Fan blade fatigue life evaluation method based on data driving

Abstract

The invention relates to the technical field of wind driven generators, in particular to a wind turbine blade fatigue life evaluation method based on data driving, which comprises the following steps: acquiring blade original data of a wind driven generator; performing feature extraction on the blade original data of the wind driven generator based on the blade fatigue prediction model to obtain a blade fatigue prediction result of the wind driven generator, and solving a stress state accumulated damage value in the blade original data of the wind driven generator; and calculating the final fatigue of the blade of the wind driven generator by combining a blade fatigue prediction result, a stress state accumulated damage value and an environment correction factor of the wind driven generator and taking error minimization of the blade fatigue prediction result and the stress state accumulated damage value as an optimization target. And completing the service life evaluation of the wind driven generator blade based on the final fatigue of the wind driven generator blade. According to the method, accurate matching of fatigue prediction and actually measured damage values can be realized, and a reliable decision basis is provided for maintenance of wind power equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbines, and in particular to a data-driven method for evaluating the fatigue life of wind turbine blades. Background Art

[0002] As a key component of wind energy conversion, wind turbine blades are subjected to complex load environments during long-term operation, including the combined effects of multiple alternating stresses such as wind load, gravity load, and centrifugal force. These complex loads cause fatigue damage to the blades, which may eventually lead to structural failure. Accurately assessing the fatigue life of blades is of great significance for ensuring the safe operation of wind turbines and formulating maintenance strategies. At present, traditional blade fatigue life assessment methods have obvious limitations in practical applications. On the one hand, traditional methods overly rely on SN curves and Miner's linear cumulative damage theory. This simplified theoretical model cannot accurately describe the nonlinear fatigue behavior of blade materials under complex working conditions. Especially under multi-axial stress states and variable amplitude loads, the linear cumulative theory often underestimates the actual degree of fatigue damage. On the other hand, when processing monitoring data, traditional methods usually use a single stress analysis method, ignoring the influence of environmental factors such as temperature changes and wind speed fluctuations on the fatigue damage process, and lack the ability to deeply mine and extract features from multi-source sensor data, resulting in a large amount of monitoring data containing blade health status information not being fully utilized, which ultimately affects the accuracy and reliability of fatigue life assessment. Summary of the invention

[0003] The purpose of the present invention is to provide a data-driven method for evaluating the fatigue life of wind turbine blades. The method uses a deep learning model to extract fatigue characteristics of multi-source monitoring data of blades, combines it with a stress state cumulative damage calculation method, accurately evaluates the fatigue state of blades, and introduces environmental correction factors of temperature and wind speed, thereby improving the adaptability of the evaluation results in complex environments, and being able to achieve an accurate match between fatigue prediction and measured damage values, so that the evaluation results are more in line with actual working conditions and provide a reliable decision-making basis for wind power equipment maintenance.

[0004] The present invention is achieved through the following technical solutions:

[0005] A data-driven method for assessing fatigue life of wind turbine blades includes the following steps:

[0006] Obtain the original data of wind turbine blades;

[0007] Based on the blade fatigue prediction model, the original data of the wind turbine blade is feature extracted to obtain the wind turbine blade fatigue prediction result, and the cumulative damage value of the stress state in the original data of the wind turbine blade is solved;

[0008] Combining the blade fatigue prediction results, the cumulative damage value of the stress state and the environmental correction factor of the wind turbine blade, and taking minimizing the error between the blade fatigue prediction results and the cumulative damage value of the stress state as the optimization goal, the final fatigue of the wind turbine blade is calculated, and the life assessment of the wind turbine blade is completed based on the final fatigue of the wind turbine blade.

[0009] Optionally, the obtaining of original blade data of the wind turbine generator specifically includes:

[0010] Collect blade strain data of wind turbines;

[0011] Collect blade load data of wind turbines;

[0012] Collect blade environmental data of wind turbines;

[0013] A blade input vector of the wind turbine is constructed based on blade strain data, blade load data and blade environment data of the wind turbine, and data of the blade input vector of the wind turbine is standardized to serve as input of a blade fatigue prediction model.

[0014] Optionally, the blade fatigue prediction model is built by a convolutional neural network model, which is specifically composed of an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a flattening layer, a fully connected layer and an output layer.

