Wind turbine generator running state monitoring method based on machine learning

Abstract

The invention relates to the technical field of data processing, in particular to a wind turbine generator running state monitoring method based on machine learning, and the method comprises the steps: carrying out the logarithmic transformation of each piece of data in a data sequence corresponding to a plurality of data indexes in a preset time period during the running of a wind turbine generator, and obtaining a corresponding logarithmic value; constructing a scatter diagram of all logarithm values; obtaining a neighborhood radius and a minimum sample number of a DBSCAN algorithm, and obtaining a clustering result of all data points in the scatter diagram according to the neighborhood radius and the minimum sample number; a plurality of boundary points are obtained according to the clustering result, the isolation degree of each boundary point is obtained according to the distribution characteristics of the data points within the neighborhood radius range of each boundary point, and abnormal points in the scatter diagram are obtained according to the isolation degree of each boundary point; and the operation state of the wind turbine generator is monitored according to the abnormal points in the scatter diagram, so that the accuracy of monitoring the operation state of the wind turbine generator by the DBSCAN algorithm is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method for monitoring the operating status of a wind turbine generator set based on machine learning. Background Art

[0002] Wind turbine operating status monitoring plays a key role in the wind energy industry. Monitoring the operating status of wind turbines can timely detect potential problems and abnormalities, so as to take preventive maintenance measures, reduce equipment damage and downtime, and improve equipment reliability and safety.

[0003] At present, various sensors are used to monitor the working status and environmental conditions of various components of wind turbines in real time. The operation data of wind turbines are collected and transmitted in real time through remote monitoring systems. The collected data is analyzed and processed in real time to obtain the health indicators of wind turbines. The health status of wind turbines is evaluated based on the health indicators. However, during the data collection and analysis process, the data will contain noise points and outliers. The DBSCAN algorithm is a powerful density clustering algorithm that divides data points into core points, boundary points, and noise points based on the density connectivity between data points. In the traditional way, only noise points are used as outliers to evaluate the health status of wind turbines.

[0004] However, in this process, some abnormal points are highly similar to normal data points in features and are covered within the neighborhood radius of the core points as boundary points. The DBSCAN algorithm cannot accurately identify such situations, which reduces the accuracy of anomaly detection.

[0005] Therefore, how to improve the accuracy of wind turbine operating status monitoring using the DBSCAN algorithm has become an urgent problem to be solved. Summary of the invention

[0006] In view of this, an embodiment of the present invention provides a method for monitoring the operating status of a wind turbine generator set based on machine learning to solve the problem of how to improve the accuracy of monitoring the operating status of a wind turbine generator set using the DBSCAN algorithm.

[0007] An embodiment of the present invention provides a method for monitoring the operating status of a wind turbine generator system based on machine learning, the method comprising the following steps: When the wind turbine is running, at least two data indicators within a preset time period are respectively obtained, a data sequence of each data indicator is obtained, each data in each data sequence is logarithmically transformed to obtain a corresponding logarithmic value, and a scatter plot of all logarithmic values ​​is constructed; When clustering all the data points in the scatter plot using the DBSCAN algorithm, obtaining a neighborhood radius and a minimum number of samples of the DBSCAN algorithm, and obtaining a clustering result of all the data points in the scatter plot according to the neighborhood radius and the minimum number of samples; At least one boundary point in the scatter plot is obtained according to the clustering result, and for any boundary point, the degree of isolation of the boundary point is obtained according to the distribution characteristics of data points within a neighborhood radius of the boundary point, the degrees of isolation of all boundary points in the scatter plot are obtained, and abnormal points in the scatter plot are obtained according to the degrees of isolation of all boundary points in the scatter plot; The operating status of the wind turbine generator set is monitored according to the abnormal points in the scatter plot.

[0008] Preferably, obtaining the clustering results of all data points in the scatter plot according to the neighborhood radius and the minimum number of samples includes: According to the neighborhood radius and the minimum number of samples, the data points in the scatter plot are divided into core points, boundary points and noise points, each boundary point and noise point in the scatter plot is regarded as a complex point, and for any core point, the complex points belonging to the core point are clustered together with the core point to obtain at least one cluster.

