A method, system, equipment, and medium for assessing the condition of a wind turbine drivetrain.

Abstract

This specification provides a method, system, device, and medium for assessing the condition of a wind turbine drivetrain. The method includes: collecting multi-source heterogeneous monitoring data of the wind turbine drivetrain, wherein the monitoring data includes at least macroscopic operating parameter data and vibration waveform data; performing vibration state analysis and automatic fault diagnosis of the mechanical components of the drivetrain based on the vibration waveform data to obtain a first assessment result; assessing the health status of the drivetrain subsystem based on the macroscopic operating parameter data using a pre-trained state estimation model to obtain a second assessment result; and fusing the first assessment result and the second assessment result to generate a comprehensive condition assessment report of the wind turbine drivetrain.

Description

Technical Field

[0001] This document relates to the field of wind turbine drivetrain health monitoring technology, and in particular to a method, system, equipment and medium for assessing the condition of wind turbine drivetrain. Background Technology

[0002] As the core path for energy transfer, the health of the wind turbine drivetrain directly affects the reliability and power generation efficiency of the entire unit. The drivetrain includes key components such as the main shaft, gearbox, couplings, and generator. It is subjected to complex alternating loads over long periods, making it prone to mechanical failures such as bearing wear, gear pitting, and shaft misalignment. Traditional maintenance methods rely primarily on periodic inspections and the experience-based judgment of operators, resulting in long planned downtime, high maintenance costs, and the inability to detect early-stage, latent faults in a timely manner, failing to meet the demands of modern, lean operation and maintenance in wind farms.

[0003] With the widespread application of condition monitoring technology in the wind power sector, two main independent technical approaches currently exist: one is the analysis method based on macroscopic operating parameters of SCADA systems, which focuses on anomaly alarms through thresholds or simple statistical models, but is not sensitive to early mechanical faults; the other is the waveform data analysis method based on vibration monitoring systems (CMS), which can effectively diagnose specific mechanical faults through spectrum, envelope spectrum, and other means, but its analysis is often conducted in isolation, lacking correlation with the overall operating conditions of the unit, and the diagnostic knowledge relies on expert experience, making it difficult to achieve automated and standardized condition assessment. These two approaches create data silos, with single evaluation dimensions, making it impossible to conduct integrated and quantitative assessments of the overall health status of the drivetrain, and even more difficult to achieve correlation mapping and early warning from microscopic component degradation to macroscopic system performance decline.

[0004] Therefore, there is an urgent need for a comprehensive assessment method for the transmission chain status of wind turbines that can effectively integrate multi-source heterogeneous monitoring data, take into account both microscopic mechanical conditions and macroscopic operating performance, and achieve automated and intelligent assessment. This method would overcome the limitations of the aforementioned one-sided assessment, delayed diagnosis, and reliance on expert experience, and provide reliable technical support for predictive maintenance. Summary of the Invention

[0005] This specification provides one or more embodiments of a method for assessing the condition of a wind turbine drivetrain, including: S1. Collect multi-source heterogeneous monitoring data of the wind turbine drive train, wherein the monitoring data includes at least macroscopic operating parameter data and vibration waveform data; S2. Based on the vibration waveform data, perform vibration state analysis and automatic fault diagnosis of the mechanical components of the transmission chain to obtain the first evaluation result; S3. Based on the macroscopic operating parameter data, the health status of the transmission chain subsystem is assessed using a pre-trained state estimation model to obtain a second assessment result; S4. Integrate the first evaluation result and the second evaluation result to generate a comprehensive status evaluation report of the wind turbine drive chain.

[0006] Furthermore, the collection of multi-source heterogeneous monitoring data of the wind turbine drivetrain specifically includes: Collect macroscopic operating parameter data at the ten-minute level from the SCADA (Supervisory Control and Data Acquisition) system; It also collects vibration waveform data from the Condition Monitoring System (CMS), wherein the sampling frequency of the vibration waveform data is not less than 12800 Hz.

