Methods, systems, equipment, and media for troubleshooting inefficiently operating wind turbine units in wind farms.

Abstract

This invention provides a method, system, computer equipment, and medium for identifying inefficiently operating wind turbines in wind farms. By acquiring periodic statistical data of SCADA operation data for all wind turbines within the wind farm, and adding the geographical coordinates of each wind turbine, the operating data of the wind turbine to be analyzed is obtained. Then, essential and necessary variables other than power generation are extracted from the operating data of the wind turbine to be analyzed, and combined to obtain corresponding multidimensional feature vectors. Cluster analysis is then performed on the multidimensional feature vectors of all wind turbines in each statistical period to obtain corresponding clusters. Inefficiency identification labels are added to the wind turbines in each cluster, and the sum of the inefficiency identification labels for each wind turbine in all statistical periods is used as the inefficiency judgment value. This technical solution for identifying inefficiently operating wind turbines comprehensively considers many environmental and control factors, efficiently and accurately identifying inefficiently operating wind turbines, thereby effectively ensuring the stable operation of the wind farm.

Description

Technical Field

[0001] This invention relates to the field of new energy wind power technology, and in particular to a method, system, computer equipment, and storage medium for troubleshooting inefficiently operating wind turbine generators in wind farms. Background Technology

[0002] Against the backdrop of vigorous development of new energy sources, with the expansion of wind farms and the increase in the number of wind turbines, the amount of operational data generated by wind turbines has also experienced explosive growth. However, among all operating wind turbines in a wind farm, turbines of the same model may experience reduced power generation efficiency due to factors such as blade contamination, suboptimal mechanical equipment, or inherent defects in the equipment itself, thus affecting the normal operation of the entire wind farm. Therefore, for wind power operators, it is crucial to quickly and accurately identify inefficient wind turbines in the wind farm through data analysis and improve their power generation performance.

[0003] Because the local environment of each operating wind turbine in a wind farm varies, such as wind speed, terrain, wind direction, wind shear, turbulence, temperature, and humidity, and because control differences exist due to factors such as power curtailment and farm group control, and because these factors lead to different operating conditions for each wind turbine, it is inaccurate to simply determine the power generation efficiency of a wind turbine based on its power curve. At the same time, environmental and control differences also make it difficult to accurately identify inefficient wind turbines.

[0004] Existing technical methods for identifying inefficient wind turbines mainly include: using a small number of selected wind turbine operating data variables to calculate the power curve of a single wind turbine and identifying inefficient wind turbines through horizontal comparison; or, using grid-connected data and power curtailment data to filter out normally grid-connected operating data of turbines, calculating the average wind speed and power generation of each turbine, and identifying turbines with lower power generation under the same standardized wind speed conditions. However, in the face of large-scale data application scenarios, existing conventional methods are not only inefficient in processing, but also fail to fully consider the differences in the local environment and active control of each operating wind turbine in complex wind farm environments, resulting in unreliable identification of inefficient wind turbines.

[0005] Therefore, there is an urgent need to provide a method for identifying inefficiently operating wind turbines that can comprehensively consider numerous environmental and control factors and variables. Summary of the Invention

[0006] The purpose of this invention is to provide a method for identifying inefficiently operating wind turbines in wind farms. This method involves adding the geographical coordinates of the operating wind turbines to the turbine operation data, creating wind turbine operation data that includes geographical location information variables. Then, the corresponding multi-dimensional feature vectors are extracted, and cluster analysis and inefficiency identification labels are assigned. The inefficiency determination value of each wind turbine is used as the basis for identifying inefficient wind turbines. This method solves the problem of low data processing efficiency in the inefficiency identification process of existing technologies, and further improves the accuracy of identifying inefficiently operating wind turbines by comprehensively considering various environmental and control factors.

[0007] To achieve the above objectives, it is necessary to provide a method, system, computer equipment, and storage medium for troubleshooting inefficiently operating wind turbine units in wind farms, addressing the aforementioned technical problems.

