A wind turbine structural component fatigue life prediction method and system

Abstract

The application discloses a wind turbine structural component fatigue life prediction method and system. The method comprises the following steps: obtaining original defect information of a wind turbine structural component, and constructing a digital twin simulation model according to the original defect information; obtaining measured load information in the running process of the wind turbine structural component; inputting the measured load information into the digital twin simulation model to obtain crack propagation information; and obtaining fatigue life evaluation information of the wind turbine structural component according to the crack propagation information. The crack propagation analysis is combined with the actual defect information and the actual running load of the wind turbine structural component, the fine simulation of the fatigue damage evolution process of the structural component is realized, the consistency of the fatigue life prediction result and the actual working state is improved, reliable basis is provided for the structural health monitoring, life evaluation and operation and maintenance decision of the wind turbine, and the operation safety of the wind turbine is improved and the operation and maintenance cost is reduced.

Description

Technical Field

[0001] This application relates to the technical field of wind turbine maintenance, and in particular to a method and system for predicting the fatigue life of wind turbine structural components. Background Technology

[0002] As wind turbines operate for extended periods, their key structural components, such as the main shaft, frame, hub, and bearing housings, are continuously subjected to complex alternating loads, making them prone to fatigue damage. This, in turn, affects the operational safety and service life of the turbine. Therefore, fatigue life prediction for wind turbine structural components is of great significance.

[0003] Existing fatigue life prediction methods typically rely on finite element analysis, fracture mechanics models, and design load spectra. These methods often employ pre-defined cracks or idealized defects as the analysis object, making it difficult to accurately reflect the actual defect state of structural components. Furthermore, the design load spectrum differs from the actual operating load of wind turbines, leading to discrepancies between predicted results and actual service conditions. In addition, existing methods generally lack the ability to dynamically correct the prediction model based on field inspection results, making it difficult to accurately reflect the crack propagation process and the remaining service life of structural components. Summary of the Invention

[0004] The purpose of this application is to overcome the above problems and provide a method and system for predicting the fatigue life of wind turbine structural components.

[0005] The technical solution of this application provides a method for predicting the fatigue life of wind turbine structural components, including: Obtain original defect information of wind turbine structural components; A digital twin simulation model is constructed based on the original defect information; Obtain the measured load information of the wind turbine structural components during operation; The measured load information is input into the digital twin simulation model to obtain crack propagation information; The fatigue life assessment parameters of the wind turbine structural components are obtained based on the crack propagation information.

[0006] Furthermore, after obtaining the crack propagation information, the process also includes: Obtain real-time crack information of the wind turbine structural components; The crack prediction deviation is determined based on the real-time crack information and the crack propagation information; If the crack prediction deviation is greater than a preset threshold, the digital twin simulation model is calibrated. Based on the calibrated digital twin simulation model, updated crack propagation information is obtained.

[0007] Furthermore, the step of constructing a digital twin simulation model based on the original defect information specifically includes: The target defect is determined based on the original defect information; The target defect is converted into a parameterized crack seed according to a preset equivalence rule; The digital twin simulation model is constructed based on the parameterized crack seed.

[0008] Furthermore, obtaining the measured load information of the wind turbine structural components during operation specifically includes: Collect load time series data of the wind turbine structural components during operation; The load time series data is cleaned, filtered, and subjected to cyclic statistical processing to obtain the measured load information.

[0009] Furthermore, the cyclic statistical processing includes: The load time series data were statistically analyzed using the rainflow counting method to obtain the number of cycles corresponding to different stress amplitudes and the number of cycles corresponding to different average stresses; The measured load information is generated based on the statistical results.

[0010] Furthermore, the step of inputting the measured load information into the digital twin simulation model to obtain crack propagation information specifically includes: The stress intensity factor in the digital twin simulation model is determined based on the measured load information; The crack propagation rate is determined based on the stress intensity factor. The crack size is updated based on the crack propagation rate to obtain the crack propagation information.

[0011] Further, determining the crack propagation rate based on the stress intensity factor includes: The crack propagation rate is determined based on the defect propagation model and the stress intensity factor. The formula for the defect propagation model is as follows: ; Where da / dN is the crack propagation rate, λ is the environmental correction factor, C and m are material constants; ΔK is the stress intensity factor; R is the stress ratio; K IC It refers to the fracture toughness of the material.

[0012] Further, the step of updating the crack size based on the crack propagation rate to obtain the crack propagation information includes: The crack growth rate is determined based on the crack propagation rate and the number of cycles corresponding to each load in the measured load information; Update the crack size based on the crack growth rate; The crack propagation information is obtained based on the updated crack size.

[0013] Furthermore, obtaining the fatigue life assessment parameters of the wind turbine structural components based on crack propagation information includes: The current crack size of the wind turbine structural component is determined based on the crack propagation information. Determine whether the current crack size meets the preset failure condition; If the conditions are not met, fatigue life assessment parameters for the wind turbine structural components are determined based on the crack propagation information.

[0014] The technical solution of this application also provides a fatigue life prediction system for wind turbine structural components, including: The defect acquisition module is used to acquire the original defect information of wind turbine structural components; The modeling module is used to construct a digital twin simulation model based on the original defect information; The load acquisition module is used to acquire measured load information of the wind turbine structural components during operation. The crack analysis module is used to input the measured load information into the digital twin simulation model to obtain crack propagation information; The life assessment module is used to obtain fatigue life assessment parameters for the wind turbine structural components based on crack propagation information.

[0015] The above technical solution has the following beneficial effects: This application discloses a method and system for predicting the fatigue life of wind turbine structural components. The method involves acquiring original defect information of the wind turbine structural components and constructing a digital twin simulation model based on this information; acquiring measured load information of the wind turbine structural components during operation; inputting the measured load information into the digital twin simulation model to obtain crack propagation information; and obtaining fatigue life assessment information of the wind turbine structural components based on the crack propagation information. By combining the actual defect information and actual operating loads of the wind turbine structural components for crack propagation analysis, a refined simulation of the fatigue damage evolution process of the structural components is achieved. This improves the consistency between fatigue life prediction results and actual operating conditions, providing a reliable basis for wind turbine structural health monitoring, life assessment, and operation and maintenance decisions. This contributes to improving the operational safety of wind turbines and reducing operation and maintenance costs. Attached Figure Description

[0016] The disclosure of this application will become more readily understood with reference to the accompanying drawings. It should be understood that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. In the drawings: Figure 1This is a flowchart of the fatigue life prediction method for wind turbine structural components in one embodiment of this application; Figure 2 This is a schematic diagram of a parameterized crack seed in one embodiment of this application; Figure 3 This is a schematic diagram of a rainflow counting method in one embodiment of this application; Figure 4 This is a crack propagation rate curve obtained based on measured load information in one embodiment of this application; Figure 5 This is a flowchart of a method for predicting the fatigue life of a wind turbine structural component, as described in one embodiment of this application. Figure 6 This is a flowchart of a method for predicting the fatigue life of wind turbine structural components in another embodiment of this application; Figure 7 This is a schematic diagram of the fatigue life prediction system for wind turbine structural components in one embodiment of this application; Figure 8 This is a schematic diagram of the hardware structure of an electronic device in one embodiment of this application. Detailed Implementation

[0017] The specific embodiments of this application will be further described below with reference to the accompanying drawings.

