Composite self-adaptive specified performance variable pitch control method for wind turbine generator

Abstract

The invention discloses a wind turbine generator composite adaptive specified performance variable pitch control method, which comprises the following steps: step 1, establishing a dynamic model of a wind turbine generator, including a pneumatic power equation for capturing wind energy of the wind turbine generator, a transmission system dynamic equation and a wind turbine generator simplified model finally used for control; 2, constructing a finite time specified performance function and a conversion error, and converting a problem with error constraint into an unconstrained problem; and step 3, combining the filtering error, and designing a composite adaptive specified performance variable pitch controller. The finite time stipulated performance control method is adopted, the problem that the traditional asymptotic control convergence speed is low is solved, the convergence time and the steady-state precision can be preset, and the dynamic response performance and the power generation efficiency of the system are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of wind power control, specifically to a composite adaptive specified performance pitch control method for wind turbine generators. Background Technology

[0002] Wind energy, as a clean and renewable energy source, is crucial for addressing the non-renewable nature of fossil fuels and their severe environmental consequences. In recent years, it has been widely applied and rapidly developed through wind power generation systems. Among these, variable pitch wind turbines can achieve more robust power output by adjusting the pitch angle, thereby achieving higher energy capture efficiency, reducing structural loads, and enhancing grid stability. Maintaining high wind speeds at rated output power is critical in all operating areas of wind turbines. This not only improves power generation efficiency and reduces operating costs but also plays a key role in ensuring power system stability and equipment lifespan reliability.

[0003] However, in high-wind-speed regions, wind power systems exhibit uncertain, nonlinear, and non-affine dynamic characteristics. They also need to cope with external disturbances and parameter uncertainties arising from complex and harsh environments, making high-performance pitch control exceptionally complex. Existing research (such as nonlinear PI control and adaptive neural network control) shares a common characteristic: it focuses only on the steady-state performance of the system. In practical applications, ensuring that the tracking error evolves according to specified transient performance is crucial for improving power generation efficiency, reducing fatigue losses, and enhancing system stability. Therefore, specified performance control has emerged as an effective control method. It constrains the tracking error by designing a specified performance function, thereby simultaneously guaranteeing optimal system performance in both transient and steady-state conditions, such as overshoot, convergence speed, and steady-state accuracy. However, most current control methods based on specified performance functions can only constrain error convergence exponentially, meaning that the system tracking error requires an infinite amount of time to enter the specified region, failing to meet the requirement of rapid convergence for high-performance control.

[0004] Unlike asymptotic control and exponentially convergent specified performance control, finite-time specified performance control combines the advantages of finite-time convergence with specified performance control. It ensures that the tracking error converges to a pre-defined boundary defined by the performance function within a finite time, thus providing faster convergence speed, higher control accuracy, and stronger anti-interference capability. Furthermore, considering the uncertainties inherent in wind turbine systems, directly using the upper bound of the approximation error when employing function approximators such as neural networks for compensation may reduce steady-state control accuracy. Therefore, designing a pitch control method for wind turbines that effectively handles uncertainties and integrates finite-time specified performance is of significant research and application value for improving the overall performance, reliability, and economy of wind power systems. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention provides a composite adaptive specified performance pitch control method for wind turbines. A novel finite-time specified performance function is designed to preset the convergence time and steady-state boundary of the system tracking error. A fuzzy logic system and a composite adaptive law are used to uniformly estimate and compensate for the unknown nonlinearity and lumped disturbance of the system, so as to ensure that the rotor tracking error of the system can always be maintained within the specified region and converge and stabilize within the specified steady-state boundary within an adjustable time.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A composite adaptive specified performance pitch control method for wind turbine generators includes the following steps:

[0008] Step 1: Establish the dynamic model of the wind turbine, including the aerodynamic power equation for capturing wind energy, the dynamic equation of the transmission system, and the simplified model of the wind turbine for final control.

[0009] Step 2: Construct a finite-time performance function and a transformation error to transform the problem with error constraints into an unconstrained problem;

[0010] Step 3: Based on the filtering error, design a composite adaptive specified performance pitch controller.

