Wind turbine bearing fault detection method, device, controller and storage medium

Abstract

This application discloses a method, apparatus, controller, and storage medium for detecting bearing faults in wind turbine generators. The method includes acquiring operating parameters of the wind turbine generator, including nacelle acceleration and the turbine's rotational speed; when the rotational speed is stable, performing high-pass filtering on the nacelle acceleration to obtain a target acceleration within a preset frequency range; extracting the frequency domain characteristics of the target acceleration; and detecting whether a bearing fault has occurred in the wind turbine generator based on the frequency domain characteristics and the rotational speed. Using the wind turbine generator bearing fault detection method, apparatus, controller, and storage medium provided in this application can improve the power generation performance and lifespan of wind turbine generators.

Description

Technical Field

[0001] This application relates to the field of wind power generation technology, specifically to a method, device, controller, and storage medium for detecting bearing faults in wind turbine generators. Background Technology

[0002] As a crucial component of wind turbines, bearing failure can significantly impact the power generation performance and lifespan of the turbine. Therefore, fault detection of wind turbine bearings is of paramount importance.

[0003] Currently, temperature detection is commonly used for bearing fault detection. Specifically, a temperature sensor collects the bearing's temperature, and when the temperature reaches a set threshold, it indicates that the bearing is overheating and has malfunctioned. However, the bearing temperature only shows a significant increase when wear is severe. This means that bearing failure can only be detected after significant wear has occurred. Consequently, wind turbines may operate in a sub-optimal state for extended periods, affecting their power generation performance and lifespan. Summary of the Invention

[0004] The purpose of this application is to provide a method, device, controller, and storage medium for detecting bearing faults in wind turbine generators, in order to solve the technical problem in the prior art that bearing faults can only be detected after the bearings have worn out severely, resulting in wind turbine generators operating in a sub-healthy state for a long time, which in turn affects the power generation performance and lifespan of the wind turbine generators.

[0005] The technical solution of this application is as follows:

[0006] Firstly, a method for detecting bearing faults in wind turbine generators is provided, including:

[0007] The operating parameters of the wind turbine are obtained, including the nacelle acceleration and the rotational speed of the wind turbine.

[0008] When the rotational speed is stable, the nacelle acceleration is high-pass filtered to obtain the target acceleration within a preset frequency range;

[0009] Extract the frequency domain features of the target acceleration;

[0010] Based on the frequency domain characteristics and the rotational speed, the bearings of the wind turbine are detected to be faulty.

[0011] Secondly, a wind turbine bearing fault detection device is provided, comprising:

[0012] The acquisition module is used to acquire the operating parameters of the wind turbine, including the nacelle acceleration and the rotational speed of the wind turbine.

[0013] The filtering module is used to perform high-pass filtering on the nacelle acceleration when the rotational speed is stable, so as to obtain the target acceleration within a preset frequency range;

[0014] An extraction module is used to extract the frequency domain features of the target acceleration;

[0015] The detection module is used to detect whether the bearing of the wind turbine has failed, based on the frequency domain characteristics and the rotational speed.

[0016] Thirdly, a controller is provided, which may include:

[0017] Processor; and memory storing computer program instructions;

[0018] The processor reads and executes the computer program instructions to implement the wind turbine bearing fault detection method as shown in any embodiment of the first aspect.

[0019] Fourthly, a readable storage medium is provided, on which computer program instructions are stored, which, when executed by a processor, implement the wind turbine bearing fault detection method as shown in any embodiment of the first aspect.

[0020] The technical solutions provided by the embodiments of this application bring at least the following beneficial effects:

[0021] This application embodiment acquires operating parameters of the wind turbine, including nacelle acceleration and turbine rotational speed. When the rotational speed is stable, the nacelle acceleration is high-pass filtered to obtain a target acceleration within a preset frequency range. The frequency domain characteristics of the target acceleration are extracted, and then, based on these characteristics and the rotational speed, the bearings of the wind turbine are detected for faults. Since the nacelle acceleration changes significantly when the bearing experiences slight wear or the presence of foreign objects, bearing failures can be detected based on nacelle acceleration, allowing for early maintenance and preventing the wind turbine from operating in a sub-optimal state for extended periods, thus improving its power generation performance and lifespan.

[0022] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0023] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application, and do not constitute an undue limitation of this application.

[0024] Figure 1 This is a spectrum diagram of cabin acceleration provided in an embodiment of this application;

[0025] Figure 2 This is a spectrum diagram of cabin acceleration provided in an embodiment of this application;

[0026] Figure 3 This is a spectrum diagram of cabin acceleration provided in an embodiment of this application;

[0027] Figure 4 This is a spectrum diagram of cabin acceleration provided in an embodiment of this application;

[0028] Figure 5 This is a flowchart illustrating a method for detecting bearing faults in a wind turbine generator provided in an embodiment of this application.

[0029] Figure 6 This is a flowchart illustrating a method for detecting bearing faults in a wind turbine generator provided in an embodiment of this application.

[0030] Figure 7 This is a flowchart illustrating a method for detecting bearing faults in a wind turbine generator provided in an embodiment of this application.

[0031] Figure 8 This is a flowchart illustrating a method for detecting bearing faults in a wind turbine generator provided in an embodiment of this application.

[0032] Figure 9 This is a schematic diagram of the structure of a wind turbine bearing fault detection device provided in an embodiment of this application;

[0033] Figure 10 This is a schematic diagram of the structure of a controller provided in an embodiment of this application. Detailed Implementation

[0034] To enable those skilled in the art to better understand the technical solutions of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings.

[0035] It should be noted that the terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0036] As is known from the background technology, the existing technology of using temperature detection method for bearing fault detection can only detect bearing failure after the bearing has been severely worn. This results in the wind turbine unit operating in a sub-healthy state for a long time, which in turn affects the power generation performance and lifespan of the wind turbine unit.

[0037] Furthermore, under normal operating conditions, there is no impact between the inner ring of the bearing and the rolling elements of the wind turbine. When the bearing of the wind turbine fails, such as due to cracks or foreign objects causing uneven friction, periodic impacts will occur between the inner ring of the bearing and the rolling elements, which will be reflected as characteristic frequencies and harmonics in the nacelle acceleration spectrum.