[0018] It is readily understood that, based on the technical solution of this application, various structural and implementation methods can be interchanged by those skilled in the art without altering the essential spirit of this application. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this application and should not be considered as the entirety of this application or as limitations or restrictions on the technical solution of the application.

[0019] The directional terms such as up, down, left, right, front, back, front, back, top, and bottom mentioned or possibly used in this specification are defined relative to the structures shown in the accompanying drawings. These are relative concepts and may therefore vary depending on their location and usage. Therefore, these or other directional terms should not be interpreted as restrictive. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meanings of the above in this application according to the specific circumstances.

[0021] like Figure 1 As shown, a flowchart of a fatigue life prediction method for wind turbine structural components according to an embodiment of this application is presented, including: Step S101: Obtain the original defect information of the wind turbine structural components; Step S102: Construct a digital twin simulation model based on the original defect information; Step S103: Obtain the measured load information of the wind turbine structural components during operation; Step S104: Input the measured load information into the digital twin simulation model to obtain crack propagation information; Step S105: Obtain fatigue life assessment information of the wind turbine structural components based on the crack propagation information.

[0022] Specifically, in step S101, the original defect information of the wind turbine structural components is obtained. This can be achieved using ultrasonic testing, radiographic testing, magnetic particle testing, or other non-destructive testing methods to inspect key metal load-bearing components such as the main shaft, frame, bearing housing, and hub of the wind turbine, thereby identifying inherent manufacturing defects introduced by processes such as welding, casting, and machining. The original defect information includes, but is not limited to, defects such as porosity, conveying defects, inclusions, and microcracks. Target defects can also be screened based on stress concentration areas, fatigue-sensitive locations, and defect morphology, and converted into parametric cracks according to preset equivalent rules, thus meeting the requirements for subsequent fracture mechanics analysis and numerical simulation calculations.

[0023] In step S102, a digital twin simulation model is constructed based on the original defect information. First, the digital twin simulation model is established by combining the geometric structure, material properties and boundary conditions of the wind turbine structural components. The original defect information obtained in step S101 is equivalent to a parametric crack. Then, the parametric crack is implanted into the digital twin simulation model to obtain a digital twin simulation model containing the initial crack.

[0024] Since the original defects obtained by nondestructive testing include irregular defects such as porosity, inclusions, tool marks, and lack of fusion, they are difficult to use directly for fracture mechanics analysis. Therefore, the original defects are converted into parametric cracks according to the preset equivalent rules to meet the calculation requirements of fracture mechanics theory and improve the convergence and stability of numerical simulation.

[0025] In step S103, the measured load information of the wind turbine structural components during operation is obtained. This measured load information is obtained by processing the load time-series data collected during the actual operation of the wind turbine structural components. Specifically, the measured load time-series data of the wind turbine structural components during operation is first collected, and then data cleaning, filtering, and cyclic statistical processing are performed sequentially to obtain the measured load information, which serves as the input for the subsequent digital twin simulation model.

[0026] In step S104, the measured load information is input into the digital twin simulation model to obtain crack propagation information. The measured load information is applied as input to the digital twin simulation model to determine the stress intensity factor of the crack. The crack propagation rate is calculated based on the stress intensity factor, and the crack growth is accumulated according to the number of cycles corresponding to each load level to update the crack size, thereby obtaining crack propagation information. When there is a large deviation between the model prediction results and the actual detection results, the model parameters can be calibrated based on real-time crack information to further improve the accuracy of crack propagation analysis.

[0027] In step S105, fatigue life assessment information of the wind turbine structural component is obtained based on crack propagation information. First, the current crack size of the wind turbine structural component is determined based on the crack propagation information, and it is judged whether it meets the preset failure conditions. If it does not meet the conditions, the life assessment parameters of the structural component are determined based on the crack propagation information. The life assessment parameters may include at least one of predicted total life, consumed life, remaining safe life, and safety margin, which are used to reflect the current health status and remaining service life of the structural component, providing a basis for structural health monitoring, fault early warning, maintenance decision-making, and life management of the wind turbine.

[0028] This embodiment presents a method for predicting the fatigue life of wind turbine structural components. The method involves acquiring original defect information of the wind turbine structural components and constructing a digital twin simulation model based on this information. It then acquires measured load information during the operation of the wind turbine structural components, inputs this measured load information into the digital twin simulation model to obtain crack propagation information, and finally obtains fatigue life assessment information for the wind turbine structural components based on this crack propagation information. By combining the actual defect information and actual operating loads of the wind turbine structural components for crack propagation analysis, a refined simulation of the fatigue damage evolution process of the structural components is achieved. This improves the consistency between fatigue life prediction results and actual operating conditions, providing a reliable basis for wind turbine structural health monitoring, life assessment, and operation and maintenance decisions. This contributes to improving the operational safety of wind turbines and reducing operation and maintenance costs.

[0029] In one embodiment, after obtaining the crack propagation information, the method further includes: Obtain real-time crack information of the wind turbine structural components; The crack prediction deviation is determined based on the real-time crack information and the crack propagation information; If the crack prediction deviation is greater than a preset threshold, the digital twin simulation model is calibrated. Based on the calibrated digital twin simulation model, updated crack propagation information is obtained.

[0030] In this embodiment, real-time crack information refers to the current actual crack state information of the wind turbine structural components obtained through non-destructive testing, including parameters such as crack size and crack location, which reflects the actual crack situation of the structural components during the current operating stage. Crack prediction deviation refers to the difference between real-time crack information and crack propagation information, used to measure the consistency between the prediction results of the digital twin simulation model and the actual crack state.

