Generator bearing degradation evaluation method and system based on metal self-repairing technology

Abstract

The invention relates to the technical field of bearing degradation analysis, and discloses a generator bearing degradation evaluation method and system based on a metal self-repairing technology. The method comprises the following steps: collecting vibration, balance and temperature data of a bearing; extracting balance and wear characteristics, and establishing a degradation index system; injecting a self-repairing material; monitoring a repairing effect, and obtaining improvement data; predicting the health state of the bearing through a memory network; and generating an evaluation report, optimizing a repair strategy, and predicting the residual life. According to the method, the health state of the bearing can be accurately evaluated, personalized optimization of a metal self-repairing material formula is achieved, whole-process monitoring and effect evaluation are conducted on the repairing process, finally, accurate prediction of the remaining service life of the bearing is achieved, and scientificity and efficiency of wind turbine generator bearing maintenance are improved.

Description

Technical Field

[0001] The present application relates to the technical field of bearing degradation analysis, and in particular to a generator bearing degradation assessment method and system based on metal self-repair technology. Background Art

[0002] With the rapid development of the wind power industry, the reliability of wind turbine main shaft bearings and generator bearings has become increasingly prominent. Wind turbine bearings are prone to wear, fatigue and dynamic balance damage when they operate for a long time under high load and variable speed conditions. According to research from the World Tribology Conference, the energy loss caused by friction during equipment operation exceeds 30%, and the damage to wind turbine main shaft bearings and generator bearings, as key components, directly affects the power generation efficiency and equipment reliability. Traditional bearing maintenance methods mainly rely on regular replacement of grease or disassembly of bearings for repair, which is not only costly, but also requires downtime for maintenance, which seriously affects the economic benefits of wind farms. In recent years, metal self-repair technology has received widespread attention as a new type of in-situ repair technology for friction surfaces. By adding nanomaterials with special properties to grease, a metal ceramic protective layer is formed under the action of friction heat and pressure to achieve self-repair of the bearing surface.

[0003] However, the existing metal self-repair technology for wind turbine bearings has obvious defects and limitations. First, there is a lack of scientific and systematic bearing health status assessment methods, which makes it impossible to accurately judge the degree of damage and repair needs of bearings; second, the formula ratio of metal self-repair materials often adopts a unified standard, and fails to make personalized adjustments according to the damage characteristics of different bearings; third, the repair process lacks effective monitoring, making it difficult to judge the repair effect and progress; finally, the remaining service life prediction method for repaired bearings is imperfect, which makes the maintenance plan lack scientific basis. These problems have led to uneven application effects of metal self-repair technology in wind turbine bearing maintenance, and its technical potential cannot be fully utilized, which has restricted the further reduction of wind farm operation and maintenance costs. Summary of the invention

[0004] The present application provides a generator bearing degradation assessment method and system based on metal self-repair technology, which is used to accurately assess the bearing health status, achieve personalized optimization of the metal self-repair material formula, and monitor and evaluate the repair process throughout the process, ultimately achieving accurate prediction of the remaining service life of the bearing, and improving the scientificity and efficiency of wind turbine bearing maintenance.

[0005] In a first aspect, the present application provides a generator bearing degradation assessment method based on metal self-repair technology, and the generator bearing degradation assessment method based on metal self-repair technology includes: collecting vibration signals, dynamic balancing parameters and temperature data of the generator bearing to obtain a time series of comprehensive parameters of the bearing operating status; according to the time series of comprehensive parameters of the bearing operating status, using a multidimensional feature extraction algorithm to extract bearing balance characteristics and wear characteristics to obtain a bearing degradation index system; according to the bearing degradation index system, injecting metal self-repairing materials into the generator bearing to obtain a repaired bearing; performing dynamic balancing tests and working status monitoring on the repaired bearing to obtain a repair stage judgment result and balance improvement data; according to the repair stage judgment result and balance improvement data, predicting and analyzing the bearing balance stability and degradation trend through a long short-term memory network to obtain a bearing health status assessment index; according to the bearing health status assessment index, generating a bearing balance and degradation assessment report and optimizing the metal self-repair strategy to obtain a bearing remaining service life prediction result.

[0006] In a second aspect, the present application provides a generator bearing degradation assessment system based on metal self-repairing technology, and the generator bearing degradation assessment system based on metal self-repairing technology includes:

[0007] The acquisition module is used to collect the vibration signal, dynamic balance parameters and temperature data of the generator bearing to obtain the time series of comprehensive parameters of the bearing operation status;

[0008] An extraction module, used to extract the balance characteristics and wear characteristics of the bearing according to the time series of the comprehensive parameters of the bearing operation state by using a multi-dimensional feature extraction algorithm to obtain a bearing degradation index system;

[0009] A repair module, used for injecting metal self-repairing material into the generator bearing according to the bearing degradation index system to obtain a repaired bearing;

[0010] A monitoring module, used to perform a dynamic balance test and work status monitoring on the repaired bearing to obtain a repair stage judgment result and balance improvement data;

[0011] A prediction module is used to predict and analyze the balance stability and degradation trend of the bearing through a long short-term memory network according to the judgment results of the repair stage and the balance improvement data, so as to obtain a bearing health status evaluation index;

[0012] A generation module is used to generate a bearing balance and degradation assessment report and optimize the metal self-repair strategy according to the bearing health status assessment index to obtain a prediction result of the remaining service life of the bearing.

[0013] In a third aspect, a computer device is provided, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the computer device executes the above-mentioned generator bearing degradation assessment method based on metal self-repair technology.

[0014] In a fourth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium, and when the computer-readable storage medium is run on a computer, the computer executes the above-mentioned generator bearing degradation assessment method based on metal self-repair technology.

[0015] In the technical solution provided by the present application, multi-dimensional parameter data of the bearing is collected through vibration sensors, balance testers and temperature sensors to form a time series of comprehensive parameters of the bearing operating status, providing a comprehensive data basis for subsequent analysis and avoiding the problem of incomplete diagnosis caused by a single parameter of the traditional monitoring method; a multi-dimensional feature extraction algorithm is used to deeply mine the balance characteristics and wear characteristics of the bearing, and a bearing degradation index system is constructed to achieve accurate quantification of the health status of the bearing, providing a scientific basis for the optimization of the metal self-repair material formula; the metal self-repair material formula is customized according to the bearing degradation index system and injected into the bearing, which solves the limitation that the traditional unified formula is not suitable for different damage characteristics and improves the repair efficiency; dynamic balancing tests and working status monitoring are performed on the repaired bearings to obtain the judgment results of the repair stage and the balance improvement data, realizing the whole-process monitoring of the repair process, and timely discovering and adjusting abnormal situations; the long short-term memory network is used to predict the balance stability and degradation trend of the bearing The analysis fully utilizes the advantages of the deep learning algorithm in time series data processing and significantly improves the prediction accuracy. The algorithm can capture the complex nonlinear relationship in the long time series of bearing state parameters, and is particularly suitable for application scenarios in wind farm environments where bearing conditions are changeable and data noise is large. The contribution of the algorithm is reflected in the upgrade of traditional prediction methods based on experience or simple statistics to the level of intelligent prediction based on deep learning, making the prediction results closer to the actual degradation law; finally, the bearing balance and degradation assessment report is generated according to the bearing health status assessment index, and the metal self-repair strategy is optimized. At the same time, the remaining service life of the bearing is predicted, which provides a scientific basis for wind farm operation and maintenance decisions, realizes the transformation from post-maintenance to predictive maintenance, and greatly reduces the unplanned downtime and maintenance costs of wind turbines caused by bearing failures, thereby improving the operational efficiency and economic benefits of wind farms; in addition, this method can achieve repair and evaluation without disassembling the bearing, which reduces the downtime and maintenance time and ensures the continuous and stable power generation of the wind farm. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0017] Figure 1 This is a schematic diagram of an embodiment of a generator bearing degradation assessment method based on metal self-repairing technology in an embodiment of the present application;

[0018] Figure 2 This is a schematic diagram of an embodiment of a generator bearing degradation assessment system based on metal self-repairing technology in an embodiment of the present application;

[0019] Figure 3 It is a schematic block diagram of the structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The embodiments of the present application provide a method and system for assessing the deterioration of a generator bearing based on metal self-repair technology. The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the specific process of the embodiment of the present application is described below. Figure 1 In the embodiment of the present application, an embodiment of a generator bearing degradation assessment method based on metal self-repairing technology includes:

[0022] Step S101, collecting vibration signals, dynamic balance parameters and temperature data of the generator bearing to obtain a time series of comprehensive parameters of the bearing operation status;

[0023] Step S102, extracting the balance characteristics and wear characteristics of the bearing according to the time series of the comprehensive parameters of the bearing operation state by using a multi-dimensional feature extraction algorithm to obtain a bearing degradation index system;

[0024] Step S103: injecting metal self-repairing material into the generator bearing according to the bearing degradation index system to obtain a repaired bearing;

[0025] Step S104, performing a dynamic balance test and working status monitoring on the repaired bearing to obtain a repair stage judgment result and balance improvement data;

[0026] Step S105: According to the judgment result of the repair stage and the balance improvement data, the balance stability and degradation trend of the bearing are predicted and analyzed through the long short-term memory network to obtain the bearing health status assessment index;

[0027] Step S106: Generate a bearing balance and degradation assessment report based on the bearing health status assessment index and optimize the metal self-repair strategy to obtain a prediction result of the remaining service life of the bearing.

