A wind turbine generator system monitoring data validity evaluation method, device and medium

Abstract

The application relates to a wind turbine monitoring data validity evaluation method, which comprises the following steps: obtaining the linear correlation of monitoring data under the normal working state of a wind turbine; obtaining the monitoring data of the wind turbine in real time; preprocessing and shearing the monitoring data, and calculating the Pearson correlation coefficient and the grey correlation degree between different data type monitoring data; comparing the Pearson correlation coefficient between the monitoring data and the grey correlation degree between the monitoring data sets, obtaining and recording the linear correlation between the monitoring data; comparing the linear correlation obtained in real time with the linear correlation under the normal working state, if the linear correlation changes, it is determined that the monitoring data is abnormal, and an alarm is given, if the linear correlation does not change, it is determined that the monitoring data is valid. Compared with the prior art, the application combines the Pearson correlation coefficient and the grey correlation degree, and can more accurately determine the validity of the monitoring data.

Description

Technical Field

[0001] This invention relates to the field of data fusion technology, and in particular to a method, equipment and medium for evaluating the effectiveness of monitoring data from wind turbine generator sets. Background Technology

[0002] Against the backdrop of global efforts to address climate change and energy transition, the shift from a traditional fossil fuel-based economy to renewable energy is inevitable. Wind power, with its abundant resources and environmental friendliness, has become a significant force in the renewable energy sector. Globally, installed wind power capacity continues to grow rapidly, with large-scale wind turbines being put into operation, highlighting their increasingly important role in the energy supply system. The operating environment of wind turbines is extremely complex, influenced by a combination of meteorological factors and their own mechanical and electrical characteristics. External conditions such as intermittent wind speed, variable wind direction, and changes in temperature and air pressure can cause significant fluctuations in operating parameters such as output power, speed, and blade stress. To ensure efficient and stable operation of wind turbines, reduce failure rates, and improve power generation efficiency, real-time monitoring and data validity assessment of their operating status are necessary to enable timely and effective maintenance and management measures. In-depth analysis of monitoring data can uncover underlying patterns and characteristics, providing strong support for optimized operation, fault diagnosis, and performance evaluation of wind turbines.

[0003] Because data may contain anomalies during collection, a single data source cannot determine the validity of the data on its own. Furthermore, data with many different parameters often have inherent logical relationships, so the validity of the data can be judged based on other parameters that are strongly correlated with it. Traditional methods have certain limitations when dealing with multivariate data with complex correlations. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art by providing a method, device and medium for evaluating the effectiveness of wind turbine generator monitoring data, and to explore the linear relationship between different parameters in order to achieve the effectiveness evaluation of wind turbine generator monitoring data.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] A method for evaluating the validity of monitoring data from wind turbine generator sets, the method comprising:

[0007] Obtain the linear correlation between various data types of monitoring data under normal operating conditions of wind turbine generator sets;

[0008] Real-time acquisition of monitoring data from wind turbine generators;

[0009] The monitoring data is preprocessed and cropped to obtain multiple monitoring datasets of equal data volume of different data types. The Pearson correlation coefficient and the grey relational degree between the monitoring data of different data types are calculated.

[0010] Compare the Pearson correlation coefficients between different monitoring data sets and the grey relational degree between different monitoring datasets. If the absolute value of the Pearson correlation coefficient between any pair of monitoring data sets is greater than a preset threshold, the absolute value of the difference between the grey relational degree value and the Pearson correlation coefficient is less than a preset deviation value, and both the Pearson correlation coefficient and the grey relational degree are positive, then it is determined that there is a significant positive linear correlation between these two pairs of monitoring data sets. If the absolute value of the Pearson correlation coefficient between any pair of monitoring data sets is greater than a preset threshold, the absolute value of the difference between the grey relational degree value and the Pearson correlation coefficient is less than a preset deviation value, and both the Pearson correlation coefficient and the grey relational degree are negative, then it is determined that there is a significant positive linear correlation between these two pairs of monitoring data sets. If there is a significant negative linear correlation between any two pairs of monitoring data, and the absolute values ​​of the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both less than the preset threshold and both the Pearson correlation coefficient and the grey relational degree are positive, then it is determined that there is a weak positive linear correlation between these two pairs of monitoring data. If the absolute values ​​of the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both less than the preset threshold and both the Pearson correlation coefficient and the grey relational degree are negative, then it is determined that there is a weak negative linear correlation between these two pairs of monitoring data. If the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both 0, then it is determined that there is no linear correlation between these two pairs of monitoring data. Record the linear correlation between each pair of monitoring data.