[0031] In this embodiment, after obtaining crack propagation information, real-time crack information of the wind turbine structural components is further acquired, and the real-time crack information is compared with the crack propagation information to calculate the crack prediction deviation. When the crack prediction deviation exceeds a preset threshold, it indicates that the current model can no longer accurately represent the actual crack propagation process of the structural components. At this time, the digital twin simulation model is calibrated to adapt the model parameters to the current actual state of the structural components, and crack propagation analysis is re-performed based on the calibrated digital twin simulation model to obtain updated crack propagation information.

[0032] Specifically, the actual crack size and location obtained from on-site non-destructive testing are acquired, and then compared with the crack size and location predicted by the digital twin simulation model to determine the crack prediction deviation. When the crack prediction deviation exceeds a preset threshold, the digital twin simulation model is calibrated. This calibration can be performed using particle swarm optimization and / or Bayesian update algorithms to adjust the model parameters.

[0033] This embodiment compares real-time crack information with crack propagation information and uses the comparison result as the basis for model correction. This allows the digital twin simulation model to continuously reflect the current actual crack propagation state of wind turbine structural components, thereby effectively reducing model prediction errors, improving the consistency between crack propagation analysis results and actual crack states, and further improving the accuracy and reliability of fatigue life assessment results. This provides more reliable data support for wind turbine structural health monitoring, remaining life assessment, and operation and maintenance decisions.

[0034] In one embodiment, calibrating the digital twin simulation model specifically includes: The model parameters in the digital twin simulation model are calibrated using the particle swarm optimization algorithm and / or the Bayesian update algorithm. The digital twin simulation model is updated based on the calibrated model parameters.

[0035] In this embodiment, the model parameters include the material constant C, the material index m, and the fracture toughness K. IC At least one of the following. When using the particle swarm optimization algorithm to calibrate model parameters, the optimization objective is to minimize the error between the crack predicted by the digital twin simulation model and the actual crack. A global search is performed on the model parameters to determine the optimal values ​​of the model parameters. When using the Bayesian update algorithm to calibrate model parameters, the model parameters are probabilistically updated based on real-time crack information to obtain more reliable model parameters. The Bayesian update algorithm can also be combined with the particle swarm optimization algorithm. First, the Bayesian update algorithm is used to determine the reasonable distribution range of the model parameters, and then the particle swarm optimization algorithm is used to iteratively optimize the model parameters to further improve the calibration accuracy of the model parameters.

[0036] After the model parameters are calibrated, the digital twin simulation model is updated based on the calibrated model parameters. This allows the updated digital twin simulation model to more accurately reflect the current crack propagation state of the wind turbine structural components, thereby improving the accuracy of crack propagation analysis and fatigue life assessment.

[0037] In another embodiment, constructing a digital twin simulation model based on the original defect information specifically includes: The target defect is determined based on the original defect information; The target defect is converted into a parameterized crack seed according to a preset equivalence rule; The digital twin simulation model is constructed based on the parameterized crack seed.

[0038] In this embodiment, the target defect refers to the defect selected from the original permission information for crack propagation analysis. This is because wind turbine units may contain various manufacturing defects, and not all defects will significantly affect fatigue life. Therefore, factors such as stress concentration areas, fatigue-sensitive areas, and defect morphology can be combined to screen the original defects and determine the target defects that need to be analyzed in detail. The parameterized crack seed refers to a crack model with regular geometric parameters, used as the initial crack in the digital twin simulation model.

[0039] This embodiment determines the target defect based on the original defect information, and then converts the target defect into a parametric crack seed according to a preset equivalent rule. The preset equivalent rule is the rule followed when converting irregular defects into parametric crack seeds to ensure that the converted crack model meets the requirements of fracture mechanics analysis and numerical simulation. Then, using the parametric crack seed as the initial crack in the digital twin simulation model, a digital twin simulation model is established by combining the geometry, material properties, and boundary conditions of the wind turbine structural component. The inputs to the digital twin simulation model include the parametric crack seed, the geometry, material properties, and boundary conditions of the wind turbine structural component; the outputs of the digital twin simulation model include the stress-strain field of the structural component and crack propagation analysis results, providing basic data for subsequent crack propagation analysis and fatigue life assessment.

[0040] This embodiment transforms complex irregular defects into parameterized crack seeds that meet the requirements of fracture mechanics analysis. While preserving the characteristics of the impact of defects on the structural stress, it improves the computational stability of the digital twin simulation model, enabling subsequent crack propagation analysis to more accurately reflect the evolution of actual fatigue damage in the structural components.

[0041] like Figure 2 As shown, in one embodiment, the parametric crack seed adopts a semi-elliptical crack form, and its crack size is determined based on the measured values ​​from flaw detection, and parametrically set in conjunction with the requirements of fracture mechanics analysis. Specifically, the ratio of crack depth a to half-length c, a / c, is 0.25–0.33; the total crack length 2c is 5%–15% of the width W of the hot spot region; and the crack depth a does not exceed 10%–20% of the structural component thickness. This ensures that the parametric crack seed can both reflect the actual defect characteristics and meet the calculation requirements of fracture mechanics analysis and numerical simulation.

[0042] As a specific embodiment, for localized corrosion defects detected in the hot spot area of ​​a wind turbine weld, it is known that non-destructive testing yielded a maximum defect extension length of 22 mm along the surface, a hot spot area width of 200 mm, and a structural component thickness of 40 mm. First, based on the hot spot area width, the allowable range for the total crack length is determined to be 10 mm to 30 mm. Since the measured defect length of 22 mm is within the allowable range, the total crack length 2c is taken as 22 mm, i.e., the crack half-length c is 11 mm. Subsequently, according to a preset equivalence rule, the ratio of crack depth to crack half-length a / c is taken as 0.3, and the crack depth a is calculated to be 3.3 mm. Since the crack depth of 3.3 mm does not exceed the allowable upper limit of 20% of the structural component thickness and meets the preset range of a / c, the defect is equivalent to a semi-elliptical parametric crack seed with a depth a of 3.3 mm and a half-length c of 11 mm, and is implanted into the weld hot spot area corresponding to the digital twin simulation model as the initial crack.

[0043] Furthermore, when the defect size obtained by nondestructive testing exceeds the allowable range of the preset equivalent rules, the crack size can be corrected according to the preset equivalent rules. For example, when the total crack length exceeds the allowable upper limit of 15% of the width of the hot spot region, the total crack length is corrected to the corresponding upper limit value, and the crack depth is re-determined according to the preset crack depth to crack half-length ratio to generate a parametric crack seed that meets the requirements of fracture mechanics analysis.

