Wind turbine layout position detection method, model training method and device

Abstract

The present application discloses a method for detecting the layout position of a wind turbine, a model training method, and an apparatus. The method divides a wind turbine in a wind farm into multiple sectors. For a target sector among the multiple sectors, the current terrain data of the target sector and the current wind parameter data of the wind farm are obtained. The current terrain data and the current wind parameter data are input into a trained risk prediction model to obtain a risk prediction result for the target sector. Based on the risk prediction result, the layout position of the wind turbine is detected. That is, the embodiment of the present application can detect whether the layout position of the wind turbine has a high-frequency vibration risk based on the predetermined correspondence between terrain data, wind parameter data, and the risk of exceeding the limit of nacelle acceleration, combined with the current terrain data of the target sector and the current wind parameter data of the wind farm, without the need for manual judgment. This avoids the influence of personal experience on the detection results and improves the accuracy of the detection results.

Description

Technical Field

[0001] The present application belongs to the technical field of wind farm site selection, and specifically relates to a layout position detection method, model training method and device for a wind turbine generator set. Background Art

[0002] In wind farms, high-frequency vibrations in wind turbines can affect their safe operation and power generation. Therefore, when selecting a site for a wind turbine, it's often necessary to check whether the turbine's location poses a high-frequency vibration risk. If so, relocating the turbine may be necessary.

[0003] Currently, the detection of whether a wind turbine has a high-frequency vibration risk at the layout location is mainly based on manual experience and the terrain of the layout location, but the accuracy is poor. Summary of the Invention

[0004] The embodiments of the present application provide a layout position detection method, a model training method, and an apparatus for a wind turbine generator set, which can improve the accuracy of detection results when detecting the layout position of a wind turbine generator set.

[0005] In a first aspect, an embodiment of the present application provides a method for detecting the layout position of a wind turbine generator set, the method comprising:

[0006] Divide the wind turbines in the wind farm into multiple sectors, and obtain current terrain data of the target sector and current wind parameter data of the wind farm for a target sector among the multiple sectors;

[0007] The current terrain data and wind parameter data are input into the trained risk prediction model to obtain a risk prediction result for the target sector. The risk prediction model is used to characterize the corresponding relationship between the terrain data, wind parameter data, and the risk of exceeding the cabin acceleration limit. The risk prediction result is used to indicate whether the target sector has a high-frequency vibration risk.

[0008] According to the risk prediction results, the layout position of the wind turbine is detected.

[0009] In a second aspect, an embodiment of the present application provides a method for training a risk prediction model, the method comprising:

[0010] Acquire a training sample, where the training sample includes historical terrain data of a target sector among multiple sectors of each wind turbine generator set in the wind farm and historical wind parameter data of the wind farm;

[0011] Determine a machine learning model for establishing a correspondence between historical terrain data, historical wind parameter data, and the risk of exceeding the cabin acceleration limit;

[0012] Train machine learning models based on historical terrain data and historical wind parameter data;

[0013] If the stopping condition is met, the training is stopped and a risk prediction model that has completed the training is obtained.

[0014] In a third aspect, an embodiment of the present application provides a layout position detection device for a wind turbine generator system, the device comprising:

[0015] A data acquisition module is used to divide the wind turbines in the wind farm into multiple sectors, and for a target sector among the multiple sectors, obtain the current terrain data of the target sector and the current wind parameter data of the wind farm;

[0016] A risk prediction result determination module is used to input current terrain data and current wind parameter data into a trained risk prediction model to obtain a risk prediction result for the target sector. The risk prediction model is used to characterize the correspondence between terrain data, wind parameter data, and the risk of exceeding the cabin acceleration limit. The risk prediction result is used to indicate whether the target sector has a high-frequency vibration risk.

[0017] The detection module is used to detect the layout position of the wind turbine according to the risk prediction results.

[0018] In a fourth aspect, an embodiment of the present application provides a risk prediction model training device, the device comprising:

[0019] A training sample acquisition module is used to acquire training samples, where the training samples include historical terrain data of a target sector among multiple sectors of each wind turbine in the wind farm and historical wind parameter data of the wind farm;

[0020] a machine learning model determination module, configured to determine a machine learning model for establishing a corresponding relationship between historical terrain data, historical wind parameter data, and the risk of exceeding a limit on cabin acceleration;

[0021] The training module is used to train the machine learning model based on historical terrain data and historical wind parameter data;

[0022] If the stopping condition is met, the training is stopped and a risk prediction model that has completed the training is obtained.

[0023] In a fifth aspect, an embodiment of the present application provides an electronic device, including:

[0024] processor;

[0025] a memory for storing computer program instructions;

[0026] When the computer program instructions are executed by a processor, the method according to the first aspect or the method according to the second aspect is implemented.

[0027] In a sixth aspect, an embodiment of the present application provides a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the method described in the first aspect or the method described in the second aspect is implemented.

[0028] The embodiments of the present application provide a method for detecting the layout position of a wind turbine, a model training method, and an apparatus. The wind turbines in a wind farm are divided into multiple sectors. For a target sector among the multiple sectors, the current terrain data of the target sector and the current wind parameter data of the wind farm are obtained. The current terrain data and the current wind parameter data are input into a trained risk prediction model to obtain a risk prediction result for the target sector. Based on the risk prediction result, the layout position of the wind turbine is detected. That is, the embodiments of the present application can detect whether there is a high-frequency vibration risk at the layout position of the wind turbine based on the predetermined correspondence between terrain data, wind parameter data, and the risk of exceeding the limit of nacelle acceleration, combined with the current terrain data of the target sector and the current wind parameter data of the wind farm, without the need for manual judgment. This can avoid the influence of personal experience on the detection results and improve the accuracy of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A flow chart of a method for detecting the layout position of a wind turbine generator system provided in an embodiment of the present application;

[0030] Figure 2 A top view of a sector provided in an embodiment of the present application;

[0031] Figure 3 A flow chart of another method for detecting the layout position of a wind turbine generator system provided in an embodiment of the present application;

[0032] Figure 4 A schematic diagram of a point location provided in an embodiment of the present application;

[0033] Figure 5 A flowchart of a risk prediction model training method provided in an embodiment of the present application;

[0034] Figure 6 A structural diagram of a wind turbine layout position detection device provided in an embodiment of the present application;

[0035] Figure 7 A structural diagram of a risk prediction model training device provided in an embodiment of the present application;

[0036] Figure 8 A structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0037] The following will be combined with the accompanying drawings in the embodiments of the present application to clearly describe the technical solutions in the embodiments of the present application. Obviously, the embodiments described are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of this application.

[0038] The terms "first," "second," and the like in the specification and claims of this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of this application can be implemented in an order other than that illustrated or described herein, and that the objects distinguished by "first," "second," and the like are generally of the same type, and do not limit the number of objects; for example, the first object can be one or more. In addition, the term "and / or" in the specification and claims refers to at least one of the connected objects, and the character " / " generally indicates that the objects connected are in an "or" relationship.

