Wind turbine component life assessment method and system based on multi-physics coupling

Abstract

This invention relates to the field of wind turbine management technology, and particularly to a method and system for assessing the lifespan of wind turbine components based on multiphysics coupling. The method includes: dividing the wind turbine rotor gearbox system into multiple fine-grained assessment units based on multiphysics coupling, according to the principle of matching component functional zoning with physical field types; acquiring multi-source sensor data for each assessment unit; determining the multiphysics coupling state threshold for each assessment unit based on the geometric characteristics and historical maintenance fault data of the wind turbine rotor gearbox system; for each assessment unit, identifying degradation candidate units based on the multiphysics coupling state threshold and multi-source sensor data; acquiring historical data corresponding to each degradation candidate unit, and using this data to perform degradation trend analysis on the degradation candidate units to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit. This method can improve the accuracy of lifespan prediction.

Description

Technical Field

[0001] This invention relates to the technical field of wind turbine management, and in particular to a method and system for assessing the lifespan of wind turbine components based on multi-physics coupling. Background Technology

[0002] As the core equipment of a wind power system, the operational reliability of wind turbines directly determines the power generation efficiency and economic benefits of a wind farm. The rotor gearbox system is a key transmission and energy conversion component of the wind turbine, undertaking the core function of converting the wind energy captured by the rotor into mechanical energy and transmitting it to the generator. Since wind power systems operate in complex and harsh natural environments and working conditions for a long time, it is necessary to conduct life assessments on wind turbines to ensure their safe and stable operation.

[0003] Existing methods for assessing the lifespan of wind turbine components often involve isolated analyses of single physical fields. For example, they may use only structural mechanics to assess impeller fatigue or only tribology to assess gearbox wear. These methods neglect the nonlinear coupling effects of multiple physical fields. For instance, fluctuations in impeller aerodynamic loads can alter gearbox stress distribution, increased gearbox oil temperature can reduce lubricant viscosity and accelerate wear, and wear vibrations can unbalance impeller dynamic loads. Existing methods cannot quantify the combined impact of multiple fields on degradation, leading to errors in lifespan prediction and inaccurate assessments of the remaining lifespan of wind turbine units. Summary of the Invention

[0004] This invention provides a method and system for assessing the lifespan of wind turbine components based on multi-physics coupling, which can effectively solve the problems in the background art.

[0005] To achieve the above objectives, in a first aspect, the present invention provides a method for assessing the lifespan of wind turbine components based on multi-physics coupling, comprising:

[0006] For wind turbine impeller gearbox systems, multiple fine-grained evaluation units based on multi-physics are divided according to the principle of matching component functional zoning with physical field type, and multi-source sensor data of each evaluation unit are acquired.

[0007] Based on the geometric characteristics and historical operation and maintenance fault data of the wind turbine rotor gearbox system, the multi-field coupling state threshold of each of the evaluation units is determined.

[0008] For each of the evaluation units, a degradation candidate unit is determined based on the multi-field coupling state threshold and the multi-source sensing data;

[0009] Historical data corresponding to each degradation candidate unit is obtained, and degradation trend analysis is performed on the degradation candidate unit to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit.

[0010] In conjunction with the first aspect, in one possible design, the evaluation unit includes physical field parameter mapping relationships, sensor data interface specifications, and a degradation feature benchmark library.

[0011] In conjunction with the first aspect, in one possible design, the degradation feature benchmark library is constructed based on component fatigue crack propagation rate and tooth surface wear data obtained from multi-physics field coupling tests, combined with material damage laws obtained from microscopic characterization techniques.

[0012] In conjunction with the first aspect, in one possible design, the evaluation units are divided as follows: on the impeller side, there are blade root stress evaluation units, blade tip vibration evaluation units, and leading edge corrosion evaluation units; on the gearbox side, there are gear meshing evaluation units, bearing contact evaluation units, and lubricating oil film evaluation units.

[0013] In conjunction with the first aspect, in one possible design, the blade root stress assessment unit and the leading edge corrosion assessment unit are associated with distributed fiber optic sensors, the blade tip vibration assessment unit is associated with an accelerometer, the gear meshing assessment unit and the bearing contact assessment unit are associated with a MEMS accelerometer, and the lubricating oil film assessment unit is associated with an oil sensor and a temperature sensor.

