Wind turbine yaw anomaly detection method, device and electronic equipment

Abstract

The present application provides a method, device and electronic device for detecting abnormal yaw of a wind turbine. The method for detecting abnormal yaw of a wind turbine includes: obtaining wind direction data, yaw frequency data and cabin position data of the wind turbine; determining the windward deflection angle of the wind turbine according to the cabin position data and the wind direction data; determining whether the wind turbine has abnormal yaw according to the windward deflection angle and the yaw frequency data. This can improve the accuracy of detecting abnormal yaw of the wind turbine.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormality detection of wind turbines, and in particular to a method, a device and an electronic device for detecting abnormal yaw of a wind turbine. Background Art

[0002] With the development of wind power generation technology, wind turbines, like hydraulic machinery, have become a power source to replace human and animal power, playing an important role in the development of productivity. Wind turbines can convert wind energy into mechanical energy. Wind turbines include a yaw system. The function of the yaw system is to quickly and smoothly align the wind direction when the direction of the wind speed vector changes, so that the wind rotor can obtain the maximum wind energy.

[0003] In actual operation, the yaw system may be abnormal, which will affect the power generation stability of the wind turbine, and may also cause serious hardware failures or even safety accidents. Therefore, relevant technologies will conduct statistical analysis on the number of yaws of wind turbines in advance to identify wind turbines with frequent yaws.

[0004] However, since the operating system of a wind turbine is relatively complex, it is difficult to reflect the yaw condition of the wind turbine by only using the number of wind turbine yaws as the wind turbine yaw data, resulting in low accuracy of the wind turbine yaw detection and warning. Summary of the invention

[0005] The present application provides a method, device and electronic equipment for detecting abnormal yaw of a wind turbine. The yaw detection of the wind turbine by the method has high accuracy.

[0006] The present application provides a method for detecting abnormal yaw of a wind turbine, comprising:

[0007] Obtain wind direction data, yaw frequency data and cabin position data of wind turbines;

[0008] Determining the wind deflection angle of the wind turbine according to the nacelle position data and the wind direction data;

[0009] Whether the wind turbine has yaw abnormality is determined according to the wind deflection angle and the yaw number data.

[0010] Further, determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data includes:

[0011] Determine the average value of the wind deflection angle of the wind turbine at multiple times within a predetermined time period;

[0012] Determining an angle threshold according to an average of the wind deflection angles of the plurality of wind turbines;

[0013] Determining the yaw frequency data of the wind turbine within the predetermined time period;

[0014] Determining a yaw number threshold value according to the yaw number data of the plurality of wind turbines;

[0015] When at least one of the average of the wind deflection angles of the wind turbine exceeds the angle threshold and the yaw frequency data of the wind turbine exceeds the yaw frequency threshold, it is determined that the wind turbine has yaw abnormality.

[0016] Furthermore, the determining of the yaw number threshold according to the yaw number data of the plurality of wind turbines comprises:

[0017] Determine an average value of the yaw number data of a plurality of the wind turbines;

[0018] Determining a standard deviation of the yaw number data of a plurality of the wind turbines;

[0019] The yaw number threshold is determined according to an average value of the yaw number data and a standard deviation of the yaw number data.

[0020] Furthermore, determining the angle threshold according to the average of the wind deflection angles of the plurality of wind turbines comprises:

[0021] Determining an average of the mean values ​​of the wind deflection angles of a plurality of the wind turbines;

[0022] Determining a standard deviation of the mean values ​​of the wind deflection angles of the plurality of wind turbines;

[0023] The angle threshold is determined according to an average value of the mean values ​​of the windward deflection angles and a standard deviation of the mean values ​​of the windward deflection angles.

[0024] Further, the predetermined time period includes a plurality of sub-time periods, and the plurality of sub-time periods have the same length;

[0025] The determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data includes:

[0026] Determining whether the plurality of wind turbines in the sub-time period have yaw abnormalities according to the wind deflection angle and the yaw frequency data in the sub-time period;

[0027] After determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data, the method further includes:

[0028] Marking the wind turbine generator having yaw abnormality in the sub-time period as a yaw abnormal wind turbine;

[0029] Counting the number of times the same abnormal yaw wind turbine is marked within the predetermined time period;

[0030] The fault level of the wind turbine with the same yaw abnormality is determined according to the number of markings and the number of sub-time periods.

[0031] Further, determining the fault level of the same wind turbine with abnormal yaw according to the number of markings and the number of sub-time periods includes:

[0032] Determine the wind turbine with abnormal yaw, whose number of marks is greater than half of the number of sub-time periods, as a wind turbine with high-risk abnormal yaw;

[0033] Determine the wind turbine with yaw abnormality whose marking times are equal to half of the number of segments in the sub-time period as a wind turbine with medium-risk yaw abnormality;

[0034] The wind turbine with abnormal yaw, whose marking times are greater than zero and less than half of the number of segments in the sub-time period, is determined as a wind turbine with low-risk yaw abnormality.

[0035] Furthermore, the method comprises:

[0036] Acquire operating data of a plurality of wind turbines in a wind farm, wherein the operating data includes the wind direction data and the wind speed data;

[0037] Classifying a plurality of wind turbines in the wind farm according to the wind direction data and the wind speed data to obtain a plurality of wind turbine clusters;

[0038] Determining the wind deflection angle of the wind turbine according to the nacelle position data and the wind direction data includes:

[0039] Determining the wind deflection angles of the multiple wind turbines in the wind turbine cluster according to the nacelle position data and the wind direction data of the multiple wind turbines in the wind turbine cluster;

[0040] The determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data includes:

[0041] According to the wind deflection angles and the yaw frequency data of the plurality of wind turbines in the wind turbine cluster, it is determined whether the wind turbines in the wind turbine cluster have abnormal yaw.