[0038] Specifically, such as Figure 1 As shown, under normal operating conditions, the nacelle acceleration spectrum of a wind turbine typically only contains low-frequency components generated by the rotor rotation frequency. Figure 2 , Figure 3 , Figure 4 As shown, when the bearings of a wind turbine exhibit failure characteristics such as wear, foreign objects, or cracks, resulting in bearing failure, an impact will occur between the inner ring of the bearing and the rolling elements, causing characteristic frequencies and harmonics to appear in the nacelle acceleration spectrum.

[0039] Based on the above findings, this application provides a method, device, controller, and storage medium for detecting bearing faults in wind turbines. This method acquires operating parameters of the wind turbine, including nacelle acceleration and turbine rotational speed. When the rotational speed is stable, the nacelle acceleration is high-pass filtered to obtain a target acceleration within a preset frequency range. The frequency domain characteristics of the target acceleration are extracted, and then, based on these characteristics and the rotational speed, the bearing fault of the wind turbine is detected. Since the nacelle acceleration changes significantly when the bearing experiences slight wear or the presence of foreign objects, bearing faults can be detected based on nacelle acceleration, allowing for early maintenance and preventing the wind turbine from operating in a sub-optimal state for extended periods, thus improving the power generation performance and lifespan of the wind turbine.

[0040] The following is a description of a wind turbine bearing fault detection method provided in the embodiments of this application.

[0041] Figure 5 This illustration shows a flowchart of a wind turbine bearing fault detection method provided in an embodiment of this application. The execution subject of this method can be the controller of the wind turbine, or it can be an industrial control computer of the wind farm, a cloud server, or other equipment. Figure 5 As shown, the wind turbine bearing fault detection method may include the following steps:

[0042] S510, obtains the operating parameters of the wind turbine.

[0043] The operating parameters may include nacelle acceleration and wind turbine rotation speed.

[0044] During wind turbine operation, vibration sensors installed inside the nacelle to detect overall turbine vibration, or vibration sensors in the Content Management System (CMS) mounted on the turbine's bearings, collect the nacelle acceleration; the turbine's rotational speed is collected by the turbine's speed sensor. Then, the wind turbine's operating parameters can be obtained, such as periodically acquiring these parameters, which may include the aforementioned collected nacelle acceleration and rotational speed.

[0045] The S520, when the engine speed is stable, performs high-pass filtering on the nacelle acceleration to obtain the target acceleration within a preset frequency range.

[0046] The preset frequency range can be a pre-set frequency range that is allowed to pass through the high-pass filter. This frequency range can be set to a relatively high frequency range so that the high-pass filter can filter out low-frequency signals generated by components such as impellers.

[0047] The target acceleration can be the cabin acceleration that falls within a preset frequency range.

[0048] After obtaining the operating parameters of the wind turbine, including nacelle acceleration and turbine speed, it is possible to determine whether the turbine speed is in a stable state. For example, the magnitude of the speed variation can be used to determine whether the turbine speed is stable. How to determine whether the wind turbine speed is in a stable state will be explained in detail below and will not be repeated here.

[0049] When the wind turbine's rotational speed is stable, high-pass filtering is applied to the nacelle acceleration to filter out accelerations outside the preset frequency range. This can be achieved by blocking or weakening nacelle accelerations outside the preset frequency range, thus obtaining the target accelerations within the preset frequency range. High-pass filtering can be performed using Butterworth digital filters, Chebyshev filters, or Bezier filters. By retaining only the target nacelle accelerations within the preset frequency range, low-frequency noise in the nacelle acceleration can be eliminated, reducing the impact of low-frequency signals on bearing fault detection.

[0050] S530 extracts the frequency domain features of the target acceleration.

[0051] Frequency domain characteristics are a coordinate system used to describe the frequency characteristics of a signal. For cabin acceleration, frequency domain characteristics can typically include the frequency and amplitude of cabin acceleration.

[0052] After obtaining the target acceleration within a preset frequency range, the frequency domain features of the target acceleration can be extracted. For example, a fast Fourier transform (FFT) can be performed on the target acceleration to extract its frequency domain features.

[0053] The S540 detects whether the bearings of a wind turbine are faulty based on frequency domain characteristics and rotational speed.

[0054] After extracting the frequency domain characteristics of the target acceleration, the bearings of the wind turbine can be detected based on the frequency domain characteristics of the nacelle acceleration and the rotational speed of the preceding wind turbine. Thus, by utilizing the characteristic that the frequency domain characteristics of nacelle acceleration show significant changes when the bearing experiences minor damage or the presence of foreign objects, bearing faults can be identified at an early stage, allowing for timely maintenance. Furthermore, improvements can be made to lubrication and operating strategies. After these improvements, the health of the bearings can be continuously monitored, lubrication checks can be performed regularly, spare parts can be prepared in advance, replacement plans can be developed, unplanned downtime can be reduced, and prolonged unplanned downtime caused by bearing seizure can be avoided.

[0055] It is understood that in the wind turbine bearing fault detection method provided in this application embodiment, after a bearing fault is detected, a warning signal can be output so that personnel can receive the warning signal. The warning signal may include at least one of bearing fault indication information and maintenance method information. This not only allows for early detection of bearing faults in the wind turbine, but also enables the issuance of a warning signal when a bearing fault is detected, allowing personnel to receive the warning signal and perform maintenance on the wind turbine bearing based on the warning signal. This further improves the reliability and availability of the wind turbine.

[0056] This application embodiment acquires operating parameters of the wind turbine, including nacelle acceleration and turbine rotational speed. When the rotational speed is stable, the nacelle acceleration is high-pass filtered to obtain a target acceleration within a preset frequency range. The frequency domain characteristics of the target acceleration are extracted, and then, based on these characteristics and the rotational speed, the bearings of the wind turbine are detected for faults. Since the nacelle acceleration changes significantly when the bearing experiences slight wear or the presence of foreign objects, bearing failures can be detected based on nacelle acceleration, allowing for early maintenance and preventing the wind turbine from operating in a sub-optimal state for extended periods, thus improving its power generation performance and lifespan.