[0039] Analysis of wind turbine operating data reveals that high-frequency vibration is a major issue affecting wind turbine operation, impacting, for example, turbine maintenance and power generation. High-frequency vibration primarily manifests as excessive nacelle acceleration. Therefore, high-frequency vibration is often associated with excessive nacelle acceleration.

[0040] Based on data from wind farms with high-frequency vibration, it's clear that only a few turbines within a wind farm experience high-frequency vibration, and that these problems occur primarily in certain months. Furthermore, analysis of the operating data for wind turbines with high-frequency vibration reveals that only certain sectors of these turbines experience excessive nacelle acceleration. This means that for wind turbines within the same wind farm, the presence of excessive nacelle acceleration in a sector is related to the terrain and wind parameter data for that sector.

[0041] Therefore, in order to detect the layout position of the wind turbine to be built and avoid the problem of excessive nacelle acceleration in certain sectors of the wind turbine to be built, the embodiment of the present application combines the terrain data, wind parameter data and the correspondence between the risks of excessive nacelle acceleration of the sector to detect the layout position of the wind turbine, thereby improving the accuracy of the detection results.

[0042] The following describes in detail the wind turbine layout position detection method, model training method and device provided by the embodiments of the present application through specific embodiments and their application scenarios in conjunction with the accompanying drawings.

[0043] Figure 1 This is a flow chart of a method for detecting the layout position of a wind turbine generator set provided in an embodiment of the present application.

[0044] like Figure 1 As shown, the layout position detection method of the wind turbine generator set may include the following steps:

[0045] S110 , dividing the wind turbines in the wind farm into multiple sectors, and acquiring current terrain data of a target sector and current wind parameter data of the wind farm for the target sector among the multiple sectors.

[0046] S120: Input the current terrain data and the current wind parameter data into the trained risk prediction model to obtain the risk prediction result of the target sector.

[0047] Among them, the risk prediction model is used to characterize the correspondence between terrain data, wind parameter data and the risk of cabin acceleration exceeding the limit, and the risk prediction result is used to characterize whether there is a high-frequency vibration risk in the target sector.

[0048] S130. Detect the layout position of the wind turbine generator system based on the risk prediction result.

[0049] In an embodiment of the present application, a wind turbine in a wind farm is divided into multiple sectors. For a target sector among the multiple sectors, current terrain data of the target sector and current wind parameter data of the wind farm are obtained; the current terrain data and current wind parameter data are input into a trained risk prediction model to obtain a risk prediction result for the target sector; and based on the risk prediction result, the layout position of the wind turbine is detected. That is, based on the predetermined correspondence between terrain data, wind parameter data, and the risk of exceeding the limit of nacelle acceleration, the embodiment of the present application can detect whether the layout position of the wind turbine has a high-frequency vibration risk, combining the current terrain data of the target sector and the current wind parameter data of the wind farm, without the need for manual judgment. This can avoid the influence of personal experience on the detection results and improve the accuracy of the detection results.

[0050] The above steps are explained in detail below:

[0051] In S110 , the sector is an area obtained by dividing the terrain where the wind turbine is located with the layout position of the wind turbine as the center of the circle.

[0052] The embodiment of the present application does not specifically limit the way of dividing sectors. For example, refer to Figure 2 , Figure 2 A top view of a sector provided in an embodiment of the present application. Figure 2 The area formed by 11.25 degrees to the left and right of the wind turbine's true north is a sector, also known as sector 0. Then, going clockwise downward, every 22.5 degrees is a sector. In other words, in the clockwise direction, the sector next to sector 0 is sector 1 (11.25 degrees - 33.75 degrees). In this way, the terrain where the wind turbine is located can be divided into 16 sectors (sector 15). Point O is the center of the circle corresponding to the sector, which is also the layout location of the wind turbine.

[0053] The target sector can be one or more sectors among multiple sectors that are at risk of high-frequency vibration. The current terrain data of the target sector is the actual terrain data currently collected for the target sector. For the same target sector, the terrain data obtained may also vary depending on the collection time. When it is necessary to detect the layout position of the wind turbine, the embodiment of the present application collects the terrain data of the target sector and uses this terrain data as the current terrain data of the target sector.

[0054] The terrain data may be data reflecting sector terrain information. For example, points reflecting sector terrain information may be determined, and the current terrain data of the target sector may be determined based on the current data of these points. The present embodiment does not limit the process of determining the points; the current data of these points may include, but is not limited to, the elevation of these points, the elevation difference between different points, the slope, and the horizontal distance.

[0055] The elevation here is the vertical distance from the point to the reference plane. The reference plane can be selected according to actual needs, for example, it can be a horizontal ground. It should be noted that in the embodiment of the present application, each point corresponds to the same reference plane. The elevation difference is the difference in elevation between two points. For example, for point A and point B, the elevation of point A is h1, and the elevation of point B is h2. Then the elevation difference between point A and point B is h1-h2. The slope is the ratio of the elevation difference between two points to the horizontal distance between the two points. For example, the horizontal distance between point A and point B is x AB , then the slope of AB is (h1-h2) / x AB , the slope of BA is (h2-h1) / x AB .

[0056] The current wind parameter data for a wind farm is the wind parameter data collected at the same time. For example, this wind parameter data can be measured by a wind tower within the wind farm. A wind tower is a tall structure used to measure wind parameter data, specifically, a tower-shaped structure used to observe and record near-ground airflow. A wind tower can be installed within a wind farm. Under this premise, the corresponding wind parameter data for different wind turbines within the same wind farm is the same.

[0057] For example, the wind parameter data may include but are not limited to the shear of wind speeds above 10m / s at hub height (shear_big), the minimum wind shear of wind speeds between 6m / s and 12m / s at hub height (shear_e_min), the average wind shear of wind speeds between 6m / s and 12m / s at hub height (shear_e_mean), the proportion of negative wind shear samples of wind speeds between 6m / s and 12m / s at hub height to all data (shear_min_p), the maximum value of turbulence intensity at a wind speed of 10m / s (Tur_max), the SD value of turbulence intensity at a wind speed of 10m / s (Tur_SD), the average value of turbulence intensity at a wind speed of 10m / s (Tur_mean), the average wind speed from cut-in to cut-out, and the average wind speed from cut-out to cut-out. Transformation value (Shear), 90% quantile of the difference in wind direction change between the highest layer and 30 meters (direction_transform), average value of characteristic turbulence from wind speed above 8m / s to cut-out wind speed (Tur_over8), ratio of maximum value to average value of 20 samples with 10-min average wind speed greater than 10m / s (Max_min_ratio), proportion of samples with standard deviation of more than 20 degrees above 8m / s (Over20_ratio), maximum wind speed of 10-min data (MAXspeed), average wind speed (Speed), number of samples in sector (Count), A value of Weibull fitting of wind speed distribution, and K value of Weibull fitting of wind speed distribution.

[0058] The quantile is a data distribution, specifically the proportion that does not exceed a certain value. For example, the 90% quantile represents the percentage of data that does not exceed a certain value. It should be noted that the aforementioned values ​​of 10 m / s, 6-12 m / s, 8 m / s, 30 meters, and 90% are merely examples and can be adjusted as needed in practice.