[0042] Furthermore, the multiple wind turbines in the wind farm are classified according to the wind direction data and the wind speed data to obtain multiple wind turbine clusters, including:

[0043] Determining the correlation between the operating condition data of every two wind turbines among the plurality of wind turbines in the wind farm;

[0044] Every two wind turbines whose correlation is greater than a correlation threshold are determined to belong to the same wind turbine cluster.

[0045] Furthermore, determining the correlation between the operating condition data of every two wind turbines among the plurality of wind turbines in the wind farm includes:

[0046] Determining a correlation between the wind speed data and a correlation between the wind direction data;

[0047] An average value of the correlation of the wind speed data and the correlation of the wind direction data is determined as the correlation of the operating condition data.

[0048] Furthermore, after determining the correlation between the operating condition data of every two wind turbines among the plurality of wind turbines in the wind farm, the method further includes:

[0049] A wind turbine whose correlation with other wind turbines in the wind farm except the wind turbine cluster is not greater than the correlation threshold is determined as a wind turbine with abnormal operating condition.

[0050] The present application provides a wind turbine yaw abnormality detection and early warning device, comprising:

[0051] An acquisition module is used to acquire wind direction data, yaw frequency data and cabin position data of the wind turbine;

[0052] A first processing module, used for determining the wind deflection angle of the wind turbine according to the nacelle position data and the wind direction data;

[0053] The second processing module is used to determine whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data.

[0054] The present application provides an electronic device, including a processor and a memory;

[0055] Memory, used to store computer programs;

[0056] The processor is used to implement any of the methods described above when executing a program stored in the memory.

[0057] The present application provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the method described in any one of the above items is implemented.

[0058] In some embodiments, a method for detecting yaw anomaly of a wind turbine of the present application determines whether the wind turbine has yaw anomaly through yaw count data and wind yaw angle that can reflect the yaw condition of the wind turbine. Compared with the related art which only considers the yaw count of the wind turbine, the present application considers more factors related to the yaw of the wind turbine, thereby improving the accuracy of yaw anomaly detection of the wind turbine. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 The figure is a schematic diagram of the flow of a method for detecting abnormal yaw of a wind turbine provided in an embodiment of the present application;

[0060] Figure 2 Shown Figure 1 The flowchart of step 130 in the method for detecting abnormal yaw of a wind turbine is shown;

[0061] Figure 3 Shown Figure 1 A detailed flow chart of a method for detecting abnormal yaw of a wind turbine is shown;

[0062] Figure 4 Shown Figure 1 A schematic diagram of a process flow when there are multiple wind turbines in the wind turbine yaw anomaly detection method shown;

[0063] Figure 5 Shown Figure 1 A heat map of correlation between multiple wind turbines in the wind turbine yaw anomaly detection method shown;

[0064] Figure 6 The figure is a flow chart of an application example of the method for detecting abnormal yaw of a wind turbine provided by an embodiment of the present application;

[0065] Figure 7 The figure shows a module schematic diagram of a device for detecting abnormal yaw of a wind turbine provided in an embodiment of the present application;

[0066] Figure 8 Shown is a schematic structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0067] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of the present specification. Instead, they are merely examples of devices and methods consistent with some aspects of one or more embodiments of the present specification as detailed in the appended claims.

[0068] It should be noted that: in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0069] In order to solve this technical problem, an embodiment of the present application provides a method for detecting yaw abnormality of a wind turbine. By using the yaw count data and the wind yaw angle that can reflect the yaw situation of the wind turbine, it is determined whether the wind turbine has yaw abnormality. Compared with the related art which only considers the yaw count of the wind turbine, the present application considers more factors related to the yaw of the wind turbine, which can improve the accuracy of yaw abnormality detection of the wind turbine.

[0070] Figure 1 The figure is a flow chart of a method for detecting abnormal yaw of a wind turbine provided in an embodiment of the present application. The method may include the following steps 110 to 130:

[0071] Step 110, obtaining wind direction data, yaw frequency data and nacelle position data of the wind turbine.

[0072] The nacelle position data, wind direction data and yaw frequency data can reflect the yaw of the wind turbine.

[0073] The wind turbine generator may be a single wind turbine generator. In this way, it is possible to detect whether the single wind turbine generator has abnormal yaw. The wind turbine generator may also be multiple wind turbine generators in a wind farm. In this way, it is possible to detect whether multiple wind turbine generators have abnormal yaw. For detailed description, please refer to the following.

[0074] Step 120, determining the wind deflection angle of the wind turbine according to the nacelle position data and the wind direction data.

[0075] The cabin position data may include the angle data of the cabin. The windward yaw angle is the angle of the cabin relative to the wind direction, that is, the angle between the cabin and the wind direction. The angle of the cabin at the initial position is 0 degrees, that is, the reference position. After the cabin yaws, it deviates from this initial position, and the cabin position data is the angle data of the current position of the cabin relative to the initial position, which is the yaw angle.

[0076] Step 130, determining whether the wind turbine has yaw abnormality according to the wind deflection angle and yaw frequency data.