[0007] Furthermore, the vibration state analysis of the transmission chain mechanical components based on the vibration waveform data includes: Time-domain analysis, frequency-domain analysis, envelope spectrum analysis, and trend analysis are performed on the vibration waveform data to extract time-domain statistical indicators and frequency-domain characteristic frequencies that reflect the microscopic health status of the mechanical components in the transmission chain. The time-domain analysis specifically involves calculating time-domain statistical indicators, including RMS, peak value, kurtosis, and skewness indices. The frequency domain analysis specifically involves identifying rotating frequency harmonics and component characteristic fault frequencies in the frequency domain through spectral transformation. The envelope spectrum analysis specifically involves performing a Hilbert transform on the vibration signal to extract the envelope signal, and then performing a spectrum analysis on the envelope signal to highlight the characteristic frequencies of bearing failure and their harmonic components. The trend analysis specifically involves observing the time-series trend of the time-domain statistical indicators or characteristic frequency amplitudes.

[0008] Furthermore, the identification of rotating frequency harmonics and component characteristic fault frequencies in the frequency domain through spectrum transformation specifically includes: The vibration time-domain signal is converted into a frequency-domain signal using Fast Fourier Transform (FFT). Based on the design parameters of bearings and gears in the transmission chain, the harmonics of their rotational frequencies and the characteristic fault frequencies of bearings and gears are marked in the spectrum.

[0009] Furthermore, based on the vibration waveform data, performing automatic fault diagnosis specifically includes: Based on the preprocessed vibration waveform data, the acceleration spectrum, velocity spectrum, and envelope spectrum of the vibration signal are calculated. Extract feature values ​​related to the preset fault type from the acceleration spectrum, velocity spectrum, and envelope spectrum; The feature values ​​are compared with preset fault thresholds to automatically diagnose the fault types of bearings and gears in the transmission chain, as well as imbalance, misalignment, and looseness faults in rotating machinery. The preset fault thresholds include vibration velocity thresholds, vibration acceleration thresholds, total envelope thresholds, and impact thresholds, and the thresholds include at least two levels: alarm value and danger value. Based on the diagnostic results, output the fault type identifier and the corresponding severity level.

[0010] Furthermore, the step of assessing the health status of the transmission chain subsystem based on the macroscopic operating parameter data using a pre-trained state estimation model to obtain a second assessment result specifically includes: Using historical normal operation macroscopic operating parameter data of the target wind turbine, a multivariate state estimation technique MSET model is trained. Real-time macroscopic operating parameter data are input into the MSET model, and the sequence probability ratio test algorithm SPRT is used to calculate the feature value representing the degree of state deviation, thereby evaluating the health score of the transmission chain system.

[0011] Furthermore, the macroscopic operating parameter data includes: main bearing temperature, gearbox oil temperature, gearbox bearing temperature, generator drive end bearing temperature, and generator non-drive end bearing temperature.

[0012] This specification provides one or more embodiments of a wind turbine drivetrain condition assessment system, including: Data acquisition module: used to collect multi-source heterogeneous monitoring data of the wind turbine drive train, the monitoring data including at least macroscopic operating parameter data and vibration waveform data; First evaluation module: used to perform vibration state analysis and automatic fault diagnosis of mechanical components in the transmission chain based on the vibration waveform data, and obtain the first evaluation result; The second evaluation module is used to evaluate the health status of the transmission chain subsystem based on the macroscopic operating parameter data and through a pre-trained state estimation model to obtain the second evaluation result. Comprehensive evaluation module: used to integrate the first evaluation result and the second evaluation result to generate a comprehensive status evaluation report of the wind turbine drive chain.

[0013] This specification provides one or more embodiments of an electronic device, including: Processor; and, A memory is configured to store computer-executable instructions, which, when executed, cause the processor to implement the steps of the wind turbine drivetrain state assessment method described above.

[0014] This specification provides one or more embodiments of a storage medium for storing computer-executable instructions that, when executed, implement the steps of the wind turbine drivetrain condition assessment method described above.

[0015] By employing the embodiments of this invention, the limitations of traditional single-data-source assessments are overcome. It can simultaneously capture performance degradation and early deterioration of mechanical components, significantly improving the comprehensiveness of condition monitoring and the sensitivity of fault identification. It achieves precise location and severity classification of specific faults, as well as quantitative evaluation of the overall operating trend of the drivetrain subsystem, balancing the depth of fault diagnosis with health management. By generating a comprehensive condition assessment report, it presents the multi-dimensional analysis results in a unified manner, providing operation and maintenance personnel with intuitive and reliable decision support. This helps reduce unplanned downtime, decrease maintenance costs, and improve the power generation reliability of wind turbines, achieving a comprehensive, accurate, and forward-looking assessment of the drivetrain's health status.