[0057] Furthermore, in this embodiment, vibration sensors installed inside the nacelle of the wind turbine for detecting overall turbine vibration, or vibration sensors from a CMS mounted on the wind turbine bearings, are used to collect the nacelle acceleration of the wind turbine, thereby enabling wind turbine bearing fault detection based on the nacelle acceleration. Thus, on the one hand, no additional equipment needs to be installed; utilizing existing equipment to collect nacelle acceleration reduces the cost of the wind turbine bearing fault detection method provided in this embodiment. On the other hand, since existing vibration sensors installed inside the nacelle or vibration sensors from the CMS are generally highly accurate, the accuracy of the collected nacelle acceleration is also high, thereby improving the accuracy of the wind turbine bearing fault detection method provided in this embodiment.

[0058] In some embodiments, the frequency domain features may include amplitude and the frequency corresponding to the amplitude. In this case, the bearing failure of the wind turbine can be detected based on the amplitude and frequency in the frequency domain features, the characteristic frequency determined based on the rotational speed, and the harmonics. Accordingly, the specific implementation of the above step S540 can be as follows:

[0059] Among the amplitude values ​​of the frequency domain features, the M largest first target amplitude values ​​are obtained, where M is a positive integer;

[0060] For M first target amplitudes, determine the first target frequency corresponding to each first target amplitude to obtain M first target frequencies;

[0061] Calculate the bearing's characteristic frequency and harmonic frequency n based on the rotational speed, where n is a positive integer;

[0062] Based on M first target frequencies, characteristic frequencies, and harmonics, the bearings of wind turbine units are detected to determine whether a fault has occurred.

[0063] Given that the frequency domain characteristics of nacelle acceleration include both amplitude and frequency, we can first determine the M largest amplitudes among all amplitude values ​​in the frequency domain characteristics, i.e., the M first target amplitudes, where M can be a positive integer, such as 2. For the aforementioned M first target amplitudes, we can determine the frequency corresponding to each first target amplitude among all frequencies in the frequency domain characteristics, i.e., the first target frequency, thus obtaining the M first target frequencies. Then, we can calculate the characteristic frequency and harmonics of the bearing based on the wind turbine's rotational speed. Finally, based on the M first target frequencies and the aforementioned characteristic frequency and harmonics, we can detect whether the wind turbine's bearing has malfunctioned.

[0064] As a concrete example, the characteristic frequency X can be calculated based on formulas (1) and (2):

[0065] X = co inne *freq gs (1)

[0066]

[0067] Among them, co inne This refers to the characteristic frequency coefficient of the bearing inner ring. This characteristic frequency coefficient can be the bearing inner ring coefficient, bearing outer ring coefficient, rolling system number, cage coefficient, etc. gs For frequency conversion, freq gs The value is the average speed under steady-state conditions / 60. n is the harmonic, n = 1, 2, 3, ..., k. The value of k is closely related to the characteristic frequency coefficient of the bearing. n*X is less than or equal to the sampling frequency of the nacelle acceleration / 2.

[0068] Taking M=2 as an example, the specific implementation of determining the M first target frequencies can be as follows: First, from all amplitudes of the cabin acceleration, find the first target amplitude with the largest amplitude, denoted as amp1, and the first target amplitude with the largest amplitude other than amp1, denoted as amp2. Then determine the first target frequencies corresponding to each amp1 and amp2, denoted as freq1 and freq2 respectively. It should be noted that, in order to avoid interference from adjacent frequency signals to the extraction of freqa (a=1,2), before extracting freq2, the amplitudes in the left and right neighborhoods of freq1 can be set to 0. A typical value for the preset number is 10.

[0069] Understandably, the frequencies corresponding to the first two amplitudes in the non-frequency domain features of freq1 and freq2, arranged strictly from largest to smallest, are considered. Specifically, interference from neighboring frequencies needs to be taken into account. For example, if amp1 = 0.8g at freq1 = 0.51Hz is the largest amplitude in the frequency domain feature, and the amplitude of 0.75g at 0.52Hz is the second largest amplitude after amp1, but 0.52Hz and freq1 are in a neighborhood, the frequency at 0.52Hz is not considered as freq2. Based on the above considerations, a method of setting the frequency domain amplitude within the neighborhood to 0 is adopted when extracting the first target amplitude.

[0070] Thus, when a wind turbine's bearings exhibit failure characteristics such as wear, foreign objects, or cracks, characteristic frequencies and harmonics will appear in the frequency domain characteristics of the nacelle's acceleration. Therefore, by using the frequency corresponding to the maximum amplitude value, and the characteristic frequencies and harmonics determined based on the wind turbine's rotational speed, the accuracy of the bearing failure detection can be further improved.

[0071] In some embodiments, if there is a first target frequency corresponding to each first target amplitude that is greater than or equal to a preset amplitude threshold among the M first target amplitudes, then the M first target frequencies can be obtained.

[0072] After determining M first target amplitudes, these M first target amplitudes can be compared with a preset amplitude threshold to determine if any of the M first target amplitudes are greater than or equal to the preset amplitude threshold. For example, the preset amplitude threshold could be 0.02, and its specific value can be set according to actual needs. If any of the M first target amplitudes are greater than or equal to the preset amplitude threshold, then the first target frequency corresponding to each first target amplitude is determined, resulting in M ​​first target frequencies. Subsequent steps are then performed to determine the bearing's characteristic frequency and harmonics based on the rotational speed, and to detect whether the wind turbine's bearing has failed based on the M first target frequencies, characteristic frequencies, and harmonics. If none of the M first target amplitudes are greater than or equal to the preset amplitude threshold (i.e., any one of the M first target amplitudes is less than the preset amplitude threshold), then the subsequent steps to determine whether the wind turbine's bearing has failed are not performed.

[0073] In this way, since the probability of a wind turbine bearing failure is very small when all M first target amplitudes are less than the preset amplitude threshold, the detection of a wind turbine bearing failure can be performed only when one of the M first target amplitudes is greater than or equal to the preset amplitude threshold. This improves the success rate and efficiency of the wind turbine bearing failure detection method and reduces unnecessary computation and resource consumption to some extent.

[0074] In some embodiments, the frequency corresponding to the maximum amplitude of the nacelle acceleration in two directions can be used to determine whether a wind turbine has malfunctioned. Accordingly, the specific implementation can be as follows:

[0075] The cabin acceleration may include a spectrum of the first cabin acceleration in a first direction and a spectrum of the second cabin acceleration in a second direction.

[0076] The first direction can be perpendicular to the second direction. The first direction and the second direction can be two mutually perpendicular directions in the coordinate system. For example, the first direction can be the x-direction and the second direction can be the y-direction.