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

[0012] According to Bates' theory, the mechanical power P captured by a wind turbine from the wind... a Represented as

[0013] (1)

[0014] in, Indicates air density; This represents wind speed; π represents pi (the mathematical constant of a circle). Indicates the radius of the wind turbine blade; Indicates the wind energy utilization coefficient; The pitch angle; The tip speed ratio is defined as:

[0015] (2)

[0016] in, Indicates the rotor speed;

[0017] The dynamic equations for the rotor and generator sides of the wind turbine are as follows:

[0018] (3)

[0019] (4)

[0020] in, This represents the aerodynamic torque of the wind turbine generator; and These are the rotor moment of inertia and the generator moment of inertia, respectively. and These are rotor damping and generator damping, respectively; and These are the high-speed shaft torque and low-speed shaft torque of the wind turbine, respectively. The electromagnetic torque of the wind turbine generator; For generator speed; and They represent and The first derivative;

[0021] Based on equations (3)-(4), and considering external interference... Let t be time. The dynamic model of the wind turbine is rewritten as follows:

[0022] (5)

[0023] in, This represents the total inertia of the rotor-side drive system of the wind turbine. This indicates the total damping of the wind turbine. This represents the equivalent electromagnetic torque of the wind turbine. This indicates the gearbox speed ratio.

[0024] Furthermore, the process of step 2 is as follows:

[0025] Construct a finite-time specified performance function of the inverse cosine function type. for:

[0026] (6)

[0027] in, The initial value of the performance function is specified for a finite time period. Let be the steady-state boundary constant, satisfying ; The maximum convergence time constant is preset. This is a preset constant for adjustable convergence speed; It is the inverse cosine function;

[0028] Define rotor speed tracking error for:

[0029] (7)

[0030] in, The desired rotor speed;

[0031] Constructing intermediate variables And select the following strictly monotonically increasing smooth transition function. for:

[0032] (8)

[0033] in, It is an exponential function;

[0034] Taking the inverse function of equation (8), we get:

[0035] (9)

[0036] in, for inverse function, The transformation error is constructed by... This transforms the original constrained problem into an unconstrained problem.

[0037] Furthermore, the process of step 3 is as follows:

[0038] Differentiating equation (9), we get:

[0039] (10)

[0040] in, Conversion error The first derivative; and They are respectively and The first derivative;

[0041] Combining equation (5), the following filter error s is designed:

[0042] (11)

[0043] in, It is a positive integer;

[0044] Differentiating equation (11) yields the filtering error. derivative for:

[0045] (12)

[0046] in, As an intermediate variable, Indicates aerodynamic torque pitch angle Find the partial derivative; It is a bounded variable; Represents an unknown smooth nonlinear function; For lumped interference; For control input;

[0047] Introducing the unknown smooth nonlinear function in the following fuzzy logic system approximation equation (12) :

[0048] (13)

[0049] in, The weight vector is the ideal weight vector, and the superscript T indicates transpose; Represents the input vector; These are the basis functions of the fuzzy logic system. Represents the approximation error of a fuzzy logic system;

[0050] The pitch controller is designed as follows:

[0051] (14)

[0052] in, It is a positive number; for The square of the second norm, Represents the L2 norm; for The estimated value; for The estimated value;

[0053] design and The composite adaptive law is:

[0054] (15)

[0055] (16)

[0056] in, and It is a positive number; and They are respectively and The first derivative.

[0057] The technical concept of this invention is:

[0058] For wind turbine systems operating in high-wind-speed regions and subject to model uncertainties and external disturbances, a novel finite-time performance function is constructed to predetermine the transient and steady-state performance boundaries of the system tracking error. Subsequently, an adaptive performance-compliant pitch controller is designed, which approximates the unknown nonlinear dynamics of the system through a fuzzy logic system and uses a composite adaptive law to uniformly estimate and compensate for nonlinear characteristics, approximation errors, and lumped disturbances. While ensuring system stability, the tracking error is strictly constrained to converge to the pre-specified performance boundaries within a finite time.

