Abnormal vibration identification method, device and wind turbine generator set

Abstract

The present invention provides an abnormal vibration identification method, device and wind turbine generator set, comprising: screening the operating data of a target wind turbine generator within a preset time period, obtaining an operating data set, obtaining a yaw state data set and a non-yaw state data set, and respectively determining a first vibration amplitude distribution and a second vibration amplitude distribution; based on the first vibration amplitude distribution and the second vibration amplitude distribution, determining abnormal vibration information of the target wind turbine generator. The present invention performs a comprehensive comparative analysis of the first vibration amplitude distribution obtained by analyzing the operating data in the yaw state and the second vibration amplitude distribution obtained by analyzing the operating data in the non-yaw state, and can determine the abnormal vibration information of the wind turbine generator based on the analysis results, effectively distinguish various causes of abnormal vibration of the yaw system, more conveniently locate the abnormal vibration caused by the yaw system, and facilitate guiding staff to achieve precise management of the wind turbine generator set.

Description

Technical Field

[0001] The present invention relates to the field of new energy technologies, and in particular to an abnormal vibration identification method, device and wind turbine generator set. Background Art

[0002] The yaw system of a wind turbine is a device that can adjust the wind rotor to face the main wind direction in order to absorb wind energy to the maximum extent. It is generally composed of a yaw bearing, a yaw drive device, a yaw brake, a yaw counter, a cable protection device, and a yaw hydraulic circuit.

[0003] The yaw system is worn or there are foreign objects such as oil leakage, surface enamel layer, and anti-rust layer on the friction disc surface that have not been cleaned up. These abnormal conditions will affect the smooth operation of the wind turbine and pose certain safety hazards.

[0004] Currently, the industry generally identifies yaw system brake pad issues based on the relationship between cumulative yaw angle and yaw brake pad wear. Algorithms used to monitor yaw system wear generally fall into two categories. One involves intelligent brake pad wear detection based on multi-sensor technology, using various switch detection modules for collaborative detection. The other involves building a data-driven detection model that combines mechanism analysis and data mining methods, making judgments based on abnormal cumulative yaw angles.

[0005] The above-mentioned abnormal vibration detection of the yaw system can only detect abnormal vibration caused by brake pad wear, but cannot effectively distinguish various causes of abnormal vibration of the yaw system, making it difficult to guide staff to achieve accurate management of wind turbines. Summary of the Invention

[0006] The present invention provides an abnormal vibration identification method, device and wind turbine generator set, which are used to solve the defect in the existing technology that the actual factors causing abnormal vibration of the wind turbine generator set cannot be accurately classified and identified. It can complete the accurate analysis and identification of the factors causing abnormal vibration of the yaw system, and facilitate the guidance of staff to achieve precise management of the wind turbine generator set.

[0007] In a first aspect, the present invention provides a method for identifying abnormal vibrations, comprising:

[0008] Filtering the operating data of the target wind turbine within a preset time period to obtain an operating data set; the operating data includes the vibration amplitude and yaw angle at any sampling time point;

[0009] Dividing the operating data set into a yaw state data set and a non-yaw state data set;

[0010] determining, based on the yaw state data set, a first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state, and determining, based on the non-yaw state data set, a second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state;

[0011] Abnormal vibration information of the target wind turbine within the preset time period is determined based on the first vibration amplitude distribution and the second vibration amplitude distribution.

[0012] According to an abnormal vibration identification method provided by the present invention, determining a first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state based on the yaw state data set, and determining a second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state based on the non-yaw state data set, includes:

[0013] determining an absolute position of a nacelle of the target wind turbine based on the yaw state data set;

[0014] Dividing a plurality of sector-shaped areas with the absolute position of the nacelle as the center, and taking each sector-shaped area as a yaw angle area;

[0015] For each vibration amplitude and yaw angle at each sampling time point in the yaw state data set, assigning the vibration amplitude to a target yaw angle region according to the target yaw angle region to which the yaw angle belongs, so as to obtain a first vibration amplitude distribution corresponding to the yaw state data set;

[0016] re-determining the absolute position of the nacelle of the target wind turbine based on the non-yaw state data set;

[0017] Dividing a plurality of sector-shaped areas with the new absolute position of the nacelle as the center, and taking each sector-shaped area as a yaw angle area;

[0018] For the vibration amplitude and yaw angle at each sampling time point in the non-yaw state data set, the vibration amplitude is placed in the target yaw angle region according to the target yaw angle region to which the yaw angle belongs, so as to obtain a second vibration amplitude distribution corresponding to the non-yaw state data set.

[0019] According to an abnormal vibration identification method provided by the present invention, the dividing of the plurality of sector-shaped areas with the absolute position of the cabin as the center comprises dividing the circular area with the absolute position of the cabin as the center by a first preset angle to obtain the plurality of sector-shaped areas;

[0020] Accordingly, dividing the plurality of sector-shaped areas with the new absolute position of the cabin as the center includes dividing the circular area with the absolute position of the cabin as the center by a second preset angle as a step size to obtain the plurality of sector-shaped areas;

[0021] The first preset angle is the same as or different from the second preset angle.

[0022] According to an abnormal vibration identification method provided by the present invention, determining abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution includes:

[0023] Determining, based on the first vibration amplitude distribution, a first number of vibration amplitudes in each yaw angle region that are greater than a first amplitude threshold and a first total number of vibration amplitudes in each yaw angle region;

[0024] taking the yaw angle regions for which the ratio between the first number and the first total number is greater than a first preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the yaw state;

[0025] Determining, based on the second vibration amplitude distribution, a second number of vibration amplitudes in each yaw angle region that are greater than a second amplitude threshold and a second total number of vibration amplitudes in each yaw angle region;

[0026] taking the yaw angle regions for which the ratio between the second number and the second total number is greater than a second preset threshold as abnormal yaw angle regions in a non-yaw state, to determine the total number of abnormal yaw angle regions in the non-yaw state;

[0027] If the total number of abnormal yaw angle regions in the yaw state is greater than a third preset threshold, and the total number of abnormal yaw angle regions in the non-yaw state is not greater than a fourth preset threshold, it is determined that the abnormal vibration information is caused by a brake pad failure;

[0028] If the total number of abnormal yaw angle areas in the yaw state is greater than a third preset threshold, the total number of abnormal yaw angle areas in the non-yaw state is greater than a fourth preset threshold, and the distribution of the abnormal yaw angle areas matches the wind frequency distribution in the abnormal terrain state, then it is determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain.