[0059] In S120, a risk prediction model is used to characterize the correspondence between terrain data, wind parameter data, and the risk of exceeding the nacelle acceleration limit. The input of the risk prediction model is the terrain data and wind parameter data, and the output is a risk prediction result. For example, the risk prediction result can be yes or no, where "yes" may indicate that there is a risk of exceeding the nacelle acceleration limit in the target sector, that is, there is a risk of high-frequency vibration of the wind turbine at that layout location; "no" may indicate that there is no risk of exceeding the nacelle acceleration limit in the target sector, that is, there is no risk of high-frequency vibration of the wind turbine at that layout location.

[0060] The embodiments of this application do not limit the structure of the risk prediction model. Any model that can determine the correspondence between terrain data, wind parameter data, and the risk of exceeding the cabin acceleration limit can be used. For example, an XGBoost model can be used, or a custom model can be constructed or selected based on actual needs. Before application, the risk prediction model can be trained. The training process of the risk prediction model can be seen in the following embodiments.

[0061] In an embodiment of the present application, the current terrain data and current wind parameter data of the target sector are input into the trained risk prediction model, and the trained risk prediction model can be used to determine whether there is a risk of cabin acceleration exceeding the limit in the target sector, that is, whether the wind turbine has a risk of high-frequency vibration in the layout position. There is no need to manually determine it based on experience, thus avoiding the unreliability and judgment errors of human reliance on experience and improving the accuracy of the detection results.

[0062] In S130, illustratively, the layout position of the wind turbine generator set may be detected based on the risk prediction result. Specifically, detecting the layout position of the wind turbine generator set based on the risk prediction result may include the following steps:

[0063] In response to the high-frequency vibration risk existing in the target sector, the target sector of the layout position of the wind turbine is closed through the yaw system, or the layout position of the wind turbine is adjusted.

[0064] In embodiments of the present application, if the risk prediction result is yes, meaning there is a high-frequency vibration risk in a target sector, in some embodiments, the yaw system can be automatically shut down in the target sector. For example, if the target sector is sector 1, and there is a high-frequency vibration risk in sector 1, the yaw system can be shut down in sector 1, i.e., the yaw system can rotate within the remaining sectors (sectors 0, 2-15).

[0065] In some embodiments, a prompt message can also be sent to the user, which may include a risk prediction result (yes) and an adjustment strategy for the layout position determined based on the current terrain data and the current wind parameter data. In this way, when the user receives the prompt message, he can adjust the layout position of the wind turbine according to the adjustment strategy without the need for manual adjustment based on experience. This simplifies the user's operation and improves the accuracy of the layout position.

[0066] When the risk prediction result is no, that is, there is no high-frequency vibration risk in the target sector, illustratively, when only high-frequency vibration is considered, the layout position can be retained to indicate that a wind turbine can be established at the layout position.

[0067] In an embodiment of the present application, the target sector can be predicted based on the trained risk prediction model, combined with the current terrain data of the target sector and the current wind parameter data of the wind field, to predict whether the target sector is at risk of high-frequency vibration. If the risk prediction result is yes, the target sector can be closed through the yaw system, or an adjustment strategy can be sent to the user, so that the user can adjust the layout position based on the adjustment strategy. This simplifies manual operations and improves adjustment efficiency and the accuracy of adjustment results.

[0068] Considering that adjacent sectors may affect each other, in order to improve the accuracy of the detection result, in some embodiments, the above S120 may include the following steps:

[0069] In response to the existence of high-frequency vibration risk in the target sector, the current terrain data of the adjacent sector adjacent to the target sector and the current wind parameter data of the wind field are input into the trained risk prediction model containing adjacent sector elements to determine whether the adjacent sector has high-frequency vibration risk.

[0070] In an embodiment of the present application, if a target sector is determined to be at risk of exceeding cabin acceleration limits, further evaluation can be performed on adjacent sectors to determine whether they are at risk of exceeding cabin acceleration limits. For example, if the target sector is sector 1, and sectors 0 and 2 are adjacent to sector 1, if sector 1 is at risk of exceeding cabin acceleration limits, further evaluation can be performed to determine whether sectors 0 and 2 are at risk of exceeding cabin acceleration limits, thereby improving the accuracy of the detection results.

[0071] In order to obtain the current terrain data of the target sector, in some embodiments, refer to Figure 3 The wind turbine layout position detection method may include the following steps:

[0072] S310: Determine a first point set and a second point set associated with the layout position according to the layout position and point determination rule of the wind turbine generator.

[0073] S320: Obtain current wind parameter data and layout position of the wind farm, current terrain data corresponding to each point in the first point set, and each point in the second point set.

[0074] S330: Input the current terrain data and the current wind parameter data into the trained risk prediction model to obtain the risk prediction result of the target sector.

[0075] S340. Detect the layout position of the wind turbine generator system based on the risk prediction result.

[0076] Among them, S330 and S340 Figure 1The processes of S120 and S130 are the same. For details, please refer to the description of S120 and S130. For the sake of brevity, they are not repeated here.

[0077] Below Figure 3 The other steps in the are detailed as follows:

[0078] In S310 , the first point set and the second point set are point sets that can reflect the terrain information of the target sector.

[0079] In some embodiments, the present application embodiment takes the example of a first point set including a first point B, a second point C, a third point D, and a fourth point E, and a second point set including a fifth point B' and a sixth point C'. To determine the first point set and the second point set, the above S310 may include the following steps:

[0080] S3101. Determine a first point in a target sector based on a relationship between a first elevation of a first candidate point and a second elevation of a layout position.

[0081] The first candidate point is a point in the target sector whose horizontal distance from the layout position meets a first preset condition.

[0082] The first elevation is the elevation of the first candidate point, and the second elevation is the elevation of the layout location. For example, the first preset condition may be that the horizontal distance between the first candidate point and the layout location is greater than or equal to d1. The value of d1 can be set based on actual needs. For example, it can be set to 100 meters, meaning that the horizontal distance between the first candidate point and the layout location is greater than or equal to 100 meters.

[0083] The relationship between the first elevation and the second elevation may include the first elevation being smaller than the second elevation, and the first elevation being not smaller than the second elevation.

[0084] Specifically, when the first elevation is not less than the second elevation, it can be determined that the first point position and the layout position coincide with each other, that is, point B and point A coincide with each other.

[0085] When the first elevation is less than the second elevation, a first point whose elevation meets the preset elevation can be determined from the first area, wherein the first area is an area between points in the target sector whose horizontal distance from the layout position meets the preset distance.

[0086] Exemplarily, the first area may be an area between points in the target sector whose horizontal distance from point A is greater than d1 and less than d2. Optionally, d2=800 meters.

[0087] In some embodiments, the first low point in the first area can be determined as the first point. The low point here is a point where the elevation has an inflection point, that is, the elevations of the points before the low point decrease in sequence, and the elevations of the points after the low point increase in sequence.