[0011] The linear correlation obtained by real-time calculation is compared with the linear correlation of monitoring data of various data types under normal working conditions. If the linear correlation between any pair of monitoring data changes, it is determined that the monitoring data is abnormal and an alarm is triggered. If the linear correlation between each monitoring data remains unchanged, the monitoring data is determined to be valid.

[0012] Furthermore, the monitoring data includes acceleration, tower tilt angle, temperature, and load.

[0013] Furthermore, the preprocessing includes outlier handling, missing value handling, and erroneous data correction.

[0014] Furthermore, the formula for calculating the Pearson correlation coefficient is as follows:

[0015]

[0016] Where PCC is the Pearson correlation coefficient, n is the sample size, and x i and y i Let x and y be the i-th observations of samples x and y, respectively. and y and y are the sample means of samples x and y, respectively.

[0017] Furthermore, the Pearson correlation coefficient has a range of [-1, 1]. If the calculated Pearson correlation coefficient is greater than or less than the range, an alarm will be triggered.

[0018] Furthermore, the calculation process for the grey relational degree between the different data type datasets includes:

[0019] Obtain multiple monitoring datasets with equal amounts of data of different data types;

[0020] Arrange the data from the same monitoring dataset in chronological order of collection time to form a data column;

[0021] If any data column is designated as the reference data column, then the remaining data columns are designated as comparison data columns. Calculate the grey relational coefficient between the data in each comparison data column and the data in the reference data column.

[0022] Calculate the average value of the grey relational coefficients between the data in each comparison data column and the data in the reference data column. The average value is the grey relational degree between the comparison data column and the reference data column, that is, the grey relational degree between the corresponding monitoring dataset and the other monitoring datasets.

[0023] Repeat the selection and calculation of the reference data columns above to obtain the grey correlation degree between the datasets of different data types.

[0024] Furthermore, the formula for calculating the grey relational coefficient is as follows:

[0025] X0 = (x1, x2, ..., x) n ),

[0026] X i =(x i1 x i2 , ..., x in ), i = 1, 2, ..., m,

[0027] Δ t(i,0) =|x0(t)-x i (t)|,

[0028]

[0029] Where X0 is the reference data column, X i This is a comparison of data columns, where n is the number of data points, m is the number of data types, ρ is the resolution coefficient, and Δ... t(i,0) For each comparison data column X i The absolute difference between the reference data column X0 and the reference data column X0 at time point t, Δ maxand Δ min These are the maximum and minimum values ​​of the absolute difference, respectively.

[0030] Furthermore, if the absolute value of the difference between the Pearson correlation coefficient and the grey relational degree between any pair of monitoring data is greater than a preset adjustment value, it is determined that there is an anomaly in the monitoring data and an alarm is triggered.

[0031] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the wind turbine generator monitoring data validity assessment method as described above.

[0032] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for evaluating the validity of wind turbine generator monitoring data as described above.

[0033] Compared with the prior art, the beneficial effects of the present invention include:

[0034] 1. This invention employs a combined analysis of Pearson correlation coefficient and grey relational coefficient to process data from different sensors in real time. Leveraging the complex correlations between wind turbine parameters, it uncovers linear relationships between different parameters and compares these relationships with those under normal operating conditions. This allows for timely detection of anomalies in the monitoring data, enabling real-time effectiveness assessment of wind turbine monitoring data. When anomalies are detected, timely alarms are triggered to alert relevant personnel and prevent potential safety hazards. This invention combines two evaluation indicators: Pearson correlation coefficient and grey relational coefficient. Pearson correlation coefficient primarily assesses the strength and direction of linear correlation, while grey relational coefficient considers the similarity and proximity between data sequences. The combination of both allows for a more accurate assessment of the effectiveness of the monitoring data, more comprehensively reflecting the linear correlation between monitoring data, and increasing computational robustness, further ensuring monitoring accuracy.