[0008] In a first aspect, embodiments of the present invention provide a method for troubleshooting inefficiently operating wind turbine generators in wind farms, the method comprising the following steps:

[0009] Acquire SCADA operation data and geographic plane coordinates of all wind turbine units within the wind farm; the SCADA operation data is wind turbine unit operation data collected according to a preset frequency and preset duration;

[0010] According to the preset statistical period, calculate the periodic statistical data corresponding to the SCADA operation data of each wind turbine.

[0011] Add the geographic plane coordinates of each wind turbine to the corresponding periodic statistics data to obtain the operating data of the wind turbine to be analyzed.

[0012] Extract the essential and necessary variables other than power generation from the wind turbine operation data to be analyzed, and combine the essential and necessary variables to obtain the corresponding multidimensional feature vector;

[0013] Cluster analysis was performed on the multidimensional feature vectors of all wind turbines in each statistical period to obtain the corresponding clusters, and inefficient identification labels were added to the wind turbines in each cluster.

[0014] The sum of the inefficiency identification tags of each wind turbine in all statistical periods is counted to obtain the corresponding inefficiency judgment value, and inefficient wind turbines are identified based on the inefficiency judgment value.

[0015] Furthermore, the periodic statistical data includes the maximum, minimum, average, and standard deviation of each SCADA operating data within each statistical period.

[0016] Furthermore, the step of extracting the essential and necessary variables other than power generation from the wind turbine operating data to be analyzed includes:

[0017] Variables directly related to power generation in the wind turbine operation data to be analyzed are selected as mandatory variables; the mandatory variables include nacelle wind speed, nacelle wind direction, pitch angle, generator speed, generator torque, power curtailment signal, fault signal, and grid connection signal.

[0018] Calculate the Pearson correlation coefficients between the other variables (excluding the mandatory variables) and the power generation in the operating data of the wind turbine to be analyzed, and select the necessary variables based on the Pearson correlation coefficients.

[0019] Furthermore, the step of selecting the necessary variables based on the Pearson correlation coefficient includes:

[0020] Calculate the absolute value of each Pearson correlation coefficient and select variables whose absolute value is greater than a preset threshold as necessary variables.

[0021] Furthermore, the step of performing cluster analysis on the multidimensional feature vectors of all wind turbine units in each statistical period to obtain the corresponding clusters includes:

[0022] The number of clusters is determined in advance based on the geographical plane coordinates of each wind turbine unit;

[0023] Based on the number of clusters, the K-means clustering algorithm is used to perform cluster analysis on the multidimensional feature vectors of all wind turbine units in each statistical period.

[0024] Furthermore, the step of adding inefficient identification tags to wind turbines in each cluster includes:

[0025] The statistical period statistics include the total number of periods and the number of wind turbine units within each cluster. Based on the total number of periods and the number of wind turbine units within each cluster, a numerical label sequence corresponding to each cluster is generated. The numerical label sequence is a sequence where the first term is 0, the last term is the ratio of the number of wind turbine units within the cluster to the total number of periods, and the number of terms is an arithmetic sequence of the number of wind turbine units within the corresponding cluster.

[0026] The wind turbines in each cluster are sorted in descending order according to their corresponding power generation, and inefficiency identification labels are added sequentially from top to bottom according to the first item to the last item of the corresponding numerical label sequence.

[0027] Furthermore, the step of identifying inefficiently operating wind turbine units based on the inefficiency determination value includes:

[0028] The inefficiency judgment values ​​of each wind turbine are sorted in descending order, and a preset proportion of inefficiency judgment values ​​are selected from top to bottom to determine the corresponding wind turbine as an inefficient wind turbine.

[0029] Secondly, embodiments of the present invention provide a system for troubleshooting inefficiently operating wind turbine generators in wind farms, the system comprising:

[0030] The data acquisition module is used to acquire SCADA operation data and geographic plane coordinates of all wind turbine units in the wind farm; the SCADA operation data is wind turbine unit operation data acquired according to a preset frequency and preset duration;

[0031] The statistical analysis module is used to calculate the periodic statistical data corresponding to the SCADA operation data of each wind turbine according to the preset statistical period.