[0014] In conjunction with the first aspect, in one possible design, based on the multi-field coupling state threshold and the multi-source sensing data, degradation candidate units are determined, including:

[0015] By comparing the multi-field coupling state threshold with the multi-source sensing data, multiple degradation risk units are obtained through screening.

[0016] For each of the aforementioned degradation risk units, multiphysics feature data are extracted;

[0017] Using the normal feature data of the corresponding unit in the degradation feature benchmark library, the multiphysics feature data is matched item by item to eliminate misjudged units caused by data fluctuations and retain degradation candidate units.

[0018] In conjunction with the first aspect, in one possible design, the multiphysics feature data includes aerodynamic field parameters, structural field parameters, and temperature field parameters.

[0019] In conjunction with the first aspect, in one possible design, the normal feature data of the corresponding unit in the degradation feature benchmark library are used to perform item-by-item matching on the multiphysics feature data, eliminating misjudged units caused by data fluctuations and retaining degradation candidate units, including:

[0020] If all key feature parameters of a risky unit exceed the normal feature data, it is determined to be a degenerate candidate unit.

[0021] If some parameters of a risk unit exceed the normal feature data but other parameters are normal, it is judged as a misjudged unit.

[0022] A short-term monitoring mode is activated for misjudged units. If the out-of-limit parameter returns to the normal range during the short-term monitoring process, the misjudgment is completely eliminated.

[0023] If the out-of-limit parameters continue to deteriorate, they will be reinstated into the list of degraded candidate units.

[0024] In conjunction with the first aspect, in one possible design, historical data corresponding to each of the aforementioned degradation candidate units is obtained, and degradation trend analysis is performed on the degradation candidate units based on this data to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit, including:

[0025] Collect historical operation and maintenance data and multi-source sensor data from various evaluation units;

[0026] For each candidate unit of degradation, its degradation history is analyzed, and combined with its historical operation and maintenance failure data, a degradation trend analysis model is constructed using a hybrid model of physical model and data-driven approach.

[0027] By establishing a degradation trend analysis model, the degradation rate of each degradation candidate unit is analyzed to determine the degradation mode;

[0028] Calculate the remaining useful life based on the degradation mode.

[0029] Secondly, the present invention also provides a wind turbine component life assessment system based on multi-physics coupling, comprising:

[0030] The module for unit division and sensor data acquisition, for wind turbine impeller gearbox system, divides multiple fine-grained evaluation units based on multi-physics field according to the principle of matching component functional zoning and physical field type, and acquires multi-source sensor data of each evaluation unit.

[0031] The threshold determination module determines the multi-field coupling state threshold of each evaluation unit based on the geometric features of the wind turbine impeller gearbox system and historical operation and maintenance fault data.

[0032] The degradation candidate unit determination module determines degradation candidate units for each of the evaluation units based on the multi-field coupling state threshold and the multi-source sensing data.

[0033] The remaining service life prediction module acquires historical data corresponding to each of the degradation candidate units, and uses this data to perform degradation trend analysis on the degradation candidate units to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit.

[0034] The technical solution of this invention can achieve the following technical effects:

[0035] This method comprehensively considers the nonlinear coupling effects of multiple physical fields by dividing the evaluation into multiple fine-grained evaluation units based on multi-physics fields. It accurately reflects the complex operating conditions of wind turbines in actual operation, avoiding the limitations of single-physics field evaluation and improving the accuracy of lifespan prediction. The evaluation units are divided according to the principle of matching component functional zoning with physical field types, facilitating refined management and evaluation for different components and physical fields, thus improving the relevance and effectiveness of the evaluation. By acquiring multi-source sensor data from each evaluation unit and combining it with material damage patterns obtained through microscopic characterization techniques to construct a degradation characteristic benchmark library, a systematic approach to data acquisition and processing is achieved. The fusion of multi-source sensor data improves the richness and accuracy of the data, while microscopic characterization techniques... The introduction of this approach enhances the scientific rigor and depth of data interpretation. Based on the geometric characteristics of the wind turbine rotor gearbox system and historical maintenance fault data, the multi-field coupling state thresholds for each evaluation unit are determined. Combined with actual operating experience, the threshold settings become more scientific and reasonable. Furthermore, as equipment operating data accumulates, the thresholds can be dynamically adjusted to adapt to changes in equipment status, improving the flexibility and accuracy of the evaluation. By comprehensively considering the nonlinear coupling effects of multiple physical fields, combined with fine-grained evaluation unit division and the system's data acquisition and processing flow, the degradation characteristics of equipment under different operating conditions can be accurately captured, thereby improving the accuracy of lifespan prediction, enabling early detection of potential faults, reducing unplanned downtime, optimizing maintenance plans, and lowering maintenance costs. Attached Figure Description

