A method and system for rapid estimation of fatigue load damage of a wind turbine

Abstract

This invention discloses a rapid estimation method and system for fatigue load damage of wind turbine generator sets. The method includes the following steps: constructing a linearized high-order aeroelastic dynamics model and a control system model of the wind turbine generator set; integrating the linearized high-order aeroelastic dynamics model and the control system model to obtain a closed-loop aerodynamic servo-elastic system model; using the three-dimensional wind speed in the geodetic coordinate system as the input of the closed-loop aerodynamic servo-elastic system model, and outputting the dynamic response of the wind turbine generator set load in the geodetic coordinate system; converting the dynamic response of the wind turbine generator set load in the geodetic coordinate system into the dynamic response of the wind turbine generator set load in the rotating coordinate system; calculating the power spectral density response in both the geodetic and rotating coordinate systems; and calculating the load fatigue damage of the wind turbine generator set. This invention achieves the integration of mechanics, control systems, and electrical systems, and multidisciplinary optimization design, which is beneficial for early optimization design and reduction of levelized cost of electricity.

Description

Technical Field

[0001] This invention relates to the technical field of fatigue load calculation for wind turbine generator sets, and in particular to a rapid estimation method and system for fatigue load damage of wind turbine generator sets. Background Technology

[0002] Currently, wind turbine generators are developing towards larger sizes, longer blades, and even multiple rotors. Whether it's onshore stationary wind turbines or offshore floating wind turbines, the complexity of the system requires that the component design of the unit be considered as part of a more holistic and integrated system rather than as independent systems. This leads to more convenient and efficient design. Multidisciplinary design optimization methods are very suitable for such holistic design. This design method has strong versatility and can be applied to models of varying complexity. One of the main challenges of a multidisciplinary design framework is ensuring that the model has enough detail to describe such a complete system while simultaneously limiting computational costs, conserving computational resources, and quickly calculating results.

[0003] Existing technologies, such as Chinese invention patent application CN114330047A, disclose a method, system, medium, and equipment for rapid assessment of offshore wind turbine loads. This method is mainly aimed at offshore fixed wind turbine units. It calculates the relationship curve between soil resistance and pile displacement based on the foundation pile structure and soil parameters below the mud surface. It also obtains the characteristic load interpolation function for each verification turbine location by fitting the load history data database of offshore wind power projects and the results of sensitivity analysis. However, this application is for rapid verification of the characteristic loads of different turbine locations in the entire wind farm, not for optimizing the turbine model load. Furthermore, it can only assess the ultimate load, not the component fatigue load. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and propose a rapid estimation method and system for fatigue load damage of wind turbine generator sets. By applying a high-order linear wind turbine generator set dynamic model, the whole system and control system can be described in detail. The model input is the three-dimensional incoming wind speed sensed by the blades, and the output is the dynamic response measured by the system's sensors. The power spectral density is calculated based on this, and then the fatigue damage of each model sensor is rapidly evaluated.

[0005] The objective of this invention is achieved through the following technical solution: a rapid estimation method for fatigue load damage of wind turbine generator sets, comprising the following steps:

[0006] S1. Collect the wind speed sampled by the wind turbine blades as they rotate, and construct a linearized high-order aeroelastic dynamics model and a control system model for the wind turbine.

[0007] S2. Correlate the input of the linearized high-order aeroelastic dynamics model with the output of the control system model, integrate the linearized high-order aeroelastic dynamics model and the control system model to obtain a closed-loop aerodynamic servo-elastic system model.

[0008] S3. Define the three-dimensional wind speed time series in the blade rotation coordinate system, and obtain the three-dimensional wind speed in the geodetic coordinate system through the transformation matrix; use the three-dimensional wind speed in the geodetic coordinate system as the input of the closed-loop aerodynamic servo elastic system model, and output the dynamic response of the wind turbine load in the geodetic coordinate system.