[0015] Optionally, the calculation formula of the input layer is:

[0016]

[0017] Where X(t) is the blade input vector of the wind turbine, and L is the sequence length;

[0018] The calculation formula of the first convolutional layer is:

[0019]

[0020] Where C1(t) is the feature map output of the first convolutional layer, k1 is the time dimension kernel of the first convolutional kernel, b1 is the bias term of the first convolutional layer, K1(i,j) is the convolution kernel weight of the jth channel at the i-th time of the first convolutional layer, i is the sliding index of the time dimension, j is the index of the feature dimension, and X(ti) is the wind turbine blade input vector at time ti;

[0021] The calculation formula of the first pooling layer is:

[0022] P1(t)=max{C1(t:t+p1)}

[0023] Among them, P1(t) is the feature map output of the first pooling layer, t is the current time point, p1 is the window size of the first pooling layer, and C1(t:t+p1) is the value of the first convolutional layer output from t to t+p1;

[0024] The calculation formula of the second convolutional layer is:

[0025]

[0026] Among them, C2(t) is the feature map output of the second convolutional layer, k2 is the time dimension kernel of the second convolutional layer, d1 is the number of output channels of the first convolutional layer, K2(i,j) is the convolution kernel weight of the jth channel at the i-th time of the second convolutional layer, and b2 is the bias term of the second convolutional layer;

[0027] The calculation formula of the second pooling layer is:

[0028] P2(t)=max{C2(t:t+p2)}

[0029] Among them, P2(t) is the feature map output of the second pooling layer, and C2(t:t+p2) is the value of the second convolutional layer output from time t to t+p2;

[0030] The calculation formula of the flattened layer is:

[0031] F(t)=flatten(P2(t))

[0032] Where F(t) is the output of the flattening layer;

[0033] The calculation formula of the fully connected layer is:

[0034] FC(t)=σ(F(t)W+b)

[0035] Among them, FC(t) is the prediction result of blade fatigue of wind turbine, σ is the Relu activation function, W is the weight matrix, b is the bias vector, d in is the dimension of the flattened input.

[0036] Optionally, the specific process of solving the stress state cumulative damage value in the original data of the wind turbine blade is as follows:

[0037] Collect blade strain data and blade load data of wind turbines;

[0038] Calculate the principal stress, shear stress and equivalent stress of the wind turbine blades according to the blade strain data and blade load data of the wind turbine;

[0039] The cycle range of equivalent stress is determined based on the equivalent stress of wind turbine blades, and effective cycle statistics are performed on the stress cycle range of wind turbine blades to solve the cumulative damage value of stress state.

[0040] Optionally, the specific calculation formula for solving the stress state cumulative damage value in the original data of the wind turbine blade is:

[0041]

[0042]

[0043] Δσ(t)=max(σ eq (t))-min(σ eq (t))

[0044]

[0045]

[0046] Among them, σ1(t) and σ2(t) are the first and second principal stresses of the blade, respectively, and τ xy (t) is the blade shear stress, E is the elastic modulus of the blade material, ν is the Poisson's ratio of the blade material, ε x (t), ε y (t) are the spanwise and chordwise strains of the blade, F x (t), F y (t) are the spanwise and chordwise forces of the blade, h is the blade section height, I is the blade section inertia moment, G is the shear modulus of the blade material, γ xy (t) is the blade shear strain, M z (t) is the blade torsional bending moment, r is the blade section radius, J is the blade section polar moment of inertia, σ eq (t) is the blade equivalent stress, Δσ(t) is the cycle range of the blade equivalent stress, N(t) is the effective number of blade stress cycles, σ th is the fatigue limit threshold of the blade, I() is the effective cycle judgment function of the blade stress, k is the number of divisions of the blade stress level, D(t) is the cumulative damage value of the blade at the current stress level, which is characterized by the cumulative damage value of the stress state, C is the fatigue strength coefficient of the blade material, and m is the negative slope of the SN curve of the blade material.

[0047] Optionally, the final fatigue degree of the blades of the wind turbine is calculated by taking the minimization of the error between the blade fatigue prediction result and the cumulative damage value of the stress state as the optimization goal, and the calculation formula is:

[0048]

[0049]

[0050] Among them, Fatigue(t) is the final fatigue degree of the blade at time t, v(t) is the real-time wind speed at time t, v0 is the reference wind speed, T0 is the reference temperature, T(t) is the blade surface temperature at time t, T is the total length of the time series, min(L) is, Fatigue actual (t) is the measured fatigue degree of the blade at time t, D actual (t) is the measured cumulative damage value.

