Double-fed wind turbine generator fault monitoring method and system based on frequency monitoring

Abstract

The invention relates to the technical field of fault monitoring of wind turbine generators, in particular to a doubly-fed wind turbine generator fault monitoring method and system based on frequency monitoring, and the method comprises the steps: obtaining the operation parameter data of a doubly-fed wind turbine generator, generating similar working condition time periods of the gear box in the doubly-fed wind turbine generator based on a clustering algorithm according to the operation parameter data; extracting target vibration data; according to the target vibration data, generating modal components and center frequencies corresponding to the modal components, and generating modal component energy; and generating a fusion modal energy value according to the modal component energy, and generating a unit fault early warning result. According to the method, only the vibration data in the time period corresponding to the similar working conditions are extracted, the interference of signal differences on the vibration characteristics under different working conditions is eliminated, the comprehensive criterion of the unit fault is carried out based on the fused modal energy value, the gearbox fault risk is early warned as accurately as possible, and the missing report is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of fault monitoring of wind turbines, and in particular to a method and system for fault monitoring of doubly-fed wind turbines based on frequency monitoring. Background Art

[0002] Doubly-fed wind turbines have become the mainstay of global onshore and offshore wind power generation thanks to their variable speed constant frequency operation capability, excellent power control performance, and high cost-performance ratio. They are currently one of the most widely used wind turbine models, especially in large-scale wind farms above the megawatt level. The gearbox is the most core and vulnerable component in a wind turbine, and its operating status directly affects the safety and operation and maintenance costs of the entire machine.

[0003] In the existing technology, gearbox fault monitoring generally adopts a method of directly extracting features based on full-time vibration data, or relies on traditional decomposition methods such as EMD and wavelet transform to extract characteristic frequency components related to the fault. Vibration data is often mixed in different working conditions. Direct full-time analysis or empirical segmentation is difficult to truly ensure the working condition consistency of the data. It is very easy to cause the same vibration feature to fail in judgment under different working conditions, resulting in false alarms. Traditional decomposition methods such as EMD and wavelet analysis signals are prone to low fault monitoring accuracy due to their susceptibility to modal aliasing.

[0004] Therefore, there is an urgent need to design a fault monitoring method and system for doubly fed wind turbines based on frequency monitoring. Summary of the Invention

[0005] Based on this, it is necessary to provide a method and system for monitoring the faults of doubly fed wind turbines based on frequency monitoring that can solve the problem of false alarms caused by the failure of judgment criteria for the same vibration feature under different working conditions in the existing technology, as well as the problem of low fault monitoring accuracy caused by the susceptibility of traditional signal decomposition methods to modal aliasing. The method can extract only vibration data of time periods corresponding to similar working conditions, eliminate the interference of signal differences under different working conditions on vibration characteristics, and perform comprehensive judgment of unit faults based on fused modal energy values ​​to achieve the most accurate possible early warning of gearbox failure risks and reduce missed reports.

[0006] The technical solutions of the present invention are as follows: A method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring, the method comprising: Acquiring operating parameter data of a doubly-fed wind turbine generator set, and generating similar operating condition time periods of a gearbox in the doubly-fed wind turbine generator set based on the operating parameter data using a clustering algorithm; extracting target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period; Generating modal components and center frequencies corresponding to the modal components according to the target vibration data, and generating modal component energies according to the modal components and the center frequencies corresponding to the modal components; A fused modal energy value is generated according to the modal component energy, and a unit fault warning result is generated according to the fused modal energy value.

[0007] Optionally, obtaining operating parameter data of the doubly-fed wind turbine generator set, and generating similar operating condition time periods of the gearbox in the doubly-fed wind turbine generator set based on the operating parameter data based on a clustering algorithm, includes: Acquiring operating parameter data of a doubly-fed wind turbine generator set, and generating a multi-dimensional operating condition characteristic vector according to the operating parameter data; Normalizing and assigning weights to the multi-dimensional operating condition feature vectors to generate weighted feature vectors; Segmenting the weighted feature vector according to preset segmentation intervals, and screening out target operating condition segments based on a clustering algorithm and preset screening conditions; A similar operating condition time period is generated according to the target operating condition section.

[0008] Optionally, the preset segment interval includes a wind speed interval and a pitch angle interval; The weighted feature vector is segmented according to a preset segmentation interval, and the target operating condition segment is screened out based on a clustering algorithm and preset screening conditions, including: The weighted feature vector is segmented according to the wind speed interval and the pitch angle interval, and the data in each segment are respectively clustered using a clustering algorithm to obtain a plurality of operating condition clusters; Filtering out target operating condition segments from each of the operating condition clusters according to preset filtering conditions; The center vector of each target operating condition segment is calculated, and the target operating condition segment is selected from the target operating condition segments according to the center vector.

[0009] Optionally, the vibration monitoring data includes first vibration data and second vibration data, the first vibration data is obtained based on a first vibration sensor, and the second vibration data is obtained based on a second vibration sensor; the first vibration sensor and the second vibration sensor are arranged opposite to each other on both sides of the input shaft bearing seat of the gearbox; Extracting target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period includes: acquiring first vibration data and second vibration data of the gearbox; generating differential vibration data based on the first vibration data and the second vibration data; Target vibration data is extracted from the differential vibration data according to the similar operating condition time period.

[0010] Optionally, generating modal components and center frequencies corresponding to the modal components according to the target vibration data, and generating modal component energies according to the modal components and the center frequencies corresponding to the modal components, includes: Decomposing the target vibration data based on a variational modal decomposition method, and generating a plurality of modal components and a center frequency corresponding to each modal component; Filtering out target modal components based on the modal components, center frequencies, and estimated main gear meshing frequencies; Modal component energy is generated according to the target modal component.

[0011] Optionally, generating a fused modal energy value according to the modal component energy, and generating a unit fault warning result according to the fused modal energy value, includes: generating a fused modal energy value according to the modal component energies; An energy growth rate and an energy fluctuation abnormality are generated according to the fused modal energy values ​​corresponding to each target vibration data; and a unit fault warning result is generated according to the energy growth rate and the energy fluctuation abnormality.

[0012] Optionally, the unit fault warning result includes a unit fault warning and a unit normal indication; Generating a unit fault warning result according to the energy growth rate and energy fluctuation abnormality includes: Determine whether the energy growth rate is greater than the healthy growth rate, and whether the energy fluctuation abnormality is greater than the healthy abnormality; if so, generate a unit fault warning; if not, generate a unit normal indication.