[0077] In the embodiment of the present application, the yaw frequency data and the windward deflection angle of the wind turbine can both reflect whether the wind turbine has abnormal yaw. Compared with the related art that only considers the yaw frequency of the wind turbine, the present application considers more factors related to the yaw of the wind turbine, which can improve the accuracy of the wind turbine yaw abnormality detection. In addition, the method of detecting and warning the frequent yaw frequency of the wind turbine in the related art is to calculate the frequent yaw probability of the wind turbine to achieve the frequent yaw frequency detection and warning of the wind turbine. This frequent yaw probability reflects the possibility of whether the wind turbine has abnormal yaw. Compared with the frequent yaw probability, the yaw frequency data and the windward deflection angle of the wind turbine are more conducive to reflecting whether the actual wind turbine has abnormal yaw, thereby improving the accuracy of the wind turbine yaw abnormality detection.

[0078] Figure 2 Shown Figure 1 FIG. 1 is a flow chart of step 130 in the method for detecting abnormal yaw of a wind turbine. Figure 2 As shown, in some embodiments of step 130 , step 130 further includes steps 131 to 135 .

[0079] Step 131, determining the average of the wind deflection angles of the wind turbine at multiple moments within a predetermined time period. The preset time period may be a time period set by the user. The moment may also be a moment set by the user as required. The preset time period may be, for example, 1 day or more, or even longer. The moment may be, for example, 1 hour or more, or even longer, which is not limited here. For example, the preset time period may be 24 hours, and the moment may be 1 hour.

[0080] The preset time period may include a whole time period. In this way, the user can determine whether the wind turbine has yaw abnormality in the whole time period. The preset time period may include multiple sub-time periods. In this way, the user can determine whether the wind turbine has yaw abnormality in multiple sub-time periods. By calculating the yaw abnormality multiple times, the wind turbine with yaw abnormality can be determined more accurately. In addition, the amount of data calculated for each sub-time period multiple times is small, so that the rate of each calculation is relatively fast. Please see below for detailed introduction.

[0081] Step 132, determining an angle threshold according to the average of the wind deflection angles of the plurality of wind turbines, where the angle threshold is used to determine whether the wind turbine has yaw abnormality.

[0082] The above step 132 can be implemented through various embodiments. In some embodiments, the above step 132 further includes the first step to the third step. The first step, in some embodiments, determines the average value of the mean values ​​of the wind deviation angles of multiple wind turbines. The mean value can reflect the changing trend of the overall data, and individual abnormal data will not affect the changing trend of the overall data. Therefore, the subsequent use of the mean value is more conducive to determining a reasonable threshold. The average value of the wind deviation angle can reflect the changing trend of the wind deviation angle. The second step is to determine the standard deviation of the mean values ​​of the wind deviation angles of multiple wind turbines. The standard deviation can reflect the fluctuation of the wind deviation angle, that is, the standard deviation reflects the amplitude of the change of the data. The smaller the average value of the wind deviation angle and the standard deviation of the wind deviation angle, the more stable the wind deviation angle is, reflecting the more stable the wind turbine is. The third step is to determine the angle threshold according to the average value of the mean value of the wind deviation angle and the standard deviation of the mean value of the wind deviation angle. The angle threshold includes an upper limit angle threshold and a lower limit angle threshold. The upper limit angle threshold It is the sum of the mean value μ1 of the wind deflection angle and twice the standard deviation σ1, that is, Angle lower threshold is the difference between the mean value μ1 of the wind deflection angle and twice the standard deviation σ1, for example In this way, the angle threshold is not a fixed threshold, but changes with the average value of the mean values ​​of the wind deflection angles and the standard deviation of the mean values ​​of the wind deflection angles, thereby improving the accuracy of determining the angle threshold.

[0083] The steps in some other embodiments of the above step 132 are similar to the steps in the above embodiment of the above step 132, except that in the second step of this other embodiment, the variance of the mean of the wind deflection angles of the plurality of wind turbines can be determined, and the variance is squared to obtain the standard deviation, which will not be described in detail here.

[0084] Step 133, determining the yaw frequency data of the wind turbine within a predetermined time period.

[0085] Step 134: determining a yaw frequency threshold value according to the yaw frequency data of the plurality of wind turbines.

[0086] The above step 134 can be implemented through various embodiments. In some embodiments, the above step 134 further includes the first step to the third step. The first step is to determine the average value of the yaw number data of multiple wind turbines. The average value of the yaw number data can reflect the changing trend of the yaw number data. The second step is to determine the standard deviation of the yaw number data of multiple wind turbines. The standard deviation can reflect the fluctuation of the yaw number data. The smaller the average value of the yaw number data and the standard deviation of the yaw number data, the more stable the yaw number data. The third step is to determine the yaw number threshold value based on the average value of the yaw number data and the standard deviation of the yaw number data. The yaw number threshold value includes the yaw number upper limit threshold value And the lower limit threshold of the number of yaw Upper limit of yaw times It is the sum of the mean value μ2 and twice the standard deviation σ2 of the yaw frequency data, that is, Lower limit threshold of yaw times It is the difference between the mean value μ2 and twice the standard deviation σ2 of the yaw frequency data, for example In this way, the average value of the yaw number data and the standard deviation of the yaw number data are used to better reflect the fluctuation of the yaw number data compared with the average value, so as to determine a more reasonable yaw number threshold value.

[0087] The steps in some other embodiments of the above step 134 are similar to the steps in the above embodiment of the above step 134, except that in the second step of the other embodiment, the variance of the yaw times data of the plurality of wind turbines can be determined, and the standard deviation can be obtained by taking the square root of the variance, which will not be described in detail here.

