Data processing method, data processing device, medium, electronic equipment and product

Abstract

The present disclosure relates to a data processing method, a data processing device, a medium, an electronic device and a product, which are applied to a wind turbine generator set. The method includes: obtaining a current operation scatter plot of the wind turbine generator set, the current operation scatter plot is used to characterize the output power at different wind speeds; inputting the current operation scatter plot into a pre-trained anomaly detection model to obtain target anomaly information output by the anomaly detection model, the target anomaly information including a target anomaly area and / or a target anomaly type. In this way, the target anomaly information in the current operation scatter plot is determined by the current operation scatter plot of the wind turbine generator set and the anomaly detection model, and the anomaly information in the current operation scatter plot of the wind turbine generator set can be quickly determined, thereby effectively improving the accuracy and efficiency of determining the abnormal situation of the wind turbine generator set, and further effectively improving the overall performance of the wind turbine generator set.

Description

Technical Field

[0001] The present disclosure relates to the technical field of wind turbines, and in particular to a data processing method, a data processing device, a medium, an electronic device and a product. Background Art

[0002] The power curve of a wind turbine reflects the corresponding relationship between the output power of the unit and the change in wind speed. It is not only a necessary design basis for characterizing the output performance of a wind turbine, but also an important evaluation indicator for measuring wind power generation capacity. The manufacturer of a wind turbine will submit the theoretical power curve of the wind turbine when providing the equipment. However, after the wind turbine is actually put into operation, as the operating time increases, the actual power curve of the wind turbine may deviate from the theoretical power curve.

[0003] In the prior art, the actual power curve of the wind turbine generator set is compared with the theoretical power curve, and the existence and type of abnormality of the wind turbine generator set are judged by manual reasoning. However, in large-scale screening application scenarios, the manual reasoning judgment method is inefficient. In addition, in large-scale screening application scenarios, there may also be a problem of reduced judgment accuracy due to fatigue caused by high-intensity work of the human brain. Summary of the invention

[0004] In order to overcome the problems existing in the related art, the present disclosure provides a data processing method, a data processing device, a medium, an electronic device and a product.

[0005] According to a first aspect of an embodiment of the present disclosure, there is provided a data processing method, which is applied to a wind turbine generator set, and the method comprises: Obtaining a current operation scatter plot of the wind turbine generator set, wherein the current operation scatter plot is used to characterize output power at different wind speeds; The current running scatter plot is input into a pre-trained anomaly detection model to obtain target anomaly information output by the anomaly detection model, where the target anomaly information includes a target anomaly region and / or a target anomaly type.

[0006] Optionally, the training process of the anomaly detection model includes: Acquire a training sample data set, wherein the training sample data set includes abnormal power scatter plot sample data, abnormal area annotation data and abnormal type annotation data of a plurality of wind turbine generator sets; The training sample data set is used as training data to perform model training on a preset initial model to obtain the anomaly detection model.

[0007] Optionally, the acquiring of a training sample data set includes abnormal power scatter plot sample data of a plurality of wind turbine generator sets, abnormal region annotation data and abnormal type annotation data, including: Acquire a plurality of historical abnormal scatter plots of the wind turbine generator sets, wherein the historical abnormal scatter plots include abnormal conditions of power output of the wind turbine generator sets at different wind speeds within a historical time period; Annotating the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set; The abnormal data set is used as a training sample data set.

[0008] Optionally, the historical anomaly scatter plot is a two-dimensional scatter plot, a first coordinate axis of the historical anomaly scatter plot represents wind speed, and a second coordinate axis of the historical anomaly scatter plot represents power, and annotating the abnormal area and abnormal type of each of the historical anomaly scatter plots to obtain the abnormal data set includes: Obtaining a standard power scatter plot, wherein the standard power scatter plot is used to characterize the rated power output by the wind turbine generator set under different wind speed conditions; Determining a target deviation between the output power of the wind turbine generator set and the rated power at the same wind speed according to the standard power scatter plot and the historical abnormal scatter plot; Taking a plurality of scattered points whose target deviation is greater than a preset deviation threshold as a first target scattered point set; Marking the area where the first target scattered point set is located as a first abnormal area; The abnormality type corresponding to the first abnormal area is marked as a poor performance abnormality.

[0009] Optionally, marking the area where the first target scattered point set is located as a first abnormal area further includes: When it is determined that there is a first scattered point in the first target scattered point set that exceeds a preset first power threshold, an area where a plurality of second scattered points in the first target scattered point set whose output powers are less than or equal to the output power of the first scattered point except the first scattered point are located is marked as a first abnormal area.

[0010] Optionally, the marking of the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set includes: Determine a second target scatter point set from the historical abnormal scatter point graph, wherein the second target scatter point set includes a plurality of scatter points constituting one or more line segments, and is used to characterize abnormal scatter points of the wind turbine generator set outputting a first fixed power at different wind speeds; taking multiple scatter points in the historical abnormal scatter diagram that are less than the first fixed power as third scatter points, wherein the third scatter points are used to represent scatter points where the target deviation between the output power of the wind turbine generator set at different wind speeds and the rated power of the wind turbine generator set is less than a preset deviation threshold; Marking the area where the second target scattered point set and the plurality of third scattered points are located as a second abnormal area; The abnormality type corresponding to the second abnormal area is marked as power-limiting abnormality.