[0009] Preferably, obtaining the isolation degree of the boundary point according to the distribution characteristics of the data points within the neighborhood radius of the boundary point includes: Obtaining a neighborhood data complexity coefficient of the boundary point according to the number of clusters contained within the neighborhood radius of the boundary point; Obtaining a weighted density feature value of the boundary point according to the Euclidean distance between the boundary point and the core point of the cluster where the boundary point is located; The degree of isolation of the boundary point is obtained by performing a weighted summation on the neighborhood data complexity coefficient and the weighted density eigenvalue.

[0010] Preferably, obtaining the neighborhood data complexity coefficient of the boundary point according to the number of clusters contained in the neighborhood radius of the boundary point includes: Each complex point within the neighborhood radius of the boundary point is taken as a target complex point, the number of the target complex points is counted, the clusters to which each target complex point belongs are counted to obtain the number of clusters, and the ratio of the number of clusters to the number of target complex points is taken as the neighborhood data complexity coefficient of the boundary point.

[0011] Preferably, obtaining the weighted density feature value of the boundary point according to the Euclidean distance between the boundary point and the core point of the cluster where the boundary point is located includes: Obtaining a neighborhood area according to the neighborhood radius, and taking the ratio between the number of the target complex points and the neighborhood area as a first variable; Taking the core point to which the boundary point belongs as the target core point, calculating the Euclidean distance between the boundary point and the target core point, and taking the difference between the neighborhood radius and the Euclidean distance as the second variable; The product of the first variable and the second variable is used as the weighted density of the boundary point, an average value of the weighted density of all boundary points in the scatter plot is calculated, and a weighted density characteristic value of the boundary point is obtained according to the average value.

[0012] Preferably, obtaining the weighted density characteristic value of the boundary point according to the average value includes: The absolute value of the difference between the weighted density of the boundary point and the average value is calculated as the first eigenvalue, and the first eigenvalue is used as an independent variable of the hyperbolic tangent function to obtain the weighted density eigenvalue of the boundary point.

[0013] Preferably, obtaining the abnormal points in the scatter plot according to the isolation degree of all boundary points in the scatter plot includes: For any isolation degree, if the isolation degree is within a preset isolation degree range, the boundary point corresponding to the isolation degree is marked as an abnormal point.

[0014] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: When a wind turbine is running, the present invention obtains at least two data indicators within a preset time period, obtains a data sequence corresponding to each data indicator, performs logarithmic transformation on each data in each data sequence, obtains a corresponding logarithmic value, and constructs a scatter plot of all logarithmic values; when clustering all data points in the scatter plot using a DBSCAN algorithm, obtains a neighborhood radius and a minimum number of samples of the DBSCAN algorithm, and obtains a clustering result of all data points in the scatter plot according to the neighborhood radius and the minimum number of samples; obtains at least one boundary point in the scatter plot according to the clustering result, and for any boundary point, obtains the degree of isolation of the boundary point according to the distribution characteristics of data points within the neighborhood radius of the boundary point, obtains the degree of isolation of all boundary points in the scatter plot, and obtains abnormal points in the scatter plot according to the degree of isolation of all boundary points in the scatter plot; and monitors the operating status of the wind turbine according to the abnormal points in the scatter plot. Among them, two data indicators are obtained when the wind turbine is running, and the data in the data indicators are logarithmically changed to enhance the distribution characteristics of data points, reduce the intervals between data points, make densely distributed data points more prominent and recognizable, and also reduce the sensitivity of the data to outliers, which is helpful for analysis and identification, and construct a scatter plot of all logarithmically changed values. At this time, the coordinate axis range of the scatter plot can effectively display the distribution of data points, and will not contain too many or too few blanks, optimize the clustering effect of the DBSCAN algorithm, use the DBSCAN algorithm on the scatter plot to obtain boundary points, analyze the degree of isolation of all boundary points and the type and distribution characteristics of data points within the neighborhood radius of the boundary points, calculate the degree of isolation of the boundary points, mark the boundary points that meet the requirements as outliers, complete the judgment of all boundary points, and improve the accuracy of the DBSCAN algorithm in monitoring the operating status of wind turbines. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0016] Figure 1 This is a method flow chart of a method for monitoring the operating status of a wind turbine generator system based on machine learning provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0017] Embodiments of the present disclosure are described in detail below, and examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, but should not be understood as limiting the present disclosure.