[0088] For example, refer to Figure 4 , Figure 4 A schematic diagram of points provided in an embodiment of the present application. Within the first region, the elevations of points preceding point B decrease, while the elevations of points following point B increase. That is, point B is the point where the elevation has an inflection point. Therefore, point B can be determined as the first point. Figure 4 For example, if the first area contains one low point (point B), in actual application, the first area may contain multiple low points. For example, if there are two low points after point B, point B (the first low point) is still determined as the first point.

[0089] S3102: Determine a second point, a third point, and a fourth point in the target sector according to the layout position and the first point.

[0090] Exemplarily, when the first elevation is less than the second elevation and the horizontal distance between the layout position and the first point is less than a first threshold, a second point whose elevation satisfies a second preset condition is determined from the second area, where the second area is an area within the target sector between points whose horizontal distances from the first point meet a third preset condition.

[0091] Determine a point in the target sector whose horizontal distance from the second point satisfies a fourth preset condition as a fourth point;

[0092] A third point is determined according to the second point and the fourth point, and the third point is located between the second point and the fourth point.

[0093] In an embodiment of the present application, for example, the first threshold value may be 800 meters. The second preset condition may be that the elevation is the highest in the second area. The third preset condition may be that the horizontal distance from the first point B is greater than d1 and less than d3, optionally, d3 = 1000 meters. The fourth preset condition may be that the horizontal distance is greater than or equal to d4, for example, d4 = 1000 meters. In actual application, the first threshold value, the second preset condition, the third preset condition, and the fourth preset condition may be adjusted as needed.

[0094] Specifically, when the horizontal distance between AB is less than the first threshold, the point with the highest elevation in the second area can be selected as the second point. Figure 4 , point C is the point with the highest elevation in the second area, so point C can be determined as the second point. For example, the point with a horizontal distance of 1000 meters from point C, that is, Figure 4 Point E in is determined as the fourth point.

[0095] When the second point and the fourth point are determined, the third point can be determined based on the second point and the fourth point.

[0096] Exemplarily, the elevation difference between the second candidate point and the third candidate point can be determined. When the elevation difference is less than a second threshold, the second candidate point is determined to be the third point. When the elevation difference is not less than the second threshold, the third point is determined to coincide with the fourth point.

[0097] Among them, the second candidate point and the third candidate point are points between the second point and the fourth point, the elevation of the second candidate point is not less than the elevation of the third candidate point, and the horizontal distance between the second candidate point and the third candidate point is a preset distance.

[0098] Specifically, the search can be performed starting from the second point with a predetermined step length. For example, starting from the second point, a point with a horizontal distance from the second point of the predetermined step length is determined, and then the elevation difference between the point and the second point is determined. If the elevation difference meets the second threshold, the point can be determined as the third point. Otherwise, starting from the point, the search can continue with the predetermined step length to determine the next point until the elevation difference between the next point and the previous point reaches the second threshold. At this time, the next point can be determined as the third point. The horizontal distance between the next point and the previous point is the predetermined step length. For example, the predetermined step length is 210 meters, and the second threshold is 21 meters.

[0099] For example, refer to Figure 4 , the horizontal distance between point D and point D' is 210 meters, and the elevation difference between point D and point D' is 21 meters, so point D can be determined as the third point.

[0100] In some embodiments, when the horizontal distance between AB is not less than a first threshold, it can be determined that the second point, the third point, and the fourth point respectively coincide with the first point.

[0101] S3103: Determine a second point set according to the layout position, the first point, the second point, and the third point.

[0102] In some embodiments, the intersection of a line connecting the layout position and the third point and a first straight line may be determined as the fifth point, where the first straight line is a straight line passing through the first point in the vertical direction.

[0103] An intersection point of a line connecting the layout position and the third point and the second straight line is determined as a sixth point, and the second straight line is a straight line passing through the second point in the vertical direction.

[0104] For example, refer to Figure 4 , you can connect point A and point D, and determine the intersection B' of the line AD and the straight line at point B in the vertical direction as the fifth point, and determine the intersection C' of the line AD and the straight line at point C in the vertical direction as the sixth point.

[0105] It should be noted that the method for determining the first point set and the second point set is not limited to the above embodiment, and any method can be applied to the embodiment of the present application as long as the determined points can reflect the terrain information of the target sector.

[0106] In S320, illustratively, the terrain data of the target sector may include but is not limited to: the elevations of A, B, C, D, and E, the AB slope, BC slope, CD slope, AC slope, and AD slope, the AB elevation difference, AC elevation difference, BC elevation difference, BB' elevation difference, and CC' elevation difference, as well as the horizontal distance between AB, the horizontal distance between AC, the horizontal distance between BC, and the horizontal distance between CD, where point A is the layout position.

[0107] In an embodiment of the present application, preferably, based on the layout position of the wind turbine and the point determination rules, the points reflecting the terrain information of the target sector can be determined, and then the terrain data of the target sector can be determined based on the data of these points. In this way, the terrain information of the target sector can be accurately determined, thereby improving the accuracy of the risk prediction results.

[0108] When using the risk prediction model to determine the risk prediction results of the target sector, the risk prediction model needs to be trained first to improve the prediction performance of the risk prediction model.

[0109] Based on this, the present application embodiment also provides a risk prediction model training method, for example, referring to Figure 5 , the training method of the risk prediction model may include the following steps:

[0110] S510: Obtain training samples.

[0111] The training samples include the historical terrain data of the target sector in the multiple sectors of each wind turbine in the wind farm and the historical wind parameter data /

[0112] S520. Determine a machine learning model for establishing a corresponding relationship between historical terrain data, historical wind parameter data, and the risk of exceeding a limit on cabin acceleration.

[0113] S530. Train a machine learning model based on historical terrain data and historical wind parameter data; if a stopping condition is met, stop training to obtain a trained risk prediction model.

[0114] In an embodiment of the present application, the historical terrain data of the target sector of the wind turbine and the historical wind parameter data of the wind farm are used to train a machine learning model to determine the correspondence between the historical terrain data, the historical wind parameter data and the risk of exceeding the cabin acceleration limit. In this way, the layout position of the wind turbine can be detected by using this correspondence without the need for manual detection, thereby avoiding the influence of personal experience on the detection results and improving the accuracy of the detection results.

[0115] The above steps are explained in detail below:

[0116] In S510, the same wind farm may include multiple wind turbines. Different wind turbines may be divided into sectors using the same division method. For example, each sector may be defined as an area 11.25 degrees to the left and right of the wind turbine's true north, and then a sector may be defined as every 22.5 degrees in a clockwise direction.

[0117] In the embodiment of the present application, the historical terrain data of the target sector of each wind turbine in the wind farm and the historical wind parameter data of the wind farm are used as training samples to train the risk prediction model, thereby increasing the diversity of samples and improving the training effect of the model.

[0118] It should be noted that for different wind turbines within the same wind farm, the sectors at risk of exceeding the nacelle acceleration limit can be different. For example, for wind turbine No. 8, sector No. 1 (11.25 degrees to 33.75 degrees) is at risk of exceeding the nacelle acceleration limit, while for wind turbine No. 11, sector No. 4 (56.25 degrees to 78.75 degrees) is at risk of exceeding the nacelle acceleration limit. Therefore, the target sectors for different wind turbines in this training sample can be different.

