Wind power bearing vibration control monitoring method and device based on piezoelectricity and metal rubber

Abstract

The invention discloses a wind power bearing vibration control monitoring method and device based on piezoelectricity and metal rubber, and belongs to the technical field of wind power bearing health management, and the method comprises the following steps: bearing vibration monitoring, characteristic frequency band division, frequency band division correction and monitoring result output. According to the method, the vibration signals of the wind power bearing are collected, the vibration characteristics of the bearing under the normal operation working condition and the abnormal operation working condition are distinguished and analyzed, and the characteristic frequency interval is further subjected to self-adaptive expansion correction under the abnormal working condition, so that the influence of complex load and rotating speed fluctuation on bearing characteristic extraction is effectively reduced; and the accuracy of bearing vibration monitoring results is improved. Besides, under the complex load condition of the offshore wind turbine generator, a characteristic frequency verification and energy gain fusion mechanism is introduced to dynamically update the characteristic frequency band of the bearing, the adaptive capacity of the monitoring method to complex working conditions is enhanced, and the vibration state of the wind power bearing is effectively monitored and controlled.

Description

Technical Field

[0001] This invention relates to the field of wind turbine bearing health management technology, and in particular to a method and device for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber. Background Technology

[0002] Currently, wind turbine generators are developing towards higher power density, longer lifespan, and adaptability to complex environments. Key components such as main shaft bearings, pitch bearings, and yaw bearings endure low-speed heavy loads, alternating loads, and multi-source coupled vibrations during operation. Bearing vibration can cause problems such as raceway fatigue, pitting and spalling, and deterioration of lubrication conditions, and may also amplify transmission chain vibrations, reducing the operational stability of the wind turbine and increasing maintenance costs. Therefore, vibration monitoring and control of wind turbine bearings has become an important technical aspect of wind power equipment operation and maintenance.

[0003] In existing technologies, vibration control and monitoring of wind turbine bearings typically follow these steps: First, vibration sensors are placed near the main shaft bearing, pitch bearing, or yaw bearing of the wind turbine generator to collect vibration signals during operation. Second, the vibration signals output by the sensors are sampled and recorded. During the acquisition process, sampling frequency, sampling period, and filtering parameters are usually set to obtain raw vibration data under bearing operating conditions. Then, the collected raw vibration data is preprocessed, including noise reduction, filtering, and signal segmentation, to reduce interference from environmental noise, structural resonance, and vibrations from other components on the bearing vibration signals. Based on this, time-domain analysis, frequency-domain analysis, or envelope analysis are performed on the processed raw vibration data to extract parameters such as vibration amplitude and characteristic frequencies. Subsequently, the extracted vibration characteristic parameters are compared with preset thresholds or empirical models to assess the bearing's operating status. When the vibration amplitude or characteristic parameters exceed the set range, an abnormal bearing condition is determined, and alarm information or operational suggestions are output. In terms of vibration control, existing technologies mostly employ passive vibration reduction methods, dissipating vibration energy by installing rubber pads, elastic supports, or damping elements in the bearing support structure. Finally, maintenance personnel, based on the provided vibration data and alarm information, combined with the results of manual inspections, perform maintenance, lubrication, or replacement of the wind turbine bearings to ensure the normal operation of the wind turbine generator set.

[0004] For example, Chinese patent application CN120293523A discloses a method, device, electronic device, and storage medium for diagnosing wind turbine bearing faults. The method includes: acquiring the cyclic frequency corresponding to the original vibration signal of the wind turbine bearing to be diagnosed; performing channelization processing on the original vibration signal using a frequency-shifting filter based on the cyclic frequency to obtain sub-channel signals, and determining the signal to be filtered based on the sub-channel signals; using the original vibration signal as a reference signal, optimizing the filter coefficients of the frequency-shifting filter based on the maximum Versoria criterion to obtain the optimal filter corresponding to the original vibration signal, and performing frequency-shifting filtering on the signal to be filtered using the optimal filter to obtain bearing fault characteristic signals; and analyzing the bearing fault characteristic signals to obtain the fault diagnosis result of the wind turbine bearing.

[0005] However, in the process of implementing the inventive technical solution in the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems:

[0006] During the operation of offshore wind turbines, the turbine structure experiences complex overall vibrations under the influence of wind, wave, and operational loads. Furthermore, this overall vibration is transmitted to the bearings via the transmission chain and support structure, inevitably superimposing a large amount of structural vibration components originating from outside the bearing itself into the bearing vibration signal. In this situation, the bearing vibration signal simultaneously contains operating information from the bearing itself and coupled vibrations from related structures caused by the external excitation environment. These two components superimpose each other in the frequency domain, making them difficult to distinguish effectively.

[0007] Especially under low-speed, heavy-load or frequent load fluctuations, the bearing body has a low vibration frequency and weak vibration energy. Its characteristic information is easily masked by the vibration of external structures with large amplitude. Existing monitoring methods based on overall vibration signal analysis are difficult to effectively distinguish between bearing body vibration and non-bearing body vibration in the frequency domain. This leads to unstable bearing vibration feature extraction, insufficient accuracy in early abnormal state identification, and significant uncertainty in monitoring results. It is difficult to continuously meet the needs of bearing operation status monitoring and abnormal identification in the complex operating environment of offshore wind power. Summary of the Invention

[0008] To address the technical problem of accurately extracting bearing vibration characteristics during the monitoring of bearing operation status in offshore wind turbines using existing technologies, this invention provides a method and device for wind turbine bearing vibration control and monitoring based on piezoelectricity and metal-rubber. The technical solution is as follows:

[0009] On the one hand, a vibration control and monitoring method for wind turbine bearings based on piezoelectricity and metal rubber is provided. This method includes: acquiring the bearing vibration signal and its operating condition identifier of the wind turbine bearing, including normal operation and abnormal operation conditions. The abnormal operation condition characterizes the condition where the output power or main shaft speed of the wind turbine is not within the corresponding limiting range; extracting the bearing median frequency from the bearing vibration signal using frequency domain features; obtaining a characteristic frequency reference interval from a pre-stored mapping relationship based on the bearing median frequency; dividing the bearing vibration signal into bearing characteristic frequency bands and non-bearing characteristic frequency bands using the characteristic frequency reference interval; when the operating condition identifier of the bearing vibration signal corresponds to an abnormal operation condition, firstly expanding and correcting the characteristic frequency reference interval to obtain a characteristic frequency verification interval; then correcting the non-bearing characteristic frequency band according to the characteristic frequency verification interval to obtain a characteristic verification frequency band and updating the non-bearing characteristic frequency band; adjusting the energy gain of the characteristic verification frequency band and merging it with the bearing characteristic frequency band for updating; performing blind source separation on the updated non-bearing characteristic frequency band and determining whether to update the bearing characteristic frequency band; and outputting the bearing vibration monitoring results based on the bearing characteristic frequency band.

