Method for predicting service life of fan blade under wake flow influence

Abstract

The invention relates to the technical field of wind turbine generators, in particular to a method for predicting the service life of a fan blade under the influence of wake flow, and the method mainly comprises the steps: obtaining the coordinates of a wake flow region according to the position of a downstream fan and the azimuth angle of an initial blade, and calculating the service life of the fan blade through a Y-3DJGF model and the coordinates of the wake flow region. And the wake flow wind speed at the segment of the downstream blade is obtained, the downstream fan load is obtained through iterative calculation and integration of an improved BEM method, and the predicted service life result of the fan blade is obtained through a random load prediction model. By means of the method, downstream four-dimensional wake flow input is obtained through the Y-3DJGF model, the improved BEM method is used for calculating the load and power of a wind turbine wind wheel, after the load is obtained, the random load prediction model is used for blade life prediction, and a system, namely a series of processes from wind to wind load to fatigue life prediction is formed. And the blade life prediction accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbines, and in particular to a method for predicting the life of wind turbine blades under the influence of wake flow. Background Art

[0002] When a wind turbine performs yaw maneuvers to optimize wind energy capture, the resulting wake effect generates a series of complex random loads on the downstream wind turbine, posing additional challenges to the blade structure. These random loads originate from the instability of the wake, including fluctuations in wind speed, increased turbulence intensity, and sudden changes in wind direction. They are highly uncertain in time and space, and have a profound impact on the fatigue life of the blades.

[0003] In order to accurately predict the life of wind turbine blades under the influence of yaw wake, modern engineering practice usually combines knowledge from multiple disciplines and adopts a comprehensive analysis method. First, a detailed yaw wake model is constructed to capture the dynamic characteristics of the wake after the yaw operation, including its impact on the wind speed distribution and turbulent structure. Subsequently, the stress distribution of these random loads on the blade structure is analyzed using the theory of material mechanics and structural dynamics, and finally the fatigue accumulation effect is used to predict the life of the wind turbine blades under a random load. However, in the above process, only the relevant data of wind speed is often calculated, and the downstream blade load and power are not involved, so the predicted life value may be inaccurate. Summary of the invention

[0004] The purpose of the present invention is to provide a method for predicting the life of wind turbine blades under the influence of wake flow to solve the above problems in the prior art.

[0005] The implementation case of the present invention is achieved through the following technical solutions:

[0006] The present invention provides a method for predicting the life of a fan blade under the influence of wake flow, comprising:

[0007] S101: Obtain basic data and wind speed of the current wind turbine to obtain the incoming wind speed in front of the wind turbine;

[0008] S102: obtaining the yaw angle of the upstream wind turbine, the position of the downstream wind turbine and the initial blade azimuth, obtaining the wake area coordinates according to the downstream wind turbine position and the initial blade azimuth, and obtaining the wake wind speed at the blade segment through the Y-3DJGF model and the wake area coordinates;

[0009] S103: defining initial values ​​of the axial induction factor and the tangential induction factor, obtaining the airflow inclination angle, acquiring the local pitch angle, and obtaining the angle of attack through the airflow inclination angle and the local pitch;

[0010] S104: Obtaining the lift coefficient and drag coefficient of the fan blade, and calculating the tip loss coefficient and the hub loss coefficient;

[0011] S105: Calculate the axial force coefficient and the tangential force coefficient according to the attack angle, and update the axial induction factor and the tangential induction factor according to the axial force coefficient and the tangential force coefficient;

[0012] S106: Determine whether the updated axial induction factor and tangential induction factor have converged. If so, obtain the thrust and torque of the downstream fan load output data by integration. If not, repeat the steps S103 to S106 for the updated axial induction factor and tangential induction factor until the axial induction factor and tangential induction factor converge.

[0013] S107: Calculate the probability density function of the load data to build a prediction model, and obtain the predicted life result of the wind turbine blade through the prediction model.