[0011] Optionally, the marking of the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set includes: Determine a third target scatter point set from the historical abnormal scatter point diagram, the third target scatter point set including scatter points where the output power of the wind turbine generator set at a first wind speed is higher than a first standard power in a preset standard power scatter point diagram, and scatter points where the output power of the wind turbine generator set at a second wind speed is lower than a second standard power in the standard power scatter point diagram, the first wind speed being lower than the second wind speed; In a case where it is determined that the output power of the fourth scattered point in the third target scattered point set is greater than the preset second power threshold for the first time, if it is determined that the third target scattered point set, the fourth scattered point and the standard power scatter point diagram form a triangular area or an irregular closed area, the triangular area or the irregular closed area is marked as a third abnormal area; The abnormality type corresponding to the third abnormal area is marked as left-biased abnormality.

[0012] Optionally, the marking of the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set includes: Determine a fourth target scatter point set from the historical abnormal scatter point graph, wherein the fourth target scatter point set includes a plurality of scatter points constituting a plurality of similar curves, and is used to characterize that the wind turbine generator set has a plurality of abnormal scatter points with different output powers at the same wind speed; The area where the fourth target scattered point set is located is marked as a fourth abnormal area, and the abnormality type corresponding to the fourth abnormal area is marked as a multi-line abnormality.

[0013] Optionally, the historical abnormal scatter plot is a two-dimensional scatter plot, a first coordinate axis of the historical abnormal scatter plot represents a pitch angle, and a second coordinate axis of the historical abnormal scatter plot represents power, which is used to characterize the output power of the wind turbine generator set under different pitch angle conditions, and the abnormal area and the abnormal type of each historical abnormal scatter plot are marked to obtain an abnormal data set, including: Determine a fifth target scatter point set from the historical abnormal scatter point graph, wherein the fifth target scatter point set includes a plurality of scatter points constituting one or more line segments and is used to characterize abnormal scatter points of the wind turbine generator set outputting a second fixed power at different pitch angles; The area where the fifth target scattered point set is located is marked as a fifth abnormal area, and the abnormality type corresponding to the fifth abnormal area is marked as pitch angle abnormality.

[0014] According to a second aspect of an embodiment of the present disclosure, there is provided a data processing device, which is applied to a wind turbine generator set, and the device includes: An acquisition module, used for acquiring a current operation scatter plot of the wind turbine generator set, wherein the current operation scatter plot is used for representing the output power at different wind speeds; An output module is used to input the current running scatter plot into a pre-trained anomaly detection model to obtain target anomaly information output by the anomaly detection model, wherein the target anomaly information includes a target anomaly area and / or a target anomaly type.

[0015] Optionally, the training process of the anomaly detection model includes: Acquire a training sample data set, wherein the training sample data set includes abnormal power scatter plot sample data, abnormal area annotation data and abnormal type annotation data of a plurality of wind turbine generator sets; The training sample data set is used as training data to perform model training on a preset initial model to obtain the anomaly detection model.

[0016] Optionally, the output module is further used to: Acquire a plurality of historical abnormal scatter plots of the wind turbine generator sets, wherein the historical abnormal scatter plots include abnormal conditions of power output of the wind turbine generator sets at different wind speeds within a historical time period; Annotating the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set; The abnormal data set is used as a training sample data set.

[0017] Optionally, the historical anomaly scatter plot is a two-dimensional scatter plot, a first coordinate axis of the historical anomaly scatter plot represents wind speed, a second coordinate axis of the historical anomaly scatter plot represents power, and the output module is further used to: Obtaining a standard power scatter plot, wherein the standard power scatter plot is used to characterize the rated power output by the wind turbine generator set under different wind speed conditions; Determining a target deviation between the output power of the wind turbine generator set and the rated power at the same wind speed according to the standard power scatter plot and the historical abnormal scatter plot; Taking a plurality of scattered points whose target deviation is greater than a preset deviation threshold as a first target scattered point set; Marking the area where the first target scattered point set is located as a first abnormal area; The abnormality type corresponding to the first abnormal area is marked as a poor performance abnormality.

[0018] Optionally, the output module is further used to: When it is determined that there is a first scattered point in the first target scattered point set that exceeds a preset first power threshold, an area where a plurality of second scattered points in the first target scattered point set whose output powers are less than or equal to the output power of the first scattered point except the first scattered point are located is marked as a first abnormal area.

[0019] Optionally, the output module is further used to: Determine a second target scatter point set from the historical abnormal scatter point graph, wherein the second target scatter point set includes a plurality of scatter points constituting one or more line segments, and is used to characterize abnormal scatter points of the wind turbine generator set outputting a first fixed power at different wind speeds; taking multiple scatter points in the historical abnormal scatter diagram that are less than the first fixed power as third scatter points, wherein the third scatter points are used to represent scatter points where the target deviation between the output power of the wind turbine generator set at different wind speeds and the rated power of the wind turbine generator set is less than a preset deviation threshold; Marking the area where the second target scattered point set and the plurality of third scattered points are located as a second abnormal area; The abnormality type corresponding to the second abnormal area is marked as power-limiting abnormality.

[0020] Optionally, the output module is further used to: Determine a third target scatter point set from the historical abnormal scatter point diagram, the third target scatter point set including scatter points where the output power of the wind turbine generator set at a first wind speed is higher than a first standard power in a preset standard power scatter point diagram, and scatter points where the output power of the wind turbine generator set at a second wind speed is lower than a second standard power in the standard power scatter point diagram, the first wind speed being lower than the second wind speed; In a case where it is determined that the output power of the fourth scattered point in the third target scattered point set is greater than the preset second power threshold for the first time, if it is determined that the third target scattered point set, the fourth scattered point and the standard power scatter point diagram form a triangular area or an irregular closed area, the triangular area or the irregular closed area is marked as a third abnormal area; The abnormality type corresponding to the third abnormal area is marked as left-biased abnormality.