[0028] It is understandable that the execution subject of the present application may be a generator bearing degradation assessment system based on metal self-repair technology, or a terminal or a server, which is not limited here. The present application embodiment is described by taking a server as the execution subject as an example.

[0029] Specifically, a triaxial acceleration vibration sensor is installed on the bearing seat of the wind turbine generator, and the sampling frequency is set to 10kHz to capture the early fault characteristics of the bearing; at the same time, the imbalance and phase angle of the bearing during rotation are measured by a balance tester; in addition, an embedded temperature sensor is arranged on the surface of the bearing seat for continuous temperature monitoring. During the acquisition process, the vibration sensor records the bearing running at a rated speed of 1500r / min for one week, the balance tester measures the dynamic balance data three times a day, and the temperature sensor collects the temperature value once an hour. After the data acquisition is completed, the vibration signal, dynamic balance parameters and temperature data are synchronously integrated according to the same timestamp to obtain the time series of the comprehensive parameters of the bearing operation status, which contains comprehensive information on the bearing operation status.

[0030] Multi-dimensional feature extraction is performed on the time series of comprehensive parameters of the bearing operation status. The vibration signal is decomposed at multiple scales through wavelet packet transform to extract the energy distribution characteristics of different frequency bands; then, the mean, variance, peak factor and waveform factor are extracted from the dynamic balancing parameters to obtain the quantitative evaluation parameters of balance; then, the temperature data and speed data are correlated and analyzed, and the temperature rise rate per unit speed is calculated as the thermodynamic wear index; then, the vibration energy distribution characteristics are envelope demodulated through Hilbert transform to extract the characteristic frequency energy of the inner ring, outer ring, rolling element and cage of the bearing; finally, the quantitative evaluation parameters of balance are constructed as the bearing balance index, and the thermodynamic wear index and the bearing fault characteristic spectrum are constructed as the bearing wear index, which together constitute the bearing degradation index system.

[0031] The formula ratio of the metal self-repairing material is determined based on the bearing degradation index system. First, magnesium hydroxysilicate is selected as the main component, boron and tin are added as auxiliary components, the material is ground to a particle size of less than 100 nanometers by ball milling, and treated with a surface modifier. Then the treated metal self-repairing material is added to the extreme pressure lithium-based grease at a ratio of 5%, and the repair grease compound is formed by high-speed stirring. During the shutdown maintenance of the wind turbine, the bearing seat oil filling port cover is removed, and the repair grease compound is injected. The injection amount is 70%-80% of the bearing cavity volume. The bearing is manually rotated several times to ensure that the grease is evenly distributed. Then the wind turbine is started and operated at low speed for 4 hours to form a repair transition period. Dynamic monitoring of the repaired bearings is carried out. The vibration signal is periodically collected and the vibration root mean square value change rate is calculated to form a vibration attenuation curve; the bearing imbalance is measured every 4 hours, and the data is recorded to form a balance change sequence; the difference between the bearing temperature and the ambient temperature is continuously monitored, the unit load temperature rise ratio is calculated, and the temperature rise change curve is generated. Based on these data, a three-dimensional feature space is constructed, and the data points are projected to form a repair trajectory. The turning points of the trajectory are identified through segmented fitting, and the repair process is divided into four stages: fine grinding, adsorption cleaning, repair layer generation, and repair layer maintenance. The repair stage discrimination results are obtained. At the same time, the balance improvement data is obtained by calculating the difference in imbalance before and after repair, the phase angle change, and the balance stabilization time.

[0032] The duration proportion and conversion rate of each stage are extracted from the repair stage judgment results to construct the repair dynamic feature vector; the balance improvement data is divided into time windows, and the balance change gradient and stability in each window are calculated to form the balance stability time series characteristics. These features are time-resampled by sliding window technology to generate a training time series sample set. Then the sample set is trained using a long short-term memory network to establish a bearing balance stability predictor, and future operating condition parameters are input to predict the change trend of the bearing balance state. According to the predicted trend and historical degradation data, the balance weight coefficient and wear weight coefficient are calculated through bidirectional weight allocation to form a comprehensive evaluation weight matrix. This matrix is ​​weighted and normalized with the current state parameters of the bearing to obtain the bearing health status assessment index.

[0033] Finally, the health status scoring standard is set according to the bearing health status assessment index, and the index is mapped to a 0-100 point range to establish a bearing health status classification table. Bearing failure cases with similar degradation modes are extracted from the historical database to construct a life mapping function. The current bearing health status assessment index is converted through this function, and the theoretical value of the remaining service life is calculated in combination with the operating condition correction parameters. The efficiency of the metal self-repairing material is analyzed according to the results of the repair stage, and the ratio of magnesium hydroxysilicate, boron and tin in the material formula is adjusted to form an optimized self-repair strategy. The bearing health status score, balance improvement data, degradation trend prediction results and optimized metal self-repair strategy are integrated into an evaluation report. Finally, according to the theoretical value of the remaining service life and the change rate of the health status score, the Weibull distribution fitting method is used to correct it and obtain the prediction result of the remaining service life of the bearing.

[0034] In the embodiment of the present application, multi-dimensional parameter data of the bearing is collected by vibration sensors, balance testers and temperature sensors to form a time series of comprehensive parameters of the bearing operating status, providing a comprehensive data basis for subsequent analysis and avoiding the problem of incomplete diagnosis caused by a single parameter of the traditional monitoring method; a multi-dimensional feature extraction algorithm is used to deeply mine the balance characteristics and wear characteristics of the bearing, and a bearing degradation index system is constructed to achieve accurate quantification of the health status of the bearing, providing a scientific basis for the optimization of the metal self-repair material formula; the metal self-repair material formula is customized according to the bearing degradation index system and injected into the bearing, which solves the limitation that the traditional unified formula is not suitable for different damage characteristics and improves the repair efficiency; the repaired bearing is subjected to dynamic balancing test and working status monitoring, and the judgment results and balance improvement data of the repair stage are obtained, so as to achieve full monitoring of the repair process and timely detect and adjust abnormal conditions; a long short-term memory network is used to predict and analyze the balance stability and degradation trend of the bearing The algorithm can capture the complex nonlinear relationship in the long time series of bearing status parameters, and is particularly suitable for wind farm environments with variable bearing working conditions and large data noise. The contribution of the algorithm is reflected in the upgrade of traditional prediction methods based on experience or simple statistics to the level of intelligent prediction based on deep learning, making the prediction results closer to the actual degradation law. Finally, the bearing balance and degradation assessment report is generated according to the bearing health status assessment index, and the metal self-repair strategy is optimized. At the same time, the remaining service life of the bearing is predicted, which provides a scientific basis for wind farm operation and maintenance decision-making, realizes the transformation from post-maintenance to predictive maintenance, greatly reduces the unplanned downtime and maintenance costs of wind turbines caused by bearing failures, and improves the operational efficiency and economic benefits of wind farms. In addition, this method can achieve repair and evaluation without disassembling the bearing, reduces the downtime and maintenance time, and ensures the continuous and stable power generation of wind farms.

[0035] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0036] A triaxial acceleration sensor is installed on the wind turbine generator to collect bearing vibration signals. The sampling frequency is set to 10kHz to obtain the original bearing vibration data.

[0037] The generator speed data and power output data are collected through the wind turbine main control system to obtain the generator operating parameters;

[0038] The balance tester is used to collect dynamic balance data of the generator bearing, measure the unbalance amount and unbalance phase angle when the bearing rotates, and obtain the dynamic balance characteristic parameters of the bearing;

[0039] The bearing seat temperature is continuously monitored by the embedded temperature sensor, and the ambient temperature is recorded at the same time to obtain the bearing temperature rise data;

[0040] The original vibration data of the bearing, the operating parameters of the generator, the characteristic parameters of the bearing dynamic balance and the temperature rise data of the bearing are synchronously integrated according to the timestamp to form a time series database containing multi-dimensional parameters, and the time series of the comprehensive parameters of the bearing operating status is obtained.