[0010] On the other hand, a wind turbine bearing vibration control and monitoring device based on piezoelectricity and metal rubber is provided, including: a bearing vibration monitoring module, a characteristic frequency band division module, a frequency band division correction module, and a monitoring result output module. The bearing vibration monitoring module collects the bearing vibration signal and its operating condition identifier from the wind turbine bearing. The operating condition identifier includes normal operating conditions and abnormal operating conditions. The abnormal operating conditions characterize the condition where the output power or main shaft speed of the wind turbine is not within the corresponding limit range. The characteristic frequency band division module extracts frequency domain features from the bearing vibration signal to obtain the bearing median frequency. Based on the bearing median frequency, a characteristic frequency reference range is obtained from a pre-stored mapping relationship to achieve the characteristic frequency control and monitoring. The bearing vibration signal is divided into bearing characteristic frequency bands and non-bearing characteristic frequency bands by a reference frequency interval. The frequency band division correction module is used when the bearing vibration signal's operating condition corresponds to an abnormal operating condition. It first expands and corrects the reference frequency interval to obtain a characteristic frequency verification interval, then corrects the non-bearing characteristic frequency band according to the characteristic frequency verification interval to obtain the characteristic verification frequency band, and then updates the non-bearing characteristic frequency band. After adjusting the energy gain of the characteristic verification frequency band, it is merged with the bearing characteristic frequency band and updated. The monitoring result output module is used to perform blind source separation on the updated non-bearing characteristic frequency band and then determine whether to update the bearing characteristic frequency band. Based on the bearing characteristic frequency band, it outputs the bearing vibration monitoring results.

[0011] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0012] The wind turbine bearing vibration control and monitoring method based on piezoelectricity and metal rubber proposed in this invention achieves synergistic technical effects across multiple levels, including signal acquisition, frequency domain division, and feature updating. This method no longer relies solely on a unified analysis of the overall vibration signal; instead, it incorporates operating condition identifiers to include changes in operating status caused by deviations in wind turbine output power and main shaft speed from their limit ranges into the monitoring logic. This allows the bearing vibration analysis to adaptively adjust to changes in operating conditions, thereby reducing the impact of complex operating conditions on the stability of vibration characteristics at the source.

[0013] Specifically, at the frequency domain processing level, this invention uses the bearing median frequency as the core parameter, constructs a characteristic frequency reference interval through pre-stored mapping relationships, and separates the bearing characteristic frequency band from the non-bearing characteristic frequency band based on this interval, thereby forming a clear vibration source differentiation mechanism in the frequency domain. Furthermore, under abnormal operating conditions, by expanding and correcting the characteristic frequency reference interval to form a characteristic frequency verification interval, frequency drift caused by factors such as speed fluctuations and sudden load changes can be effectively covered, avoiding the false exclusion of effective bearing vibration components and improving the adaptability of the monitoring method under unsteady operating conditions.

[0014] Furthermore, this invention enhances the potential bearing vibration information by adjusting the energy gain of the feature verification frequency band and fusing it with the bearing feature frequency band, significantly improving the identifiability of bearing vibration features in the frequency domain. Simultaneously, by combining blind source separation processing of non-bearing feature frequency bands, interference from non-bearing vibrations such as structural vibration and transmission system vibration is further reduced, improving the accuracy of bearing feature frequency band update judgment. Finally, the bearing vibration monitoring results are output based on the updated bearing feature frequency band, making the monitoring conclusions more comprehensively reflect the actual operating state of the bearing.

[0015] In summary, through the synergistic effect of the above-mentioned technical means, this invention can achieve stable extraction and continuous tracking of bearing vibration characteristics in the complex and variable operating environment of offshore wind power, significantly improve the accuracy of early abnormal bearing identification and the consistency of monitoring results, effectively reduce the uncertainty caused by changes in operating conditions and multi-source vibration coupling, and has high engineering application value and promotion significance. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1A flowchart of a wind turbine bearing vibration control and monitoring method based on piezoelectricity and metal rubber provided in an embodiment of this application;

[0018] Figure 2 A flowchart illustrating the process of obtaining the characteristic frequency reference range for a wind turbine bearing vibration control and monitoring method based on piezoelectricity and metal rubber, as provided in this application embodiment.

[0019] Figure 3 The interval expansion correction logic diagram of the wind turbine bearing vibration control and monitoring method based on piezoelectricity and metal rubber provided in the embodiments of this application is shown. Detailed Implementation

[0020] Embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of the present disclosure are shown in the drawings, it should be understood that embodiments of the present disclosure may be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure.

[0021] It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure. In the description of the embodiments of this disclosure, the term "comprising" and similar terms should be understood as open-ended inclusion, i.e., "including but not limited to". The term "based on" should be understood as "at least partially based on". The term "one embodiment" or "this embodiment" should be understood as "at least one embodiment". The terms "first", "second", etc., may refer to different or the same objects.

[0022] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0023] As a typical complex electromechanical coupling system, the long-term operational reliability of wind turbines largely depends on the health of key components in the drivetrain. Among these, the main shaft bearings and related support bearings bear the combined effects of wind loads, structural loads, and torsional loads, making them among the components most prone to fatigue damage and failure. Especially in offshore wind power applications, where the turbine operates under conditions of high humidity, high salt spray, strong wind loads, and random changes in operating conditions, any abnormality in the bearings often leads to a significant increase in vibration levels, affecting the safe operation of the entire turbine and even causing shutdowns or major equipment damage. Therefore, effective vibration monitoring and condition assessment of wind turbine bearings are of significant engineering importance.

[0024] Currently, the condition monitoring of wind turbine bearings mostly adopts methods based on vibration signal analysis. This involves extracting time-domain, frequency-domain, or time-frequency-domain features from the collected overall vibration signals to determine the bearing's operating status. However, in actual operation, the main shaft speed and output power of wind turbines fluctuate frequently with changes in wind conditions. The bearing vibration signal typically contains multiple non-bearing vibration components, such as gear meshing vibration, structural resonance vibration, and electromagnetic excitation vibration. This results in vibration signals exhibiting strong non-stationarity, low signal-to-noise ratio, and severe multi-source coupling.

[0025] Furthermore, with the continuous increase in the single-unit capacity of wind turbines, the structural size and stress complexity of bearings are constantly increasing. Early anomalies often manifest as vibrations with small amplitudes and dispersed distribution, easily masked by strong background vibrations. Existing monitoring methods struggle to accurately extract the vibration characteristics of the bearing itself while effectively suppressing non-bearing interference vibrations under complex operating environments, resulting in significant uncertainties in the monitoring results. This makes it difficult to meet the needs of bearing condition monitoring and anomaly identification under long-term, stable operation conditions of offshore wind turbines. Therefore, there is an urgent need for a wind turbine bearing vibration monitoring method that can effectively distinguish between bearing-based vibrations and non-bearing-based vibrations in the frequency domain, while incorporating changes in operating conditions, and possessing good adaptability and stability. This would improve the reliability of bearing operating condition assessment and the ability to identify early anomalies.