[0013] Optionally, a double-fed wind turbine fault monitoring system based on frequency monitoring is also provided, the system comprising: A similar operating condition generation module is used to obtain operating parameter data of the doubly-fed wind turbine generator set and generate similar operating condition time periods of the gearbox in the doubly-fed wind turbine generator set based on the operating parameter data based on a clustering algorithm; a vibration data screening module, configured to extract target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period; a modal energy generation module, configured to generate modal components and center frequencies corresponding to the modal components according to the target vibration data, and to generate modal component energy according to the modal components and the center frequencies corresponding to the modal components; The fault warning generation module is used to generate a fused modal energy value according to the modal component energy, and generate a unit fault warning result according to the fused modal energy value.

[0014] Optionally, the similar operating condition generation module is also used to: obtain operating parameter data of the doubly fed wind turbine generator set, and generate a multidimensional operating condition feature vector based on the operating parameter data; normalize and assign weights to the multidimensional operating condition feature vector, and generate a weighted feature vector; segment the weighted feature vector according to preset segmentation intervals, and filter out the target operating condition segment based on the clustering algorithm and preset filtering conditions; and generate a similar operating condition time period based on the target operating condition segment.

[0015] Optionally, the similar operating condition generation module is also used to: segment the weighted characteristic vector according to preset segmentation intervals, and filter out target operating condition segments based on a clustering algorithm and preset filtering conditions, including: segmenting the weighted characteristic vector according to the wind speed interval and the pitch angle interval, and performing fine-grained clustering on the data in each segment using a clustering algorithm to obtain multiple operating condition clusters; filtering out target operating condition segments from each of the operating condition clusters according to preset filtering conditions; calculating the center vector of each of the target operating condition segments, and filtering out the target operating condition segments from each of the target operating condition segments based on the center vector.

[0016] Optionally, the vibration monitoring data includes first vibration data and second vibration data, the first vibration data is obtained based on a first vibration sensor, and the second vibration data is obtained based on a second vibration sensor; the first vibration sensor and the second vibration sensor are arranged opposite to each other on both sides of the input shaft bearing seat of the gearbox; the vibration data screening module is also used to: obtain the first vibration data and the second vibration data of the gearbox; generate differential vibration data based on the first vibration data and the second vibration data; and extract target vibration data from the differential vibration data based on the similar operating time period.

[0017] Optionally, the modal energy generation module is also used to: decompose the target vibration data based on the variational modal decomposition method, and generate multiple modal components and the center frequency corresponding to each modal component; filter out the target modal component according to each of the modal components, the center frequency and the estimated main gear meshing frequency; and generate modal component energy according to the target modal component.

[0018] Optionally, the fault warning generation module is also used to: generate a fused modal energy value based on the modal component energy; generate an energy growth rate and an energy fluctuation abnormality based on the fused modal energy value corresponding to each target vibration data, and generate a unit fault warning result based on the energy growth rate and the energy fluctuation abnormality.

[0019] Optionally, the unit fault warning result includes a unit fault warning indication and a unit non-fault indication; the fault warning generation module is also used to: determine whether the energy growth rate is greater than the healthy growth rate, and the energy fluctuation abnormality is greater than the healthy abnormality; if the judgment is yes, generate a unit fault warning; if the judgment is no, generate a unit normal indication.

[0020] Optionally, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring are implemented.

[0021] Optionally, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring are implemented.

[0022] The present invention achieves the following technical effects: The above-mentioned frequency-monitoring-based doubly-fed wind turbine fault monitoring method and system obtains the operating parameter data of the doubly-fed wind turbine, and generates a similar operating condition time period of the gearbox in the doubly-fed wind turbine based on the operating parameter data based on a clustering algorithm; extracts target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period; generates modal components and the center frequencies corresponding to each modal component according to the target vibration data, and generates modal component energy according to the modal components and the center frequencies corresponding to each modal component; generates a fused modal energy value according to the modal component energy, and generates a unit fault warning result according to the fused modal energy value. In order to provide basic data for subsequent analysis of the gearbox, this application realizes frequency-based operating status analysis under the same operating conditions, thereby improving the accuracy of subsequent fault monitoring, fully considering the different effects of different operating conditions such as wind speed, rotational speed, and power on the vibration health of the gearbox, thereby obtaining the operating parameter data of the doubly fed wind turbine generator set, and generating similar operating condition time periods of the gearbox in the doubly fed wind turbine generator set based on the operating parameter data based on the clustering algorithm, thereby accurately identifying the time periods when the gearbox is in similar operating conditions, effectively avoiding the interference of different operating conditions on the vibration data characteristics, and improving the reliability and Targetedly, the target vibration data is then extracted from the vibration monitoring data of the gearbox according to the similar working condition time period. By only extracting the vibration data of the time period corresponding to the similar working condition, the interference of the signal difference under different working conditions on the vibration characteristics is eliminated. Then, the modal components and the center frequencies corresponding to each modal component are generated according to the target vibration data, and the modal component energy is generated according to the modal components and the center frequencies corresponding to each modal component, so as to separate the physical characteristic modal components with clear center frequencies, and calculate the energy for each mode, accurately extract the gear meshing frequency and its harmonics, and effectively avoid frequency aliasing or information loss. Finally, a fused modal energy value is generated according to the modal component energy, and a unit fault warning result is generated according to the fused modal energy value. By making a comprehensive judgment on the unit fault based on the fused modal energy value, the gearbox fault risk can be warned as accurately as possible, the missed reports can be reduced, and the timeliness of the unit fault monitoring and operation and maintenance response can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 1 is a flow chart of a method for monitoring a fault of a doubly-fed wind turbine generator system based on frequency monitoring in one embodiment; Figure 2 A schematic diagram of a process for generating a similar operating condition time period in one embodiment; Figure 3 A schematic diagram of a process for screening a target operating condition section in one embodiment; Figure 4 A schematic diagram of a process for extracting target vibration data in one embodiment; Figure 5A schematic diagram of a process for generating modal component energy in one embodiment; Figure 6 A schematic diagram of a process for generating a unit fault warning result in one embodiment; Figure 7 A schematic diagram of a process for generating a normal indication of a unit in one embodiment; Figure 8 FIG. 4 is a structural block diagram of a doubly-fed wind turbine fault monitoring system based on frequency monitoring in one embodiment. DETAILED DESCRIPTION

[0024] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0025] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0026] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0027] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0028] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0029] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0030] In one embodiment, a terminal is provided, which is used to: obtain operating parameter data of a doubly fed wind turbine generator set, generate a similar operating condition time period of a gearbox in the doubly fed wind turbine generator set based on the operating parameter data based on a clustering algorithm; extract target vibration data from vibration monitoring data of the gearbox based on the similar operating condition time period; generate modal components and center frequencies corresponding to each modal component based on the target vibration data, and generate modal component energy based on the modal components and the center frequencies corresponding to each modal component; generate a fused modal energy value based on the modal component energy, and generate a unit fault warning result based on the fused modal energy value.