[0041] Specifically, the three-axis acceleration sensor is installed on the wind turbine generator to fully capture the vibration characteristics of the bearing in three directions (radial, axial and tangential). The three-axis acceleration sensor usually uses a piezoelectric sensor to convert mechanical vibration into an electrical signal. The sensor installation position is selected on the generator bearing seat, as close to the bearing body as possible without affecting the normal operation of the bearing. The sampling frequency is set to 10kHz, which is based on the Nyquist sampling theorem to ensure that the characteristic frequency of the bearing fault can be captured. These frequencies are usually in the range of several thousand hertz. The collected data is stored as a time domain vibration waveform to form the original vibration data of the bearing, which contains an information matrix of two dimensions: amplitude and time. The generator speed sensor and power monitoring device are integrated in the main control system of the wind turbine. The speed sensor is usually based on the Hall effect principle or the photoelectric principle and is installed at the end of the generator shaft. It generates one or more pulse signals per revolution. The accurate speed value is obtained by calculating the number of pulses per unit time. The power output data is collected by current and voltage sensors, and the real-time power value is obtained after multiplication. These two types of data are sampled at a frequency of not less than 1Hz by the data acquisition card of the main control system, recorded in the form of time series, and constitute the generator operating parameters. These parameters are crucial for subsequent analysis, since the vibration characteristics and temperature behavior of the bearing are directly related to the speed and load.

[0042] A balance tester is a device specially used to measure the dynamic balance state of rotating parts. A balance tester with a phase measurement function is required to collect dynamic balance data for generator bearings. During the test, a phase mark point is fixed on the generator shaft. The tester captures the reference position for each revolution through a photoelectric or magnetic sensor, and measures the vibration force of the support point through a force sensor. The imbalance amount indicates the degree of uneven mass distribution of the rotating body, and the unit is gram·millimeter (g·mm). The imbalance phase angle indicates the angular position of the maximum imbalance mass point relative to the reference point, and the unit is degree (°). The measured imbalance amount and imbalance phase angle constitute the dynamic balance characteristic parameters of the bearing. These parameters directly reflect the balance state of the bearing and are important indicators for evaluating the quality and health of the bearing.

[0043] The embedded temperature sensor uses a thermocouple or PT100 temperature sensor, which is installed on the outer surface of the bearing seat, as close to the working area of ​​the bearing as possible. At the same time, an ambient temperature sensor is set in the cabin to record the ambient temperature. The temperature data acquisition frequency is usually set to once per minute to form a continuous temperature time series. By calculating the difference between the bearing seat temperature and the ambient temperature, the bearing temperature rise data is obtained, which eliminates the influence of ambient temperature fluctuations on the measurement results and more accurately reflects the friction heat generation inside the bearing. The bearing temperature rise is an important indicator of the health status of the bearing. Abnormal temperature rise often indicates that the bearing has problems such as abnormal friction, poor lubrication or overload. After the above four types of data (bearing original vibration data, generator operating parameters, bearing dynamic balance characteristic parameters and bearing temperature rise data) are collected, they need to be synchronized and integrated. First, a timestamp in a unified format is added to each data point, accurate to the millisecond level. Then, through data interpolation processing, the data with different sampling frequencies are resampled to a unified time point, so that all data are aligned in the time dimension. The integrated data set contains multi-dimensional parameters to form a time series database, that is, the time series of comprehensive parameters of the bearing operating status. This time series is the data basis for subsequent bearing health status assessment. By associating parameters of different dimensions, it can comprehensively reflect the operating status and degree of deterioration of the bearing.

[0044] Taking a 1.5MW wind turbine in a wind farm as an example, the data collection process of its main shaft bearing (model FD-239 / 750CA / W33) is described. The triaxial acceleration sensors are installed on the top, side and end faces of the bearing seat. The sampling frequency is set to 10kHz, and vibration data is collected for 10 seconds every hour. The main control system records the generator speed and power output at a frequency of 1Hz. The balance tester measures the bearings during regular maintenance and records the imbalance and phase angle. The temperature sensor collects the bearing seat temperature and ambient temperature once a minute. These data are synchronously integrated according to the timestamp to form a comprehensive time series containing multi-dimensional parameters such as vibration amplitude, spectrum characteristics, speed, power, imbalance, phase angle, temperature, etc., which records the operating status of the bearing under different working conditions.

[0045] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0046] The vibration signal in the time series of comprehensive parameters of the bearing operation status is decomposed at multiple scales through wavelet packet transform to obtain the vibration energy distribution characteristics.

[0047] The mean, variance, peak factor and waveform factor of dynamic balancing parameters are extracted from the time series of comprehensive parameters of bearing operation status to obtain quantitative evaluation parameters of balance.

[0048] The temperature data in the time series of the comprehensive parameters of the bearing operation status and the generator speed data are correlated and analyzed to calculate the temperature rise rate per unit speed and obtain the thermodynamic wear index.

[0049] The vibration energy distribution characteristics are subjected to envelope demodulation through Hilbert transform, and the characteristic frequency energy of the inner ring, outer ring, rolling element and cage of the bearing is extracted to obtain the characteristic spectrum of the bearing fault.

[0050] The bearing balance index is constructed based on the quantitative evaluation parameters of balance, and the bearing wear index is constructed based on the thermodynamic wear index and the bearing fault characteristic spectrum, and the bearing double degradation evaluation benchmark is obtained;

[0051] The bearing balance index and bearing wear index are normalized and weight coefficients are set. A comprehensive evaluation standard is generated through weighted fusion to obtain the bearing degradation index system.

[0052] Specifically, the vibration signal in the time series of the comprehensive parameters of the bearing operation status is processed. Wavelet packet transform is a time-frequency domain analysis method. Compared with the traditional Fourier transform, it can have good resolution in both the time domain and the frequency domain. The process of wavelet packet transform of the vibration signal includes: selecting a suitable wavelet basis function (such as Daubechies wavelet), determining the number of decomposition layers (usually 3-5 layers), and then decomposing the vibration signal through high-pass and low-pass filters in turn to form a tree structure. Each node represents a signal in a specific frequency band, and the energy value of each node is calculated to form the vibration energy distribution characteristics. This method can effectively extract the vibration characteristics of the bearing in different frequency bands and provide more detailed information for fault diagnosis. Extracting the statistical characteristics of dynamic balancing parameters from the time series of the comprehensive parameters of the bearing operation status is an important means to evaluate the balance of the bearing. The mean reflects the average level of the imbalance, and the calculation method is to calculate the arithmetic mean of the imbalance data within a certain time window. The variance describes the degree of fluctuation of the imbalance, and the calculation method is the sum of the squares of the differences between each measured value and the mean divided by the number of samples. The peak factor is the ratio of the maximum imbalance to the root mean square value, which reflects the peak characteristics of the imbalance degree. The waveform factor is the ratio of the root mean square value to the average value, which characterizes the complexity of the unbalance waveform. These four parameters together constitute the quantitative evaluation parameters of balance, which comprehensively describe the dynamic balance state of the bearing.

[0053] The calculation of thermodynamic wear index depends on the correlation analysis of temperature data and speed data. The calculation formula is: in, Indicates the bearing temperature rise (the difference between the bearing temperature and the ambient temperature), is the bearing temperature, is the ambient temperature, is the generator speed. When calculating, first pair the temperature data with the speed data, then calculate the ratio of temperature rise to speed at each time point, and finally take the average value within a certain time window as the unit speed temperature rise rate. This indicator can reflect the intensity of bearing friction heat generation and is an important parameter for evaluating the bearing wear status.

[0054] Hilbert transform is an effective tool for extracting bearing fault characteristics. It converts real signals into analytical signals to obtain the instantaneous amplitude of the signal. The vibration energy distribution characteristics are Hilbert transformed to obtain its envelope signal, and then the envelope signal is Fourier transformed to obtain the envelope spectrum. From the envelope spectrum, the characteristic frequencies and amplitudes of the inner ring, outer ring, rolling element and cage of the bearing can be identified. These characteristic frequencies are related to the bearing geometry parameters and speed. The calculation formula is as follows: in, , , , They are the characteristic frequencies of the inner ring, outer ring, rolling element and cage, is the number of rolling elements, is the relative speed, is the rolling element diameter, is the pitch diameter, is the contact angle. The energy values ​​corresponding to these characteristic frequencies are extracted to form the characteristic spectrum of bearing faults.

[0055] Bearing balance index The construction is based on the quantitative evaluation parameters of balance, and the calculation formula is: in, is the mean value of the imbalance, is the variance of the imbalance, is the crest factor, is the form factor, , , , is the reference value of the corresponding parameter (usually taken from the data of a healthy bearing), , , , is the weight coefficient of each parameter.