[0119] Historical terrain data refers to the terrain data of the target sector within a historical time period, and historical wind parameter data refers to the wind parameter data of the wind farm within the same historical time period. The wind parameter data is the same for different wind turbines within the same wind farm. The details of the terrain data and wind parameter data can be found in the above embodiments and will not be detailed here for the sake of brevity.

[0120] In S520, the machine learning model is used to establish a correspondence between historical terrain data, historical wind parameter data, and the risk of exceeding the cabin acceleration limit. This embodiment of the present application does not limit the type of machine learning model; any model that can establish a correspondence between historical terrain data, historical wind parameter data, and the risk of exceeding the cabin acceleration limit can be used.

[0121] For example, a machine learning model can be selected from existing models. In order to ensure the reliability and stability of the machine learning model, a 5-fold cross-validation method can be used for model selection, which can ensure the reliability and stability of the selected machine learning model. For example, XGBoost can be selected as the machine learning model to be trained.

[0122] In S530, historical terrain data and historical wind parameter data may be input into the machine learning model, which may then output a sample risk prediction result. The sample risk prediction result may be determined by the machine learning model based on sample cabin accelerations in the target sector. For example, the a% quantile of the sample cabin accelerations in the target sector may be determined. If the a% quantile is not less than a preset value, the sample risk prediction result is positive; otherwise, the sample risk prediction result is negative. For example, a = 90, and the preset value is 0.049.

[0123] The sample cabin acceleration can be determined by a machine learning model based on historical terrain data and historical wind parameter data. The specific determination process is not limited in the embodiment of this application.

[0124] The stopping condition is a condition for stopping the training of the machine learning model. For example, the stopping condition may be when the training cycle reaches a preset number of times, or when the sample nacelle acceleration output by the machine learning model and the nacelle acceleration loss value of the wind turbine over a historical time period become stable. In this way, a fully trained risk prediction model is obtained.

[0125] In order to improve the effect of the risk prediction model, before using the training samples to train the machine learning model, the parameters of the machine learning model can be optimized first, and then the machine learning model with optimized parameters can be trained with the training samples, which can improve the training effect of the model.

[0126] Based on this, in some embodiments, after S520, the method may further include the following steps:

[0127] The parameters of the machine learning model are optimized using a grid search method to obtain the value of the model evaluation indicator (Area Under Curve, AUC) for evaluating the predictive performance of the machine learning model.

[0128] Among them, grid search is a parameter adjustment method that can select a smaller finite set of model hyperparameters for search, and then arrange and combine the possible values ​​of these hyperparameters to generate all possible combination results to obtain a "grid".

[0129] AUC is defined as the area under the ROC curve. The ROC curve is a curve drawn based on a series of different binary classification methods (cutoff values ​​or thresholds), with the true positive rate (sensitivity) as the vertical axis and the false positive rate (1-specificity) as the horizontal axis. It reflects the ability of the model to classify unbalanced samples.

[0130] A set of hyperparameters corresponds to an AUC value. The larger the AUC value, the better the classification effect of the model and the better the performance. In an embodiment of the present application, the above-mentioned training samples can be divided into a training set and a validation set, and the above-mentioned machine learning model can be trained using the training set to obtain the AUC value of each group of hyperparameters for the training set (training AUC), and then the validation set is used for validation to obtain the AUC value of each group of hyperparameters for the validation set (validation AUC), and the hyperparameter with the largest AUC value in the validation AUC is used as the parameter optimization result of the above-mentioned machine learning model.

[0131] After parameter optimization, the training samples can be used to train the machine learning model after parameter optimization to obtain a trained risk prediction model. This can improve the training effect of the risk prediction model and thus improve the accuracy of the risk prediction results.

[0132] To improve the training efficiency of the model, in some embodiments, the above S530 may include the following steps:

[0133] Determine a first target sector according to the wind direction corresponding to the historical wind parameter data within a predetermined time, where the first target sector is a sector within the target sector;

[0134] Preprocessing the historical wind parameter data to obtain first historical wind parameter data;

[0135] A machine learning model is trained based on the historical terrain data of the first target sector and the first historical wind parameter data.

[0136] In the embodiment of the present application, the wind direction of the historical wind parameter data may belong to the target sector or may not belong to the target sector. If the wind direction of the historical wind parameter data belongs to the target sector within a predetermined time, the sector to which the wind direction belongs can be determined as the first target sector.

[0137] For example, if the target sectors are sectors 0 and 1, and the wind direction of the historical wind parameter data within the predetermined time period belongs to sector 1, sector 1 can be determined as the first target sector. For another example, if the wind direction of the historical wind parameter data within the predetermined time period belongs to sector 0, sector 0 can be determined as the first target sector.

[0138] If the wind direction of the historical wind parameter data within the predetermined time does not belong to the target sector, as an option, the target sector can be abandoned, that is, the training sample does not include the historical terrain data and historical wind parameter data of the target sector.

[0139] In an embodiment of the present application, the target sector is screened using the wind direction of the historical wind parameter data within a predetermined time to obtain a first target sector that matches the wind direction. In this way, when the above-mentioned machine learning model is trained based on the terrain data and wind parameter data of the first target sector, the accuracy of the model training results can be improved.

[0140] In some embodiments, considering that the wind parameter data has many dimensions, the wind parameter data can be subjected to dimensionality reduction processing, which can improve the training efficiency of the model.

[0141] Taking into account that some wind parameter data have little correlation with terrain data, ie, have little impact on the high-frequency vibration risk of the unit, in some embodiments, the historical wind parameter data may be preprocessed to obtain first historical wind parameter data.

[0142] Exemplarily, feature screening may be performed on historical wind parameter data to extract historical wind parameter data having specific wind parameter features; and the extracted historical wind parameter data may be standardized to obtain first historical wind parameter data.

[0143] Specific wind parameter features may be features that are highly correlated with terrain data. For example, specific wind parameter features may include but are not limited to shear_big, shear_e_mean, Tur_max, Tur_mean, speed, Over20_ratio, direction_transform, Max_min_ratio, etc. The meaning of each parameter can be found in the above embodiments.

[0144] In order to avoid the influence of extreme data on the training results, illustratively, the embodiment of the present application standardizes the historical wind parameter data of the extracted specific wind parameter features to obtain the first historical wind parameter data.

[0145] For example, the extracted historical wind parameter data can be standardized using the following formula:

[0146]

[0147] Among them, x * is the historical wind parameter data after the standardization of a specific wind parameter feature, that is, the first historical wind parameter data, x is the historical wind parameter data before the standardization of the specific wind parameter feature, x max and x min They are respectively the historical maximum and minimum values ​​of the specific wind parameter feature within the predetermined time.

[0148] After the first target sector and the first historical wind parameter data are determined, the terrain data of the first target sector can be merged with the first historical wind parameter data, and then the merged data can be used to train the above-mentioned machine learning model, which can improve the training efficiency and training effect of the model.

[0149] In an embodiment of the present application, a machine learning model is established and trained using training samples, so that the trained machine learning model (risk prediction model) can automatically determine whether there is a risk of excessive cabin acceleration in the corresponding sector based on the input terrain data and wind parameter data, thereby realizing automated detection of the layout position and avoiding the unreliability and judgment errors of human judgment based on experience.