[0021] Optionally, the output module is further used to: Determine a fourth target scatter point set from the historical abnormal scatter point graph, wherein the fourth target scatter point set includes a plurality of scatter points constituting a plurality of similar curves, and is used to characterize that the wind turbine generator set has a plurality of abnormal scatter points with different output powers at the same wind speed; The area where the fourth target scattered point set is located is marked as a fourth abnormal area, and the abnormality type corresponding to the fourth abnormal area is marked as a multi-line abnormality.

[0022] Optionally, the historical abnormal scatter plot is a two-dimensional scatter plot, a first coordinate axis of the historical abnormal scatter plot represents a pitch angle, and a second coordinate axis of the historical abnormal scatter plot represents power, which is used to characterize the output power of the wind turbine generator set under different pitch angle conditions, and the output module is further used to: Determine a fifth target scatter point set from the historical abnormal scatter point graph, wherein the fifth target scatter point set includes a plurality of scatter points constituting one or more line segments and is used to characterize abnormal scatter points of the wind turbine generator set outputting a second fixed power at different pitch angles; The area where the fifth target scattered point set is located is marked as a fifth abnormal area, and the abnormality type corresponding to the fifth abnormal area is marked as pitch angle abnormality.

[0023] According to a third aspect of an embodiment of the present disclosure, there is provided a non-temporary computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the methods described in the first aspect are implemented.

[0024] According to a fourth aspect of an embodiment of the present disclosure, there is provided an electronic device, including: a memory having a computer program stored thereon; A processor is used to execute the computer program in the memory to implement the steps of any one of the methods in the first aspect.

[0025] According to a fifth aspect of an embodiment of the present disclosure, a computer program product is provided, including a computer program, which implements the steps of any one of the methods in the first aspect when executed by a processor.

[0026] Through the above technical solution, the current operation scatter plot of the wind turbine generator set is obtained, and the current operation scatter plot is used to characterize the output power at different wind speeds; the current operation scatter plot is input into a pre-trained anomaly detection model to obtain the target anomaly information output by the anomaly detection model, and the target anomaly information includes a target anomaly area and / or a target anomaly type. In this way, the target anomaly information in the current operation scatter plot is determined by the current operation scatter plot of the wind turbine generator set and the anomaly detection model, and the anomaly information in the current operation scatter plot of the wind turbine generator set can be quickly determined, thereby effectively improving the accuracy and efficiency of determining the abnormal conditions of the wind turbine generator set, and further effectively improving the overall performance of the wind turbine generator set.

[0027] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings: Figure 1 is a flow chart of a data processing method according to an exemplary embodiment; Figure 2 is based on Figure 1 A flowchart of an anomaly detection model training process shown in the illustrated embodiment; Figure 3 is based on Figure 2 A flow chart of a data processing method shown in the illustrated embodiment; Figure 4 is based on Figure 3 A flow chart of a data processing method shown in the illustrated embodiment; Figure 5 is based on Figure 3 A flow chart of another data processing method shown in the illustrated embodiment; Figure 6 is based on Figure 3 A flowchart of another data processing method shown in the illustrated embodiment; Figure 7 is based on Figure 3 A flowchart of another data processing method shown in the illustrated embodiment; Figure 8 is based on Figure 3 A flowchart of another data processing method shown in the illustrated embodiment; Fig. 9 is a block diagram of a data processing device according to an exemplary embodiment; Fig.10It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0029] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0030] Before introducing the specific implementation methods of the present disclosure in detail, the application scenarios of the present disclosure are first described as follows. The present disclosure can be applied to the application scenarios of determining abnormal information of wind turbines. The power curve of a wind turbine reflects the corresponding relationship between the output power of the unit and the change in wind speed. It is not only a necessary design basis for characterizing the output performance of the wind turbine, but also an important evaluation indicator for measuring the wind power generation capacity. The manufacturer of the wind turbine will submit the theoretical power curve of the wind turbine when providing the equipment. However, after the wind turbine is actually operated, as the operating time increases, the actual power curve of the wind turbine may deviate from the theoretical power curve. In the prior art, the actual power curve of the wind turbine is compared with the theoretical power curve, and whether the wind turbine is abnormal and the type of abnormality are determined by manual reasoning. However, in large-scale screening application scenarios, the manual reasoning judgment method has the problem of low efficiency. In addition, in large-scale screening application scenarios, there may also be a problem of reduced judgment accuracy due to fatigue caused by high-intensity work of the human brain.

[0031] In order to solve the above technical problems, the present disclosure provides a data processing method, a data processing device, a medium, an electronic device and a product. By obtaining the current operation scatter plot of the wind turbine generator set, the current operation scatter plot is used to characterize the output power at different wind speeds; the current operation scatter plot is input into a pre-trained anomaly detection model to obtain the target abnormality information output by the anomaly detection model, and the target abnormality information includes a target abnormal area and / or a target abnormality type. In this way, by determining the target abnormality information in the current operation scatter plot through the current operation scatter plot of the wind turbine generator set and the anomaly detection model, the abnormal information in the current operation scatter plot of the wind turbine generator set can be quickly determined, and the accuracy and efficiency of determining the abnormal conditions of the wind turbine generator set can be effectively improved, thereby effectively improving the overall performance of the wind turbine generator set.

[0032] The specific implementation of the present disclosure will be described in detail below with reference to the specific drawings.