[0009] S4. Convert the dynamic response of the wind turbine load in the geodetic coordinate system to the dynamic response of the wind turbine load in the rotating coordinate system;

[0010] S5. Based on the dynamic response of the wind turbine load in the geodetic coordinate system and the dynamic response of the wind turbine load in the rotating coordinate system, calculate the power spectral density response in the geodetic coordinate system and the rotating coordinate system.

[0011] S6. Based on the power spectral density response in the geodetic coordinate system and the rotating coordinate system, calculate the load fatigue damage of the wind turbine generator.

[0012] Furthermore, step S1 includes:

[0013] The linearized high-order aeroelastic dynamics model of wind turbine generator is represented by state-space equations, as shown in formulas (1a) to (1c). This linearized high-order aeroelastic dynamics model includes unsteady aerodynamic information of the blades and structural dynamics information. The unsteady aerodynamic information of the blades includes eddies, dynamic stall, dynamic inflow and wake.

[0014] ;

[0015] ;

[0016] ;

[0017] in, State vector The differential of its state variables, including unsteady aerodynamic information of the blades and structural dynamics information, dimension. ; For input to the control system model, The disturbance input to the control system model, dimension ; , This represents the three-dimensional wind speed input into the model. For horizontal wind speed, For cross wind speed and The vertical wind speed is recorded; the three-dimensional wind speed is obtained by an anemometer installed on the blades. and It is the output of the dynamic model, dimension. and ; For the dynamic system matrix, The drive matrix is ​​the input to the control system. and For the observable matrix of the dynamic model, and For the feedforward matrix of the dynamic model, , , , , , , , , The input matrix represents three wind speeds;

[0018] The control system model adopts state-space equations, as shown in equations (2a) to (2b):

[0019] ;

[0020] ;

[0021] in, For the state vector of the control system model The dimension is the differential of the state vector. ; The dimension is the input signal to the control system model. ; It is the output of the control system model, dimension. ; For the control system model matrix, The driving matrix is ​​the input to the control system model. The observable matrix of the control system model. This is the feedforward matrix of the control system model.

[0022] Furthermore, step S2 includes:

[0023] Input to the linearized high-order aeroelastic dynamics model With the output of the control system model Related, the output of the linearized higher-order aeroelastic dynamics model Input to the control system model Related, as shown in formula (3):

[0024] ;

[0025] Based on formulas (3), (1a), (2a) and (1b), a closed-loop aerodynamic servo-elastic system is integrated. The input of the closed-loop aerodynamic servo-elastic system is the three-dimensional wind speed at any operating point. Its system and output equations are shown in formulas (4a) to (4b):

[0026] ;

[0027] ;

[0028] The state vector of the pneumatic servo elastic system is , The derivative of the state vector, Output the state vector; The identity matrix has the following dimensions. The closed-loop pneumatic servo elastic system and its input matrix are shown in equations (5a) to (5e):

[0029] ;

[0030] ;

[0031] ;

[0032] ;

[0033] ;

[0034] The output and feedforward matrix of the closed-loop pneumatic servo elastic system are shown in equations (6a) to (6e):

[0035] ;

[0036] ;

[0037] ;

[0038] ;

[0039] .

[0040] Furthermore, step S3 includes:

[0041] Define the input timing of the three-dimensional wind speed in the blade rotating coordinate system as follows: , as well as , Input timing for horizontal wind speed, Input timing for crosswind speed and The vertical wind speed is input as a time series, and the three-dimensional wind speed in the geodetic coordinate system is obtained through the transformation matrix (7). , as well as , The horizontal wind speed is given in the geodetic coordinate system. The lateral wind speed is given in the geodetic coordinate system. The vertical wind speed is given in the geodetic coordinate system.

[0042] ;

[0043] The wind speed signals obtained by Fourier transform of the three-dimensional wind speed in the geodetic coordinate system are shown in formulas (8a) to (8c):

[0044] ;

[0045] ;

[0046] ;

[0047] The transfer function of the closed-loop system matrix of the wind turbine generator is obtained as shown in Equation (9). The input is the three-dimensional wind speed (8a) to (8c), and the output is the dynamic response of the load in the complex domain as shown in Equation (10):

[0048] ;

[0049] in It is an imaginary number. Angular frequency;

[0050] The dynamic response of the wind turbine generator load in the geodetic coordinate system is shown in Equation (10):

[0051] .