[0056] Bearing wear index The construction of depends on the thermodynamic wear index and the bearing fault characteristic spectrum, and the calculation formula is: in, is the rate of temperature rise per unit speed, , , , are the energy values ​​of the characteristic frequencies of the inner ring, outer ring, rolling element and cage, , , , , is the reference value of the corresponding parameter, , , , , It is the weight coefficient of each parameter. The bearing balance index and bearing wear index together constitute the bearing double degradation evaluation benchmark.

[0057] Finally, the bearing balance index and bearing wear index are normalized and weighted to obtain the bearing degradation index system. , the calculation formula is: in, and are the minimum and maximum values ​​of the bearing balance index, and are the minimum and maximum values ​​of the bearing wear index, and is the weight coefficient of the bearing balance index and the bearing wear index, satisfying + = 1.

[0058] Taking the main shaft bearing (FD-239 / 750CA / W33) of a 1.5MW wind turbine in a wind farm as an example, the data processing process is as follows: First, the collected 10kHz vibration signal is decomposed by wavelet packet, and the db4 wavelet basis function is selected to decompose it into 4 layers to obtain the energy distribution of 16 frequency bands. Then, the mean value of the unbalance calculated from the dynamic balancing test data is 32g·mm, the variance is 15.6, the peak factor is 3.4, and the waveform factor is 1.8. Then, the difference between the bearing temperature (average 65℃) and the ambient temperature (average 20℃) and the average speed of the generator (1200r / min) are correlated and analyzed, and the unit speed temperature rise rate is calculated to be 0.0375℃ / (r / min). Subsequently, the vibration signal is envelope demodulated by Hilbert transform, and the energy values ​​of the inner ring characteristic frequency (98.6Hz), outer ring characteristic frequency (65.2Hz), rolling element characteristic frequency (32.8Hz) and cage characteristic frequency (5.2Hz) are extracted. Based on these data, the bearing balance index was calculated to be 0.68, the bearing wear index was 0.72, and after normalization and weighted fusion (weights were 0.4 and 0.6 respectively), the bearing degradation index system value was 0.704. This index directly reflects the health status of the bearing and provides a scientific basis for the injection strategy of metal self-repairing materials.

[0059] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0060] According to the bearing balance index and bearing wear index in the bearing degradation index system, the formula ratio of the metal self-repairing material is determined to prepare nano-scale self-repairing particles containing hydroxy magnesium silicate, boron and tin components;

[0061] The self-repairing particles are ground to a particle size of less than 100 nanometers by ball milling and treated with a surface modifier to prepare a metal self-repairing material with high dispersibility;

[0062] Add the metal self-repairing material to the extreme pressure lithium-based grease at a ratio of 5%, mix evenly through a high-speed stirring device, and form a repairing grease complex;

[0063] During the shutdown maintenance of the wind turbine, remove the oil filling cap on the generator bearing seat and inject the repair grease compound into the bearing cavity. The injection amount is controlled at 70% to 80% of the bearing cavity.

[0064] By manually rotating the bearing for several turns, the repair grease compound is evenly distributed on the bearing rolling element and raceway surface to form an initial repair layer;

[0065] The bearing seat is restored to its original state and the wind turbine is started, so that the generator runs at a low speed for four hours to form a repair transition period and obtain a repaired bearing.

[0066] Specifically, according to the bearing balance index and bearing wear index in the bearing degradation index system, the formula ratio of the metal self-repairing material needs to be determined. This process first analyzes the bearing wear state: when the bearing wear index is high and the balance index is low, it indicates that the main problem of the bearing is surface wear. At this time, the proportion of magnesium hydroxysilicate in the self-repairing material should be increased; when the balance index is high and the wear index is low, it indicates that the main problem of the bearing is imbalance. At this time, the proportion of boron and tin should be increased to improve the fluidity and adhesion of the material. Hydroxy magnesium silicate Mg3(Si2O5)(OH)4 is a silicate mineral with a layered crystal structure. Under the action of friction heat and pressure, micro-sintering and micro-metallurgical processes can occur to form a metal ceramic repair layer. Boron can improve the hardness and wear resistance of the material, while tin helps to improve the adhesion and film-forming properties of the material. The typical ratio of the three components is 75% magnesium hydroxysilicate, 15% boron, and 10% tin, but it will be adjusted according to the specific bearing degradation conditions. The preparation of nano-scale self-repairing particles adopts ball milling, which is a mechanical crushing method that can effectively reduce the particle size. The specific operation is to mix the raw materials in a certain proportion and put them into a ball mill, add an appropriate amount of grinding media (usually steel balls or zirconia balls of different diameters), set the ball-to-material ratio (mass ratio of balls to materials) to 5:1, the rotation speed is 300-400 rpm, and the grinding time is 4-6 hours. During the grinding process, the material is gradually crushed into nano-scale particles under the impact, friction and extrusion of the grinding media. After grinding, the particle size distribution is detected by a laser particle size analyzer to ensure that the particle size of all particles is less than 100 nanometers. Then use a surface modifier for treatment. The surface modifier usually uses alkyl silane compounds, which can form an organic coating layer on the surface of the particles to prevent particle agglomeration and improve its dispersibility and stability in grease. The treatment method is to mix the nanoparticles and the surface modifier in a mass ratio of 1:0.05, stir and react in a solvent for 2 hours, and then filter, wash and dry to obtain a metal self-healing material with high dispersibility. Adding metal self-healing materials to grease is a key step in achieving bearing self-repair. Extreme pressure lithium-based grease is selected as the carrier because it has good compressive resistance, water stability and mechanical stability, and is suitable for the working conditions of the main shaft and generator bearings of wind turbines. The addition ratio is set to 5%, which is the optimal concentration verified by a large number of experiments. It can ensure sufficient repair ability without affecting the basic performance of the grease. The mixing process uses a high-speed stirring device with a speed set to 1500-2000 rpm and a stirring time of 30 minutes to ensure that the self-healing material is evenly dispersed in the grease to form a stable repair grease complex. After mixing, the cone penetration test is used to confirm that the consistency level of the compound has not changed significantly, ensuring that it still meets the bearing lubrication requirements.

[0067] Re-lubricating the bearing grease during wind turbine downtime maintenance is a necessary measure to ensure safety. First, clean the area around the bearing seat to prevent impurities from entering. Then remove the oil filling cap on the generator bearing seat, which is usually located on the top or side of the bearing seat. Check the condition of the original grease. If there is obvious discoloration, hardening or contamination, you need to remove the old grease first. Use a special grease gun to inject the repair grease compound into the bearing cavity, and control the injection amount to 70% to 80% of the bearing cavity. Controlling the injection amount is very important: less than 70% will result in insufficient lubrication, causing overheating and premature wear of the bearing; more than 80% will increase the running resistance of the bearing and generate additional heat. The injection process should be maintained at a uniform and slow speed to avoid bubbles or uneven distribution.

[0068] After the injection is completed, the bearing needs to be rotated manually for several turns, usually 10-15 turns. This process has two purposes: one is to promote the uniform distribution of the repair grease compound on the rolling element and raceway surface of the bearing; the other is to make the self-repairing material initially contact the worn part to form an initial repair layer. The manual rotation should be kept stable, and the rotation time per turn is about 2-3 seconds. It should not be too fast (affecting the distribution of grease) or too slow (affecting the formation of the initial repair layer). After the rotation is completed, observe the operation of the bearing to confirm that there is no abnormal resistance or noise.

[0069] After restoring the bearing seat to its original state, the bearing needs to be operated at a low speed to form a repair transition period. Low-speed operation means controlling the generator speed within the range of 30%-50% of the rated speed. The purpose of this is to reduce the workload of the bearing and allow the self-repairing material enough time to work on the friction surface. The operating time is set to 4 hours. During this time, under the action of friction heat and pressure, the self-repairing material will gradually form a stable repair layer. After the transition period, the bearing can resume normal operation, and the repair layer will continue to improve in subsequent operations, ultimately achieving the ideal repair effect.

[0070] Taking the main shaft bearing of a 1.5MW wind turbine in a wind farm as an example, the bearing model is FD-239 / 750CA / W33. The bearing wear index is 0.72 and the balance index is 0.65 through bearing degradation assessment. According to these indicators, the formula of the self-repairing material is determined to be 78% magnesium silicate hydroxy, 12% boron, and 10% tin. The material was ground for 8 hours by ball milling, and the particle size test showed that 85% of the particles were less than 80 nanometers. After being treated with an alkyl silane surface modifier, the treated self-repairing material was added to No. 2 extreme pressure lithium-based grease at a ratio of 5%, and stirred at high speed for 40 minutes to form a repair grease compound. During the planned shutdown maintenance of the wind turbine, the bearing seat oil filling port cover was removed and about 2.5 kg of the repair grease compound was injected, accounting for 75% of the bearing cavity. After manually rotating the bearing 12 times, the bearing seat was restored and the wind turbine was controlled to run at a low speed of 500 rpm for 4 hours. Tests one week after the repair showed that the bearing temperature had dropped by 8°C, the vibration amplitude had dropped by 25%, and the balance state had improved significantly, confirming the effectiveness of this method in wind turbine bearing maintenance.