[0044] This embodiment converts irregular defects such as pores, inclusions, lack of fusion, and corrosion pits obtained from on-site detection into semi-elliptical parametric crack seeds that meet the requirements of fracture mechanics analysis. While retaining the main geometric and stress characteristics of the defects, it improves the computability of the digital twin simulation model, making subsequent crack propagation analysis and fatigue life prediction more consistent with actual engineering conditions.

[0045] Furthermore, the digital twin simulation model supports the application of external load boundary conditions and can jointly solve the stress-strain field and crack propagation behavior, thus providing a foundation for subsequent crack propagation analysis.

[0046] In another embodiment, constructing a digital twin simulation model based on the original defect information further includes: A digital twin simulation model is constructed based on the structural characteristics of the wind turbine components and the available computational resources. Mesh discretization is performed on the digital twin simulation model; Local adaptive mesh refinement is applied to the crack hotspot area to achieve finer mesh subdivision; Non-critical areas are divided using a coarser mesh.

[0047] In this embodiment, a digital twin simulation model of the wind turbine components is established based on the structural characteristics and computational resource conditions, and a local adaptive refinement strategy is used to discretize the mesh. Specifically, the crack hotspot region is finely meshed to improve the computational accuracy of crack stress field and crack propagation analysis; non-critical regions are coarser meshed to reduce the number of meshes and lower computational resource consumption.

[0048] This embodiment adopts a local adaptive encryption strategy, which effectively balances computational efficiency and computational resource consumption while ensuring the computational accuracy of the digital twin simulation model, thereby improving the computational stability and engineering applicability of the digital twin simulation model in the fatigue life prediction of wind turbine structural components.

[0049] In one embodiment, obtaining the measured load information of the wind turbine structural components during operation specifically includes: Collect load time series data of the wind turbine structural components during operation; The load time series data is cleaned, filtered, and subjected to cyclic statistical processing to obtain the measured load information.

[0050] In this embodiment, the load time series data refers to the load data continuously collected from the wind turbine structural components during actual operation. It records the load changes experienced by the structural components at different operating times in chronological order, accurately reflecting the stress state of the wind turbine during operation. "Cleaning" refers to preprocessing the load time series data to remove abnormal, invalid, or missing data, thereby improving data quality. "Filtering" refers to noise suppression of the load time series data, reducing the impact of environmental interference and measurement errors on the load data. "Cyclic statistical processing" refers to performing cyclic load statistics on the load time series data, extracting different load cyclic characteristics, and forming measured load information that characterizes the actual operating conditions.

[0051] In this embodiment, since the wind turbine is affected by factors such as wind speed changes and start-stop control during actual operation, the load on the structural components exhibits continuous variation. Therefore, load time series data during the operation of the structural components is collected, cleaned, and filtered to improve data quality. Subsequently, the processed load time series data undergoes cyclic statistical processing to convert the continuously changing load data into measured load information that can characterize the actual cyclic load features.

[0052] This embodiment enables the measured load information to more accurately reflect the load variation law of wind turbine structural components during actual service, providing a reliable load input for subsequent crack propagation analysis and improving the consistency between fatigue life prediction results and actual operating conditions.

[0053] In one embodiment, the cyclic statistical processing includes: The load time series data were statistically analyzed using the rainflow counting method to obtain the number of cycles corresponding to different stress amplitudes and the number of cycles corresponding to different average stresses; The measured load information is generated based on the statistical results.

[0054] Specifically, the rainflow counting method is used to perform cyclic identification on the load time series data after cleaning and filtering. The continuously changing load signal is decomposed into multiple load cycles, and the stress amplitude, average stress, and cycle number corresponding to each load cycle are statistically analyzed to generate the measured load spectrum (σ). i , σ m n i This refers to the measured load information. The measured load spectrum includes different stress amplitudes σ. i Mean stress σ m and the corresponding number of loops n i It is used to characterize the cyclic load characteristics of wind turbine structural components under actual operating conditions.

[0055] This embodiment employs the rainflow counting method to perform cyclic statistical processing on load time series data, transforming complex and continuously changing load data into statistically significant cyclic load information. While preserving the actual load characteristics, it effectively reduces the computational complexity of directly using the original load data for analysis, making the generated measured load information more consistent with the actual load conditions of wind turbine structural components. This provides a more accurate and reliable load input for subsequent crack propagation analysis, thereby further improving the accuracy and reliability of fatigue life prediction results.

[0056] Specifically, the load time series data is first filtered and denoised, and extreme values ​​are extracted to obtain a load peak-valley sequence arranged in chronological order. Then, the rainflow counting method is used to identify the load peak-valley sequence in cycles. Complete load cycles are identified step-by-step according to preset cycle determination rules, and residual extreme values ​​that cannot form complete cycles are retained as half-cycles. The number of cycles corresponding to a complete load cycle is recorded as 1, and the number of cycles corresponding to a half-cycle is recorded as 0.5.

[0057] After identifying each load cycle, the stress amplitude and average stress are calculated based on the maximum and minimum loads corresponding to that load cycle. The formula for calculating the stress amplitude is as follows: ; in, S is the stress amplitude, Smax is the maximum load, and Smin is the minimum load.

[0058] The formula for calculating the mean stress is as follows: ; in, The average stress is denoted as .

[0059] Subsequently, all load cycles were statistically analyzed based on stress amplitude and average stress. The stress amplitude, average stress, and number of cycles corresponding to each load cycle were summarized to generate the measured load spectrum (σ). i , σ m n i ), where σ i σ represents the stress amplitude. m n represents the average stress. i The measured load spectrum represents the number of cycles and is used to characterize the cyclic load characteristics of wind turbine structural components during actual operation. It also serves as the load input for subsequent crack propagation analysis and fatigue life prediction.

[0060] like Figure 3 As shown, the load time series includes multiple peak and trough points arranged in chronological order. When using the rainflow counting method, cyclic identification begins with the first peak point, forming the first complete load cycle, and the corresponding extreme points are removed from the load sequence. Subsequently, the same cyclic identification process is performed on the remaining load sequences to form the second complete load cycle. Peak and trough points that do not form a closed cycle are retained as half-cycles, awaiting subsequent load data to participate in cyclic identification again. After completing all cyclic identifications, the stress amplitude, average stress, and cycle number corresponding to each load cycle are statistically analyzed, and a measured load spectrum is generated.

