] The mechanical transmission system of a wind turbine generator consists of a wind rotor, a low-speed shaft, a gearbox, a high-speed shaft and a generator. The three-mass-block spring-damper system model is used to describe the coordination of the mechanical transmission: i) Since the wind rotor has a large mass, its moment of inertia is also large, so it is regarded as a mass block; ii) The low-speed shaft has a certain degree of flexibility and can be equivalent to a mass block; iii) In view of the different speeds of the two meshing gears in the gearbox, this model distributes the moment of inertia of the gearbox to the high-speed shaft and the low-speed shaft, and the moment of inertia of the high-speed shaft and the generator is combined into a mass block. Figure 1 This is a simplified diagram of a three-mass spring-damper system.

[0144] Dynamic analysis and calculation of wind wheel:

[0145]

[0146] T s =T K1 +T D1 #(4)

[0147]

[0148]

[0149]

[0150] J r is the moment of inertia of the wind wheel, T m is the mechanical torque of the wind wheel, T s is the low speed shaft torque, K rs is the spring constant between the wind wheel and the low-speed shaft, D rs is the damping coefficient between the wind wheel and the low-speed shaft. r is the wind wheel speed, ω s is the low speed shaft speed, It is the difference between the wind wheel speed and the low speed shaft speed.

[0151] Dynamic calculation of low-speed shaft:

[0152]

[0153]

[0154]

[0155]

[0156] J s is the moment of inertia of the low-speed shaft, K sg is the spring constant between the low-speed shaft and the generator, D sg is the damping coefficient between the low-speed shaft and the generator, ω g is the generator speed, N gear is the gearbox transmission ratio. is the first-order derivative of the low-speed shaft speed with respect to time, is the difference between the low speed shaft speed and the generator speed.

[0157] Dynamic analysis of generator:

[0158]

[0159] J g is the moment of inertia of the generator, T e is the generator torque, is the first-order derivative of the generator speed with respect to time.

[0160] (3) Generator torque control model

[0161] Since the generator torque control model adopts the first-order inertia model, its expression is:

[0162]

[0163] P e =μω g T e #(14)

[0164] τ e is the inertia time of the generator, T e is the torque setting value of the generator, μ is the generator efficiency, is the torque reference value of the generator, First derivative of the generator torque setpoint with respect to time.

[0165] (4) Pitch angle servo system model

[0166] The pitch angle servo system is used to adjust the pitch angle of the wind turbine to meet the needs of wind turbine operation and output power. The system can be simplified into a first-order inertial model as shown below:

[0167]

[0168] τ β is the inertia time of the pitch angle servo system, β is the pitch angle setting value, β refis the pitch angle reference value.

[0169] (5) Tower

[0170] M t =hF t #(16)

[0171] M t is the tower bending moment, and h is the tower height.

[0172] Step 2: Establish a decoupled control framework for wind turbine rotor speed and pitch angle

[0173] 2.1) Power reference tracking

[0174] The wind farm needs to have the following active power control capabilities: 1) Delta mode, which requires the required amount of power to be kept as a backup; 2) Rate constraint, the output power should not change at a rate greater than the required rate; 3) Absolute production constraint, the wind power should not be greater than the given reference power; 4) Automatic frequency control, the wind farm adjusts the active power according to the system frequency change. This invention does not involve the related issues of frequency regulation, but only considers the control of wind turbines tracking the reference power under the wind farm active power command, such as Figure 2 A block diagram of a wind turbine.

[0175] When the wind turbine is in the maximum power point tracking mode, the generator rotor speed and pitch angle are controlled to make the wind turbine operate at the optimal power coefficient. If the wind turbine needs to operate in the power reference tracking state, it only needs to adjust C p Just make it deviate from the optimal value. C in the power reference tracking state p It can be calculated by the following formula:

[0176]

[0177] is the reference power coefficient under power reference tracking, is the optimal power factor in the maximum power point tracking mode, is the reference power value, is the maximum power at the current wind speed.