[0031] The terminal may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices.

[0032] In one embodiment, Figure 1 As shown, a method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring is provided, the method comprising: Step S100: acquiring operating parameter data of a doubly-fed wind turbine generator set, and generating similar operating condition time periods of a gearbox in the doubly-fed wind turbine generator set based on the operating parameter data using a clustering algorithm; Step S200: extracting target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period; Step S300: generating modal components and center frequencies corresponding to the modal components according to the target vibration data, and generating modal component energies according to the modal components and the center frequencies corresponding to the modal components; Step S400: generating a fused modal energy value according to the modal component energy, and generating a unit fault warning result according to the fused modal energy value.

[0033] In this embodiment, in order to provide basic data for subsequent analysis of the gearbox, frequency-based operating status analysis is implemented under the same operating conditions, thereby improving the accuracy of subsequent fault monitoring. Fully consider the different effects of different operating conditions such as wind speed, rotational speed, and power on the vibration health of the gearbox, thereby obtaining the operating parameter data of the doubly fed wind turbine generator set, and generating similar operating condition time periods of the gearbox in the doubly fed wind turbine generator set based on the operating parameter data based on a clustering algorithm, thereby accurately identifying the time period when the gearbox is in similar operating conditions, effectively avoiding the interference of different operating conditions on the vibration data characteristics, and improving the reliability of subsequent feature extraction and abnormality judgment. The method is precise and targeted, and then the target vibration data is extracted from the vibration monitoring data of the gearbox according to the similar working condition time period. By only extracting the vibration data of the time period corresponding to the similar working condition, the interference of the signal difference under different working conditions on the vibration characteristics is eliminated. Then, the modal components and the center frequencies corresponding to each modal component are generated according to the target vibration data, and the modal component energy is generated according to the modal components and the center frequencies corresponding to each modal component, so as to separate the physical characteristic modal components with clear center frequencies, and calculate the energy for each mode, accurately extract the gear meshing frequency and its harmonics, and effectively avoid frequency aliasing or information loss. Finally, a fused modal energy value is generated according to the modal component energy, and a unit fault warning result is generated according to the fused modal energy value. By performing a comprehensive judgment of the unit fault based on the fused modal energy value, the gearbox fault risk can be warned as accurately as possible, the missed reports can be reduced, and the timeliness of the unit fault monitoring and operation response can be improved.

[0034] In one embodiment, Figure 2 As shown, step S100: obtaining operating parameter data of the doubly-fed wind turbine generator set, and generating similar operating condition time periods of the gearbox in the doubly-fed wind turbine generator set based on the operating parameter data based on a clustering algorithm, including: Step S110: acquiring operating parameter data of the doubly-fed wind turbine generator system, and generating a multi-dimensional operating condition characteristic vector according to the operating parameter data; Step S120: normalizing and assigning weights to the multi-dimensional operating condition feature vector to generate a weighted feature vector; Step S130: segmenting the weighted feature vector according to preset segmentation intervals, and filtering out target operating condition segments based on a clustering algorithm and preset screening conditions; Step S140: generating a similar operating condition time period according to the target operating condition section.

[0035] In this embodiment, in step S110, operating parameter data of the doubly-fed wind turbine generator system is acquired. The operating parameter data includes raw operating parameter data such as wind speed, main shaft speed, generator speed, active power, and pitch angle. Based on the raw operating parameters, the transmission ratio and main shaft acceleration are calculated to form a multi-dimensional operating condition feature vector. The transmission ratio reflects the actual speed change characteristics and transmission state of the gearbox and is obtained by dividing the generator speed by the main shaft speed. The main shaft acceleration is obtained by calculating the differential of the main shaft speed and is used to capture abnormalities such as sudden operating condition changes and dynamic responses.

[0036] In step S120, the operating condition feature vector is normalized to have a mean of 0 and a variance of 1 using z-score or min-max normalization. Principal component analysis or correlation analysis is then used, combined with historical vibration response data, to assign a weight to each feature to obtain a weighted feature vector.

[0037] Then, the weighted feature vector is segmented according to the preset segmentation intervals, and the target operating condition segment is screened out based on the clustering algorithm and the preset screening conditions, and the time period corresponding to the final operating condition segment is extracted and a similar operating condition time period is generated.

[0038] For example, the multi-dimensional operating condition feature vector is expressed as ,in, They represent wind speed, main shaft speed, generator speed, active power, pitch angle, transmission ratio and main shaft acceleration respectively.

[0039] The weighted feature vector is expressed as . In the vector to They represent the weights of wind speed, main shaft speed, generator speed, active power, pitch angle, transmission ratio and main shaft acceleration respectively. to They are 0.2, 0.2, 0.25, 0.15, 0.05, 0.1 and 0.05 respectively.

[0040] In this embodiment, not only wind speed, main shaft speed, generator speed, active power, and pitch angle are collected, but also the transmission ratio and main shaft acceleration are calculated to capture more detailed working condition differences. Through normalization and weight configuration, the clustering results are made more in line with the actual gearbox. The weights can also be adjusted to improve flexibility and adaptability to different gearboxes.

[0041] It should be noted that the examples of the above-mentioned multi-dimensional operating condition characteristic vectors and weighted characteristic vectors are only for illustrating the technical solution of the present application, and are not limiting. In practice, different numbers of parameters can be monitored based on the resources that can be allocated by the managers of the doubly fed wind turbine generator set. This application does not impose any limitations on this.