[0026] Firstly, such as Figure 1 The diagram shows a flowchart of a wind turbine bearing vibration control and monitoring method based on piezoelectricity and metal rubber provided in an embodiment of this application. The method includes the following steps: bearing vibration monitoring, characteristic frequency band division, frequency band division correction, and monitoring result output.

[0027] As the first step in the vibration control and monitoring method for wind turbine bearings based on piezoelectricity and metal rubber, bearing vibration monitoring specifically involves: acquiring bearing vibration signals and their operating condition identifiers through a Supervisory Control and Data Acquisition (SCADA) system. These operating condition identifiers include normal and abnormal operating conditions. Abnormal operating conditions characterize situations where the wind turbine's output power or main shaft speed is outside the corresponding limit range; conversely, normal operating conditions are defined as when both the wind turbine's output power and main shaft speed are within the corresponding limit range. By simultaneously acquiring the wind turbine bearing vibration signals and their operating condition identifiers, and clearly distinguishing between normal and abnormal operating conditions, vibration signal analysis is no longer isolated from the operating state. This method explicitly incorporates the unsteady-state effects introduced by the wind turbine's output power or main shaft speed deviating from the limit range into the monitoring logic, effectively avoiding misjudgments of vibration characteristics caused by changes in operating conditions, improving the correlation between vibration analysis and actual operating conditions, and providing a reliable basis for subsequent adaptive adjustment of frequency domain characteristics.

[0028] As a further solution, since offshore wind turbines operate in open, complex, and highly uncertain meteorological environments for extended periods, external environmental factors such as wind speed, ambient air pressure, and ambient noise intensity can directly or indirectly affect the vibration signal acquisition process. On the one hand, changes in wind speed and air pressure can cause fluctuations in the overall load and structural response of the turbine, resulting in significant amplitude and spectral fluctuations in the bearing vibration signal. On the other hand, the high intensity of ambient noise at sea can easily introduce background interference signals through structural transmission or sensor coupling, reducing the signal-to-noise ratio of the bearing vibration signal. In this situation, if a fixed sampling frequency and fixed filtering parameters are still used for vibration signal acquisition, problems such as insufficient sampling resolution, effective features being submerged by noise, or excessive filtering leading to the loss of effective bearing vibration information can easily occur, thus affecting the accuracy of subsequent frequency domain feature extraction and state judgment. Therefore, before acquiring the bearing vibration signal and its operating condition indicators for wind turbine bearings, the following steps are also included:

[0029] First, meteorological environmental parameters characterizing the operating environment of offshore wind turbines are acquired through a monitoring and data acquisition system. These parameters include wind speed, ambient air pressure, and ambient noise intensity. Next, each meteorological environmental parameter is weighted and coupled with its corresponding predefined meteorological environmental thresholds to obtain the monitoring environmental interference coefficient. Specifically, the constraint expression for the monitoring environmental interference coefficient is as follows: In the formula, Indicates the environmental interference coefficient during monitoring. Indicates the weight of ambient wind speed. Indicates the weight of ambient air pressure. Indicates environmental noise weight. This indicates that a wind speed reference threshold has been set. Indicates meteorological wind speed. This indicates the setting of a reference air pressure threshold. Indicates ambient air pressure. This indicates that a noise reference intensity threshold is set. Indicates the intensity of ambient noise.

[0030] Next, it is determined whether the acquired environmental interference coefficient is within the set interference control range. If so, the bearing vibration signal is acquired using the initial fixed parameters of the piezoelectric sensor, which include the initial sampling frequency and the initial filter intensity. Otherwise, the sampling adjustment frequency and filter adjustment intensity corresponding to the environmental interference coefficient are obtained from the pre-stored mapping relationship to update the initial fixed parameters.

[0031] It should be added that during the operation of offshore wind turbines, the external meteorological environment has a significant impact on the structural response and vibration signal acquisition quality of the turbine. To comprehensively and effectively characterize the degree of interference from the operating environment on bearing vibration monitoring, this invention selects meteorological wind speed, ambient air pressure, and ambient noise intensity as meteorological environmental parameters. Specifically, meteorological wind speed mainly reflects the influence of aerodynamic load changes on bearing vibration response; ambient air pressure characterizes macroscopic meteorological conditions and their combined effect on the structure and vibration baseline; and ambient noise intensity characterizes the impact of external random interference on the signal-to-noise ratio of the vibration signal. By comprehensively introducing these three types of meteorological environmental parameters, the operating environment of offshore wind turbines can be effectively characterized from multiple dimensions such as load changes, environmental conditions, and noise interference, providing a reliable environmental basis for the adaptive adjustment of bearing vibration signal acquisition parameters and subsequent condition monitoring and analysis.

[0032] In addition, the meteorological environment setting thresholds include setting wind speed reference thresholds, setting air pressure reference thresholds, and setting noise reference intensity thresholds. Both the meteorological environment setting thresholds and the interference control range are obtained from the bearing monitoring database. The bearing monitoring database is a database specifically created to store core configuration information when designing the vibration control and monitoring method for wind turbine bearings based on piezoelectricity and metal rubber. This database stores various setting values ​​and mapping relationships necessary for the operation of the method, such as the meteorological environment setting thresholds. The initial settings of these values ​​are not arbitrarily specified. Technicians can manually set, adjust, or finely tune them at any time according to the specific performance of the method in actual tests.

[0033] It should also be noted that the pre-stored mapping relationship takes the environmental interference coefficient as input and the sampling adjustment frequency and filter adjustment intensity as output. The setting principle is as follows: as the environmental interference coefficient increases, the sampling adjustment frequency increases accordingly to enhance the ability to capture detailed characteristics of bearing vibration; simultaneously, the filter adjustment intensity increases synchronously to suppress the influence of environmental noise and random interference on the vibration signal. When the environmental interference coefficient gradually decreases, the sampling adjustment frequency and filter adjustment intensity decrease accordingly, ensuring the integrity of effective information during signal acquisition while avoiding oversampling or over-filtering.

[0034] By quantifying the interference coefficient of the monitoring environment and mapping it to the sampling adjustment frequency and the filtering adjustment intensity, this mechanism can suppress the adverse effects of complex meteorological environment in advance during the vibration signal acquisition stage, so that the acquired bearing vibration signal maintains high consistency and stability under different environmental conditions, providing a reliable data basis for subsequent bearing vibration feature extraction and condition monitoring analysis.