[0077] A spectrum diagram can be used to represent the relationship between the frequency and amplitude of nacelle acceleration.

[0078] At this point, the specific implementation method for detecting whether the bearing of a wind turbine has failed, based on M first target frequencies, characteristic frequencies, and harmonics, can be as follows:

[0079] Calculate the first nth harmonic based on the characteristic frequency and harmonics.

[0080] Based on the first nth harmonic, determine the first target nth harmonic corresponding to the maximum amplitude within the first frequency interval of the spectrum diagram of the first cabin acceleration, and the second target nth harmonic corresponding to the maximum amplitude within the second frequency interval of the spectrum diagram of the second cabin acceleration;

[0081] If, among the M first target frequencies, there exists a first target frequency that is the same as the nth harmonic of the first target frequency or the nth harmonic of the second target frequency, it is determined that the bearing of the wind turbine unit has failed.

[0082] The first frequency range is the frequency range whose frequency deviation from the first nth harmonic (nX) is within a first deviation range, i.e., the frequency range whose frequency deviation from nX is within the first deviation range. The second frequency range is the frequency range whose frequency deviation from the first nth harmonic is within a second deviation range, i.e., the frequency range whose frequency deviation from nX is within the second deviation range. The first deviation range may be the same as or different from the second deviation range.

[0083] The first nth harmonic is the nth harmonic calculated based on the characteristic frequency X and harmonic n calculated above.

[0084] The nacelle acceleration can include a first nacelle acceleration in a first direction and a second nacelle acceleration in a second direction. At this point, the bearing n-harmonic frequency nX, i.e., the first n-harmonic frequency, can be calculated based on the aforementioned characteristic frequency X and harmonic frequency n. The maximum amplitude within the first frequency range is determined in the spectrum of the first nacelle acceleration, and then the frequency corresponding to this maximum amplitude is determined, i.e., the first target n-harmonic frequency, also known as the n-harmonic frequency in the first direction. The aforementioned formulas for calculating the first target n-harmonic frequency can be as shown in formulas (3) and (4).

[0085] nX amp =max(amp[nX-ε,nX+ε]) (3)

[0086] nX-ε <freq max <nX+ε (4)

[0087] Wherein nX amp The maximum amplitude within a preset frequency range around frequency nX in the spectrum of the acceleration of the first cabin; ε represents the aforementioned preset frequency range; freq max This represents the nth harmonic of the first target.

[0088] Similarly, the maximum amplitude within the second frequency range can be determined from the acceleration spectrum of the second cabin, and then the frequency corresponding to this maximum amplitude can be determined, i.e., the second target n-harmonic, also known as the n-harmonic of the second direction. The method for determining the second target n-harmonic is similar to the method for determining the first target n-harmonic, and will not be elaborated here.

[0089] After determining the first target n-harmonic frequency and the second target n-harmonic frequency, it can be determined whether there is a first target frequency among the M first target frequencies that is the same as the first target n-harmonic frequency or the second target n-harmonic frequency. That is, it can be determined whether there is an n-harmonic frequency in the first direction or the second direction among the first target frequencies. If there is a first target frequency among the M first target frequencies that is the same as the first target n-harmonic frequency or the second target n-harmonic frequency, it is determined that the wind turbine bearing has failed. In other words, if at least one of the first target n-harmonic frequency or the second target n-harmonic frequency exists among the M first target frequencies, then the wind turbine bearing is considered to have failed.

[0090] In this way, by determining the first target n-harmonic frequency and the second target n-harmonic frequency based on the first n-harmonic frequency calculated from the characteristic frequency and the harmonic frequency, the determined first target n-harmonic frequency and the second target n-harmonic frequency can be more accurate. Then, based on the first target n-harmonic frequency, the second target n-harmonic frequency, and M first target frequencies, it can be determined whether the bearing of the wind turbine has failed, thereby further improving the accuracy of the wind turbine bearing detection results.

[0091] In some embodiments, the bearing failure of the wind turbine can be detected when there are no interfering frequencies in both the third and fourth frequency ranges. Accordingly, before determining that the bearing failure of the wind turbine has occurred when there is a first target frequency among the M first target frequencies that is the same as the nth harmonic of the first target or the nth harmonic of the second target, the following steps can also be performed:

[0092] Determine whether the first interference frequency exists in the third and fourth frequency intervals respectively.

[0093] The third frequency range is the frequency range in which the frequency deviation from the first target (n times the frequency) is within the third deviation range, and the fourth frequency range is the frequency range in which the frequency deviation from the second target (n times the frequency) is within the fourth deviation range. The third deviation range and the fourth deviation range can be the same or different.

[0094] In this case, if among the M first target frequencies there exists a first target frequency that is the same as the nth harmonic of the first target frequency or the nth harmonic of the second target frequency, the specific implementation method for determining that the bearing of the wind turbine has failed can be as follows:

[0095] If there is no first interference frequency in either the third or fourth frequency range, and if among the M first target frequencies there is a first target frequency that is the same as the nth harmonic of the first target or the nth harmonic of the second target, then the bearing of the wind turbine unit is determined to have failed.

[0096] A third frequency interval, defined as the third deviation range, and a fourth frequency interval, defined as the fourth deviation range, can be determined, with the deviation range from the first target's nth harmonic being the third deviation range. Then, it can be determined whether an interference frequency, i.e., a first interference frequency, exists within the third and fourth frequency intervals. If no first interference frequency exists within either the third or fourth frequency interval, it is determined whether any of the M first target frequencies is the same as either the first or second target's nth harmonic. If no first interference frequency exists within either the third or fourth frequency interval, and a first target frequency is present among the M first target frequencies that is the same as either the first or second target's nth harmonic, then the wind turbine's bearing has failed. If a first interference frequency exists within either the third or fourth frequency interval, the step of detecting whether the wind turbine's bearing has failed is not performed.