[0018] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure 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 interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0019] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0020] See also Figure 1 , is a method flow chart of a method for monitoring the operating status of a wind turbine generator system based on machine learning provided in Embodiment 1 of the present invention, such as Figure 1 As shown, the method may include: Step S101, when the wind turbine is running, respectively obtain at least two data indicators within a preset time period, obtain a data sequence corresponding to each data indicator, perform logarithmic transformation on each data in each data sequence to obtain a corresponding logarithmic value, and construct a scatter plot of all logarithmic values.

[0021] Because the amount of data on the operating status of wind turbines is large, the use of scatter plots can help analyze the relationship between different operating parameters of wind turbines. At the same time, scatter plots can also help identify and locate abnormal data points. However, if a scatter plot is directly constructed and time is used as the X-axis and data size is used as the Y-axis, the scatter plot will become very crowded, covering up some distribution characteristics of the data and the correlation between the data. In addition, the data points in the scatter plot change linearly as a whole, and the density change is not obvious, which may cause problems of over-segmentation or unreasonable division when the DBSCAN algorithm is used to cluster the data in the scatter plot. Therefore, when the wind turbine is running, at least two data indicators in the preset time period are obtained respectively, and a data series corresponding to each data indicator is obtained. Each data in each data series is logarithmically transformed to obtain the corresponding logarithmic value, and a scatter plot of all logarithmic values ​​is constructed.

[0022] The specific process for obtaining the data series of each data indicator is as follows: First, consider the correlation between two data indicators in the physical sense, and obtain the two data indicators that are required and correlated in the physical sense, such as the relationship between speed and power: the higher the speed, the greater the power; the relationship between temperature and humidity: the higher the temperature, the lower the humidity, etc. Set the preset time period to 1 hour, and there is no restriction here. The implementer can set it according to the implementation scenario, and obtain the data sequence corresponding to each data indicator within 1 hour according to these two data indicators. In order to ensure the accuracy and consistency of each data in the data sequence, after obtaining all data sequences, each data in all data sequences is preprocessed to achieve the purpose of data cleaning for all data sequences, wherein data preprocessing includes but is not limited to: data cleaning, filling missing values, deleting samples or features where missing values ​​are located, detecting and processing outliers, and deleting duplicate samples or features; data normalization, ensuring that the data meets specific standards or specifications, such as unit consistency, unified data format, etc.; data conversion, such as normalizing or standardizing the data to ensure that different indicators are comparable. Among them, data preprocessing technology belongs to the prior art and will not be repeated here.

[0023] Among them, each data in each data sequence is logarithmically transformed to obtain the corresponding logarithmic value, and a scatter plot of all logarithmic values ​​is constructed. The specific process is: In order to enhance the changing trend of data in the data series, reduce the data spacing, make the densely distributed data more prominent and recognizable, and help the subsequent analysis and outlier identification, each data in the data series corresponding to each data indicator is logarithmically transformed with a base of 10 to obtain the corresponding logarithmic value, and a scatter plot of all logarithmic values ​​is constructed. The calculation formula of the logarithmic value is:

[0024] Among them, y represents the value after logarithmic transformation, that is, the logarithmic value, and x represents the data in the data sequence.

[0025] It is worth noting that if the two data indicators to be obtained are physically correlated, such as the relationship between speed and power, the relationship between temperature and humidity, etc., then these two data indicators are used as the X-axis and Y-axis of the scatter plot. If there is only one data indicator to be obtained, or two or more data indicators to be obtained are not physically correlated, then time is used as the X-axis of the scatter plot, and the corresponding logarithmic value in the data sequence corresponding to the data indicator is used as the Y-axis of the scatter plot.

[0026] Step S102, when clustering all data points in the scatter plot using the DBSCAN algorithm, obtain the neighborhood radius and the minimum number of samples of the DBSCAN algorithm, and obtain the clustering result of all data points in the scatter plot according to the neighborhood radius and the minimum number of samples.