[0036] Figure 1 This is a flowchart of the present invention;

[0037] Figure 2 This is a structural diagram of a wind turbine component life assessment system based on multiphysics coupling. Detailed Implementation

[0038] This application will now be described with reference to the accompanying drawings.

[0039] like Figure 1 As shown, the wind turbine component life assessment method based on multiphysics coupling of the present invention specifically includes the following steps:

[0040] S1. For wind turbine impeller gearbox system, multiple fine-grained evaluation units based on multi-physics field are divided according to the principle of matching component functional zoning with physical field type, and multi-source sensor data of each evaluation unit are acquired.

[0041] S2. Based on the geometric characteristics and historical operation and maintenance fault data of the wind turbine impeller gearbox system, determine the multi-field coupling state threshold of each of the evaluation units;

[0042] S3. For each of the evaluation units, based on the multi-field coupling state threshold and the multi-source sensing data, determine the degradation candidate units;

[0043] S4. Obtain historical data corresponding to each of the degradation candidate units, and use this data to perform degradation trend analysis on the degradation candidate units to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit.

[0044] In this embodiment, the method comprehensively considers the nonlinear coupling effects of multiple physical fields by dividing the evaluation into multiple fine-grained evaluation units based on multi-physics fields. This accurately reflects the complex operating conditions of wind turbines in actual operation, avoiding the limitations of single-physics field evaluation and improving the accuracy of lifespan prediction. The evaluation units are divided according to the principle of matching component functional zoning with physical field types, facilitating refined management and evaluation for different components and physical fields, thus improving the targeting and effectiveness of the evaluation. By acquiring multi-source sensor data from each evaluation unit and combining it with material damage patterns obtained from microscopic characterization techniques to construct a degradation feature benchmark library, a systematic approach to data acquisition and processing is achieved. The fusion of multi-source sensor data improves the richness and accuracy of the data, while the microscopic... The introduction of characterization techniques enhances the scientific rigor and depth of data interpretation. Based on the geometric characteristics of the wind turbine rotor gearbox system and historical maintenance fault data, the multi-field coupling state thresholds for each evaluation unit are determined. Combined with actual operating experience, the threshold settings are made more scientific and reasonable. Furthermore, as equipment operating data accumulates, the thresholds can be dynamically adjusted to adapt to changes in equipment status, improving the flexibility and accuracy of the evaluation. By comprehensively considering the nonlinear coupling effects of multiple physical fields, combined with fine-grained evaluation unit division and the system's data acquisition and processing flow, the degradation characteristics of equipment under different operating conditions can be accurately captured, thereby improving the accuracy of life prediction, enabling early detection of potential faults, reducing unplanned downtime, optimizing maintenance plans, and lowering maintenance costs.

[0045] In some embodiments of the present invention, for step S1, for the wind turbine impeller gearbox system, multiple evaluation units based on multi-physics fine-grained are divided according to the principle of matching component functional zoning with physical field type, and multi-source sensing data of each evaluation unit are acquired.

[0046] The evaluation unit includes physical field parameter mapping relationships, sensor data interface specifications, and a degradation feature benchmark library;

[0047] The rotor gearbox system of a wind turbine is subjected to the combined action of multiple physical fields, and there are nonlinear coupling effects between these physical fields. By establishing the mapping relationship between the physical field parameters, the interaction between different physical fields can be quantified.

[0048] The fusion of multi-source sensor data requires a unified data interface standard to ensure seamless integration of data from different types of sensors; standardized interface design can reduce the complexity of data processing and improve system compatibility and scalability.