[0061] Specifically, such as Figure 3 As shown, the horizontal axis represents the magnitude of stress, and the vertical axis represents the direction of time progression. The load time series includes multiple peak points and valley points arranged in chronological order. Figure 3 In the diagram, 1 to 10 represent extreme points in the load history, and the dashed line represents the raindrop flow path. When using the rainflow counting method for cycle identification, raindrops start flowing downwards from the inside of each peak point; when a raindrop encounters a peak higher than the initial peak during its flow, it stops flowing; when a raindrop encounters a raindrop that has already fallen above it, it also stops flowing; when a raindrop completes a closed flow path, it is determined that a complete load cycle has been formed, and the stress amplitude, average stress, and number of cycles corresponding to the load cycle are recorded. At the same time, the corresponding extreme points are removed from the load sequence, and the above cycle identification process is repeated for the remaining load sequences.

[0062] The cycle identification begins from extreme point 1. The raindrop flows from 1 to 2, then falls vertically from 2 to 3, and continues to flow from 3 to 4. When the raindrop encounters the raindrop falling from above at extreme point 4, it stops flowing, thus forming the first complete load cycle 1~2~3~4. The stress amplitude, average stress, and number of cycles corresponding to this cycle are recorded. Then, extreme points 1, 2, 3, and 4 are removed from the load sequence.

[0063] Subsequently, the remaining extreme points 5 to 10 were repeatedly identified. The raindrop started from extreme point 5, flowed through 5→6 and fell vertically to 7, and continued to flow along 7→8. When the raindrop stopped flowing at extreme point 8, a second complete load cycle 5~6~7~8 was formed, and the corresponding stress amplitude, average stress and number of cycles were recorded. At the same time, extreme points 5, 6, 7 and 8 were removed from the load sequence.

[0064] Finally, only extreme points 9 and 10 remain. Since they cannot form a closed loop, they are determined to be half-cycles and retained in subsequent load time series to continue participating in rainflow pairing. Ultimately, this embodiment identifies two complete load cycles and one half-cycle, and statistically analyzes the stress amplitude, average stress, and cycle number for each load cycle to generate the measured load spectrum (σ). i , σ m n i ), where σ i σ represents the stress amplitude. m n represents the average stress. i The number of cycles is indicated by the measured load spectrum, which serves as the load input for subsequent crack propagation analysis and fatigue life prediction.

[0065] In another embodiment, inputting the measured load information into the digital twin simulation model to obtain crack propagation information specifically includes: The stress intensity factor in the digital twin simulation model is determined based on the measured load information; The crack propagation rate is determined based on the stress intensity factor. The crack size is updated based on the crack propagation rate to obtain the crack propagation information.

[0066] In this embodiment, the stress intensity factor refers to a parameter in fracture mechanics used to characterize the strength of the stress field at the crack tip, reflecting the stress state of the crack under external load. The crack propagation rate refers to the rate at which a crack grows under cyclic loading, describing how quickly the crack expands with each cyclic load. Crack propagation information refers to the evolution of the crack size under load, reflecting the development process of fatigue damage in the structural component.

[0067] In this embodiment, the measured load information is first used as the load input to the digital twin simulation model. The digital twin simulation model then calculates the stress intensity factor at the crack tip under the measured load information, based on the structural geometry, material properties, and boundary conditions. This stress factor reflects the stress state at the crack tip under the current load condition. The crack propagation rate is then determined based on the stress intensity factor, and the crack size is updated according to the crack propagation rate, thereby obtaining crack propagation information.

[0068] This embodiment establishes the correlation between measured load information, stress intensity factor, crack propagation rate, and crack size to achieve dynamic simulation of the crack propagation process. This enables crack propagation analysis to fully reflect the fatigue damage evolution process of wind turbine structural components under actual operating conditions, improves the consistency between crack propagation analysis results and actual service conditions, and provides a reliable basis for subsequent fatigue life assessment.

[0069] In one embodiment, determining the crack propagation rate based on the stress intensity factor includes: Construct a defect extension model based on the stress intensity factor; Crack propagation rate is determined based on a defect propagation model.

[0070] In this embodiment, a defect propagation model is constructed based on the stress intensity factor, combined with fracture mechanics and fatigue crack propagation criteria. This defect propagation model describes the crack propagation behavior under cyclic loading and calculates the crack propagation rate based on the stress intensity factor, providing a basis for subsequent crack size updates.

[0071] Specifically, the extended finite element method or adaptive mesh re-division technique can be used for calculation, with the stress intensity factor as the driving parameter. A defect propagation model can be established by combining at least one of the Paris formula or Forman formula to describe the crack propagation law under cyclic loading. Subsequently, the crack propagation rate is determined based on the defect propagation model, and the crack size is updated based on the crack propagation rate, thereby obtaining crack propagation information.

[0072] This embodiment establishes a defect propagation model and correlates measured load information with the crack propagation process to achieve dynamic analysis of crack propagation. This makes the analysis results of crack propagation more consistent with the actual load conditions of wind turbine structural components, thereby improving the accuracy and reliability of crack propagation prediction results.

[0073] In one embodiment, determining the crack propagation rate based on the stress intensity factor includes: The crack propagation rate is determined based on the defect propagation model and the stress intensity factor. The formula for the defect propagation model is as follows: ; Where da / dN is the crack propagation rate, λ is the environmental correction factor, C is the first material constant, m is the second material constant; ΔK is the stress intensity factor; R is the stress ratio; K IC It refers to the fracture toughness of the material.

[0074] The defect propagation model in this embodiment is a pre-established mathematical model used to describe the propagation law of cracks under cyclic loading. The crack propagation rate is calculated based on the stress intensity factor, providing a calculation basis for crack propagation analysis.

[0075] In this embodiment, after obtaining the stress intensity factor, the stress ratio R is determined based on the measured load information, and the stress intensity factor is used as a driving parameter input into the defect propagation model. This is combined with the environmental correction coefficient λ, the first material constant C, the second material constant m, the stress ratio R, and the material fracture toughness K. IC The crack propagation rate is calculated. The environmental correction factor λ is used to correct for the effects of environmental factors such as temperature, humidity, and wind erosion on crack propagation; the stress ratio R reflects the effect of cyclic loading on the stress state; and the first material constants C and m, as well as the material fracture toughness K, are also considered. IC Used to characterize the crack propagation characteristics of structural component materials.

[0076] Among them, the environmental correction coefficient λ is used to correct for the influence of environmental factors such as temperature, humidity, and wind and sand corrosion on the crack propagation rate; the first material constant C is a proportionality coefficient used to characterize the material's resistance to fatigue crack propagation; the second material constant m is a power law exponent used to characterize the sensitivity of the material's crack propagation rate to the stress intensity factor amplitude ΔK; R is the stress ratio used to reflect the influence of cyclic loading on the stress state; K IC Fracture toughness is used to characterize a material's ability to resist the unstable propagation of cracks.