[0042] In another embodiment, in step S120, a method of assigning a weight to each feature by using a correlation analysis method combined with historical vibration response data is exemplarily described: First, collect historical operating condition characteristic data and historical vibration response data within a preset historical period. The historical period can be set to one month, six months, or longer, and the more data collected, the better. Historical operating condition characteristic data includes wind speed, main shaft speed, generator speed, active power, and pitch angle. Historical vibration response data includes vibration energy, RMS vibration, and main meshing frequency component energy. When collecting data, a one-to-one correspondence between operating condition parameters and vibration response characteristics at the same time is required to facilitate subsequent sequence operations.

[0043] Then, outliers in the operating parameters or vibration characteristics that are invalid or outside the reasonable physical range are removed to clean up the anomalies. This can be done through interpolation, forward / backward filling, or simply discarding missing samples to handle missing values. The system then ensures that all parameter timestamps are strictly aligned to ensure the integrity of each piece of data. The processed historical operating characteristic data and historical vibration response data are then normalized.

[0044] Next, for each operating parameter, a correlation analysis method is used to calculate the correlation coefficient between it and the target vibration response characteristic. The correlation coefficient is calculated according to the Pearson correlation coefficient to obtain the correlation coefficient between each operating parameter and the target vibration response characteristic. The correlation coefficient is then normalized after taking the absolute value to obtain the weight of each operating parameter. The calculation formula of the Pearson correlation coefficient is prior art and should be known to those skilled in the art, so it will not be elaborated in this application. The formula for normalizing the correlation coefficient after taking the absolute value is as follows: ,in, is the weight of each working condition parameter, is the correlation coefficient between the i-th operating condition parameter and the vibration response characteristics. M is the total number of operating condition parameters. It represents the sum of the absolute values ​​of the correlation coefficients between all operating parameters and vibration response characteristics, and x is the index number of the operating parameter.

[0045] After obtaining the weights of the various operating condition parameters, the weights can be assigned to the normalized operating condition feature vectors that are subsequently actually detected to obtain a weighted feature vector.

[0046] Therefore, by setting the weights of the operating condition parameters, we fully consider the different effects of different operating condition characteristics such as wind speed, rotational speed, and power on the vibration health of the gearbox. This allows the characteristics with different influences to play a greater role in the subsequent operating condition similarity calculation and cluster analysis, making the clustering results more physically realistic and discriminative, avoiding the role of irrelevant or weakly correlated characteristics influencing the role of truly important characteristics, and improving the accuracy of the operating condition segmentation. In addition, through weight distribution, the subsequent cluster comparison analysis will pay more attention to those characteristics that are most sensitive to vibration, which can detect early anomalies and trend changes in the gearbox earlier and more accurately, increase the focus on key features, and through the combination of weight distribution and normalization, eliminate the impact of dimensionality while further eliminating the impact of large numerical features on distance or similarity through weight distribution.

[0047] In one embodiment, Figure 3 As shown, the preset segment interval includes a wind speed interval and a pitch angle interval; Step S130: Segmenting the weighted feature vector according to preset segmentation intervals, and filtering out target operating condition segments based on a clustering algorithm and preset screening conditions, including: Step S131: segmenting the weighted feature vector according to the wind speed interval and the pitch angle interval, and performing fine-grained clustering on the data in each segment using a clustering algorithm to obtain multiple operating condition clusters; Step S132: Filtering out target operating condition segments from each of the operating condition clusters according to preset filtering conditions; Step S133: Calculate the center vector of each target operating condition segment, and select a target operating condition segment from each target operating condition segment according to the center vector.

[0048] In this embodiment, in step S131, the wind speed interval and the pitch angle interval are both pre-set. For example, all data are divided into multiple segmented intervals according to the wind speed interval and the pitch angle interval. The wind speed interval segmentation includes but is not limited to: 0-5m / s, 5-10m / s, 10-20m / s, and the pitch angle interval includes but is not limited to: 0-10°, 10-20°, 20-90°. The data in each segment is used as a subset. Then, a clustering algorithm such as K-means, Gaussian Mixture Model or density clustering DBSCAN is used, the weighted feature vector is input, the Euclidean distance between data points is calculated, the optimal number of clusters is automatically selected, and the data is divided into several working condition clusters.

[0049] In step S132, when screening the target operating condition segment, the operating condition segments that are continuous in time and whose duration meets the set threshold are screened out from the samples in the operating condition cluster, and non-continuous and isolated sample points are eliminated. That is, the preset screening condition is that the operating condition segments are continuous in time and whose duration meets the set threshold. The set threshold is pre-set, such as 5 minutes.

[0050] In step S133, the Euclidean distance between the characteristic vector of the target operating condition segment and the center vector of the cluster to which it belongs is calculated, and several consecutive sampling moments with the smallest distance are selected as the representative operating condition segment of the operating condition cluster, that is, the final operating condition segment.

[0051] Finally, the final operating condition section is set as the target operating condition section, thereby improving the clustering accuracy by first setting the preset segmentation intervals of the wind speed interval and the pitch angle interval, and then clustering the features.

[0052] Furthermore, the wind speed range is set primarily because wind speed is the primary indicator of wind energy input intensity, directly affecting the load and unit dynamic response, and therefore the gearbox state. The pitch angle range is set because the pitch angle is the primary control parameter of the wind turbine control strategy, directly affecting blade stress and power generation.

[0053] When setting the wind speed interval, the fixed interval method can be used, specifically, the partitioning is directly performed according to the above-mentioned 0-5m / s, 5-10m / s, and 10-20m / s. Alternatively, the adaptive quantile method can be used to segment the wind speed data according to the 25%, 50%, and 75% quantiles of the data distribution. This method is suitable for wind farms with highly uneven wind speed distribution. For the setting of the pitch angle interval, it can also be segmented according to the actual adjustment points, such as the blade adjustment control has discrete angles, or subdivided according to the interval with frequent pitch action. Of course, the segmentation thresholds of the wind speed interval and the pitch angle interval can also be dynamically adjusted according to the actual wind turbine operation distribution, or set by operation and maintenance experience, so as to adapt to different wind farms and models.

[0054] Based on the settings of wind speed range and pitch angle range, each data point is classified according to the range of wind speed and pitch angle to form a two-dimensional segmented grid of wind speed and pitch angle. For example, wind speed is 0-5m / s and pitch angle is 0-10°, wind speed is 5-10m / s and pitch angle is 10-20°, etc. Each combination is a segmented unit.