[0027] First, the neighborhood radius and the minimum number of samples of the DBSCAN algorithm are determined through empirical adjustment or verification technology, and then the DBSCAN algorithm is run on the scatter plot to obtain the clustering results of all data points in the scatter plot for screening outliers.

[0028] Among them, the clustering results of all data points in the scatter plot include: According to the neighborhood radius and the minimum number of samples, the data points in the scatter plot are divided into core points, boundary points and noise points, each boundary point and noise point in the scatter plot is regarded as a complex point, and for any core point, the complex points belonging to the core point are clustered together with the core point to obtain at least one cluster.

[0029] The DBSCAN algorithm divides all data points in the scatter plot into core points, boundary points, and noise points according to the neighborhood radius and the minimum number of samples, where the data points whose number of sample points within the neighborhood radius is greater than or equal to the minimum number of samples are core points, data points that are not core points but within the neighborhood radius of a core point are boundary points, and data points that are neither core points nor boundary points are noise points. Each boundary point and each noise point in the scatter plot is regarded as a complex point, and each core point represents a cluster. For any core point, the complex point belonging to the core point is a cluster member of the cluster where the core point is located. Similarly, all data points in the scatter plot are divided into at least one cluster. It is worth noting that the DBSCAN algorithm belongs to the prior art and will not be described in detail here.

[0030] Step S103, obtaining at least one boundary point in the scatter plot according to the clustering result, obtaining the degree of isolation of any boundary point according to the distribution characteristics of data points within the neighborhood radius of the boundary point, obtaining the degree of isolation of all boundary points in the scatter plot, and obtaining abnormal points in the scatter plot according to the degree of isolation of all boundary points in the scatter plot.

[0031] There may be some abnormal points in the boundary points. These abnormal points are highly similar to normal data points in features. When the DBSCAN algorithm is running, they are covered in the neighborhood radius of the core point, making it impossible for the DBSCAN algorithm to identify them. That is, these abnormal points are not classified as noise points, but are classified as boundary points, resulting in a decrease in the accuracy of abnormality detection. Therefore, in the example of the present invention, according to step S102, the clustering results of all data points in the scatter plot can be obtained, and then each boundary point in the scatter plot is obtained according to the clustering results. The degree of isolation of each boundary point is calculated by analyzing the type and distribution characteristics of the data points within the neighborhood radius of the boundary point, and the boundary points that meet the requirements are marked as abnormal points, and the discrimination of all boundary points is completed, thereby improving the accuracy of abnormality detection of the DBSCAN algorithm.

[0032] Among them, taking any boundary point as an example, the specific process of calculating the isolation degree of the boundary point is: (1) Obtaining the neighborhood data complexity coefficient of the boundary point according to the number of clusters contained in the neighborhood radius of the boundary point, including: Each complex point within the neighborhood radius of the boundary point is taken as a target complex point, the number of the target complex points is counted, the clusters to which each target complex point belongs are counted to obtain the number of clusters, and the ratio of the number of clusters to the number of target complex points is taken as the neighborhood data complexity coefficient of the boundary point.

[0033] In one embodiment, for the i-th boundary point in the scatter plot, the number of all data points within the neighborhood radius of the i-th boundary point except the core point, that is, the number of target complex points, is counted, and the number of clusters is obtained by counting the clusters where each target complex point is located. The neighborhood data complexity coefficient of the i-th boundary point is obtained according to the number of target complex points and the number of clusters. The calculation formula of the neighborhood data complexity coefficient of the i-th boundary point is:

[0034] in, represents the complexity coefficient of the neighborhood data of the i-th boundary point, Indicates the number of clusters corresponding to the target complex point corresponding to the i-th boundary point, Indicates the number of target complex points corresponding to the i-th boundary point.

[0035] It should be noted that The larger it is, the more clusters there are in the neighborhood of the i-th boundary point. The closer it is to 1, that is, the closer the neighborhood data complexity coefficient of the i-th boundary point is to 1, indicating that the boundary point may be closer to other clusters or belong to the boundary area of ​​multiple clusters, and the greater the possibility of being an outlier or unclear data point.