[0059] Compared with the prior art, the beneficial effects of the present invention are:

[0060] 1. The present invention adopts a finite-time performance control method, which overcomes the problem of slow convergence speed of traditional asymptotic control. It can pre-set the convergence time and steady-state accuracy, and significantly improves the dynamic response performance and power generation efficiency of the system.

[0061] 2. The composite adaptive law designed in this invention achieves unified compensation for system nonlinearity, fuzzy approximation error and external disturbances, avoids the problem of reduced control accuracy caused by directly using the upper bound of the error, and enhances the robustness and steady-state control accuracy of the system.

[0062] 3. Compared with traditional PID control, exponential convergence performance control and other methods, this invention has faster finite-time convergence characteristics, higher power control accuracy and stronger anti-interference ability, and is more suitable for the complex operating conditions of actual wind turbine units. Attached Figure Description

[0063] Figure 1 This is a control flowchart of the present invention;

[0064] Figure 2 This is a diagram of turbulent wind speeds;

[0065] Figure 3 Rotor speed diagram;

[0066] Figure 4 For tracking error graph;

[0067] Figure 5 Pitch angle diagram;

[0068] Figure 6 This is a diagram showing the power generation output. Detailed Implementation

[0069] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0070] Reference Figures 1-6 A composite adaptive specified performance pitch control method for wind turbine generators includes the following steps:

[0071] Step 1: Establish the dynamic model of the wind turbine, including its aerodynamic power equation for capturing wind energy, the dynamic equation of the transmission system, and the simplified model of the wind turbine used for control. The process is as follows:

[0072] According to Bates' theory, the mechanical power P captured by a wind turbine from the wind... a It can be represented as

[0073] (1)

[0074] in, Indicates air density; This represents wind speed; π represents pi (the mathematical constant of a circle). Indicates the radius of the wind turbine blade; Indicates the wind energy utilization coefficient; The pitch angle; The tip speed ratio is defined as:

[0075] (2)

[0076] in, Indicates the rotor speed;

[0077] The dynamic equations for the rotor and generator sides of the wind turbine are as follows:

[0078] (3)

[0079] (4)

[0080] in, This represents the aerodynamic torque of the wind turbine generator; and These are the rotor moment of inertia and the generator moment of inertia, respectively. and These are rotor damping and generator damping, respectively; and These are the high-speed shaft torque and low-speed shaft torque of the wind turbine, respectively. The electromagnetic torque of the wind turbine generator; For generator speed; and They represent and The first derivative.

[0081] Based on equations (3)-(4), and considering external interference... Let t be time. The dynamic model of the wind turbine is rewritten as follows:

[0082] (5)

[0083] in, This represents the total inertia of the rotor-side drive system of the wind turbine. This indicates the total damping of the wind turbine. This represents the equivalent electromagnetic torque of the wind turbine. This indicates the gearbox speed ratio.

[0084] Step 2: Construct a finite-time performance function and a transformation error to transform the problem with error constraints into an unconstrained problem. The process is as follows:

[0085] Construct a finite-time specified performance function of the inverse cosine function type. for

[0086] (6)

[0087] in, The initial value of the performance function is specified for a finite time period. Let be the steady-state boundary constant, satisfying ; The maximum convergence time constant is preset. This is a preset constant for adjustable convergence speed; It is an inverse cosine function.

[0088] Define rotor speed tracking error for:

[0089] (7)

[0090] in, The desired rotor speed.

[0091] To ensure speed tracking error Satisfy constraints Construct new intermediate variables And select the following strictly monotonically increasing smooth transition function. for

[0092] (8)

[0093] in, It is an exponential function;

[0094] Taking the inverse function of equation (8), we can obtain:

[0095] (9)

[0096] in, for inverse function, The transformation error is constructed by... This transforms the original constrained problem into an unconstrained problem.

[0097] Step 3: Based on the filtering error, design a composite adaptive specified performance pitch controller to achieve rapid error convergence. The process is as follows:

[0098] Differentiating equation (9), we get:

[0099] (10)

[0100] in, Conversion error The first derivative; and They are respectively and The first derivative.