[0033] Figure 1 is a flow chart of a data processing method according to an exemplary embodiment. Figure 1 As shown, the method comprises the following steps: Step 101, obtaining a current operation scatter plot of the wind turbine generator set.

[0034] The current operation scatter plot is used to characterize the output power at different wind speeds or pitch angles.

[0035] In this step, the current operation scatter plot may be a two-dimensional scatter plot. In one representation, the first coordinate axis of the current operation scatter plot may represent wind speed, and the second coordinate axis of the current operation scatter plot may represent power, which is used to characterize the output power at different wind speeds. In another representation, the first coordinate axis of the current operation scatter plot may represent pitch angle, and the second coordinate axis of the current operation scatter plot may represent power, which characterizes the output power at different pitch angles.

[0036] It should be noted that the operation data of each wind turbine generator set can be obtained regularly from the SCADA (Supervisory Control and Data Acquisition) system of the wind farm, and the operation data includes the output power of the wind turbine generator set at different wind speeds or pitch angles during the current operation period. The operation data is cleared by using interpolation technology and missing value filling strategy to obtain the current operation data, and the current operation data is converted into a current operation scatter plot through visualization operation.

[0037] Step 102: input the current running scatter plot into a pre-trained anomaly detection model to obtain target anomaly information output by the anomaly detection model.

[0038] The target abnormality information includes a target abnormality area and / or a target abnormality type.

[0039] In this step, the basic model of the target detection algorithm can be improved to obtain a pre-trained anomaly detection model, and the current running scatter plot is input into the pre-trained anomaly detection model to obtain the target anomaly area and / or target anomaly type output by the anomaly detection model.

[0040] It should be noted that the basic model can be Yolov8 (You Only Look Once, the 8th version of the target detection algorithm), which includes a Conv module (Convolutional Module), a traditional upsampling module (using the nearest neighbor interpolation method or the bilinear interpolation technique to infer the value of the new pixel point according to the spatial relationship between pixels) and MPDIoU Loss (Modified Point Distance Intersection over Union Loss), and the pre-trained anomaly detection model optimizes the Conv module to an ODConv module (Oriented Deformable Convolution Module); optimizes the upsampling module to a CARAFE upsampling operator (Content-Aware ReAssembly of Features); and optimizes MPDIoU Loss to a composite loss function of MPDIoU Loss and Inner-IoU (Inner Intersection over Union). Among them, the ODConv module is used to extract the features of the current running scatter plot, convert the current running scatter plot into an attention feature map including multiple attention weights in the spatial dimension, input channel dimension and output channel dimension, and the CARAFE upsampling operator is used to improve the high-resolution information in the attention feature map and generate a target feature map. The composite loss function of MPDIoU Loss and Inner-IoU in the pre-trained anomaly detection model can significantly improve the model's positioning accuracy for abnormal areas.

[0041] The above technical scheme determines the target abnormal information in the current operation scatter diagram of the wind turbine generator set through the current operation scatter diagram of the wind turbine generator set and the abnormality detection model, and can quickly determine the abnormal information in the current operation scatter diagram of the wind turbine generator set, thereby effectively improving the accuracy and efficiency of determining the abnormal conditions of the wind turbine generator set, and further effectively improving the overall performance of the wind turbine generator set.

[0042] Figure 2 is based on Figure 1 The embodiment shown is a flowchart of an anomaly detection model training process, such as Figure 2 As shown, the training process of the anomaly detection model includes: Step 201: Obtain a training sample data set.

[0043] The training sample data set includes abnormal power scatter plot sample data, abnormal area annotation data and abnormal type annotation data of a plurality of wind turbine generator sets.

[0044] In this step, the abnormal power scatter plot sample data may be a scatter plot of abnormal output power of the wind turbine generator set at different wind speeds in the historical event segment; the abnormal area annotation data may be the area where the abnormal scatter points in the abnormal power scatter plot sample data are located, and the abnormal type annotation data may be the abnormal type corresponding to the abnormal area in the abnormal power scatter plot sample data. The abnormal type may be one of poor performance abnormality, multi-line abnormality, power restriction abnormality, left deviation abnormality and pitch angle abnormality.

[0045] Step 202: Using the training sample data set as training data, perform model training on a preset initial model to obtain the anomaly detection model.

[0046] The preset initial model is obtained by optimizing the convolution module, upsampling module and loss function of the Yolov8 model. The convolution module is optimized to the ODConv module; the upsampling module is optimized to the CARAFE upsampling operator; and the MPDIoU Loss is optimized to the composite loss function of MPDIoU Loss and Inner-IoU. The ODConv module is used to extract the features of the current running scatter plot, convert the current running scatter plot into an attention feature map including multiple attention weights in the spatial dimension, input channel dimension and output channel dimension, and the CARAFE upsampling operator is used to improve the high-resolution information in the attention feature map and generate the target feature map.

[0047] The above technical solution trains the preset initial model according to the training sample data set to obtain an anomaly detection model, which can provide data support for the subsequent determination of the target abnormal area and / or target abnormal type.

[0048] Figure 3 is based on Figure 2 The embodiment shown is a flow chart of a data processing method, such as Figure 3 As shown, Figure 2 The step 201 of obtaining a training sample data set, wherein the training sample data set includes a plurality of abnormal power scatter plot sample data of the wind turbine generator set, abnormal region annotation data and abnormal type annotation data, may include: Step 2011, obtaining historical abnormality scatter plots of the plurality of wind turbine generator sets.

[0049] The historical abnormal scatter plot includes abnormal situations of the output power of the wind turbine generator set at different wind speeds or pitch angles within a historical time period.

[0050] Step 2012: annotate the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set.