[0036] (2) Obtaining a weighted density feature value of the boundary point according to the Euclidean distance between the boundary point and the core point of the cluster to which the boundary point belongs, including: Obtaining a neighborhood area according to the neighborhood radius, and taking the ratio between the number of the target complex points and the neighborhood area as a first variable; Taking the core point to which the boundary point belongs as the target core point, calculating the Euclidean distance between the boundary point and the target core point, and taking the difference between the neighborhood radius and the Euclidean distance as the second variable; The product of the first variable and the second variable is used as the weighted density of the boundary point, an average value of the weighted density of all boundary points in the scatter plot is calculated, and a weighted density characteristic value of the boundary point is obtained according to the average value.

[0037] In one implementation, the area covered by the neighborhood radius is calculated, the Euclidean distance between the ith boundary point and the core point to which the ith boundary point belongs is calculated, and the weighted density of the ith boundary point is obtained according to the area and the Euclidean distance, wherein the Euclidean distance is a prior art and will not be described in detail here. Then, the calculation formula for the weighted density of the ith boundary point is:

[0038] in, represents the weighted density of the i-th boundary point, represents the number of target complex points corresponding to the i-th boundary point, represents the neighborhood radius, represents the area covered by the neighborhood radius, Represents the Euclidean distance between the ith boundary point and the core point to which the ith boundary point belongs.

[0039] It should be noted that It represents the comprehensive situation of the Euclidean distance between the i-th boundary point and the core point and the data density within the neighborhood radius of the i-th boundary point. The smaller the weighted density, the farther the i-th boundary point is from the core point, and the fewer data points there are within the neighborhood radius of the i-th boundary point, which means it is more likely to be an outlier.

[0040] Similarly, the weighted density of all boundary points in the scatter plot is calculated to obtain the average value of the weighted density, and then the weighted density eigenvalue of the ith boundary point is obtained according to the average value of the weighted density and the weighted density of the ith boundary point.

[0041] Wherein, obtaining the weighted density characteristic value of the boundary point according to the average value includes: The absolute value of the difference between the weighted density of the boundary point and the average value is calculated as the first eigenvalue, and the first eigenvalue is used as an independent variable of the hyperbolic tangent function to obtain the weighted density eigenvalue of the boundary point.

[0042] In one embodiment, the weighted density characteristic value of the ith boundary point is obtained according to the average value of the weighted density of all boundary points in the scatter plot and the weighted density of the ith boundary point, and the calculation formula of the weighted density characteristic value is:

[0043] in, represents the weighted density eigenvalue of the i-th boundary point, tanh() represents the hyperbolic tangent function, represents the weighted density of the i-th boundary point, represents the weighted density of the jth boundary point, represents the number of all boundary points in the scatter plot, Represents the absolute value symbol.

[0044] It should be noted that represents the average value of the weighted density of all boundary points in the scatter plot, The larger the value, the greater the difference between the i-th boundary point and the overall density of all boundary points in the scatter plot. The farther the i-th boundary point is from the core point to which it belongs, the more likely it is an outlier. The tanh() function is used to convert The value of is fixed between [0, 1].

[0045] (3) Performing a weighted summation on the neighborhood data complexity coefficient and the weighted density eigenvalue to obtain the degree of isolation of the boundary point.

[0046] In one embodiment, the degree of isolation of the i-th boundary point is obtained according to the neighborhood data complexity coefficient and the weighted density feature value of the i-th boundary point, and the calculation formula of the degree of isolation is:

[0047] in, represents the degree of isolation of the i-th boundary point, represents the weighted density eigenvalue of the i-th boundary point, represents the complexity coefficient of the neighborhood data of the i-th boundary point, represents the first weight value, Represents the second weight value.