[0035] The second step in the piezoelectric and metal-rubber-based wind turbine bearing vibration control and monitoring method, namely characteristic frequency band division, involves the following: First, frequency domain feature extraction is performed on the bearing vibration signal to obtain the bearing median frequency. This allows frequency band division to move beyond fixed empirical frequencies or single characteristic frequency points, instead basing it on the energy distribution of the vibration signal under current operating conditions. This approach dynamically reflects the changing trends of bearing vibration characteristics, improves the adaptability and robustness of characteristic frequency selection, and enhances the stability of frequency domain analysis under complex loads and variable speed conditions. Second, a characteristic frequency reference interval is obtained from a pre-stored mapping relationship based on the bearing median frequency. Then, the bearing vibration signal is divided into bearing characteristic frequency bands and non-bearing characteristic frequency bands using the characteristic frequency reference interval. By mapping the bearing median frequency to the pre-stored characteristic frequency reference interval, the initial division of the bearing characteristic frequency band and non-bearing characteristic frequency band is achieved. This process constructs a clear structural partition in the frequency domain, effectively reducing the interference of non-bearing vibrations such as gear meshing, structural resonance, and external excitation on bearing feature extraction, and providing clear frequency band boundaries for subsequent feature enhancement and verification.

[0036] As a further consideration, since wind turbines operate under unstable conditions of varying wind speeds and loads for extended periods, the radial and axial loads on the bearings continuously fluctuate with changes in wind conditions, power output, and structural state. Under different load conditions, the internal contact stiffness, damping characteristics, and energy transfer paths of the bearings all change, leading to a shift in the distribution of vibration energy in the frequency spectrum. Even if the bearing itself does not exhibit structural abnormalities, the main energy concentration range of its vibration spectrum may shift due to load changes, resulting in a deviation in the bearing median frequency calculated based on the vibration spectrum.

[0037] If the bearing median frequency, which does not consider the load effect, is directly used in the subsequent vibration characteristic frequency band division and state analysis, it is easy to misjudge the spectral drift caused by load changes as bearing abnormalities, or to mask the true abnormal frequency characteristics under high load conditions. This leads to inaccurate frequency band positioning and insufficient feature extraction stability, thus affecting the reliability of bearing operating condition assessment. Therefore, it is necessary to introduce a load-related correction mechanism after obtaining the bearing median frequency for further coupling correction. Specifically, after obtaining the bearing median frequency, the following should also be included:

[0038] An equivalent load correction factor is introduced to couple and correct the obtained bearing median frequency before updating. The equivalent load correction factor compensates for the influence of vibration spectrum distribution variations under different load conditions on the bearing median frequency. The equivalent load correction factor is the result of proportionalizing the bearing equivalent load and the reference equivalent load, followed by interactive calculation with the damping characteristic adjustment coefficient. The bearing equivalent load is the result of weighting the bearing radial load and the bearing axial load separately, followed by coupled square root calculation. Coupling correction involves multiplying the equivalent load correction factor and the bearing median frequency. The constraint expression for the equivalent load correction factor is as follows: ; In the formula, Indicates the equivalent load on the bearing. Indicates radial load weight. Indicates the radial load of the bearing. Indicates the axial load weight. Indicates the axial load of the bearing. This represents the equivalent load correction factor. Indicates the reference equivalent load. This represents the damping characteristic adjustment coefficient.

[0039] It should be added that the reference equivalent load, radial load weight, and axial load weight are all obtained from the bearing monitoring database. The damping characteristic adjustment coefficient is set by professionals according to industry standards. For example, in one exemplary embodiment, the damping characteristic adjustment coefficient can be set according to the bearing structure type, installation method, and vibration damping characteristics, and its value range can be a positive real number. For example, when the bearing structure damping is large and the vibration response to load changes is relatively smooth, the damping characteristic adjustment coefficient can be set to a small value, such as 0.3 to 0.6, so that the equivalent load correction factor shows a slow adjustment trend with load changes, thereby avoiding excessive correction to the bearing median frequency. When the bearing structure damping is small and the load change has a more sensitive effect on the vibration spectrum distribution, the damping characteristic adjustment coefficient can be set to a medium to large value, such as 0.8 to 1.2, so that the equivalent load correction factor has a more obvious response capability to load changes, thereby enhancing the compensation effect for the overall spectrum drift.

[0040] Therefore, the introduction of this mechanism enables the bearing median frequency to maintain good stability and physical consistency under different operating load conditions, providing a more accurate and reliable reference benchmark for subsequent dynamic division of characteristic frequency bands, energy assessment and anomaly identification based on the median frequency, thereby improving the adaptability of wind power bearing vibration monitoring methods to complex working conditions and the accuracy of anomaly identification.

[0041] Among them, such as Figure 2 The diagram shown is a flowchart for obtaining the characteristic frequency reference range of the wind turbine bearing vibration control and monitoring method based on piezoelectricity and metal rubber provided in this application embodiment. (Refer to...) Figure 2 It should also be noted that the specific method for obtaining the characteristic frequency reference interval is as follows:

[0042] First, the upper and lower boundary values ​​of the bearing median frequency are retrieved from the pre-stored mapping relationship. Next, it is determined whether the difference between the upper and lower boundary values ​​is within a set offset range. If so, a characteristic frequency reference interval is constructed based on the closed intervals corresponding to the lower and upper boundary values. Otherwise, the upper and lower boundary offsets of the upper and lower boundary values ​​are obtained from the bearing median frequency, respectively. The upper boundary offset is the absolute difference between the upper and lower boundary values, and the lower boundary offset is the absolute difference between the lower and upper boundary values. Then, according to the pre-stored mapping relationship, the upper and lower boundary buffer values ​​corresponding to the upper and lower boundary offsets are obtained. The characteristic frequency reference interval is constructed based on the closed intervals corresponding to the obtained lower and upper boundary buffer values.

[0043] In bearing vibration analysis, the setting of the characteristic frequency range directly affects the accuracy of subsequent frequency domain energy statistics, feature extraction, and anomaly identification results. Due to factors such as load fluctuations, speed changes, and environmental disturbances in actual wind turbine operation, even when the bearing median frequency is used as a frequency domain reference, the distribution of bearing vibration energy in the spectrum may still deviate within a certain range. If the characteristic frequency range is set too narrowly, it can easily lead to the truncation of effective characteristic energy; if the range is set too wide, it will introduce too many irrelevant frequency components, reducing feature distinguishability.

[0044] For the reasons mentioned above, this method introduces a dynamic determination and buffering mechanism based on pre-stored mapping relationships when constructing the characteristic frequency reference interval. By querying the upper and lower boundary values ​​of the bearing median frequency from the pre-stored mapping relationships and determining whether the difference between the two is within the set offset range, the corresponding closed interval can be directly used as the characteristic frequency reference interval when the spectrum distribution is relatively stable, thereby ensuring the accuracy of frequency band positioning.

[0045] When the difference between the upper and lower frequency boundary values ​​exceeds the set offset range, it indicates that the current bearing vibration spectrum distribution has a significant asymmetry or offset trend. In this case, by calculating the upper and lower boundary offsets of the upper and lower frequency boundary values ​​relative to the bearing median frequency, and further introducing corresponding upper and lower boundary buffer values ​​based on pre-stored mapping relationships, the original boundaries can be adaptively expanded or contracted, thereby constructing a characteristic frequency reference interval that better matches the current vibration characteristic distribution.