[0016] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in one or more embodiments of this specification or in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this specification. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart of a wind turbine drivetrain condition assessment method provided for one or more embodiments of this specification; Figure 2 A schematic diagram of the composition of a wind turbine drivetrain condition assessment system provided for one or more embodiments of this specification; Figure 3 This is a schematic diagram of the structure of an electronic device provided for one or more embodiments of this specification. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of this document.

[0020] Method Implementation Examples According to embodiments of the present invention, a method for assessing the condition of a wind turbine drivetrain is provided. Figure 1 A flowchart illustrating a method for assessing the condition of a wind turbine drivetrain, provided for one or more embodiments of this specification, is shown below. Figure 1 As shown, the wind turbine drivetrain condition assessment method according to an embodiment of the present invention specifically includes: S1. Collect multi-source heterogeneous monitoring data of the wind turbine drive chain, wherein the monitoring data includes at least macroscopic operating parameter data and vibration waveform data.

[0021] Collecting and integrating monitoring data of different types and sources from the wind turbine drivetrain, i.e., multi-source heterogeneous monitoring data. Here, "multi-source" refers to data originating from multiple independent monitoring systems, and "heterogeneous" refers to data with fundamental differences in structure, frequency, and physical meaning. Specifically, this step mainly involves the simultaneous collection of two types of data: The first category is macroscopic operating parameter data provided by the SCADA (Supervisory Control and Data Acquisition) system. The SCADA system collects and records macroscopic parameters reflecting the overall operating status of the unit at fixed time intervals (ten minutes in this embodiment). These parameters mainly include temperature data such as main bearing temperature, gearbox front and rear bearing temperatures, gearbox oil temperature, generator front and rear bearing temperatures, and winding temperature; speed and power data such as generator speed, main shaft speed, and active / reactive power; and environmental data, including wind speed and cabin temperature.

[0022] The second category is vibration waveform data collected by the Condition Monitoring System (CMS). The CMS system continuously collects raw waveform signals of vibration acceleration at a sampling frequency of no less than 12800 Hz using accelerometers installed at key parts of the transmission chain.

[0023] During the data acquisition process, based on the specified wind farm, turbine number, and time range, SCADA ten-minute data records and CMS raw vibration waveform data for the corresponding time period are simultaneously retrieved from the wind farm's central database or the corresponding data interface. These two types of data are then preprocessed, including standardizing the data to the same time reference point and format.

[0024] S2. Based on the vibration waveform data, perform vibration state analysis and automatic fault diagnosis of the mechanical components of the transmission chain to obtain the first evaluation result.

[0025] Specifically, it is divided into two stages: the first stage is to perform vibration state analysis of the mechanical components of the transmission chain based on the vibration waveform data; the second stage is to perform automatic fault diagnosis of the mechanical components of the transmission chain based on the vibration waveform data.

[0026] In the vibration state analysis phase, four analyses are performed in parallel on the preprocessed vibration waveform data, including time-domain analysis, frequency-domain analysis, envelope spectrum analysis, and trend analysis. First, time-domain analysis is performed, which involves directly performing statistical calculations on the time-series data of the original vibration signal to calculate various time-domain indices, specifically including mean, maximum, minimum, effective value (RMS), peak value, waveform, impulse, margin, skewness, and kurtosis. Second, frequency-domain analysis is performed, the core of which is to use Fast Fourier Transform (FFT) to convert the time-domain vibration signal to the frequency domain, obtaining a spectrum displaying each frequency component and its amplitude. Based on this, according to bearing parameters—that is, the geometric design parameters of the bearings in the transmission chain, such as the inner ring, outer ring, rolling elements, cage, and gears, such as the number of balls, pitch diameter, contact angle, and number of teeth—the corresponding theoretical fault characteristic frequencies and their harmonics are automatically calculated. These characteristic frequency lines are automatically marked on the spectrum, allowing for rapid observation of abnormal energy peaks at specific fault frequencies, thereby locating the faulty component. Envelope spectrum analysis is performed by applying a Hilbert transform to the original vibration signal to extract the envelope signal whose amplitude changes over time. This envelope signal is then subjected to FFT analysis to obtain the envelope spectrum, which clearly highlights the characteristic frequencies of bearing faults and their harmonic components, significantly improving the detection rate of early faults. Finally, trend analysis is conducted by plotting the amplitudes of the aforementioned time-domain indicators or key frequency-domain characteristic frequencies as trend curves in chronological order. By analyzing the trend of these curves, the evolution of the component's condition can be determined, enabling predictive early warning.

