Method and device for detecting bearing wear of a wind turbine generator

Abstract

Provided are a bearing wear detection method and device for a wind turbine. The bearing wear detection method comprises: obtaining operation data of a wind turbine, wherein the operation data comprises a rotational speed of the wind turbine and a nacelle acceleration; determining a data characteristic index corresponding to each rotational speed interval, wherein the data characteristic index of each rotational speed interval is used to represent a characteristic of the operation data in which the rotational speed belongs to the rotational speed interval; and detecting whether the generator bearing wear of the wind turbine is abnormal according to the data characteristic index of each rotational speed interval.

Description

Technical Field

[0001] This disclosure generally relates to the field of power technology, and more specifically, to a method and apparatus for detecting bearing wear in wind turbine generators. Background Technology

[0002] Wind turbine generator bearings are a crucial transmission component of wind turbine generators. However, because wind turbine generators are installed at high locations and are exposed to harsh environments such as humidity, corrosion, and strong winds for extended periods, the mechanical structure of the generator bearings is prone to damage. The main function of generator bearings is to support the generator rotor, enabling it to rotate, while controlling its axial and radial movement, reducing the coefficient of friction during its movement, and ensuring its rotational accuracy.

[0003] However, due to factors such as initial installation, prolonged poor lubrication, and harsh working environments, generator bearings are prone to wear, leading to abnormal generator vibration, reduced generator lifespan, and unstable wind turbine operation. Furthermore, abnormal vibration can cause frequent turbine shutdowns and even safety accidents. Additionally, it increases the coefficient of friction during operation, reducing rotational accuracy and impacting power generation performance. Moreover, due to the operating environment, severe bearing wear not only increases spare parts and maintenance costs for the wind farm but also results in significant power generation losses due to prolonged downtime during spare parts replacement and frequent turbine shutdowns during operation. Summary of the Invention

[0004] An exemplary embodiment of this disclosure provides a method and apparatus for detecting bearing wear in wind turbine generators, which can conveniently and effectively detect whether the bearing wear of wind turbine generators is abnormal.

[0005] According to a first aspect of the present disclosure, a method for detecting bearing wear in a wind turbine is provided, comprising: acquiring operating data of the wind turbine, wherein the operating data includes the rotational speed and nacelle acceleration of the wind turbine; determining data feature indicators corresponding to each rotational speed range, wherein the data feature indicators for each rotational speed range are used to characterize the features of operating data whose rotational speed belongs to that rotational speed range; and detecting whether the wear of the generator bearing of the wind turbine is abnormal based on the data feature indicators for each rotational speed range.

[0006] Optionally, the step of determining the data characteristic indicators corresponding to each speed range includes: aggregating the operating data according to the speed to obtain aggregated data, wherein the aggregated data includes the speed and the statistical value of the nacelle acceleration corresponding to the speed; for each speed range, determining the data characteristic indicators of the speed range based on the aggregated data of the speed range belonging to the speed range.

[0007] Optionally, the step of aggregating the operating data according to the rotational speed to obtain aggregated data includes: adjusting the precision of the rotational speed in the operating data; and aggregating the operating data with the same rotational speed in the adjusted operating data to obtain aggregated data.

[0008] Optionally, the statistical values ​​of the cabin acceleration include: the standard deviation Y of the cabin acceleration in the y-direction and the standard deviation X of the cabin acceleration in the x-direction.

[0009] Optionally, each speed range includes at least two of the following speed ranges: a first speed range, a second speed range, and a third speed range; wherein, the first speed range is the stable speed range of the wind turbine that is least affected by the turbine system; the second speed range is the speed range from the wind turbine's speed during stable grid-connected operation to the speed at which the power reaches the rated power; and the third speed range is the speed range from the wind turbine's grid-connected speed to the speed corresponding to the cut-out wind speed.

[0010] Optionally, the data characteristic indicators corresponding to the first speed range include: the amount of data with standard deviation Y exceeding a first standard deviation threshold; and / or, the data characteristic indicators corresponding to the second speed range include at least one of the following: the coefficient of variation of standard deviation Y, the amount of data with standard deviation Y exceeding a second standard deviation threshold, the absolute value of kurtosis of standard deviation Y, the maximum value of standard deviation Y, the difference between the maximum and minimum values ​​of standard deviation Y, the speed of the aggregated data with the largest standard deviation Y, the maximum difference of standard deviation Y, and the ratio of the maximum difference to the coefficient of variation; and / or, the data characteristic indicators corresponding to the third speed range include at least one of the following: the correlation coefficient between speed and standard deviation Y, the correlation coefficient between speed and standard deviation X, the correlation coefficient between speed and standard deviation deviation, the ratio between the amount of data with standard deviation deviation greater than 0 and the amount of data with standard deviation deviation less than or equal to 0, the ratio between the sum of standard deviation deviations greater than a first deviation threshold and the sum of standard deviation deviations less than a second deviation threshold, the proportion of data with standard deviation deviation greater than 0, and the maximum length of consecutive standard deviation deviations greater than 0, wherein the standard deviation deviation is the difference between the standard deviation Y and the standard deviation X of the same aggregated data.

[0011] Optionally, the step of detecting whether the wear of the generator bearing of the wind turbine is abnormal based on the data characteristic indicators of each speed range includes: determining whether the data characteristic indicators of each speed range meet the corresponding preset conditions; and determining that the wear of the generator bearing of the wind turbine is abnormal in response to the data characteristic indicators of at least two speed ranges meeting the corresponding preset conditions.

[0012] Optionally, the preset conditions corresponding to the first speed range include the following sub-conditions: the amount of data with standard deviation Y exceeding the first standard deviation threshold is greater than or equal to N1; and / or, the preset conditions corresponding to the second speed range include at least one of the following sub-conditions: the coefficient of variation of standard deviation Y is greater than or equal to the coefficient of variation threshold; the amount of data with standard deviation Y exceeding the second standard deviation threshold is greater than N2; the absolute value of the kurtosis of standard deviation Y is greater than the kurtosis threshold; the maximum value of standard deviation Y is greater than or equal to the maximum value threshold; the difference between the maximum and minimum values ​​of standard deviation Y is greater than or equal to the difference threshold; the speed of the aggregated data with the largest standard deviation Y is greater than the first speed threshold; the speed of the aggregated data with the largest standard deviation Y is less than the second speed threshold; the maximum difference value of standard deviation Y is less than the maximum difference value threshold; and the ratio of the maximum difference value to the coefficient of variation is less than or equal to the first ratio threshold. ; and / or, the preset conditions corresponding to the third speed range include at least one of the following sub-conditions: the correlation coefficient between speed and standard deviation Y is greater than the first coefficient threshold, the correlation coefficient between speed and standard deviation X is less than or equal to the second coefficient threshold, the correlation coefficient between speed and standard deviation deviation is greater than the third coefficient threshold, the ratio between the amount of data with standard deviation deviation greater than 0 and the amount of data with standard deviation deviation less than or equal to 0 is greater than the second ratio threshold, the ratio between the sum of standard deviation deviations greater than the first deviation threshold and the sum of standard deviation deviations less than the second deviation threshold is greater than the third ratio threshold, the proportion of data with standard deviation deviation greater than 0 is greater than or equal to the proportion threshold, and the maximum length of consecutive standard deviation deviations greater than 0 is greater than or equal to the length threshold, wherein the standard deviation deviation is the difference between the standard deviation Y and the standard deviation X of the same aggregated data, and N1 and N2 are positive integers.