[0077] The first material constant C and the second material constant m can be obtained through material fatigue crack propagation tests. Specifically, cyclic loading can be applied to a standard specimen to measure the relationship between the crack size a and the number of cycles N, calculate the crack propagation rate da / dN and the corresponding stress intensity factor amplitude ΔK, and determine the first material constant C and the second material constant m based on the fitting relationship between da / dN and ΔK.

[0078] This embodiment comprehensively considers factors such as operating environment, material properties, and load conditions to make the calculated crack propagation rate more consistent with the crack propagation law of wind turbine structural components during actual service, providing a reliable basis for subsequent crack size updates.

[0079] like Figure 4As shown, the crack propagation rate curve (da / dN-ΔK curve) is obtained based on measured load information, where the horizontal axis represents the stress intensity factor ΔK and the vertical axis represents the crack propagation rate da / dN. This crack propagation rate curve can intuitively reflect the law of crack propagation rate variation with the stress intensity factor, providing a basis for crack size updates and fatigue life assessment.

[0080] In one embodiment, updating the crack size based on the crack propagation rate to obtain the crack propagation information includes: The crack growth rate is determined based on the crack propagation rate and the number of cycles corresponding to each load in the measured load information; Update the crack size based on the crack growth rate; The crack propagation information is obtained based on the updated crack size.

[0081] In this embodiment, crack growth refers to the increase in size of the crack under a certain cyclic load, used to characterize the extent of crack propagation during load cycling. The number of cycles refers to the number of repetitions of each load level in the measured load information, used to reflect the frequency of different loads occurring during the actual operation of the wind turbine.

[0082] In this embodiment, the number of cycles corresponding to each load in the measured load information is combined to determine the crack growth amount under each load level based on the crack propagation rate, and the crack growth amount at each level is accumulated to update the current crack size, thus obtaining the crack propagation information.

[0083] This embodiment combines the crack propagation rate with the number of cycles under actual operating conditions, enabling the crack size update process to fully consider the cumulative effect of long-term repeated action of different loads on crack propagation, realizing the dynamic evolution of the crack propagation process, and making the obtained crack propagation information more consistent with the crack propagation law of wind turbine structural components in actual service, thereby improving the accuracy of crack propagation analysis results and fatigue life prediction results.

[0084] In another embodiment, obtaining the fatigue life assessment parameters of the wind turbine structural component based on crack propagation information includes: The current crack size of the wind turbine structural component is determined based on the crack propagation information. Determine whether the current crack size meets the preset failure condition; If the conditions are not met, fatigue life assessment parameters for the wind turbine structural components are determined based on the crack propagation information.

[0085] In this embodiment, the current crack size refers to the crack size of the wind turbine structural component at the current stage, determined based on crack propagation information, reflecting the actual state of crack propagation at the current moment. Preset failure conditions are pre-set failure criteria for the structural component, such as determining that the component has reached a failure state when the current crack size reaches the critical crack size, or when the stress and strain of the structural component do not meet safety specifications. Fatigue life assessment parameters are parameters used to characterize the service status and remaining service capacity of the wind turbine structural component, including predicted total life, consumed life, remaining safe life, and safety margin.

[0086] In this embodiment, the current crack size is first determined based on crack propagation information, and it is then determined whether the current crack size meets preset failure conditions. When the current crack size meets the preset failure conditions, it indicates that the crack has propagated to the point of affecting the safe operation of the structural component, and the structural component is determined to have reached a failure state; when the preset failure conditions are not met, it indicates that the structural component still has the ability to continue service. Further, fatigue life assessment parameters are determined based on crack propagation information to evaluate the current life state of the structural component.

[0087] This embodiment combines crack propagation analysis results with failure determination, enabling fatigue life assessment to be based on the actual evolution of cracks. This avoids assessments based solely on theoretical life or empirical data, improving the consistency between fatigue life assessment results and actual service conditions. It provides a reliable basis for wind turbine structural health monitoring, condition-based maintenance, and operation and maintenance decisions.

[0088] In one embodiment, fatigue life assessment parameters include predicted total life, consumed life, remaining safe life, and safety margin indicators.

[0089] Specifically, predicting the total lifetime N total The total number of load cycles required for the crack to propagate from its initial size to the critical crack size; the consumed lifetime N. consumed The number of load cycles that the structural member has withstood, calculated based on historical measured load spectra; remaining safe life t remaining Determine according to the following formula: ; Where f is the load cycle frequency, and the remaining safe life is used to characterize the expected safe operating time of a wind turbine structural component from its current state to its failure state.

[0090] The safety margin index MI is determined according to the following formula: ; Among them, a c Let a be the critical crack size. nowThis is the current crack size output by the digital twin simulation model. The safety margin index is used to characterize the current structural safety status of the wind turbine structural components. When the safety margin index is closer to 1, it indicates that the structural component is further away from the failure state and has a higher safety margin. When the safety margin index approaches 0, it indicates that the current crack size is gradually approaching the critical crack size, and the structural component is close to the failure state.

[0091] This embodiment comprehensively predicts the total lifespan, consumed lifespan, remaining safe lifespan, and safety margin indicators to quantitatively evaluate the current fatigue damage state and remaining service capacity of wind turbine structural components. It can not only predict the remaining safe operating time of structural components, but also dynamically reflect the structural safety margin, providing a more reliable assessment basis for structural health monitoring and operation and maintenance decisions.

[0092] like Figure 5 As shown, a flowchart of a fatigue life prediction method for wind turbine structural components in one embodiment of this application is illustrated, specifically including: Step S501: Obtain the original defect information of the wind turbine structural components; Step S502: Determine the target defect based on the original defect information; Step S503: Convert the target defect into a parameterized crack seed according to a preset equivalent rule; Step S504: Construct the digital twin simulation model based on the parameterized crack seed; Step S505: Collect load time series data of the wind turbine structural components during operation; Step S506: Clean, filter, and perform cyclic statistical processing on the load time series data to obtain the measured load information; Step S507: Determine the stress intensity factor in the digital twin simulation model based on the measured load information; Step S508: Determine the crack propagation rate based on the stress intensity factor; Step S509: Determine the crack growth rate based on the crack propagation rate and the number of cycles corresponding to each load in the measured load information; Step S510: Update the crack size based on the crack growth rate; Step S511: Obtain the crack propagation information based on the updated crack size; Step S512: Obtain real-time crack information of the wind turbine structural components; Step S513: Determine the crack prediction deviation based on the real-time crack information and the crack propagation information; Step S514: If the crack prediction deviation is greater than a preset threshold, calibrate the digital twin simulation model; Step S515: Based on the calibrated digital twin simulation model, obtain the updated crack propagation information; Step S516: Determine the current crack size of the wind turbine structural component based on the crack propagation information; Step S517: Determine whether the current crack size meets the preset failure condition; Step S518: If not satisfied, determine the fatigue life assessment parameters of the wind turbine structural component based on the crack propagation information.