[0071] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0072] After the metal self-repairing material is injected into the bearing, the vibration signal is periodically collected, and the vibration root mean square value change rate is calculated through spectrum analysis to form a vibration attenuation curve;

[0073] Use a balance tester to measure the unbalance of the bearing every four hours, record the data to form a balance change sequence, and obtain the dynamic balance improvement trend;

[0074] The temperature sensor continuously monitors the difference between the bearing temperature and the ambient temperature, calculates the unit load temperature rise ratio, and generates a temperature rise curve;

[0075] According to the vibration attenuation curve, dynamic balance improvement trend and temperature rise change curve, a three-dimensional feature space is constructed, and the data points are projected into the three-dimensional feature space to form a repair trajectory;

[0076] The repair trajectory is segmented and fitted to identify the turning points in the trajectory. According to the turning points, the repair process is divided into the fine grinding stage, the adsorption and cleaning stage, the repair layer generation stage and the repair layer maintenance stage to obtain the repair stage discrimination result.

[0077] The imbalance difference, phase angle change and balance stabilization time before and after repair are calculated, and the balance improvement data is obtained by combining the stable segment data in the dynamic balance improvement trend.

[0078] Specifically, after the metal self-repairing material is injected into the bearing, the vibration signal needs to be collected periodically. The monitoring period is usually set to once every 4 hours, the sampling duration is 10 seconds, and the sampling frequency is 10kHz. The collected vibration time domain signal is converted into a spectrum by fast Fourier transform (FFT), and then the vibration root mean square value is calculated, which reflects the magnitude of the vibration energy. The vibration root mean square value change rate is calculated by subtracting the root mean square value at the initial moment from the root mean square value at the current moment, and then dividing it by the root mean square value at the initial moment. The resulting data sequence forms a vibration attenuation curve. This curve intuitively reflects the trend of the bearing vibration energy over time and is an important indicator for evaluating the repair effect of metal self-repairing materials. Balance test is an important means to evaluate the dynamic balance state of the bearing. The balance tester is used to measure the unbalance of the bearing every four hours. The measurement needs to be carried out under the same speed conditions to ensure the comparability of the data. The tester measures the vibration force of the support point through a force sensor, and calculates the unbalance and phase angle in combination with the phase information provided by the phase sensor. These data are arranged in chronological order to form a balance change sequence. By performing trend analysis on this sequence and calculating the rate of change and direction of change between consecutive time points, the dynamic balance improvement trend is obtained. This trend can reflect the change of the bearing balance state over time and is an important basis for judging the impact of metal self-repairing materials on the bearing balance performance.

[0079] Temperature monitoring is a direct means of evaluating the friction state of the bearing. The bearing temperature and the ambient temperature are continuously monitored by the temperature sensor at intervals of 1 minute, and the difference between the two is calculated to obtain the bearing temperature rise. At the same time, the current load data (usually expressed in torque or power) is obtained from the main control system of the wind turbine, and the unit load temperature rise ratio is calculated, that is, the bearing temperature rise divided by the current load. This indicator eliminates the influence of load changes on the bearing temperature and more accurately reflects the friction state of the bearing. The calculated unit load temperature rise ratio is arranged in chronological order to generate a temperature rise change curve, which is an intuitive indicator for evaluating the influence of metal self-repairing materials on bearing friction heat. The above three curves (vibration attenuation curve, dynamic balance improvement trend and temperature rise change curve) each reflect different aspects of the bearing repair process. In order to comprehensively analyze the repair effect, it is necessary to construct a three-dimensional feature space with the vibration root mean square value change rate as the X-axis, the balance improvement rate as the Y-axis, and the unit load temperature rise ratio change rate as the Z-axis. The data points at different times are projected into this three-dimensional space in chronological order to form a continuous repair trajectory. This trajectory intuitively shows the coordinated changes of various performance indicators of the bearing during the repair process, providing a basis for the subsequent stage division.

[0080] Segmented fitting of the repair trajectory is a key step in identifying different stages of repair. First, the trajectory is mathematically described using methods such as piecewise linear fitting or polynomial fitting, and then the turning points in the trajectory are identified by calculating the derivative or curvature change of the fitting curve. These turning points correspond to the state changes in the repair process and are the basis for dividing the repair stages. According to the working mechanism of metal self-repair technology, the repair process can be divided into four stages: fine grinding stage (characterized by a slight increase in the root mean square value of vibration, a slight deterioration in balance, and an increase in temperature rise), adsorption and cleaning stage (vibration begins to decrease, balance tends to stabilize, and temperature rise continues to increase), repair layer generation stage (vibration rapidly decreases, balance improves significantly, and temperature rise begins to decrease) and repair layer maintenance stage (all indicators tend to stabilize). By comparing the matching degree of the characteristics of each segment of the trajectory with the characteristics of the above stages, the time boundary of the repair stage is determined, and the repair stage discrimination result is obtained.

[0081] The calculation of balance improvement data involves multiple indicators, which are mainly achieved through the following formula: in, Indicates the difference in unbalance. and are the imbalance amounts before and after repair, respectively; represents the phase angle change, and Especially the phase angle before and after restoration; Indicates the equilibrium stability time, is the time point when a steady state is reached. is the point in time when repair begins; is the balance improvement index, , , and are weight coefficients, corresponding to the weights of unbalance improvement, phase angle change, stabilization time and stability respectively; is the reference stabilization time; and They are the balance stability indicators before and after repair, respectively, and are usually expressed by the standard deviation of the balance.

[0082] Taking the main shaft bearing of a 1.5MW wind turbine as an example, a 7-day monitoring was carried out after the metal self-repairing material was injected. On the first day, the vibration root mean square value slightly increased from 4.5mm / s to 4.8mm / s, the imbalance increased from 38g·mm to 42g·mm, and the unit load temperature rise ratio increased from 0.12℃ / kNm to 0.14℃ / kNm. These data changes indicate that the bearing is in the fine grinding stage. On the second day, the vibration root mean square value began to drop to 4.3mm / s, the imbalance stabilized at around 40g·mm, and the unit load temperature rise ratio continued to rise to 0.15℃ / kNm, indicating that it entered the adsorption cleaning stage. From the third to the fifth day, the vibration root mean square value dropped rapidly to 3.2mm / s, the imbalance dropped to 26g·mm, the phase angle changed from the original 135° to 95°, and the unit load temperature rise ratio began to drop to 0.11℃ / kNm, indicating that the bearing entered the repair layer generation stage. On the sixth and seventh days, all indicators tended to be stable, the vibration root mean square value was stable at about 3.0mm / s, the imbalance fluctuated around 25g·mm, and the unit load temperature rise ratio was stable at 0.10℃ / kNm, indicating that the bearing entered the repair layer maintenance stage. According to the formula, the imbalance difference was 13g·mm, the phase angle changed by 40°, the balance stability time was 96 hours, and the balance stability was improved by 65%. The comprehensive calculation showed that the balance improvement index was 0.72, which intuitively reflects the significant improvement effect of metal self-repairing materials on the balance performance of bearings.

[0083] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0084] Extract the duration proportion and conversion rate of each stage from the repair stage discrimination results to construct the repair dynamic feature vector;

[0085] The balance improvement data is divided into time windows, and the balance change gradient and stability within each time window are calculated to form the balance stability time series characteristics;

[0086] The dynamic feature vector and the time series features of the balance stability are resampled in time through the sliding window technology to generate a time series sample set for training;

[0087] The long short-term memory network is used to train the training time series sample set, a bearing balance stability predictor is established, and future operating condition parameters are input to predict the change trend of the bearing balance state;

[0088] According to the bearing balance state change trend and historical degradation data, the balance weight coefficient and wear weight coefficient are calculated by the two-way weight allocation method to form a comprehensive evaluation weight matrix;

[0089] The comprehensive evaluation weight matrix is ​​weighted multiplied by the current state parameters of the bearing, and the bearing health status assessment index is obtained after normalization.