[0013] Optionally, the step of determining whether the data characteristic indicators of each speed range meet the corresponding preset conditions includes: for each speed range, when a preset number of sub-conditions of the speed range are met, determining that the data characteristic indicators of the speed range meet the corresponding preset conditions.

[0014] Optionally, it further includes: displaying detection results, wherein the detection results are used to indicate whether the wear of the generator bearing of the wind turbine is abnormal, and / or the operating characteristics of the wind turbine when the wear of the generator bearing is abnormal.

[0015] According to a second aspect of the present disclosure, a bearing wear detection device for a wind turbine is provided, comprising: a data acquisition unit configured to acquire operating data of the wind turbine, wherein the operating data includes the rotational speed and nacelle acceleration of the wind turbine; a feature determination unit configured to determine data feature indicators corresponding to each rotational speed range, wherein the data feature indicators for each rotational speed range are used to characterize the features of operating data whose rotational speed belongs to that rotational speed range; and a wear detection unit configured to detect whether the wear of the generator bearing of the wind turbine is abnormal based on the data feature indicators of each rotational speed range.

[0016] Optionally, the feature determination unit is configured to: aggregate the operating data according to the rotational speed to obtain aggregated data, wherein the aggregated data includes the rotational speed and the statistical value of the nacelle acceleration corresponding to the rotational speed; and for each rotational speed range, determine the data feature index of the rotational speed range based on the aggregated data of the rotational speed range belonging to the rotational speed range.

[0017] Optionally, the feature determination unit is configured to: adjust the accuracy of the rotation speed in the operating data; and aggregate the operating data with the same rotation speed in the adjusted operating data to obtain aggregated data.

[0018] Optionally, the statistical values ​​of the cabin acceleration include: the standard deviation Y of the cabin acceleration in the y-direction and the standard deviation X of the cabin acceleration in the x-direction.

[0019] Optionally, each speed range includes at least two of the following speed ranges: a first speed range, a second speed range, and a third speed range; wherein, the first speed range is the stable speed range of the wind turbine that is least affected by the turbine system; the second speed range is the speed range from the wind turbine's speed during stable grid-connected operation to the speed at which the power reaches the rated power; and the third speed range is the speed range from the wind turbine's grid-connected speed to the speed corresponding to the cut-out wind speed.

[0020] Optionally, the data characteristic indicators corresponding to the first speed range include: the amount of data with standard deviation Y exceeding a first standard deviation threshold; and / or, the data characteristic indicators corresponding to the second speed range include at least one of the following: the coefficient of variation of standard deviation Y, the amount of data with standard deviation Y exceeding a second standard deviation threshold, the absolute value of kurtosis of standard deviation Y, the maximum value of standard deviation Y, the difference between the maximum and minimum values ​​of standard deviation Y, the speed of the aggregated data with the largest standard deviation Y, the maximum difference of standard deviation Y, and the ratio of the maximum difference to the coefficient of variation; and / or, the data characteristic indicators corresponding to the third speed range include at least one of the following: the correlation coefficient between speed and standard deviation Y, the correlation coefficient between speed and standard deviation X, the correlation coefficient between speed and standard deviation deviation, the ratio between the amount of data with standard deviation deviation greater than 0 and the amount of data with standard deviation deviation less than or equal to 0, the ratio between the sum of standard deviation deviations greater than a first deviation threshold and the sum of standard deviation deviations less than a second deviation threshold, the proportion of data with standard deviation deviation greater than 0, and the maximum length of consecutive standard deviation deviations greater than 0, wherein the standard deviation deviation is the difference between the standard deviation Y and the standard deviation X of the same aggregated data.

[0021] Optionally, the wear detection unit is configured to: determine whether the data characteristic indicators of each speed range meet the corresponding preset conditions; and in response to the data characteristic indicators of at least two speed ranges meeting the corresponding preset conditions, determine that the generator bearing of the wind turbine is abnormally worn.

[0022] Optionally, the preset conditions corresponding to the first speed range include the following sub-conditions: the amount of data with standard deviation Y exceeding the first standard deviation threshold is greater than or equal to N1; and / or, the preset conditions corresponding to the second speed range include at least one of the following sub-conditions: the coefficient of variation of standard deviation Y is greater than or equal to the coefficient of variation threshold; the amount of data with standard deviation Y exceeding the second standard deviation threshold is greater than N2; the absolute value of the kurtosis of standard deviation Y is greater than the kurtosis threshold; the maximum value of standard deviation Y is greater than or equal to the maximum value threshold; the difference between the maximum and minimum values ​​of standard deviation Y is greater than or equal to the difference threshold; the speed of the aggregated data with the largest standard deviation Y is greater than the first speed threshold; the speed of the aggregated data with the largest standard deviation Y is less than the second speed threshold; the maximum difference value of standard deviation Y is less than the maximum difference value threshold; and the ratio of the maximum difference value to the coefficient of variation is less than or equal to the first ratio threshold. ; and / or, the preset conditions corresponding to the third speed range include at least one of the following sub-conditions: the correlation coefficient between speed and standard deviation Y is greater than the first coefficient threshold, the correlation coefficient between speed and standard deviation X is less than or equal to the second coefficient threshold, the correlation coefficient between speed and standard deviation deviation is greater than the third coefficient threshold, the ratio between the amount of data with standard deviation deviation greater than 0 and the amount of data with standard deviation deviation less than or equal to 0 is greater than the second ratio threshold, the ratio between the sum of standard deviation deviations greater than the first deviation threshold and the sum of standard deviation deviations less than the second deviation threshold is greater than the third ratio threshold, the proportion of data with standard deviation deviation greater than 0 is greater than or equal to the proportion threshold, and the maximum length of consecutive standard deviation deviations greater than 0 is greater than or equal to the length threshold, wherein the standard deviation deviation is the difference between the standard deviation Y and the standard deviation X of the same aggregated data, and N1 and N2 are positive integers.

[0023] Optionally, the wear detection unit is configured to: for each speed range, when a preset number of sub-conditions for that speed range are met, determine that the data characteristic indicators of that speed range meet the corresponding preset conditions.