[0052] Furthermore, step S4 includes:

[0053] In the frequency domain, the load response in the geodetic coordinate system is transformed into that in the rotating coordinate system. Equation (7) is expressed as an exponential function in the complex domain, as shown in Equation (12):

[0054] ;

[0055] in, , , ;

[0056] Decompose matrix B into symmetric, cosine, and sine forms;

[0057] Based on formulas (10) and (12), the dynamic response of the wind turbine load in the time-domain geodetic coordinate system is expressed. Dynamic response of wind turbine load converted to time-domain blade rotating coordinate system As shown in formula (13):

[0058] ;

[0059] in, for The result after Fourier transformation; Equation (13) is also transformed into the load dynamic response in the frequency domain by Fourier transformation, as shown in Equation (14):

[0060] .

[0061] Furthermore, step S5 includes:

[0062] The expression for the power spectral density response in the geodetic coordinate system is shown in equation (11):

[0063] ;

[0064] The expression for the power spectral density response in the rotating coordinate system is shown in equation (15):

[0065] .

[0066] Furthermore, step S6 includes:

[0067] Load fatigue damage based on power spectral density response is calculated using formulas (11) and (15), employing a combination of rainflow counting and empirical formulas; the SN curve is defined as shown in formula (16):

[0068] ;

[0069] in The amplitude of the input load signal. The number of loops. and All are constants; from the power spectral density and scalar obtained To estimate rainflow counts, and with scalars representing the complete power spectrum, The characteristic frequency is shown in formula (17):

[0070] ;

[0071] Define bandwidth coefficient as well as As shown in formula (18):

[0072] ;

[0073] Fatigue damage rate D is the damage rate of a narrow bandwidth process. Damage rate with amplitude counting The linear weighted result is shown in formula (19):

[0074] ;

[0075] in These are weighting coefficients. Represented as With second-order bandwidth coefficient index The product of functions is shown in formula (20):

[0076] ;

[0077] The narrow bandwidth impairment coefficient is calculated as shown in formula (21):

[0078] ;

[0079] in For gamma function, The critical stress level, The average uptake rate is shown in formula (22):

[0080] ;

[0081] Obtained by fitting empirical data and used as a bandwidth parameter and The function is shown in formula (23):

[0082] .

[0083] Furthermore, step S6 includes:

[0084] To estimate the bending moment for which gravity distribution is the main source of fatigue damage, the gravity term needs to be added to the power spectral density response, as shown in equation (24):

[0085] ;

[0086] in, For the blade length, The mass distribution of the blade and its distribution along the blade span The curve function, This is the acceleration due to gravity.

[0087] A rapid estimation system for fatigue load damage of wind turbine generator sets, used to implement the aforementioned rapid estimation method for fatigue load damage of wind turbine generator sets, includes:

[0088] The wind speed acquisition module is used to collect the wind speed sampled by the wind turbine blades as they rotate.

[0089] The dynamics modeling module is used to construct a linearized high-order aeroelastic dynamics model of a wind turbine generator.

[0090] The control system modeling module is used to build the control system model of the wind turbine generator set;

[0091] The closed-loop aero-servo-elastic system modeling module is used to integrate the linearized high-order aeroelastic dynamics model and the control system model to obtain the closed-loop aero-servo-elastic system model.

[0092] The dynamic response calculation module takes the three-dimensional wind speed in the geodetic coordinate system as the input to the closed-loop aerodynamic servo elastic system model and outputs the dynamic response of the wind turbine load in the geodetic coordinate system; it also converts the dynamic response of the wind turbine load in the geodetic coordinate system into the dynamic response of the wind turbine load in the rotating coordinate system.

[0093] The power spectral density response calculation module calculates the power spectral density response in both the geodetic and rotating coordinate systems based on the dynamic response of the wind turbine load in the geodetic coordinate system and the dynamic response of the wind turbine load in the rotating coordinate system.