[0097] The specific implementation method for determining whether a first interference frequency exists in the third frequency interval can be as follows: First, identify whether a first interference frequency exists in the frequency interval to the left of the first target n-fold frequency in the third frequency interval. The frequency interval to the left of the first target n-fold frequency can be a frequency interval whose frequency value is less than the first target n-fold frequency and whose difference from the first target n-fold frequency is less than or equal to the third deviation range. The first interference frequency can be nX... amp / co, where co is a constant, such as 3. Then, the maximum amplitude value in the frequency interval to the left of the first target n-fold frequency in this third frequency interval can be determined, denoted as left. amp If left amp <nX amp If / co, it is assumed that the first interference frequency does not exist in the frequency interval to the left of the first target's nth harmonic in the third frequency interval. Similarly, the specific implementation methods for determining whether the first interference frequency exists in the frequency interval to the right of the first target's nth harmonic in the third frequency interval, and for determining whether the first interference frequency exists in the frequency intervals to the left and right of the second target's nth harmonic in the fourth frequency interval, are similar to the specific implementation methods for determining whether the first interference frequency exists in the frequency interval to the left of the first target's nth harmonic in the third frequency interval, and will not be elaborated here.

[0098] It should be noted that interference on both sides of the harmonic can also be judged by the area under the curve in the vicinity of the nth harmonic or by the amplitude of each frequency component in the vicinity. The principle of this method is similar to the method of judging whether the first interference frequency is in the third and fourth frequency intervals, and will not be elaborated here.

[0099] Therefore, considering that the presence of a first interference frequency in the third or fourth frequency range might affect the accuracy of wind turbine bearing fault detection results, a bearing fault in the wind turbine is determined only if the first interference frequency is absent in both the third and fourth frequency ranges, and if among the M first target frequencies, there is a first target frequency that is the same as the nth harmonic of either the first or second target frequency. This further improves the accuracy of wind turbine bearing fault detection results.

[0100] In some embodiments, the frequency domain features may include a spectrum of the first cabin acceleration in a first direction and a spectrum of the second cabin acceleration in a second direction, wherein the first direction is perpendicular to the second direction. The specific implementation of step S540 above can be as follows:

[0101] Calculate the bearing's characteristic frequency and harmonic frequency n based on the rotational speed, where n is a positive integer;

[0102] Based on the characteristic frequency and harmonics, the third target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the first cabin acceleration, and the fourth target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the second cabin acceleration are determined.

[0103] Based on n third target n-fold frequencies and n fourth target n-fold frequencies, detect whether the bearings of the wind turbine unit have failed.

[0104] First, the characteristic frequency and harmonics of the bearing can be calculated based on the rotational speed. After determining the characteristic frequency and harmonics of the bearing, the third target n-harmonic corresponding to the maximum amplitude of each of the n harmonics in the spectrum of the first nacelle acceleration can be determined based on the characteristic frequency and harmonics, thus obtaining n third target n-harmonics. Similarly, the fourth target n-harmonic corresponding to the maximum amplitude of each of the n harmonics in the spectrum of the second nacelle acceleration can be determined, thus obtaining n fourth target n-harmonics. The bearings of the wind turbine can then be detected for faults based on these n third target n-harmonics and n fourth target n-harmonics. It is understood that the specific implementation of determining the third target n-harmonic corresponding to the maximum amplitude of each of the n harmonics in the spectrum of the first nacelle acceleration, and the fourth target n-harmonic corresponding to the maximum amplitude of each of the n harmonics in the spectrum of the second nacelle acceleration, is similar to the specific implementation of determining the first target n-harmonic corresponding to the maximum amplitude in the spectrum of the first nacelle acceleration, and the second target n-harmonic corresponding to the maximum amplitude in the spectrum of the second nacelle acceleration, as described in the above embodiment, and will not be repeated here.

[0105] It should be noted that in this embodiment, the frequency corresponding to the maximum amplitude value at each harmonic is determined. For example, assuming n is 6, the frequency corresponding to the maximum amplitude value at harmonic 1, the frequency corresponding to the maximum amplitude value at harmonic 2, ..., the frequency corresponding to the maximum amplitude value at harmonic 6 is determined.

[0106] In this way, based on the maximum amplitude of the first and second nacelle accelerations at each harmonic, corresponding to the n third target n harmonics and the n fourth target n harmonics, the frequency characteristics at different harmonics are comprehensively considered, thereby further improving the accuracy of wind turbine bearing testing results.

[0107] In some embodiments, the specific implementation of detecting whether the bearing of a wind turbine has failed based on n third target n-harmonic frequencies and n fourth target n-harmonic frequencies can be as follows:

[0108] Determine whether a second interference frequency exists in the i-th fifth frequency interval and the i-th sixth frequency interval, respectively.

[0109] Where i is a positive integer less than or equal to n. The i-th fifth frequency interval is the frequency interval whose frequency deviation from the n-th harmonic of the i-th third target is within the fifth deviation range, and the i-th sixth frequency interval is the frequency interval whose frequency deviation from the n-th harmonic of the i-th fourth target is within the sixth deviation range.

[0110] If a second interference frequency exists in the i-th fifth frequency interval, remove the i-th third target n-fold frequency from the n third target n-fold frequencies. If a second interference frequency exists in the i-th sixth frequency interval, remove the i-th fourth target n-fold frequency from the n fourth target n-fold frequencies to obtain the first frequency set.

[0111] Based on the first frequency set, detect whether the bearings of the wind turbine unit have failed.

[0112] After obtaining n third target n-fold frequencies and n fourth target n-fold frequencies, it can be determined whether there is an interfering frequency, i.e., whether there is a second interfering frequency, in the i-th fifth frequency interval and the i-th sixth frequency interval. This second interfering frequency can be the same as or different from the first interfering frequency. It should be noted that the aforementioned determination of whether there is a second interfering frequency needs to be performed for each third target n-fold frequency and each fourth target n-fold frequency until the existence of a second interfering frequency in each fifth and sixth frequency interval is determined. Furthermore, the specific implementation method for determining whether there is a second interfering frequency in the i-th fifth and i-th sixth frequency intervals is similar to the implementation method for determining whether there is a first interfering frequency in the third and fourth frequency intervals, and will not be repeated here.

[0113] If a second interference frequency exists within the i-th fifth frequency interval, the i-th third target n-fold frequency can be removed from the n third target n-fold frequencies. Similarly, if a second interference frequency exists within the i-th sixth frequency interval, the i-th fourth target n-fold frequency can be removed from the n fourth target n-fold frequencies. A set is formed from all the remaining third and fourth target n-fold frequencies, resulting in a first frequency set. This first frequency set is then used to detect whether the wind turbine's bearings have malfunctioned.