[0049] The degradation feature benchmark library is constructed based on component fatigue crack propagation rate and tooth surface wear data obtained from multiphysics coupling tests, combined with material damage laws obtained from microscopic characterization techniques. Different degradation stages are simulated on a multiphysics coupling test platform to obtain component fatigue crack propagation rate and tooth surface wear data. Material damage laws are analyzed using microscopic characterization techniques to establish the correlation between degradation features and microstructural changes. The test data and microscopic analysis results are integrated into the degradation feature benchmark library to provide a comparison standard for real-time monitoring.

[0050] The evaluation units are divided as follows: on the impeller side, there are blade root stress evaluation units, blade tip vibration evaluation units, and leading edge corrosion evaluation units; on the gearbox side, there are gear meshing evaluation units, bearing contact evaluation units, and lubricating oil film evaluation units.

[0051] The blade root stress assessment unit is used to monitor the stress distribution at the blade root and is linked to distributed fiber optic sensors to acquire blade root stress data in real time.

[0052] The blade tip vibration assessment unit is used to monitor the vibration at the blade tip, and is linked to an accelerometer to capture the dynamic response of the blade tip;

[0053] The leading edge corrosion assessment unit is used to assess the degree of corrosion at the impeller leading edge. It is associated with distributed fiber optic sensors to monitor strain and temperature changes on the leading edge surface, indirectly reflecting the corrosion situation.

[0054] The gear meshing evaluation unit is used to monitor vibration and shock during gear meshing, and is linked to a MEMS accelerometer to capture the dynamic characteristics of gear meshing;

[0055] The bearing contact evaluation unit is used to evaluate the stress and wear in the bearing contact area, and is linked to MEMS accelerometers and temperature sensors to monitor the bearing's operating status.

[0056] The lubricating oil film evaluation unit is used to monitor the thickness and temperature of the lubricating oil film, correlate with oil sensors and temperature sensors, evaluate the lubrication effect, and prevent wear caused by poor lubrication.

[0057] In this embodiment, by dividing the system into multiple fine-grained evaluation units based on physical field type and component function, the monitoring of the wind turbine rotor-gearbox system becomes more refined and targeted. Each unit focuses on specific physical field parameters, enabling the system to better capture the dynamic changes and degradation trends of each component. By establishing a unified data interface standard, data from different types of sensors can be seamlessly integrated, providing comprehensive and accurate system status information and ensuring effective monitoring of the interactions between each physical field. The establishment of a degradation feature benchmark library, combined with experimental data and microscopic analysis, provides a comparative standard for real-time monitoring. The system can compare real-time acquired data with historical degradation features during actual operation, promptly identifying degradation issues in the wind turbine. Potential faults; by dividing the system into specific assessment units, it can comprehensively monitor the status changes of key components of the wind turbine; each unit has specialized sensors and monitoring methods, which can specifically analyze and evaluate the health status of each component, improving the monitoring accuracy of the entire system; by establishing a mapping relationship between physical field parameters, it can quantify the nonlinear coupling effect between different physical fields; it can not only accurately capture the status of each component in the system, but also predict the degradation trend of each component under different operating conditions, enhancing the reliability and life prediction of the wind turbine; by monitoring the degradation characteristics and abnormal changes of each component in the early stage, it can provide predictive information for the operation and maintenance of the wind turbine, thereby avoiding sudden failures, improving operation and maintenance efficiency, and reducing downtime and maintenance costs.

[0058] In some embodiments of the present invention, for step S2, based on the geometric features of the wind turbine impeller gearbox system and historical operation and maintenance fault data, the multi-field coupling state threshold of each of the evaluation units is determined to distinguish the normal state and degradation risk state of the evaluation unit.

[0059] Collect geometric characteristic parameters of wind turbine impellers and gearboxes, including impeller blade length, width, thickness, and torsion angle; gearbox gear module, number of teeth, meshing angle; bearing diameter, contact angle, etc.

[0060] Extract historical fault records, including fault type, occurrence time, fault location, and maintenance records; acquire historical multi-source sensor data;

[0061] The collected data is cleaned to remove noise and outliers, and aligned according to timestamps to ensure data consistency and comparability;

[0062] A multiphysics coupling model of the impeller and gearbox system was established using finite element analysis software.

[0063] Using geometric features as model parameters, the influence of geometric features on multiphysics coupling effects is analyzed through parametric modeling.