[0094] The load fatigue damage calculation module calculates the load fatigue damage of wind turbine generators based on the power spectral density response in both the geodetic coordinate system and the rotating coordinate system.

[0095] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0096] 1. This invention integrates a high-fidelity linearized dynamic model of a wind turbine generator set with a control system model in a closed loop, thereby achieving the integration and multidisciplinary optimization design of mechanics, control system, and electrical system. The output results are reliable and accurate, which is beneficial for the early optimization design of wind turbine generator sets and the reduction of the cost per kilowatt-hour.

[0097] 2. This invention can avoid a large amount of preliminary load calculation work. For example, the existing technology requires several hours or even a day to complete a full set of load calculations for wind turbine generators, while the present invention can complete the rapid assessment of fatigue load in less than 10 minutes, and then carry out the next step of design decision-making and technical solution comparison. Attached Figure Description

[0098] Figure 1The flowchart shows the calculation process for a rapid estimation method of fatigue load damage in wind turbine generator sets. Detailed Implementation

[0099] The present invention will be further described below with reference to specific embodiments.

[0100] Example 1

[0101] See Figure 1 The diagram shows a rapid estimation method for fatigue load damage of wind turbine generator sets provided in this embodiment, which includes the following steps:

[0102] S1. Collect the wind speed sampled by the wind turbine blades as they rotate, and construct a linearized high-order aeroelastic dynamics model and a control system model for the wind turbine. The linearized high-order aeroelastic dynamics model of the wind turbine is represented by state-space equations, as shown in formulas (1a) to (1c). The linearized high-order aeroelastic dynamics model includes unsteady aerodynamic information of the blades and structural dynamics information. The unsteady aerodynamic information of the blades includes eddies, dynamic stall, dynamic inflow and wake.

[0103] ;

[0104] ;

[0105] ;

[0106] in, State vector The differential of its state variables, including unsteady aerodynamic information of the blades and structural dynamics information, dimension. ; For input to the control system model, The disturbance input to the control system model, dimension ; , This represents the three-dimensional wind speed input into the model. For horizontal wind speed, For cross wind speed and The vertical wind speed is recorded; the three-dimensional wind speed is obtained by an anemometer installed on the blades. and It is the output of the dynamic model, dimension. and ; For the dynamic system matrix, The drive matrix is ​​the input to the control system. and For the observable matrix of the dynamic model, and For the feedforward matrix of the dynamic model, , , , , , , , , The input matrix represents three wind speeds;

[0107] The control system model adopts state-space equations, as shown in equations (2a) to (2b):

[0108] ;

[0109] ;

[0110] in, For the state vector of the control system model The dimension is the differential of the state vector. ; The dimension is the input signal to the control system model. ; It is the output of the control system model, dimension. ; For the control system model matrix, The driving matrix is ​​the input to the control system model. The observable matrix of the control system model. This is the feedforward matrix of the control system model.

[0111] S2, Process 2: Correlate the input of the linearized higher-order aeroelastic dynamics model with the output of the control system model, integrate the linearized higher-order aeroelastic dynamics model and the control system model to obtain a closed-loop aerodynamic servo-elastic system model, including: the input of the linearized higher-order aeroelastic dynamics model... With the output of the control system model Related, the output of the linearized higher-order aeroelastic dynamics model Input to the control system model Related, as shown in formula (3):

[0112] ;

[0113] Based on formulas (3), (1a), (2a) and (1b), a closed-loop aerodynamic servo-elastic system is integrated. The input of the closed-loop aerodynamic servo-elastic system is the three-dimensional wind speed at any operating point. Its system and output equations are shown in formulas (4a) to (4b):

[0114] ;

[0115] ;

[0116] The state vector of the pneumatic servo elastic system is , The derivative of the state vector, Output the state vector; The identity matrix has the following dimensions. The closed-loop pneumatic servo elastic system and its input matrix are shown in equations (5a) to (5e):

[0117] ;

[0118] ;

[0119] ;

[0120] ;

[0121] ;

[0122] The output and feedforward matrix of the closed-loop pneumatic servo elastic system are shown in equations (6a) to (6e):

[0123] ;

[0124] ;

[0125] ;

[0126] ;

[0127] .