[0014] Preferably, the wake wind speed at the blade segment is obtained by using the Y-3DJGF model and the wake area coordinates, including:

[0015]

[0016] Where u(x,y,z) is the wake wind speed at the blade segment, x,y,z are the coordinates of the wake area, and u hub is the incoming wind speed at hub height, and is the double Gaussian factor in the y direction, y min is the horizontal distance from the minimum value of the Gaussian function to the center, σ y is the characteristic wake velocity in the horizontal direction, y d is the offset of the middle line of the middle wake in the yaw wake model, α 0 is the wind speed and the rear yaw angle, C T is the thrust coefficient, β is the yaw angle of the upstream wind turbine, r 0 is the initial wake radius, k w is the comprehensive expansion coefficient, C is the 3DJGF model parameter, α is the wind shear index, Z hub is the wheel hub height.

[0017] Preferably, obtaining the airflow inclination angle and obtaining the angle of attack through the airflow inclination angle and the local pitch comprises:

[0018]

[0019] In the formula, is the airflow inclination angle, v ∞ is the incoming wind speed, a is the axial induction factor, a′ is the tangential induction factor, … is the angular velocity of the wind turbine, and θ is the local pitch angle.

[0020] Preferably, the calculation of the tip loss coefficient and the hub loss coefficient includes:

[0021] F=F t ×F h

[0022]

[0023] Where F is the loss coefficient, F t is the tip loss coefficient, F h is the hub loss coefficient, B is the number of blades, R is the radius of the wind wheel, and R hub is the hub radius, is the inflow angle.

[0024] Preferably, the calculating of the axial force coefficient and the tangential force coefficient by the angle of attack comprises:

[0025] C n =C l cosε+C d sinε

[0026] C t =C l sinε+C d cosε

[0027] In the formula, C n is the axial force coefficient, C t is the tangential force coefficient, C l is the lift coefficient, C d is the resistance coefficient.

[0028] Preferably, updating the axial induction factor and the tangential induction factor by the axial force coefficient and the tangential force coefficient comprises:

[0029] The tangential force is calculated and it is determined whether the tangential force is greater than or equal to the set threshold. If so, the axial induction factor is updated by the first formula; if not, the axial induction factor is updated by the second formula.

[0030] Preferably, the calculating the tangential force and determining whether the tangential force satisfies a set threshold comprises:

[0031]

[0032] The first formula includes:

[0033]

[0034] The second formula includes:

[0035]

[0036] Updated tangential induction factors include:

[0037]

[0038] In the formula, σ r is the solidity of the defined chord length.

[0039] Preferably, the step of obtaining the downstream fan load output data thrust and torque by integration includes:

[0040]

[0041] Where T is thrust, Q is torque, ρ is air density, V 0 is the incoming wind speed.

[0042] Preferably, calculating the load data probability density function to obtain an output value comprises:

[0043] S m N=C

[0044]

[0045] In the formula, μ(S) is the membership function, f(S) is the probability density function, S is the random stress, S 0 is the fatigue damage threshold S ε , S -1 is the fatigue limit, m is the SN curve coefficient, N is the number of hours of effective wind speed in a year, and E is the output value of the fatigue life prediction model.

[0046] Preferably, the step of outputting the output value to the prediction model to obtain the predicted life result of the wind turbine blade includes:

[0047]

[0048] Where Y is the predicted life of the wind turbine blades, and ω is the wind wheel speed.