[0051] Optionally, the constraints of the optimization objective are:

[0052]

[0053] Among them, σ y is the yield strength of the blade material, σ u is the ultimate strength of the blade material, s f is the safety factor of the blade, D cr is the critical fatigue damage value, v cut-in is the wind turbine cut-in wind speed, v cut-out is the fan cut-out wind speed, T min is the minimum allowable operating temperature, T max is the maximum allowable operating temperature.

[0054] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0055] The present invention adopts a deep learning model to extract the fatigue characteristics of multi-source monitoring data of blades, combines it with the stress state cumulative damage calculation method to accurately evaluate the fatigue state of blades, and introduces environmental correction factors of temperature and wind speed, which improves the adaptability of the evaluation results in complex environments. It can achieve accurate matching of fatigue prediction and measured damage value, make the evaluation results more in line with actual working conditions, and provide a reliable decision-making basis for wind power equipment maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic flow chart of the data-driven wind turbine blade fatigue life assessment method provided by the present invention. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the drawings in the present invention. Obviously, what is described is only a part of the present invention, not all of it. Generally, the components of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0058] like Figure 1As shown, the present invention provides one embodiment: a data-driven method for evaluating the fatigue life of a wind turbine blade, the method comprising the following steps:

[0059] Obtain the original data of wind turbine blades;

[0060] Based on the blade fatigue prediction model, the original data of the wind turbine blade is feature extracted to obtain the wind turbine blade fatigue prediction result, and the cumulative damage value of the stress state in the original data of the wind turbine blade is solved;

[0061] Combining the blade fatigue prediction results, the cumulative damage value of the stress state and the environmental correction factor of the wind turbine blade, and taking minimizing the error between the blade fatigue prediction results and the cumulative damage value of the stress state as the optimization goal, the final fatigue of the wind turbine blade is calculated, and the life assessment of the wind turbine blade is completed based on the final fatigue of the wind turbine blade.

[0062] In a specific implementation of this embodiment, the obtaining of the original data of the blades of the wind turbine generator specifically includes:

[0063] Collect blade strain data of wind turbines;

[0064] Collect blade load data of wind turbines;

[0065] Collect blade environmental data of wind turbines;

[0066] A blade input vector of the wind turbine is constructed based on blade strain data, blade load data and blade environment data of the wind turbine, and data of the blade input vector of the wind turbine is standardized to serve as input of a blade fatigue prediction model.

[0067] Specifically, the blade strain data of wind turbines include: ε x (t), ε y (t) and γ xy (t), the blade load data of wind turbines include: F x (t), F y (t) and M z (t); The blade environmental data of the wind turbine include: v(t), T(t) and ρ(t).

[0068] In this embodiment, the blade fatigue prediction model is built by a convolutional neural network model, which is specifically composed of an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a flattening layer, a fully connected layer and an output layer.

[0069] Optionally, the calculation formula of the input layer is:

[0070]

[0071] Where X(t) is the blade input vector of the wind turbine, and L is the sequence length;

[0072] The calculation formula of the first convolutional layer is:

[0073]

[0074] Where C1(t) is the feature map output of the first convolutional layer, k1 is the time dimension kernel of the first convolutional kernel, b1 is the bias term of the first convolutional layer, K1(i,j) is the convolution kernel weight of the jth channel at the i-th time of the first convolutional layer, i is the sliding index of the time dimension, j is the index of the feature dimension, and X(ti) is the wind turbine blade input vector at time ti;

[0075] The calculation formula of the first pooling layer is:

[0076] P1(t)=max{C1(t:t+p1)}

[0077] Among them, P1(t) is the feature map output of the first pooling layer, t is the current time point, p1 is the window size of the first pooling layer, and C1(t:t+p1) is the value of the first convolutional layer output from t to t+p1;

[0078] The calculation formula of the second convolutional layer is:

[0079]