[0128] S3, Process 1: Define the three-dimensional wind speed time series in the blade rotating coordinate system using wind sampling, and obtain the three-dimensional wind speed in the geodetic coordinate system through a transformation matrix; use the three-dimensional wind speed in the geodetic coordinate system as the input to the closed-loop aerodynamic servo elastic system model, and output the dynamic response of the wind turbine load in the geodetic coordinate system, including:

[0129] Define the input timing of the three-dimensional wind speed in the blade rotating coordinate system as follows: , as well as , Input timing for horizontal wind speed, Input timing for crosswind speed and The vertical wind speed is input as a time series, and the three-dimensional wind speed in the geodetic coordinate system is obtained through the transformation matrix (7). , as well as , The horizontal wind speed is given in the geodetic coordinate system. The lateral wind speed is given in the geodetic coordinate system. The vertical wind speed is given in the geodetic coordinate system.

[0130] ;

[0131] The wind speed signals obtained by Fourier transform of the three-dimensional wind speed in the geodetic coordinate system are shown in formulas (8a) to (8c):

[0132] ;

[0133] ;

[0134] ;

[0135] The transfer function of the closed-loop system matrix of the wind turbine generator is obtained as shown in Equation (9). The input is the three-dimensional wind speed (8a) to (8c), and the output is the dynamic response of the load in the complex domain as shown in Equation (10):

[0136] ;

[0137] in It is an imaginary number. Angular frequency;

[0138] Step 3: The dynamic response of the wind turbine load in the geodetic coordinate system is shown in Equation (10):

[0139] .

[0140] S4. Convert the dynamic response of the wind turbine load in the geodetic coordinate system to the dynamic response of the wind turbine load in the rotating coordinate system, including:

[0141] In the frequency domain, the load response in the geodetic coordinate system is transformed into that in the rotating coordinate system. Equation (7) is expressed as an exponential function in the complex domain, as shown in Equation (12):

[0142] ;

[0143] in, , , ;

[0144] Decompose matrix B into symmetric, cosine, and sine forms;

[0145] Based on formulas (10) and (12), the dynamic response of the wind turbine load in the time-domain geodetic coordinate system is expressed. Dynamic response of wind turbine load converted to time-domain blade rotating coordinate system As shown in formula (13):

[0146] ;

[0147] in, for The result after Fourier transformation; Equation (13) is also transformed into the load dynamic response in the frequency domain by Fourier transformation, as shown in Equation (14):

[0148] .

[0149] S5. Based on the dynamic response of the wind turbine load in the geodetic coordinate system and the dynamic response of the wind turbine load in the rotating coordinate system, calculate the power spectral density response in both the geodetic and rotating coordinate systems, including:

[0150] The expression for the power spectral density response in the geodetic coordinate system is shown in equation (11):

[0151] ;

[0152] The expression for the power spectral density response in the rotating coordinate system is shown in equation (15):

[0153] .

[0154] S6. Based on the power spectral density response in the geodetic coordinate system and the rotating coordinate system, calculate the load fatigue damage of the wind turbine generator set, including:

[0155] Load fatigue damage based on power spectral density response is calculated using formulas (11) and (15), employing a combination of rainflow counting and empirical formulas; the SN curve is defined as shown in formula (16):

[0156] ;

[0157] in The amplitude of the input load signal. The number of loops. and All are constants; from the power spectral density and scalar obtained To estimate rainflow counts, and with scalars representing the complete power spectrum, The characteristic frequency is shown in formula (17):

[0158] ;

[0159] Define bandwidth coefficient as well as As shown in formula (18):

[0160] ;

[0161] Fatigue damage rate D is the damage rate of a narrow bandwidth process. Damage rate with amplitude counting The linear weighted result is shown in formula (19):

[0162] ;

[0163] in These are weighting coefficients. Approximately expressed as With second-order bandwidth coefficient index The product of functions is shown in formula (20):

[0164] ;

[0165] The narrow bandwidth impairment coefficient is calculated as shown in formula (21):

[0166] ;

[0167] in For gamma function, The critical stress level, The average uptake rate is shown in formula (22):

[0168] ;

[0169] Obtained by fitting empirical data and used as a bandwidth parameter and The function is shown in formula (23):

[0170] .