[0178] 2.2) Decoupling control strategy of wind turbine rotor speed and pitch angle

[0179] It is generally believed that the low-speed shaft torque T s and tower bending moment M t The fatigue load of the wind turbine is calculated by reducing T s and M t The standard deviation of can reduce the fatigue degree of wind turbine. It can be found that the corresponding parameters are directly related to the fluctuation of wind turbine rotor speed and pitch angle. Figure 3 A wind turbine rotor speed and pitch angle decoupled control block diagram is shown, in which an active power controller simultaneously generates reference values ​​of the wind turbine rotor speed and pitch angle.

[0180] Linear modeling and controller design based on this strategy are performed and the block diagram is used to achieve reference power tracking while reducing wind turbine fatigue.

[0181] Step 3: Linearize the wind turbine model

[0182] The analytical theory of linear systems can be used to describe the characteristics of nonlinear systems and reduce the difficulty of control. Therefore, the wind turbine model needs to be linearized. The small signal method is one of the linearization methods of the wind turbine model, that is, the linear description of the relevant model is obtained by constructing a linear increment of the nonlinear model at a specific working point.

[0183] 3.1) Small signal method

[0184] The time-domain state-space model, also known as eigenvalue analysis, is a classic method for small-signal stability analysis of power systems.

[0185] in:

[0186]

[0187]

[0188]

[0189]

[0190]

[0191]

[0192]

[0193] The general steps of eigenvalue analysis are as follows:

[0194] c) Obtain the steady-state operating point of the power grid through power flow calculation;

[0195] d) Yield at the stable point:

[0196]

[0197] Where Δx and Δu are the increment vectors of state variables and algebraic variables, respectively.

[0198]

[0199] Simplify (26) into the system small signal state space model at the steady-state operating point:

[0200]

[0201] Where A is the state matrix of the system model, given by

[0202]

[0203] 3.2) The incremental linear model of the wind turbine system is shown below.

[0204] Aerodynamic model:

[0205]

[0206] in and Respectively represent T m Yes r , partial derivatives of v and β.

[0207] Mechanical transmission system:

[0208]

[0209]

[0210]

[0211]

[0212]

[0213] Generator torque control model:

[0214]

[0215] Pitch angle servo system model:

[0216]

[0217] Tower:

[0218]

[0219] in and Respectively represent M t Yes r , partial derivatives of v and β.

[0220] 3.3) Constructing state space expressions

[0221]

[0222] in:

[0223]

[0224]

[0225] d=Δv#(41)

[0226]

[0227]

[0228]

[0229]

[0230]

[0231]

[0232]

[0233] Step 4: Design of Active Power Controller

[0234] 4.1) Prediction Model

[0235] In the framework of decoupling control of wind turbine rotor speed and pitch angle, equation (39) can describe the dynamic characteristics of wind turbine. In order to make it a predictable model in model predictive control, it is necessary to discretize equation (39). The discrete state space model is:

[0236]

[0237] Among them A d , B d , C d , D d 、E d and W d Respectively expressed as:

[0238]

[0239]

[0240]

[0241] C d =C#(53)

[0242] D d =D#(54)

[0243] W d=W#(55)

[0244] 4.2) Optimization goals

[0245] The control target of the controller designed in the present invention is to reduce the fatigue load of the unit based on the reference power tracking. Therefore, the cost function of the controller can be expressed as:

[0246]

[0247] in:

[0248]

[0249] N p Expressed as the control interval, Q o is the optimization weight for penalizing output deviation, Q c is the optimized weight for penalizing the input variable, y ref is the reference value of the output variable.

[0250] From the optimization problem, it can be seen that minimizing the fatigue load means that the cost function includes T s and F t Minimize the amount of change.

[0251] The constraints include: power reference tracking, power output limit, mechanical transmission system torque limit, tower thrust limit, generator torque limit and pitch angle limit. The specific constraints are as follows:

[0252]

[0253] in is the minimum value of input power change, is the maximum value of the input power variation, and are the maximum value of input bending moment variation and the maximum value of input spindle torque variation respectively; and Δβ min are the minimum value of the output generator torque change and the minimum value of the output pitch angle change, respectively. and Δβ max They are the maximum value of the output generator torque change and the maximum value of the output pitch angle change respectively.

[0254] In addition, the control interval N p is 10.