[0048] It should be noted that setting is 0.65, set is 0.35, there is no limit here, the implementer can set it according to the implementation scenario, and then The value range of is (0, 1), The closer it is to 1, the more noise points appear in the neighborhood radius of the ith boundary point and the more clusters there are within the neighborhood radius. The farther the ith boundary point is from the core point and the fewer data points there are within the neighborhood radius, the more likely the ith boundary point is to be distributed in a sparsely populated area and is more likely to be an outlier. Similarly, the degree of isolation of each boundary point is analyzed to obtain the degree of isolation of all boundary points in the scatter plot. Furthermore, the outliers in the scatter plot are obtained based on the degree of isolation of all boundary points in the scatter plot, including: For any isolation degree, if the isolation degree is within a preset isolation degree range, the boundary point corresponding to the isolation degree is marked as an abnormal point.

[0049] In one embodiment, the preset isolation degree range is set to (0.75, 1), which is not limited here. The implementer can set it according to the implementation scenario. For any boundary point, if the abnormality degree of the boundary point is within (0.75, 1), the boundary point is marked as an abnormal point. Similarly, the isolation degree of each boundary point is analyzed to obtain all abnormal points in the scatter plot.

[0050] Step S104: monitoring the operating status of the wind turbine generator system according to the abnormal points in the scatter plot.

[0051] According to step S103, abnormal points are screened out from all boundary points in the scatter plot, and then the noise points and abnormal points in the scatter plot are used as abnormal values ​​when the wind turbine is running, and the core points in the scatter plot and the boundary points not marked as abnormal points are used as normal values. The operating status of the wind turbine is monitored according to the abnormal values ​​and normal values, and the monitored data and analysis results are visualized, and intuitive reports and charts are provided. Corresponding measures are taken according to the monitoring results to ensure the safe and efficient operation of the wind turbine.

[0052] It is worth noting that the focus of the present invention is how to screen out abnormal points from the boundary points obtained by clustering using the DBSCAN algorithm, monitor the operating status data of the wind turbine according to the abnormal values ​​and take corresponding measures, which belongs to the prior art and will not be repeated here.

[0053] In summary, the present invention provides a method for monitoring the operating status of a wind turbine based on machine learning, including obtaining at least two data indicators within a preset time period respectively when the wind turbine is running, obtaining a data sequence corresponding to each data indicator, performing a logarithmic transformation on each data in each of the data sequences to obtain a corresponding logarithmic value, and constructing a scatter plot of all logarithmic values; when clustering all data points in the scatter plot using the DBSCAN algorithm, obtaining a neighborhood radius and a minimum number of samples of the DBSCAN algorithm, and obtaining a clustering result of all data points in the scatter plot according to the neighborhood radius and the minimum number of samples; obtaining at least one boundary point in the scatter plot according to the clustering result, and for any boundary point, obtaining the degree of isolation of the boundary point according to the distribution characteristics of data points within the neighborhood radius of the boundary point, obtaining the degree of isolation of all boundary points in the scatter plot, and obtaining abnormal points in the scatter plot according to the degree of isolation of all boundary points in the scatter plot; and monitoring the operating status of the wind turbine according to the abnormal points in the scatter plot. Among them, two data indicators are obtained when the wind turbine is running, and the data in the data indicators are logarithmically changed to enhance the distribution characteristics of data points, reduce the intervals between data points, make densely distributed data points more prominent and recognizable, and also reduce the sensitivity of the data to outliers, which is helpful for analysis and identification, and construct a scatter plot of all logarithmically changed values. At this time, the coordinate axis range of the scatter plot can effectively display the distribution of data points, and will not contain too many or too few blanks, optimize the clustering effect of the DBSCAN algorithm, use the DBSCAN algorithm on the scatter plot to obtain boundary points, analyze the degree of isolation of all boundary points and the type and distribution characteristics of data points within the neighborhood radius of the boundary points, calculate the degree of isolation of the boundary points, mark the boundary points that meet the requirements as outliers, complete the judgment of all boundary points, and improve the accuracy of the DBSCAN algorithm in monitoring the operating status of wind turbines.

[0054] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such 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 invention, and should be included in the protection scope of the present invention.