[0032] The location addition module is used to add the geographic plane coordinates of each wind turbine to the corresponding periodic statistics data to obtain the operating data of the wind turbine to be analyzed.

[0033] The variable extraction module is used to extract the mandatory and necessary variables other than power generation from the wind turbine operation data to be analyzed, and to combine the mandatory and necessary variables to obtain the corresponding multidimensional feature vector.

[0034] The clustering analysis module is used to perform clustering analysis on the multidimensional feature vectors of all wind turbines in each statistical period to obtain the corresponding clusters, and add inefficient identification labels to the wind turbines in each cluster.

[0035] The inefficiency identification module is used to calculate the sum of inefficiency identification tags for each wind turbine in all statistical periods, obtain the corresponding inefficiency judgment value, and identify inefficient wind turbines based on the inefficiency judgment value.

[0036] Thirdly, embodiments of the present invention also provide a computer device, including 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 above-described method.

[0037] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0038] The present application provides a method, system, computer equipment, and storage medium for identifying inefficiently operating wind turbines in a wind farm. The method involves acquiring preset duration SCADA operation data and geographic coordinates of all wind turbines in the wind farm; calculating periodic statistical data corresponding to the SCADA operation data of each wind turbine according to a preset statistical period; adding the geographic coordinates of each wind turbine to the corresponding periodic statistical data to obtain the wind turbine operation data to be analyzed; extracting essential and necessary variables (excluding power generation) from the wind turbine operation data to be analyzed, combining them to obtain corresponding multidimensional feature vectors; performing cluster analysis on the multidimensional feature vectors of all wind turbines in each statistical period to obtain corresponding clusters; adding inefficiency identification labels to the wind turbines in each cluster; and summing the inefficiency identification labels of each wind turbine in all statistical periods to obtain the corresponding inefficiency judgment value, thereby identifying inefficiently operating wind turbines. Compared with existing technologies, this method for identifying inefficient wind turbines in wind farms requires a single data source, is simple in method, and is suitable for big data application scenarios. It is not only low in cost, but also solves the problem of low data processing efficiency in the process of identifying inefficient units. Furthermore, by comprehensively considering many environmental and control factors, it can further improve the accuracy of identifying inefficient wind turbines, and provide a reliable and effective guarantee for the stable operation of wind farms. Attached Figure Description

[0039] Figure 1 This is a schematic diagram illustrating the application scenario of the method for investigating inefficiently operating wind turbine units in wind farms, as described in this embodiment of the invention.

[0040] Figure 2 This is a flowchart illustrating the method for investigating inefficiently operating wind turbine units in a wind farm, as described in this embodiment of the invention.

[0041] Figure 3 This is a schematic diagram of the structure of the wind turbine troubleshooting system for inefficient wind farm operation in an embodiment of the present invention;

[0042] Figure 4 This is an internal structural diagram of the computer device in an embodiment of the present invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and beneficial effects of this application clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Obviously, the embodiments described below are only part of the embodiments of the present invention and are used to illustrate the present invention, but are not intended to limit the scope of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0044] The method for troubleshooting inefficiently operating wind turbine units in wind farms provided by this invention can be applied to, for example... Figure 1 The terminal or server shown. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be a standalone server or a server cluster composed of multiple servers. This invention is a method for identifying inefficient wind turbines based on SCADA (Supervisory Control and Data Acquisition) system data collected from wind turbine control, operation, and surrounding environment. The SCADA system typically records hundreds of sampling points at different frequencies, such as 1 second or several seconds. For example, the server can acquire operating data of each wind turbine in the wind farm collected by the SCADA system at a preset frequency for a preset duration, and combine this data with the geographical location information of each wind turbine. The method of this invention is used to identify and investigate inefficient wind turbines in any wind farm. The results are then used for further research by the server or transmitted to the terminal for user access. The following embodiments will provide a detailed description of the wind turbine inefficient operation investigation method of this invention.