[0024] Optionally, it further includes: a result display unit configured to display detection results, wherein the detection results are used to indicate whether the wear of the generator bearing of the wind turbine is abnormal, and / or the operating characteristics of the wind turbine when the wear of the generator bearing is abnormal.

[0025] According to a third aspect of the present disclosure, a computer-readable storage medium storing a computer program is provided, which, when executed by a processor, causes the processor to perform the bearing wear detection method for a wind turbine as described above.

[0026] According to a fourth aspect of the present disclosure, a bearing wear detection device for a wind turbine is provided. The bearing wear detection device includes: a processor; and a memory storing a computer program, which, when executed by the processor, causes the processor to perform the bearing wear detection method for a wind turbine as described above.

[0027] The bearing wear detection method and apparatus for wind turbines according to the exemplary embodiments of this disclosure utilize the characteristic that the bearing wear process leads to abnormal vibration of the unit. It directly uses the unit's operating data to detect the wear of the generator bearings without the need for additional detection equipment. This method can conveniently and effectively detect whether the generator bearings of wind turbines are worn abnormally, providing strong theoretical support for on-site operation and maintenance personnel and effectively reducing operation and maintenance costs.

[0028] Further aspects and / or advantages of the general concept of this disclosure will be set forth in part in the description which follows, and in part will be clear from the description or may be learned by practice of the general concept of this disclosure. Attached Figure Description

[0029] The above and other objects and features of exemplary embodiments of this disclosure will become clearer from the following description taken in conjunction with the accompanying drawings, which exemplarily illustrate the embodiments, wherein:

[0030] Figure 1 A flowchart illustrating a bearing wear detection method for a wind turbine according to an exemplary embodiment of the present disclosure;

[0031] Figure 2 A schematic diagram illustrating characteristic wear of a generator bearing according to an exemplary embodiment of the present disclosure is shown.

[0032] Figure 3 A structural block diagram of a bearing wear detection device for a wind turbine generator according to an exemplary embodiment of the present disclosure is shown.

[0033] Figure 4 A structural block diagram of a bearing wear detection device for a wind turbine generator according to another exemplary embodiment of the present disclosure is shown. Detailed Implementation

[0034] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings, examples of which are illustrated in the drawings, wherein the same reference numerals always refer to the same parts. The embodiments will now be described with reference to the accompanying drawings in order to explain this disclosure.

[0035] This disclosure proposes using the operating data of the wind turbine unit to achieve generator bearing wear detection. As an example, a processor can read the wind turbine unit's operating data from a data source. Then, based on the wind turbine unit's generator speed, wind speed, nacelle x-axis acceleration, nacelle y-axis acceleration, and other operating data, the processor performs data acquisition, data cleaning, data processing, feature extraction, and anomaly detection to effectively detect wind turbine generator bearing anomalies. Furthermore, the processor can provide the bearing wear detection results to a display device for display, thereby providing support to on-site maintenance personnel. For example, the wind turbine unit's controller or the wind farm's controller can read the wind turbine unit's operating data from a SCADA data source and use this data to perform generator bearing wear detection. The bearing wear detection results can also be provided to a display device in the control room for display.

[0036] Figure 1 A flowchart illustrating a bearing wear detection method for a wind turbine according to an exemplary embodiment of the present disclosure is shown.

[0037] Reference Figure 1 In step S10, the operating data of the wind turbine is obtained.

[0038] The operating data includes the wind turbine's rotational speed and nacelle acceleration. For example, the rotational speed may be the generator speed of the wind turbine. For example, the nacelle acceleration may specifically include: nacelle acceleration in the x-direction and / or nacelle acceleration in the y-direction. For example, the x-direction may refer to the lateral direction of the nacelle, and the y-direction may refer to the longitudinal direction of the nacelle.

[0039] As an example, step S10 may include: reading multiple real-time data points of the wind turbine over a period of time, and then cleaning the read real-time data to obtain the operating data.

[0040] As an example, multiple real-time data points of the wind turbine over a period of time can be read from the SCADA data source.

[0041] As an example, the data fields to be read may include: generator speed, wind speed, nacelle x-axis acceleration, and nacelle y-axis acceleration.

[0042] As an example, the length of the time period can be adjusted according to the data acquisition frequency; for example, at least 200,000 real-time data entries can be read.

[0043] As an example, the steps for cleaning the read real-time data may include: removing abnormal data (e.g., data including null values) from the read real-time data due to problems such as acquisition and transmission; then, when the amount of jump data in the real-time data after removing abnormal data meets certain conditions, selecting real-time data from the real-time data after removing abnormal data where both the x-direction nacelle acceleration and the y-direction nacelle acceleration are within a reasonable and reliable range (e.g., the default range can be -0.45 to 0.45, which can be adjusted according to the actual situation and specific needs).

[0044] As an example, the filtered data must meet the requirement of having sufficient data above a certain wind speed value, and the amount of filtered data must reach a certain percentage of the original data (for example, it is generally recommended not to be less than 30%). For example, the amount of filtered data with wind speeds exceeding a first wind speed threshold should not be less than a first preset data amount, and the amount of filtered data with wind speeds exceeding a second wind speed threshold should not be less than a second preset data amount.

[0045] In step S20, based on the operating data, the data characteristic indicators corresponding to each speed range are determined.

[0046] The data feature indexes for each speed range are used to characterize the features of the operating data that belong to that speed range.

[0047] As an example, each speed range is preset. For instance, the speed range can be preset according to the specific operating conditions of the wind turbine.

[0048] As an example, each speed range may include at least two of the following speed ranges: a first speed range, a second speed range, and a third speed range.

[0049] As an example, the first speed range can be the stable speed range of the wind turbine that is least affected by the turbine system. In other words, within this speed range, the speed is least affected by systems such as the electrical system and yaw system. For example, this speed range can be the speed range corresponding to a relatively stable full-load state and a state near full-load. The first speed range can be obtained based on experience and the stable operation of the wind turbine.

[0050] As an example, the second speed range can be the speed range from the wind turbine's speed during stable grid-connected operation to the speed when the power reaches the rated power.

[0051] As an example, the third speed range can be the speed range from the grid-connected speed of the wind turbine to the speed corresponding to the cut-off wind speed. As an example, the difference between the maximum and minimum values ​​in the third speed range must be greater than a certain value (e.g., greater than 4). As an example, the amount of aggregated data for speeds belonging to this speed range must reach a certain amount.

[0052] As an example, the first speed range, the second speed range, and the third speed range may partially overlap. As an example, the first speed range may be the narrowest of all speed ranges, and the third speed range may be the widest of all speed ranges. For example, the third speed range may include the second speed range, and the second speed range may include the first speed range. For example, when the wind turbine is a direct-drive wind turbine, the range of the first speed range may be 9-11 rpm, the range of the second speed range may be 7-15 rpm, and the range of the third speed range may be 5-18 rpm.