[0080] Among them, C2(t) is the feature map output of the second convolutional layer, k2 is the time dimension kernel of the second convolutional layer, d1 is the number of output channels of the first convolutional layer, K2(i,j) is the convolution kernel weight of the jth channel at the i-th time of the second convolutional layer, and b2 is the bias term of the second convolutional layer;

[0081] The calculation formula of the second pooling layer is:

[0082] P2(t)=max{C2(t:t+p2)}

[0083] Among them, P2(t) is the feature map output of the second pooling layer, and C2(t:t+p2) is the value of the second convolutional layer output from time t to t+p2;

[0084] The calculation formula of the flattened layer is:

[0085] F(t)=flatten(P2(t))

[0086] Where F(t) is the output of the flattening layer;

[0087] The calculation formula of the fully connected layer is:

[0088] FC(t)=σ(F(t)W+b)

[0089] Among them, FC(t) is the prediction result of blade fatigue of wind turbine, σ is the Relu activation function, W is the weight matrix, b is the bias vector, d in is the dimension of the flattened input.

[0090] In the implementation, based on the original monitoring data of wind turbine blades, including strain data, load data and environmental data, normalization preprocessing is first performed to standardize the data to a uniform scale. Subsequently, the sliding time window method is used to segment the data with a window length L and a step size s to form a time series sample set, and feature extraction is performed through a two-level convolution-pooling structure: the first level uses 32 3×9 convolution kernels for feature extraction, and after the tanh activation function, the dimension is reduced by a 2×1 maximum pooling layer; the second level uses 64 3×32 convolution kernels to further extract high-level features, and also undergoes activation and pooling operations. Finally, the feature map is flattened and mapped to a 128-dimensional feature space through a fully connected layer to obtain the blade fatigue feature representation FC(t). The entire process realizes the automatic extraction of fatigue features from original multi-source data, providing a reliable feature basis for subsequent fatigue calculation.

[0091] In this embodiment, the specific process of solving the stress state cumulative damage value in the original data of the wind turbine blade is as follows:

[0092] Collect blade strain data and blade load data of wind turbines;

[0093] Calculate the principal stress, shear stress and equivalent stress of the wind turbine blades according to the blade strain data and blade load data of the wind turbine;

[0094] The cycle range of equivalent stress is determined based on the equivalent stress of wind turbine blades, and effective cycle statistics are performed on the stress cycle range of wind turbine blades to solve the cumulative damage value of stress state.

[0095] The specific calculation formula for solving the stress state cumulative damage value in the original data of the wind turbine blade is:

[0096]

[0097]

[0098] Δσ(t)=max(σ eq (t))-min(σ eq (t))

[0099]

[0100] Among them, σ1(t) and σ2(t) are the first and second principal stresses of the blade, respectively, and τ xy (t) is the blade shear stress, E is the elastic modulus of the blade material, ν is the Poisson's ratio of the blade material, ε x (t), ε y (t) are the spanwise and chordwise strains of the blade, F x (t), F y (t) are the spanwise and chordwise forces of the blade, h is the blade section height, I is the blade section inertia moment, G is the shear modulus of the blade material, γ xy (t) is the blade shear strain, M z (t) is the blade torsional bending moment, r is the blade section radius, J is the blade section polar moment of inertia, σ eq (t) is the blade equivalent stress, Δσ(t) is the cycle range of the blade equivalent stress, N(t) is the effective number of blade stress cycles, σ th is the fatigue limit threshold of the blade, I() is the effective cycle judgment function of the blade stress, k is the number of divisions of the blade stress level, D(t) is the cumulative damage value of the blade at the current stress level, which is characterized by the cumulative damage value of the stress state, C is the fatigue strength coefficient of the blade material, and m is the negative slope of the SN curve of the blade material.

[0101] In the implementation, based on the blade strain data and blade load data of the wind turbine, the strain-stress relationship is established using Hooke's law, and combined with the geometric characteristics of the blade section, the principal stresses σ1(t), σ2(t) and shear stress τ are calculated. xy (t). Then the equivalent stress σ is calculated based on the von Mises criterion eq (t), the rain flow counting method is used to count the stress cycles, and the stress amplitude Δσ(t) and the effective number of cycles N(t) are obtained. On this basis, the Miner linear cumulative damage theory is applied, combined with the material SN curve characteristics, namely the slope m and the material constant C, to calculate the cumulative damage value D(t) at each moment. The cumulative damage calculation method based on continuous damage mechanics effectively combines strain-stress analysis with fatigue theory to achieve quantitative evaluation of blade fatigue state.