[0045] In one embodiment, such as Figure 2 As shown, a method for troubleshooting inefficiently operating wind turbine units in a wind farm is provided, including the following steps:

[0046] S11. Obtain SCADA operation data and geographic plane coordinates of all wind turbines in the wind farm; the SCADA operation data is wind turbine operation data collected according to a preset frequency and preset duration; wherein, as mentioned above, the wind turbine operation data is the original SCADA system data collected, including wind turbine control, operation, and surrounding environment data, and will be processed and filtered as needed to obtain relevant data that is helpful for identifying and analyzing inefficient wind turbines; it should be noted that the preset frequency and preset duration of wind turbine operation data collection can be selected according to actual needs, such as obtaining annual or monthly collection frequency f of 1Hz operation data, and no specific limitation is made here;

[0047] S12. Calculate the periodic statistical data corresponding to the SCADA operation data of each wind turbine according to the preset statistical period. The periodic statistical data includes the maximum, minimum, average, and standard deviation of each SCADA operation data within each statistical period. The preset statistical period can be reasonably selected according to actual application needs, as well as the preset frequency and preset duration of the acquired SCADA operation data. If the preset duration is 1 month, the preset frequency is 1Hz, and the preset statistical period is 10 minutes, then the SCADA operation data of each wind turbine for one month needs to be divided into multiple periods in 10-minute intervals. Existing mathematical statistical methods are used to perform statistical analysis on the data within each 10-minute interval to obtain the maximum, minimum, average, and standard deviation of each variable in the SCADA operation data within each 10-minute interval for subsequent analysis. It should be noted that the periodic statistical data includes not only the maximum, minimum, average, and standard deviation of each SCADA operation data within the corresponding period, but also the number of each wind turbine and the start time of the corresponding statistical period, which facilitates subsequent cluster analysis of the operation data of each turbine.

[0048] S13. Add the geographic plane coordinates of each wind turbine to the corresponding periodic statistical data to obtain the wind turbine operation data to be analyzed. The geographic plane coordinates can be understood as the location information of the wind turbine. Each wind turbine has two values: the horizontal coordinate and the vertical coordinate. Add them to the periodic statistical data obtained above so that each periodic statistical data of each wind turbine contains the corresponding geographic location information, which is convenient for determining the corresponding number of clusters in subsequent cluster analysis.

[0049] S14. Extract the mandatory and necessary variables other than power generation from the wind turbine operating data to be analyzed, and combine the mandatory and necessary variables to obtain the corresponding multidimensional feature vector; wherein, the multidimensional feature vector can be understood as a row vector that simultaneously includes mandatory and necessary variables; specifically, the step of extracting the mandatory and necessary variables other than power generation from the wind turbine operating data to be analyzed includes:

[0050] Variables directly related to power generation in the wind turbine operation data to be analyzed are selected as mandatory variables; the mandatory variables include nacelle wind speed, nacelle wind direction, pitch angle, generator speed, generator torque, power curtailment signal, fault signal, and grid connection signal.

[0051] Calculate the Pearson correlation coefficients between the power generation and all variables other than the mandatory variables in the wind turbine operating data to be analyzed, and then select the necessary variables based on the Pearson correlation coefficients. Specifically, the step of selecting the necessary variables based on the Pearson correlation coefficients includes:

[0052] Calculate the absolute value of each Pearson correlation coefficient and select variables whose absolute value is greater than a preset threshold as necessary variables. The preset threshold can be determined according to actual needs and is not specifically limited here. For example, if the preset threshold is 0.6, that is, if the absolute value of the correlation coefficient between a certain variable and power generation calculated according to the Pearson correlation coefficient formula is greater than 0.6, then the variable is considered a necessary variable and can be used for subsequent cluster analysis.

[0053] S15. Perform cluster analysis on the multidimensional feature vectors of all wind turbines in each statistical period to obtain the corresponding clusters, and add inefficient identification labels to the wind turbines in each cluster. In principle, cluster analysis can be implemented using any existing clustering algorithm. However, in order to better fit the characteristics of real wind farm data and ensure the efficiency of cluster analysis, this embodiment preferably uses the K-means clustering algorithm to perform cluster analysis on the multidimensional feature vectors of wind turbines.