[0053] As an example, step S20 may include: aggregating the operating data according to the rotational speed to obtain aggregated data; and for each rotational speed range, determining the data characteristic index of that rotational speed range based on the aggregated data where the rotational speed belongs to that range. The aggregated data includes the rotational speed and the statistical value of the nacelle acceleration corresponding to that rotational speed.

[0054] As an example, the step of aggregating the operating data according to the rotational speed to obtain aggregated data may include: adjusting the precision of the rotational speed in the operating data; and aggregating the operating data with the same rotational speed in the adjusted operating data to obtain aggregated data.

[0055] As an example, the step of adjusting the accuracy of the rotational speed in the operating data may include: rounding the rotational speed value in the operating data so that the rotational speed value retains one decimal place.

[0056] As an example, the step of aggregating operational data with the same rotational speed in the adjusted operational data to obtain aggregated data may include: aggregating all operational data with the same rotational speed in the adjusted operational data to obtain one aggregated data point corresponding to that rotational speed. This aggregated data point includes: the rotational speed and a statistical value obtained by statistically analyzing the nacelle acceleration of all operational data with that rotational speed. It should be understood that each aggregated data point includes one rotational speed, and different aggregated data points include different rotational speeds.

[0057] As an example, the type of the above statistical value may be standard deviation, and it should be understood that it may also be other types of statistical values, which this disclosure does not limit. As an example, the statistical value of the nacelle acceleration may include: the standard deviation Y of the nacelle acceleration in the y-direction and the standard deviation X of the nacelle acceleration in the x-direction. For example, each aggregated data may include: rotational speed, the standard deviation Y of the nacelle acceleration in the y-direction of all operating data with that rotational speed, and the standard deviation X of the nacelle acceleration in the x-direction of all operating data with that rotational speed.

[0058] As an example, for each speed range, the step of determining the data characteristic index of that speed range based on the aggregated data of speeds belonging to that speed range may include: for each speed range, firstly, filtering out the aggregated data of speeds belonging to that speed range, and then, based on the filtered aggregated data, determining the data characteristic index of that speed range.

[0059] In step S30, based on the data characteristic indicators of each speed range, the wear of the generator bearing of the wind turbine is detected to be abnormal.

[0060] As an example, the data characteristic indicators corresponding to the first speed range may include, but are not limited to, the amount of data whose standard deviation Y exceeds the first standard deviation threshold. Here, the amount of data N5 whose standard deviation Y exceeds the first standard deviation threshold may refer to the number of aggregated data whose speed belongs to the first speed range and whose standard deviation Y exceeds the first standard deviation threshold p7.

[0061] As an example, the data characteristic indicators corresponding to the second speed range may include, but are not limited to, at least one of the following: the coefficient of variation of the standard deviation Y, the amount of data in the standard deviation Y that exceeds the second standard deviation threshold, the absolute value of the kurtosis of the standard deviation Y, the maximum value of the standard deviation Y, the difference between the maximum and minimum values ​​of the standard deviation Y, the speed of the aggregated data with the largest standard deviation Y, the maximum difference of the standard deviation Y, and the ratio of the maximum difference to the coefficient of variation.

[0062] Here, the coefficient of variation of standard deviation Y can refer to: the coefficient of variation Coeff of the standard deviation Y of all aggregated data whose rotational speeds belong to the second rotational speed range; the number of data N4 whose standard deviation Y exceeds the second standard deviation threshold can refer to: the number of aggregated data whose rotational speeds belong to the second rotational speed range and whose standard deviation Y exceeds the second standard deviation threshold p8; the absolute value of the kurtosis of standard deviation Y, math.abs(yPeak), can refer to: the absolute value of the kurtosis of the standard deviation Y of all aggregated data whose rotational speeds belong to the second rotational speed range; the maximum value of standard deviation Y, yMax, can refer to: the maximum value among all aggregated data whose rotational speeds belong to the second rotational speed range; the difference between the maximum and minimum values ​​of standard deviation Y, yRange, can refer to: the difference between the maximum and minimum values ​​among all aggregated data whose rotational speeds belong to the second rotational speed range; the rotational speed of the aggregated data with the largest standard deviation Y, rpmMax, can refer to: the rotational speed of the aggregated data with the largest standard deviation Y among all aggregated data whose rotational speeds belong to the second rotational speed range. For example, the formula for calculating the coefficient of variation can be: Let A = (a1, a2, a3, ... an), Coeff = A.std / A.mean.

[0063] Regarding the maximum difference of standard deviation Y, all aggregated data with rotational speeds belonging to the second rotational speed range can be sorted in ascending order according to rotational speed. Then, the difference between the standard deviations Y of two adjacent aggregated data can be obtained as the difference value Diff of standard deviation Y. The maximum difference of standard deviation Y, yDiffMax, can refer to the maximum value among all the obtained difference values ​​Diff.

[0064] As an example, the data characteristic indicators corresponding to the third speed range may include, but are not limited to, at least one of the following: the correlation coefficient between speed and standard deviation Y, the correlation coefficient between speed and standard deviation X, the correlation coefficient between speed and standard deviation deviation, the ratio between the amount of data with standard deviation deviation greater than 0 and the amount of data with standard deviation deviation less than or equal to 0, the ratio between the sum of standard deviation deviations greater than the first deviation threshold and the sum of standard deviation deviations less than the second deviation threshold, the proportion of data with standard deviation deviation greater than 0, and the maximum length of consecutive standard deviation deviations greater than 0, wherein the standard deviation deviation YSubtractX is the difference between the standard deviation Y and the standard deviation X of the same aggregated data.

[0065] Here, the correlation coefficient *corCoefY* between rotational speed and standard deviation *Y* can refer to the correlation coefficient between the rotational speed and standard deviation *Y* for all aggregated data where the rotational speed falls within the third rotational speed range; the correlation coefficient *corCoefX* between rotational speed and standard deviation *X* can refer to the correlation coefficient between the rotational speed and standard deviation *X* for all aggregated data where the rotational speed falls within the third rotational speed range; the correlation coefficient *corCoef* between rotational speed and standard deviation deviation can refer to the correlation coefficient between the rotational speed and standard deviation deviation for all aggregated data where the rotational speed falls within the third rotational speed range; and the ratio *posDivNeg* between the number of aggregated data where the rotational speed falls within the third rotational speed range and the standard deviation deviation is greater than 0 and the number of aggregated data where the rotational speed falls within the third rotational speed range and the standard deviation deviation is less than or equal to 0. The ratio of the number of aggregated data points with a standard deviation less than or equal to 0; the ratio of the sum of standard deviations greater than the first deviation threshold to the sum of standard deviations less than the second deviation threshold. `Integral` can refer to the ratio of the sum of standard deviations of all aggregated data points with a speed within the third speed range and a standard deviation greater than the first deviation threshold (p24), to the sum of standard deviations of all aggregated data points with a speed within the third speed range and a standard deviation less than the second deviation threshold (p24); the percentage of data points with a standard deviation greater than 0. `rpmIncrease` can refer to the ratio of the number of aggregated data points with a speed within the third speed range and a standard deviation greater than 0 to the total number of aggregated data points with a speed within the third speed range. For example, the second deviation threshold can be a negative of the first deviation threshold.