[0064] Simulate the dynamic changes of multiphysics fields under different working conditions and analyze how geometric features affect the coupling relationship between physical fields;

[0065] Based on historical operation and maintenance failure data, identify the main failure modes and analyze the correlation between each failure mode and multiphysics coupling effects;

[0066] By using a multiphysics coupling model, the contribution of each physical field to the fault mode under different geometric characteristics is quantified;

[0067] Extracting degradation features related to failure modes from historical multi-source sensor data;

[0068] Based on historical operation and maintenance failure data and degradation characteristics, define the normal state and degradation risk state of the component;

[0069] By combining a multiphysics coupling model, the range of physical field parameters in the normal state and the degradation risk state under different geometric characteristics is analyzed;

[0070] Considering the influence of geometric features, thresholds are determined for components with different geometric parameters.

[0071] In this embodiment, by combining the geometric features of the impeller and gearbox with historical operation and maintenance fault data, the main fault modes can be accurately identified and analyzed, improving the accuracy of wind turbine fault detection. By establishing a multi-physics coupling model, the contribution of each physical field to the fault mode can be quantified, and the influence of geometric features on the coupling effect between physical fields can be analyzed in depth, improving the comprehensiveness and accuracy of component health status assessment. Based on the fusion of the multi-physics coupling model and historical data, the normal state and degradation risk state of the component can be clearly distinguished, thereby predicting possible equipment failures in advance and reducing downtime and maintenance costs. By considering the impact of different geometric features on the component state, personalized state assessment thresholds can be set for different wind turbines, avoiding overly coarse judgments and improving the reliability of fault diagnosis. By collecting and integrating multi-source sensor data, historical fault records, and other multi-dimensional information, multi-angle data support can be provided for fault diagnosis and risk assessment, improving the operation and maintenance efficiency and safety of wind turbines.

[0072] In some embodiments of the present invention, for step S3, for each of the evaluation units, a degradation candidate unit is determined based on the multi-field coupling state threshold and the multi-source sensing data;

[0073] By comparing the multi-field coupling state threshold with the multi-source sensing data, multiple degradation risk units are obtained through screening.

[0074] For each degradation risk unit, multiphysics feature data is extracted; the multiphysics feature data includes aerodynamic field parameters, structural field parameters, and temperature field parameters.

[0075] Using the normal feature data of the corresponding unit in the degradation feature benchmark library, the multiphysics feature data is matched item by item to eliminate misjudged units caused by data fluctuations and retain degradation candidate units.

[0076] The real-time sensing data of each evaluation unit is compared with the preset multi-field coupling state threshold to quickly mark the units that exceed the limit.

[0077] If a single field in a certain unit exceeds the limit, the correlation verification of adjacent physical field data is triggered;

[0078] When the blade root stress exceeds the limit, check the blade tip vibration data simultaneously. If the vibration amplitude increases synchronously, it indicates that the aerodynamic load fluctuation is causing stress concentration, thus confirming the risk.

[0079] When the gearbox oil temperature exceeds the limit, check the gear meshing frequency components. If there is a sudden increase in energy in a specific frequency band, it indicates that the wear has intensified, leading to the temperature rise, and the risk is confirmed.

[0080] The unit is marked as a degradation risk unit only when both layers of verification pass, to avoid misjudgment based on a single parameter;

[0081] For each degradation risk unit, multiphysics feature data is extracted; the multiphysics feature data includes aerodynamic field parameters, structural field parameters, and temperature field parameters;

[0082] The aerodynamic field parameters of the blade root stress assessment unit include impeller speed, angle of attack, and aerodynamic load fluctuation frequency; the structural field parameters include stress amplitude, stress cycle number, and stress-temperature coupling coefficient; and the temperature field parameters include the temperature gradient and thermal stress coefficient in the blade root region.

[0083] The aerodynamic field parameters of the gear meshing evaluation unit include impeller output torque fluctuation; the structural field parameters include gear meshing force and contact fatigue damage index; and the temperature field parameters include gearbox oil temperature and tooth surface friction heat generation rate.

[0084] Using the normal feature data of the corresponding unit in the degradation feature benchmark library, the multiphysics feature data is matched item by item to eliminate misjudged units caused by data fluctuations, and retain degradation candidate units, including:

[0085] The degradation characteristic benchmark library contains the distribution range of characteristic parameters of each evaluation unit under normal and degradation conditions;

[0086] A risk unit is identified as a degenerate candidate unit only when all of its key characteristic parameters exceed the normal range.