[0088] Step 135, when there is a situation that the mean value of the windward deflection angle of the wind turbine exceeds the angle threshold, and the yaw number data of the wind turbine exceeds at least one of the yaw number threshold, the wind turbine yaw abnormality is determined. The angle threshold and the yaw number threshold determined by using the windward deflection angle and the yaw number data are not fixed thresholds. The angle threshold and the yaw number threshold are thresholds that change with the windward deflection angle and the yaw number data, respectively, which are in line with the operating conditions of the wind turbine and improve the accuracy of yaw abnormality detection.

[0089] Figure 3 Shown Figure 1 The detailed flow chart of the wind turbine yaw anomaly detection method is shown in FIG. Figure 3 As shown, the wind turbine yaw anomaly detection method may include steps 210 to 260.

[0090] Step 210, obtain wind direction data, yaw frequency data and nacelle position data of multiple wind turbines in a predetermined time period, the predetermined time period includes multiple sub-time periods, and the time lengths of the multiple sub-time periods are the same. For example, the preset time period can be 24 hours, the sub-time period can be 6 hours, the number of sub-time periods is 4, and the number of cycles from step 220 to step 260 is 4 times in total, so that the wind turbines with abnormal yaw in one day can be marked up to 4 times, so that the amount of data processed each time is small, and the total amount of marked data is also small, which improves processing efficiency.

[0091] Step 220, determining the wind deflection angles of the plurality of wind turbines within the sub-time period according to the nacelle position data and the wind direction data.

[0092] Step 230, determining whether the yaw of the plurality of wind turbines in the sub-time period is abnormal based on the wind deflection angle and the yaw frequency data in the sub-time period.

[0093] After determining whether the multiple wind turbines in the sub-time period have yaw abnormalities in this step 230, if the wind turbines do not have yaw abnormalities, the numbers of the wind turbines with normal yaws can be recorded, or the numbers of the wind turbines with normal yaws can be omitted. If the wind turbine has yaw abnormalities, the subsequent steps 240 to 260 are executed to facilitate the subsequent warning to the user and preventive protection.

[0094] Step 240: Mark the wind turbines with yaw abnormality in the sub-time period as wind turbines with yaw abnormality.

[0095] In step 240, the number of times the same yaw abnormal wind turbine is marked within a predetermined time period is counted, and the same yaw abnormal wind turbine is marked according to the unique identifier of the yaw abnormal wind turbine. The unique identifier may be the serial number of the yaw abnormal wind turbine. In this way, the same yaw abnormal wind turbine can be accurately marked to avoid omission or error.

[0096] Step 250, counting the number of times the same wind turbine with abnormal yaw is marked within a predetermined time period.

[0097] Step 260, determining the fault level of the same abnormal yaw wind turbine according to the number of markings and the number of segments of the multiple sub-time periods. After this step 260, the method further includes performing abnormal yaw warning for the fault level. This is convenient for reminding users and taking timely protection.

[0098] The above step 260 can be implemented through various embodiments. In some embodiments, the above step 260 includes determining the yaw abnormal wind turbine with a number of marks greater than half of the number of segments of multiple sub-time periods as a high-risk yaw abnormal wind turbine. Determine the yaw abnormal wind turbine with a number of marks equal to half of the number of segments of the sub-time period as a medium-risk yaw abnormal wind turbine. Determine the yaw abnormal wind turbine with a number of marks greater than zero and less than half of the number of segments of the sub-time period as a low-risk yaw abnormal wind turbine. In this way, by marking the yaw abnormal wind turbine, it is more conducive to reflecting the abnormal degree of the yaw abnormal wind turbine, while reducing the errors caused by a single statistical data, and improving the accuracy of determining different fault levels of the yaw abnormal wind turbine. In other embodiments, the above step 260 includes determining the probability of abnormal occurrence of the same yaw abnormal wind turbine by the proportion of the number of marks to the number of segments of multiple sub-time periods. Determine the yaw abnormal wind turbine with a probability of abnormal occurrence greater than 1 / 2 as a high-risk yaw abnormal wind turbine. Determine the yaw abnormal wind turbine with a probability of abnormal occurrence equal to 1 / 2 as a medium-risk yaw abnormal wind turbine. A yaw abnormal wind turbine with an abnormal probability of less than 1 / 2 is determined as a low-risk yaw abnormal wind turbine. For example, the number of sub-time periods is 20, and the number of markings is 12. The abnormal probability of the same yaw abnormal wind turbine is equal to 1 / 2, and the yaw abnormal wind turbine is a medium-risk yaw abnormal wind turbine.

[0099] In this embodiment, when the yaw abnormal wind turbines are counted multiple times in multiple sub-time periods within the same predetermined time period, it is found that one or some yaw abnormal wind turbines keep appearing, indicating that the degree of yaw abnormality of the yaw abnormal wind turbine is relatively serious. In addition, the deviation caused by single statistical data is avoided, making the detection of yaw abnormal wind turbines more accurate.

[0100] The method for detecting and warning the frequent yaw of wind turbines in the related art analyzes the wind turbines in the wind farm. However, the environmental conditions of wind turbines are different, and the working data are quite different, such as the wind turbines at the top of the mountain and the wind turbines at the foot of the mountain. Therefore, the yaw frequency of wind turbines of multiple wind turbines in the wind farm is detected at the same time, resulting in low accuracy of yaw detection and warning of wind turbines.

[0101] Figure 4 Shown Figure 1 The flow chart of the wind turbine yaw anomaly detection method shown in the figure is when there are multiple wind turbines. When the wind turbines are multiple wind turbines in a wind farm, the wind turbine yaw anomaly detection method includes steps 310 to 360, which are described in detail as follows:

[0102] Combination Figure 1 and Figure 4 As shown, in step 310, the operating data of multiple wind turbines in the wind farm are obtained, and the operating data include wind direction data and wind speed data.