[0049] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0050] The method provided by the present invention mainly includes obtaining the wake area coordinates according to the downstream fan position and the initial blade azimuth angle, obtaining the wake wind speed at the blade segment through the Y-3DJGF model and the wake area coordinates, defining the initial values ​​of the axial induction factor and the tangential induction factor; updating the axial induction factor and the tangential induction factor through iterative calculation, judging whether the updated axial induction factor and the tangential induction factor converge, obtaining the downstream fan load output data thrust and torque through integration, calculating the probability density function to obtain the construction of the random load prediction model, and obtaining the predicted life result of the fan blade through the downstream fan load. Through the above method, the downstream four-dimensional wake input is obtained through the Y-3DJGF model, and the wind turbine rotor and single blade load and power are calculated using the improved BEM method, while considering the downstream blade load and power, and after obtaining the load, the blade life prediction is performed, forming a system, that is, a series of processes from wind to wind load to fatigue life prediction, which improves the accuracy of blade life prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0052] Figure 1 The process of the present invention is shown in FIG. Figure 1 ;

[0053] Figure 2 The process of the present invention is shown in FIG. Figure 2 ;

[0054] Figure 3 It is a schematic diagram of the random stress change process and its probability description;

[0055] Figure 4 The yaw angle variation curve of the blade predicted life span with natural wind input at point a;

[0056] Figure 5 Predict the longitudinal position change curve of blade life for natural wind input;

[0057] Figure 6 The horizontal position change curve of the predicted life span of the blade under 5D natural wind input. DETAILED DESCRIPTION

[0058] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0059] The division of modules in this application is a logical division. There may be other division methods when implemented in actual applications. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0060] The modules or submodules described independently may be physically separated or not: they may be implemented by software or hardware, and some modules or submodules may be implemented by software, and the processor may call the software to implement the functions of these modules or submodules, and other modules or submodules may be implemented by hardware, such as by hardware circuits. In addition, some or all of the modules may be selected according to actual needs to achieve the purpose of the present application.

[0061] Please refer to Figure 1-2 The present invention provides a method for predicting the life of a fan blade under the influence of wake flow, comprising:

[0062] S101: Obtain basic data and wind speed of the current wind turbine to obtain the incoming wind speed in front of the wind turbine;

[0063] Because the incoming wind speed is not uniform, the wind speed at the center of the hub should not be simply used to replace the incoming wind speed. Instead, the following wind shear model should be used to simulate the incoming wind speed characteristics in front of the wind turbine.

[0064] The incoming wind speed in front of the wind turbine includes:

[0065]

[0066] In the formula, v ∞ is the incoming wind speed in front of the wind turbine, U ref At the reference height Z ref Measured wind speed, Z ref is the reference height for measuring wind speed, Z hub is the hub height, Z is the difference between the calculated wind speed position and the hub height, and α is the wind shear index, which is determined by the local atmospheric environment.

[0067] S102: obtaining the yaw angle of the upstream wind turbine, the position of the downstream wind turbine and the initial blade azimuth, obtaining the wake area coordinates according to the downstream wind turbine position and the initial blade azimuth, and obtaining the wake wind speed at the blade segment through the Y-3DJGF model and the wake area coordinates;

[0068] S103: defining initial values ​​of the axial induction factor and the tangential induction factor, obtaining the airflow inclination angle, acquiring the local pitch angle, and obtaining the angle of attack through the airflow inclination angle and the local pitch;

[0069] S104: Obtaining the lift coefficient and drag coefficient of the fan blade, and calculating the tip loss coefficient and the hub loss coefficient;

[0070] S105: Calculate the axial force coefficient and the tangential force coefficient according to the attack angle, and update the axial induction factor and the tangential induction factor according to the axial force coefficient and the tangential force coefficient;

[0071] S106: Determine whether the updated axial induction factor and tangential induction factor have converged. If so, obtain the thrust and torque of the downstream fan load output data by integration. If not, repeat the steps S103 to S106 for the updated axial induction factor and tangential induction factor until the axial induction factor and tangential induction factor converge.

[0072] S107: Calculate the probability density function of the load data to build a prediction model, and obtain the predicted life result of the wind turbine blade through the prediction model.