[0255] Step 5: Feedback Correction

[0256] Due to the inevitable errors and model mismatch in the state measurement process, there is an inevitable error between the output obtained by the linearized model and the actual output. Therefore, feedback correction is essential in the control process. In feedback correction, we use the tracking of the power command value to complete the correction. The method is as follows:

[0257] ΔP fc (k+1|k)=ΔP(k+1|k)+ζ[ΔP(k)-ΔP(k|k-1)]#(57)

[0258] ζ is the correction factor, P fc To track the power command value in feedback correction, ΔP(k+1|k) is the power change at time k predicted at time k+1, ΔP(k) is the power change at time k, and ΔP(k|k-1) is the power change at time k predicted at time k-1.

[0259] The above embodiments are only preferred embodiments of the present invention and are not limitations of the technical solutions of the present invention. Any technical solution that can be implemented on the basis of the above embodiments without creative work should be deemed to fall within the scope of protection of the patent of the present invention.

Claims

1. A wind turbine decoupling control method for coordinating active power tracking and fatigue load optimization, characterized in that: The following steps are involved: Step 1: Establish a nonlinear model of the wind turbine generator, including an aerodynamic model, a mechanical transmission system analysis model, a generator torque control model, a pitch angle servo system model, and a tower model; Step 2: Establish a decoupled control framework for wind turbine rotor speed and pitch angle to achieve reference power tracking and reduce wind turbine fatigue; Step 3: Linearize the nonlinear model of the wind turbine; obtain the linear description of the relevant model by constructing the linear increment of the nonlinear model; Step 4: Design of active power controller; Step 5: Feedback correction.

2. The wind turbine decoupling control method for coordinated active power tracking and fatigue load optimization according to claim 1 is characterized in that: The step 1 is specifically as follows: 1.1) Aerodynamic model The following mathematical model is used to describe its aerodynamic characteristics: P m =0.5πρR 2 C p (l,b)v 3 #(1) F t =0.5πR 2 C t (l,b)v 2 #(2) P m and F t are the mechanical power and thrust of the wind wheel respectively, ρ is the air density, R is the radius of the wind wheel blade, λ is the tip speed ratio, C p is the wind turbine mechanical power coefficient, C t is the wind rotor thrust coefficient, v is the wind speed, and β is the wind turbine pitch angle; 1.2) Mechanical transmission system analysis model The mechanical transmission system of a wind turbine includes a wind wheel, a low-speed shaft, a gearbox, a high-speed shaft and a generator. The three-mass spring-damper system model is used to describe the mechanical transmission system: Dynamic analysis and calculation of wind wheel: T s =T K1 +T D1 #(4) J r is the moment of inertia of the wind wheel, T m is the mechanical torque of the wind wheel, T s is the low speed shaft torque, K rs is the spring constant between the wind wheel and the low-speed shaft, D rs is the damping coefficient between the wind wheel and the low-speed shaft, ω r is the wind wheel speed, ω s is the low speed shaft speed, is the difference between the wind wheel speed and the low-speed shaft speed; The difference between the rotation angles of the main shaft and the low-speed shaft of the wind wheel The dynamic modeling of the low-speed shaft is as follows: J s is the moment of inertia of the low-speed shaft, K sg is the spring constant between the low-speed shaft and the generator, D sg is the damping coefficient between the low-speed shaft and the generator, ω g is the generator speed, N gear is the gearbox transmission ratio, is the first-order derivative of the low-speed shaft speed with respect to time, is the difference between the low-speed shaft speed and the generator speed, is the difference in rotational angle between the low-speed shaft and the generator shaft. Dynamic analysis of generator: J g is the moment of inertia of the generator, T e is the generator torque, is the first-order derivative of the generator speed with respect to time; 1.3) Generator torque control model The generator torque control model adopts the first-order inertia model, and its expression is: P e =μω g T e #(14) τ e is the inertia time of the generator, T e is the torque setting value of the generator, μ is the generator efficiency, is the torque reference value of the generator, The first derivative of the generator torque setting with respect to time; 1.4) Pitch angle servo system model The pitch angle servo system is simplified to a first-order inertial model as shown below: τ β is the inertia time of the pitch angle servo system, β is the pitch angle setting value, β ref is the pitch angle reference value; 1.5) Tower model M t =hF t #(16) M t is the tower bending moment, h is the tower height, F t is the tower thrust.