[0150] Taking into account the mutual influence between adjacent sectors, in order to improve the training effect of the model, in some embodiments, as an option, historical terrain data of sectors adjacent to the first target sector can also be obtained to train the above-mentioned machine learning model.

[0151] Specifically, the above-mentioned “training a machine learning model based on historical terrain data and historical wind parameter data” may include the following steps:

[0152] Determine a first target sector according to the wind direction corresponding to the historical wind parameter data within a predetermined time, where the first target sector is a sector within the target sector;

[0153] Preprocessing the historical wind parameter data to obtain first historical wind parameter data;

[0154] A machine learning model including adjacent sector elements is trained based on the historical terrain data of the first target sector, the historical terrain data of adjacent sectors adjacent to the first target sector, and the first historical wind parameter data.

[0155] For example, the first target sector is sector 1, and the sectors adjacent to the first target sector include sector 0 and sector 2, then the historical terrain data of sector 0 and sector 2 can be obtained. Then, the historical terrain data of sector 0 (the sector adjacent to the first target sector), sector 1 (the first target sector) and sector 2 (the sector adjacent to the first target sector) and the first historical wind parameter data are used to train the above-mentioned machine learning model, so as to improve the training effect of the model. Based on the same inventive concept, the embodiment of the present application also provides a layout position detection device for a wind turbine generator set, which is described below in combination with Figure 6 The layout position detection device of the wind turbine generator set provided in the embodiment of the present application is described in detail.

[0156] Figure 6 This is a structural diagram of a layout position detection device for a wind turbine generator set provided in an embodiment of the present application.

[0157] like Figure 6 As shown, the layout position detection device of the wind turbine generator set may include:

[0158] The data acquisition module 61 is used to divide the wind turbines in the wind farm into multiple sectors, and for a target sector among the multiple sectors, obtain the current terrain data of the target sector and the current wind parameter data of the wind farm;

[0159] A risk prediction result determination module 62 is configured to input current terrain data and current wind parameter data into a trained risk prediction model to obtain a risk prediction result for the target sector. The risk prediction model is configured to characterize the correspondence between terrain data, wind parameter data, and the risk of exceeding the cabin acceleration limit. The risk prediction result is configured to characterize whether the target sector has a high-frequency vibration risk.

[0160] The detection module 63 is used to detect the layout position of the wind turbine generator system according to the risk prediction result.

[0161] The layout position detection method for a wind turbine provided in an embodiment of the present application divides a wind turbine in a wind farm into multiple sectors. For a target sector among the multiple sectors, the current terrain data of the target sector and the current wind parameter data of the wind farm are obtained; the current terrain data and the current wind parameter data are input into a trained risk prediction model to obtain a risk prediction result for the target sector; and based on the risk prediction result, the layout position of the wind turbine is detected. That is, the embodiment of the present application can detect whether the layout position of the wind turbine has a high-frequency vibration risk based on the predetermined correspondence between terrain data, wind parameter data, and the risk of exceeding the limit of nacelle acceleration, combined with the current terrain data of the target sector and the current wind parameter data of the wind farm, without the need for manual judgment, thereby avoiding the influence of personal experience on the detection results and improving the accuracy of the detection results.

[0162] In some embodiments, the detection module 63 is specifically configured to:

[0163] In response to the high-frequency vibration risk existing in the target sector, the target sector of the layout position of the wind turbine is closed through the yaw system, or the layout position of the wind turbine is adjusted.

[0164] In some embodiments, the detection module 63 is specifically configured to:

[0165] In response to the existence of high-frequency vibration risk in the target sector, the current terrain data of the adjacent sector adjacent to the target sector and the current wind parameter data of the wind field are input into the trained risk prediction model containing adjacent sector elements to determine whether the adjacent sector has high-frequency vibration risk.

[0166] In some embodiments, the layout position detection device of the wind turbine generator set may further include:

[0167] a point set determination module configured to, after the acquisition module 61 divides the wind turbines in the wind farm into multiple sectors, determine, for a target sector among the multiple sectors, a first point set and a second point set associated with the layout position of the wind turbines based on the layout position and point determination rules before acquiring current terrain data of the target sector and current wind parameter data of the wind farm;

[0168] The acquisition module 61 is specifically used to:

[0169] Obtain current terrain data corresponding to the layout position, each point in the first point set, and each point in the second point set.

[0170] In some embodiments, the first set of points includes a first point, a second point, a third point, and a fourth point;

[0171] The point set determination module includes:

[0172] a first determining unit, configured to determine a first point within the target sector based on a relationship between a first elevation of the first candidate point and a second elevation of the layout position, the first candidate point being a point within the target sector whose horizontal distance from the layout position satisfies a first preset condition;

[0173] a second determining unit, configured to determine a second point, a third point, and a fourth point in the target sector according to the layout position and the first point;

[0174] The third determining unit is configured to determine a second point set according to the layout position, the first point, the second point, and the third point.

[0175] In some embodiments, the second determining unit is specifically configured to:

[0176] If the first elevation is less than the second elevation and the horizontal distance between the layout position and the first point is less than the first threshold, a second point whose elevation meets a second preset condition is determined from the second area, where the second area is an area within the target sector between points whose horizontal distances from the first point meet a third preset condition.

[0177] Determine a point in the target sector whose horizontal distance from the second point satisfies a fourth preset condition as a fourth point;

[0178] A third point is determined according to the second point and the fourth point, and the third point is located between the second point and the fourth point.

[0179] In some embodiments, the second set of points includes a fifth point and a sixth point;

[0180] The third determining unit is specifically configured to:

[0181] Determine the intersection of the line connecting the layout position and the third point and the first straight line as the fifth point, where the first straight line is a straight line passing through the first point in the vertical direction;

[0182] An intersection point of a line connecting the layout position and the third point and the second straight line is determined as a sixth point, and the second straight line is a straight line passing through the second point in the vertical direction.

[0183] The layout position detection device of the wind turbine generator set provided in the embodiment of the present application can achieve Figures 1-4 To avoid repetition, the various processes in the embodiment of the method for detecting the layout position of a wind turbine generator set will not be described again.

[0184] Based on the same inventive concept, the present application embodiment also provides a risk prediction model training device. Figure 7 The training device for the risk prediction model provided in the embodiment of the present application is described in detail.

[0185] Figure 7 A structural diagram of a risk prediction model training device provided in an embodiment of the present application.

[0186] like Figure 7 As shown, the training device of the risk prediction model may include:

[0187] A training sample acquisition module 71 is configured to acquire training samples, where the training samples include historical terrain data of a target sector among multiple sectors of each wind turbine in the wind farm and historical wind parameter data of the wind farm;

[0188] a machine learning model determination module 72 for determining a machine learning model for establishing a correspondence between historical terrain data, historical wind parameter data, and the risk of exceeding a limit on nacelle acceleration;

[0189] A training module 73 is used to train a machine learning model based on historical terrain data and historical wind parameter data;

[0190] If the stopping condition is met, the training is stopped and a risk prediction model that has completed the training is obtained.