[0035] 2. This invention also limits the value of the Pearson correlation coefficient and the difference between the grey relational coefficient and the Pearson correlation coefficient, so as to more comprehensively detect whether the monitoring data is abnormal and issue an alarm in a timely manner. Attached Figure Description

[0036] Figure 1 This is a flowchart for evaluating the validity of data in this invention;

[0037] Figure 2 This is a heatmap of the Pearson correlation coefficients between the three variables in the tower section according to an embodiment of the present invention;

[0038] Figure 3This is a grey relational analysis heatmap of the tower section three acceleration 2 as reference data column and each variable in one embodiment of the present invention;

[0039] Figure 4 This is a heatmap of Pearson correlation coefficients among various variables in the three-acceleration anomaly of the tower section in one embodiment of the present invention;

[0040] Figure 5 This is a grey relational analysis heatmap of the three accelerations 1 and 2 (which are reference data columns) in a tower section of the present invention, and the variables. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0042] Example 1

[0043] The present invention aims to disclose a method for evaluating the validity of monitoring data of wind turbine generator sets. Since wind turbine generator sets are affected by the interaction of various meteorological factors and their own mechanical and electrical characteristics, there are many inherent logical relationships between various data of wind turbine generator sets, such as output power, speed, blade stress and other operating parameters. Therefore, the validity of each monitoring data of wind turbine generator sets can be judged based on other parameter data that are strongly correlated with it.

[0044] Methods for evaluating the validity of monitoring data from wind turbine generator sets, such as Figure 1 As shown, it includes:

[0045] Step S1: Obtain the linear correlation between various data types of monitoring data under normal operating conditions of the wind turbine generator set.

[0046] The process of obtaining the linear correlation between various data types of monitoring data under normal operating conditions of a wind turbine generator includes:

[0047] Based on historical operation records, acquire monitoring data of various data types under normal operating conditions of the wind turbine generator set;

[0048] The monitoring data is preprocessed and cropped to obtain multiple monitoring datasets of equal volume with different data types;

[0049] Calculate the Pearson correlation coefficient between monitoring data of different data types and the grey relational degree between datasets of different data types;

[0050] Compare the Pearson correlation coefficients among the monitoring data and the grey relational degree among the monitoring datasets, and record the linear correlation between the samples.

[0051] After obtaining monitoring data on the normal operating status of the wind turbine generator, the acquisition frequency, accuracy, and data format of monitoring data of different data types are analyzed, the data source is determined and recorded, and the cutting is performed based on the records after data preprocessing.

[0052] The formula for calculating the Pearson correlation coefficient is:

[0053]

[0054] Where PCC is the Pearson correlation coefficient, n is the sample size, and x i and y i Let x and y be the i-th observations of samples x and y, respectively. and y and y are the sample means of samples x and y, respectively.

[0055] The process of calculating the grey relational degree between datasets of different data types includes:

[0056] Obtain multiple monitoring datasets with equal amounts of data of different data types;

[0057] Arrange the data from the same monitoring dataset in chronological order of collection time to form a data column;

[0058] If any data column is designated as the reference data column, then the remaining data columns are designated as comparison data columns. Calculate the grey relational coefficient between the data in each comparison data column and the data in the reference data column.

[0059] Calculate the average value of the grey relational coefficients between the data in each comparison data column and the data in the reference data column. The average value is the grey relational degree between the comparison data column and the reference data column, that is, the grey relational degree between the corresponding monitoring dataset and the other monitoring datasets.

[0060] Repeat the selection and calculation of the reference data columns above to obtain the grey correlation degree between the datasets of different data types.

[0061] The formula for calculating the grey relational coefficient is:

[0062] X0 = (x1, x2, ..., x) n ),

[0063] X i =(x i1 x i2 , ..., x in ), i = 1, 2, ..., m,

[0064] Δ t(i,0) =|x0(t)-x i (t)|,

[0065]

[0066] Where X0 is the reference data column, X i This is a comparison of data columns, where n is the number of data points, m is the number of data types, ρ is the resolution coefficient, and Δ... t(i,0) For each comparison data column X i The absolute difference between the reference data column X0 and the reference data column X0 at time point t, Δ max and Δ min These are the maximum and minimum values ​​of the absolute difference, respectively.