[0054] Specifically, the step of performing cluster analysis on the multidimensional feature vectors of all wind turbine units in each statistical period to obtain the corresponding clusters includes:

[0055] Based on the geographical plane coordinates of each wind turbine, the number of clusters is determined in advance; among them, choosing an appropriate number of clusters K is crucial when executing the K-means clustering algorithm. Existing methods for selecting the K value include: (1) Simple setting method, that is, directly using the value obtained by dividing the sample size n by 2 and then taking the square root as the K value; (2) Elbow method, when the selected K value is less than the actual number of clusters, the cost value will decrease significantly as the K value increases; when the selected K value is greater than the actual number of clusters, the cost value will not change so significantly as the K value increases, and the correct K value is at this inflection point; (3) Interval statistics method, by randomly generating the same number of random samples as the original sample size in the rectangular area (or cubic area in high dimension) where the sample is located according to a uniform distribution, and performing K-Means clustering on this random sample, thereby obtaining the number of sample points within a class. The distance Dk between the samples is collected repeatedly, and a suitable measure is introduced as the interval measure Gapk. The Monte Carlo method is used to find the K value that makes Gapk reach its maximum value, which is the optimal number of clusters. (4) Silhouette coefficient method: By calculating the average distance from the sample point to other samples in the same cluster and the average distance from the sample point to all samples in other clusters, the similarity between the sample and its cluster is measured, i.e., cohesion. The optimal number of clusters is selected as the number of clusters with high cohesion of all sample points. (5) Canopy algorithm: By using coarse clustering in advance, the initial number of clusters and cluster centers are determined for the K-means algorithm. All five methods can be used to select the number of clusters, but each has its best applicable scenario. Therefore, in actual application, users can select the method to determine the number of clusters according to the existing data scale and specific prediction scenario. In this embodiment, considering real-world wind farms, due to geographical location factors, there is often natural grouping with consistency. Preferably, the number of clusters is predetermined based on the geographical plane coordinates of each wind turbine, i.e., the distance between geographical locations. For example, wind turbines within a preset range are treated as one category. The number of clusters is obtained while ensuring a balanced number of wind turbines of each category, thus achieving a good fit with the K-means algorithm. While ensuring the accuracy of clustering, it also simplifies the workload of cluster analysis to a certain extent.

[0056] Based on the cluster number, the K-means clustering algorithm is used to perform cluster analysis on the multidimensional feature vectors of all wind turbines in each statistical period. Once the cluster number is determined using the above method, cluster analysis can be performed on the multidimensional feature vectors of all wind turbines in each statistical period according to the following steps:

[0057] 1) Based on the determined number of clusters K, randomly select the multidimensional feature vectors of K wind turbine units from each statistical period as the initial cluster centers; assuming there are 100 wind turbine units operating in the wind farm and the preset number of clusters K is 5, then randomly select 5 (K) data points from the original data as the initial cluster centers, and the number of samples in each cluster is not less than 100 / 2K, that is, there are 10 wind turbine units in each cluster;

[0058] 2) Calculate the distance between each multidimensional feature vector and the K cluster centers for each statistical period, and assign each multidimensional feature vector to the nearest cluster point. After clustering, calculate the average value ci for all clusters grouped into the same cluster point, and use this average value as the new cluster center.

[0059]

[0060] Where, N i This refers to the number of sample data points contained in the i-th cluster, where x represents the sample data.

[0061] 3) Repeat the iteration multiple times until certain requirements are met, then stop clustering and label the obtained cluster centers as C = {c1, c2, c3…} to obtain the corresponding cluster families;

[0062] In this embodiment, the K-means clustering algorithm is used to perform clustering analysis on the multidimensional feature vectors of wind turbine units in each statistical period. This not only ensures the accuracy of the clustering results, but also ensures good adaptability when the number of wind turbine units in a large wind farm increases, leading to an increase in the amount of operating data. In other words, it ensures that the method of this invention has strong generalization ability.