[0066] Regarding the maximum length of consecutive standard deviations greater than 0, yMoreXMaxLen, all aggregated data with rotational speeds belonging to the third rotational speed range can be sorted in ascending order by rotational speed. Then, the aggregated data with the most consecutive rows of standard deviations greater than 0 can be determined. The rotational speed span of these aggregated data is the maximum length of consecutive standard deviations greater than 0.

[0067] As an example, step S30 may include: determining whether the data characteristic indicators of each speed range meet the corresponding preset conditions; and in response to the data characteristic indicators of at least two speed ranges meeting the corresponding preset conditions, determining that the generator bearing of the wind turbine is abnormally worn.

[0068] As an example, the preset conditions corresponding to the first speed range may include, but are not limited to, the following sub-conditions: the amount of data N5 whose standard deviation Y exceeds the first standard deviation threshold p7 is greater than or equal to N1. N1 is a positive integer.

[0069] As an example, the preset conditions corresponding to the second speed range may include, but are not limited to, at least one of the following sub-conditions: the coefficient of variation (Coeff) of the standard deviation Y is greater than or equal to the coefficient of variation threshold p9; the amount of data (N4) in the standard deviation Y exceeding the second standard deviation threshold p8 is greater than N2; the absolute value of the kurtosis of the standard deviation Y, math.abs(yPeak), is greater than the kurtosis threshold p12; the maximum value of the standard deviation Y, yMax, is greater than or equal to the maximum value threshold p11; the difference between the maximum and minimum values ​​of the standard deviation Y, yRange, is greater than or equal to the difference threshold p10; the speed (rpmMax) of the aggregated data with the largest standard deviation Y is greater than the first speed threshold p13; the speed (rpmMax) of the aggregated data with the largest standard deviation Y is less than the second speed threshold p14; the maximum difference value of the standard deviation Y, yDiffMax, is less than the maximum difference value threshold p15; and the ratio of the maximum difference value to the coefficient of variation, yDiffMax / Coeff, is less than or equal to the first ratio threshold p16. N2 is a positive integer.

[0070] As an example, the preset conditions corresponding to the third speed range may include, but are not limited to, at least one of the following sub-conditions: the correlation coefficient between speed and standard deviation Y (corcoefY) is greater than the first coefficient threshold p20; the correlation coefficient between speed and standard deviation X (corcoefX) is less than or equal to the second coefficient threshold p19; the correlation coefficient between speed and standard deviation deviation (corCoef) is greater than the third coefficient threshold p18; the ratio between the amount of data with standard deviation deviation greater than 0 and the amount of data with standard deviation deviation less than or equal to 0 (posDivNeg) is greater than the second ratio threshold p21; the ratio between the sum of standard deviation deviations greater than the first deviation threshold and the sum of standard deviation deviations less than the second deviation threshold (integral) is greater than the third ratio threshold p22; the proportion of data with standard deviation deviation greater than 0 (rpmIncrease) is greater than or equal to the proportion threshold p23; and the maximum length of consecutive standard deviation deviations greater than 0 (yMoreXMaxLen) is greater than or equal to the length threshold p17, where the standard deviation deviation is the difference between the standard deviation Y and the standard deviation X of the same aggregated data.

[0071] It should be understood that abnormal generator bearing wear can refer to the degree of wear of the generator bearing reaching a specific level. It should also be understood that the threshold values ​​involved in the sub-conditions of the above exemplary embodiments can be set to appropriate values ​​according to actual conditions and specific needs, thereby limiting the specific level. For example, when the specific level is limited to a slight degree by setting the threshold, a wind turbine generator bearing wear early warning can be achieved, informing the on-site wind turbine generator bearings that slight wear has already occurred, providing sufficient time for on-site maintenance personnel to carry out planned maintenance.

[0072] As an example, the steps for determining whether the data characteristic indicators of each speed range meet the corresponding preset conditions include: for each speed range, when a preset number of sub-conditions for that speed range are met, it is determined that the data characteristic indicators of that speed range meet the corresponding preset conditions. That is, when a preset number of data characteristic indicators for that speed range meet their respective sub-conditions, it is determined that the data characteristic indicators of that speed range meet the corresponding preset conditions. It should be understood that the preset numbers for different speed ranges may be the same or different.

[0073] Furthermore, as an example, the bearing wear detection method for a wind turbine according to an exemplary embodiment of this disclosure may further include: displaying detection results, wherein the detection results are used to indicate whether the wear of the generator bearings of the wind turbine is abnormal, and / or, the operating characteristics of the wind turbine when abnormal wear of the generator bearings occurs. For example, Figure 2 A schematic diagram showing the characteristics of generator bearing wear is provided. The gray box shows the unit's operating characteristics when the generator bearing wear is abnormal.

[0074] This disclosure utilizes the nacelle acceleration characteristics in different speed ranges to analyze the nacelle acceleration variation characteristics in the x and y directions under different speed ranges of the unit. Combining the characteristics of multiple speed ranges, it accurately locates the bearing wear characteristics in the entire speed range during wind turbine operation, promptly detects generator bearing wear problems, and provides support for on-site operation and maintenance personnel.

[0075] By maintaining the bearings in a timely manner and adjusting the rotation accuracy promptly, the service life and power generation performance of the generator can be improved.

[0076] In addition, accurate detection of generator bearing wear can reduce frequent shutdowns caused by unstable operation of the wind turbine.

[0077] This disclosure describes a method for detecting bearing wear in wind turbine generators by collecting real-time operational data. This method is applicable to all megawatt-class wind turbine generators. Key parameters involved can be adjusted based on the specific project and generator unit.

[0078] This disclosure provides a low-cost, high-efficiency method for detecting wear on wind turbine generator bearings. Based on data such as generator speed, wind speed, nacelle x-axis acceleration, and nacelle y-axis acceleration during wind turbine operation, the method achieves effective detection of abnormalities in wind turbine generator bearings through data acquisition, data cleaning, data processing, feature extraction, and anomaly detection. On the one hand, it can detect units that are still operating despite abnormal bearing wear, effectively reducing the operation and maintenance costs of wind turbines, enabling planned maintenance of generator bearings, preventing further damage, and avoiding safety accidents caused by accelerated bearing wear. On the other hand, it can also provide early warning of wind turbine generator bearing wear, notifying on-site personnel in advance of abnormal bearing wear and providing sufficient time for planned maintenance.

[0079] Figure 3 A structural block diagram of a bearing wear detection device for a wind turbine generator according to an exemplary embodiment of the present disclosure is shown.