[0191] In an embodiment of the present application, the historical terrain data of the target sector of the wind turbine and the historical wind parameter data of the wind farm are used to train a machine learning model to determine the correspondence between the historical terrain data, the historical wind parameter data and the risk of exceeding the cabin acceleration limit. In this way, the layout position of the wind turbine can be detected by using this correspondence without the need for manual detection, thereby avoiding the influence of personal experience on the detection results and improving the accuracy of the detection results.

[0192] In some embodiments, the training module 73 includes:

[0193] a determining unit, configured to determine a first target sector according to a wind direction corresponding to historical wind parameter data within a predetermined time, the first target sector being a sector within the target sector;

[0194] A preprocessing unit, configured to preprocess the historical wind parameter data to obtain first historical wind parameter data;

[0195] A training unit is used to train a machine learning model based on historical terrain data and first historical wind parameter data of a first target sector.

[0196] In some embodiments, the pre-processing unit is specifically configured to:

[0197] Perform feature screening on historical wind parameter data to extract historical wind parameter data with specific wind parameter characteristics;

[0198] The extracted historical wind parameter data is standardized to obtain the first historical wind parameter data.

[0199] In some embodiments, the risk prediction model training device may further include:

[0200] The parameter optimization module is used to optimize the parameters of the machine learning model using a grid search method after the training module 71 trains the machine learning model based on the historical terrain data of the first target sector and the first historical wind parameter data, so as to obtain the value of the model detection indicator AUC for evaluating the predictive performance of the machine learning model.

[0201] In some embodiments, the training module 73 is specifically configured to:

[0202] Determine a first target sector according to the wind direction corresponding to the historical wind parameter data within a predetermined time, where the first target sector is a sector within the target sector;

[0203] Preprocessing the historical wind parameter data to obtain first historical wind parameter data;

[0204] According to the historical terrain data of the first target sector, the historical terrain data of the adjacent sectors adjacent to the first target sector and the first historical wind parameter data, a machine learning model including adjacent sector elements is trained. The training device of the risk prediction model provided by the embodiment of the present application can achieve Figure 5 To avoid repetition, the various processes in the embodiment of the training method of the risk prediction model shown will not be described again here.

[0205] Based on the same inventive concept, an embodiment of the present application further provides an electronic device, which can be a mobile electronic device or a non-mobile electronic device. For example, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, an in-vehicle electronic device, a wearable device, an ultra-mobile personal computer (UMPC), a netbook, or a personal digital assistant (PDA), etc., and the non-mobile electronic device can be a server, a network attached storage (NAS), a personal computer (PC), a television (TV), an ATM, or an kiosks, etc., which are not specifically limited in the embodiments of the present application.

[0206] like Figure 8 As shown, the electronic device may include a processor 81 and a memory 82 for storing computer program instructions.

[0207] The processor 81 may include a central processing unit (CPU) or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0208] The memory 82 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 82 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. In one example, the memory 82 may include removable or non-removable (or fixed) media, or the memory 82 may be a non-volatile solid-state memory. In one example, the memory 82 may be a read-only memory (ROM). In one example, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or a flash memory, or a combination of two or more of these.

[0209] The processor 81 reads and executes the computer program instructions stored in the memory 82 to implement Figure 1-Figure 5 The method in the embodiment shown in FIG. Figure 1-Figure 5The corresponding technical effects achieved by executing the method in the illustrated embodiment are not described in detail here for the sake of brevity.

[0210] In one example, the electronic device may further include a communication interface 83 and a bus 84. Figure 8 As shown, the processor 81, the memory 82, and the communication interface 83 are connected via a bus 84 and communicate with each other.

[0211] The communication interface 83 is mainly used to implement communication between various modules, devices and / or equipment in the embodiments of the present application.

[0212] Bus 84 includes hardware, software or both, and each component of electronic equipment is coupled to each other.For example, but not limitation, bus 84 may include accelerated graphics port (Accelerated Graphics Port, AGP) or other graphics bus, enhanced industry standard architecture (Extended Industry Standard Architecture, EISA) bus, front side bus (Front Side Bus, FSB), hyper transport (Hyper Transport, HT) interconnection, industry standard architecture (Industry Standard Architecture, ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 84 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the application considers any suitable bus or interconnection.

[0213] The electronic device divides the wind turbines in the wind farm into multiple sectors. For a target sector among the multiple sectors, the current terrain data of the target sector and the current wind parameter data of the wind farm are obtained, and then the layout position detection method of the wind turbines in the embodiment of the present application can be executed, thereby realizing the combination of Figures 1-4 The layout position detection method of the wind turbine described Figure 6 A layout position detection device for a wind turbine generator system is described.

[0214] After obtaining the training samples, the electronic device can also execute the training method of the risk prediction model in the embodiment of the present application, thereby realizing the combination of Figure 5 The training method of the risk prediction model described and Figure 7 Describe the training setup for the risk prediction model.

[0215] In addition, in conjunction with the wind turbine layout position detection method or risk prediction model training method in the above-mentioned embodiments, embodiments of the present application may provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by a processor, any of the wind turbine layout position detection methods or risk prediction model training methods in the above-mentioned embodiments is implemented.

[0216] It should be understood that the present application is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted here. In the above embodiments, several specific steps are described and illustrated as examples. However, the method process of the present application is not limited to the specific steps described and illustrated. Those skilled in the art can make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present application.

[0217] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium that can store or transmit information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0218] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or systems based on a series of steps or devices. However, this application is not limited to the order of the above steps. In other words, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0219] Aspects of the present invention are described above with reference to the flowchart and / or block diagram of the method, device (system) and computer program product according to the embodiment of the present invention.It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer program instructions.These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed via the processor of the computer or other programmable data processing device enable the implementation of the function / action specified in one or more boxes of the flowchart and / or block diagram.Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit.It is also understood that each box in the block diagram and / or flowchart and the combination of the boxes in the block diagram and / or flowchart can also be implemented by the dedicated hardware that performs the specified function or action, or can be implemented by the combination of dedicated hardware and computer instructions.

[0220] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms without departing from the purpose of this application and the scope of protection of the claims, all of which are within the protection of this application.

Claims

1. A method for detecting the layout position of a wind turbine generator set, characterized in that: The method comprises: Divide the wind turbines of the wind farm into multiple sectors, and for a target sector among the multiple sectors, obtain current terrain data of the target sector and current wind parameter data of the wind farm; Inputting the current terrain data and the current wind parameter data into a trained risk prediction model to obtain a risk prediction result for the target sector, wherein the risk prediction model is used to characterize a correspondence between the terrain data, the wind parameter data, and a risk of exceeding a limit on the cabin acceleration, and the risk prediction result is used to characterize whether the target sector has a high-frequency vibration risk; detecting the layout position of the wind turbine generator system according to the risk prediction result; The sector is an area obtained by dividing the terrain where the wind turbine is located with the layout position of the wind turbine as the center of the circle; Detecting the layout position of the wind turbine generator system according to the risk prediction result includes: In response to the high-frequency vibration risk existing in the target sector, closing the target sector of the layout position of the wind turbine generator set through a yaw system, or adjusting the layout position of the wind turbine generator set; The adjusting the layout position of the wind turbine generator set includes: An adjustment strategy for the layout position is determined based on the current terrain data and the current wind parameter data, prompt information including a risk prediction result for the layout position and the adjustment strategy is generated, and the prompt information is sent to a user so that the user adjusts the layout position based on the adjustment strategy.