[0029] According to an abnormal vibration identification method provided by the present invention, determining, based on the first vibration amplitude distribution, a first number of vibration amplitudes in each yaw angle region that are greater than a first amplitude threshold and a first total number of vibration amplitudes in each yaw angle region, includes:

[0030] Based on the first vibration amplitude distribution, a plurality of vibration intervals are divided into equal intervals according to a first fixed value to form a vibration interval group; the plurality of vibration intervals include an abnormal vibration amplitude interval and a normal vibration amplitude interval, the minimum value of the abnormal vibration amplitude interval is greater than the first amplitude threshold, and the maximum value of the normal vibration amplitude interval is less than or equal to the first amplitude threshold;

[0031] In each yaw angle region, determining the vibration interval to which each vibration amplitude belongs, and using the number of vibration amplitudes in each vibration interval as the weight of the vibration interval;

[0032] determining the first number and a first total number of the vibration amplitudes in each yaw angle region, where the first number is the sum of weights of all abnormal vibration amplitude intervals;

[0033] Accordingly, determining, based on the second vibration amplitude distribution, a second number of vibration amplitudes in each yaw angle region that are greater than a second amplitude threshold and a second total number of vibration amplitudes in each yaw angle region includes:

[0034] Based on the second vibration amplitude distribution, a plurality of vibration intervals are divided into equal intervals according to a second fixed value to form a vibration interval group; the plurality of vibration intervals include an abnormal vibration amplitude interval and a normal vibration amplitude interval, the minimum value of the abnormal vibration amplitude interval is greater than the second amplitude threshold, and the maximum value of the normal vibration amplitude interval is less than or equal to the second amplitude threshold;

[0035] In each yaw angle region, determining the vibration interval to which each vibration amplitude belongs, and using the number of vibration amplitudes in each vibration interval as the weight of the vibration interval;

[0036] determining the second number and a second total number of the vibration amplitudes in each yaw angle region, the second number being the sum of the weights of all abnormal vibration amplitude intervals;

[0037] The first constant value is the same as or different from the second constant value.

[0038] According to an abnormal vibration identification method provided by the present invention, the first amplitude threshold is the same as the second amplitude threshold, the first preset threshold is the same as the second preset threshold, and the third preset threshold is the same as the fourth preset threshold.

[0039] According to an abnormal vibration identification method provided by the present invention, determining abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution includes:

[0040] Based on the first vibration amplitude distribution, the vibration amplitude in each yaw angle region is divided into bins, and a fatigue estimation value corresponding to each bin is calculated;

[0041] Determine the first number of bins in which fatigue estimation values ​​in all bins are greater than a first fatigue threshold and the first total number of bins in each yaw angle region;

[0042] taking the yaw angle regions where the ratio between the first number of bins and the first total number of bins is greater than a fifth preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the yaw state;

[0043] Based on the second vibration amplitude distribution, the vibration amplitude in each yaw angle region is divided into bins, and a fatigue estimation value corresponding to each bin is calculated;

[0044] Determine the number of second bins in which fatigue estimation values ​​in all bins are greater than a second fatigue threshold and the second total number of bins in each yaw angle region;

[0045] taking the yaw angle regions in which the ratio between the second number of bins and the second total number of bins is greater than a sixth preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the non-yaw state;

[0046] If the total number of abnormal yaw angle regions in the yaw state is greater than a seventh preset threshold, and the total number of abnormal yaw angle regions in the non-yaw state is not greater than an eighth preset threshold, determining that the abnormal vibration information is caused by a brake pad failure;

[0047] If the total number of abnormal yaw angle areas in the yaw state is greater than the seventh preset threshold, the total number of abnormal yaw angle areas in the non-yaw state is greater than the eighth preset threshold, and the distribution of the abnormal yaw angle areas matches the wind frequency distribution in the abnormal terrain state, then it is determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain.

[0048] According to an abnormal vibration identification method provided by the present invention, the first fatigue threshold is the same as the second fatigue threshold, the fifth preset threshold is the same as the sixth preset threshold, and the seventh preset threshold is the same as the eighth preset threshold.

[0049] In a second aspect, the present invention further provides an abnormal vibration identification device, comprising:

[0050] A data acquisition unit is used to screen the operating data of the target wind turbine within a preset time period and obtain an operating data set; the operating data includes the vibration amplitude and yaw angle at any sampling time point;

[0051] A data cleaning unit, configured to divide the operating data set into a yaw state data set and a non-yaw state data set;

[0052] a data calibration unit, configured to determine, based on the yaw state data set, a first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state, and to determine, based on the non-yaw state data set, a second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state;

[0053] The vibration analysis unit is configured to determine abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution.

[0054] In a third aspect, the present invention provides a wind turbine group comprising at least one target wind turbine and a vibration processor, and further comprising a memory and a program or instruction stored in the memory and executable on the vibration processor, wherein the program or instruction, when executed by the vibration processor, performs the abnormal vibration identification method as described in any one of the first aspects.

[0055] In a fourth aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the abnormal vibration identification method as described above is implemented.

[0056] In a fifth aspect, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the above-described abnormal vibration identification methods.

[0057] The abnormal vibration identification method, device and wind turbine provided by the present invention conduct a comprehensive comparative analysis by analyzing the first vibration amplitude distribution obtained by analyzing the operating data in the yaw state and the second vibration amplitude distribution obtained by analyzing the operating data in the non-yaw state. The abnormal vibration information of the wind turbine can be determined based on the analysis results, and various causes of abnormal vibration of the yaw system can be effectively distinguished, which makes it more convenient to better locate the abnormal vibration caused by the yaw system, and facilitates guiding staff to achieve precise management of the wind turbine. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0059] Figure 1 This is one of the flow charts of the abnormal vibration identification method provided by the present invention;

[0060] Figure 2 This is the second flow chart of the abnormal vibration identification method provided by the present invention;

[0061] Figure 3 It is a structural schematic diagram of the abnormal vibration identification device provided by the present invention;

[0062] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention; DETAILED DESCRIPTION

[0063] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0064] It should be noted that, in the description of the embodiments of the present invention, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "include a ..." do not exclude the presence of other identical elements in the process, method, article or device comprising the elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0065] The terms "first," "second," and the like in this application are used to distinguish similar objects, and are not used to describe a specific order or precedence. It should be understood that such terms are interchangeable where appropriate, so that embodiments of this application can be implemented in an order other than that illustrated or described herein. Furthermore, the terms "first," "second," and the like generally distinguish objects of a class and do not limit the number of objects; for example, the first object can be one or more. Furthermore, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates an "or" relationship between the connected objects.

[0066] Abnormal vibration of the yaw system is often caused by changes in the physical properties of its components, such as abnormal wear of the brake pads or foreign matter between the brake pads and the brake disc. However, it can also be caused by the operating environment of the wind turbine, such as special terrain.

[0067] The abnormal vibration identification method for the yaw system of a wind turbine provided by the present invention collects the operating data of the wind turbine and compares the operating data characteristics in the yaw state and the non-yaw state to accurately distinguish the cause of the abnormal vibration of the yaw system, so as to guide the staff to achieve precise management of the wind turbine.

[0068] Figure 1 This is one of the flow charts of the abnormal vibration identification method provided by the present invention, such as Figure 1 As shown, including but not limited to the following steps:

[0069] Step 101: Filtering the operating data of a target wind turbine within a preset time period to obtain an operating data set; the operating data includes the vibration amplitude and yaw angle at any sampling time point.