[0027] In the automatic fault diagnosis stage, based on the preprocessed vibration waveform data, the acceleration spectrum, velocity spectrum, and envelope spectrum of the vibration signal are calculated. Feature values ​​related to preset fault types are extracted from the acceleration spectrum, velocity spectrum, and envelope spectrum. Preset fault types include faults in various parts of the bearing, gear wear / tooth breakage, rotor imbalance, shaft misalignment, and mechanical loosening. These feature values ​​are compared with a preset fault threshold library, which references the vibration diagnosis standard VDI_3834. In this embodiment, the preset fault thresholds include vibration velocity threshold, vibration acceleration threshold, total envelope value threshold, and impact threshold. For key parameters such as vibration velocity, vibration acceleration, total envelope value, and impact value, at least two threshold levels, alarm and danger, are set. When the feature value corresponding to a certain fault mode exceeds the alarm threshold, it is initially determined that there is a suspicion of this type of fault. If it exceeds a higher danger threshold, it is determined that the fault has developed to a severe level. Based on the diagnosis results, the fault type identifier and the corresponding severity level are output.

[0028] S3. Based on the macroscopic operating parameter data, the health status of the transmission chain subsystem is assessed using a pre-trained state estimation model to obtain a second assessment result.

[0029] Using historical normal operation macroscopic operating parameter data of the target wind turbine, a multivariate state estimation (MSET) model is trained. The macroscopic operating parameter data includes: main bearing temperature, gearbox oil temperature, gearbox bearing temperature, generator drive-end bearing temperature, and generator non-drive-end bearing temperature. The MSET model is trained offline, learning health-related patterns. The specific execution process is as follows: First, a sufficiently long period of macroscopic operating parameter data, confirmed as normal operation, is selected from the historical database of the target wind turbine as the training set. This data comprehensively covers key temperature rise points in the drivetrain, mainly including main bearing temperature, gearbox oil temperature, gearbox bearing temperature, generator drive-end bearing temperature, and generator non-drive-end bearing temperature. The MSET algorithm constructs a memory matrix representing the health state from these parameter values ​​at all times within this period.

[0030] After completing offline training and obtaining the MSET model for this specific unit, the online health assessment phase begins. Specifically, the real-time collected temperature data is input into the trained MSET model. The MSET model identifies several historical health state points from the historical memory matrix that are most similar to the current input vector. Using a weighted interpolation algorithm, it calculates the estimated health temperature state of the drivetrain under the current operating conditions. Next, the actual temperature measurements are compared with the MSET model's estimates to obtain a set of residuals. These residuals are then analyzed online using the Sequence Probability Ratio Test (SPRT) algorithm. The residual sequence is used to calculate eigenvalues ​​or SPRT scores representing the degree of deviation from the status. These scores visually represent the statistical significance of the current real-time data deviating from the historical health pattern: a score closer to 0 indicates a closer operating status to the historical health baseline; a positive and continuously increasing score indicates that the system is beginning to exhibit detectable abnormal deviations. Finally, these SPRT eigenvalues ​​are mapped to an intuitive health score, which is output as the second assessment result.

[0031] S4. Integrate the first evaluation result and the second evaluation result to generate a comprehensive status evaluation report of the wind turbine drive chain.

[0032] The results from the vibration analysis step, the automatic fault diagnosis step, and the health status assessment step are integrated. The results from different sources are weighted and comprehensively evaluated. The CRITIC method is used to determine the feature weight ratios in each result, with the weight ratios of each result tentatively assumed to be equal. A comprehensive health status assessment report is generated, including the overall health status of the drivetrain, a list of component faults, and trends in key parameters. To ensure the accuracy of the health score, pre-training is performed based on one year of historical data.

[0033] The completed evaluation results for a specific wind field and a specific time period are cached. When an evaluation request of the same scope is received, the cached results are directly invoked.