[0114] Therefore, considering that the presence of a second interference frequency within any of the fifth or sixth frequency intervals might affect the accuracy of wind turbine bearing fault detection results, a first frequency set is formed solely from the third and fourth target n-harmonic frequencies, which do not contain a second interference frequency. This first frequency set is used to determine whether a wind turbine bearing has failed. This approach further improves the accuracy of wind turbine bearing fault detection results.

[0115] In some embodiments, the specific implementation of detecting whether the bearing of a wind turbine has failed based on a first frequency set can be as follows:

[0116] Obtain the preset threshold and the amplitude corresponding to each frequency in the first frequency set;

[0117] Determine a second target amplitude in the first frequency set whose amplitude is less than a preset threshold;

[0118] Remove the second target frequency corresponding to the second target amplitude from the first frequency set to obtain the second frequency set;

[0119] The second frequency set is deduplicated to obtain the third frequency set.

[0120] Based on the third frequency set, detect whether the bearings of the wind turbine unit have failed.

[0121] After obtaining the first frequency set, a preset threshold c can be acquired. c represents a pre-defined minimum allowable amplitude value, which can be set according to actual conditions. The amplitude corresponding to each frequency in the first frequency set can also be obtained. Then, from all amplitude values ​​corresponding to each frequency in the first frequency set, amplitudes smaller than c can be selected as the second target amplitude. All frequencies corresponding to the second target amplitude are removed from the first frequency set to obtain the second frequency set. Next, the second frequency set can be deduplicated to obtain a third frequency set. Based on this third frequency set, the bearings of the wind turbine can be checked for faults.

[0122] In this way, by using a third set of frequencies that have been processed after removing frequencies with amplitudes less than a preset threshold and performing past repetition, the detection of whether the bearings of wind turbines have failed can reduce unnecessary computation to a certain extent, thereby improving the efficiency of the wind turbine bearing failure detection method.

[0123] In some embodiments, the specific implementation of detecting whether the bearing of a wind turbine has failed based on a third frequency set can be as follows:

[0124] Obtain a preset set of ratios and any two frequencies from the third set of frequencies;

[0125] Calculate the ratio of any two frequencies;

[0126] If a target ratio belonging to a preset ratio set exists among all ratios, it is determined that the bearing of the wind turbine unit has failed.

[0127] The preset ratio set can be a set of ratios pre-set based on historical data or experience, denoted as ob. list .

[0128] When detecting whether a wind turbine bearing has failed based on a first frequency set, a preset ratio set and any two frequencies from a third frequency set can be obtained. The ratio of these two frequencies is calculated to obtain the ratio of any two frequencies in the third frequency set. Then, it can be determined whether any ratio, i.e., the target ratio, exists among all the ratios calculated based on the frequencies in the third frequency set. If the target ratio, belonging to the preset ratio set, exists among all the ratios calculated based on the frequencies in the third frequency set, the wind turbine bearing can be considered to have failed.

[0129] It should be noted that when calculating the ratio of any two frequencies in the third frequency set, the frequency with the larger value should be compared to the frequency with the smaller value. Furthermore, the elements in the preset ratio set are a series of elements greater than 1 with specific meanings; these elements can indicate that at least two harmonics exist in the bearing's nacelle acceleration signal. The method provided in this embodiment is executed when the number of elements in the third frequency set is greater than 1.

[0130] Thus, if any ratio between any two frequencies in the third frequency set belongs to a preset ratio set, it indicates that at least two harmonics exist in the third frequency set, suggesting a bearing failure in the wind turbine. This further improves the accuracy of the wind turbine bearing failure detection method.

[0131] In some embodiments, the AC component of the cabin acceleration after high-pass filtering can be determined as the target acceleration. Accordingly, the specific implementation method is as follows:

[0132] Remove the DC component of the nacelle acceleration to obtain the AC component of the nacelle acceleration;

[0133] When the rotational speed is stable, the AC component of the nacelle acceleration is high-pass filtered to obtain the target AC component of the nacelle acceleration that belongs to the preset frequency range.

[0134] The target AC component is defined as the target acceleration.

[0135] The cabin acceleration signal usually contains DC and AC components. It is necessary to extract the AC component parasitic on the DC component, that is, to remove the DC signal from the cabin acceleration. Specifically, the mean value of the cabin acceleration can be calculated first, and the DC component removal is completed by subtracting the mean value from the cabin acceleration at each sampling time, as shown in formulas (5) and (6).

[0136] s(t)=x(t)-x0 (5)

[0137]

[0138] Where s(t) is the AC component of cabin acceleration; x(t) is cabin acceleration; t represents the sampling time; x0 is the mean cabin acceleration; and T is the sampling period.

[0139] After removing the DC component of the nacelle acceleration to obtain the AC component, a high-pass filter can be applied to this AC component to obtain the AC component within a preset frequency range, i.e., the target AC component, which can then be identified as the target acceleration. The specific implementation principle of high-pass filtering for the AC component of the nacelle acceleration is similar to that of high-pass filtering for the nacelle acceleration described above, and will not be repeated here for the sake of simplicity.

[0140] In some embodiments, the stability of the rotational speed can be determined based on the rotational speed fluctuation coefficient. The corresponding processing can be as follows:

[0141] Calculate the fluctuation coefficient of rotational speed;

[0142] If the fluctuation coefficient meets the preset fluctuation conditions, the rotational speed is determined to be in a stable state.

[0143] Among them, the fluctuation coefficient can be used to indicate the range of change in rotational speed, such as the difference between the standard deviation, coefficient of variation, extreme value, heatstroke, and median.

[0144] The system can calculate a fluctuation coefficient to indicate the magnitude of changes in rotational speed and determine whether this coefficient meets preset fluctuation conditions. If the fluctuation coefficient meets the preset conditions, the rotational speed is determined to be in a stable state. Conversely, if the fluctuation coefficient does not meet the preset conditions, the rotational speed is determined to be in an unstable state.

[0145] As a specific example, when the volatility coefficient is the standard deviation, the standard deviation can be calculated according to formula (7).

[0146] gs_std = max(gs) - min(gs) (7)

[0147] Where gs represents rotational speed and gs_std represents standard deviation.

[0148] When the fluctuation coefficient is the standard deviation, the preset fluctuation condition can be that the standard deviation is less than the preset standard deviation threshold 'a', such as 'a' being 0.1. When the standard deviation is less than 'a', the rotational speed can be considered to be in a stable state.

[0149] When the volatility coefficient is the coefficient of variation, the coefficient of variation can be calculated according to formula (8).