[0063] After completing the cluster analysis for all statistical periods through the above steps, it is necessary to add corresponding inefficiency identification labels to the wind turbines in each cluster obtained for each statistical period; specifically, the step of adding inefficiency identification labels to the wind turbines in each cluster includes:

[0064] The system collects the total number of statistical periods and the number of wind turbines within each cluster. Based on these figures, it generates a numerical label sequence corresponding to each cluster. The numerical label sequence begins with 0, ends with the ratio of the number of wind turbines within the cluster to the total number of periods, and the number of terms is an arithmetic progression of the number of wind turbines within the corresponding cluster. For example, if there are n sets of statistical period data for each wind turbine, and m wind turbines within a certain cluster, then an arithmetic progression of 0 to m / n containing m terms is generated as the numerical label sequence. This sequence is used to label each wind turbine in the cluster according to the following steps.

[0065] The wind turbines in each cluster are sorted in descending order of their corresponding power generation, and inefficiency identification labels are added sequentially from top to bottom according to the first item to the last item of the corresponding numerical label sequence. Specifically, it can be understood that the wind turbines in each cluster are assigned larger inefficiency identification labels according to the principle that the lower the power generation, the larger the inefficiency identification label. The labeling is completed according to the numerical label sequence obtained above.

[0066] S16. Calculate the sum of the inefficiency identification tags of each wind turbine unit in all statistical periods to obtain the corresponding inefficiency judgment value, and identify inefficient operating wind turbine units based on the inefficiency judgment value. The high or low inefficiency judgment value does not represent that the power generation efficiency of the corresponding wind turbine unit is the lowest or highest throughout the entire preset time period, but rather indicates the power generation rate under most circumstances. Based on this, this embodiment preferably uses the inefficiency judgment value to determine a preset proportion of wind turbine units as inefficient operating wind turbine units. Specifically, the step of identifying inefficient operating wind turbine units based on the inefficiency judgment value includes:

[0067] The inefficiency judgment values ​​of each wind turbine are sorted in descending order, and a preset proportion of inefficiency judgment values ​​are selected from top to bottom to determine the corresponding wind turbine as an inefficient wind turbine.

[0068] This application provides a method for identifying inefficient wind turbines by statistically analyzing wind turbine operation data with added geographic plane coordinates according to a preset statistical period, extracting multidimensional feature vectors, and then using K-means to cluster the multidimensional feature vectors of all wind turbines in each statistical period. For each cluster, the wind turbines are sorted according to their power generation and assigned an arithmetic progression inefficiency identification label. The sum of the inefficiency identification labels for each wind turbine is then used as the inefficiency judgment value. Based on this, a preset proportion of wind turbines with higher inefficiency judgment values ​​are selected as inefficient operating wind turbines. This method requires a single data source, is simple, and is suitable for big data applications. It is not only low-cost but also highly efficient in the identification process. Furthermore, by comprehensively considering various environmental and control factors, it ensures the accuracy of the identification results, thus providing a reliable guarantee for the stable operation of wind farms.

[0069] It should be noted that although the steps in the flowchart above are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order requirement for the execution of these steps, and they can be executed in other orders.

[0070] In one embodiment, such as Figure 3 As shown, a system for troubleshooting inefficiently operating wind turbine generators in a wind farm is provided. The system includes:

[0071] The data acquisition module is used to acquire SCADA operation data and geographic plane coordinates of all wind turbine units in the wind farm; the SCADA operation data is wind turbine unit operation data acquired according to a preset frequency and preset duration;

[0072] The statistical analysis module is used to calculate the periodic statistical data corresponding to the SCADA operation data of each wind turbine according to the preset statistical period.

[0073] The location addition module is used to add the geographic plane coordinates of each wind turbine to the corresponding periodic statistics data to obtain the operating data of the wind turbine to be analyzed.

[0074] The variable extraction module is used to extract the mandatory and necessary variables other than power generation from the wind turbine operation data to be analyzed, and to combine the mandatory and necessary variables to obtain the corresponding multidimensional feature vector.