[0034] The beneficial effects of this invention are as follows: By employing the embodiments of this invention, the limitations of traditional single-data-source assessments are overcome. It can simultaneously capture performance degradation and early deterioration of mechanical components, significantly improving the comprehensiveness of condition monitoring and the sensitivity of fault identification. It achieves precise location and severity classification of specific faults, as well as quantitative evaluation of the overall operating trend of the drivetrain subsystem, balancing the depth of fault diagnosis with health management. By generating a comprehensive condition assessment report, it presents the multi-dimensional analysis results in a unified manner, providing operation and maintenance personnel with intuitive and reliable decision support. This helps reduce unplanned downtime, decrease maintenance costs, and improve the power generation reliability of wind turbines, achieving a comprehensive, accurate, and forward-looking assessment of the drivetrain's health status.

[0035] System Implementation Examples According to embodiments of the present invention, a wind turbine drivetrain condition assessment system is provided. Figure 2 A schematic diagram illustrating the composition of a wind turbine drivetrain condition assessment system provided in one or more embodiments of this specification, as shown below. Figure 2 As shown, the wind turbine drivetrain condition assessment system according to an embodiment of the present invention specifically includes: Data acquisition module 20: used to acquire multi-source heterogeneous monitoring data of the wind turbine drive train, the monitoring data including at least macroscopic operating parameter data and vibration waveform data; First evaluation module 22: used to perform vibration state analysis and automatic fault diagnosis of transmission chain mechanical components based on the vibration waveform data, and obtain the first evaluation result; Second evaluation module 24: Used to evaluate the health status of the transmission chain subsystem based on the macroscopic operating parameter data and through a pre-trained state estimation model to obtain a second evaluation result; Comprehensive evaluation module 26: used to integrate the first evaluation result and the second evaluation result to generate a comprehensive status evaluation report of the wind turbine drive chain.

[0036] The embodiments of the present invention are system embodiments corresponding to the above method embodiments. The specific operation of each module can be understood by referring to the description of the method embodiments, and will not be repeated here.

[0037] Device Example 1 This invention provides an electronic device, such as... Figure 3 As shown, it includes: a memory 30, a processor 32, and a computer program stored in the memory 30 and executable on the processor 32. When the computer program is executed by the processor 32, it performs the following method steps: S1. Collect multi-source heterogeneous monitoring data of the wind turbine drive train, wherein the monitoring data includes at least macroscopic operating parameter data and vibration waveform data; S2. Based on the vibration waveform data, perform vibration state analysis and automatic fault diagnosis of the mechanical components of the transmission chain to obtain the first evaluation result; S3. Based on the macroscopic operating parameter data, the health status of the transmission chain subsystem is assessed using a pre-trained state estimation model to obtain a second assessment result; S4. Integrate the first evaluation result and the second evaluation result to generate a comprehensive status evaluation report of the wind turbine drive chain.

[0038] Device Example 2 This invention provides a computer-readable storage medium storing an information transmission implementation program. When executed by a processor 32, the program performs the following method steps: S1. Collect multi-source heterogeneous monitoring data of the wind turbine drive train, wherein the monitoring data includes at least macroscopic operating parameter data and vibration waveform data; S2. Based on the vibration waveform data, perform vibration state analysis and automatic fault diagnosis of the mechanical components of the transmission chain to obtain the first evaluation result; S3. Based on the macroscopic operating parameter data, the health status of the transmission chain subsystem is assessed using a pre-trained state estimation model to obtain a second assessment result; S4. Integrate the first evaluation result and the second evaluation result to generate a comprehensive status evaluation report of the wind turbine drive chain.

[0039] The computer-readable storage media described in this embodiment include, but are not limited to, ROM, RAM, disk, or optical disk.

[0040] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for assessing the condition of a wind turbine drivetrain, characterized in that, include: S1. Collect multi-source heterogeneous monitoring data of the wind turbine drive train, wherein the monitoring data includes at least macroscopic operating parameter data and vibration waveform data; S2. Based on the vibration waveform data, perform vibration state analysis and automatic fault diagnosis of the mechanical components of the transmission chain to obtain the first evaluation result; S3. Based on the macroscopic operating parameter data, the health status of the transmission chain subsystem is assessed using a pre-trained state estimation model to obtain a second assessment result; S4. Integrate the first evaluation result and the second evaluation result to generate a comprehensive status evaluation report of the wind turbine drive chain.