[0067] Step S2: Acquire real-time monitoring data of the wind turbine generator set.

[0068] The monitoring data includes acceleration, tower tilt angle, temperature, and load.

[0069] Step S3: Preprocess and cut the monitoring data to obtain multiple monitoring datasets of equal data volume for different data types, and calculate the Pearson correlation coefficient and the grey relational degree between the monitoring data of different data types.

[0070] Preprocessing includes outlier handling, missing value handling, and error correction.

[0071] The calculation of the Pearson correlation coefficient is the same as the formula written in step 1 above.

[0072] The Pearson correlation coefficient ranges from [-1, 1]. An alarm will be triggered if the calculated Pearson correlation coefficient is greater than or less than the range.

[0073] The calculation process and formula for the grey relational degree between datasets of different data types are the same as those described in step 1 above.

[0074] Step S4: Compare the Pearson correlation coefficient between each monitoring data set and the grey relational degree between each monitoring data set. If the absolute value of the Pearson correlation coefficient between any pair of monitoring data sets is greater than a preset threshold, the absolute value of the difference between the grey relational degree value and the Pearson correlation coefficient is less than a preset deviation value, and both the Pearson correlation coefficient and the grey relational degree are positive, then it is determined that there is a significant positive linear correlation between these two pairs of monitoring data sets. If the absolute value of the Pearson correlation coefficient between any pair of monitoring data sets is greater than a preset threshold, the absolute value of the difference between the grey relational degree value and the Pearson correlation coefficient is less than a preset deviation value, and both the Pearson correlation coefficient and the grey relational degree are negative, then it is determined that there is a significant positive linear correlation between these two pairs of monitoring data sets. If the absolute values ​​of the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both less than the preset threshold and both are positive, then a weak positive linear correlation is determined to exist between these two pairs of monitoring data. If the absolute values ​​of the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both less than the preset threshold and both are negative, then a weak negative linear correlation is determined to exist between these two pairs of monitoring data. If the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both 0, then no linear correlation is determined to exist between these two pairs of monitoring data.

[0075] If the absolute value of the difference between the Pearson correlation coefficient and the grey relational degree between any pair of monitoring data is greater than a preset adjustment value, then the monitoring data is determined to be abnormal and an alarm is triggered.

[0076] Step S5: Record the linear correlation between the monitoring data.

[0077] Step S6: Compare the linear correlation obtained by real-time calculation with the linear correlation of monitoring data of multiple data types under normal working conditions. If the linear correlation between any pair of monitoring data changes, it is determined that the monitoring data is abnormal and an alarm is triggered. If the linear correlation between each monitoring data remains unchanged, it is determined that the monitoring data is valid.

[0078] In this embodiment, a |PCC| greater than 0.8 between any two variables is considered a significant linear correlation. Under normal operating conditions and with accurate monitoring data, if the PCC between two variables is greater than 0.8, a significant positive linear correlation exists; if the PCC is less than -0.8, a significant negative linear correlation exists. If, in a future set of monitoring data, the PCC value between the two variables remains greater than 0.8 or less than -0.8, the data is considered highly valid. If the correlation between the two variables changes, i.e., the calculated PCC is less than 0.8 or greater than -0.8, it indicates that the data for the two variables may be abnormal.

[0079] The following are examples based on real-world application scenarios:

[0080] First, monitoring data from various sensors on the wind turbine generator set were collected. Then, the monitoring data was processed: since different types of data were collected at different frequencies (100Hz for acceleration sensors, 5Hz for tilt sensors, 1Hz for temperature sensors, and 50Hz for load sensors), one data point was taken from every 100 acceleration data points, one data point was taken from every 5 tower tilt data points, and one data point was taken from every 50 load data points to obtain datasets with equal sample sizes. Then, the Pearson correlation coefficients of each variable were calculated to obtain the Pearson correlation coefficient matrix. Finally, a Pearson correlation heatmap was generated. Figure 2 The Pearson correlation heatmap for section three of the tower shows significant positive linear correlations between acceleration 1 and acceleration 2, tilt angle 3 and load 5, tilt angle 3 and load 6, temperature 1 and temperature 2, load 1 and load 2, load 1 and load 3, load 2 and load 3, load 5 and load 6, load 5 and load 7, and load 6 and load 7. Taking acceleration 1 and acceleration 2 in section three of the tower as an example, acceleration 2 is set as the reference data column, and the other variables are used as comparison data columns. Ten data points are taken for each variable as samples to calculate the grey relational coefficient with acceleration 2 as the reference data column. Figure 3 This is a heatmap of grey relational analysis between acceleration 1 and acceleration 2 (reference data column) and various variables in the tower section. The graph shows that the grey relational coefficients for acceleration 1 and acceleration 2 are all 1 across all ten sample data points, indicating a grey relational degree of 1. This is similar to the Pearson correlation coefficient between acceleration 1 and acceleration 2, thus verifying a significant positive linear correlation between them. Significant negative linear correlations exist between acceleration 1 and temperature 1, acceleration 1 and temperature 2, acceleration 2 and temperature 1, acceleration 2 and temperature 2, load 1 and load 5, load 1 and load 6, load 1 and load 7, load 2 and load 5, load 2 and load 6, load 2 and load 7, load 3 and load 5, load 3 and load 6, load 3 and load 7, and load 4 and load 8. Similarly, taking acceleration 1 and temperature 1 as examples, acceleration 1 is the reference data column. Ten data points are taken as samples for each variable to calculate the grey relational coefficient with acceleration 1 as the reference data column. Then, the grey relational degree of each comparison data column is calculated. The grey relational degree of acceleration 1 and temperature 1 is similar to its Pearson correlation coefficient, both being less than 0.8. Therefore, it can be verified that there is a significant negative linear correlation between acceleration 1 and temperature 1. Figure 4The Pearson correlation heatmap for the three acceleration data points in the tower section shows an anomaly. The graph reveals that the Pearson correlation coefficient between acceleration 1 and acceleration 2 decreased from 1 to 0.68. Acceleration 1 and acceleration 2 changed from a pair of variables with a significant positive linear correlation to a pair of variables with a negative correlation. Similarly, using acceleration 2 as the reference data column and the other variables as comparison data columns, ten data points were taken from each variable to calculate the grey relational coefficient with acceleration 2 as the reference data column. (See...) Figure 5 . Figure 5 and Figure 3 The comparison shows that the grey relational coefficient also decreased significantly, and the grey relational degree decreased from 1 to 0.6834, which is similar to the data of the Pearson correlation coefficient. This verifies the calculation results of the Pearson correlation coefficient, proving that acceleration 1 and acceleration 2, which were originally strongly correlated variables, have become weakly correlated, while the correlation of the other strongly correlated variables remains basically unchanged. Therefore, it can be judged that there is an anomaly in the data of acceleration 2 in this set of monitoring data, and the data of the other strongly correlated variables are highly effective.

[0081] In summary, if the correlation coefficients between strongly correlated variables in the monitoring data fluctuate little, and the grey relational degree values ​​are similar to the Pearson correlation coefficient values, the data can be considered to have high validity. Conversely, if the correlation coefficients between certain variables deviate significantly, such as a strong linear correlation becoming a weak linear correlation, or a positive linear correlation becoming a negative linear correlation, it indicates that the data set may contain anomalies. This innovative method can help engineers better understand the intrinsic relationships between various parameters of wind turbine generators, establish accurate validity assessment models, and achieve validity evaluation of wind turbine generator monitoring data.

[0082] Example 2

[0083] Based on Embodiment 1, this embodiment provides an electronic device, including: one or more processors and a memory, wherein the memory stores one or more programs, the one or more programs including instructions for executing the aforementioned method for evaluating the validity of wind turbine generator monitoring data.