[0103] The operating condition data is used to reflect the working condition of the environment in which the wind turbine is located. The working condition of the corresponding wind turbine varies greatly depending on whether the environment in which the wind turbine is located is at the top of a mountain or at the foot of a mountain. For example, the operating condition data includes the operating condition data of the wind turbine at the top of a mountain and the operating condition data of the wind turbine at the foot of a mountain. Subsequently, according to different operating condition data, multiple wind turbine clusters are classified, which makes it easier to distinguish wind turbines in different working conditions, so that it is more accurate to further determine whether the wind turbine has yaw abnormality.

[0104] Step 320: classify multiple wind turbines in the wind farm according to the wind direction data and the wind speed data to obtain multiple wind turbine clusters. By distinguishing wind turbines with different operating condition data in this way, the accuracy of abnormal operating conditions of wind turbines can be improved.

[0105] The above step 320 can adopt a variety of embodiments. In some embodiments, the above step 320 includes classifying multiple wind turbines in the wind farm using a correlation algorithm based on wind direction data and wind speed data to obtain multiple wind turbine clusters. In this way, wind turbines with similar working conditions can be determined by the correlation algorithm, which is more conducive to determining wind turbines with abnormal working conditions.

[0106] Furthermore, the above step 320 further includes step 1, determining the correlation between the operating data of every two wind turbines among the multiple wind turbines in the wind farm. Step 2, determining the two wind turbines whose correlation is greater than the correlation threshold as belonging to the same wind turbine cluster. By using the correlation in this way to process one-dimensional linear data, the amount of data processed is small, and the processing efficiency is improved. Among them, the correlation threshold can be determined according to user needs. The larger the correlation threshold, the closer the operating data of every two wind turbines, the stronger the correlation, and the more accurate the subsequent analysis. The smaller the correlation threshold. The correlation threshold can be greater than 0.6 and less than 0.9. Exemplarily, the correlation threshold is 0.7, so that two wind turbines with a correlation greater than 0.7 can be classified into the same wind turbine cluster.

[0107] After the above step 1, in some embodiments, the method may include a third step, determining the wind turbines whose correlation with other wind turbines in the wind farm except the wind turbine cluster is not greater than the correlation threshold as wind turbines with abnormal working conditions. In this way, the maintenance of the wind turbines in the entire wind farm is taken into account, the abnormal conditions of the wind turbines in the wind farm are comprehensively analyzed, the efficiency of the maintenance of the wind turbines in the wind farm is improved, and it is also convenient to directly monitor the wind turbines with abnormal working conditions in the later stage. After the step of determining the wind turbines with abnormal working conditions, the method also includes marking the wind turbines with abnormal working conditions, and the marking of the wind turbines with abnormal working conditions can be used to facilitate the search for the wind turbines with abnormal working conditions in the later stage. For the above-mentioned wind turbines with abnormal working conditions, the method may also include abnormal working condition warning for the wind turbines with abnormal working conditions. This is convenient to remind users and timely protect the wind turbines with abnormal working conditions. After the above step 1, in other embodiments, the method also includes deleting the wind turbines whose correlation with other wind turbines in the wind farm except the wind turbine cluster is not greater than the correlation threshold from the wind turbines in the wind farm, reducing the data for subsequent processing.

[0108] The above step 1 further includes step 1, determining the correlation of wind speed data and wind direction data. Step 2, determining the average of the correlation of wind speed data and wind direction data as the correlation of working condition data.

[0109] Among them, the following Pearson correlation coefficient calculation formula is used to determine the correlation of wind speed data and wind direction data:

[0110]

[0111]

[0112]

[0113]

[0114] Where ρ X，Y represents the correlation coefficient between the two units, cov represents the covariance, X and Y represent the data sets of the two wind turbines in the predetermined time period (here refers to the two types of data: wind speed data and wind direction data), μ X , μ Y They represent the mean of the data of the two wind turbines in the predetermined time period, E represents the expectation, σ X , σ Y represents the standard deviation of the data of the two wind turbines within a predetermined time period, X i , Y i Respectively represent the i-th data related to the data of the two wind turbines, Indicates the average value of data sample X within a predetermined time period, It represents the sample average value of data sample Y within a predetermined time period. The value range of i is (1, n), n represents the number of wind turbines, and ∑ represents the sum.

[0115] The following mean calculation formula is used to determine the mean of the correlation between the wind speed data and the wind direction data as the correlation of the working condition data:

[0116]

[0117] In the formula, It represents the correlation of wind turbine operating data, which is the average of the correlation of wind speed data and wind direction data. They represent the correlation of wind speed data and wind direction data of wind turbines respectively, and are obtained by the Pearson correlation coefficient calculation formula introduced above.

[0118] In the embodiment of the above step 320, the average of the correlation of wind speed data and the correlation of wind direction data is used as the correlation of operating condition data. This can effectively reflect the operating conditions of multiple wind turbines through the correlation of operating condition data, thereby improving the accuracy of determining the operating conditions of multiple wind turbines.

[0119] In some other embodiments of the above step 320, the above step 320 may include using a clustering algorithm to classify multiple wind turbines in a wind farm according to wind direction data and wind speed data to obtain multiple wind turbine clusters. In this way, wind turbines with similar working conditions can be determined by the clustering algorithm, and it is easier to determine wind turbines with yaw working conditions, and then it is easier to determine wind turbines with abnormal yaw. Among them, the clustering algorithm may include a k-meas algorithm. No detailed examples are given here.