[0087] If some parameters exceed the limits but other parameters are normal, it is judged as a misjudged unit;

[0088] A short-term monitoring mode is activated for misjudged units. If the out-of-limit parameters return to the normal range, the misjudgment is completely eliminated.

[0089] If the parameters exceeding the limit continue or worsen, the unit will be reinstated to the list of degraded candidate units and its risk level will be upgraded.

[0090] In this embodiment, by combining multi-field coupling state thresholds with multi-source sensor data, degradation candidate units can be accurately identified, and multiple degradation risk units can be gradually screened based on data comparison, improving the accuracy and reliability of system diagnosis. An association verification mechanism for adjacent physical field data is introduced to ensure that each risk unit not only meets the standard under a single physical field but also requires collaborative verification across multiple physical fields, reducing the risk of misjudgment based on a single parameter. By matching each item with normal feature data in the degradation feature benchmark library, misjudgments caused by data fluctuations can be effectively eliminated, while retaining genuine degradation candidate units. Simultaneously, for misjudged units, the system automatically activates a short-term monitoring mode to further verify and correct the judgment results, improving the flexibility and adaptability of intelligent diagnosis. Comprehensive analysis using multi-physical field feature data from aerodynamic, structural, and temperature fields can comprehensively assess the degradation state of each evaluation unit, providing accurate diagnostic results. By quickly marking out-of-limit units and verifying them with the support of multi-physical field data, misjudgments caused by fluctuations in data from a single sensor are avoided, thereby improving the operation and maintenance efficiency of wind turbine units and ensuring the stability and safety of wind power equipment.

[0091] In some embodiments of the present invention, for step S4, historical data corresponding to each of the degradation candidate units is obtained, and degradation trend analysis is performed on the degradation candidate units to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit.

[0092] Collect historical operation and maintenance data and multi-source sensor data from various evaluation units to ensure data integrity and accuracy; preprocess this data, including noise reduction, missing value imputation, and standardization, to ensure data quality.

[0093] For each candidate degradation unit, its degradation history is analyzed, and combined with its historical operation and maintenance failure data, a degradation trend analysis model is constructed using a hybrid model of physical model and data-driven approach. Regression models in machine learning, including random forest regression and support vector machine regression, are used to model the degradation trend, or the degradation process is simulated using finite element analysis. Different degradation models are established for components such as impellers and gearboxes based on their physical characteristics and historical data.

[0094] By establishing a degradation trend analysis model, the degradation rate of each degradation candidate unit is analyzed to determine the degradation mode;

[0095] Based on the analyzed degradation patterns, the remaining service life is calculated. For each degradation unit, the difference from the degradation threshold is calculated by comparing historical data and the current state, thereby predicting the remaining service life of the wind turbine components.

[0096] In this embodiment, by combining regression models from machine learning with finite element analysis, the degradation process of wind turbine components can be modeled and simulated more accurately, thereby improving the accuracy of remaining service life prediction. Historical operation and maintenance data and multi-source sensor data are integrated to ensure data comprehensiveness and high quality, better reflecting the actual performance of wind turbine components under different operating conditions. Data preprocessing steps such as denoising, missing value imputation, and standardization ensure data integrity and accuracy, thus avoiding the impact of data quality issues on the analysis results. Specific degradation models are established for different wind turbine components to better adapt to the physical characteristics of each component, improving the precision of the analysis. Degradation trend analysis provides a deeper understanding of the degradation rate and patterns of each component, improving the accuracy of wind turbine fault prediction and maintenance decisions. Accurate prediction of remaining service life and early warning can optimize maintenance and replacement plans, avoid unnecessary downtime and sudden failures, ultimately extending equipment life and reducing operation and maintenance costs.

[0097] like Figure 2 As shown, the present invention also provides a wind turbine component life assessment system based on multi-physics coupling, which specifically includes the following modules;

[0098] The module for unit division and sensor data acquisition, for wind turbine impeller gearbox system, divides multiple fine-grained evaluation units based on multi-physics field according to the principle of matching component functional zoning and physical field type, and acquires multi-source sensor data of each evaluation unit.