[0046] In this way, the method can improve the adaptability of the characteristic frequency range to spectral shift and asymmetric distribution while ensuring the accuracy of frequency band positioning. It avoids feature omission or noise introduction caused by fixed frequency band or single offset strategy, and provides a more stable and reliable frequency basis for subsequent frequency domain feature update and bearing status identification.

[0047] It should be understood that, in one exemplary embodiment, when the amplitude difference between the obtained upper and lower frequency boundary values ​​exceeds a set threshold, a boundary buffer adjustment strategy can be introduced based on the degree of offset. When the offset is small, the corresponding buffer adjustment amplitude is small to avoid excessive bandwidth expansion; when the offset is moderate, the upper or lower frequency boundary can be appropriately expanded to cover the spectral offset caused by operating condition disturbances; when the offset is large, a larger boundary buffer value can be set to ensure that the main spectrum and its side energy are included within the characteristic frequency reference range. In this way, a graded response to different degrees of offset can be achieved, making the bandwidth correction process coherent and controllable. It should also be understood that the construction method of the above mapping relationship is not limited to a fixed lookup table method; it can also be implemented using continuous function relationships, piecewise linear interpolation, or model generation based on statistical laws. This mapping relationship can be continuously updated or optimized according to the operating conditions of the wind turbine and the structural characteristics of the bearing. Its specific form does not constitute a limitation on the scope of protection of this invention.

[0048] As a further approach, in the early stages of bearing degradation, the energy changes corresponding to its characteristic frequencies are often relatively small and easily masked by high-energy interference in non-bearing frequency bands. If the bearing characteristic frequency band and non-bearing characteristic frequency bands are not distinguished and analyzed, judging solely based on absolute energy magnitude is insufficient to accurately reflect the changing trends of the bearing's structural state, thus affecting the sensitivity and reliability of health assessment. For these reasons, it is necessary to further divide the bearing vibration signal into bearing characteristic frequency bands and non-bearing characteristic frequency bands, building upon the initial characteristic frequency baseline division, and constructing an evaluation index reflecting the bearing's health status through the energy distribution relationship between the two. Specifically, the bearing vibration signal is divided into bearing characteristic frequency bands and non-bearing characteristic frequency bands based on the characteristic frequency baseline interval, and this process includes:

[0049] First, obtain the first frequency band energy value and the second frequency band energy value corresponding to the bearing characteristic frequency band and the non-bearing characteristic frequency band, respectively, and record the sum of the first frequency band energy value and the second frequency band energy value as the total frequency band energy; then, record the proportion of the first frequency band energy value to the total frequency band energy as the bearing health index, and determine the numerical relationship between the bearing health index and the set health warning threshold; when the bearing health index is greater than the health warning threshold, send a bearing health warning prompt.

[0050] When a bearing is in a healthy state, the vibration energy within its characteristic frequency band is typically low, and the bearing health index is within a small range. As bearing wear, fatigue, or localized defects gradually develop, the energy within the bearing's characteristic frequency band gradually accumulates, increasing its proportion in the total frequency band energy, thus causing the bearing health index to show a continuously rising trend. By comparing this health index with a pre-set health warning threshold, the process of the bearing's condition evolving from normal to abnormal can be effectively identified. Compared to judgment methods based on single frequency band energy or absolute vibration amplitude, this method normalizes the energy of different frequency bands through energy proportions, effectively reducing the impact of speed changes, load fluctuations, and overall vibration level changes on the assessment results, and improving the comparability and stability of bearing health assessment results under different operating conditions. Therefore, this analysis can achieve a relatively quantitative assessment of the bearing's health status in complex operating environments, enhance the ability to identify early bearing anomalies, and provide timely and reliable decision-making basis for bearing operation and maintenance by sending health warning prompts when the bearing health index exceeds the set health warning threshold.

[0051] The third step in the piezoelectric and metal-rubber-based wind turbine bearing vibration control and monitoring method, namely frequency band division correction, involves the following: First, when the bearing vibration signal's operating condition corresponds to an abnormal operating condition, the characteristic frequency reference interval is first expanded and corrected to obtain a characteristic frequency verification interval. Then, the non-bearing characteristic frequency band is corrected according to the characteristic frequency verification interval to obtain the characteristic verification frequency band, which is then updated. Specifically, the frequency band outside the characteristic frequency verification interval is the non-bearing characteristic frequency band, and the frequency band within the characteristic frequency verification interval but not containing the bearing characteristic frequency band is the characteristic verification frequency band. Next, the characteristic verification frequency band undergoes energy gain adjustment and is then merged with the bearing characteristic frequency band and updated. For example, Figure 3 The diagram shown is a range expansion correction logic diagram for the wind turbine bearing vibration control and monitoring method based on piezoelectricity and metal rubber provided in this application embodiment. (Refer to...) Figure 3 The characteristic frequency reference interval is expanded and corrected to obtain the characteristic frequency verification interval. The specific steps are as follows:

[0052] First, the absolute deviations of the wind turbine's output power and main shaft speed from the median values ​​within the corresponding limit ranges are obtained. The absolute deviations include the absolute power deviation and the absolute speed deviation. Second, the result of the normalization and coupling processing of the absolute deviations is input into a pre-stored mapping table to obtain the range expansion ratio. The normalization and coupling processing involves normalizing the absolute deviations separately and then adding them together. Next, the characteristic frequency reference range is amplified based on the obtained range expansion ratio to obtain the characteristic frequency verification range.

[0053] It is important to understand that by expanding and correcting the characteristic frequency reference range to obtain the characteristic frequency verification range, the spectral offset caused by speed fluctuations and power changes can be appropriately covered while ensuring the accuracy of the reference frequency band positioning. On the one hand, this avoids the problem of characteristic frequency truncation or missed detection when operating conditions fluctuate significantly; on the other hand, it prevents excessive expansion of the frequency band and the introduction of too many irrelevant frequency components when operating conditions are stable. Therefore, the adaptability and verification reliability of the characteristic frequency range under different operating conditions are improved, providing a more robust frequency domain foundation for subsequent bearing health status confirmation and early warning judgment.

[0054] It should also be understood that the mapping table here is used to describe the mapping relationship between the result of normalized coupling processing of the absolute deviation and the interval expansion ratio. In a specific embodiment, the mapping table can be divided into multiple level intervals according to the magnitude of the comprehensive deviation value, such as intervals with small deviation, medium deviation, and large deviation, and different interval expansion ratios are set for different levels. Specifically, when the comprehensive deviation value is small, the corresponding interval expansion ratio is small, and only the characteristic frequency reference interval is slightly enlarged; when the comprehensive deviation value is at a medium level, the corresponding interval expansion ratio is moderately increased to compensate for a certain degree of spectral shift; when the comprehensive deviation value is large, the corresponding interval expansion ratio is further increased to cover the frequency drift range that may occur under significant operating condition fluctuations.