[0090] Specifically, extracting the duration ratio and conversion rate of each stage from the repair stage discrimination results is the basic step to construct the repair dynamic feature vector. The repair stage discrimination results include the start and end time points of the four stages: fine grinding stage, adsorption cleaning stage, repair layer generation stage, and repair layer maintenance stage. The duration ratio of each stage refers to the proportion of the duration of each stage to the total repair time, which is calculated by dividing the duration of each stage by the total repair time. The conversion rate describes the speed of transition from one stage to the next stage, which is calculated by dividing the change in relevant parameters (such as vibration root mean square value and imbalance) before and after the stage conversion by the time required for conversion. The combination of these two types of indicators forms a repair dynamic feature vector, which intuitively reflects the dynamic process of metal self-repairing materials in bearings and is an important basis for predicting future repair effects. The balance improvement data contains information such as imbalance difference, phase angle change, and balance stabilization time, and needs to be segmented and processed according to time windows. The time window refers to dividing the entire monitoring period into several equal time periods, and each window contains a certain number of continuous data points. For each time window, the balance gradient is calculated, that is, the rate of change of the balance parameter within the window. The calculation method is to subtract the balance parameter value at the end of the window from the parameter value at the beginning of the window, and then divide it by the time span of the window. At the same time, the standard deviation of the balance data within the window is calculated as a measure of the stability of the balance. These two types of indicators are arranged in chronological order to form a time series feature of balance stability. This feature intuitively shows the change law of the bearing balance state over time, providing a time series data basis for subsequent prediction analysis.

[0091] Sliding window technology is a commonly used time series processing method for resampling and feature extraction of time series data. In this method, the application of sliding window is to convert the repair dynamic feature vector and balance stability time series feature into a data format suitable for machine learning algorithm input. In specific implementation, the window size (for example, 24 hours) and the sliding step size (for example, 4 hours) are set. The window starts from the beginning of the time series and slides back one step each time. All feature values ​​in the window are extracted to form a sample. As the window slides continuously, multiple samples are generated to form a training time series sample set. This sample set contains the dynamic change information of the bearing state during the repair process and is the data basis for training the prediction model. The long short-term memory network (LSTM) is a special recurrent neural network that is good at processing long-term dependencies in time series data. In the bearing health status assessment, the input of the LSTM network is the previously generated training time series sample set, and the output is the predicted value of the bearing balance state at a future moment. The network structure includes an input layer, an LSTM layer, and an output layer. The number of neurons in the LSTM layer is usually set to 32-128, which is adjusted according to the data complexity. The training process uses the back propagation algorithm, the loss function selects the mean square error (MSE), and the optimization algorithm uses Adam. After the training is completed, the future operating parameters (such as the expected load level, operating speed, etc.) are input into the trained network to obtain the prediction results of the change trend of the bearing balance state, including the predicted values ​​of key indicators such as imbalance and phase angle in the future (such as 7 days).

[0092] The two-way weight allocation method is an algorithm that dynamically adjusts the weights of evaluation indicators based on historical data and predicted trends. The health status of a bearing is affected by two main aspects: balance and wear. The importance of the two aspects varies with the working conditions and use stages of the bearing. The two-way weight allocation method achieves a dynamic trade-off between the two aspects by calculating the balance weight coefficient and the wear weight coefficient. When calculating, the change amplitude and stability in the change trend of the bearing balance state are first analyzed. Indicators with large change amplitudes and poor stability should be given higher weights. Then, combined with the correlation between each indicator and the bearing failure in the historical degradation data, indicators with high correlation are given higher weights. The two-part calculation results are combined to form the balance weight coefficient and the wear weight coefficient to form a comprehensive evaluation weight matrix. This matrix reflects the contribution of different indicators to the health status of the bearing under the current working conditions and is the key to comprehensively evaluating the health status of the bearing. The comprehensive evaluation weight matrix is ​​weighted and multiplied with the current state parameters of the bearing to obtain the original evaluation value, which is then mapped to the interval of 0-100 through normalization to obtain the bearing health status evaluation index. The normalization method is to subtract the historical minimum value from the original evaluation value, divide it by the difference between the historical maximum and minimum values, and then multiply it by 100. This index intuitively reflects the health status of the bearing. The higher the value, the better the bearing condition. It is an important basis for bearing maintenance decisions.

[0093] Taking the main shaft bearing of a 1.5MW wind turbine in a wind farm as an example, the repair stage discrimination results analysis found that the fine grinding stage accounted for 15% of the total repair time, the adsorption cleaning stage accounted for 20%, the repair layer generation stage accounted for 45%, and the repair layer maintenance stage accounted for 20%. The conversion rates of each stage were 0.06 / hour, 0.12 / hour, and 0.03 / hour, respectively. These data were combined into the repair dynamic feature vector (0.15, 0.20, 0.45, 0.20, 0.06, 0.12, 0.03). The balance improvement data was divided into 24-hour time windows, and the balance gradients of each window were calculated to be -1.2g·mm / day, -2.5g·mm / day, -3.8g·mm / day, -2.0g·mm / day, and -0.5g·mm / day, respectively, and the corresponding balance stability (standard deviation) was 3.2, 2.8, 1.9, 1.5, and 1.2, respectively. The sliding window technology was used with a window size of 72 hours and a step size of 24 hours to generate a training sample set. The bearing balance stability predictor was obtained through LSTM network training to predict the balance state change trend in the next 7 days. Combined with historical degradation data, the balance weight coefficient was calculated to be 0.62 and the wear weight coefficient was 0.38, forming a comprehensive evaluation weight matrix. After weighted calculation and normalization, the bearing health status assessment index was 82, indicating that the bearing was in good condition and was expected to operate normally for a long time, confirming the effectiveness of metal self-repair technology for wind turbine bearing maintenance.

[0094] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0095] Set the health status scoring standard according to the bearing health status assessment index, map the index value to a scoring range of zero to one hundred points, and establish a bearing health status grading table;

[0096] Extract bearing failure cases with similar degradation patterns from the historical database, extract the corresponding relationship between failure time points and health status assessment indexes, and construct a life mapping function;

[0097] The current bearing health status assessment index is converted through the life mapping function, and the theoretical value of the remaining service life of the bearing is calculated by combining the operating condition correction parameters;

[0098] Analyze the efficiency of metal self-repairing materials based on the results of the repair stage, and adjust the ratio of magnesium hydroxysilicate, boron and tin in the metal self-repairing material formula based on the balance improvement data to form an optimized metal self-repairing strategy;

[0099] Integrate the bearing health status score, balance improvement data, degradation trend prediction results and optimized metal self-repair strategy to generate a bearing balance and degradation assessment report;

[0100] According to the theoretical value of the remaining service life and the change rate of the bearing health status score, the correction calculation is performed through the Weibull distribution fitting method to obtain the prediction result of the remaining service life of the bearing.

[0101] Specifically, the health status scoring standard is set according to the bearing health status assessment index, and the index value is mapped to a scoring range of zero to one hundred points. This mapping process is based on the bearing performance requirements and historical operation data of the wind power industry. Usually, 0-40 points are defined as the danger zone (indicating that the bearing has been seriously deteriorated and needs to be replaced immediately), 41-60 points are defined as the warning zone (indicating that the bearing has begun to deteriorate significantly and needs to be closely monitored), 61-80 points are defined as the attention zone (indicating that the bearing performance has declined but is still within an acceptable range), and 81-100 points are defined as the health zone (indicating that the bearing is in good condition). The scoring standard takes into account multiple factors such as the vibration characteristics, balance performance, temperature rise, and noise level of the bearing. The weight of each factor is adjusted according to the operating characteristics of the wind turbine. These area divisions and corresponding treatment suggestions are organized into a bearing health status classification table to provide an intuitive reference for operation and maintenance decisions. Extracting bearing failure cases with similar degradation modes from the historical database is a key step in establishing a life prediction model. The judgment criteria for similar degradation modes include the same or similar bearing models, similar working conditions, and similar degradation characteristics (such as vibration spectrum characteristics, temperature rise mode, etc.). For each failure case, the corresponding relationship between the failure time point and the health status assessment index is extracted, and the change curve of the health status assessment index from healthy to failure is recorded. By analyzing the data of multiple failure cases, the statistical relationship between the health status assessment index and the remaining service life is identified, and the life mapping function is constructed using regression analysis methods (such as polynomial regression, exponential regression, etc.). This function takes the health status assessment index as the independent variable and the remaining service life as the dependent variable to establish a mathematical relationship between the two.

[0102] The current bearing health status assessment index is converted through the life mapping function to obtain a preliminary estimate of the remaining service life. Then, it is adjusted in combination with the operating condition correction parameters, which take into account the impact of current and expected load levels, speed range, ambient temperature, start-stop frequency and other factors on the bearing life. The correction method is to multiply the preliminary estimate by the correction coefficient of each operating condition parameter to obtain the theoretical value of the remaining service life of the bearing taking into account the actual operating conditions. This theoretical value more accurately reflects the expected life of the bearing under actual conditions and provides a scientific basis for operation and maintenance planning.