[0080] like Figure 3 As shown, the bearing wear detection device for a wind turbine according to an exemplary embodiment of the present disclosure includes: a data acquisition unit 10, a feature determination unit 20, and a wear detection unit 30.

[0081] Specifically, the data acquisition unit 10 is configured to acquire the operating data of the wind turbine, wherein the operating data includes the rotational speed and nacelle acceleration of the wind turbine.

[0082] The feature determination unit 20 is configured to determine the data feature index corresponding to each speed range, wherein the data feature index of each speed range is used to characterize the features of the operating data whose speed belongs to that speed range.

[0083] The wear detection unit 30 is configured to detect whether the wear of the generator bearing of the wind turbine is abnormal based on the data characteristic indicators of each speed range.

[0084] As an example, the feature determination unit 20 can be configured to: aggregate the operating data according to the rotational speed to obtain aggregated data, wherein the aggregated data includes the rotational speed and the statistical value of the nacelle acceleration corresponding to the rotational speed; and for each rotational speed range, determine the data feature index of the rotational speed range based on the aggregated data of the rotational speed range belonging to the rotational speed range.

[0085] As an example, the feature determination unit 20 may be configured to: adjust the accuracy of the rotation speed in the running data; and aggregate the running data with the same rotation speed in the adjusted running data to obtain aggregated data.

[0086] As an example, the statistical values ​​of the cabin acceleration may include: the standard deviation Y of the cabin acceleration in the y-direction and the standard deviation X of the cabin acceleration in the x-direction.

[0087] As an example, each speed range may include at least two of the following speed ranges: a first speed range, a second speed range, and a third speed range; wherein, the first speed range is the stable speed range of the wind turbine that is least affected by the turbine system; the second speed range is the speed range from the speed of the wind turbine during stable grid-connected operation to the speed at which the power reaches the rated power; and the third speed range is the speed range from the grid-connected speed of the wind turbine to the speed corresponding to the cut-out wind speed.

[0088] As an example, the data characteristic indicators corresponding to the first speed range may include: the amount of data whose standard deviation Y exceeds the first standard deviation threshold.

[0089] As an example, the data characteristic indicators corresponding to the second speed range may include at least one of the following: the coefficient of variation of the standard deviation Y, the amount of data in the standard deviation Y that exceeds the second standard deviation threshold, the absolute value of the kurtosis of the standard deviation Y, the maximum value of the standard deviation Y, the difference between the maximum and minimum values ​​of the standard deviation Y, the speed of the aggregated data with the largest standard deviation Y, the maximum difference of the standard deviation Y, and the ratio of the maximum difference to the coefficient of variation.

[0090] As an example, the data feature indicators corresponding to the third speed range may include at least one of the following: the correlation coefficient between speed and standard deviation Y, the correlation coefficient between speed and standard deviation X, the correlation coefficient between speed and standard deviation deviation, the ratio between the amount of data with standard deviation deviation greater than 0 and the amount of data with standard deviation deviation less than or equal to 0, the ratio between the sum of standard deviation deviations greater than the first deviation threshold and the sum of standard deviation deviations less than the second deviation threshold, the proportion of data with standard deviation deviation greater than 0, and the maximum length of consecutive standard deviation deviations greater than 0, wherein the standard deviation deviation is the difference between the standard deviation Y and the standard deviation X of the same aggregated data.

[0091] As an example, the wear detection unit 30 can be configured to: determine whether the data characteristic indicators of each speed range meet the corresponding preset conditions; and in response to the data characteristic indicators of at least two speed ranges meeting the corresponding preset conditions, determine that the generator bearing of the wind turbine is abnormally worn.

[0092] As an example, the preset conditions corresponding to the first speed range may include the following sub-conditions: the amount of data with a standard deviation Y exceeding the first standard deviation threshold is greater than or equal to N1. N1 is a positive integer.

[0093] As an example, the preset conditions corresponding to the second speed range may include at least one of the following sub-conditions: the coefficient of variation of the standard deviation Y is greater than or equal to the coefficient of variation threshold; the amount of data with standard deviation Y exceeding the second standard deviation threshold is greater than N2; the absolute value of the kurtosis of the standard deviation Y is greater than the kurtosis threshold; the maximum value of the standard deviation Y is greater than or equal to the maximum value threshold; the difference between the maximum and minimum values ​​of the standard deviation Y is greater than or equal to the difference threshold; the speed of the aggregated data with the largest standard deviation Y is greater than the first speed threshold; the speed of the aggregated data with the largest standard deviation Y is less than the second speed threshold; the maximum difference value of the standard deviation Y is less than the maximum difference value threshold; and the ratio of the maximum difference value to the coefficient of variation is less than or equal to the first ratio threshold. N2 is a positive integer.

[0094] As an example, the preset conditions corresponding to the third speed range may include at least one of the following sub-conditions: the correlation coefficient between speed and standard deviation Y is greater than the first coefficient threshold; the correlation coefficient between speed and standard deviation X is less than or equal to the second coefficient threshold; the correlation coefficient between speed and standard deviation deviation is greater than the third coefficient threshold; the ratio between the amount of data with standard deviation deviation greater than 0 and the amount of data with standard deviation deviation less than or equal to 0 is greater than the second ratio threshold; the ratio between the sum of standard deviation deviations greater than the first deviation threshold and the sum of standard deviation deviations less than the second deviation threshold is greater than the third ratio threshold; the proportion of data with standard deviation deviation greater than 0 is greater than or equal to the proportion threshold; and the maximum length of consecutive standard deviation deviations greater than 0 is greater than or equal to the length threshold, wherein the standard deviation deviation is the difference between the standard deviation Y and the standard deviation X of the same aggregated data.

[0095] As an example, the wear detection unit 30 can be configured to determine that the data characteristic index of the speed range meets the corresponding preset condition when a preset number of sub-conditions of the speed range are met for each speed range.

[0096] As an example, the bearing wear detection device for a wind turbine according to an exemplary embodiment of the present disclosure may further include: a result display unit (not shown), configured to display detection results, wherein the detection results are used to indicate whether the wear of the generator bearing of the wind turbine is abnormal, and / or the operating characteristics of the wind turbine when the wear of the generator bearing is abnormal.

[0097] As an example, the bearing wear detection device for a wind turbine according to an exemplary embodiment of the present disclosure may be installed in the controller of the wind turbine or the controller of the wind farm.

[0098] Figure 4 A structural block diagram of a bearing wear detection device for a wind turbine generator according to another exemplary embodiment of the present disclosure is shown.

[0099] like Figure 4 As shown, in addition to the data acquisition unit 10, feature determination unit 20, and wear detection unit 30, the bearing wear detection device for wind turbines according to another exemplary embodiment of the present disclosure also includes a result display unit 40.