[0070] In order to clearly demonstrate the abnormal vibration identification method provided by the present invention, any target wind turbine in a wind turbine group is used as a research object for description.

[0071] The present invention collects operating data within a preset time period before and after the abnormal vibration occurs, and uses the operating data to complete the analysis of the cause of the abnormal vibration.

[0072] The preset time period can be set according to the requirements of recognition accuracy, such as three days, one month or three months, etc., which is not specifically limited in the present invention.

[0073] The operating data collected by the present invention is determined by reading the measured operating data of each wind turbine in the wind turbine group from the wind farm SCADA system. This data primarily includes the vibration amplitude and yaw angle of the yaw system of the target wind turbine at each sampling time point within a preset time period. It may also include other data such as the yaw position, true north position, nacelle position, and wind frequency data at each sampling time point. The relevant operating data collected at each sampling time point can be grouped together to construct an operating data list.

[0074] It should be noted that the present invention does not require additional sensors to collect operating data.

[0075] Step 102: Divide the operating data set into a yaw state data set and a non-yaw state data set.

[0076] Furthermore, the present invention divides the operating data collected at each sampling time point according to the operating state of the yaw system, including:

[0077] If the yaw system is in the yaw state at a certain sampling time point, that is, the yaw system is in the adjustment stage of adjusting the yaw angle, then the relevant operating data collected at the sampling time point is used as the yaw state data, and the yaw state data set is constructed from all the yaw state data collected in the yaw state.

[0078] If the yaw system is in a non-yaw state at a certain sampling time point, that is, the yaw system is in a stage where the yaw angle is not adjusted (which can be understood as the yaw system is in a non-operating state), then the relevant operating data collected at this sampling time point will be used as the non-yaw state data, and the non-yaw state data set will be constructed from all the non-yaw state data collected in the non-yaw state.

[0079] Step 103: Based on the yaw state data set, determine the first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state, and based on the non-yaw state data set, determine the second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state.

[0080] It should be noted that the present invention first independently analyzes the yaw state data set in the yaw state and the non-yaw state data set in the non-yaw state to obtain the first vibration amplitude distribution and the second vibration amplitude distribution respectively.

[0081] Taking the analysis of the yaw state data set in the yaw state to obtain the first vibration amplitude distribution as an example, it mainly includes:

[0082] The absolute position of the nacelle of the target wind turbine is taken as the center point, and the 360° area where the center point is located is divided according to the preset angle. Each sector area obtained is regarded as a yaw angle area. For example, if the preset angle is 10 degrees, each yaw angle area refers to a sector area with an angle of 10 degrees and the absolute position of the nacelle as the center. In this way, each target wind turbine corresponds to 36 yaw angle areas.

[0083] Furthermore, the operation data corresponding to each sampling time point in the yaw state data set is determined to be placed in the yaw angle region to which the yaw angle of the operation data belongs, based on the yaw angle region to which the yaw angle of the operation data belongs.

[0084] The vibration amplitude of the operating data is assigned to the yaw angle region in the following manner:

[0085] Draw a yaw angle area distribution diagram of the target wind turbine generator in the yaw state. If the yaw angle of the operating data collected at any sampling time point belongs to a certain yaw angle area (hereinafter referred to as the target yaw angle area), then the sampling time point is plotted in the target yaw angle area as a marked point. The larger the vibration amplitude at the sampling time point, the farther the marked point is from the center point.

[0086] The yaw angle area where the marked point corresponding to each sampling point is located represents the yaw angle corresponding to the sampling point, and the distance between the marked point and the center point represents the vibration amplitude corresponding to the sampling point.

[0087] In this way, the operating data of all sampling time points in the yaw state data set can be plotted in the yaw angle area distribution diagram in the yaw state in the form of marked points to obtain the first vibration amplitude distribution.

[0088] Similarly, referring to the above method, the operating data of all sampling time points in the non-yaw state data set can be plotted as marked points in the yaw angle area distribution diagram in the non-yaw state to obtain the second vibration amplitude distribution, which will not be described in detail here.

[0089] Step 104: Determine abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution.

[0090] If the abnormal vibration of the yaw system is caused by an abnormal brake pad, such as brake pad wear or the presence of foreign matter, then as the yaw angle changes, the abnormal vibration data in the yaw state will appear in all or most yaw angle areas, but in the non-yaw state (when the yaw system is not in operation), the abnormal vibration data will not appear in the yaw angle area, or abnormal vibration data will only appear in a few yaw angle areas.

[0091] If the abnormal vibration of the yaw system is caused by an external environment, such as abnormal terrain, then as the yaw angle changes, the abnormal vibration data will appear in all or most yaw angle regions, whether in the yaw state or the non-yaw state.

[0092] Abnormal vibration data refers to data in which the vibration amplitude in the operating data is greater than a certain amplitude threshold.

[0093] In view of this, the present invention analyzes the first vibration amplitude distribution and the second vibration amplitude distribution. If it is determined that abnormal vibration data exists in most yaw angle areas in the first vibration amplitude distribution, and abnormal vibration data does not exist in most yaw angle areas in the second vibration amplitude distribution, then it is determined that the abnormal vibration information of the target wind turbine within the preset time period is abnormal vibration caused by brake pad failure; if it is determined that abnormal vibration data exists in most yaw angle areas in the first vibration amplitude distribution, and abnormal vibration data also exists in most yaw angle areas in the second vibration amplitude distribution, then it is determined that the abnormal vibration information of the target wind turbine within the preset time period is abnormal vibration caused by abnormal terrain.

[0094] The abnormal vibration identification method provided by the present invention performs a comprehensive comparative analysis by analyzing a first vibration amplitude distribution obtained by analyzing the operating data in the yaw state and a second vibration amplitude distribution obtained by analyzing the operating data in the non-yaw state. The abnormal vibration information of the wind turbine can be determined based on the analysis results, and various causes of abnormal vibration of the yaw system can be effectively distinguished, making it more convenient to better locate the abnormal vibration caused by the yaw system, and facilitating the guidance of staff to achieve precise management of the wind turbine set.

[0095] Based on the content of the above embodiment, as an optional embodiment, determining the first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state based on the yaw state data set, and determining the second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state based on the non-yaw state data set includes:

[0096] determining an absolute position of a nacelle of the target wind turbine based on the yaw state data set;

[0097] Dividing a plurality of sector-shaped areas with the absolute position of the nacelle as the center, and taking each sector-shaped area as a yaw angle area;

[0098] For each vibration amplitude and yaw angle at each sampling time point in the yaw state data set, assigning the vibration amplitude to a target yaw angle region according to the target yaw angle region to which the yaw angle belongs, so as to obtain a first vibration amplitude distribution corresponding to the yaw state data set;

[0099] re-determining the absolute position of the nacelle of the target wind turbine based on the non-yaw state data set;

[0100] Dividing a plurality of sector-shaped areas with the new absolute position of the nacelle as the center, and taking each sector-shaped area as a yaw angle area;

[0101] For the vibration amplitude and yaw angle at each sampling time point in the non-yaw state data set, the vibration amplitude is placed in the target yaw angle region according to the target yaw angle region to which the yaw angle belongs, so as to obtain a second vibration amplitude distribution corresponding to the non-yaw state data set.