2. The method according to claim 1, characterized in that, The collection of multi-source heterogeneous monitoring data of the wind turbine drivetrain specifically includes: Collect macroscopic operating parameter data at the ten-minute level from the SCADA (Supervisory Control and Data Acquisition) system; It also collects vibration waveform data from the Condition Monitoring System (CMS), wherein the sampling frequency of the vibration waveform data is not less than 12800 Hz.

3. The method according to claim 1, characterized in that, The vibration state analysis of the transmission chain mechanical components based on the vibration waveform data includes: Time-domain analysis, frequency-domain analysis, envelope spectrum analysis, and trend analysis are performed on the vibration waveform data to extract time-domain statistical indicators and frequency-domain characteristic frequencies that reflect the microscopic health status of the mechanical components in the transmission chain. The time-domain analysis specifically involves calculating time-domain statistical indicators, including RMS, peak value, kurtosis, and skewness indices. The frequency domain analysis specifically involves identifying rotating frequency harmonics and component characteristic fault frequencies in the frequency domain through spectral transformation. The envelope spectrum analysis specifically involves performing a Hilbert transform on the vibration signal to extract the envelope signal, and then performing a spectrum analysis on the envelope signal to highlight the characteristic frequencies of bearing failure and their harmonic components. The trend analysis specifically involves observing the time-series trend of the time-domain statistical indicators or characteristic frequency amplitudes.

4. The method according to claim 3, characterized in that, The identification of rotating frequency harmonics and component characteristic fault frequencies in the frequency domain through spectrum transformation specifically includes: The vibration time-domain signal is converted into a frequency-domain signal using Fast Fourier Transform (FFT). Based on the design parameters of bearings and gears in the transmission chain, the harmonics of their rotational frequencies and the characteristic fault frequencies of bearings and gears are marked in the spectrum.

5. The method according to claim 1, characterized in that, Based on the vibration waveform data, the automatic fault diagnosis specifically includes: Based on the preprocessed vibration waveform data, the acceleration spectrum, velocity spectrum, and envelope spectrum of the vibration signal are calculated. Extract feature values ​​related to the preset fault type from the acceleration spectrum, velocity spectrum, and envelope spectrum; The feature values ​​are compared with preset fault thresholds to automatically diagnose the fault types of bearings and gears in the transmission chain, as well as imbalance, misalignment, and looseness faults in rotating machinery. The preset fault thresholds include vibration velocity thresholds, vibration acceleration thresholds, total envelope thresholds, and impact thresholds, and the thresholds include at least two levels: alarm value and danger value. Based on the diagnostic results, output the fault type identifier and the corresponding severity level.

6. The method according to claim 1, characterized in that, The step of assessing the health status of the transmission chain subsystem based on the macroscopic operating parameter data and obtaining the second assessment result through a pre-trained state estimation model specifically includes: Using historical normal operation macroscopic operating parameter data of the target wind turbine, a multivariate state estimation technique MSET model is trained; Real-time macroscopic operating parameter data are input into the MSET model, and the sequence probability ratio test algorithm SPRT is used to calculate the feature value representing the degree of state deviation, thereby evaluating the health score of the drive chain system.

7. The method according to claim 1, characterized in that, The macroscopic operating parameter data include: main bearing temperature, gearbox oil temperature, gearbox bearing temperature, generator drive end bearing temperature, and generator non-drive end bearing temperature.

8. A wind turbine drivetrain condition assessment system, characterized in that, include: Data acquisition module: used to collect multi-source heterogeneous monitoring data of the wind turbine drive train, the monitoring data including at least macroscopic operating parameter data and vibration waveform data; First evaluation module: used to perform vibration state analysis and automatic fault diagnosis of mechanical components in the transmission chain based on the vibration waveform data, and obtain the first evaluation result; The second evaluation module is used to evaluate the health status of the transmission chain subsystem based on the macroscopic operating parameter data and through a pre-trained state estimation model to obtain the second evaluation result. Comprehensive evaluation module: used to integrate the first evaluation result and the second evaluation result to generate a comprehensive status evaluation report of the wind turbine drive chain.

9. An electronic device, characterized in that, include: processor; as well as, A memory configured to store computer-executable instructions, which, when executed, cause the processor to implement the steps of the wind turbine drivetrain condition assessment method as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, Used to store computer-executable instructions, which, when executed, implement the steps of the wind turbine drivetrain condition assessment method as described in any one of claims 1 to 7.

Images