[0073] The method provided by the present invention mainly includes obtaining the wake area coordinates according to the downstream fan position and the initial blade azimuth angle, obtaining the wake wind speed at the blade segment through the Y-3DJGF model and the wake area coordinates, defining the initial values ​​of the axial induction factor and the tangential induction factor; judging whether the updated axial induction factor and the tangential induction factor converge, obtaining the downstream fan load output data through integration, outputting the output value to the random load prediction model, and obtaining the predicted life results of the fan blades. Through the above method, the upstream four-dimensional wake input is obtained through the Y-3DJGF model, and the improved BEM method is used to calculate the wind turbine rotor and single blade load and power, while considering the downstream blade load and power. After the load is obtained, the blade life prediction is performed, forming a system, that is, a series of processes from wind to wind load to fatigue life prediction, thereby improving the accuracy of blade life prediction.

[0074] In an exemplary embodiment of the present invention, the Y-3DJGF model is a yaw full wake model improved by using a double Gaussian function based on the 3DJGF wake model proposed by Gao, and performs well in both the far and near wake regions.

[0075] Through the Y-3DJGF model and the wake area coordinates, the wake wind speed at the blade segment is obtained, including:

[0076]

[0077] Where u(x,y,z) is the wake wind speed at the blade segment, x,y,z are the coordinates of the wake area, and u hub is the incoming wind speed at hub height, and is the double Gaussian factor in the y direction, y min is the horizontal distance from the minimum value of the Gaussian function to the center, σ y is the characteristic wake velocity in the horizontal direction, y d is the offset of the middle line of the middle wake in the yaw wake model, α 0 is the wind speed and the rear yaw angle, C T is the thrust coefficient, β is the yaw angle of the upstream wind turbine, r 0 is the initial wake radius, k w is the comprehensive expansion coefficient, C is the 3DJGF model parameter, α is the wind shear index, z hub is the wheel hub height.

[0078] Specifically, Where W 1 and W 2 is the empirical coefficient, k z is the 3DJGF model parameter, k y is the horizontal wake expansion coefficient.

[0079]

[0080] Where a is the axial induction factor, σ z is the characteristic wake width in the vertical direction, r z is the wake radius, and is the double Gaussian factor in the z direction.

[0081]

[0082] In the formula, z min is the vertical distance from the Gaussian minimum to the rotor center.

[0083] In an exemplary embodiment of the present invention, the axial induction factor a and the tangential induction factor a′ are defined according to Betz theory and the rotating wake model. The axial force acting on the blade element of the wind wheel surface is obtained from the momentum theory as follows:

[0084]

[0085] Where F is the axial force on the blade element of the wind wheel, v w is the wake wind speed, ρ is the air density, and r is the radius of the blade element at its current position.

[0086] The torque acting on the blade ring is:

[0087] dM=4πρΩv ∞ (1-a)a′r 3 dr

[0088] Where Ω is the angular velocity of the wind turbine, M is the torque acting on the blade ring, and a′ is the tangential induction factor.

[0089] The blade element theory assumes that the wind turbine blade is divided into many blade elements along the span direction, and does not consider the mutual interference between adjacent blade elements in the blade span direction. The force on the blade element is only determined by the aerodynamic performance of the blade element itself. The relative velocity of the airflow passing through the blade element is:

[0090]

[0091] Where w is the relative velocity of the airflow passing through the blade element.

[0092] The axial force acting on the N blade elements on the blade surface is:

[0093]

[0094] According to the tangential force, the wind wheel torque generated by the aerodynamic force on N blade elements is:

[0095]

[0096] In the formula, F n is the axial force acting on the N blade elements on the blade element surface, N is the number of blades, c is the geometric chord length, C n is the axial force coefficient, F t is the tip loss coefficient, C t is the tangential force coefficient.

[0097] The axial force and torque in the rotation of the wind rotor are calculated respectively using the blade element theory and momentum theory, and then they are combined and solved using the axial induction factor and tangential induction factor as the link. Finally, the required wind rotor torque, axial force and other data are calculated based on the values ​​after iterative solution of the induction factor.