[0120] Step 330, obtaining wind direction data, yaw frequency data and nacelle position data of wind turbines in multiple wind turbine clusters.

[0121] This step 330 may include obtaining the wind direction data, yaw count data and cabin position data of the wind turbines of a single cluster from multiple wind turbine clusters each time, and performing subsequent processing until the wind direction data, yaw count data and cabin position data of all clusters are obtained multiple times, and the number of acquisitions is the same as the number of wind turbine clusters, so that the amount of data processed each time is small.

[0122] Figure 4 Step 330 is similar to Figure 1 Step 110 in Figure 1 In step 110, Figure 4 In step 330, the object to be acquired is the wind turbines in the plurality of wind turbine clusters. In this way, it can be determined whether the plurality of wind turbines have yaw abnormalities.

[0123] Step 340 , determining the wind deflection angles of the multiple wind turbines in the wind turbine cluster according to the nacelle position data and wind direction data of the multiple wind turbines in the wind turbine cluster.

[0124] Step 350, determining whether the wind turbine has yaw abnormality based on the wind deflection angle and yaw frequency data.

[0125] Step 360, based on the wind deflection angles and yaw times of the multiple wind turbines in the wind turbine cluster, determine whether the wind turbines in the wind turbine cluster have yaw abnormalities. In this way, it is possible to determine whether the wind turbines in each wind turbine cluster have yaw abnormalities on the basis of distinguishing the wind turbines, which can improve the accuracy of yaw abnormality detection, and the whole process is performed according to each wind turbine cluster, so that the amount of data processed each time is relatively small and the processing efficiency is high.

[0126] Figure 5 Shown Figure 1 A heat map of the correlation between multiple wind turbines in the wind turbine yaw anomaly detection method is shown. Figure 6 The figure is a flow chart of an application example of the method for detecting yaw anomaly of a wind turbine provided in an embodiment of the present application. Figure 5 The correlation is represented by the degree of grayscale, and the higher the correlation, the darker the grayscale.

[0127] Taking 124 wind turbines in a 4MW offshore wind farm as an example, the embodiments of the present application are described in detail.

[0128] Step 410, extracting wind direction data, yaw frequency data and wind speed data of 124 wind turbines in the wind farm within a preset time period. The preset time period includes a six-hour sub-time period from 2019-03-03 00:00:00 to 2019-03-03 06:00:00.

[0129] Step 420, determine the average of the correlation between the wind speed data and the wind direction data of every two wind turbines in the plurality of wind turbines in the sub-time period as the correlation between the two wind turbines and the operating condition data to form a correlation coefficient matrix, and give Figure 5 Heat map of the correlation between multiple wind turbines.

[0130] Step 430, based on the correlation between the operating data of every two wind turbines in the wind farm, two wind turbines with a correlation greater than 0.7 are determined to belong to the same wind turbine cluster, and wind turbines that have not formed a cluster are determined to be wind turbines with abnormal operating conditions. Table 1 shows the wind turbine clusters of 124 wind turbines and the wind turbines that have not formed a cluster. For example, Figure 5The wind turbine 1 in the first column is taken as the reference, and the wind turbines with a correlation greater than 0.7 among the 10 rows of wind turbines in the first column, such as wind turbine 1, wind turbine 2, wind turbine 3, wind turbine 4, wind turbine 5, wind turbine 33, wind turbine 34, wind turbine 35, and wind turbine 36, are determined to be the same wind turbine cluster. For example, this wind turbine cluster is called wind turbine cluster 1. Then from Figure 5 In the above example, the correlation between all rows and columns in wind turbine cluster 1 can be deleted. After that, the wind turbine 6 in column 1 is taken as the standard to determine the wind turbines in the 10 rows in column 1 whose correlation with wind turbine 6 in column 1 is greater than 0.7. For details, please refer to wind turbine cluster 2. By analogy, the wind turbine cluster and the wind turbines that have not formed a cluster are obtained based on the correlation of 124 wind turbines. The amount of data processed in this way is reduced, which can improve the processing efficiency.

[0131] Table 1 Wind turbine clusters and wind turbines that do not form a cluster

[0132]

[0133] Step 440, based on multiple wind turbine clusters, determine the number of wind turbine yaws within 6 hours for each wind turbine in the multiple wind turbines in each wind turbine cluster and the wind deflection angle for each hour within 6 hours. The number of wind turbine yaws within 6 hours includes the number of wind turbine yaws recorded once every hour. In this case, the number of wind turbine yaws here is the cumulative sum of the number of wind turbine yaws in 6 hours. The number of wind turbine yaws includes the number of wind turbine yaws within 6 hours and the number of wind turbine yaws is data recorded once every 6 hours. In this case, the number of wind turbine yaws here is the number of wind turbine yaws in 6 hours.

[0134] Step 450, determining an angle threshold value based on the mean of the wind deflection angles of multiple wind turbines. For example, a wind turbine cluster 1 including 9 wind turbines is used for illustration. First, the mean of the wind deflection angles of the 9 wind turbines is averaged to obtain an average value μ1 of the mean of the wind deflection angles. Next, the standard deviation of the mean of the wind deflection angles of the 9 wind turbines is calculated to obtain a standard deviation σ1 of the mean of the wind deflection angles. Then, the angle threshold value is obtained based on the average value μ1 and the standard deviation σ1.