[0055] For example, if the wind speed range is 0-14 m / s and the pitch angle range is 0-16°, the wind speed intervals can be set to 0-4, 4-8, 8-12, and 12-16 m / s, and the pitch angle intervals can be set to 0-6, 6-12, and 12-18°. This creates a total of 4x3 segments, for a total of 12 segments. If the wind speed is 5.6 m / s and the pitch angle is 8.4°, it should be classified into the segment [4-8, 6-12].

[0056] Furthermore, the steps of using clustering algorithm to perform fine-grained clustering are explained by taking the wind speed of 4-8m / s and pitch of 6-12° as an example. 2500 data subsets are extracted from this segment, and K-means clustering is performed on the weighted feature vectors of these 2500 data. Different cluster numbers P are tried respectively. k Value, calculate the silhouette coefficient of each cluster, and select the P corresponding to the silhouette coefficient with the highest value k If you try P k =2, 3, 4, 5, calculate the silhouette coefficient of each cluster, assuming P k =3 when the silhouette coefficient is the highest, so P is selected k =3, divided into 3 working condition clusters. Then get the cluster label of each sample. For example, 1500 belong to cluster 1, 600 belong to cluster 2, and 400 belong to cluster 3. It should be noted that in the K-means clustering method, the number of clusters is generally represented by K. In order to distinguish it from the modal component K in the variational mode decomposition method below, the cluster number P is used in this embodiment. k express.

[0057] Next, the target operating condition segment is filtered from each of the aforementioned operating condition clusters according to the preset filtering criteria. Taking Cluster 1, which contains 1500 data points, as an example, examining the timestamps reveals a continuous period of 480 samples every 5 seconds between 8:30 and 9:10. The remaining data is scattered, such as a few minutes, which does not meet the minimum continuous duration requirement of 10 minutes. Therefore, only the 40-minute period between 8:30 and 9:10 is retained as the valid operating condition segment, and all other isolated points are removed, thus filtering out the target operating condition segment.

[0058] Then, the Euclidean distance between each data and the central eigenvector of cluster 1 is calculated in the continuous segment of 08:30-09:10, and the data are sorted from small to large according to the distance. The 100 continuous data with the smallest distance are selected as the representative operating condition segment of cluster 1, and the representative operating condition segments in other clusters are screened at the same time. That is, the target operating condition segment screened out is the representative operating condition segment corresponding to each cluster, and the representative operating condition segments corresponding to a cluster are similar working states.

[0059] Furthermore, in step S140 , the generated similar operating condition time period is also a time period corresponding to each representative operating condition section corresponding to a cluster, so as to ensure that the working states corresponding to the similar operating condition time periods are the same.

[0060] Therefore, by acquiring the operating parameter data of the doubly-fed wind turbine, and then performing normalization, weight assignment, segmentation, clustering, and filtering, we can obtain the time periods corresponding to similar gearbox operating states under specific circumstances. This provides basic data for subsequent gearbox analysis, enabling frequency-based operating state analysis under the same operating conditions, thereby improving the accuracy of subsequent fault monitoring. Specific circumstances include those when wind speed, rotational speed, load, pitch angle, etc. are approximately constant.

[0061] In one embodiment, the vibration monitoring data includes first vibration data and second vibration data, the first vibration data being acquired based on a first vibration sensor, and the second vibration data being acquired based on a second vibration sensor; the first vibration sensor and the second vibration sensor being disposed opposite each other on opposite sides of the input shaft bearing seat of the gearbox. Exemplarily, the first vibration sensor and the second vibration sensor are mounted symmetrically along the axis of the input shaft, with the first vibration sensor being 100 mm from the centerline of the bearing seat, and the second vibration sensor being 100 mm from the centerline of the bearing seat. The mounting surface is located on the outer circumference of the bearing seat to ensure close contact with the surface of the input shaft bearing seat, thereby improving vibration transmission quality.

[0062] Furthermore, the first vibration sensor and the second vibration sensor adopt a three-axis acceleration sensor, with the axis of the gearbox input shaft as the reference. The X axis of the first vibration sensor and the second vibration sensor is parallel to the axis to ensure efficient collection of axial vibration. The Y and Z axes of the first vibration sensor and the second vibration sensor are clearly orthogonal and installed in a plane perpendicular to the axis, which can accurately capture the radial and tangential vibration characteristics generated by the gear meshing point. During installation, an installation positioning fixture is used to assist installation, and the angle is calibrated using a level and a laser rangefinder. The angle deviation is controlled within plus or minus 2 degrees to ensure accuracy. During installation, a dedicated magnetic base or bolt fastening method is used to install the sensor to prevent the adhesive from loosening due to long-term aging and affecting the signal quality.

[0063] like Figure 4 As shown, step S200: extracting target vibration data from the vibration monitoring data of the gearbox according to the similar working condition time period, including: Step S210: Acquire first vibration data and second vibration data of the gearbox; Step S220: generating differential vibration data according to the first vibration data and the second vibration data; Step S230: extracting target vibration data from the differential vibration data according to the similar operating condition time period.

[0064] In this embodiment, in the gearbox, the early signal of gear wear is weak, the vibration energy is low, and it is often masked by environmental vibration or noise. By placing the first vibration sensor and the second vibration sensor on both sides of the input shaft bearing seat of the gearbox and generating differential vibration data based on the first vibration data and the second vibration data, the common-mode vibration noise interference from the wind turbine tower or environmental vibration is effectively suppressed, thereby highlighting the weak vibration signal of the local gear meshing point, improving the sensor's sensitivity to early gear wear failures, and effectively extracting the vibration characteristics of the near-field vibration generated in the gear wear area. The far-field common-mode interference is weakened and the signal-to-noise ratio is significantly improved, which solves the problem in the existing technology that single sensor monitoring is easily mixed with a large amount of background interference signals, resulting in spectrum confusion and increased difficulty in identifying characteristic frequencies.

[0065] In addition, the input shaft of the gearbox usually bears a large load, and the wear of the gear meshing is mostly concentrated in this area. The bearing seat directly supports the bearing, and the vibration signal transmission path is shortest and the loss is minimal. Therefore, the first vibration sensor and the second vibration sensor are arranged opposite to each other on both sides of the bearing seat of the input shaft of the gearbox, so that the measurement point position is aligned with the position with a high probability of fault occurrence under actual operating conditions, thereby providing reliable data support through the setting of the sensor installation position.