[0102] This embodiment provides a method for determining a first vibration amplitude distribution according to a yaw state data set, and determining a second vibration amplitude distribution according to a non-yaw state data set.

[0103] Taking the determination of the first vibration amplitude distribution based on the yaw state data set as an example, the following is explained:

[0104] First, the absolute position of the nacelle of the target wind turbine can be determined according to the yaw position, true north position, etc. of the target wind turbine in the yaw state data set, which will not be elaborated in the present invention.

[0105] Taking the absolute position of the cabin as the center point (or called the center), multiple sector-shaped areas are evenly divided according to the preset angle, and each sector-shaped area is used as a yaw angle area. For example, if the preset angle is 10°, 36 yaw angle areas can be divided. The vertex of each yaw angle area is the center point, and the angle is 10°.

[0106] Furthermore, for the vibration amplitude and yaw angle collected at each sampling time point in the yaw state data set, if the yaw angle at a certain sampling time point belongs to a certain yaw angle area (called a target yaw angle area), the vibration amplitude at the sampling time point is placed within the target yaw angle area.

[0107] Optionally, each yaw angle region is divided into a plurality of vibration intervals at equal intervals according to a certain value to construct a vibration interval group.

[0108] For example, for any yaw angle region, assuming a (unit: g), the yaw angle region can be divided into the following nine vibration intervals: 0-a, a-2a, 2a-3a, 3a-4a, 4a-5a, 5a-6a, 6a-7a, 7a-8a, and greater than 8a. The above nine vibration intervals constitute a vibration interval group.

[0109] Placing the vibration amplitude at the sampling time point within the target yaw angle region may be performed by representing the vibration amplitude at the sampling time point in a corresponding vibration interval in the form of a marked point.

[0110] For example, if the polarization angle collected at a certain sampling time point is 36° and the vibration amplitude is 0.038g, then the annotation points of the operating data related to the sampling time point can be added in the vibration interval 5a-7a in the yaw angle area corresponding to 30°-40°.

[0111] Based on the above method, the operating data corresponding to each sampling time point in the yaw state data set is sequentially represented in the form of marked points to obtain the first vibration amplitude distribution corresponding to the yaw state data set.

[0112] Similarly, a similar method can be used to obtain the second vibration amplitude distribution corresponding to the non-yaw state data set, which will not be described in detail here.

[0113] It should be noted that when generating the second vibration amplitude distribution based on the operating data collected at each sampling time point in the non-yaw state data set, the absolute position of the cabin is re-determined and the yaw angle area distribution in the non-yaw state is re-established, that is, the analysis of the yaw state data set and the non-yaw state data set can be performed independently.

[0114] The abnormal vibration identification method provided by the present invention provides a method for independently analyzing the operating data of a wind turbine in a yaw state and a non-yaw state. It does not require additional sensors to collect data and does not cost anything extra, and can achieve better positioning of abnormal vibrations in the yaw system.

[0115] Based on the content of the above embodiment, as an optional embodiment, dividing the plurality of sector-shaped areas with the absolute position of the cabin as the center includes dividing the circular area with the absolute position of the cabin as the center by a first preset angle to obtain the plurality of sector-shaped areas;

[0116] Accordingly, dividing the plurality of sector-shaped areas with the new absolute position of the cabin as the center includes dividing the circular area with the absolute position of the cabin as the center by a second preset angle as a step size to obtain the plurality of sector-shaped areas;

[0117] The first preset angle is the same as or different from the second preset angle.

[0118] The step of dividing the plurality of sector-shaped areas with the absolute position of the cabin as the center includes dividing a circular area with the absolute position of the cabin as the center with a first preset angle as a step length to obtain the plurality of sector-shaped areas;

[0119] Accordingly, dividing the plurality of sector-shaped areas with the new absolute position of the cabin as the center includes dividing the circular area with the absolute position of the cabin as the center by a second preset angle as a step size to obtain the plurality of sector-shaped areas;

[0120] The present invention provides a method for dividing a sector area with the absolute position of the cabin as the center, including a method for dividing the yaw angle area in the yaw state and a method for dividing the yaw angle area in the non-yaw state, and both can be performed independently.

[0121] Taking the method of dividing the yaw angle area in the yaw state as an example, after determining the absolute position of the cabin, it is used as the center point (also called the center), and the circular area centered on the absolute position of the cabin is divided at a first preset angle, such as 10° as a step size. 36 yaw angle areas in the yaw state can be obtained.

[0122] It should be noted that the second preset angle used when dividing the yaw angle regions in the non-yaw state may be the same as or different from the first preset angle used when dividing the yaw angle regions in the yaw state. Correspondingly, the number of yaw angle regions in the yaw state may be the same as or different from the number of yaw angle regions in the non-yaw state.

[0123] For example, when the first preset angle is set to 10°, the second preset angle can be set to 10° or 5°, etc., which is not specifically limited in the present invention. Correspondingly, the number of yaw angle regions in the non-yaw state is 72.

[0124] In the abnormal vibration identification method provided by the present invention, a modeling method is provided for vibration anomaly analysis based on the data characteristics of the operating data of the wind turbine generator set, which can establish an intuitive correlation between the vibration amplitude and the yaw angle, and finally the operating data collected by each sampling point can be placed in each yaw angle area. In this way, the abnormal vibration results can be quickly extracted according to the distribution of each operating data in each yaw angle area. There is no need to add additional sensors to collect data, there is no additional cost, and the recognition accuracy is high and the practicability is strong.

[0125] Based on the content of the above embodiment, as an optional embodiment, determining the abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution includes:

[0126] Determining, based on the first vibration amplitude distribution, a first number of vibration amplitudes in each yaw angle region that are greater than a first amplitude threshold and a first total number of vibration amplitudes in each yaw angle region;

[0127] taking the yaw angle regions for which the ratio between the first number and the first total number is greater than a first preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the yaw state;

[0128] Determining, based on the second vibration amplitude distribution, a second number of vibration amplitudes in each yaw angle region that are greater than a second amplitude threshold and a second total number of vibration amplitudes in each yaw angle region;

[0129] taking the yaw angle regions for which the ratio between the second number and the second total number is greater than a second preset threshold as abnormal yaw angle regions in a non-yaw state, to determine the total number of abnormal yaw angle regions in the non-yaw state;

[0130] If the total number of abnormal yaw angle regions in the yaw state is greater than a third preset threshold, and the total number of abnormal yaw angle regions in the non-yaw state is not greater than a fourth preset threshold, it is determined that the abnormal vibration information is caused by a brake pad failure;

[0131] If the total number of abnormal yaw angle areas in the yaw state is greater than a third preset threshold, the total number of abnormal yaw angle areas in the non-yaw state is greater than a fourth preset threshold, and the distribution of the abnormal yaw angle areas matches the wind frequency distribution in the abnormal terrain state, then it is determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain.