[0100] The data acquisition unit 10 may include: a data reading unit 101, a parameter reading unit 102, and a data cleaning unit 103.

[0101] The feature determination unit 20 may include a data preprocessing unit 201 and a feature extraction unit 202.

[0102] Specifically, the data reading unit 101 is configured to read multiple real-time data points from the wind turbine over a period of time from a data source (e.g., a SCADA data source). As an example, the data fields read may include: generator speed, wind speed, nacelle x-axis acceleration, and nacelle y-axis acceleration.

[0103] The parameter reading unit 102 is configured to read the parameters used throughout the entire wear detection process. For example, it can acquire the parameters set by the user via the display screen. The parameters used throughout the wear detection process may include specific parameters used in data cleaning, data preprocessing, feature extraction, and wear detection processes.

[0104] The data cleaning unit 103 is configured to clean the read real-time data to obtain operational data. For example, it can be specifically configured to: remove abnormal data (e.g., data including null values) caused by problems such as acquisition and transmission from the read real-time data; then, when the amount of jump data in the real-time data after removing abnormal data meets certain conditions, select real-time data from the real-time data after removing abnormal data where both the x-direction cabin acceleration and the y-direction cabin acceleration are within a reasonable and reliable range (e.g., the default range can be -0.45 to 0.45, which can be adjusted according to actual conditions and specific needs).

[0105] The data preprocessing unit 201 is configured to adjust the accuracy of the rotation speed in the running data; and to aggregate the running data with the same rotation speed in the adjusted running data to obtain aggregated data.

[0106] The feature extraction unit 202 is configured to determine the data feature indicators of each speed range based on the aggregated data of speeds belonging to that speed range.

[0107] The wear detection unit 30 is configured to detect whether the wear of the generator bearing of the wind turbine is abnormal based on the data characteristic indicators of each speed range.

[0108] The result display unit 40 is configured to display the detection results, wherein the detection results are used to indicate whether the wear of the generator bearing of the wind turbine is abnormal, and / or the operating characteristics of the wind turbine when the wear of the generator bearing is abnormal.

[0109] As an example, the data acquisition unit 10, the feature determination unit 20, and the wear detection unit 30 can be located in the processor, and the result display unit 40 can be located in the display.

[0110] It should be understood that the specific processing performed by the bearing wear detection device for wind turbines according to the exemplary embodiments of this disclosure has been referenced. Figures 1 to 2 A detailed description has been provided, and the relevant details will not be repeated here.

[0111] It should be understood that the various units in the bearing wear detection device for wind turbines according to exemplary embodiments of this disclosure can be implemented as hardware components and / or software components. Those skilled in the art can implement the various units, for example, using field-programmable gate arrays (FPGAs) or application-specific integrated circuits (ASICs), based on the processes performed by each defined unit.

[0112] An exemplary embodiment of this disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform a bearing wear detection method for a wind turbine as described in the exemplary embodiment above. This computer-readable storage medium is any data storage device capable of storing data read from a computer system. Examples of computer-readable storage media include: read-only memory, random access memory, read-only optical disc, magnetic tape, floppy disk, optical data storage device, and carrier waves (such as data transmission via the Internet through wired or wireless transmission paths).

[0113] A bearing wear detection device for a wind turbine according to an exemplary embodiment of the present disclosure includes a processor (not shown) and a memory (not shown), wherein the memory stores a computer program that, when executed by the processor, causes the processor to perform a bearing wear detection method for a wind turbine as described in the exemplary embodiment above. As an example, the bearing wear detection device for a wind turbine according to an exemplary embodiment of the present disclosure may be installed in the controller of a wind turbine or the controller of a wind farm.

[0114] While some exemplary embodiments of this disclosure have been shown and described, those skilled in the art will understand that modifications may be made to these embodiments without departing from the principles and spirit of this disclosure, which are defined by the claims and their equivalents.

Claims

1. A method of bearing wear detection for a wind turbine generator, the method comprising: The method comprises: obtaining operation data of a wind turbine, wherein the operation data comprises a rotating speed of the wind turbine and a nacelle acceleration; determining data characteristic indexes corresponding to each rotating speed interval, wherein the data characteristic indexes of each rotating speed interval are used to represent characteristics of the operation data of which the rotating speed belongs to the rotating speed interval; detecting whether the generator bearing wear of the wind turbine is abnormal according to the data characteristic indexes of each rotating speed interval, wherein the rotating speed intervals comprise a first rotating speed interval, a second rotating speed interval and a third rotating speed interval; wherein the first rotating speed interval is a stable rotating speed interval in which the wind turbine is least affected by a system of the wind turbine; the second rotating speed interval is a rotating speed interval from a rotating speed when the wind turbine is stably operated in parallel to a rotating speed when power reaches a rated power; the third rotating speed interval is a rotating speed interval from a parallel rotating speed of the wind turbine to a rotating speed corresponding to a cut-out wind speed, wherein the data characteristic indexes corresponding to the first rotating speed interval comprise a data amount in which a standard deviation Y of the nacelle y-direction acceleration exceeds a first standard deviation threshold value; the data characteristic indexes corresponding to the second rotating speed interval comprise a coefficient of variation of the standard deviation Y, a data amount in which the standard deviation Y exceeds a second standard deviation threshold value, an absolute value of kurtosis of the standard deviation Y, a maximum value of the standard deviation Y, a difference between the maximum value and a minimum value of the standard deviation Y, a rotating speed of aggregated data in which the standard deviation Y is maximum, a differential maximum value of the standard deviation Y, and a ratio of the differential maximum value to the coefficient of variation; the data characteristic indexes corresponding to the third rotating speed interval comprise a correlation coefficient of the rotating speed and the standard deviation Y, a correlation coefficient of the rotating speed and a standard deviation X of the nacelle x-direction acceleration, a correlation coefficient of the rotating speed and a standard deviation deviation, a ratio between a data amount in which the standard deviation deviation is greater than 0 and a data amount in which the standard deviation deviation is less than or equal to 0, a ratio between a sum of the standard deviation deviations greater than a first deviation threshold value and a sum of the standard deviation deviations less than a second deviation threshold value, a proportion of the data amount in which the standard deviation deviation is greater than 0, and a maximum length in which the standard deviation deviation is continuously greater than 0, wherein the standard deviation deviation is a difference between the standard deviation Y and the standard deviation X of the same aggregated data, and the aggregated data is obtained by aggregating the operation data according to the rotating speed.

2. The bearing wear detection method of claim 1, wherein, The step of determining the data characteristic indexes corresponding to each rotating speed interval comprises: aggregating the operation data according to the rotating speed to obtain aggregated data, wherein the aggregated data comprises a rotating speed and a statistical value of a nacelle acceleration corresponding to the rotating speed; for each rotating speed interval, determining the data characteristic indexes of the rotating speed interval based on the aggregated data of which the rotating speed belongs to the rotating speed interval.