[0051] In this step, the area where the abnormal scatter points in each historical abnormal scatter point map are located is marked as an abnormal area, and the abnormal type is determined according to the aggregation characteristics of the scatter points in the abnormal area. The abnormal type can be one of poor performance abnormality, multi-line abnormality, power limit abnormality, left deviation abnormality and pitch angle abnormality.

[0052] Step 2013: Use the abnormal data set as a training sample data set.

[0053] The abnormal data set includes a plurality of historical abnormal scatter plots of the wind turbine generator sets, the abnormal area in each of the historical abnormal scatter plots, and the abnormal type corresponding to the abnormal area.

[0054] The above technical solution obtains an abnormal data set by marking the abnormal area and the abnormal type in each of the historical abnormal scatter plots of multiple wind turbines, thereby providing data support for subsequent training of the abnormality detection model.

[0055] Figure 4 is based on Figure 3 The embodiment shown is a flow chart of a data processing method, such as Figure 4 As shown, Figure 2 The step 2012 of marking the abnormal area and abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set may include: S11, obtain a standard power scatter plot.

[0056] The standard power scatter plot is used to characterize the rated power output by the wind turbine generator set under different wind speed conditions.

[0057] S12: determining a target deviation between the output power of the wind turbine generator set and the rated power at the same wind speed according to the standard power scatter plot and the historical abnormal scatter plot.

[0058] The target deviation may be the difference between the output power of the wind turbine generator set and the rated power at the same wind speed. The historical abnormal scatter plot is a two-dimensional scatter plot, wherein the first coordinate axis of the historical abnormal scatter plot represents the wind speed, and the second coordinate axis of the historical abnormal scatter plot represents the power.

[0059] S13, taking a plurality of scattered points whose target deviation is greater than a preset deviation threshold as a first target scattered point set.

[0060] The deviation threshold is a value that increases in steps, and the target deviation of the multiple scattered points in the first target scattered point set gradually increases as the wind speed increases.

[0061] S14: Mark the area where the first target scattered point set is located as a first abnormal area.

[0062] In this step, if it is determined that there is a first scattered point in the first target scattered point set that exceeds a preset first power threshold, an area where a plurality of second scattered points in the first target scattered point set whose output power is less than or equal to the output power of the first scattered point except the first scattered point are located is marked as a first abnormal area.

[0063] S15: Mark the abnormality type corresponding to the first abnormal area as a poor performance abnormality.

[0064] The above technical scheme determines the first target scatter point set according to the standard power scatter point diagram and the historical abnormal scatter point diagram, determines the first abnormal area according to the first target scatter point set, and determines that the abnormal type corresponding to the first abnormal area is poor performance, which provides a basis for timely corrective maintenance of the wind turbine generator set in the first abnormal area in response to the poor performance, thereby effectively improving the power generation of the wind turbine generator set in the first abnormal area.

[0065] Figure 5 is based on Figure 3 The embodiment shown is a flowchart of another data processing method, such as Figure 4 As shown, Figure 2 The step 2012 of marking the abnormal area and abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set may also include: S21, determining a second target scatter point set from the historical abnormal scatter point graph.

[0066] The second target scattered point set includes a plurality of scattered points forming one or more line segments, and is used to characterize abnormal scattered points of the wind turbine generator set outputting a first fixed power at different wind speeds.

[0067] S22: taking a plurality of scatter points in the historical abnormal scatter point graph that are smaller than the first fixed power as third scatter points.

[0068] The third scatter point is used to represent a scatter point where the target deviation between the output power of the wind turbine generator set at different wind speeds and the rated power of the wind turbine generator set is less than a preset deviation threshold. The historical abnormal scatter plot is a two-dimensional scatter plot, the first coordinate axis of the historical abnormal scatter plot represents wind speed, and the second coordinate axis of the historical abnormal scatter plot represents power.

[0069] S23: Mark the area where the second target scattered point set and the plurality of third scattered points are located as a second abnormal area.

[0070] S24: Mark the abnormality type corresponding to the second abnormal area as power-limiting abnormality.

[0071] The above technical scheme determines a second target scatter point set based on the historical abnormal scatter point diagram, determines a second abnormal area based on the second target scatter point set, and determines that the abnormality type corresponding to the second abnormal area is a power restriction abnormality, which provides a basis for timely correcting and maintaining the wind turbines in the second abnormal area in response to the power restriction abnormality, thereby effectively improving the power generation of the wind turbines in the second abnormal area.

[0072] Figure 6 is based on Figure 3 The embodiment shown is a flowchart of another data processing method, as shown in FIG. Figure 4 As shown, Figure 2 The step 2012 of marking the abnormal area and abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set may also include: S31, determining a third target scatter point set from the historical abnormal scatter point graph.

[0073] The third target scatter point set includes scatter points where the output power of the wind turbine generator set at the first wind speed is higher than the first standard power in the preset standard power scatter point diagram, and scatter points where the output power at the second wind speed is lower than the second standard power in the standard power scatter point diagram, and the first wind speed is lower than the second wind speed. The historical abnormal scatter point diagram is a two-dimensional scatter point diagram, the first coordinate axis of the historical abnormal scatter point diagram represents wind speed, and the second coordinate axis of the historical abnormal scatter point diagram represents power.

[0074] S32: When it is determined that the output power of the fourth scatter point in the third target scatter point set is greater than the preset second power threshold for the first time, if it is determined that the third target scatter point set, the fourth scatter point and the standard power scatter point diagram form a triangular area or an irregular closed area, the triangular area or the irregular closed area is marked as a third abnormal area.