[0132] Figure 2 This is the second flow chart of the abnormal vibration identification method provided by the present invention, as shown in FIG. Figure 2 As shown, after the operation data of the target wind turbine is collected and cleaned as provided in the above embodiment, the acquired operation data set is divided into a yaw state data set and a non-yaw state data set according to the yaw state and the non-yaw state.

[0133] Then, a first vibration amplitude distribution is determined based on all the operating data in the yaw state data set, and a second vibration amplitude distribution is determined based on all the operating data in the non-yaw state data set.

[0134] Furthermore, the first vibration amplitude distribution and the second vibration amplitude distribution are analyzed independently. In this embodiment, the analysis of the first vibration amplitude distribution is taken as an example for description.

[0135] After the absolute position of the nacelle corresponding to the target wind turbine is determined according to all operating data in the yaw state data set, the circular area where the nacelle is located is divided into sectors with the absolute position of the nacelle as the center to obtain multiple yaw angle areas.

[0136] Then, each set of operating data in the yaw state data set (the yaw angle and vibration amplitude collected at each sampling time point are regarded as a group) is placed in the target yaw angle area according to the target yaw angle area to which the yaw angle belongs, and the first vibration amplitude distribution is obtained.

[0137] It should be noted that after obtaining the first vibration amplitude distribution, if it is determined that there are marked points corresponding to the operating data in the N0 yaw angle areas, the operating data collected within the preset time period is considered valid; otherwise, the operating data collected within the preset time period is considered invalid, so as to prevent the abnormal vibration identification result from being affected due to the unrepresentative sampling of the operating data.

[0138] Assuming that the total number of yaw angle areas divided is 36 and the value of N0 is 25, if it is determined that there are marked points corresponding to the operating data in more than 25 yaw angle areas, the subsequent recognition logic will continue to be executed; if not, the operating data within another preset time period will be re-acquired.

[0139] Assuming the first amplitude threshold is 0.03 (unit: g), for any yaw angle region, if the first total number of annotated points (determined by the yaw angle of the operating data) is 50, and the number of corresponding annotated points with vibration amplitudes greater than 0.03 is 20, then the ratio of the first number associated with that yaw angle region to the first total number can be calculated to be 40%. Repeat the above steps until 36 ratios associated with yaw angle regions are obtained (the ratio is 0 if there are no annotated points in the yaw angle region).

[0140] Assuming that the first preset threshold is 20%, it can be determined that all yaw angle areas with a ratio between the first number and the first total number greater than 20% are abnormal yaw angle areas. In this way, the total number of 50 yaw angle areas judged as abnormal yaw angle areas can be determined (denoted as K1).

[0141] Similarly, the above method can be used to determine the total number of abnormal yaw angle regions (denoted as K2) in all corresponding yaw angle regions according to the second vibration amplitude distribution.

[0142] Assume that the total number K1 of abnormal yaw angle areas among the 50 yaw angle areas determined by the yaw state data set is 30, and the total number K2 of abnormal yaw angle areas among the 50 yaw angle areas determined by the non-yaw state data set is 10.

[0143] It should be noted that, in the abnormal vibration identification method provided by the present invention, the yaw state data set and the non-yaw state data set can be processed independently to obtain the corresponding first vibration amplitude distribution and second vibration amplitude distribution, respectively; furthermore, by analyzing the first vibration amplitude distribution and the second vibration amplitude distribution, respectively, the total number of all the yaw angle areas related to each of them that are judged to be abnormal yaw angle areas is obtained.

[0144] Therefore, the first amplitude threshold and the second amplitude threshold may be the same or different, the first preset threshold and the second preset threshold may be the same or different, and the third preset threshold and the fourth preset threshold may also be the same or different.

[0145] However, in order to simplify the judgment logic, generally speaking, the first amplitude threshold can be equal to the second amplitude threshold, the first preset threshold can be equal to the second preset threshold, and the third preset threshold can be equal to the fourth preset threshold. This is equivalent to processing the yaw state data set and the non-yaw state data set in the same way.

[0146] Assume that the third preset threshold is equal to the fourth preset threshold, both are set to 15 (in units). Since K1=30 is greater than the third preset threshold, but K2=10 is not greater than the fourth preset threshold, it is determined that the abnormal vibration information is caused by brake pad failure.

[0147] Assuming K1=30 and K2=20, K1 is greater than the third preset threshold, and K2 is greater than the fourth preset threshold. In this case, it can be determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain.

[0148] It should be noted that when it is determined that the abnormal vibration information is caused by abnormal terrain, the distribution of marked points in the polarized state and the distribution of marked points in the non-polarized state can be analyzed to see whether they have obvious wind frequency distribution characteristics under the abnormal terrain state, so as to verify whether the identification result that the abnormal vibration information obtained is caused by abnormal terrain is correct.

[0149] By determining the distribution of the marked points in the polarized state and the distribution of the marked points in the non-polarized state and the obvious wind frequency distribution characteristics in the abnormal terrain state, it can be determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain.

[0150] It should be noted that if the total number of abnormal yaw angle areas in the yaw state is greater than the third preset threshold, and the total number of abnormal yaw angle areas in the non-yaw state is greater than the fourth preset threshold, but the distribution of marked points in the polarization state and the distribution of marked points in the non-polarization state do not have obvious wind frequency distribution characteristics under abnormal terrain conditions, it is not considered that the abnormal vibration of the yaw system is caused by abnormal terrain, and the staff needs to use other means to detect the specific cause of the abnormal vibration of the yaw system.

[0151] The abnormal vibration identification method provided by the present invention can distinguish the situation where the vibration of some sector-shaped areas (sectors for short) is too large due to special terrain factors. Under normal circumstances, the abnormal vibration caused by brake pad failure will be distributed in most sectors, while the abnormal vibration caused by special terrain will be distributed in some special sectors, and there will be obvious wind frequency distribution characteristics in a certain wind direction. Based on the above principle, the present invention analyzes the operating data under yaw and non-yaw conditions, and judges under preset conditions that the amplitude of the operating data in the yaw state is abnormal, while the amplitude of the operating data in the non-yaw state is normal, which is closer to the abnormal vibration caused by brake pad failure; when it is judged that the amplitude of the operating data in the non-yaw state is abnormal, and the amplitude of the operating data in the non-yaw state is also abnormal, which is closer to the abnormal vibration caused by abnormal terrain, not only can the abnormal vibration caused by brake pad failure be better located, but also the abnormal vibration caused by special terrain can be effectively distinguished.