[0098] Obtaining the airflow inclination angle and the angle of attack through the airflow inclination angle and the local pitch includes:

[0099]

[0100] In the formula, is the airflow inclination angle, v ∞ is the incoming wind speed, a is the axial induction factor, a′ is the tangential induction factor, Ω is the angular velocity of the wind turbine, and θ is the local pitch angle.

[0101] In an exemplary embodiment of the present invention, in the assumption of blade element theory, the gas only flows in layers along the axial direction and does not interfere with each other. In reality, there is a pressure difference between the two sides of the blade, and secondary flow will be generated along the span direction of the blade, so the output torque of the wind turbine will be reduced. Prandt l correction is performed for this assumption. The assumption of infinite number of blades is corrected for the blade element momentum theory, and a correction factor F is defined, which includes tip loss and hub loss, greatly improving the accuracy of the blade element momentum theory in wind turbine load calculation.

[0102] Specifically, the calculation of the tip loss coefficient and the hub loss coefficient includes:

[0103] F=F t ×F h

[0104]

[0105] Where F is the loss coefficient, F t is the tip loss coefficient, F h is the hub loss coefficient, B is the number of blades, R is the radius of the wind wheel, and R hub is the hub radius, φ is the inflow angle.

[0106] The iterative equation of the induction factor can be obtained as:

[0107]

[0108] In an exemplary embodiment of the present invention, the calculation of the axial force coefficient and the tangential force coefficient by the angle of attack includes:

[0109] C n =C l cosε+C d sinε

[0110] C t =C l sinε+C d cosε

[0111] In the formula, C n is the axial force coefficient, C t is the tangential force coefficient, C l is the lift coefficient, C d is the resistance coefficient.

[0112] In an exemplary embodiment of the present invention, updating the axial induction factor and the tangential induction factor by the axial force coefficient and the tangential force coefficient comprises:

[0113] The tangential force is calculated and it is determined whether the tangential force is greater than or equal to the set threshold. If so, the axial induction factor is updated by the first formula; if not, the axial induction factor is updated by the second formula.

[0114] When the wind turbine is under heavy load, it is in turbulent working state, and the classical momentum theory is no longer applicable. At present, there are several main methods for calculating the induction factor, including Glauert method, Buhl method, AeroDyn method and Jerson method. Glauert first combined the blade element theory with the momentum theorem, applied the blade element momentum theory to airfoil design and aerodynamic calculation, and corrected the value when the axial induction factor is large. The Glauert method is the method with the longest use time and the widest use range, but its simulation results are more sensitive to the influence of the tip loss coefficient F, and produce obvious numerical discontinuity when F<1, with large errors. The AeroDyn method corrects the problem that the results are sensitive to the tip loss coefficient F, and uses F as the basis for judging the high load of the wind turbine. The Buhl method refits the experimental data and proposes a new method for calculating the induction factor. The results are consistent with the axial induction factor calculated by the classical BEM method when a=0.4, which solves the problem of numerical instability when considering tip loss and has higher accuracy. The Jerson method defines the rotor induction factor a and the wake induction factor b, takes into account the general form of wake influence, proposes a new correction formula for low tip speed ratio, combines the advantages of Glauert correction, and makes corrections. In contrast, the Buhl method is more suitable for the wind turbine wake conditions at general wind speeds in this paper.

[0115] Specifically, calculating the tangential force and determining whether the tangential force meets the set threshold includes:

[0116]

[0117] The first formula includes:

[0118]

[0119] The second formula includes:

[0120]

[0121] Updated tangential induction factors include:

[0122]

[0123] In the formula, σ r is the solidity of the defined chord length.

[0124] In an exemplary embodiment of the present invention, the downstream fan load output data thrust and torque are obtained by integration, including:

[0125]

[0126] Where T is thrust, Q is torque, ρ is air density, V 0 is the incoming wind speed, and v ∞ Consistent.