[0066] Differential vibration data is generated from the first vibration data and the second vibration data based on the following formula: ,in, is the differential vibration data, is the first vibration data, The second vibration data is obtained by differential processing, where sensors located close to each other capture roughly the same far-field interference signals. Subtracting the two effectively cancels them out, thus highlighting the local vibration characteristics.

[0067] After acquiring the differential vibration data, target vibration data is extracted from the differential vibration data according to the similar operating condition time period, so as to perform subsequent fault monitoring and analysis based on the target vibration data.

[0068] In one embodiment, after a fixed maintenance period, the common-mode rejection effectiveness of the first and second vibration sensors is tested, and the test results are used to determine whether to adjust their positions. The fixed maintenance period is a pre-set period, such as one month. This fixed maintenance period eliminates the need for continuous testing, saving computing resources.

[0069] First, the first vibration data and the second vibration data Remove the DC component, trend term, and acquisition noise, then perform fast Fourier transform to obtain the complex amplitude and . and Respectively, in frequency Then calculate the amplitude of the differential signal ,pass Then calculate the common mode rejection ratio:

[0070] Where, Common mode rejection ratio, which is used to express the frequency Under this condition, the energy ratio of the differential signal to the common mode signal is The higher it is, the more effective it is in suppressing common-mode interference.

[0071] By comparing the common-mode rejection threshold with the preset common-mode rejection threshold, it is determined whether it is less than the common-mode rejection threshold. If it is less than the common-mode rejection threshold, it indicates that the common-mode rejection effect has deteriorated, and a self-test / maintenance reminder is triggered. The common-mode rejection threshold is set in advance by those skilled in the art and is not limited or exemplified without application.

[0072] Therefore, in this embodiment, the evaluation and self-calibration steps are set to ensure that the fault warning is not interfered with by non-equipment problems such as the environment, structural conduction, installation changes, etc., thereby greatly reducing false alarms.

[0073] In one embodiment, Figure 5 As shown, step S300: generating modal components and center frequencies corresponding to the modal components according to the target vibration data, and generating modal component energy according to the modal components and the center frequencies corresponding to the modal components, including: Step S310: decomposing the target vibration data based on a variational modal decomposition method, and generating a plurality of modal components and a center frequency corresponding to each modal component; Step S320: Filtering out target modal components according to the modal components, center frequencies, and estimated main gear meshing frequencies; Step S330: Generate modal component energy according to the target modal component.

[0074] In this embodiment, in step S310, the target vibration data Perform variational modal decomposition (VMD), set the VMD decomposition mode number to K, decompose the target vibration data into K modal components, and obtain the center frequency f of each modal component. Variational modal decomposition is an adaptive signal decomposition technique that can adaptively decompose non-stationary complex signals into a set of modal components with different center frequencies. It is a mature technology in the prior art, so the formula will not be elaborated in this application.

[0075] In step S320, the main gear meshing frequency is estimated based on the gearbox structural parameters and operating conditions. and its several harmonics , and filter out the center frequency and and its several harmonics The closest modal component is set as the target modal component. The gear speed is first monitored, and then the main gear meshing frequency is estimated based on the product of the speed and the number of teeth. The multi-order harmonic frequencies are generally ,in, is the harmonic order ( =1, 2, 3...).

[0076] Therefore, using VMD, the characteristic components of the gear meshing frequency and its harmonic frequencies can be accurately extracted, effectively avoiding the modal aliasing problem and solving the problems of low stability and poor accuracy caused by the use of traditional EMD or wavelet analysis in existing technologies.

[0077] Next, in step S330, each filtered target modal component is subjected to the following formula: Calculating modal component energy : , where [t0, t1] is the analysis time window, k represents the index of the target modal component, represents the modal component energy of the kth target modal component.

[0078] Furthermore, each of the similar working condition time periods corresponds to one target vibration data, and each of the target vibration data corresponds to multiple modal component energies. .

[0079] In one embodiment, Figure 6 As shown, step S400: generating a fused modal energy value according to the modal component energy, and generating a unit fault warning result according to the fused modal energy value, specifically includes: Step S410: Based on the following formula, the energy of multiple modal components corresponding to the target vibration data is calculated. Generate fusion modal energy value : ,in, is the fusion modal energy value, is the number of target modal components, where one target vibration data corresponds to one fusion modal energy value .

[0080] Step S420: generating an energy growth rate and an energy fluctuation abnormality according to the fusion modal energy value corresponding to each target vibration data; Step S430: generating a unit fault warning result according to the energy growth rate and the energy fluctuation abnormality.

[0081] In this embodiment, in step S420, the energy growth rate represents the growth trend of the fusion value and is generated based on the following formula: ; Where, is the energy growth rate, is the index of the fused modal energy value, is the total number of fused modal energy values, is the mean of the index of the fused modal energy value, For the The fused modal energy value, is the mean value of the fused modal energy value. The calculation formula is .calculate The role of is to decentralize and ensure the numerical stability of trend fitting.

[0082] The energy growth rate is a statistical characteristic of the gearbox's overall operating status. It's insensitive to single anomalies or occasional signal fluctuations, accurately capturing risk signals of gradual anomalies within large amounts of data, reducing false alarm rates. Many mechanical failures show a steady increase in key characteristics before they occur, but they don't necessarily reach the alarm threshold. The energy growth rate can detect these subtle but continuous changes in advance, providing early warning and securing maintenance windows. This is particularly useful for gradual hazards such as bearing or gear wear, increased clearance, and lubrication degradation.

[0083] The energy fluctuation anomaly degree is generated based on the following formula: , where is the energy fluctuation anomaly corresponding to the g-th fusion modal energy value, for The standard deviation of the fused modal energy values. The energy fluctuation anomaly is used to measure the degree of outliers relative to the global mean and fluctuation range. It can quickly monitor acute faults such as sudden impacts, instability, and component breakage. When the trend is steadily increasing, sudden and sharp deviations from the mean at individual points can also serve as supplementary evidence for secondary judgments, achieving highly sensitive detection of sudden events in local time periods.

[0084] Therefore, using the growth trend slope for anomaly screening can effectively monitor the continuous changes in equipment status, and has the advantages of robustness and early warning. Using energy fluctuation anomalies to screen local anomalies can keenly capture sudden and extreme state changes and achieve accurate positioning of single-point risks. Through the combination of the two, it combines the advantages of trend and local health monitoring, providing all-round, multi-level anomaly diagnosis capabilities for the gearbox in the doubly fed wind turbine.