[0152] Based on the content of the above embodiment, as an optional embodiment, determining, based on the first vibration amplitude distribution, a first number of vibration amplitudes in each yaw angle region that are greater than a first amplitude threshold and a first total number of vibration amplitudes in each yaw angle region includes:

[0153] Based on the first vibration amplitude distribution, a plurality of vibration intervals are divided into equal intervals according to a first fixed value to form a vibration interval group; the plurality of vibration intervals include an abnormal vibration amplitude interval and a normal vibration amplitude interval, the minimum value of the abnormal vibration amplitude interval is greater than the first amplitude threshold, and the maximum value of the normal vibration amplitude interval is less than or equal to the first amplitude threshold;

[0154] In each yaw angle region, determining the vibration interval to which each vibration amplitude belongs, and using the number of vibration amplitudes in each vibration interval as the weight of the vibration interval;

[0155] determining the first number and a first total number of the vibration amplitudes in each yaw angle region, where the first number is the sum of weights of all abnormal vibration amplitude intervals;

[0156] Accordingly, determining, based on the second vibration amplitude distribution, a second number of vibration amplitudes in each yaw angle region that are greater than a second amplitude threshold and a second total number of vibration amplitudes in each yaw angle region includes:

[0157] Based on the second vibration amplitude distribution, a plurality of vibration intervals are divided into equal intervals according to a second fixed value to form a vibration interval group; the plurality of vibration intervals include an abnormal vibration amplitude interval and a normal vibration amplitude interval, the minimum value of the abnormal vibration amplitude interval is greater than the second amplitude threshold, and the maximum value of the normal vibration amplitude interval is less than or equal to the second amplitude threshold;

[0158] In each yaw angle region, determining the vibration interval to which each vibration amplitude belongs, and using the number of vibration amplitudes in each vibration interval as the weight of the vibration interval;

[0159] determining the second number and a second total number of the vibration amplitudes in each yaw angle region, the second number being the sum of the weights of all abnormal vibration amplitude intervals;

[0160] The first constant value is the same as or different from the second constant value.

[0161] In this embodiment, in order to illustrate the determination of the first total number, for any yaw angle region, assuming that the first constant is a (unit g), the yaw angle region can be divided into the following 9 vibration intervals: 0-a, a-2a, 2a-3a, 3a-4a, 4a-5a, 5a-6a, 6a-7a, 7a-8a, and greater than 8a. The above 9 vibration intervals constitute a vibration interval group.

[0162] For example, if the polarization angle collected at a certain sampling time point is 36° and the vibration amplitude is 0.038g, then the annotation points of the operating data related to the sampling time point can be added in the vibration interval 5a-7a in the yaw angle area corresponding to 30°-40°.

[0163] Based on the above method, the operating data corresponding to each sampling time point in the yaw state data set is sequentially represented in the form of marked points to obtain the first vibration amplitude distribution corresponding to the yaw state data set.

[0164] Based on the first vibration amplitude distribution, the number of marked points in each vibration interval in each yaw angle region (ie, the number of vibration amplitudes in each vibration interval) can be determined and used as the weight corresponding to each vibration interval.

[0165] In this way, the weight corresponding to each vibration interval in each yaw angle region can be determined.

[0166] Furthermore, the present invention may define a vibration amplitude interval whose minimum value is greater than a first amplitude threshold as an abnormal vibration amplitude interval, and define a vibration amplitude interval whose maximum value is less than or equal to the first amplitude threshold as a normal vibration amplitude interval.

[0167] Assuming that the first amplitude threshold is 0.03 (unit: g), the four vibration intervals of 0-a, a-2a, 2a-3a, and 3a-4a among the above nine vibration intervals are normal vibration amplitude intervals, while 4a-5a, 5a-6a, 6a-7a, 7a-8a, and greater than 8a are abnormal vibration amplitude intervals. As for the vibration interval 4a-5a, it can be divided into a normal vibration amplitude interval based on the degree of proximity between the first amplitude threshold of 0.03 and the minimum value and maximum value of the vibration interval.

[0168] After the normal vibration amplitude interval and the abnormal vibration amplitude interval are divided, the sum of the weights of each abnormal vibration amplitude interval in each yaw angle region may be counted as the first quantity.

[0169] Assuming that the weight of the abnormal vibration amplitude interval 5a-6 is 2, the weight of 6a-7a is 4, the weight of 7a-8a is 1, and the weight greater than 8a is 3, the first number is determined to be 10.

[0170] In a similar manner, the sum of the weights of all abnormal vibration amplitude intervals may be calculated based on the second vibration amplitude distribution as the second quantity.

[0171] In addition, the first total number and the second total number of vibration amplitudes in each yaw angle region refer to the total number of marked points distributed in each yaw angle region.

[0172] It should be noted that the step size taken when dividing the first vibration amplitude distribution and the second vibration amplitude distribution into vibration intervals, that is, the first constant and the second constant can be the same or different. When the values ​​are the same, the vibration interval groups obtained by division are the same.

[0173] Based on the content of the above embodiment, as an optional embodiment, determining the abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution includes:

[0174] Based on the first vibration amplitude distribution, the vibration amplitude in each yaw angle region is divided into bins, and a fatigue estimation value corresponding to each bin is calculated;

[0175] Determine the first number of bins in which fatigue estimation values ​​in all bins are greater than a first fatigue threshold and the first total number of bins in each yaw angle region;

[0176] taking the yaw angle regions where the ratio between the first number of bins and the first total number of bins is greater than a fifth preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the yaw state;

[0177] Based on the second vibration amplitude distribution, the vibration amplitude in each yaw angle region is divided into bins, and a fatigue estimation value corresponding to each bin is calculated;

[0178] Determine the number of second bins in which fatigue estimation values ​​in all bins are greater than a second fatigue threshold and the second total number of bins in each yaw angle region;

[0179] taking the yaw angle regions in which the ratio between the second number of bins and the second total number of bins is greater than a sixth preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the non-yaw state;

[0180] If the total number of abnormal yaw angle regions in the yaw state is greater than a seventh preset threshold, and the total number of abnormal yaw angle regions in the non-yaw state is not greater than an eighth preset threshold, determining that the abnormal vibration information is caused by a brake pad failure;

[0181] If the total number of abnormal yaw angle areas in the yaw state is greater than the seventh preset threshold, the total number of abnormal yaw angle areas in the non-yaw state is greater than the eighth preset threshold, and the distribution of the abnormal yaw angle areas matches the wind frequency distribution in the abnormal terrain state, then it is determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain.

[0182] The abnormal vibration identification method provided by the present invention can not only judge the vibration distribution of different sectors in the yaw state and the non-yaw state according to the vibration amplitude, and then count the proportion of different vibration interval amplitudes of each sector in the yaw state and the non-yaw state, but also make judgments based on the cumulative value of vibration fatigue.

[0183] Specifically, based on the vibration amplitude distribution (the first vibration amplitude distribution or the second vibration amplitude distribution), the vibration amplitude in each yaw angle area is divided into bins, and fatigue estimation can be performed according to the vibration amplitude distribution in each bin, and finally, vibration abnormality analysis is performed based on the obtained fatigue estimation value. The fatigue estimation adopted by the present invention can be implemented using commonly used algorithms, and the present invention does not make specific limitations on this.