[0171] To estimate the bending moment for which gravity distribution is the main source of fatigue damage, the gravity term needs to be added to the power spectral density response, as shown in equation (24):

[0172] ;

[0173] in, For the blade length, The mass distribution of the blade and its distribution along the blade span The curve function, This is the acceleration due to gravity.

[0174] Example 2

[0175] This embodiment discloses a rapid estimation system for fatigue load damage of wind turbine generator sets, used to implement the rapid estimation method for fatigue load damage of wind turbine generator sets described in Embodiment 1, including:

[0176] The wind speed acquisition module is used to collect the wind speed sampled by the wind turbine blades as they rotate.

[0177] The dynamics modeling module is used to construct a linearized high-order aeroelastic dynamics model of a wind turbine generator.

[0178] The control system modeling module is used to build the control system model of the wind turbine generator set;

[0179] The closed-loop aero-servo-elastic system modeling module is used to integrate the linearized high-order aeroelastic dynamics model and the control system model to obtain the closed-loop aero-servo-elastic system model.

[0180] The dynamic response calculation module takes the three-dimensional wind speed in the geodetic coordinate system as the input to the closed-loop aerodynamic servo elastic system model and outputs the dynamic response of the wind turbine load in the geodetic coordinate system; it also converts the dynamic response of the wind turbine load in the geodetic coordinate system into the dynamic response of the wind turbine load in the rotating coordinate system.

[0181] The power spectral density response calculation module calculates the power spectral density response in both the geodetic and rotating coordinate systems based on the dynamic response of the wind turbine load in the geodetic coordinate system and the dynamic response of the wind turbine load in the rotating coordinate system.

[0182] The load fatigue damage calculation module calculates the load fatigue damage of wind turbine generators based on the power spectral density response in both the geodetic coordinate system and the rotating coordinate system.

[0183] Example 3

[0184] This embodiment discloses a non-transitory computer-readable medium storing instructions that, when executed by a processor, perform the steps of the rapid estimation method for fatigue load damage of a wind turbine generator set according to Embodiment 1.

[0185] In this embodiment, the non-transitory computer-readable medium can be a disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), USB flash drive, portable hard drive, etc.

[0186] Example 4

[0187] This embodiment discloses a computing device, including a processor and a memory for storing processor-executable programs. When the processor executes the program stored in the memory, it implements the rapid estimation method for fatigue load damage of wind turbine generator sets as described in Embodiment 1.

[0188] The computing device described in this embodiment may be a desktop computer, laptop computer, smartphone, PDA handheld terminal, tablet computer, programmable logic controller (PLC), or other terminal device with processor functionality.

[0189] The above-described embodiments are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Therefore, any changes made in accordance with the shape and principle of the present invention should be covered within the protection scope of the present invention.

Claims

1. A rapid estimation method for fatigue load damage of wind turbine generator sets, characterized in that, Includes the following steps: S1. Collect the wind speed sampled by the wind turbine blades as they rotate, and construct a linearized high-order aeroelastic dynamics model and a control system model for the wind turbine. S2. Correlate the input of the linearized high-order aeroelastic dynamics model with the output of the control system model, integrate the linearized high-order aeroelastic dynamics model and the control system model to obtain a closed-loop aerodynamic servo-elastic system model. S3. Define the three-dimensional wind speed time series in the blade rotation coordinate system, and obtain the three-dimensional wind speed in the geodetic coordinate system through the transformation matrix; use the three-dimensional wind speed in the geodetic coordinate system as the input of the closed-loop aerodynamic servo elastic system model, and output the dynamic response of the wind turbine load in the geodetic coordinate system. S4. Convert the dynamic response of the wind turbine load in the geodetic coordinate system to the dynamic response of the wind turbine load in the rotating coordinate system; S5. Based on the dynamic response of the wind turbine load in the geodetic coordinate system and the dynamic response of the wind turbine load in the rotating coordinate system, calculate the power spectral density response in the geodetic coordinate system and the rotating coordinate system. S6. Based on the power spectral density response in the geodetic coordinate system and the rotating coordinate system, calculate the load fatigue damage of the wind turbine generator.