[0127] An exemplary embodiment of the present invention is a fatigue life prediction model, which takes the fatigue life prediction of wind turbine blades as the research content. Based on the classic linear Miner cumulative damage theory, when the structure is subjected to S 1 , S 2 , …, S k When the complex load is composed of k different stresses, and it is assumed that the fatigue life corresponding to each specific stress S under the single action is N i , if the number of times each stress level acts in the load history is n 1 , n 2 ,…,n k , at this time, the cumulative damage of the structure caused by the random load history can be accurately expressed by the following method:

[0128]

[0129] Where D is the cumulative damage of the structure caused by random load history.

[0130] According to engineering practice, the uncertainty characteristics of stress caused by random loads are described by probability distribution functions, such as Figure 3 shown.

[0131] When the stress value can be described by a random variable with a probability density function of f(S), the stress value of each random load falls on S. i Nearby area The probability P(S i )for:

[0132] P(S i )=f(S i )ΔS

[0133] When the random load acts n times, the stress value falls on S i Nearby area The number of times ω(S i ) is:

[0134] ω(S i )=nP(S i )=nf(S i )ΔS

[0135] In the formula, ω(S i ) is the number of times, f(S i ) is the probability density function of the load composed of the ith stress, and ΔS is the infinitesimal element (small change).

[0136] According to the linear Miner cumulative damage law, the stress falls on S i Nearby area The cumulative amount of structural fatigue damage caused by the corresponding load can be expressed as:

[0137]

[0138] Where, ΔD i (S i ) is the cumulative amount of structural fatigue damage caused by the corresponding load.

[0139] When a component is subjected to a high stress level under actual working conditions, its initial fatigue limit (S d ) is no longer suitable as a limit for judging whether subsequent stress can cause fatigue damage to the material. Under actual working conditions, its initial fatigue limit (S d ) is less applicable as the limit for subsequent stress damage judgment. Therefore, the fatigue damage threshold (S ε ), set it slightly lower than (S d ). The specific analysis is as follows:

[0140] Low stress area: The stress level is lower than (S ε ), the component damage is negligible and the fuzzy damage degree λ=0.

[0141] Fuzzy damage area: stress between (S ε ) and (S d ), there is uncertainty in the damage, and λ is between 0 and 1, which is determined by the membership function, forming a damage fuzzy region.

[0142] High stress area: stress exceeds (S d ), damage is certain to occur, λ=1.

[0143] The fatigue damage caused by random cyclic loads to wind turbine blades is cumulative, so a relatively large membership function is selected, as follows:

[0144]

[0145] Based on a large amount of fatigue test data, the fuzzy uncertainty of damage exhibited by the material at different stress levels is fully considered, and the fatigue damage threshold is set to S ε =85%S dFor wind turbine blades, the stress range that causes fatigue damage is clearly defined, and its lower limit is the fatigue damage threshold value S ε , the upper limit is the fatigue limit S -1 , and the tensile strength of the material is marked as S b .

[0146] Under random load, it is assumed that the component is subjected to k+x load cycles, where the stress caused by the load of γk times is located at S ε and fatigue limit, that is, in the fuzzy area that may cause damage; and the stress under the remaining (1-γ)k loads exceeds the fatigue limit. The stress caused by the x-times load is located in S ε The following is not considered. The fatigue load damage of the component under this random load history is:

[0147]

[0148] The relationship between the stress S and the number N that a wind turbine blade is subjected to during its service is called the SN curve.

[0149] S m N=C

[0150] When the cumulative damage of fatigue failure of the structure is D = 1, the fatigue life of the structure under random load is E, and the stress probability distribution is divided into k intervals. It is assumed that the number of sub-intervals in the fuzzy area is γk, and the fuzzy damage degree is 0<γ<1. When k→∞, it can be obtained:

[0151]

[0152] Therefore, the fatigue life prediction model of the blade under the action of random stress S can be obtained from the above formula:

[0153]

[0154] In the formula, μ(S) is the membership function, f(S) is the probability density function, S is the random stress, S 0 is the fatigue damage threshold S ε , S -1 is the fatigue limit, m is the SN curve coefficient, N is the number of hours of effective wind speed in a year, and E is the output value of the fatigue life prediction model.