[0085] In one embodiment, Figure 7 As shown, step S430: generating a unit fault warning result according to the energy growth rate and energy fluctuation abnormality, including: Step S431: determining whether the energy growth rate is greater than the healthy growth rate, or whether the energy fluctuation abnormality is greater than the healthy abnormality; Step S432: If the answer is yes, generate a unit failure warning; Step S433: If the judgment is negative, a unit normal indication is generated.

[0086] In this embodiment, the unit fault warning includes three situations: a first routine warning, a second routine warning, and a highest confidence warning. In step S432, if the judgment is yes, there are several situations. The first is that the energy growth rate is greater than the healthy growth rate, but the energy fluctuation abnormality is less than or equal to the healthy abnormality; in this case, a first routine warning is generated. The second is that the energy growth rate is less than or equal to the healthy growth rate, and the energy fluctuation abnormality is greater than the healthy abnormality; in this case, a second routine warning is generated. The third is that the energy growth rate is greater than the healthy growth rate, and the energy fluctuation abnormality is greater than the healthy abnormality; in this case, a highest confidence warning is generated.

[0087] During the first routine warning, it indicates that the health status of the gearbox has continued to deteriorate over multiple consecutive time periods, which usually indicates progressive faults such as wear and failure. At this time, check the energy change curves of each major modal component, such as the main frequency and harmonics, to eliminate the overall trend deviation caused by abnormal individual components. In addition, check whether there are any systematic changes in operating conditions such as load, temperature, wind speed, etc. during this period to eliminate false trends caused by changes in operating conditions.

[0088] The second regular warning indicates that the gearbox has experienced a sudden or extreme abnormality during a specific time period. This could be due to a sudden failure, shock, misoperation, or an extreme operating condition. At this point, the original differential vibration waveform and time / frequency domain data are examined for strong shocks, spikes, or abnormal noise. Detailed spectrum analysis can also be performed to detect the presence of new frequency components, broadband noise, or enhanced characteristic frequencies to determine whether this is a true equipment failure or an occasional interference event.

[0089] When the highest confidence warning is issued, both trend anomaly and single-point fluctuation anomaly are met, and a comprehensive inspection is performed at this time.

[0090] In step S433, if the judgment is no, that is, it is judged that the energy growth rate is less than or equal to the healthy growth rate, and the energy fluctuation abnormality is less than or equal to the healthy abnormality, then a normal unit indication is generated.

[0091] In this embodiment, the health growth rate and the health abnormality are pre-set based on the data of the gearbox corresponding to the historical health period. The historical health period refers to a historical time interval in which the equipment has been confirmed to be normal and in good operating condition. The data during this period can be regarded as a health baseline for statistical comparison of subsequent trends and fluctuation anomalies. When the sequence is calculated as above The energy growth rate corresponding to the historical health period was calculated by the method of and standard deviation Finally passed The health growth rate is calculated. The health abnormality is generally set to 2.

[0092] In one embodiment, Figure 8 As shown, a double-fed wind turbine fault monitoring system based on frequency monitoring is also provided, and the system includes: A similar operating condition generation module is used to obtain operating parameter data of the doubly-fed wind turbine generator set and generate similar operating condition time periods of the gearbox in the doubly-fed wind turbine generator set based on the operating parameter data based on a clustering algorithm; a vibration data screening module, configured to extract target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period; a modal energy generation module, configured to generate modal components and center frequencies corresponding to the modal components according to the target vibration data, and to generate modal component energy according to the modal components and the center frequencies corresponding to the modal components; The fault warning generation module is used to generate a fused modal energy value according to the modal component energy, and generate a unit fault warning result according to the fused modal energy value.

[0093] In one embodiment, the similar operating condition generation module is also used to: obtain operating parameter data of the doubly fed wind turbine generator set, and generate a multidimensional operating condition feature vector based on the operating parameter data; normalize and assign weights to the multidimensional operating condition feature vector, and generate a weighted feature vector; segment the weighted feature vector according to a preset segmentation interval, and filter out a target operating condition segment based on a clustering algorithm and preset filtering conditions; and generate a similar operating condition time period based on the target operating condition segment.

[0094] In one embodiment, the similar operating condition generation module is also used to: segment the weighted characteristic vector according to preset segmentation intervals, and filter out target operating condition segments based on a clustering algorithm and preset filtering conditions, including: segmenting the weighted characteristic vector according to the wind speed interval and the pitch angle interval, and performing fine-grained clustering on the data in each segment using a clustering algorithm to obtain multiple operating condition clusters; filtering out target operating condition segments from each of the operating condition clusters according to preset filtering conditions; calculating the center vector of each of the target operating condition segments, and filtering out the target operating condition segments from each of the target operating condition segments based on the center vector.

[0095] In one embodiment, the vibration monitoring data includes first vibration data and second vibration data, the first vibration data is obtained based on a first vibration sensor, and the second vibration data is obtained based on a second vibration sensor; the first vibration sensor and the second vibration sensor are arranged opposite to each other on both sides of the input shaft bearing seat of the gearbox; the vibration data screening module is also used to: obtain the first vibration data and the second vibration data of the gearbox; generate differential vibration data based on the first vibration data and the second vibration data; and extract target vibration data from the differential vibration data based on the similar operating time period.

[0096] In one embodiment, the modal energy generation module is further used to: decompose the target vibration data based on the variational modal decomposition method, and generate multiple modal components and the center frequency corresponding to each modal component; filter out the target modal component according to each of the modal components, the center frequency and the estimated main gear meshing frequency; and generate modal component energy according to the target modal component.

[0097] In one embodiment, the fault warning generation module is also used to: generate a fused modal energy value based on the modal component energy; generate an energy growth rate and an energy fluctuation abnormality based on the fused modal energy value corresponding to each target vibration data, and generate a unit fault warning result based on the energy growth rate and the energy fluctuation abnormality.

[0098] In one embodiment, the unit fault warning result includes a unit fault warning indication and a unit non-fault indication; the fault warning generation module is also used to: determine whether the energy growth rate is greater than the healthy growth rate, and the energy fluctuation abnormality is greater than the healthy abnormality; if the judgment is yes, generate a unit fault warning; if the judgment is no, generate a unit normal indication.

[0099] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0100] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0101] It should be noted that the information interaction, execution process and other contents between the above modules are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0102] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0103] An embodiment of the present application also provides a network device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor implements the steps of any of the above-mentioned method embodiments when executing the computer program.