2. The rapid estimation method for fatigue load damage of wind turbine generator sets according to claim 1, characterized in that, Step S1 includes: The linearized high-order aeroelastic dynamics model of wind turbine generator is represented by state-space equations, as shown in formulas (1a) to (1c). This linearized high-order aeroelastic dynamics model includes unsteady aerodynamic information of the blades and structural dynamics information. The unsteady aerodynamic information of the blades includes eddies, dynamic stall, dynamic inflow and wake. ； ； ； in, State vector The differential of its state variables, including unsteady aerodynamic information of the blades and structural dynamics information, dimension. ; For input to the control system model, The disturbance input to the control system model, dimension ; , This represents the three-dimensional wind speed input into the model. For horizontal wind speed, For cross wind speed and The vertical wind speed is recorded; the three-dimensional wind speed is obtained by an anemometer installed on the blades. and It is the output of the dynamic model, dimension. and ; For the dynamic system matrix, The drive matrix is ​​the input to the control system. and For the observable matrix of the dynamic model, and For the feedforward matrix of the dynamic model, , , , , , , , , The input matrix represents three wind speeds; The control system model adopts state-space equations, as shown in equations (2a) to (2b): ； ； in, For the state vector of the control system model The dimension is the differential of the state vector. ; The dimension is the input signal to the control system model. ; It is the output of the control system model, dimension. ; For the control system model matrix, The driving matrix is ​​the input to the control system model. The observable matrix of the control system model. This is the feedforward matrix of the control system model.