3. The wind turbine decoupling control method for coordinated active power tracking and fatigue load optimization according to claim 1 is characterized in that: The step 2 is specifically as follows: 2.1) Power reference tracking When the wind turbine is in the maximum power point tracking mode, the generator rotor speed and pitch angle are controlled to make the wind turbine operate at the optimal power coefficient; when the wind turbine needs to operate in the power reference tracking state, the wind rotor mechanical power coefficient C is adjusted. p It can be deviated from the optimal value; C in the power reference tracking state p Calculated by the following formula: is the reference power coefficient under power reference tracking, is the optimal power factor in the maximum power point tracking mode, is the reference power value, is the maximum power at the current wind speed; 2.2) Decoupling control strategy of wind turbine rotor speed and pitch angle Set the generator power reference value The wind speed v is input into the rotor speed and pitch angle control system, and the generator speed reference value is output. and pitch angle reference value β ref , and then the generator speed reference value Pitch angle reference value β ref The inputs are respectively into the reference torque control system and the pitch angle servo system, and finally the generator torque value and the pitch angle are output; the linear modeling and controller based on this strategy can realize power reference tracking and reduce the fatigue of the wind turbine.

4. The wind turbine decoupling control method for coordinated active power tracking and fatigue load optimization according to claim 1 is characterized in that: The step 3 is as follows: 3.1) Small signal analysis method In order to obtain the incremental expression of the fan control model, the small signal analysis method is used; The analysis is as follows: a) Obtain the steady-state operating point of the system through power flow calculation; b) Perform a first-order Taylor expansion at the steady-state operating point to obtain its incremental expression: Where Δx and Δu are the increment vectors of state variables and algebraic variables respectively; Simplify equation (26) into the system small signal state space model at the steady-state operating point: Where A is the state matrix of the system model, 3.2) The incremental linear model of the wind turbine system is as follows: Aerodynamic model: in and Respectively represent T m Yes r , partial derivatives of v and β, where Δ is the incremental expression; Mechanical transmission system: Generator torque control model: Pitch angle servo system model: Tower: in and Respectively represent M t Yes r , partial derivatives of v and β; 3.3) Constructing state space expressions in:

5. The wind turbine decoupling control method for coordinated active power tracking and fatigue load optimization according to claim 4 is characterized in that: The step 4 is specifically as follows: 4.1) Prediction Model Formula (38) is discretized, and the discrete state space model is as follows: Among them A d , B d , C d , D d 、E d and W d Respectively expressed as: C d =C #(53) D d =D #(54) IN d =In #(55) t s is the sampling period of the discrete system, in seconds; 4.2) Optimization goals The cost function of the controller is expressed as: in: N p Expressed as the control interval, Q o is the optimization weight for penalizing output deviation, Q c is the optimized weight for penalizing the input variable, y ref is the reference value of the output variable; The constraints include power reference tracking, power output limit, mechanical transmission system torque limit, tower thrust limit, generator torque limit and pitch angle limit. The specific constraints are as follows: in is the constrained minimum value of the input power variation, is the maximum value of the input power variation constraint, and They are the maximum constraint values ​​of the input bending moment variation and the maximum constraint values ​​of the input spindle torque variation respectively; and Δβ min are the constrained minimum value of the output generator torque change and the constrained minimum value of the output pitch angle change, respectively. and Δβ max They are respectively the constrained maximum value of the output generator torque change and the constrained maximum value of the output pitch angle change.

6. The wind turbine decoupling control method for coordinated active power tracking and fatigue load optimization according to claim 1 is characterized in that: The step 5 is specifically as follows: The correction is done by tracking the power command value as follows: ΔP fc (k+1|k)=ΔP(k+1|k)+ζ[ΔP(k)-ΔP(k|k-1)] #(57) ζ is the correction factor, P fc To track the power command value in feedback correction, ΔP(k+1|k) is the power change at time k predicted at time k+1, ΔP(k) is the power change at time k, and ΔP(k|k-1) is the power change at time k predicted at time k-1.

Images