[0075] The clustering analysis module is used to perform clustering analysis on the multidimensional feature vectors of all wind turbines in each statistical period to obtain the corresponding clusters, and add inefficient identification labels to the wind turbines in each cluster.

[0076] The inefficiency identification module is used to calculate the sum of inefficiency identification tags for each wind turbine in all statistical periods, obtain the corresponding inefficiency judgment value, and identify inefficient wind turbines based on the inefficiency judgment value.

[0077] Specific limitations regarding the system for troubleshooting inefficient wind turbines in wind farms can be found in the limitations of the method for troubleshooting inefficient wind turbines in wind farms described above, and will not be repeated here. Each module in the aforementioned system for troubleshooting inefficient wind turbines in wind farms can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0078] Figure 4 An internal structural diagram of a computer device is shown in one embodiment. This computer device may specifically be a terminal or a server. Figure 4As shown, the computer device includes a processor, memory, network interface, display, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a method for troubleshooting inefficiently operating wind turbine units in a wind farm. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0079] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computing devices may include more or fewer components than those shown in the figure, or combine certain components, or have the same component arrangement.

[0080] In one embodiment, a computer device is provided, including 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 method described above.

[0081] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0082] In summary, the present invention provides a method, system, computer equipment, and storage medium for identifying inefficiently operating wind turbines in a wind farm. The method involves acquiring preset-duration SCADA operating data and geographic coordinates of all wind turbines within the wind farm; calculating periodic statistical data corresponding to the SCADA operating data of each wind turbine according to a preset statistical period; adding the geographic coordinates of each wind turbine to the corresponding periodic statistical data to obtain the operating data of the wind turbine to be analyzed; and extracting essential and necessary variables (excluding power generation) from the operating data of the wind turbine to be analyzed, combining them to obtain corresponding multidimensional feature vectors, and then applying them to... This method involves clustering the multidimensional feature vectors of all wind turbines within each statistical period to obtain corresponding clusters. After adding inefficiency identification labels to the wind turbines in each cluster, the sum of the inefficiency identification labels for each wind turbine across all statistical periods is calculated to obtain the corresponding inefficiency judgment value. This method identifies inefficiently operating wind turbines based on a single data source, is simple, and is suitable for big data applications. It is not only low-cost to implement but also solves the problem of low data processing efficiency in the process of identifying inefficient turbines. Furthermore, by comprehensively considering various environmental and control factors, it can further improve the accuracy of identifying inefficiently operating wind turbines, thereby effectively ensuring the stable operation of wind farms.

[0083] The various embodiments in this specification are described in a progressive manner. For directly identical or similar parts of the embodiments, refer to each other. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. It should be noted that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification.

[0084] The embodiments described above are merely preferred embodiments of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various improvements and substitutions without departing from the technical principles of this invention, and these improvements and substitutions should also be considered within the scope of protection of this application. Therefore, the scope of protection of this patent application should be determined by the scope of the claims.

Claims

1. A method for troubleshooting inefficiently operating wind turbine units in a wind farm, characterized in that, The method includes the following steps: Acquire SCADA operation data and geographic plane coordinates of all wind turbine units within the wind farm; the SCADA operation data is wind turbine unit operation data collected according to a preset frequency and preset duration; According to the preset statistical period, the periodic statistical data corresponding to the SCADA operation data of each wind turbine are calculated respectively; the periodic statistical data includes the maximum value, minimum value, average value and standard deviation of each SCADA operation data in each statistical period; Add the geographic plane coordinates of each wind turbine to the corresponding periodic statistics data to obtain the operating data of the wind turbine to be analyzed. Extract the essential and necessary variables other than power generation from the wind turbine operation data to be analyzed, and combine the essential and necessary variables to obtain the corresponding multidimensional feature vector; Cluster analysis is performed on the multidimensional feature vectors of all wind turbines within each statistical period to obtain corresponding clusters, and inefficient identification labels are added to the wind turbines in each cluster; the number of clusters is determined based on the geographical planar coordinates of each wind turbine; the step of adding inefficient identification labels to the wind turbines in each cluster includes: The statistical period statistics include the total number of periods and the number of wind turbine units within each cluster. Based on the total number of periods and the number of wind turbine units within each cluster, a numerical label sequence corresponding to each cluster is generated. The numerical label sequence is a sequence where the first term is 0, the last term is the ratio of the number of wind turbine units within the cluster to the total number of periods, and the number of terms is an arithmetic sequence of the number of wind turbine units within the corresponding cluster. The wind turbines in each cluster are sorted in descending order according to their corresponding power generation. In accordance with the principle that the lower the power generation, the larger the inefficiency identification label is assigned, inefficiency identification labels are added sequentially from top to bottom according to the first item to the last item of the corresponding numerical label sequence. The sum of the inefficiency identification tags of each wind turbine in all statistical periods is counted to obtain the corresponding inefficiency judgment value, and inefficient wind turbines are identified based on the inefficiency judgment value.