3. The rapid estimation method for fatigue load damage of wind turbine generator sets according to claim 2, characterized in that, Step S2 includes: Input to the linearized high-order aeroelastic dynamics model With the output of the control system model Related, the output of the linearized higher-order aeroelastic dynamics model Input to the control system model Related, as shown in formula (3): ； Based on formulas (3), (1a), (2a) and (1b), a closed-loop aerodynamic servo-elastic system is integrated. The input of the closed-loop aerodynamic servo-elastic system is the three-dimensional wind speed at any operating point. Its system and output equations are shown in formulas (4a) to (4b): ； ； The state vector of the pneumatic servo elastic system is , The derivative of the state vector, Output the state vector; The identity matrix has the following dimensions. The closed-loop pneumatic servo elastic system and its input matrix are shown in equations (5a) to (5e): ； ； ； ； ； The output and feedforward matrix of the closed-loop pneumatic servo elastic system are shown in equations (6a) to (6e): ； ； ； ； 。 4. The rapid estimation method for fatigue load damage of a wind turbine generator set according to claim 3, characterized in that, Step S3 includes: Define the input timing of the three-dimensional wind speed in the blade rotating coordinate system as follows: , as well as , Input timing for horizontal wind speed, Input timing for crosswind speed and The vertical wind speed is input as a time series, and the three-dimensional wind speed in the geodetic coordinate system is obtained through the transformation matrix (7). , as well as , The horizontal wind speed is given in the geodetic coordinate system. The lateral wind speed is given in the geodetic coordinate system. The vertical wind speed is given in the geodetic coordinate system. ； The wind speed signals obtained by Fourier transform of the three-dimensional wind speed in the geodetic coordinate system are shown in formulas (8a) to (8c): ； ； ； The transfer function of the closed-loop system matrix of the wind turbine generator is obtained as shown in Equation (9). The input is the three-dimensional wind speed (8a) to (8c), and the output is the dynamic response of the load in the complex domain as shown in Equation (10): ； in It is an imaginary number. Angular frequency; The dynamic response of the wind turbine generator load in the geodetic coordinate system is shown in Equation (10): 。 5. The rapid estimation method for fatigue load damage of a wind turbine generator set according to claim 4, characterized in that, Step S4 includes: In the frequency domain, the load response in the geodetic coordinate system is transformed into that in the rotating coordinate system. Equation (7) is expressed as an exponential function in the complex domain, as shown in Equation (12): ； in, , , ; Decompose matrix B into symmetric, cosine, and sine forms; Based on formulas (10) and (12), the dynamic response of the wind turbine load in the time-domain geodetic coordinate system is expressed. Dynamic response of wind turbine load converted to time-domain blade rotating coordinate system As shown in formula (13): ； in, for The result after Fourier transformation; Equation (13) is also transformed into the load dynamic response in the frequency domain by Fourier transformation, as shown in Equation (14): 。 6. The rapid estimation method for fatigue load damage of a wind turbine generator set according to claim 5, characterized in that, Step S5 includes: The expression for the power spectral density response in the geodetic coordinate system is shown in equation (11): ； The expression for the power spectral density response in the rotating coordinate system is shown in equation (15): 。 7. The rapid estimation method for fatigue load damage of a wind turbine generator set according to claim 6, characterized in that, Step S6 includes: Load fatigue damage based on power spectral density response is calculated using formulas (11) and (15), employing a combination of rainflow counting and empirical formulas; the SN curve is defined as shown in formula (16): ； in The amplitude of the input load signal. The number of loops. and All are constants; from the power spectral density and scalar obtained To estimate rainflow counts, and with scalars representing the complete power spectrum, The characteristic frequency is shown in formula (17): ； Define bandwidth coefficient as well as As shown in formula (18): ； Fatigue damage rate D is the damage rate of a narrow bandwidth process. Damage rate with amplitude counting The linear weighted result is shown in formula (19): ； in These are weighting coefficients. Represented as With second-order bandwidth coefficient index The product of functions is shown in formula (20): ； The narrow bandwidth impairment coefficient is calculated as shown in formula (21): ； in For gamma function, The critical stress level, The average uptake rate is shown in formula (22): ； Obtained by fitting empirical data and used as a bandwidth parameter and The function is shown in formula (23): 。 8. The rapid estimation method for fatigue load damage of a wind turbine generator set according to claim 7, characterized in that, Step S6 includes: To estimate the bending moment for which gravity distribution is the main source of fatigue damage, the gravity term needs to be added to the power spectral density response, as shown in equation (24): ； in, For the blade length, The mass distribution of the blade and its distribution along the blade span The curve function, This is the acceleration due to gravity.

9. A rapid estimation system for fatigue load damage of wind turbine generator sets, characterized in that, A method for rapidly estimating fatigue load damage of a power generator set according to any one of claims 1-8 includes: The wind speed acquisition module is used to collect the wind speed sampled by the wind turbine blades as they rotate. The dynamics modeling module is used to construct a linearized high-order aeroelastic dynamics model of a wind turbine generator. The control system modeling module is used to build the control system model of the wind turbine generator set; The closed-loop aero-servo-elastic system modeling module is used to integrate the linearized high-order aeroelastic dynamics model and the control system model to obtain the closed-loop aero-servo-elastic system model. The dynamic response calculation module takes the three-dimensional wind speed in the geodetic coordinate system as the input to the closed-loop aerodynamic servo elastic system model and outputs the dynamic response of the wind turbine load in the geodetic coordinate system; it also converts the dynamic response of the wind turbine load in the geodetic coordinate system into the dynamic response of the wind turbine load in the rotating coordinate system. The power spectral density response calculation module calculates the power spectral density response in both the geodetic and rotating coordinate systems based on the dynamic response of the wind turbine load in the geodetic coordinate system and the dynamic response of the wind turbine load in the rotating coordinate system. The load fatigue damage calculation module calculates the load fatigue damage of wind turbine generators based on the power spectral density response in both the geodetic coordinate system and the rotating coordinate system.

Images