[0102] Optionally, the final fatigue degree of the blades of the wind turbine is calculated by taking the minimization of the error between the blade fatigue prediction result and the cumulative damage value of the stress state as the optimization goal, and the calculation formula is:

[0103]

[0104]

[0105] Among them, Fatigue(t) is the final fatigue degree of the blade at time t, v(t) is the real-time wind speed at time t, v0 is the reference wind speed, T0 is the reference temperature, T(t) is the blade surface temperature at time t, T is the total length of the time series, min(L) is, Fatigue actual (t) is the measured fatigue degree of the blade at time t, D actual (t) is the measured cumulative damage value.

[0106] In the specific implementation, this embodiment constructs an error minimization objective function, which includes the absolute error term between the predicted fatigue and the actual fatigue, and the logarithmic error term between the cumulative damage value and the measured damage value. In the optimization solution process, by introducing the blade stress safety factor constraint, fatigue damage critical value constraint, wind speed operating range constraint and temperature working range constraint, it is ensured that the optimization results meet the actual engineering needs. The fatigue calculation formula finally obtained comprehensively considers the influence of model output, cumulative damage value, wind speed correction factor and temperature correction factor. Based on the calculated final fatigue of the blade, the damage increment per unit time ΔD=k'·Fatigue(t) is calculated, where k' is the proportional coefficient. According to the known critical damage value D of the blade material cr And the current cumulative damage value, get the remaining blade life

[0107]

[0108] More specifically, the constraints of the optimization objective are:

[0109]

[0110] Among them, σ y is the yield strength of the blade material, σ u is the ultimate strength of the blade material, s f is the safety factor of the blade, D cr is the critical fatigue damage value, v cut-in is the wind turbine cut-in wind speed, v cut-out is the fan cut-out wind speed, T min is the minimum allowable operating temperature, T max is the maximum allowable operating temperature.

[0111] The above is only the preferred embodiment of the present invention and is 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 data-driven method for assessing the fatigue life of wind turbine blades, characterized in that: The steps of the method include: Obtain the original data of wind turbine blades; Based on the blade fatigue prediction model, the original data of the wind turbine blade is feature extracted to obtain the wind turbine blade fatigue prediction result, and the cumulative damage value of the stress state in the original data of the wind turbine blade is solved; Combining the blade fatigue prediction results, the cumulative damage value of the stress state and the environmental correction factor of the wind turbine blade, and taking minimizing the error between the blade fatigue prediction results and the cumulative damage value of the stress state as the optimization goal, the final fatigue of the wind turbine blade is calculated, and the life assessment of the wind turbine blade is completed based on the final fatigue of the wind turbine blade.

2. The data-driven wind turbine blade fatigue life assessment method according to claim 1 is characterized in that: The obtaining of the original data of the blades of the wind turbine generator specifically includes: Collect blade strain data of wind turbines; Collect blade load data of wind turbines; Collect blade environmental data of wind turbines; A blade input vector of the wind turbine is constructed based on blade strain data, blade load data and blade environment data of the wind turbine, and data of the blade input vector of the wind turbine is standardized to serve as input of a blade fatigue prediction model.

3. The data-driven wind turbine blade fatigue life assessment method according to claim 2 is characterized in that: The blade fatigue prediction model is built by a convolutional neural network model, which is specifically composed of an input layer, a first convolutional layer, a first pooling layer, a second convolutional layer, a second pooling layer, a flattening layer, a fully connected layer and an output layer.