Claims

1. A method for monitoring the operating status of a wind turbine generator system based on machine learning, characterized in that: The wind turbine operating status monitoring method based on machine learning includes: When the wind turbine is running, at least two data indicators within a preset time period are respectively obtained, a data sequence of each data indicator is obtained, each data in each data sequence is logarithmically transformed to obtain a corresponding logarithmic value, and a scatter plot of all logarithmic values ​​is constructed; When clustering all the data points in the scatter plot using the DBSCAN algorithm, obtaining a neighborhood radius and a minimum number of samples of the DBSCAN algorithm, and obtaining a clustering result of all the data points in the scatter plot according to the neighborhood radius and the minimum number of samples; At least one boundary point in the scatter plot is obtained according to the clustering result, and for any boundary point, the degree of isolation of the boundary point is obtained according to the distribution characteristics of data points within a neighborhood radius of the boundary point, the degrees of isolation of all boundary points in the scatter plot are obtained, and abnormal points in the scatter plot are obtained according to the degrees of isolation of all boundary points in the scatter plot; The operating status of the wind turbine generator set is monitored according to the abnormal points in the scatter plot.

2. The method for monitoring the operating status of a wind turbine generator system based on machine learning according to claim 1, characterized in that: The step of obtaining the clustering results of all data points in the scatter plot according to the neighborhood radius and the minimum number of samples includes: According to the neighborhood radius and the minimum number of samples, the data points in the scatter plot are divided into core points, boundary points and noise points, each boundary point and noise point in the scatter plot is regarded as a complex point, and for any core point, the complex points belonging to the core point are clustered together with the core point to obtain at least one cluster.

3. The method for monitoring the operating status of a wind turbine generator system based on machine learning according to claim 2, characterized in that: The step of obtaining the isolation degree of the boundary point according to the distribution characteristics of the data points within the neighborhood radius of the boundary point comprises: Obtaining a neighborhood data complexity coefficient of the boundary point according to the number of clusters contained within the neighborhood radius of the boundary point; Obtaining a weighted density feature value of the boundary point according to the Euclidean distance between the boundary point and the core point of the cluster where the boundary point is located; The degree of isolation of the boundary point is obtained by performing a weighted summation on the neighborhood data complexity coefficient and the weighted density eigenvalue.

4. The method for monitoring the operating status of a wind turbine generator system based on machine learning according to claim 3, characterized in that: The obtaining of the neighborhood data complexity coefficient of the boundary point according to the number of clusters contained in the neighborhood radius of the boundary point includes: Each complex point within the neighborhood radius of the boundary point is taken as a target complex point, the number of the target complex points is counted, the clusters to which each target complex point belongs are counted to obtain the number of clusters, and the ratio of the number of clusters to the number of target complex points is taken as the neighborhood data complexity coefficient of the boundary point.

5. The method for monitoring the operating status of a wind turbine generator system based on machine learning according to claim 4, characterized in that: The step of obtaining the weighted density feature value of the boundary point according to the Euclidean distance between the boundary point and the core point of the cluster where the boundary point is located includes: Obtaining a neighborhood area according to the neighborhood radius, and taking the ratio between the number of the target complex points and the neighborhood area as a first variable; Taking the core point to which the boundary point belongs as the target core point, calculating the Euclidean distance between the boundary point and the target core point, and taking the difference between the neighborhood radius and the Euclidean distance as the second variable; The product of the first variable and the second variable is used as the weighted density of the boundary point, an average value of the weighted density of all boundary points in the scatter plot is calculated, and a weighted density characteristic value of the boundary point is obtained according to the average value.

6. The method for monitoring the operating status of a wind turbine generator system based on machine learning according to claim 5, characterized in that: The step of obtaining the weighted density characteristic value of the boundary point according to the average value includes: The absolute value of the difference between the weighted density of the boundary point and the average value is calculated as the first eigenvalue, and the first eigenvalue is used as an independent variable of the hyperbolic tangent function to obtain the weighted density eigenvalue of the boundary point.

7. The method for monitoring the operating status of a wind turbine generator system based on machine learning according to claim 1, characterized in that: The obtaining of abnormal points in the scatter plot according to the isolation degree of all boundary points in the scatter plot comprises: For any isolation degree, if the isolation degree is within a preset isolation degree range, the boundary point corresponding to the isolation degree is marked as an abnormal point.

Images