[0101] Combining equation (5), the following filter error s is designed:

[0102] (11)

[0103] in, It is a positive integer.

[0104] Differentiating equation (11) yields the filtering error. derivative for:

[0105] (12)

[0106] in, , Indicates aerodynamic torque pitch angle Find the partial derivative; It is a bounded variable; Represents an unknown smooth nonlinear function; For lumped interference; To control the input.

[0107] Introducing the unknown smooth nonlinear function in the following fuzzy logic system approximation equation (12) :

[0108] (13)

[0109] in, The weight vector is the ideal weight vector, and the superscript T indicates transpose; Represents the input vector; These are the basis functions of the fuzzy logic system. It represents the approximation error of a fuzzy logic system.

[0110] Based on the above design, the pitch controller is designed as follows:

[0111] (14)

[0112] in, It is a positive number; for The square of the second norm, Represents the L2 norm; for The estimated value; for The estimated value.

[0113] design and The composite adaptive law is:

[0114] (15)

[0115] (16)

[0116] in, and It is a positive number; and They are respectively and The initial value satisfies and , and They are respectively and The first derivative.

[0117] Analyze system performance and construct Lyapunov functions. for:

[0118] (17)

[0119] in, and To estimate the error; It is an unknown positive constant that satisfies .

[0120] Differentiating equation (17), and substituting equations (5), (12), (14), (15), and (16) into the equation, we obtain the Lyapunov function. :

[0121] (18)

[0122] in, This indicates taking the minimum value among the three. It is the residual; according to the uniformly eventually bounded stability theorem, the system tracking error can always be maintained within the specified region and within an adjustable time. It converges and stabilizes within the specified steady-state boundary.

[0123] To verify the tracking performance of the pitch control strategy of this invention, the Matlab / Simulink wind power simulation module jointly developed by Aalborg University and Risø National Laboratory in Denmark was used. This module originated from a project funded by the Danish Energy Agency and is currently a widely used simulation platform for wind turbine technology analysis and verification. The model parameters are set as follows:

[0124] This invention verifies the feasibility of the composite adaptive specified performance pitch angle control strategy (M1 strategy) through comparative studies; an adaptive neural network strategy (M2 strategy) is selected as a comparison scheme, in which the filtering error is designed as follows: The neural network pitch controller is designed as The weight update law is designed as follows To ensure fair comparison, the parameters for filtering error, pitch controller, and update law in the M2 strategy are consistent with the parameters set in formulas (11), (14), and (15). The corresponding parameter values ​​are as follows: , , , , , The parameters of the finite-time performance function are selected as follows: The initial value of the rotor speed is set as follows: ; and The initial values ​​are set as follows: and External interference settings: The simulation runtime range is... , The sampling time was 0.001s, and the number of samples was [number missing]. When the required number of samples is reached, the fan rotor speed and blade pitch angle are output; if the required number of samples is not reached, the rotor speed and error are recalculated.

[0125] Figure 2 A turbulent wind speed profile with an average wind speed of 18 m / s and a turbulence intensity of 10% is shown. Figure 3 Rotor speeds under different control strategies are described. Response curve; from Figure 3 As can be seen, compared with the M2 strategy, the M1 strategy proposed in this paper can track the rated rotor speed faster. ; Figure 4 The tracking error under different control strategies is given. Changes; from Figure 4 It can be seen that the tracking error under the M1 strategy proposed in this paper... The convergence time is approximately 10 seconds, and the tracking accuracy reaches [value missing]. In contrast, the M2 strategy has a convergence time of approximately 30 seconds and a tracking accuracy of [missing information]. Furthermore, the proposed M1 strategy can strictly limit the tracking error within the region defined by the finite-time performance function, while the M2 strategy exceeds this preset boundary. This shows that, with the help of the finite-time performance control strategy, the proposed M1 strategy can bring better control performance to the wind turbine than the M2 strategy. Figure 5 and Figure 6 The pitch angles under different control strategies are shown respectively. and power generation ;from Figure 5 It can be seen that the pitch angles of the two contrasting strategies The responses were largely consistent; however, from Figure 6 It can be seen that the power output of strategy M1 is more stable than that of strategy M2.