[0155] In an exemplary embodiment of the present invention, the predicted life result of the wind turbine blade is obtained by using the prediction model, including:

[0156]

[0157] Where Y is the predicted life of the wind turbine blades, and ω is the wind wheel speed.

[0158] The present invention also provides a specific implementation example.

[0159] like Figure 4 As shown in Figure 2, when the downstream horizontal position is the same (point a without yaw full wake), the unit is at different yaw angles (β = 0°, β = 5°, β = 10°, β = 15°), and the unit life prediction comparison with the change of longitudinal position is shown in Figure 2. Figure 4 As shown in the figure, compared with the life prediction at β = 15°, the life loss of the wind rotor gradually increases with the decrease of the yaw angle. The average relative losses of the life predictions of β = 10°, β = 5°, and β = 0° are 9.32%, 16.30%, and 24.77%, respectively. For every 5° increase in the yaw angle, the predicted life of the downstream wind turbine increases by about 7.39%. The predicted life increases by 5.16% when the yaw angle is compared with 0°, and by 9.32% when the yaw angle is compared with 15°. The relative increase is more obvious when the yaw angle is large.

[0160] like Figure 5 As shown in Figure 2, when the downstream horizontal position is the same (point a without yaw full wake), the unit is at different yaw angle positions (X = 5D, X = 6D, X = 7D, X = 8D), and the unit life prediction comparison with the yaw angle change is shown in Figure 2. Figure 5 As shown in the figure, it can be seen that for every 1D increase in the longitudinal position, the predicted life of the downstream wind turbine increases by about 8.93%.

[0161] like Figure 6 As shown in Figure 2, when the downstream radial position is the same (x = 5D), the unit is at different horizontal positions (point a before the full wake yaw, point b after the yaw; point c before the full wake yaw, point d after the yaw), and the unit life prediction comparison with the yaw angle change is shown in Figure 2. Figure 6 As shown in the figure. Because during the yaw action, the center line of the yaw wake is offset (points a and c do not move, and points b and d change with the yaw angle), the center line gradually approaches point c and moves away from point a during the action. Therefore, point a is affected by the wake and the predicted life is reduced, and the predicted life is increased. Point c is affected by the wake and the predicted life is reduced. The influence of points b and d is because the point position changes with the yaw angle and is always affected by the same position of the wake.

[0162] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0163] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0164] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for predicting the life of a wind turbine blade under the influence of wake, characterized in that: include: S101: Obtain basic data and wind speed of the current wind turbine to obtain the incoming wind speed in front of the wind turbine; S102: obtaining the yaw angle of the upstream wind turbine, the position of the downstream wind turbine and the initial blade azimuth, obtaining the wake area coordinates according to the downstream wind turbine position and the initial blade azimuth, and obtaining the wake wind speed at the blade segment through the Y-3DJGF model and the wake area coordinates; S103: defining initial values ​​of the axial induction factor and the tangential induction factor, obtaining the airflow inclination angle, acquiring the local pitch angle, and obtaining the angle of attack through the airflow inclination angle and the local pitch; S104: Obtaining the lift coefficient and drag coefficient of the fan blade, and calculating the tip loss coefficient and the hub loss coefficient; S105: Calculate the axial force coefficient and the tangential force coefficient according to the attack angle, and update the axial induction factor and the tangential induction factor according to the axial force coefficient and the tangential force coefficient; S106: Determine whether the updated axial induction factor and tangential induction factor have converged. If so, obtain the downstream fan load output data thrust and torque by integration. If not, re-perform steps S103 to S106 on the updated axial induction factor and tangential induction factor until the axial induction factor and tangential induction factor converge. S107: Calculate the probability density function of the load data to build a prediction model, and obtain the predicted life result of the wind turbine blade through the prediction model.