[0099] The threshold determination module determines the multi-field coupling state threshold of each evaluation unit based on the geometric features of the wind turbine impeller gearbox system and historical operation and maintenance fault data.

[0100] The degradation candidate unit determination module determines degradation candidate units for each of the evaluation units based on the multi-field coupling state threshold and the multi-source sensing data.

[0101] The remaining service life prediction module acquires historical data corresponding to each of the degradation candidate units, and uses this data to perform degradation trend analysis on the degradation candidate units to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit.

[0102] In this embodiment, the system comprehensively considers the nonlinear coupling effects of multiple physical fields by dividing the system into multiple fine-grained evaluation units based on multi-physics fields. This accurately reflects the complex operating conditions of wind turbines in actual operation, avoiding the limitations of single-physics field evaluation and improving the accuracy of lifespan prediction. The evaluation units are divided according to the principle of matching component functional zoning with physical field types, facilitating refined management and evaluation for different components and physical fields, thus improving the targeting and effectiveness of the evaluation. By acquiring multi-source sensor data from each evaluation unit and combining it with material damage patterns obtained through microscopic characterization techniques to construct a degradation characteristic benchmark library, a systematic approach to data acquisition and processing is achieved. The fusion of multi-source sensor data improves the richness and accuracy of the data, while the microscopic... The introduction of characterization techniques enhances the scientific rigor and depth of data interpretation. Based on the geometric characteristics of the wind turbine rotor gearbox system and historical maintenance fault data, the multi-field coupling state thresholds for each evaluation unit are determined. Combined with actual operating experience, the threshold settings are made more scientific and reasonable. Furthermore, as equipment operating data accumulates, the thresholds can be dynamically adjusted to adapt to changes in equipment status, improving the flexibility and accuracy of the evaluation. By comprehensively considering the nonlinear coupling effects of multiple physical fields, combined with fine-grained evaluation unit division and the system's data acquisition and processing flow, the degradation characteristics of equipment under different operating conditions can be accurately captured, thereby improving the accuracy of life prediction, enabling early detection of potential faults, reducing unplanned downtime, optimizing maintenance plans, and lowering maintenance costs.

[0103] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for assessing the lifespan of wind turbine components based on multiphysics coupling, characterized in that, include: For wind turbine impeller gearbox systems, multiple fine-grained evaluation units based on multi-physics are divided according to the principle of matching component functional zoning with physical field type, and multi-source sensor data of each evaluation unit are acquired. The evaluation units are divided as follows: on the impeller side, there are blade root stress evaluation units, blade tip vibration evaluation units, and leading edge corrosion evaluation units; on the gearbox side, there are gear meshing evaluation units, bearing contact evaluation units, and lubricating oil film evaluation units. Based on the geometric characteristics and historical operation and maintenance fault data of the wind turbine impeller and gearbox system, the multi-field coupling state threshold of each evaluation unit is determined. This includes: collecting geometric characteristic parameters of the wind turbine impeller and gearbox, including the length, width, thickness, and torsion angle of the impeller blades, and the module, number of teeth, meshing angle, bearing diameter, and contact angle of the gearbox gears; extracting historical operation and maintenance fault data and historical multi-source sensor data; and performing data cleaning, noise reduction, outlier removal, and timestamp alignment. A multi-physics coupling model of the impeller and gearbox system is established using finite element analysis software, with the geometric characteristic parameters used as input to the multi-physics coupling model, and parameterized modeling is performed. This study analyzes the impact of geometric features on multiphysics coupling effects, simulates the dynamic changes of multiphysics and the coupling relationships between physical fields under different operating conditions, identifies major failure modes based on historical maintenance failure data, analyzes the correlation between failure modes and multiphysics coupling effects, quantifies the contribution of each physical field to failure modes under different geometric features using a multiphysics coupling model, and extracts degradation features related to failure modes from historical multi-source sensor data. Based on historical maintenance failure data, degradation features, and the multiphysics coupling model, the study analyzes the range of physical field parameters for normal and degradation risk states under different geometric features, and determines the multiphysics coupling state threshold for different geometric parameter components, considering the influence of geometric features. For each of the evaluation units, a degradation candidate unit is determined based on the multi-field coupling state threshold and the multi-source sensing data; Historical data corresponding to each degradation candidate unit is obtained, and degradation trend analysis is performed on the degradation candidate unit to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit.