[0055] It should also be noted that the feature verification frequency band, after energy gain adjustment, is fused with and updated with the bearing feature frequency band. The specific steps are as follows:

[0056] First, the energy values ​​of the first frequency band and the energy values ​​of the third frequency band corresponding to the feature verification frequency band are obtained. The ratio of the energy values ​​of the first frequency band and the energy values ​​of the third frequency band is normalized and recorded as the gain factor. Then, the amplitude of the feature verification frequency band is coupled and adjusted based on the gain factor. The coupling adjustment here is a product operation. The feature verification frequency band after coupling adjustment is superimposed and fused with the bearing feature frequency band in the frequency domain to update the bearing feature frequency band.

[0057] This mechanism enables the smooth evolution and updating of the bearing characteristic frequency band under dynamic operating conditions. It not only improves the bearing characteristic frequency band's adaptability to spectral shifts and energy diffusion, but also avoids the problem of introducing too many irrelevant frequency components due to direct frequency band expansion. This enhances the continuity, stability, and reliability of bearing vibration characteristic characterization, providing a more accurate frequency domain basis for subsequent bearing health status assessment and early warning judgment.

[0058] Overall, this step, when detecting abnormal operating conditions, expands and corrects the characteristic frequency reference range to obtain the characteristic frequency verification range. This allows the frequency band division strategy to adapt to frequency drift caused by abnormal operating conditions such as power fluctuations and speed deviations. This method avoids the problem of the true bearing characteristics being mistakenly assigned to non-bearing frequency bands due to frequency deviations, enhancing the applicability and reliability of the monitoring method under extreme or unsteady operating conditions. Based on this, by adjusting the energy gain of the characteristic verification frequency band and merging it with the original bearing characteristic frequency band, potential effective bearing vibration components can be enhanced and compensated. This method improves the energy proportion and identifiability of bearing vibration characteristics without introducing additional noise, making early weak abnormal signals easier to detect and facilitating early warning of bearing conditions.

[0059] As the fourth step in the wind turbine bearing vibration control and monitoring method based on piezoelectricity and metal rubber, namely the output of monitoring results, specifically: after blind source separation of the updated non-bearing characteristic frequency band, it is determined whether to update the bearing characteristic frequency band, and the bearing vibration monitoring results are output according to the bearing characteristic frequency band.

[0060] As a further approach, after dividing the bearing vibration signal into frequency bands, the non-bearing characteristic frequency band is, in principle, used to characterize vibration components that do not belong to the bearing's characteristic frequency range. However, in the complex operating environment of wind turbines, due to speed fluctuations, load changes, and spectral spread effects, some bearing characteristic components may experience frequency drift or energy diffusion, thus falling into the non-bearing characteristic frequency band. If this part of the frequency band is completely regarded as a non-bearing component based solely on the frequency interval division result, bearing-related vibration information may be missed. Therefore, after performing blind source separation on the non-bearing characteristic frequency band, it is determined whether to update the bearing characteristic frequency band, specifically as follows:

[0061] The reconstruction error after blind source separation of the non-bearing characteristic frequency band is obtained. When the reconstruction error is less than the error threshold, the bearing characteristic frequency band is updated; otherwise, it is not updated. By performing blind source separation on the non-bearing characteristic frequency band and determining whether to update the bearing characteristic frequency band based on the reconstruction error, when the reconstruction error is less than the preset error threshold, it indicates that the signal components in the non-bearing characteristic frequency band have good separability and structural stability, and may still contain effective feature information related to bearing vibration. In this case, updating the bearing characteristic frequency band helps to reintegrate bearing features that have temporarily fallen into the non-bearing characteristic frequency band due to spectral drift or energy diffusion into the bearing characteristic frequency band, thereby improving the completeness of bearing feature extraction.

[0062] Conversely, when the reconstruction error after blind source separation is greater than or equal to the error threshold, it indicates that the signal source aliasing in the non-bearing characteristic frequency band is severe or that random disturbance components account for a high proportion, and its internal structure is unstable, making it difficult to reliably distinguish potential bearing characteristic components. In this case, not updating the bearing characteristic frequency band can effectively avoid introducing irrelevant vibration or noise components into the bearing characteristic frequency band, thereby maintaining the purity and stability of the bearing characteristic frequency band.

[0063] Therefore, this method further verifies the non-bearing characteristic frequency bands and controls the updating behavior of the bearing characteristic frequency bands based on the blind source separation and reconstruction error. While ensuring the integrity of the bearing characteristics, it suppresses the risk of false updates, making the evolution process of the bearing characteristic frequency bands more reliable and controllable. Overall, it improves the accuracy and engineering applicability of the bearing health monitoring method under complex working conditions.

[0064] As a further solution, the bearing vibration monitoring results are output based on the bearing characteristic frequency band. This further includes: sensing the bearing vibration based on the positive piezoelectric effect of the piezoelectric sensor to obtain a vibration feedback signal; and further, using a PID algorithm to generate a control signal based on the vibration feedback signal, and driving the piezoelectric actuator to output a damping force using the inverse piezoelectric effect to actively control the bearing vibration.

[0065] In the vibration control of wind turbine bearings, active control based on piezoelectric sensors and piezoelectric actuators can achieve rapid and precise vibration suppression for specific frequency components. However, bearing vibration signals are typically characterized by wide bandwidth, complex composition, and significant transient impacts. Relying solely on active control methods may face problems such as limited control bandwidth, high energy consumption, or decreased control stability under high-frequency random vibration, broadband noise, or sudden impact loads.

[0066] For the reasons mentioned above, it is necessary to introduce a passive vibration reduction structure with inherent damping and broadband energy absorption characteristics outside the active control system to mitigate and dissipate bearing vibration at the front end. Metal rubber, as a porous metallic elastic damping material, possesses excellent nonlinear damping characteristics and frequency selectivity. Its internal metal mesh structure can physically filter and passively attenuate high-frequency components and random disturbances during vibration transmission, thereby improving the overall characteristics of the bearing vibration signal.

[0067] In this embodiment, bearing vibration is first passively modulated via a metal-rubber structure. The metal-rubber acts as elastic support and damping energy dissipation in the bearing vibration transmission path. The friction and micro-deformation between its multiple layers of metal wires effectively attenuate high-frequency noise, impact components, and non-stationary disturbances in the bearing vibration, thereby achieving physical filtering and passive vibration reduction of the vibration signal. In this way, the vibration amplitude and complexity entering the active control system can be reduced without relying on external energy input.

[0068] Based on this, the vibration of the bearing modulated by the metal rubber is sensed using the direct piezoelectric effect of the piezoelectric sensor to obtain a vibration feedback signal. Since the high-frequency interference and random components in the vibration signal have been effectively weakened by the metal rubber, the obtained vibration feedback signal has a higher signal-to-noise ratio and stability, which is beneficial for the accurate calculation of subsequent control algorithms. Furthermore, a PID algorithm is used to generate a control signal based on the vibration feedback signal, and drives the piezoelectric actuator to output a damping force using the inverse piezoelectric effect, thereby implementing active compensation control for the main characteristic frequency components of the bearing vibration.