[0103] Analyzing the efficiency of metal self-repairing materials according to the results of repair stage discrimination is the basis for optimizing the repair strategy. The evaluation of the efficiency includes the duration of each repair stage, the rate and amplitude of bearing performance improvement in each stage. If the fine grinding stage and adsorption cleaning stage are long, and the repair layer generation stage is short, it indicates that the current self-repairing material has strong grinding and cleaning capabilities but weak film-forming capabilities, and the proportion of film-forming components needs to be increased. Combined with the balance improvement data, the indicators such as the improvement effect of imbalance, phase angle stability and balance stability time are specifically analyzed to determine the impact of the self-repairing material on the dynamic balance performance of the bearing. According to these analysis results, the proportion of magnesium hydroxysilicate, boron and tin in the metal self-repairing material formula is adjusted. Magnesium hydroxysilicate mainly provides the basic material of the repair layer, boron enhances the hardness and wear resistance of the material, and tin improves the adhesion and fluidity of the material. When the demand for balance improvement is greater than the demand for wear repair, the proportion of tin is increased; when the demand for wear repair is more prominent, the proportion of magnesium hydroxysilicate is increased; when the hardness of the repair layer needs to be increased, the proportion of boron is increased. Through this targeted adjustment, an optimized metal self-repairing strategy is formed to improve the repair effect.

[0104] The bearing health status score, balance improvement data, deterioration trend prediction results and optimized metal self-repair strategy are integrated to generate a bearing balance and deterioration assessment report. The report includes basic bearing information (model, installation position, operating time, etc.), current health status score and its position in the classification table, balance performance analysis (including imbalance change trend, phase angle stability, etc.), deterioration trend prediction (including health status index change prediction for a period of time in the future), repair effect evaluation (including performance at each stage and overall improvement effect) and recommended self-repair material formula and maintenance suggestions. This report intuitively presents the health status and repair effect of the bearing, providing a comprehensive reference for subsequent maintenance decisions.

[0105] Finally, according to the theoretical value of the remaining service life and the change rate of the bearing health status score, the Weibull distribution fitting method is used to perform correction calculations to obtain a more accurate prediction result of the remaining service life of the bearing. Weibull distribution is a classic probability distribution that describes the failure law of equipment and is suitable for the life prediction of mechanical components such as bearings. When calculating, the shape parameters and scale parameters of the Weibull distribution are first determined based on historical data, and then the current rate of change of the health status score is substituted into the distribution function to calculate the time when the bearing reaches the preset fault threshold, and the corrected remaining service life prediction value is obtained. This prediction result based on statistical methods takes into account the randomness of bearing degradation and is closer to the actual situation than a simple deterministic prediction.

[0106] Taking the main shaft bearing of a 1.5MW wind turbine in a wind farm as an example, after applying the metal self-repair technology, the health status assessment index of the bearing is 78 points, which corresponds to the upper layer of the attention area in the health status classification table. Five bearing failure cases of similar models and working conditions were extracted from the historical database, and the relationship between the health status index and the remaining life of these cases was analyzed to construct a life mapping function. The health status assessment index of the bearing was substituted into the function, and the actual operating conditions of the current wind turbine (including average wind speed, full load operation time ratio, etc.) were considered. The theoretical value of the remaining service life was calculated to be 9500 hours. By analyzing the results of the repair stage, it was found that the repair layer generation stage of the bearing was short but the effect was obvious, indicating that the film formation speed of the self-repairing material was fast, but the thickness of the repair layer still had room for improvement. Combined with the balance improvement data (the imbalance was reduced by 35%, the phase angle tended to be stable, and the balance stability time was short), the self-repairing material formula was adjusted to increase the proportion of magnesium silicate from 75% to 78%, reduce the proportion of boron from 15% to 12%, and keep the proportion of tin unchanged at 10%. The above information is integrated to generate an evaluation report, and then corrected by Weibull distribution (taking into account the steady downward trend of the health status score), the corrected remaining useful life prediction result is 8,800 hours.

[0107] The above describes the generator bearing degradation assessment method based on metal self-repairing technology in the embodiment of the present application. The following describes the generator bearing degradation assessment system based on metal self-repairing technology in the embodiment of the present application. Figure 2 In the embodiment of the present application, an embodiment of a generator bearing degradation assessment system based on metal self-repairing technology includes:

[0108] The acquisition module is used to collect the vibration signal, dynamic balance parameters and temperature data of the generator bearing to obtain the time series of comprehensive parameters of the bearing operation status;

[0109] An extraction module, used to extract the balance characteristics and wear characteristics of the bearing according to the time series of the comprehensive parameters of the bearing operation state by using a multi-dimensional feature extraction algorithm to obtain a bearing degradation index system;

[0110] A repair module, used for injecting metal self-repairing material into the generator bearing according to the bearing degradation index system to obtain a repaired bearing;

[0111] A monitoring module, used to perform a dynamic balance test and work status monitoring on the repaired bearing to obtain a repair stage judgment result and balance improvement data;

[0112] A prediction module is used to predict and analyze the balance stability and degradation trend of the bearing through a long short-term memory network according to the judgment results of the repair stage and the balance improvement data, so as to obtain a bearing health status evaluation index;

[0113] A generation module is used to generate a bearing balance and degradation assessment report and optimize the metal self-repair strategy according to the bearing health status assessment index to obtain a prediction result of the remaining service life of the bearing.

[0114] Through the coordinated cooperation of the above-mentioned components, multi-dimensional parameter data of bearings are collected through vibration sensors, balance testers and temperature sensors to form a time series of comprehensive parameters of bearing operation status, providing a comprehensive data basis for subsequent analysis and avoiding the problem of incomplete diagnosis caused by single parameters of traditional monitoring methods; multi-dimensional feature extraction algorithms are used to deeply mine the balance characteristics and wear characteristics of bearings, build a bearing degradation index system, achieve accurate quantification of bearing health status, and provide a scientific basis for the optimization of metal self-repair material formulas; according to the bearing degradation index system, the metal self-repair material formula is customized and injected into the bearing, which solves the limitation that the traditional unified formula is not suitable for different damage characteristics and improves the repair efficiency; dynamic balancing tests and working status monitoring are carried out on the repaired bearings to obtain the judgment results of the repair stage and the balance improvement data, realize the whole process monitoring of the repair process, and can timely discover and adjust abnormal situations; long short-term memory networks are used to analyze the balance stability and degradation trend of bearings Predictive analysis fully utilizes the advantages of this deep learning algorithm in time series data processing and significantly improves the prediction accuracy. The algorithm can capture the complex nonlinear relationship in the long time series of bearing status parameters, and is particularly suitable for application scenarios in wind farm environments where bearing conditions are changeable and data noise is large. The contribution of the algorithm is reflected in the upgrade of traditional prediction methods based on experience or simple statistics to the level of intelligent prediction based on deep learning, making the prediction results closer to the actual degradation law; finally, the bearing balance and degradation assessment report is generated according to the bearing health status assessment index and the metal self-repair strategy is optimized. At the same time, the remaining service life of the bearing is predicted, which provides a scientific basis for wind farm operation and maintenance decisions, realizes the transformation from post-maintenance to predictive maintenance, and greatly reduces the unplanned downtime and maintenance costs of wind turbines caused by bearing failures, thereby improving the operational efficiency and economic benefits of wind farms; in addition, this method can achieve repair and evaluation without disassembling the bearing, reducing the downtime and maintenance time and ensuring the continuous and stable power generation of the wind farm.

[0115] Reference Figure 3 In an embodiment of the present invention, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface and a database connected through a system bus. Among them, the processor designed by the computer is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0116] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0117] An embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0118] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided by the present invention and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM.

[0119] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0121] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A generator bearing degradation assessment method based on metal self-repair technology, characterized in that: The generator bearing degradation assessment method based on metal self-repair technology includes: Collect the vibration signal, dynamic balance parameters and temperature data of the generator bearing to obtain the time series of comprehensive parameters of the bearing operation status; According to the time series of the comprehensive parameters of the bearing operation status, a multi-dimensional feature extraction algorithm is used to extract the balance characteristics and wear characteristics of the bearing, so as to obtain a bearing degradation index system; According to the bearing degradation index system, injecting metal self-repairing material into the generator bearing to obtain a repaired bearing; Performing dynamic balancing test and working status monitoring on the repaired bearing to obtain repair stage identification results and balance improvement data; According to the judgment results of the repair stage and the balance improvement data, the bearing balance stability and degradation trend are predicted and analyzed through a long short-term memory network to obtain a bearing health status assessment index; According to the bearing health status assessment index, a bearing balance and degradation assessment report is generated and the metal self-repair strategy is optimized to obtain a prediction result of the remaining service life of the bearing.