[0184] It should be noted that when using fatigue estimation values ​​to perform vibration anomaly analysis, the judgment logic provided in the above embodiment can be adopted, that is, abnormal yaw angle areas are screened out from all yaw angle areas based on the fatigue estimation values, and then whether the total number of abnormal yaw angle areas is greater than a preset threshold is used as a judgment condition to determine whether the preset conditions are met in the yaw state or the non-yaw state.

[0185] If the total number of abnormal yaw angle regions in the determined yaw state is greater than a seventh preset threshold, and the total number of abnormal yaw angle regions in the non-yaw state is greater than an eighth preset threshold, it is determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain;

[0186] If the total number of abnormal yaw angle areas in the determined yaw state is greater than the seventh preset threshold, but the total number of abnormal yaw angle areas in the non-yaw state is not greater than the eighth preset threshold, it is determined that the abnormal vibration information is abnormal vibration caused by a brake pad failure.

[0187] As an optional embodiment, when the first fatigue threshold is selected to be the same as the second fatigue threshold, the fifth preset threshold is selected to be the same as the sixth preset threshold, and the seventh preset threshold is selected to be the same as the eighth preset threshold, that is, the same processing method is used for the yaw state data set and the non-yaw state data set, the processing steps and the amount of calculation for identification can be simplified.

[0188] Figure 3 FIG. 1 is a schematic diagram of the structure of the abnormal vibration identification device provided by the present invention. Figure 3 As shown, it mainly includes but is not limited to a data acquisition unit 31, a data cleaning unit 32, a data calibration unit 33 and a vibration analysis unit 34, wherein:

[0189] The data acquisition unit 31 is mainly used to screen the operating data of the target wind turbine within a preset time period and obtain an operating data set; the operating data includes the vibration amplitude and yaw angle at any sampling time point;

[0190] The data cleaning unit 32 is mainly used to divide the operating data set into a yaw state data set and a non-yaw state data set;

[0191] a data calibration unit 33, mainly used to determine, based on the yaw state data set, a first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state, and to determine, based on the non-yaw state data set, a second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state;

[0192] The vibration analysis unit 34 is mainly used to determine abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution.

[0193] It should be noted that the abnormal vibration identification device provided in the embodiment of the present invention can execute the abnormal vibration identification method described in any of the above embodiments during specific operation, which will not be described in detail in this embodiment.

[0194] The abnormal vibration identification device provided by the present invention performs a comprehensive comparative analysis of a first vibration amplitude distribution obtained by analyzing the operating data in the yaw state and a second vibration amplitude distribution obtained by analyzing the operating data in the non-yaw state. It can determine the abnormal vibration information of the wind turbine based on the analysis results, effectively distinguish various causes of abnormal vibration of the yaw system, and more conveniently locate the abnormal vibration caused by the yaw system, so as to facilitate guiding staff to achieve precise management of the wind turbine set.

[0195] Figure 4 Schematic diagram of the structure of the electronic device provided by the present invention, such as Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 may call logic instructions in the memory 430 to execute an abnormal vibration identification method, which includes: screening operating data of a target wind turbine within a preset time period to obtain an operating data set; the operating data includes vibration amplitude and yaw angle at any sampling time point; dividing the operating data set into a yaw state data set and a non-yaw state data set; determining a first vibration amplitude distribution of the target wind turbine within each yaw angle region in the yaw state based on the yaw state data set, and determining a second vibration amplitude distribution of the target wind turbine within each yaw angle region in the non-yaw state based on the non-yaw state data set; and determining abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution.

[0196] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0197] On the other hand, the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the abnormal vibration identification method provided by the above methods, the method including: screening the operating data of the target wind turbine within a preset time period to obtain an operating data set; the operating data includes the vibration amplitude and yaw angle at any sampling time point; dividing the operating data set into a yaw state data set and a non-yaw state data set; based on the yaw state data set, determining the first vibration amplitude distribution of each yaw angle area of ​​the target wind turbine in the yaw state, and based on the non-yaw state data set, determining the second vibration amplitude distribution of the target wind turbine in each yaw angle area in the non-yaw state; based on the first vibration amplitude distribution and the second vibration amplitude distribution, determining the abnormal vibration information of the target wind turbine within the preset time period.

[0198] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the abnormal vibration identification method provided in the above-mentioned embodiments, the method comprising: screening the operating data of the target wind turbine within a preset time period to obtain an operating data set; the operating data comprises the vibration amplitude and yaw angle at any sampling time point; dividing the operating data set into a yaw state data set and a non-yaw state data set; based on the yaw state data set, determining a first vibration amplitude distribution within each yaw angle region of the target wind turbine in the yaw state, and based on the non-yaw state data set, determining a second vibration amplitude distribution within each yaw angle region of the target wind turbine in the non-yaw state; based on the first vibration amplitude distribution and the second vibration amplitude distribution, determining the abnormal vibration information of the target wind turbine within the preset time period.

[0199] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0200] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0201] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for identifying abnormal vibration, characterized in that: include: Filtering the operating data of the target wind turbine within a preset time period to obtain an operating data set; the operating data includes the vibration amplitude and yaw angle at any sampling time point; Dividing the operating data set into a yaw state data set and a non-yaw state data set; determining, based on the yaw state data set, a first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state, and determining, based on the non-yaw state data set, a second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state; Abnormal vibration information of the target wind turbine within the preset time period is determined based on the first vibration amplitude distribution and the second vibration amplitude distribution.

2. The abnormal vibration identification method according to claim 1, characterized in that: The determining, based on the yaw state data set, a first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state, and determining, based on the non-yaw state data set, a second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state, includes: determining an absolute position of a nacelle of the target wind turbine based on the yaw state data set; Dividing a plurality of sector-shaped areas with the absolute position of the nacelle as the center, and taking each sector-shaped area as a yaw angle area; For each vibration amplitude and yaw angle at each sampling time point in the yaw state data set, assigning the vibration amplitude to a target yaw angle region according to the target yaw angle region to which the yaw angle belongs, so as to obtain a first vibration amplitude distribution corresponding to the yaw state data set; re-determining the absolute position of the nacelle of the target wind turbine based on the non-yaw state data set; Dividing a plurality of sector-shaped areas with the new absolute position of the nacelle as the center, and taking each sector-shaped area as a yaw angle area; For the vibration amplitude and yaw angle at each sampling time point in the non-yaw state data set, the vibration amplitude is placed in the target yaw angle region according to the target yaw angle region to which the yaw angle belongs, so as to obtain a second vibration amplitude distribution corresponding to the non-yaw state data set.

3. The abnormal vibration identification method according to claim 2, characterized in that: The step of dividing the plurality of sector-shaped areas with the absolute position of the cabin as the center comprises: dividing the circular area with the absolute position of the cabin as the center by a first preset angle to obtain the plurality of sector-shaped areas; Accordingly, dividing the plurality of sector-shaped areas with the new nacelle absolute position as the center includes: dividing the circular area with the new nacelle absolute position as the center with a second preset angle as a step size to obtain the plurality of sector-shaped areas; the new nacelle absolute position is a nacelle position of the target wind turbine re-determined based on the non-yaw state data set; The first preset angle is the same as or different from the second preset angle.