3. The bearing wear detection method of claim 2, wherein, The step of aggregating the operation data according to the rotating speed to obtain aggregated data comprises: adjusting the precision of the rotating speed in the operation data; aggregating the operation data with the same rotating speed in the adjusted operation data to obtain the aggregated data.

4. The bearing wear detection method of claim 2, wherein, The statistical value of the nacelle acceleration comprises a standard deviation Y of the nacelle y-direction acceleration and a standard deviation X of the nacelle x-direction acceleration.

5. The bearing wear detection method of claim 2, wherein, The step of detecting whether the generator bearing wear of the wind turbine is abnormal according to the data characteristic indexes of each rotating speed interval comprises: determining whether the data characteristic indexes of each rotation speed interval meet the corresponding preset conditions; in response to the data characteristic indexes of at least two rotation speed intervals meeting the corresponding preset conditions, determining that the generator bearing of the wind turbine is abnormally worn.

6. The bearing wear detection method according to claim 5, characterized in that, the preset condition corresponding to the first rotation speed interval comprises the following sub-conditions: the data quantity of the standard deviation Y exceeding the first standard deviation threshold value is greater than or equal to N1; and / or, the preset condition corresponding to the second rotation speed interval comprises at least one of the following sub-conditions: the coefficient of variation of the standard deviation Y is greater than or equal to the coefficient of variation threshold value, the data quantity of the standard deviation Y exceeding the second standard deviation threshold value is greater than N2, the absolute value of the kurtosis of the standard deviation Y is greater than the kurtosis threshold value, the maximum value of the standard deviation Y is greater than or equal to the maximum value threshold value, the difference between the maximum value and the minimum value of the standard deviation Y is greater than or equal to the difference threshold value, the rotation speed of the aggregated data with the maximum standard deviation Y is greater than the first rotation speed threshold value, the rotation speed of the aggregated data with the maximum standard deviation Y is less than the second rotation speed threshold value, the differential maximum value of the standard deviation Y is less than the differential maximum value threshold value, and the ratio of the differential maximum value to the coefficient of variation is less than or equal to the first ratio threshold value; and / or, the preset condition corresponding to the third rotation speed interval comprises at least one of the following sub-conditions: the correlation coefficient of the rotation speed and the standard deviation Y is greater than the first coefficient threshold value, the correlation coefficient of the rotation speed and the standard deviation X is less than or equal to the second coefficient threshold value, the correlation coefficient of the rotation speed and the standard deviation deviation is greater than the third coefficient threshold value, the ratio between the data quantity of the standard deviation deviation greater than 0 and the data quantity of the standard deviation deviation less than or equal to 0 is greater than the second ratio threshold value, the ratio between the sum of the standard deviation deviations greater than the first deviation threshold value and the sum of the standard deviation deviations less than the second deviation threshold value is greater than the third ratio threshold value, the proportion of the data quantity of the standard deviation deviation greater than 0 is greater than or equal to the proportion threshold value, and the maximum length of the standard deviation deviation continuously greater than 0 is greater than or equal to the length threshold value, wherein the standard deviation deviation is the difference between the standard deviation Y and the standard deviation X of the same aggregated data, wherein N1 and N2 are positive integers, wherein the statistical values of the nacelle acceleration include the standard deviation Y of the nacelle y-direction acceleration and the standard deviation X of the nacelle x-direction acceleration.

7. The bearing wear detection method according to claim 5 or 6, characterized in that, The step of determining whether the data characteristic indexes of each rotation speed interval meet the corresponding preset conditions comprises: for each rotation speed interval, when a preset number of sub-conditions of the rotation speed interval are met, determining that the data characteristic index of the rotation speed interval meets the corresponding preset condition.

8. The bearing wear detection method of claim 1, wherein, Further comprising: displaying the detection result, wherein the detection result is used to indicate whether the generator bearing wear of the wind turbine is abnormal, and / or the operating characteristics of the wind turbine when the generator bearing wear is abnormal.

9. A bearing wear detection apparatus for a wind turbine generator, comprising: Comprising: a data acquisition unit configured to acquire operating data of a wind turbine, wherein the operating data comprises the rotation speed and the nacelle acceleration of the wind turbine; a characteristic determination unit configured to determine the data characteristic index corresponding to each rotation speed interval, wherein the data characteristic index of each rotation speed interval is used to characterize the features of the operating data with the rotation speed belonging to the rotation speed interval; The wear detection unit is configured to detect whether the generator bearing wear of the wind turbine is abnormal according to the data feature indicators of the respective speed intervals, The respective speed intervals include a first speed interval, a second speed interval, and a third speed interval. The first speed interval is a stable speed interval in which the wind turbine is least affected by the turbine system. The second speed interval is a speed interval from a speed at which the wind turbine is stably connected to a power grid to a speed at which the power reaches a rated power. The third speed interval is a speed interval from a speed at which the wind turbine is connected to the power grid to a speed corresponding to a cut-out wind speed. The data feature indicators corresponding to the first speed interval include a data amount in which a standard deviation Y of a cabin y-direction acceleration exceeds a first standard deviation threshold value. The data feature indicators corresponding to the second speed interval include a coefficient of variation of the standard deviation Y, a data amount in which the standard deviation Y exceeds a second standard deviation threshold value, an absolute value of kurtosis of the standard deviation Y, a maximum value of the standard deviation Y, a difference between the maximum value and a minimum value of the standard deviation Y, a speed of aggregated data in which the standard deviation Y is maximum, a differential maximum value of the standard deviation Y, and a ratio of the differential maximum value to the coefficient of variation. The data feature indicators corresponding to the third speed interval include a correlation coefficient of the speed and the standard deviation Y, a correlation coefficient of the speed and a standard deviation X of a cabin x-direction acceleration, a correlation coefficient of the speed and a standard deviation deviation, a ratio between a data amount in which the standard deviation deviation is greater than 0 and a data amount in which the standard deviation deviation is less than or equal to 0, a ratio between a sum of the standard deviation deviations greater than a first deviation threshold value and a sum of the standard deviation deviations less than a second deviation threshold value, a proportion of the data amount in which the standard deviation deviation is greater than 0, and a maximum length in which the standard deviation deviation is continuously greater than 0. The standard deviation deviation is a difference between the standard deviation Y and the standard deviation X of the same aggregated data, wherein the aggregated data is obtained by aggregating the operation data according to the speed.

10. A computer readable storage medium storing a computer program, characterized in that, The computer program, when executed by the processor, causes the processor to perform the bearing wear detection method of the wind turbine according to any one of claims 1 to 8.

11. A bearing wear detection apparatus for a wind turbine generator unit, characterized by, The bearing wear detection device includes: a processor; a memory storing a computer program, wherein the computer program, when executed by the processor, causes the processor to perform the bearing wear detection method of the wind turbine according to any one of claims 1 to 8.

Images