[0075] S33: Mark the abnormality type corresponding to the third abnormal area as left-biased abnormality.

[0076] The above technical scheme determines the third target scatter point set according to the historical abnormal scatter point diagram, determines the third abnormal area according to the third target scatter point set, and determines that the abnormality type corresponding to the third abnormal area is a left-biased abnormality, which provides a basis for timely correction and maintenance of the wind turbines in the third abnormal area in response to the left-biased abnormality, thereby effectively improving the power generation of the wind turbines in the third abnormal area.

[0077] Figure 7 is based on Figure 3 The embodiment shown is a flowchart of another data processing method, as shown in FIG. Figure 4 As shown, Figure 2 The step 2012 of marking the abnormal area and abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set may also include: S41, determining a fourth target scatter point set from the historical abnormal scatter point graph.

[0078] The fourth target scatter point set includes a plurality of scatter points constituting a plurality of similar curves, which are used to characterize that the wind turbine generator set has a plurality of abnormal scatter points with different output powers at the same wind speed. The historical abnormal scatter point graph is a two-dimensional scatter point graph, wherein the first coordinate axis of the historical abnormal scatter point graph represents wind speed, and the second coordinate axis of the historical abnormal scatter point graph represents power.

[0079] S42: Mark the area where the fourth target scattered point set is located as a fourth abnormal area, and mark the abnormality type corresponding to the fourth abnormal area as a multi-line abnormality.

[0080] The above technical scheme determines the fourth target scatter point set according to the historical abnormal scatter point diagram, determines the fourth abnormal area according to the fourth target scatter point set, and determines that the abnormal type of the fourth abnormal area is a multi-line abnormality, which provides a basis for timely correction and maintenance of the wind turbines in the fourth abnormal area in response to the multi-line abnormality, thereby effectively improving the power generation of the wind turbines in the fourth abnormal area.

[0081] Figure 8 is based on Figure 3 The embodiment shown is a flowchart of another data processing method, as shown in FIG. Figure 4 As shown, Figure 2 The step 2012 of marking the abnormal area and abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set may also include: S51, determining a fifth target scatter point set from the historical abnormal scatter point graph.

[0082] The fifth target scatter point set includes a plurality of scatter points forming one or more line segments, which are used to characterize abnormal scatter points of the wind turbine generator set outputting a second fixed power at different pitch angles. The historical abnormal scatter point graph is a two-dimensional scatter point graph, wherein the first coordinate axis of the historical abnormal scatter point graph represents the pitch angle, and the second coordinate axis of the historical abnormal scatter point graph represents the power, which is used to characterize the output power of the wind turbine generator set under different pitch angle conditions.

[0083] S52: Mark the area where the fifth target scattered point set is located as a fifth abnormal area, and mark the abnormality type corresponding to the fifth abnormal area as pitch angle abnormality.

[0084] The above technical scheme determines the fifth target scatter point set according to the historical abnormal scatter point diagram, determines the fifth abnormal area according to the fifth target scatter point set, and determines that the abnormal type of the fifth abnormal area is the pitch angle abnormality, which provides a basis for timely correcting and maintaining the wind turbines in the fifth abnormal area in response to the pitch angle abnormality, thereby effectively improving the power generation of the wind turbines in the fifth abnormal area.

[0085] Fig. 9 is a block diagram of a data processing device according to an exemplary embodiment; see Fig. 9 , the data processing device may include: An acquisition module 901 is used to acquire a current operation scatter plot of the wind turbine generator set, wherein the current operation scatter plot is used to represent output power at different wind speeds; The output module 902 is used to input the current running scatter plot into a pre-trained anomaly detection model to obtain target anomaly information output by the anomaly detection model, where the target anomaly information includes a target anomaly area and / or a target anomaly type.

[0086] Through the above technical solution, the current operation scatter plot of the wind turbine generator set is obtained, and the current operation scatter plot is used to characterize the output power at different wind speeds; the current operation scatter plot is input into a pre-trained anomaly detection model to obtain the target anomaly information output by the anomaly detection model, and the target anomaly information includes a target anomaly area and / or a target anomaly type. In this way, the target anomaly information in the current operation scatter plot is determined by the current operation scatter plot of the wind turbine generator set and the anomaly detection model, and the anomaly information in the current operation scatter plot of the wind turbine generator set can be quickly determined, thereby effectively improving the accuracy and efficiency of determining the abnormal conditions of the wind turbine generator set, and further effectively improving the overall performance of the wind turbine generator set.

[0087] Optionally, the training process of the anomaly detection model includes: Acquire a training sample data set, wherein the training sample data set includes abnormal power scatter plot sample data, abnormal area annotation data and abnormal type annotation data of a plurality of wind turbine generator sets; The training sample data set is used as training data to perform model training on a preset initial model to obtain the anomaly detection model.

[0088] Optionally, the output module 902 is further used to: Acquire a plurality of historical abnormal scatter plots of the wind turbine generator sets, wherein the historical abnormal scatter plots include abnormal conditions of power output of the wind turbine generator sets at different wind speeds within a historical time period; Annotating the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set; The abnormal data set is used as a training sample data set.

[0089] Optionally, the historical anomaly scatter plot is a two-dimensional scatter plot, a first coordinate axis of the historical anomaly scatter plot represents wind speed, and a second coordinate axis of the historical anomaly scatter plot represents power. The output module 902 is further used for: Obtaining a standard power scatter plot, wherein the standard power scatter plot is used to characterize the rated power output by the wind turbine generator set under different wind speed conditions; Determining a target deviation between the output power of the wind turbine generator set and the rated power at the same wind speed according to the standard power scatter plot and the historical abnormal scatter plot; Taking a plurality of scattered points whose target deviation is greater than a preset deviation threshold as a first target scattered point set; Marking the area where the first target scattered point set is located as a first abnormal area; The abnormality type corresponding to the first abnormal area is marked as a poor performance abnormality.