[0069] Based on the above implementation methods, considering the significant hysteresis nonlinearity of the metal rubber material itself and the complex and variable operating conditions of wind turbines, a single PID control algorithm is unlikely to maintain optimal control performance across the entire frequency band. Therefore, the control algorithm is further expanded by adopting a composite control strategy of "feedforward compensation combined with fuzzy adaptive PID".

[0070] Specifically, the control unit first acquires the real-time rotational speed signal of the wind turbine bearing, pre-calculates its fault characteristic frequencies (such as the inner ring passing frequency, outer ring passing frequency, etc.) based on the bearing's geometric parameters, and generates a reference signal that is in the same frequency but out of phase with the dominant vibration frequency component, which serves as the input to the feedforward channel. This feedforward channel can directly drive the piezoelectric actuator to output targeted vibration damping force before vibration errors occur, thus canceling out periodic vibration components that are strongly correlated with rotational speed in advance, thereby significantly reducing the adjustment pressure on the feedback control loop.

[0071] Building upon this foundation, a fuzzy adaptive PID algorithm is employed to replace the traditional PID controller for the feedback loop. This algorithm monitors the vibration feedback signal modulated by the metal rubber in real time, calculates the current vibration error and its rate of change, and inputs this information into a fuzzy inference engine. Based on a pre-defined fuzzy rule base, the fuzzy inference engine dynamically adjusts the proportional, integral, and derivative coefficients of the PID controller online according to the current nonlinear stiffness and damping state of the metal rubber. For example, when a large error is detected due to impact vibration, the algorithm automatically increases the proportional coefficient to improve the system's rapid response capability; when the vibration tends to stabilize but small fluctuations exist, the algorithm appropriately decreases the proportional coefficient and increases the integral and derivative coefficients to eliminate steady-state errors and suppress system oscillations. This parameter self-adjustment mechanism effectively overcomes the model uncertainty caused by the amplitude-frequency dependence of the metal rubber, ensuring that the active and passive control systems maintain optimal robustness and control accuracy under various operating conditions.

[0072] By combining the passive vibration damping of metal rubber with the active vibration suppression of piezoelectric actuators, an integrated active-passive vibration control mechanism can be formed. The metal rubber is used for front-end suppression and energy dissipation of broadband vibrations and random disturbances, while the piezoelectric actuator is used for fine-tuning of key vibration components within the bearing's characteristic frequency band. The two complement each other in the frequency domain and control mechanism, effectively reducing the load and energy consumption of the active control system and improving the overall stability and reliability of vibration control.

[0073] Therefore, this solution can not only output bearing vibration monitoring results based on the vibration characteristic information in the bearing characteristic frequency band, but also realize real-time control of bearing vibration based on monitoring. Through the synergistic effect of metal rubber and piezoelectric actuator, it can achieve a combination of passive attenuation and active suppression of bearing vibration, which significantly improves the adaptability and engineering application value of bearing vibration monitoring and control system under complex operating conditions.

[0074] Secondly, embodiments of this application also provide a wind turbine bearing vibration control and monitoring device based on piezoelectricity and metal rubber. The device includes: a bearing vibration monitoring module, a characteristic frequency band division module, a frequency band division correction module, and a monitoring result output module.

[0075] The bearing vibration monitoring module is used to collect the bearing vibration signal and its operating condition identifier of the wind turbine bearing. The operating condition identifier includes normal operating condition and abnormal operating condition. The abnormal operating condition is used to characterize the condition where the output power or main shaft speed of the wind turbine is not within the corresponding limit range.

[0076] The characteristic frequency band division module is used to extract the frequency domain features of the bearing vibration signal to obtain the bearing median frequency. Based on the bearing median frequency, the characteristic frequency reference interval is obtained from the pre-stored mapping relationship. The bearing vibration signal is divided into bearing characteristic frequency bands and non-bearing characteristic frequency bands using the characteristic frequency reference interval.

[0077] The frequency band division correction module is used to first extend and correct the characteristic frequency reference interval to obtain the characteristic frequency verification interval when the bearing vibration signal corresponds to an abnormal operating condition. Then, it corrects the non-bearing characteristic frequency band according to the characteristic frequency verification interval to obtain the characteristic verification frequency band and updates the non-bearing characteristic frequency band. After adjusting the energy gain of the characteristic verification frequency band, it merges it with the bearing characteristic frequency band and updates it.

[0078] The monitoring result output module is used to determine whether to update the bearing characteristic frequency band after blind source separation of the updated non-bearing characteristic frequency band, and output the bearing vibration monitoring results based on the bearing characteristic frequency band.

[0079] In summary, this invention introduces a frequency band adaptive adjustment mechanism driven by operating condition identification, and combines it with various signal processing techniques such as bearing median frequency mapping, characteristic frequency range expansion correction, energy gain fusion, and blind source separation to effectively distinguish between bearing body vibration and non-bearing body vibration in the frequency domain. Compared with existing monitoring methods based on overall vibration signal analysis, this invention significantly improves the stability and reliability of bearing vibration feature extraction, enhances the ability to identify early abnormal bearing conditions, reduces the uncertainty of monitoring results, and can better adapt to the complex and variable operating environment of offshore wind turbines, continuously meeting the engineering application needs of wind turbine bearing operating status monitoring and anomaly identification.

[0080] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the above functions can be divided into different functional modules to complete all or part of the functions described above.

[0081] In the embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0082] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units, located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0083] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0084] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the solution, or all or part of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0085] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber, characterized in that, Includes the following steps: Collect the bearing vibration signal and its operating condition identifier of the wind turbine bearing. The operating condition identifier includes normal operating condition and abnormal operating condition. The abnormal operating condition is used to characterize the condition that the output power or main shaft speed of the wind turbine is not within the corresponding limit range. Frequency domain feature extraction is performed on the bearing vibration signal to obtain the bearing median frequency. Based on the bearing median frequency, the characteristic frequency reference interval is obtained from the pre-stored mapping relationship. The bearing vibration signal is divided into bearing characteristic frequency band and non-bearing characteristic frequency band using the characteristic frequency reference interval. When the bearing vibration signal corresponds to an abnormal operating condition, the characteristic frequency reference range is first extended and corrected to obtain the characteristic frequency verification range. Then, the non-bearing characteristic frequency band is corrected according to the characteristic frequency verification range to obtain the characteristic verification frequency band and then the non-bearing characteristic frequency band is updated. After the energy gain of the characteristic verification frequency band is adjusted, it is merged with the bearing characteristic frequency band and updated. After performing blind source separation on the updated non-bearing characteristic frequency band, it is determined whether to update the bearing characteristic frequency band, and the bearing vibration monitoring results are output based on the bearing characteristic frequency band.