2. The generator bearing degradation assessment method based on metal self-repairing technology according to claim 1 is characterized in that: The vibration signal, dynamic balance parameter and temperature data of the generator bearing are collected to obtain a time series of comprehensive parameters of the bearing operation status, including: A triaxial acceleration sensor is installed on the wind turbine generator to collect bearing vibration signals. The sampling frequency is set to 10kHz to obtain the original bearing vibration data. The generator speed data and power output data are collected through the wind turbine main control system to obtain the generator operating parameters; The balance tester is used to collect dynamic balance data of the generator bearing, measure the unbalance amount and unbalance phase angle when the bearing rotates, and obtain the dynamic balance characteristic parameters of the bearing; The bearing seat temperature is continuously monitored by the embedded temperature sensor, and the ambient temperature is recorded at the same time to obtain the bearing temperature rise data; The original vibration data of the bearing, the operating parameters of the generator, the dynamic balance characteristic parameters of the bearing and the temperature rise data of the bearing are synchronously integrated according to the timestamp to form a time series database containing multi-dimensional parameters, and the time series of the comprehensive parameters of the bearing operating status is obtained.

3. The generator bearing degradation assessment method based on metal self-repairing technology according to claim 1 is characterized in that: According to the time series of the comprehensive parameters of the bearing operation status, a multi-dimensional feature extraction algorithm is used to extract the balance characteristics and wear characteristics of the bearing to obtain a bearing degradation index system, including: The vibration signal in the time series of the comprehensive parameters of the bearing operation state is decomposed at multiple scales by wavelet packet transform to obtain the vibration energy distribution characteristics; Extracting the mean, variance, peak factor and waveform factor of dynamic balancing parameters from the time series of the comprehensive parameters of the bearing operation status to obtain quantitative evaluation parameters of balance; Perform correlation analysis on the temperature data in the time series of the comprehensive parameters of the bearing operation status and the generator speed data, calculate the temperature rise rate per unit speed, and obtain the thermodynamic wear index; Performing envelope demodulation processing on the vibration energy distribution characteristics through Hilbert transform, extracting characteristic frequency energy of the bearing inner ring, outer ring, rolling element and cage, and obtaining a bearing fault characteristic spectrum; A bearing balance index is constructed according to the balance quantitative evaluation parameter, a bearing wear index is constructed according to the thermodynamic wear index and the bearing fault characteristic spectrum, and a bearing double degradation evaluation benchmark is obtained; The bearing balance index and the bearing wear index are normalized and weight coefficients are set, and a comprehensive evaluation standard is generated through weighted fusion to obtain the bearing degradation index system.

4. The generator bearing degradation assessment method based on metal self-repairing technology according to claim 1 is characterized in that: The method of injecting a metal self-repairing material into a generator bearing according to the bearing degradation index system to obtain a repaired bearing comprises: According to the bearing balance index and the bearing wear index in the bearing degradation index system, the formula ratio of the metal self-repairing material is determined to prepare nano-scale self-repairing particles containing hydroxy magnesium silicate, boron and tin components; The self-repairing particles are ground to a particle size of less than 100 nanometers by a ball milling process, and treated with a surface modifier to prepare a metal self-repairing material with high dispersibility; The metal self-repairing material is added to the extreme pressure lithium-based grease at a ratio of 5%, and mixed evenly by a high-speed stirring device to form a repairing grease complex; During the shutdown maintenance of the wind turbine, the oil filling port cover on the generator bearing seat is removed, and the repair grease compound is injected into the bearing cavity, and the injection amount is controlled to be 70% to 80% of the bearing cavity; The bearing is manually rotated several times to make the repair grease compound evenly distributed on the rolling element and raceway surface of the bearing to form an initial repair layer; The bearing seat is restored to its original state and the wind turbine is started, so that the generator runs at a low speed for four hours to form a repair transition period, and the repaired bearing is obtained.

5. The generator bearing degradation assessment method based on metal self-repairing technology according to claim 1 is characterized in that: The dynamic balancing test and working status monitoring of the repaired bearing are performed to obtain the repair stage identification result and balance improvement data, including: After the metal self-repairing material is injected into the bearing, the vibration signal is periodically collected, and the vibration root mean square value change rate is calculated through spectrum analysis to form a vibration attenuation curve; Use a balance tester to measure the unbalance of the bearing every four hours, record the data to form a balance change sequence, and obtain the dynamic balance improvement trend; The temperature sensor continuously monitors the difference between the bearing temperature and the ambient temperature, calculates the unit load temperature rise ratio, and generates a temperature rise curve; According to the vibration attenuation curve, dynamic balance improvement trend and temperature rise change curve, a three-dimensional feature space is constructed, and the data points are projected into the three-dimensional feature space to form a repair trajectory; Performing segmented fitting on the repair trajectory, identifying turning points in the trajectory, and dividing the repair process into a fine grinding stage, an adsorption cleaning stage, a repair layer generation stage, and a repair layer maintenance stage according to the turning points, to obtain the repair stage discrimination result; The imbalance difference, phase angle change and balance stabilization time before and after the repair are calculated, and the balance improvement data is obtained by combining the stable segment data in the dynamic balance improvement trend.

6. The generator bearing degradation assessment method based on metal self-repairing technology according to claim 1 is characterized in that: According to the judgment results of the repair stage and the balance improvement data, the bearing balance stability and degradation trend are predicted and analyzed through the long short-term memory network to obtain the bearing health status evaluation index, including: Extract the duration proportion and conversion rate of each stage from the repair stage discrimination result to construct a repair dynamic feature vector; Dividing the balance improvement data into time windows, calculating the balance change gradient and stability within each time window, and forming a balance stability time series feature; Performing time resampling on the repair dynamic feature vector and the balance stability time series feature by using a sliding window technique to generate a time series sample set for training; The training time series sample set is trained using a long short-term memory network to establish a bearing balance stability predictor, and future operating condition parameters are input to predict the bearing balance state change trend; According to the bearing balance state change trend and historical degradation data, a balance weight coefficient and a wear weight coefficient are calculated by a two-way weight allocation method to form a comprehensive evaluation weight matrix; The comprehensive evaluation weight matrix is ​​weighted multiplied by the current state parameter of the bearing, and the bearing health status assessment index is obtained after normalization.

7. The generator bearing degradation assessment method based on metal self-repairing technology according to claim 1 is characterized in that: The method generates a bearing balance and degradation assessment report based on the bearing health status assessment index and optimizes the metal self-repair strategy to obtain a bearing remaining service life prediction result, including: According to the bearing health status assessment index, a health status scoring standard is set, the index value is mapped to a scoring range of zero to one hundred points, and a bearing health status grading table is established; Extract bearing failure cases with similar degradation patterns from the historical database, extract the corresponding relationship between failure time points and health status assessment indexes, and construct a life mapping function; The current bearing health status assessment index is converted by the life mapping function, and the remaining theoretical service life of the bearing is calculated in combination with the operating condition correction parameter; Analyze the efficiency of the metal self-repairing material according to the judgment result of the repair stage, and adjust the ratio of magnesium hydroxysilicate, boron and tin in the metal self-repairing material formula in combination with the balance improvement data to form an optimized metal self-repairing strategy; Integrate the bearing health status score, balance improvement data, degradation trend prediction results and optimized metal self-repair strategy to generate a bearing balance and degradation assessment report; According to the theoretical value of the remaining service life and the change rate of the bearing health status score, a correction calculation is performed using the Weibull distribution fitting method to obtain the prediction result of the remaining service life of the bearing.

8. A generator bearing degradation assessment system based on metal self-repairing technology, used to implement the generator bearing degradation assessment method based on metal self-repairing technology as described in any one of claims 1 to 7, characterized in that: The generator bearing degradation assessment system based on metal self-repairing technology includes: The acquisition module is used to collect the vibration signal, dynamic balance parameters and temperature data of the generator bearing to obtain the time series of comprehensive parameters of the bearing operation status; An extraction module, used to extract the balance characteristics and wear characteristics of the bearing according to the time series of the comprehensive parameters of the bearing operation state by using a multi-dimensional feature extraction algorithm to obtain a bearing degradation index system; A repair module, used for injecting metal self-repairing material into the generator bearing according to the bearing degradation index system to obtain a repaired bearing; A monitoring module, used to perform a dynamic balance test and work status monitoring on the repaired bearing to obtain a repair stage judgment result and balance improvement data; A prediction module is used to predict and analyze the balance stability and degradation trend of the bearing through a long short-term memory network according to the judgment results of the repair stage and the balance improvement data, so as to obtain a bearing health status evaluation index; A generation module is used to generate a bearing balance and degradation assessment report and optimize the metal self-repair strategy according to the bearing health status assessment index to obtain a prediction result of the remaining service life of the bearing.

9. A computer device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program that can be run on the processor, and is characterized in that when the processor executes the computer program, the generator bearing degradation assessment method based on metal self-repair technology described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the processor is caused to execute the generator bearing degradation assessment method based on metal self-repairing technology as claimed in any one of claims 1 to 7.

Images