[0135] Step 460, determining a yaw count threshold value based on the yaw count data of the plurality of wind turbines. For example, take the wind turbine cluster 1 including 9 wind turbines as an example. First, average the yaw count data of the 9 wind turbines to obtain an average value μ2 of the yaw count data. Next, calculate the standard deviation of the yaw count data of the 9 wind turbines to obtain a standard deviation σ2 of the yaw count data. Then, obtain the angle threshold value based on the average value μ2 and the standard deviation σ2.

[0136] Step 470, the wind turbines whose average wind deflection angle exceeds the angle threshold and / or whose yaw number data exceeds the yaw number threshold are determined as yaw abnormal wind turbines. The underlined and bold parts in Table 2 below are marked as yaw abnormal wind turbines and their data. For example, yaw abnormal wind turbine 5 in wind turbine cluster 1, yaw abnormal wind turbine 23 in wind turbine cluster 6, yaw abnormal wind turbine 64 and yaw abnormal wind turbine 66 in wind turbine cluster 3, and yaw abnormal wind turbine 92 in wind turbine cluster 4.

[0137] Table 2 Wind turbines with abnormal yaw in the wind turbine cluster

[0138]

[0139]

[0140]

[0141]

[0142] Step 480: Mark the wind turbines with yaw abnormality in the sub-time period as yaw abnormal wind turbines, and mark the wind turbines with abnormal operating conditions.

[0143] In step 490, the preset time period also includes a six-hour sub-time period from 2019-03-03 06:00:00 to 2019-03-03 12:00:00, a six-hour sub-time period from 2019-03-03 12:00:00 to 2019-03-03 18:00:00, and a six-hour sub-time period from 2019-03-03 18:00:00 to 2019-03-03 00:00:00. According to the sub-time periods of the preset time period, return to continue to execute steps 420 to 480, repeat the execution of steps 420 to 480 according to the above three sub-time periods, until the sub-time periods are used up, and continue to execute step 510.

[0144] Step 510: Count the number of times the same wind turbine with abnormal yaw is marked within the predetermined time period.

[0145] Step 520, determine the fault level of the same yaw abnormal wind turbine according to the number of markings and the number of segments of multiple sub-time periods. If the yaw abnormal wind turbine is marked for more than or equal to 3 times, the yaw abnormal wind turbine is determined as a high-risk yaw abnormal wind turbine; if the yaw abnormal wind turbine is marked for 2 times, the yaw abnormal wind turbine is determined as a medium-risk yaw abnormal wind turbine; if the yaw abnormal wind turbine is marked for 1 time, the yaw abnormal wind turbine is determined as a low-risk yaw abnormal wind turbine. Table 3 gives the fault levels of high-risk yaw abnormal wind turbines, medium-risk yaw abnormal wind turbines, and low-risk yaw abnormal wind turbines.

[0146] Table 3 Fault levels and numbers of abnormal yaw fans

[0147]

[0148] Figure 7 Shown is a module schematic diagram of a wind turbine yaw anomaly detection device provided in an embodiment of the present application.

[0149] like Figure 7 As shown, the wind turbine yaw abnormality detection and early warning device comprises:

[0150] The acquisition module 21 is used to acquire the wind direction data, yaw frequency data and cabin position data of the wind turbine.

[0151] The first processing module 22 is used to determine the wind deflection angle of the wind turbine according to the nacelle position data and the wind direction data.

[0152] The second processing module 23 is used to determine whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw frequency data.

[0153] The implementation process of the functions and effects of each unit in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and will not be repeated here.

[0154] Figure 8 The figure shows a schematic diagram of the structure of an electronic device 30 provided by an embodiment of the present application. The electronic device 30 may include a processor 31, a memory 33 storing machine executable instructions, and a communication interface 32. The processor 31 and the memory 33 may communicate via a system bus 34. In addition, by reading and executing the machine executable instructions corresponding to the data pull or data return logic in the memory 33, the processor 31 may execute the method described above.

[0155] The memory 33 mentioned herein may be any electronic, magnetic, optical or other physical storage device that may contain or store information, such as executable instructions, data, etc. For example, the machine-readable storage medium may be: RAM (RadomAccess Memory), volatile memory, non-volatile memory, flash memory, storage drive (such as hard disk drive), solid state drive, any type of storage disk (such as optical disk, DVD, etc.), or similar storage medium, or a combination thereof.

[0156] In some embodiments, a machine-readable storage medium is also provided. Figure 8 The memory 33 in the machine readable storage medium stores machine executable instructions, and the method described above is implemented when the machine executable instructions are executed by the processor. For example, the machine readable storage medium can be ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage device, etc.

[0157] The present application also provides a computer program stored in a machine-readable storage medium, such as Figure 8 The memory 33 in the memory, and when the processor executes the computer program, it causes the processor 31 to execute the method described above.

[0158] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

[0159] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

Claims

1. A method for detecting abnormal yaw of a wind turbine, characterized in that: include: Obtain wind direction data, yaw frequency data and cabin position data of wind turbines; Determining the wind deflection angle of the wind turbine according to the nacelle position data and the wind direction data; Determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data; Wherein, determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data includes: Determine the average value of the wind deflection angle of the wind turbine at multiple times within a predetermined time period; Determining an angle threshold according to an average of the wind deflection angles of the plurality of wind turbines; Determining the yaw frequency data of the wind turbine within the predetermined time period; Determining a yaw number threshold value according to the yaw number data of the plurality of wind turbines; When at least one of the average of the wind deflection angles of the wind turbine exceeds the angle threshold and the yaw frequency data of the wind turbine exceeds the yaw frequency threshold, it is determined that the wind turbine has yaw abnormality.