2. The method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber as described in claim 1, characterized in that, The acquisition of bearing vibration signals and operating condition indicators for wind turbine bearings also includes, prior to: Meteorological environmental parameters are acquired to characterize the operating environment of offshore wind turbines, including meteorological wind speed, ambient air pressure, and ambient noise intensity. After deviating from the meteorological environmental parameters and their corresponding meteorological environmental thresholds, weighted coupling is performed to obtain the monitoring environmental interference coefficient. Determine whether the acquired monitoring environment interference coefficient is within the set interference control range. If so, collect the bearing vibration signal using the initial fixed parameters of the piezoelectric sensor. The initial fixed parameters include the initial sampling frequency and the initial filter intensity. Otherwise, the sampling adjustment frequency and filter adjustment intensity corresponding to the monitoring environment interference coefficient are obtained from the pre-stored mapping relationship to update the initial fixed parameters.

3. The method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber as described in claim 1, characterized in that, After obtaining the bearing median frequency, the process further includes: An equivalent load correction factor is introduced to couple and correct the obtained bearing median frequency before updating. The equivalent load correction factor is used to compensate for the influence of vibration spectrum distribution changes under different load conditions on the bearing median frequency. The equivalent load correction factor is the result of proportional processing of the bearing equivalent load and the reference equivalent load, followed by interactive calculation with the damping characteristic adjustment coefficient. The equivalent load of the bearing is the result of weighting the radial load and axial load of the bearing respectively and then performing a coupled square root operation. The limiting expression for the equivalent load correction factor is: ； ； In the formula, Indicates the equivalent load on the bearing. Indicates radial load weight. Indicates the radial load of the bearing. Indicates the axial load weight. Indicates the axial load of the bearing. This represents the equivalent load correction factor. Indicates the reference equivalent load. This represents the damping characteristic adjustment coefficient.

4. The method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber as described in claim 1, characterized in that, The specific method for obtaining the characteristic frequency reference interval is as follows: Query the upper and lower boundary values ​​of the bearing median frequency from the pre-stored mapping relationship; Determine whether the difference between the upper and lower frequency boundary values ​​is within the set offset range. If so, construct the characteristic frequency reference interval based on the closed intervals corresponding to the lower and upper frequency boundary values. Otherwise, obtain the upper and lower boundary values ​​of the frequency and their corresponding upper and lower boundary offsets to the bearing median frequency, respectively. Based on the pre-stored mapping relationship, the upper boundary offset and lower boundary offset corresponding to the upper boundary buffer value and lower boundary buffer value are obtained respectively. Based on the closed intervals corresponding to the obtained lower boundary buffer value and upper boundary buffer value, the characteristic frequency reference interval is constructed.

5. The method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber as described in claim 1, characterized in that, The process of dividing the bearing vibration signal into bearing characteristic frequency bands and non-bearing characteristic frequency bands based on a characteristic frequency reference interval further includes: Obtain the first frequency band energy value and the second frequency band energy value corresponding to the bearing characteristic frequency band and the non-bearing characteristic frequency band respectively, and record the sum of the first frequency band energy value and the second frequency band energy value as the total frequency band energy; The proportion of the energy value of the first frequency band to the total frequency band energy is recorded as the bearing health index, and the numerical relationship between the bearing health index and the set health warning threshold is determined. When the bearing health index exceeds the health warning threshold, a bearing health warning message will be sent.

6. The method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber as described in claim 1, characterized in that, The specific steps for expanding and correcting the characteristic frequency reference interval to obtain the characteristic frequency verification interval are as follows: The absolute deviations of the wind turbine's output power and main shaft speed from the median values ​​within the corresponding limit ranges are obtained respectively, and the absolute deviations include the absolute power deviation and the absolute speed deviation. The result of normalizing and coupling the absolute deviation is input into a pre-stored mapping table to obtain the interval expansion ratio; The characteristic frequency benchmark interval is enlarged based on the obtained interval expansion ratio to obtain the characteristic frequency verification interval.

7. The method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber as described in claim 5, characterized in that, The specific steps for adjusting the energy gain of the feature verification frequency band and fusing it with the bearing feature frequency band, and then updating it, are as follows: Obtain the energy value of the first frequency band and the energy value of the third frequency band corresponding to the feature verification frequency band. The result of normalizing the ratio of the energy value of the first frequency band to the energy value of the third frequency band is recorded as the gain factor. The amplitude of the feature verification frequency band is coupled and adjusted based on the gain factor. The frequency band of the feature verification frequency band after coupling adjustment is superimposed and fused with the bearing feature frequency band in the frequency domain to update the bearing feature frequency band.

8. The method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber as described in claim 1, characterized in that, The step of determining whether to update the bearing characteristic frequency band after performing blind source separation on the updated non-bearing characteristic frequency band is as follows: The reconstruction error after blind source separation of the non-bearing characteristic frequency band is obtained. When the reconstruction error is less than the error threshold, the bearing characteristic frequency band is updated; otherwise, it is not updated.

9. The method for vibration control and monitoring of wind turbine bearings based on piezoelectricity and metal rubber as described in claim 1, characterized in that, The process of outputting bearing vibration monitoring results based on the bearing's characteristic frequency band further includes: Vibration feedback signals are obtained by sensing bearing vibration based on the positive piezoelectric effect of piezoelectric sensors; A PID algorithm is used to generate a control signal based on the vibration feedback signal, and drive the piezoelectric actuator to output a damping force using the inverse piezoelectric effect to actively control the bearing vibration.

10. A wind turbine bearing vibration control and monitoring device based on piezoelectricity and metal rubber, characterized in that, include: Bearing vibration monitoring module, characteristic frequency band division module, frequency band division correction module, and monitoring result output module: The bearing vibration monitoring module is used to collect the bearing vibration signal and its operating condition identifier of the wind turbine bearing. The operating condition identifier includes normal operating condition and abnormal operating condition. The abnormal operating condition is used to characterize the condition that the output power or main shaft speed of the wind turbine is not within the corresponding limit range. The characteristic frequency band division module is used to extract the frequency domain features of the bearing vibration signal to obtain the bearing median frequency, and to obtain the characteristic frequency reference interval from the pre-stored mapping relationship based on the bearing median frequency. The bearing vibration signal is then divided into bearing characteristic frequency bands and non-bearing characteristic frequency bands using the characteristic frequency reference interval. The frequency band division correction module is used to first extend and correct the characteristic frequency reference interval to obtain the characteristic frequency verification interval when the bearing vibration signal's operating condition identifier corresponds to an abnormal operating condition. Then, it corrects the non-bearing characteristic frequency band according to the characteristic frequency verification interval to obtain the characteristic verification frequency band and updates the non-bearing characteristic frequency band. Finally, it adjusts the energy gain of the characteristic verification frequency band and merges it with the bearing characteristic frequency band and updates it. The monitoring result output module is used to determine whether to update the bearing characteristic frequency band after blind source separation of the updated non-bearing characteristic frequency band, and output the bearing vibration monitoring result based on the bearing characteristic frequency band.

Images