[0084] At the hardware level, the electronic device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into memory and then runs it to implement the aforementioned method for evaluating the validity of wind turbine generator monitoring data. Of course, in addition to software implementation, this invention does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0085] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0086] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0087] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for evaluating the validity of monitoring data from wind turbine generator sets, characterized in that, The method includes: Obtain the linear correlation between various data types of monitoring data under normal operating conditions of wind turbine generator sets; Real-time acquisition of monitoring data from wind turbine generators; The monitoring data is preprocessed and cropped to obtain multiple monitoring datasets of equal data volume of different data types. The Pearson correlation coefficient and the grey relational degree between the monitoring data of different data types are calculated. Compare the Pearson correlation coefficients between different monitoring data sets and the grey relational degree between different monitoring datasets. If the absolute value of the Pearson correlation coefficient between any pair of monitoring data sets is greater than a preset threshold, the absolute value of the difference between the grey relational degree value and the Pearson correlation coefficient is less than a preset deviation value, and both the Pearson correlation coefficient and the grey relational degree are positive, then it is determined that there is a significant positive linear correlation between these two pairs of monitoring data sets. If the absolute value of the Pearson correlation coefficient between any pair of monitoring data sets is greater than a preset threshold, the absolute value of the difference between the grey relational degree value and the Pearson correlation coefficient is less than a preset deviation value, and both the Pearson correlation coefficient and the grey relational degree are negative, then it is determined that there is a significant positive linear correlation between these two pairs of monitoring data sets. If there is a significant negative linear correlation between any two pairs of monitoring data, and the absolute values ​​of the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both less than the preset threshold and both the Pearson correlation coefficient and the grey relational degree are positive, then it is determined that there is a weak positive linear correlation between these two pairs of monitoring data. If the absolute values ​​of the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both less than the preset threshold and both the Pearson correlation coefficient and the grey relational degree are negative, then it is determined that there is a weak negative linear correlation between these two pairs of monitoring data. If the Pearson correlation coefficient and the grey relational degree between any two pairs of monitoring data are both 0, then it is determined that there is no linear correlation between these two pairs of monitoring data. Record the linear correlation between each pair of monitoring data. The linear correlation obtained by real-time calculation is compared with the linear correlation of monitoring data of various data types under normal working conditions. If the linear correlation between any pair of monitoring data changes, it is determined that the monitoring data is abnormal and an alarm is triggered. If the linear correlation between each monitoring data remains unchanged, the monitoring data is determined to be valid. The monitoring data includes acceleration, tower tilt angle, temperature, and load; The preprocessing includes outlier handling, missing value handling, and erroneous data correction; The formula for calculating the Pearson correlation coefficient is as follows: ， Where PCC is the Pearson correlation coefficient. n For sample size, and Samples x and samples y The i One observation value, and Samples and samples The sample mean; The range of values ​​for the Pearson correlation coefficient is: If the calculated Pearson correlation coefficient is greater than or less than the range, an alarm will be triggered. The calculation process for the grey relational degree between the datasets of different data types includes: Obtain multiple monitoring datasets with equal amounts of data of different data types; Arrange the data from the same monitoring dataset in chronological order of collection time to form a data column; If any data column is designated as the reference data column, then the remaining data columns are designated as comparison data columns. Calculate the grey relational coefficient between the data in each comparison data column and the data in the reference data column. Calculate the average value of the grey relational coefficients between the data in each comparison data column and the data in the reference data column. The average value is the grey relational degree between the comparison data column and the reference data column, that is, the grey relational degree between the corresponding monitoring dataset and the other monitoring datasets. Repeat the selection and calculation of the reference data columns above to obtain the grey correlation degree between the datasets of different data types; The formula for calculating the grey relational coefficient is: X 0 = ( x 1， x 2，…， x n )， X i = ( x i1 ， x i2 ，…， x in )， i = 1，2，…， m ， ， ， in, X 0 is the reference data column. X i These are comparison data columns. n It's the amount of data. m It refers to the amount of data type. ρ It is the resolution coefficient. For each comparison data column X i With reference data column X 0 at time point t The absolute difference, Δ max and Δ min These are the maximum and minimum values ​​of the absolute difference, respectively.

2. The method for evaluating the validity of monitoring data of wind turbine generator sets according to claim 1, characterized in that, If the absolute value of the difference between the Pearson correlation coefficient and the grey relational degree between any pair of monitoring data is greater than a preset adjustment value, then the monitoring data is determined to be abnormal and an alarm is triggered.

3. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method for evaluating the validity of wind turbine generator monitoring data as described in any one of claims 1-2.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the method for evaluating the validity of wind turbine generator monitoring data as described in any one of claims 1-2.

Images