[0150]

[0151] Among them, C v σ represents the coefficient of variation, μ represents the average rotational speed, and σ represents the standard deviation of rotational speed.

[0152] When the fluctuation coefficient is the coefficient of variation, the preset fluctuation condition can be that the coefficient of variation is less than a preset coefficient of variation threshold. When the coefficient of variation is less than the preset coefficient of variation threshold, the rotational speed can be considered to be in a stable state.

[0153] The following is combined with Figure 6 This application describes a method for detecting bearing faults in wind turbine generators, as provided in an embodiment. Figure 6 As shown in the embodiments of this application, the wind turbine bearing fault detection method can be implemented through the following steps:

[0154] S610, obtains the operating parameters of the wind turbine.

[0155] The operating parameters include nacelle acceleration and wind turbine rotation speed.

[0156] S620 calculates the rotational speed fluctuation coefficient.

[0157] S630, is the speed stable?

[0158] If the rotational speed is stable, proceed with steps S640-S690; otherwise, terminate the process.

[0159] S640 calculates the bearing's characteristic frequency and harmonics based on the rotational speed.

[0160] S650 removes the DC component of nacelle acceleration.

[0161] The S660 performs a high-pass filter on the AC component of the cabin acceleration to obtain the target acceleration.

[0162] The S670 extracts the frequency domain features of the target acceleration using FFT.

[0163] The S680 detects whether the bearings of a wind turbine are faulty based on frequency domain characteristics and rotational speed.

[0164] S690 outputs the detection results.

[0165] The specific implementation principles and technical effects of each of the above steps are similar to the wind turbine bearing fault detection methods provided in the above method embodiments, and will not be repeated here for the sake of brevity.

[0166] Figure 7 This application illustrates a method for detecting bearing failures in wind turbines, based on M first target frequencies, characteristic frequencies, and harmonics. The specific implementation process of this method for detecting whether a bearing in a wind turbine has failed is shown below. Figure 7 As shown, the method may include the following steps:

[0167] S710, determine M first target amplitudes.

[0168] S720, determine whether there is a first target amplitude that is greater than or equal to a preset amplitude threshold among the M first target amplitudes.

[0169] If any of the M first target amplitudes is greater than or equal to a preset amplitude threshold, proceed to step S730. Otherwise, proceed to step S770 to end the process.

[0170] S730, for M first target amplitudes, determine the first target frequency corresponding to each first target amplitude to obtain M first target frequencies. Then, based on the characteristic frequency and harmonics, calculate the first nth harmonic, determine the first target nth harmonic corresponding to the maximum amplitude within the first frequency interval of the spectrum diagram of the first cabin acceleration, and the second target nth harmonic corresponding to the maximum amplitude within the second frequency interval of the spectrum diagram of the second cabin acceleration.

[0171] S740 determines whether the first interference frequency exists in the third and fourth frequency ranges, respectively.

[0172] If the first interference frequency is not present in either the third or fourth frequency range, proceed to step S750. Otherwise, proceed to step S770 to end the process.

[0173] S750, determine whether there exists a first target frequency among M first target frequencies that is the same as the nth harmonic of the first target or the nth harmonic of the second target.

[0174] If, among the M first target frequencies, there exists a first target frequency that is the same as an nth harmonic of the first target frequency or an nth harmonic of the second target frequency, then proceed to step S760. Otherwise, proceed to step S770 to end the process.

[0175] S760 indicates a bearing failure in the wind turbine unit.

[0176] The specific implementation principles and technical effects of each of the above steps are similar to the wind turbine bearing fault detection methods provided in the above method embodiments, and will not be repeated here for the sake of brevity.

[0177] Figure 8 This application illustrates a method for detecting bearing failures in wind turbines, which involves detecting whether a bearing in a wind turbine has failed based on n third frequencies and n fourth frequencies. Figure 8 As shown, the method may include the following steps:

[0178] S810, based on the characteristic frequency and harmonics, determines the third target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the first cabin acceleration, and the fourth target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the second cabin acceleration.

[0179] S820, if a second interference frequency exists in the i-th fifth frequency interval, remove the i-th third target n-fold frequency from the n third target n-fold frequencies; if a second interference frequency exists in the i-th sixth frequency interval, remove the i-th fourth target n-fold frequency from the n fourth target n-fold frequencies to obtain the first frequency set.

[0180] S830, remove the second target frequency corresponding to the second target amplitude from the first frequency set to obtain the second frequency set.

[0181] S840 performs deduplication on all frequencies in the second frequency set to obtain the third frequency set.

[0182] S850, is the number of elements in the third frequency set greater than 1?

[0183] If the number of elements in the third frequency set is greater than 1, proceed to step S860. Otherwise, proceed to step S880 to end the process.

[0184] S860, Does there exist a target ratio that belongs to the preset ratio set among all ratios?

[0185] If a target ratio belonging to the preset ratio set exists among all ratios, proceed to step S870. Otherwise, proceed to step S880 to end the process.

[0186] S870 indicates that the wind turbine's bearings have failed.

[0187] The specific implementation principles and technical effects of each of the above steps are similar to the wind turbine bearing fault detection methods provided in the above method embodiments, and will not be repeated here for the sake of brevity.

[0188] Based on the same inventive concept, this application also provides a wind turbine bearing fault detection device.

[0189] Figure 9 This is a schematic diagram illustrating the structure of a wind turbine bearing fault detection device according to an exemplary embodiment. Figure 9 As shown, the wind turbine bearing fault detection device 900 may specifically include:

[0190] The acquisition module 910 is used to acquire the operating parameters of the wind turbine, including the nacelle acceleration and the wind turbine rotation speed.

[0191] The filter module 920 is used to perform high-pass filtering on the nacelle acceleration when the rotation speed is stable, so as to obtain the target acceleration within a preset frequency range.

[0192] Extraction module 930 is used to extract the frequency domain features of the target acceleration;

[0193] The detection module 940 is used to detect whether the bearings of the wind turbine are faulty based on frequency domain characteristics and rotational speed.

[0194] In some embodiments, frequency domain features may include amplitude and the frequency corresponding to the amplitude;

[0195] The detection module 940 may include:

[0196] The first acquisition unit is used to acquire the M largest first target amplitudes among the amplitudes of the frequency domain features, where M is a positive integer;

[0197] The first determining unit is used to determine the first target frequency corresponding to each of the M first target amplitudes, thereby obtaining the M first target frequencies;

[0198] The second determining unit is used to calculate the characteristic frequency and harmonic frequency n of the bearing based on the rotational speed, where n is a positive integer.