2. The method for detecting yaw anomaly of a wind turbine according to claim 1, characterized in that: Determining the yaw number threshold value according to the yaw number data of the plurality of wind turbines comprises: Determine an average value of the yaw number data of a plurality of the wind turbines; Determining a standard deviation of the yaw number data of a plurality of the wind turbines; The yaw number threshold is determined according to an average value of the yaw number data and a standard deviation of the yaw number data.

3. The method for detecting yaw anomaly of a wind turbine according to claim 2, characterized in that: Determining the angle threshold according to the average of the wind deflection angles of the plurality of wind turbines comprises: Determining an average of the mean values ​​of the wind deflection angles of a plurality of the wind turbines; Determining a standard deviation of the mean values ​​of the wind deflection angles of the plurality of wind turbines; The angle threshold is determined according to an average value of the mean values ​​of the windward deflection angles and a standard deviation of the mean values ​​of the windward deflection angles.

4. The method for detecting yaw anomaly of a wind turbine according to claim 1, characterized in that: The predetermined time period includes a plurality of sub-time periods, and the plurality of sub-time periods have the same length; The determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data includes: Determining whether the plurality of wind turbines in the sub-time period have yaw abnormalities according to the wind deflection angle and the yaw frequency data in the sub-time period; After determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data, the method further includes: Marking the wind turbine generator having yaw abnormality in the sub-time period as a yaw abnormal wind turbine; Counting the number of times the same abnormal yaw wind turbine is marked within the predetermined time period; The fault level of the wind turbine with the same yaw abnormality is determined according to the number of markings and the number of sub-time periods.

5. The method for detecting yaw anomaly of a wind turbine according to claim 4, characterized in that: The determining, according to the number of markings and the number of sub-time periods, of the fault level of the same wind turbine with abnormal yaw includes: Determine the wind turbine with abnormal yaw, whose number of marks is greater than half of the number of sub-time periods, as a wind turbine with high-risk abnormal yaw; Determine the wind turbine with yaw abnormality whose marking times are equal to half of the number of segments in the sub-time period as a wind turbine with medium-risk yaw abnormality; The wind turbine with abnormal yaw, whose marking times are greater than zero and less than half of the number of segments in the sub-time period, is determined as a wind turbine with low-risk yaw abnormality.

6. The method for detecting yaw anomaly of a wind turbine according to claim 1, characterized in that: The method comprises: Acquire operating data of a plurality of wind turbines in a wind farm, wherein the operating data includes the wind direction data and the wind speed data; Classifying a plurality of wind turbines in the wind farm according to the wind direction data and the wind speed data to obtain a plurality of wind turbine clusters; Determining the wind deflection angle of the wind turbine according to the nacelle position data and the wind direction data includes: Determining the wind deflection angles of the multiple wind turbines in the wind turbine cluster according to the nacelle position data and the wind direction data of the multiple wind turbines in the wind turbine cluster; The determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data includes: According to the wind deflection angles and the yaw frequency data of the plurality of wind turbines in the wind turbine cluster, it is determined whether the wind turbines in the wind turbine cluster have abnormal yaw.

7. The method for detecting yaw anomaly of a wind turbine according to claim 6, characterized in that: The method of classifying a plurality of wind turbines in the wind farm according to the wind direction data and the wind speed data to obtain a plurality of wind turbine clusters includes: Determining the correlation between the operating condition data of every two wind turbines among the plurality of wind turbines in the wind farm; Every two wind turbines whose correlation is greater than a correlation threshold are determined to belong to the same wind turbine cluster.

8. The method for detecting yaw anomaly of a wind turbine according to claim 7, characterized in that: Determining the correlation between the operating condition data of every two wind turbines among the plurality of wind turbines in the wind farm comprises: Determining a correlation between the wind speed data and a correlation between the wind direction data; An average value of the correlation of the wind speed data and the correlation of the wind direction data is determined as the correlation of the operating condition data.

9. The method for detecting yaw anomaly of a wind turbine according to claim 7, characterized in that: After determining the correlation between the operating condition data of every two wind turbines among the plurality of wind turbines in the wind farm, the method further includes: A wind turbine whose correlation with other wind turbines in the wind farm except the wind turbine cluster is not greater than the correlation threshold is determined as a wind turbine with abnormal operating condition.

10. A wind turbine yaw abnormality detection and warning device, characterized in that: include: An acquisition module is used to acquire wind direction data, yaw frequency data and cabin position data of the wind turbine; A first processing module, used for determining the wind deflection angle of the wind turbine according to the nacelle position data and the wind direction data; A second processing module is used to determine whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data; Wherein, determining whether the wind turbine has abnormal yaw according to the wind deflection angle and the yaw number data includes: Determine the average value of the wind deflection angle of the wind turbine at multiple times within a predetermined time period; Determining an angle threshold according to an average of the wind deflection angles of the plurality of wind turbines; Determining the yaw frequency data of the wind turbine within the predetermined time period; Determining a yaw number threshold value according to the yaw number data of the plurality of wind turbines; When at least one of the average of the wind deflection angles of the wind turbine exceeds the angle threshold and the yaw frequency data of the wind turbine exceeds the yaw frequency threshold, it is determined that the wind turbine has yaw abnormality.

11. An electronic device, characterized in that: including a processor and a memory; Memory, used to store computer programs; A processor, for implementing the method according to any one of claims 1 to 9 when executing a program stored in a memory.

12. A computer-readable storage medium, characterized in that: A program is stored thereon, and when the program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

Images