4. The abnormal vibration identification method according to claim 1, characterized in that: The determining, based on the first vibration amplitude distribution and the second vibration amplitude distribution, abnormal vibration information of the target wind turbine within the preset time period includes: Determining, based on the first vibration amplitude distribution, a first number of vibration amplitudes in each yaw angle region that are greater than a first amplitude threshold and a first total number of vibration amplitudes in each yaw angle region; taking the yaw angle regions for which the ratio between the first number and the first total number is greater than a first preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the yaw state; Determining, based on the second vibration amplitude distribution, a second number of vibration amplitudes in each yaw angle region that are greater than a second amplitude threshold and a second total number of vibration amplitudes in each yaw angle region; taking the yaw angle regions for which the ratio between the second number and the second total number is greater than a second preset threshold as abnormal yaw angle regions in a non-yaw state, to determine the total number of abnormal yaw angle regions in the non-yaw state; If the total number of abnormal yaw angle regions in the yaw state is greater than a third preset threshold, and the total number of abnormal yaw angle regions in the non-yaw state is not greater than a fourth preset threshold, it is determined that the abnormal vibration information is caused by a brake pad failure; If the total number of abnormal yaw angle areas in the yaw state is greater than a third preset threshold, the total number of abnormal yaw angle areas in the non-yaw state is greater than a fourth preset threshold, and the distribution of the abnormal yaw angle areas matches the wind frequency distribution in the abnormal terrain state, then it is determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain.

5. The abnormal vibration identification method according to claim 4, characterized in that: The determining, based on the first vibration amplitude distribution, a first number of vibration amplitudes in each yaw angle region that are greater than a first amplitude threshold and a first total number of vibration amplitudes in each yaw angle region includes: Based on the first vibration amplitude distribution, a plurality of vibration intervals are divided into equal intervals according to a first fixed value to form a vibration interval group; the plurality of vibration intervals include an abnormal vibration amplitude interval and a normal vibration amplitude interval, the minimum value of the abnormal vibration amplitude interval is greater than the first amplitude threshold, and the maximum value of the normal vibration amplitude interval is less than or equal to the first amplitude threshold; In each yaw angle region, determining the vibration interval to which each vibration amplitude belongs, and using the number of vibration amplitudes in each vibration interval as the weight of the vibration interval; determining the first number and a first total number of the vibration amplitudes in each yaw angle region, where the first number is the sum of weights of all abnormal vibration amplitude intervals; Accordingly, determining, based on the second vibration amplitude distribution, a second number of vibration amplitudes in each yaw angle region that are greater than a second amplitude threshold and a second total number of vibration amplitudes in each yaw angle region includes: Based on the second vibration amplitude distribution, a plurality of vibration intervals are divided into equal intervals according to a second fixed value to form a vibration interval group; the plurality of vibration intervals include an abnormal vibration amplitude interval and a normal vibration amplitude interval, the minimum value of the abnormal vibration amplitude interval is greater than the second amplitude threshold, and the maximum value of the normal vibration amplitude interval is less than or equal to the second amplitude threshold; In each yaw angle region, determining the vibration interval to which each vibration amplitude belongs, and using the number of vibration amplitudes in each vibration interval as the weight of the vibration interval; determining the second number and a second total number of the vibration amplitudes in each yaw angle region, the second number being the sum of the weights of all abnormal vibration amplitude intervals; The first constant value is the same as or different from the second constant value.

6. The abnormal vibration identification method according to claim 4, characterized in that: The first amplitude threshold is the same as the second amplitude threshold, the first preset threshold is the same as the second preset threshold, and the third preset threshold is the same as the fourth preset threshold.

7. The abnormal vibration identification method according to claim 1, characterized in that: The determining, based on the first vibration amplitude distribution and the second vibration amplitude distribution, abnormal vibration information of the target wind turbine within the preset time period includes: Based on the first vibration amplitude distribution, the vibration amplitude in each yaw angle region is divided into bins, and a fatigue estimation value corresponding to each bin is calculated; Determine the first number of bins in which fatigue estimation values ​​in all bins are greater than a first fatigue threshold and the first total number of bins in each yaw angle region; taking the yaw angle regions where the ratio between the first number of bins and the first total number of bins is greater than a fifth preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the yaw state; Based on the second vibration amplitude distribution, the vibration amplitude in each yaw angle region is divided into bins, and a fatigue estimation value corresponding to each bin is calculated; Determine the number of second bins in which fatigue estimation values ​​in all bins are greater than a second fatigue threshold and the second total number of bins in each yaw angle region; taking the yaw angle regions in which the ratio between the second number of bins and the second total number of bins is greater than a sixth preset threshold as abnormal yaw angle regions in the yaw state, so as to determine the total number of abnormal yaw angle regions in the non-yaw state; If the total number of abnormal yaw angle regions in the yaw state is greater than a seventh preset threshold, and the total number of abnormal yaw angle regions in the non-yaw state is not greater than an eighth preset threshold, determining that the abnormal vibration information is caused by a brake pad failure; If the total number of abnormal yaw angle areas in the yaw state is greater than the seventh preset threshold, the total number of abnormal yaw angle areas in the non-yaw state is greater than the eighth preset threshold, and the distribution of the abnormal yaw angle areas matches the wind frequency distribution in the abnormal terrain state, then it is determined that the abnormal vibration information is abnormal vibration caused by abnormal terrain.

8. The abnormal vibration identification method according to claim 7, characterized in that: The first fatigue threshold is the same as the second fatigue threshold, the fifth preset threshold is the same as the sixth preset threshold, and the seventh preset threshold is the same as the eighth preset threshold.

9. An abnormal vibration identification device, characterized in that: include: A data acquisition unit is used to screen the operating data of the target wind turbine within a preset time period and obtain an operating data set; the operating data includes the vibration amplitude and yaw angle at any sampling time point; A data cleaning unit, configured to divide the operating data set into a yaw state data set and a non-yaw state data set; a data calibration unit, configured to determine, based on the yaw state data set, a first vibration amplitude distribution of the target wind turbine in each yaw angle region in the yaw state, and to determine, based on the non-yaw state data set, a second vibration amplitude distribution of the target wind turbine in each yaw angle region in the non-yaw state; The vibration analysis unit is configured to determine abnormal vibration information of the target wind turbine within the preset time period based on the first vibration amplitude distribution and the second vibration amplitude distribution.

10. A wind turbine generator system, characterized in that: The device comprises at least one target wind turbine and a vibration processor, and further comprises a memory and a program or instruction stored in the memory and executable on the vibration processor, wherein the program or instruction, when executed by the vibration processor, executes the abnormal vibration identification method according to any one of claims 1 to 8.

Images