4. The data-driven wind turbine blade fatigue life assessment method according to claim 3 is characterized in that: The calculation formula of the input layer is: Where X(t) is the blade input vector of the wind turbine, and L is the sequence length; The calculation formula of the first convolutional layer is: Where C1(t) is the feature map output of the first convolutional layer, k1 is the time dimension kernel of the first convolutional kernel, b1 is the bias term of the first convolutional layer, K1(i,j) is the convolution kernel weight of the jth channel at the i-th time of the first convolutional layer, i is the sliding index of the time dimension, j is the index of the feature dimension, and X(ti) is the wind turbine blade input vector at time ti; The calculation formula of the first pooling layer is: P1(t)=max{C1(t:t+p1)} Among them, P1(t) is the feature map output of the first pooling layer, t is the current time point, p1 is the window size of the first pooling layer, and C1(t:t+p1) is the value of the first convolutional layer output from t to t+p1; The calculation formula of the second convolutional layer is: Among them, C2(t) is the feature map output of the second convolutional layer, k2 is the time dimension kernel of the second convolutional layer, d1 is the number of output channels of the first convolutional layer, K2(i,j) is the convolution kernel weight of the jth channel at the i-th time of the second convolutional layer, and b2 is the bias term of the second convolutional layer; The calculation formula of the second pooling layer is: P2(t)=max{C2(t:t+p2)} Among them, P2(t) is the feature map output of the second pooling layer, and C2(t:t+p2) is the value of the second convolutional layer output from time t to t+p2; The calculation formula of the flattened layer is: F(t)=flatten(P2(t)) Where F(t) is the output of the flattening layer; The calculation formula of the fully connected layer is: FC(t)=σ(F(t)W+b) Among them, FC(t) is the prediction result of blade fatigue of wind turbine, σ is the Re lu activation function, W is the weight matrix, b is the bias vector, d in is the dimension of the flattened input.

5. The data-driven wind turbine blade fatigue life assessment method according to claim 4 is characterized in that: The specific process of solving the stress state cumulative damage value in the original data of the wind turbine blade is as follows: Collect blade strain data and blade load data of wind turbines; Calculate the principal stress, shear stress and equivalent stress of the wind turbine blades according to the blade strain data and blade load data of the wind turbine; The cycle range of equivalent stress is determined based on the equivalent stress of wind turbine blades, and effective cycle statistics are performed on the stress cycle range of wind turbine blades to solve the cumulative damage value of stress state.

6. The data-driven method for evaluating the fatigue life of wind turbine blades according to claim 5, characterized in that: The specific calculation formula for solving the stress state cumulative damage value in the original data of the wind turbine blade is: Δσ(t)=max(σ eq (t))-min(σ eq (t)) Among them, σ1(t) and σ2(t) are the first and second principal stresses of the blade, respectively, and τ xy (t) is the blade shear stress, E is the elastic modulus of the blade material, ν is the Poisson's ratio of the blade material, ε x (t), ε y (t) are the spanwise and chordwise strains of the blade, F x (t), F y (t) are the blade span and chord forces, h is the blade section height, I is the blade section inertia moment, G is the blade material shear modulus, γ xy (t) is the blade shear strain, M z (t) is the blade torsional bending moment, r is the blade section radius, J is the blade section polar moment of inertia, σ eq (t) is the blade equivalent stress, Δσ(t) is the cycle range of the blade equivalent stress, N(t) is the effective number of blade stress cycles, σ th is the fatigue limit threshold of the blade, I() is the effective cycle judgment function of the blade stress, k is the number of divisions of the blade stress level, D(t) is the cumulative damage value of the blade at the current stress level, which is characterized by the cumulative damage value of the stress state, C is the fatigue strength coefficient of the blade material, and m is the negative slope of the SN curve of the blade material.

7. The data-driven wind turbine blade fatigue life assessment method according to claim 6, characterized in that: The optimization goal is to minimize the error between the blade fatigue prediction result and the cumulative damage value of the stress state, and calculate the final fatigue of the blade of the wind turbine generator, and the calculation formula is: Among them, Fatigue(t) is the final fatigue degree of the blade at time t, v(t) is the real-time wind speed at time t, v0 is the reference wind speed, T0 is the reference temperature, T(t) is the blade surface temperature at time t, T is the total length of the time series, min(L) is, Fatigue actual (t) is the measured fatigue degree of the blade at time t, D actual (t) is the measured cumulative damage value.

8. The data-driven wind turbine blade fatigue life assessment method according to claim 7, characterized in that: The constraints of the optimization objective are: Among them, σ y is the yield strength of the blade material, σ u is the ultimate strength of the blade material, s f is the safety factor of the blade, D cr is the critical fatigue damage value, v cut-in is the wind turbine cut-in wind speed, v cut-out is the fan cut-out wind speed, T min is the minimum allowable operating temperature, T max is the maximum allowable operating temperature.

Images