2. The method for investigating inefficiently operating wind turbine units in a wind farm as described in claim 1, characterized in that, The steps for extracting the essential and necessary variables, excluding power generation, from the operating data of the wind turbine to be analyzed include: Variables directly related to power generation in the wind turbine operation data to be analyzed are selected as mandatory variables; the mandatory variables include nacelle wind speed, nacelle wind direction, pitch angle, generator speed, generator torque, power curtailment signal, fault signal, and grid connection signal. Calculate the Pearson correlation coefficients between the other variables (excluding the mandatory variables) and the power generation in the operating data of the wind turbine to be analyzed, and select the necessary variables based on the Pearson correlation coefficients.

3. The method for investigating inefficiently operating wind turbine units in a wind farm as described in claim 2, characterized in that, The step of selecting the necessary variables based on the Pearson correlation coefficient includes: Calculate the absolute value of each Pearson correlation coefficient and select variables whose absolute value is greater than a preset threshold as necessary variables.

4. The method for investigating inefficiently operating wind turbine units in a wind farm as described in claim 1, characterized in that, The step of performing cluster analysis on the multidimensional feature vectors of all wind turbine units in each statistical period to obtain the corresponding clusters includes: Based on the number of clusters, the K-means clustering algorithm is used to perform cluster analysis on the multidimensional feature vectors of all wind turbine units in each statistical period.

5. The method for investigating inefficiently operating wind turbine units in a wind farm as described in claim 1, characterized in that, The step of identifying inefficiently operating wind turbine units based on the inefficiency determination value includes: The inefficiency judgment values ​​of each wind turbine are sorted in descending order, and a preset proportion of inefficiency judgment values ​​are selected from top to bottom to determine the corresponding wind turbine as an inefficient wind turbine.

6. A system for troubleshooting inefficiently operating wind turbine units in a wind farm, characterized in that, The system, employing the method for troubleshooting inefficiently operating wind turbine generators in wind farms as described in claim 1, comprises: The data acquisition module is used to acquire SCADA operation data and geographic plane coordinates of all wind turbine units in the wind farm; the SCADA operation data is wind turbine unit operation data acquired according to a preset frequency and preset duration; The statistical analysis module is used to calculate the periodic statistical data corresponding to the SCADA operation data of each wind turbine according to the preset statistical period. The location addition module is used to add the geographic plane coordinates of each wind turbine to the corresponding periodic statistics data to obtain the operating data of the wind turbine to be analyzed. The variable extraction module is used to extract the mandatory and necessary variables other than power generation from the wind turbine operation data to be analyzed, and to combine the mandatory and necessary variables to obtain the corresponding multidimensional feature vector. The clustering analysis module is used to perform clustering analysis on the multidimensional feature vectors of all wind turbines in each statistical period to obtain the corresponding clusters, and add inefficient identification labels to the wind turbines in each cluster. The inefficiency identification module is used to calculate the sum of inefficiency identification tags for each wind turbine in all statistical periods, obtain the corresponding inefficiency judgment value, and identify inefficient wind turbines based on the inefficiency judgment value.

7. A computer 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 according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images