[0104] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned various method embodiments can be implemented.

[0105] An embodiment of the present application provides a computer program product. When the computer program product is run on a mobile terminal, the mobile terminal can implement the steps in the above-mentioned various method embodiments when executing the computer program product.

[0106] If the integrated unit is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process steps in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a camera / terminal device, recording medium, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signals, telecommunication signals, and software distribution media. Examples include USB flash drives, removable hard drives, magnetic disks, or optical disks. In some jurisdictions, based on legislation and patent practice, computer-readable media cannot be electric carrier signals or telecommunication signals.

[0107] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0108] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0109] In the embodiments provided in this application, it should be understood that the disclosed devices / network equipment and methods can be implemented in other ways. For example, the device / network equipment embodiments described above are merely illustrative. For example, the division of the modules or units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0110] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0111] The above-described 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. 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 various embodiments of the present application, and should all be included in the scope of protection of the present application.

[0112] An embodiment of the present application also provides a computer device, which includes: at least one processor, a memory, and a computer program stored in the memory and executable on the at least one processor, wherein the processor implements the steps in any embodiment of the above method when executing the computer program.

[0113] The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above description is an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components than described above, or a combination of certain components, or different components. For example, the computer device may also include input / output devices, network access devices, etc.

[0114] The processor may be a central processing unit (CPU), and the processor 0 may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0115] In some embodiments, the memory may be an internal storage unit of the computer device, such as a hard disk or memory of the computer device. In other embodiments, the memory may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc. equipped with the computer device. Furthermore, the memory may include both an internal storage unit of the computer device and an external storage device. The memory is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory may also be used to temporarily store data that has been output or is about to be output.

[0116] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0117] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and such modifications and improvements are intended to fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring, characterized in that: The method comprises: Acquiring operating parameter data of a doubly-fed wind turbine generator set, and generating similar operating condition time periods of a gearbox in the doubly-fed wind turbine generator set based on the operating parameter data using a clustering algorithm; extracting target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period; Generating modal components and center frequencies corresponding to the modal components according to the target vibration data, and generating modal component energies according to the modal components and the center frequencies corresponding to the modal components; A fused modal energy value is generated according to the modal component energy, and a unit fault warning result is generated according to the fused modal energy value.

2. The method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring according to claim 1, characterized in that: Acquiring operating parameter data of a doubly-fed wind turbine generator set, and generating similar operating condition time periods of a gearbox in the doubly-fed wind turbine generator set based on the operating parameter data using a clustering algorithm, including: Acquiring operating parameter data of a doubly-fed wind turbine generator set, and generating a multi-dimensional operating condition characteristic vector according to the operating parameter data; Normalizing and assigning weights to the multi-dimensional operating condition feature vectors to generate weighted feature vectors; Segmenting the weighted feature vector according to preset segmentation intervals, and screening out target operating condition segments based on a clustering algorithm and preset screening conditions; A similar operating condition time period is generated according to the target operating condition section.

3. The method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring according to claim 2, characterized in that: The preset segment intervals include wind speed intervals and pitch angle intervals; The weighted feature vector is segmented according to a preset segmentation interval, and the target operating condition segment is screened out based on a clustering algorithm and preset screening conditions, including: The weighted feature vector is segmented according to the wind speed interval and the pitch angle interval, and the data in each segment are respectively clustered using a clustering algorithm to obtain a plurality of operating condition clusters; Filtering out target operating condition segments from each of the operating condition clusters according to preset filtering conditions; The center vector of each target operating condition segment is calculated, and the target operating condition segment is selected from the target operating condition segments according to the center vector.

4. The method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring according to claim 1, characterized in that: The vibration monitoring data includes first vibration data and second vibration data, the first vibration data is obtained based on a first vibration sensor, and the second vibration data is obtained based on a second vibration sensor; the first vibration sensor and the second vibration sensor are arranged opposite to each other on both sides of the input shaft bearing seat of the gearbox; Extracting target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period includes: acquiring first vibration data and second vibration data of the gearbox; generating differential vibration data based on the first vibration data and the second vibration data; Target vibration data is extracted from the differential vibration data according to the similar operating condition time period.

5. The method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring according to claim 1, characterized in that: Generating modal components and center frequencies corresponding to the modal components according to the target vibration data, and generating modal component energies according to the modal components and the center frequencies corresponding to the modal components, including: Decomposing the target vibration data based on a variational modal decomposition method, and generating a plurality of modal components and a center frequency corresponding to each modal component; Filtering out target modal components based on the modal components, center frequencies, and estimated main gear meshing frequencies; Modal component energy is generated according to the target modal component.

6. The method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring according to claim 1, characterized in that: Generating a fused modal energy value according to the modal component energy, and generating a unit fault warning result according to the fused modal energy value, including: generating a fused modal energy value according to the modal component energies; generating an energy growth rate and an energy fluctuation abnormality degree according to the fused modal energy value corresponding to each target vibration data; A unit fault warning result is generated according to the energy growth rate and the energy fluctuation abnormality.

7. The method for monitoring faults of a doubly-fed wind turbine generator system based on frequency monitoring according to claim 1, characterized in that: The unit fault warning result includes a unit fault warning and a unit normal indication; Generating a unit fault warning result according to the energy growth rate and energy fluctuation abnormality includes: Determining whether the energy growth rate is greater than a healthy growth rate, and the energy fluctuation abnormality is greater than a healthy abnormality; If the judgment is yes, a unit fault warning is generated; If the judgment is no, a unit normal indication is generated.

8. A double-fed wind turbine fault monitoring system based on frequency monitoring, characterized in that: The system comprises: A similar operating condition generation module is used to obtain operating parameter data of the doubly-fed wind turbine generator set and generate similar operating condition time periods of the gearbox in the doubly-fed wind turbine generator set based on the operating parameter data based on a clustering algorithm; a vibration data screening module, configured to extract target vibration data from the vibration monitoring data of the gearbox according to the similar operating condition time period; a modal energy generation module, configured to generate modal components and center frequencies corresponding to the modal components according to the target vibration data, and to generate modal component energy according to the modal components and the center frequencies corresponding to the modal components; The fault warning generation module is used to generate a fused modal energy value according to the modal component energy, and generate a unit fault warning result according to the fused modal energy value.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images