2. The method for predicting the life of a wind turbine blade under the influence of wake according to claim 1 is characterized in that: The wake wind speed at the blade segment obtained by the Y-3DJGF model and the wake area coordinates includes: Where u(x,y,z) is the wake wind speed at the blade segment, x,y,z are the coordinates of the wake area, and u hub is the incoming wind speed at hub height, and is the double Gaussian factor in the y direction, y min is the horizontal distance from the minimum value of the Gaussian function to the center, σ y is the characteristic wake velocity in the horizontal direction, y d is the offset of the middle line of the middle wake in the yaw wake model, α0 is the wind speed and the rear yaw angle, C T is the thrust coefficient, β is the upstream wind turbine yaw angle, r0 is the initial wake radius, k w is the comprehensive expansion coefficient, C is the 3DJGF model parameter, α is the wind shear index, z hub is the wheel hub height.

3. The method for predicting the life of a wind turbine blade under the influence of wake according to claim 2 is characterized in that: The obtaining of the airflow inclination angle and the angle of attack through the airflow inclination angle and the local pitch comprises: In the formula, is the airflow inclination angle, v ∞ is the incoming wind speed, a is the axial induction factor, a ′ is the tangential induction factor, Ω is the angular velocity of the wind turbine, and θ is the local pitch angle.

4. The method for predicting the life of a wind turbine blade under the influence of wake according to claim 3 is characterized in that: The calculation of the tip loss coefficient and the hub loss coefficient includes: F=F t ×F h Where F is the loss coefficient, F t is the tip loss coefficient, F h is the hub loss coefficient, B is the number of blades, R is the radius of the wind wheel, and R hub is the hub radius, is the inflow angle.

5. The method for predicting the life of a wind turbine blade under the influence of wake according to claim 4, characterized in that: The calculation of the axial force coefficient and the tangential force coefficient by the angle of attack includes: C n =C l cosε+C d sinε C t =C l sinε+C d cosε In the formula, C n is the axial force coefficient, C t is the tangential force coefficient, C l is the lift coefficient, C d is the resistance coefficient.

6. A method for predicting the life of a wind turbine blade under the influence of wake according to claim 5, characterized in that: The updating of the axial induction factor and the tangential induction factor by the axial force coefficient and the tangential force coefficient comprises: The tangential force is calculated and it is determined whether the tangential force is greater than or equal to the set threshold. If so, the axial induction factor is updated by the first formula; if not, the axial induction factor is updated by the second formula.

7. A method for predicting the life of a wind turbine blade under the influence of wake according to claim 6, characterized in that: The calculating of the tangential force and judging whether the tangential force satisfies a set threshold value includes: The first formula includes: The second formula includes: Updated tangential induction factors include: In the formula, σ r is the solidity of the defined chord length.

8. The method for predicting the life of a wind turbine blade under the influence of wake according to claim 7, characterized in that: The downstream fan load output data thrust and torque obtained by integration include: Where T is thrust, Q is torque, ρ is air density, and V0 is incoming wind speed.

9. A method for predicting the life of a wind turbine blade under the influence of wake according to claim 8, characterized in that: The calculating the load data probability density function to obtain the output value comprises: S m N=C The prediction model is: In the formula, μ(S) is the membership function, f(S) is the probability density function, S is the random stress, and S0 is the fatigue damage threshold S ε , S -1 is the fatigue limit, N is the number of hours of effective wind speed in a year, and E is the output value of the fatigue life prediction model.

10. A method for predicting the life of a wind turbine blade under the influence of wake according to claim 9, characterized in that: The predicted life results of the fan blades obtained by the prediction model include: Where Y is the predicted life of the wind turbine blades, and ω is the wind wheel speed.

Images