[0093] like Figure 6 The diagram illustrates a workflow for predicting the fatigue life of wind turbine structural components in another embodiment of this application. Specifically, in step S601, non-destructive testing is performed on key metal load-bearing structural components such as the wind turbine main shaft, frame, hub, and bearing housing to obtain actual crack data within the components. Non-destructive testing can acquire information such as crack size and location, and the obtained actual crack data is used as the basis for subsequent digital twin modeling and model calibration. This ensures that crack analysis is based on the actual defect state of the structural components, improving the reliability of the fatigue life prediction results.

[0094] In step S602, a digital twin model containing cracks is established. Based on the real crack data obtained in step S601, and combined with the geometric structure, material properties, and boundary conditions of the wind turbine structural components, a digital twin model containing cracks is established. The real crack data can be equivalently implanted into the digital twin model as parametric crack seeds. Furthermore, a local adaptive mesh refinement strategy is used to refine the mesh in the crack hotspot region, thereby improving the calculation accuracy of the stress field at the crack tip and crack propagation analysis, thus increasing the model's solution efficiency while ensuring computational accuracy.

[0095] In step S603, the measured load spectrum is acquired and processed. This involves collecting load time series data during the operation of wind turbine structural components, and cleaning, filtering, and performing cyclic statistical processing on the load time series data to obtain the measured load spectrum. The rainflow counting method is used to count the number of cycles corresponding to different stress amplitudes and average stresses, generating the measured load spectrum, which is the measured load information, used to accurately characterize the cyclic load characteristics of wind turbine structural components under actual operating conditions.

[0096] In step S604, a crack evolution model driven by measured load spectrum is used. The measured load spectrum is used as the input load for the digital twin model. The stress ratio is determined based on the measured load information, and the stress intensity factor at the crack tip is calculated. Furthermore, a crack propagation model, also known as a defect propagation model, is constructed by combining fracture mechanics theory and fatigue crack propagation criteria. The crack propagation rate is determined through this model, and the crack size is updated by combining the cycle number corresponding to each load level, thus obtaining crack propagation information and realizing the dynamic simulation of the crack propagation process in wind turbine structural components.

[0097] In step S605, nondestructive testing and online model calibration are performed. Specifically, the actual crack size and location obtained from periodic nondestructive testing are acquired and compared with the crack state predicted by the digital twin model. When the crack prediction deviation exceeds a preset threshold, the model parameters in the digital twin model are calibrated using a particle swarm optimization algorithm and / or a Bayesian update algorithm. The calibrated model parameters are then used to update the digital twin model, enabling the model to continuously reflect the current actual crack propagation state of the wind turbine structural components and improving the accuracy of crack propagation analysis results.

[0098] In step S606, the current crack size is determined based on crack propagation information, and it is determined whether the current crack size meets the preset failure conditions. If the preset failure conditions are not met, fatigue life assessment parameters such as predicted total life, consumed life, remaining safe life, and safety margin are further determined to comprehensively assess the current health status and remaining service capacity of the wind turbine structural components. If the preset failure conditions are met, the structural component is determined to have reached a failure state, providing a basis for wind turbine structural health monitoring, maintenance plan formulation, and operation and maintenance decisions.

[0099] like Figure 7 As shown, embodiments of this application also provide a fatigue life prediction system for wind turbine structural components, comprising: The defect acquisition module 701 is used to acquire the original defect information of the wind turbine structural components; Modeling module 702 is used to construct a digital twin simulation model based on the original defect information; The load acquisition module 703 is used to acquire the measured load information of the wind turbine structural components during operation. The crack analysis module 704 is used to input the measured load information into the digital twin simulation model to obtain crack propagation information; The life assessment module 705 is used to obtain fatigue life assessment parameters of the wind turbine structural components based on crack propagation information.

[0100] This embodiment of a fatigue life prediction system for wind turbine structural components includes a defect acquisition module 701, a modeling module 702, a load acquisition module 703, a crack analysis module 704, and a life assessment module 705. These modules are interconnected to form a complete fatigue life prediction process. The defect acquisition module 701 acquires the original defect information of the wind turbine structural components and sends the acquired original defect information to the modeling module 702. The modeling module 702 establishes a digital twin simulation model based on the original defect information and provides the established digital twin simulation model to the crack analysis module 704. The load acquisition module 703 acquires the measured load information of the wind turbine structural components during operation and inputs the measured load information into the crack analysis module 704. Based on the digital twin simulation model and the measured load information, the crack analysis module 704 analyzes the crack propagation process of the structural components, obtains crack propagation information, and sends the crack propagation information to the life assessment module 705. The life assessment module 705 determines the fatigue life assessment parameters of the wind turbine structural components based on the crack propagation information, thereby completing the fatigue life prediction of the wind turbine structural components.

[0101] The crack analysis module 704 can also calculate the stress intensity factor at the crack tip based on the digital twin simulation model, and calculate the crack propagation rate in combination with the defect propagation model. The crack size is updated based on the crack propagation rate and the number of cycles in the measured load information, thereby obtaining crack propagation information.

[0102] The life assessment module 705 can also determine the current crack size based on crack propagation information, and combine preset failure conditions to determine fatigue life assessment parameters such as predicted total life, consumed life, remaining safe life, and safety margin, so as to comprehensively evaluate the current service status of wind turbine structural components.

[0103] This embodiment integrates non-destructive testing, digital twin modeling, measured load acquisition, crack propagation analysis, and fatigue life assessment by setting up a defect acquisition module 701, a modeling module 702, a load acquisition module 703, a crack analysis module 704, and a life assessment module 705. This achieves modular management and collaborative processing of the entire fatigue life prediction process for wind turbine structural components. Specifically, the defect acquisition module 701 establishes an analytical basis based on structural component defects; the modeling module 702 constructs a digital twin simulation model based on actual defects; the crack analysis module 704 drives crack propagation analysis based on measured loads, making the crack propagation process more consistent with actual service conditions; and the life assessment module 705 calculates fatigue life assessment parameters based on crack propagation information, achieving an accurate assessment of the remaining service capacity of wind turbine structural components. This improves the consistency between crack propagation analysis and fatigue life prediction results and actual operating conditions, and forms a complete closed loop of defect detection, crack analysis, life prediction, and operation and maintenance assessment, providing reliable data support for wind turbine structural health monitoring, condition-based maintenance, and operation and maintenance decisions.