[0090] Optionally, the output module 902 is further used to: When it is determined that there is a first scattered point in the first target scattered point set that exceeds a preset first power threshold, an area where a plurality of second scattered points in the first target scattered point set whose output powers are less than or equal to the output power of the first scattered point except the first scattered point are located is marked as a first abnormal area.

[0091] Optionally, the output module 902 is further used to: Determine a second target scatter point set from the historical abnormal scatter point graph, wherein the second target scatter point set includes a plurality of scatter points constituting one or more line segments, and is used to characterize abnormal scatter points of the wind turbine generator set outputting a first fixed power at different wind speeds; taking multiple scatter points in the historical abnormal scatter diagram that are less than the first fixed power as third scatter points, wherein the third scatter points are used to represent scatter points where the target deviation between the output power of the wind turbine generator set at different wind speeds and the rated power of the wind turbine generator set is less than a preset deviation threshold; Marking the area where the second target scattered point set and the plurality of third scattered points are located as a second abnormal area; The abnormality type corresponding to the second abnormal area is marked as power-limiting abnormality.

[0092] Optionally, the output module 902 is further used to: Determine a third target scatter point set from the historical abnormal scatter point diagram, the third target scatter point set including scatter points where the output power of the wind turbine generator set at a first wind speed is higher than a first standard power in a preset standard power scatter point diagram, and scatter points where the output power of the wind turbine generator set at a second wind speed is lower than a second standard power in the standard power scatter point diagram, the first wind speed being lower than the second wind speed; In a case where it is determined that the output power of the fourth scattered point in the third target scattered point set is greater than the preset second power threshold for the first time, if it is determined that the third target scattered point set, the fourth scattered point and the standard power scatter point diagram form a triangular area or an irregular closed area, the triangular area or the irregular closed area is marked as a third abnormal area; The abnormality type corresponding to the third abnormal area is marked as left-biased abnormality.

[0093] Optionally, the output module 902 is further used to: Determine a fourth target scatter point set from the historical abnormal scatter point graph, wherein the fourth target scatter point set includes a plurality of scatter points constituting a plurality of similar curves, and is used to characterize that the wind turbine generator set has a plurality of abnormal scatter points with different output powers at the same wind speed; The area where the fourth target scattered point set is located is marked as a fourth abnormal area, and the abnormality type corresponding to the fourth abnormal area is marked as a multi-line abnormality.

[0094] Optionally, the historical abnormal scatter plot is a two-dimensional scatter plot, a first coordinate axis of the historical abnormal scatter plot represents a pitch angle, and a second coordinate axis of the historical abnormal scatter plot represents power, which is used to characterize the output power of the wind turbine generator set under different pitch angle conditions. The output module 902 is further used to: Determine a fifth target scatter point set from the historical abnormal scatter point graph, wherein the fifth target scatter point set includes a plurality of scatter points constituting one or more line segments and is used to characterize abnormal scatter points of the wind turbine generator set outputting a second fixed power at different pitch angles; The area where the fifth target scattered point set is located is marked as a fifth abnormal area, and the abnormality type corresponding to the fifth abnormal area is marked as pitch angle abnormality.

[0095] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0096] Fig.10 FIG. 1 is a block diagram of an electronic device 1000 according to an exemplary embodiment. Fig.10As shown, the electronic device 1000 may include: a processor 1001 , a memory 1002 . The electronic device 1000 may also include one or more of a multimedia component 1003 , an input / output (I / O) interface 1004 , and a communication component 1005 .

[0097] The processor 1001 is used to control the overall operation of the electronic device 1000 to complete all or part of the steps in the above-mentioned data processing method. The memory 1002 is used to store various types of data to support the operation of the electronic device 1000. For example, these data may include instructions for any application or method used to operate on the electronic device 1000, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 1002 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 1003 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, which is used to receive external audio signals. The received audio signal may be further stored in the memory 1002 or sent through the communication component 1005. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 1004 provides an interface between the processor 1001 and other interface modules, and the above-mentioned other interface modules may be keyboards, mice, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 1005 is used for wired or wireless communication between the electronic device 1000 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, 4G, NB-IOT, eMTC, or other 5G, etc., or a combination of one or more of them, is not limited here. Therefore, the corresponding communication component 1005 may include: Wi-Fi module, Bluetooth module, NFC module, etc.

[0098] In an exemplary embodiment, the electronic device 1000 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned data processing method.

[0099] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by the processor, the steps of the above-mentioned data processing method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 1002 including program instructions, and the above-mentioned program instructions can be executed by the processor 1001 of the electronic device 1000 to complete the above-mentioned data processing method.

[0100] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0101] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0102] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. A data processing method, characterized in that: Applied to a wind turbine generator set, the method comprises: Obtaining a current operation scatter plot of the wind turbine generator set, wherein the current operation scatter plot is used to characterize output power at different wind speeds or pitch angles; The current running scatter plot is input into a pre-trained anomaly detection model to obtain target anomaly information output by the anomaly detection model, where the target anomaly information includes a target anomaly region and / or a target anomaly type.