2. The method according to claim 1, characterized in that The detecting the layout position of the wind turbine generator system according to the risk prediction result further includes: In response to the existence of high-frequency vibration risk in the target sector, the current terrain data of the adjacent sector adjacent to the target sector and the current wind parameter data of the wind field are input into a trained risk prediction model containing adjacent sector elements to determine whether the adjacent sector has high-frequency vibration risk.

3. The method according to claim 1, characterized in that The wind turbines in the wind farm are divided into a plurality of sectors. Before obtaining current terrain data of a target sector and current wind parameter data of the wind farm from the target sector, the method further comprises: Determining a first point set and a second point set associated with the layout position according to the layout position and point determination rule of the wind turbine generator system; The acquiring of current terrain data of the target sector includes: Current terrain data corresponding to the layout position, each point in the first point set, and each point in the second point set is obtained.

4. The method according to claim 3, characterized in that The first point set includes a first point, a second point, a third point and a fourth point; The determining, according to the layout position and the point determination rule, a first point set and a second point set associated with the layout position includes: determining a first point within the target sector based on a relationship between a first elevation of the first candidate point and a second elevation of the layout position, the first candidate point being a point within the target sector whose horizontal distance from the layout position satisfies a first preset condition; Determining the second point, the third point, and the fourth point within the target sector according to the layout position and the first point; The second point location set is determined according to the layout position, the first point location, the second point location, and the third point location.

5. The method according to claim 4, characterized in that The determining the second point, the third point, and the fourth point in the target sector according to the layout position and the first point includes: If the first elevation is less than the second elevation and the horizontal distance between the layout position and the first point is less than a first threshold, determining a second point within a second area whose elevation satisfies a second preset condition, the second area being an area within the target sector between points whose horizontal distance from the first point satisfies a third preset condition; Determine a point in the target sector whose horizontal distance from the second point satisfies a fourth preset condition as the fourth point; The third point is determined according to the second point and the fourth point, and the third point is located between the second point and the fourth point.

6. The method according to claim 4, characterized in that The second point set includes a fifth point and a sixth point; The determining the second point set according to the layout position, the first point, the second point, and the third point includes: Determine an intersection of a line connecting the layout position and the third point and a first straight line as the fifth point, where the first straight line is a straight line passing through the first point in a vertical direction; An intersection point of a line connecting the layout position and the third point and a second straight line is determined as the sixth point, where the second straight line is a straight line passing through the second point in the vertical direction.

7. The method according to claim 1, characterized in that The method comprises: Acquire a training sample, wherein the training sample includes historical terrain data of a target sector among multiple sectors of each wind turbine generator set in a wind farm and historical wind parameter data of the wind farm; Determine a machine learning model for establishing a correspondence between historical terrain data, historical wind parameter data, and the risk of exceeding the cabin acceleration limit; Training the machine learning model based on the historical terrain data and the historical wind parameter data; If the stopping condition is met, the training is stopped and a risk prediction model that has completed the training is obtained.

8. The method according to claim 7, characterized in that The training of the machine learning model according to the historical terrain data and the historical wind parameter data includes: Determining a first target sector according to a wind direction corresponding to historical wind parameter data within a predetermined time, where the first target sector is a sector within the target sector; Preprocessing the historical wind parameter data to obtain first historical wind parameter data; The machine learning model is trained based on the historical terrain data of the first target sector and the first historical wind parameter data.

9. The method according to claim 8, characterized in that The preprocessing of the historical wind parameter data to obtain first historical wind parameter data includes: Performing feature screening on the historical wind parameter data to extract historical wind parameter data having specific wind parameter features; The extracted historical wind parameter data is standardized to obtain the first historical wind parameter data.

10. The method according to claim 7, characterized in that After determining the machine learning model for establishing a correspondence between historical terrain data, historical wind parameter data, and the risk of exceeding a limit on the nacelle acceleration, the method further includes: The parameters of the machine learning model are optimized using a grid search method to obtain the value of the model evaluation indicator AUC, which is used to evaluate the predictive performance of the machine learning model.

11. The method according to claim 7, characterized in that The training of the machine learning model according to the historical terrain data and the historical wind parameter data further includes: Determining a first target sector according to a wind direction corresponding to historical wind parameter data within a predetermined time, where the first target sector is a sector within the target sector; Preprocessing the historical wind parameter data to obtain first historical wind parameter data; The machine learning model including adjacent sector elements is trained based on the historical terrain data of the first target sector, the historical terrain data of the adjacent sectors adjacent to the first target sector, and the first historical wind parameter data.

12. A layout position detection device for a wind turbine generator set, characterized in that: The device comprises: a data acquisition module, configured to divide the wind turbines in the wind farm into a plurality of sectors, and acquire, for a target sector among the plurality of sectors, current terrain data of the target sector and current wind parameter data of the wind farm; a risk prediction result determination module, configured to input the current terrain data and the current wind parameter data into a trained risk prediction model to obtain a risk prediction result for the target sector, wherein the risk prediction model is configured to characterize a correspondence between the terrain data, the wind parameter data, and a risk of exceeding a limit on cabin acceleration, and the risk prediction result is configured to characterize whether the target sector has a high-frequency vibration risk; A detection module, configured to detect the layout position of the wind turbine generator system according to the risk prediction result; The sector is an area obtained by dividing the terrain where the wind turbine is located with the layout position of the wind turbine as the center of the circle; The detection module is specifically used to: In response to the high-frequency vibration risk existing in the target sector, closing the target sector of the layout position of the wind turbine generator set through a yaw system, or adjusting the layout position of the wind turbine generator set; The adjusting the layout position of the wind turbine generator set includes: An adjustment strategy for the layout position is determined based on the current terrain data and the current wind parameter data, prompt information including a risk prediction result for the layout position and the adjustment strategy is generated, and the prompt information is sent to a user so that the user adjusts the layout position based on the adjustment strategy.

13. The device according to claim 12, characterized in that The device comprises: A training sample acquisition module is used to acquire training samples, wherein the training samples include historical terrain data of a target sector among multiple sectors of each wind turbine in the wind farm and historical wind parameter data of the wind farm; a machine learning model determination module, configured to determine a machine learning model for establishing a corresponding relationship between historical terrain data, historical wind parameter data, and the risk of exceeding a limit on cabin acceleration; A training module, configured to train the machine learning model based on the historical terrain data and the historical wind parameter data; If the stopping condition is met, the training is stopped and a risk prediction model that has completed the training is obtained.

14. An electronic device, characterized in that: include: processor; a memory for storing computer program instructions; When the computer program instructions are executed by the processor, the method according to any one of claims 1 to 11 is implemented.

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

Images