2. The method for assessing the lifespan of wind turbine components based on multiphysics coupling according to claim 1, characterized in that, The evaluation unit includes physical field parameter mapping relationships, sensor data interface specifications, and a degradation feature benchmark library.

3. The method for assessing the lifespan of wind turbine components based on multiphysics coupling according to claim 2, characterized in that, The degradation feature benchmark library is constructed based on component fatigue crack propagation rate and tooth surface wear data obtained from multi-physics field coupling tests, combined with material damage laws obtained from microscopic characterization techniques.

4. The method for assessing the lifespan of wind turbine components based on multiphysics coupling according to claim 1, characterized in that, The blade root stress assessment unit and the leading edge corrosion assessment unit are associated with distributed fiber optic sensors, and the blade tip vibration assessment unit is associated with an accelerometer; the gear meshing assessment unit and the bearing contact assessment unit are associated with a MEMS accelerometer, and the lubricating oil film assessment unit is associated with an oil sensor and a temperature sensor.

5. The method for assessing the lifespan of wind turbine components based on multiphysics coupling according to claim 2, characterized in that, Based on the multi-field coupling state threshold and the multi-source sensing data, degradation candidate units are determined, including: By comparing the multi-field coupling state threshold with the multi-source sensing data, multiple degradation risk units are obtained through screening. For each of the aforementioned degradation risk units, multiphysics feature data are extracted; Using the normal feature data of the corresponding unit in the degradation feature benchmark library, the multiphysics feature data is matched item by item to eliminate misjudged units caused by data fluctuations and retain degradation candidate units.

6. The wind turbine component life assessment method based on multiphysics coupling according to claim 5, characterized in that, The multiphysics field characteristic data includes aerodynamic field parameters, structural field parameters, and temperature field parameters.

7. The method for assessing the lifespan of wind turbine components based on multiphysics coupling according to claim 5, characterized in that, Using the normal feature data of the corresponding unit in the degradation feature benchmark library, the multiphysics feature data is matched item by item to eliminate misjudged units caused by data fluctuations, and retain degradation candidate units, including: If all key feature parameters of a risky unit exceed the normal feature data, it is determined to be a degenerate candidate unit. If some parameters of a risk unit exceed the normal feature data but other parameters are normal, it is judged as a misjudged unit. A short-term monitoring mode is activated for misjudged units. If the out-of-limit parameter returns to the normal range during the short-term monitoring process, the misjudgment is completely eliminated. If the out-of-limit parameters continue to deteriorate, they will be reinstated into the list of degraded candidate units.

8. The method for assessing the lifespan of wind turbine components based on multiphysics coupling according to claim 1, characterized in that, Historical data corresponding to each of the degradation candidate units is obtained, and degradation trend analysis is performed on the degradation candidate units based on this data to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit, including: Collect historical operation and maintenance data and multi-source sensor data from various evaluation units; For each candidate unit of degradation, its degradation history is analyzed, and combined with its historical operation and maintenance failure data, a degradation trend analysis model is constructed using a hybrid model of physical model and data-driven approach. By establishing a degradation trend analysis model, the degradation rate of each degradation candidate unit is analyzed to determine the degradation mode; Calculate the remaining useful life based on the degradation mode.

9. A wind turbine component life assessment system based on multiphysics coupling, wherein the system is applied to the wind turbine component life assessment method based on multiphysics coupling as described in claim 1, characterized in that, The system includes: The module for unit division and sensor data acquisition, for wind turbine impeller gearbox system, divides multiple fine-grained evaluation units based on multi-physics field according to the principle of matching component functional zoning and physical field type, and acquires multi-source sensor data of each evaluation unit. The threshold determination module determines the multi-field coupling state threshold of each evaluation unit based on the geometric features of the wind turbine impeller gearbox system and historical operation and maintenance fault data. The degradation candidate unit determination module determines degradation candidate units for each of the evaluation units based on the multi-field coupling state threshold and the multi-source sensing data. The remaining service life prediction module acquires historical data corresponding to each of the degradation candidate units, and uses this data to perform degradation trend analysis on the degradation candidate units to obtain the remaining service life of the wind turbine component corresponding to the degradation candidate unit.

Images