2. The data processing method according to claim 1, characterized in that: The training process of the anomaly detection model includes: Acquire a training sample data set, wherein the training sample data set includes abnormal power scatter plot sample data, abnormal area annotation data and abnormal type annotation data of a plurality of wind turbine generator sets; The training sample data set is used as training data to perform model training on a preset initial model to obtain the anomaly detection model.

3. The data processing method according to claim 2, characterized in that: The acquiring of a training sample data set, wherein the training sample data set includes a plurality of abnormal power scatter plot sample data of the wind turbine generator set, abnormal region annotation data and abnormal type annotation data, includes: Acquire a plurality of historical abnormal scatter plots of the wind turbine generator sets, wherein the historical abnormal scatter plots include abnormal conditions of power output of the wind turbine generator sets at different wind speeds or pitch angles within a historical time period; Annotating the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set; The abnormal data set is used as a training sample data set.

4. The data processing method according to claim 3, characterized in that: The historical anomaly scatter plot is a two-dimensional scatter plot, the first coordinate axis of the historical anomaly scatter plot represents wind speed, and the second coordinate axis of the historical anomaly scatter plot represents power. The abnormal area and abnormal type of each historical anomaly scatter plot are marked to obtain an abnormal data set, including: Obtaining a standard power scatter plot, wherein the standard power scatter plot is used to characterize the rated power output by the wind turbine generator set under different wind speed conditions; Determining a target deviation between the output power of the wind turbine generator set and the rated power at the same wind speed according to the standard power scatter plot and the historical abnormal scatter plot; Taking a plurality of scattered points whose target deviation is greater than a preset deviation threshold as a first target scattered point set; Marking the area where the first target scattered point set is located as a first abnormal area; The abnormality type corresponding to the first abnormal area is marked as a poor performance abnormality.

5. The data processing method according to claim 4, characterized in that: The step of marking the area where the first target scattered point set is located as a first abnormal area further includes: When it is determined that there is a first scattered point in the first target scattered point set that exceeds a preset first power threshold, an area where a plurality of second scattered points in the first target scattered point set whose output powers are less than or equal to the output power of the first scattered point except the first scattered point are located is marked as a first abnormal area.

6. The data processing method according to claim 3, characterized in that: The marking of the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set includes: Determine a second target scatter point set from the historical abnormal scatter point graph, wherein the second target scatter point set includes a plurality of scatter points constituting one or more line segments, and is used to characterize abnormal scatter points of the wind turbine generator set outputting a first fixed power at different wind speeds; taking multiple scatter points in the historical abnormal scatter diagram that are less than the first fixed power as third scatter points, wherein the third scatter points are used to represent scatter points where the target deviation between the output power of the wind turbine generator set at different wind speeds and the rated power of the wind turbine generator set is less than a preset deviation threshold; Marking the area where the second target scattered point set and the plurality of third scattered points are located as a second abnormal area; The abnormality type corresponding to the second abnormal area is marked as power-limiting abnormality.

7. The data processing method according to claim 3, characterized in that: The marking of the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set includes: Determine a third target scatter point set from the historical abnormal scatter point diagram, the third target scatter point set including scatter points where the output power of the wind turbine generator set at a first wind speed is higher than a first standard power in a preset standard power scatter point diagram, and scatter points where the output power of the wind turbine generator set at a second wind speed is lower than a second standard power in the standard power scatter point diagram, the first wind speed being lower than the second wind speed; In a case where it is determined that the output power of the fourth scattered point in the third target scattered point set is greater than the preset second power threshold for the first time, if it is determined that the third target scattered point set, the fourth scattered point and the standard power scatter point diagram form a triangular area or an irregular closed area, the triangular area or the irregular closed area is marked as a third abnormal area; The abnormality type corresponding to the third abnormal area is marked as left-biased abnormality.

8. The data processing method according to claim 3, characterized in that: The marking of the abnormal area and the abnormal type of each of the historical abnormal scatter plots to obtain an abnormal data set includes: Determine a fourth target scatter point set from the historical abnormal scatter point graph, wherein the fourth target scatter point set includes a plurality of scatter points constituting a plurality of similar curves, and is used to characterize that the wind turbine generator set has a plurality of abnormal scatter points with different output powers at the same wind speed; The area where the fourth target scattered point set is located is marked as a fourth abnormal area, and the abnormality type corresponding to the fourth abnormal area is marked as a multi-line abnormality.

9. The data processing method according to claim 3, characterized in that: The historical abnormal scatter plot is a two-dimensional scatter plot, the first coordinate axis of the historical abnormal scatter plot represents the pitch angle, and the second coordinate axis of the historical abnormal scatter plot represents the power, which is used to characterize the output power of the wind turbine generator set under different pitch angle conditions, and the abnormal area and the abnormal type of each historical abnormal scatter plot are marked to obtain an abnormal data set, including: Determine a fifth target scatter point set from the historical abnormal scatter point graph, wherein the fifth target scatter point set includes a plurality of scatter points constituting one or more line segments and is used to characterize abnormal scatter points of the wind turbine generator set outputting a second fixed power at different pitch angles; The area where the fifth target scattered point set is located is marked as a fifth abnormal area, and the abnormality type corresponding to the fifth abnormal area is marked as pitch angle abnormality.

10. A data processing device, characterized in that: Applied to a wind turbine generator set, the device comprises: An acquisition module, used for acquiring a current operation scatter plot of the wind turbine generator set, wherein the current operation scatter plot is used for representing the output power at different wind speeds; An output module is used to input the current running scatter plot into a pre-trained anomaly detection model to obtain target anomaly information output by the anomaly detection model, wherein the target anomaly information includes a target anomaly area and / or a target anomaly type.

11. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

12. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 9.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images