[0104] This application also provides a storage medium that stores computer instructions, which, when executed by a computer, are used to perform a method for predicting the fatigue life of wind turbine structural components as described in any of the preceding embodiments.

[0105] Figure 8 An electronic device according to this application is shown, comprising: At least one processor 801; and, A memory 802 is communicatively connected to at least one processor 801; wherein, The memory 802 stores instructions that can be executed by at least one processor 801, which enables the at least one processor 801 to perform all the steps of a wind turbine structural component fatigue life prediction method in any of the foregoing method embodiments.

[0106] Figure 8 Taking the 801 processor as an example: The electronic device may also include an input device 803 and an output device 804.

[0107] The processor 801, memory 802, input device 803 and output device 804 can be connected by a bus or other means. The figure shows an example of connection by bus.

[0108] The memory 802, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to a method for predicting the fatigue life of wind turbine structural components in an embodiment of this application. Figure 1 , Figure 5 and Figure 6 The method flow is shown. The processor 801 executes various functional applications and data processing by running non-volatile software programs, instructions, and modules stored in the memory 802, thereby realizing a method for predicting the fatigue life of wind turbine structural components in the above embodiment.

[0109] The memory 802 may include a program storage area and a data storage area. The program storage area may store an operating system and an application program required for at least one function. The data storage area may store data created based on the use of a method for predicting the fatigue life of wind turbine structural components. Furthermore, the memory 802 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory 802 may optionally include memory remotely located relative to the processor 801, and these remote memories may be connected via a network to an apparatus performing a method for predicting the fatigue life of wind turbine structural components. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0110] The input device 803 can receive user clicks and generate signal inputs related to user settings and function control for the fatigue life prediction method of wind turbine structural components. The output device 804 may include a display screen or other display device.

[0111] One or more modules are stored in memory 802, and when run by one or more processors 801, a method for predicting the fatigue life of wind turbine structural components in any of the above method embodiments is executed.

[0112] Embodiments of this application also provide a computer program product, including a computer program / instructions that, when executed by a processor, implement a method for predicting the fatigue life of a wind turbine structural component as described in any of the preceding embodiments.

[0113] As needed, the above technical solutions can be combined to achieve the best technical effect.

[0114] The above description is merely the principle and preferred embodiment of this application. It should be noted that for those skilled in the art, implementation methods obtained by appropriately combining the technical solutions disclosed in different embodiments are also included within the technical scope of this invention. Based on the principle of this application, several other modifications can also be made, which should also be considered within the protection scope of this application.

Claims

1. A method for predicting the fatigue life of wind turbine structural components, characterized in that, include: Obtain original defect information of wind turbine structural components; A digital twin simulation model is constructed based on the original defect information; Obtain the measured load information of the wind turbine structural components during operation; The measured load information is input into the digital twin simulation model to obtain crack propagation information; The fatigue life assessment parameters of the wind turbine structural components are obtained based on the crack propagation information.

2. The method for predicting the fatigue life of wind turbine structural components according to claim 1, characterized in that, After obtaining the crack propagation information, the process also includes: Obtain real-time crack information of the wind turbine structural components; The crack prediction deviation is determined based on the real-time crack information and the crack propagation information; If the crack prediction deviation is greater than a preset threshold, the digital twin simulation model is calibrated. Based on the calibrated digital twin simulation model, updated crack propagation information is obtained.

3. The method for predicting the fatigue life of wind turbine structural components according to claim 1, characterized in that, The construction of the digital twin simulation model based on the original defect information specifically includes: The target defect is determined based on the original defect information; The target defect is converted into a parameterized crack seed according to a preset equivalence rule; The digital twin simulation model is constructed based on the parameterized crack seed.

4. The method for predicting the fatigue life of wind turbine structural components according to claim 1, characterized in that, The acquisition of measured load information during the operation of the wind turbine structural components specifically includes: Collect load time series data of the wind turbine structural components during operation; The load time series data is cleaned, filtered, and subjected to cyclic statistical processing to obtain the measured load information.

5. The method for predicting the fatigue life of wind turbine structural components according to claim 4, characterized in that, The cyclic statistical processing includes: The load time series data were statistically analyzed using the rainflow counting method to obtain the number of cycles corresponding to different stress amplitudes and the number of cycles corresponding to different average stresses; The measured load information is generated based on the statistical results.

6. The method for predicting the fatigue life of wind turbine structural components according to claim 1, characterized in that, The step of inputting the measured load information into the digital twin simulation model to obtain crack propagation information specifically includes: The stress intensity factor in the digital twin simulation model is determined based on the measured load information; The crack propagation rate is determined based on the stress intensity factor. The crack size is updated based on the crack propagation rate to obtain the crack propagation information.

7. The method for predicting the fatigue life of wind turbine structural components according to claim 6, characterized in that, The step of determining the crack propagation rate based on the stress intensity factor includes: Based on the stress intensity factor, the crack propagation rate is determined as follows: ； Where da / dN is the crack propagation rate, λ is the environmental correction factor, C is the first material constant, m is the second material constant; ΔK is the stress intensity factor; R is the stress ratio; K IC It refers to the fracture toughness of the material.

8. The method for predicting the fatigue life of wind turbine structural components according to claim 6, characterized in that, The step of updating the crack size according to the crack propagation rate to obtain the crack propagation information includes: The crack growth rate is determined based on the crack propagation rate and the number of cycles corresponding to each load in the measured load information; Update the crack size based on the crack growth rate; The crack propagation information is obtained based on the updated crack size.

9. The method for predicting the fatigue life of wind turbine structural components according to claim 1, characterized in that, The fatigue life assessment parameters for the wind turbine structural components obtained based on crack propagation information include: The current crack size of the wind turbine structural component is determined based on the crack propagation information. Determine whether the current crack size meets the preset failure condition; If the conditions are not met, fatigue life assessment parameters for the wind turbine structural components are determined based on the crack propagation information.

10. A fatigue life prediction system for wind turbine structural components, characterized in that: include: The defect acquisition module is used to acquire the original defect information of wind turbine structural components; The modeling module is used to construct a digital twin simulation model based on the original defect information; The load acquisition module is used to acquire measured load information of the wind turbine structural components during operation. The crack analysis module is used to input the measured load information into the digital twin simulation model to obtain crack propagation information; The life assessment module is used to obtain fatigue life assessment parameters for the wind turbine structural components based on crack propagation information.

Images