[0126] In summary, a composite adaptive specified performance pitch control method for wind turbines can effectively handle system nonlinearity and external disturbances. This controller exhibits significant advantages in convergence speed, steady-state accuracy, transient performance constraints, and robustness, and has important engineering application value in the high-precision power control and safe operation of wind turbines.

[0127] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A composite adaptive specified performance pitch control method for wind turbine generators, characterized in that, Includes the following steps: Step 1: Establish the dynamic model of the wind turbine, including the aerodynamic power equation for capturing wind energy, the dynamic equation of the transmission system, and the simplified model of the wind turbine for final control. Step 2: Construct a finite-time performance function and a transformation error to transform the problem with error constraints into an unconstrained problem; Step 3: Based on the filtering error, design a composite adaptive specified performance pitch controller.

2. The wind turbine composite adaptive specified performance pitch control method according to claim 1, characterized in that, The process of step 1 is as follows: According to Bates' theory, the mechanical power P captured by a wind turbine from the wind... a Represented as (1) in, Indicates air density; This represents wind speed; π represents pi (the mathematical constant of a circle). Indicates the radius of the wind turbine blade; Indicates the wind energy utilization coefficient; The pitch angle; The tip speed ratio is defined as: （2） in, Indicates the rotor speed; The dynamic equations for the rotor and generator sides of the wind turbine are as follows: （3） （4） in, This represents the aerodynamic torque of the wind turbine generator; and These are the rotor moment of inertia and the generator moment of inertia, respectively. and These are rotor damping and generator damping, respectively; and These are the high-speed shaft torque and low-speed shaft torque of the wind turbine, respectively. The electromagnetic torque of the wind turbine generator; For generator speed; and They represent and The first derivative; Based on equations (3)-(4), and considering external interference... Let t be time. The dynamic model of the wind turbine is rewritten as follows: （5） in, This represents the total inertia of the rotor-side drive system of the wind turbine. This indicates the total damping of the wind turbine. This represents the equivalent electromagnetic torque of the wind turbine generator. This indicates the gearbox speed ratio.

3. The wind turbine composite adaptive specified performance pitch control method according to claim 2, characterized in that, The process of step 2 is as follows: Construct a finite-time specified performance function of the inverse cosine function type. for: （6） in, Specify the initial value of the performance function for a finite time period. Let be the steady-state boundary constant, satisfying ; The maximum convergence time constant is preset. This is a preset constant for adjustable convergence speed; It is the inverse cosine function; Define rotor speed tracking error for: （7） in, The desired rotor speed; Constructing intermediate variables And select the following strictly monotonically increasing smooth transition function. for: （8） in, It is an exponential function; Taking the inverse function of equation (8), we get: （9） in, for inverse function, The transformation error is constructed by... This transforms the original constrained problem into an unconstrained problem.

4. The wind turbine composite adaptive specified performance pitch control method according to claim 3, characterized in that, The process of step 3 is as follows: Differentiating equation (9), we get: （10） in, Conversion error The first derivative; and They are respectively and The first derivative; Combining equation (5), the following filter error s is designed: （11） in, It is a positive integer; Differentiating equation (11) yields the filtering error. derivative for: （12） in, As an intermediate variable, Indicates aerodynamic torque pitch angle Find the partial derivative; It is a bounded variable; Represents an unknown smooth nonlinear function; For lumped interference; For control input; Introducing the unknown smooth nonlinear function in the following fuzzy logic system approximation equation (12) : （13） in, The weight vector is the ideal weight vector, and the superscript T indicates transpose; Represents the input vector; These are the basis functions of the fuzzy logic system; Represents the approximation error of a fuzzy logic system; The pitch controller is designed as follows: （14） in, It is a positive number; for The square of the second norm, Represents the L2 norm; for The estimated value; for The estimated value; design and The composite adaptive law is: （15） （16） in, and It is a positive number; and They are respectively and The first derivative.

Images