[0199] The first detection unit is used to detect whether the bearings of the wind turbine unit have failed, based on M first target frequencies, characteristic frequencies, and harmonics.

[0200] In some embodiments, the cabin acceleration includes a spectrum of a first cabin acceleration in a first direction and a spectrum of a second cabin acceleration in a second direction, wherein the first direction is perpendicular to the second direction;

[0201] The first detection unit may include:

[0202] The computational subunit can be used to calculate the first nth harmonic based on the characteristic frequency and harmonics;

[0203] The first determining subunit can be used to determine, based on the first n-fold frequency, the first target n-fold frequency corresponding to the maximum amplitude in the first frequency interval of the spectrum diagram of the first cabin acceleration, and the second target n-fold frequency corresponding to the maximum amplitude in the second frequency interval of the spectrum diagram of the second cabin acceleration.

[0204] The second determining subunit can be used to determine that the bearing of the wind turbine has failed when there is a first target frequency among M first target frequencies that is the same as the nth harmonic of the first target or the nth harmonic of the second target.

[0205] In some embodiments, the wind turbine bearing fault detection device 900 may further include:

[0206] The third determining unit can be used to determine whether a first interference frequency exists in the third frequency range and the fourth frequency range respectively. The third frequency range is the frequency range in which the frequency deviation from the first target n-fold frequency is within the first deviation range, and the fourth frequency range is the frequency range in which the frequency deviation from the second target n-fold frequency is within the second deviation range.

[0207] The second defined subunit can be specifically used for:

[0208] If there is no first interference frequency in either the third or fourth frequency range, and if among the M first target frequencies there is a first target frequency that is the same as the nth harmonic of the first target or the nth harmonic of the second target, then the bearing of the wind turbine unit is determined to have failed.

[0209] In some embodiments, the frequency domain features include a spectrum of a first cabin acceleration in a first direction and a spectrum of a second cabin acceleration in a second direction, wherein the first direction is perpendicular to the second direction;

[0210] The detection module 940 may include:

[0211] The second determining unit calculates the characteristic frequency and harmonic frequency n of the bearing based on the rotational speed, where n is a positive integer.

[0212] The fourth determining unit is used to determine, based on the characteristic frequency and harmonics, the third target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the first cabin acceleration, and the fourth target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the second cabin acceleration;

[0213] The second detection unit is used to detect whether the bearings of the wind turbine have failed, based on n harmonics of the third target and n harmonics of the fourth target.

[0214] In some embodiments, the second detection unit may include:

[0215] The sub-unit is determined to determine whether a second interference frequency exists in the i-th fifth frequency interval and the i-th sixth frequency interval, respectively, where i is a positive integer less than or equal to n;

[0216] The sub-unit is removed to remove the i-th third target n-fold frequency from the n third frequencies when there is a second interference frequency in the i-th fifth frequency interval, and to remove the i-th fourth target n-fold frequency from the n fourth target n-fold frequencies when there is a second interference frequency in the i-th sixth frequency interval, so as to obtain the first frequency set.

[0217] The detection subunit is used to detect whether the bearings of the wind turbine unit have failed, based on the first frequency set.

[0218] In some embodiments, the detection subunit may include:

[0219] The amplitude acquisition subunit is used to acquire a preset threshold and the amplitude corresponding to each frequency in the first frequency set;

[0220] The amplitude determination subunit is used to determine the second target amplitude in the first frequency set whose amplitude is less than a preset threshold.

[0221] The frequency removal subunit is used to remove the second target frequency corresponding to the second target amplitude in the first frequency set to obtain the second frequency set.

[0222] The deduplication subunit is used to deduplicat all frequencies in the second frequency set to obtain the third frequency set.

[0223] The fault detection subunit is used to detect whether the bearings of the wind turbine have failed, based on the third frequency set.

[0224] In some embodiments, the fault detection subunit is specifically used for:

[0225] Obtain a preset set of ratios and any two frequencies from the third set of frequencies;

[0226] Calculate the ratio of any two frequencies;

[0227] If a target ratio belonging to a preset ratio set exists among all ratios, it is determined that the bearing of the wind turbine unit has failed.

[0228] In some embodiments, the wind turbine bearing fault detection device 900 may further include:

[0229] The DC component removal module is used to remove the DC component of the nacelle acceleration and obtain the AC component of the nacelle acceleration.

[0230] The filter module 920 may include:

[0231] The filtering subunit performs high-pass filtering on the AC component of the cabin acceleration to obtain the target AC component of the cabin acceleration that belongs to the preset frequency range.

[0232] The fifth determining unit is used to determine the target AC component as the target acceleration.

[0233] In some embodiments, the wind turbine bearing fault detection device 900 may further include:

[0234] The calculation module is used to calculate the fluctuation coefficient of the rotational speed, which indicates the magnitude of the change in rotational speed.

[0235] The judgment module is used to determine that the rotational speed is in a stable state when the fluctuation coefficient meets the preset fluctuation conditions.

[0236] The wind turbine bearing fault detection device provided in this embodiment can be used to execute the wind turbine bearing fault detection methods provided in the above-described method embodiments. The implementation methods and technical effects are similar, and will not be described again here.

[0237] Based on the same inventive concept, embodiments of this application also provide a controller, such as... Figure 10 As shown, the controller may include a processor 1001 and a memory 1002 storing computer program instructions.

[0238] Specifically, the processor 1001 may include a central processing unit (CPU), an application specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of the present invention.

[0239] Memory 1002 may include a large-capacity storage device for data or instructions. For example, and not limitingly, memory 1002 may include a hard disk drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 1002 may include removable or non-removable (or fixed) media. Where appropriate, memory 1002 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 1002 is a non-volatile solid-state memory. In a particular embodiment, memory 502 includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0240] The processor 1001 reads and executes computer program instructions stored in the memory 1002 to implement any of the wind turbine bearing fault detection methods in the above embodiments.

[0241] In one example, the main controller may also include a communication interface 1003 and a bus 1010. Wherein, for example... Figure 10 As shown, the processor 1001, memory 1002, and communication interface 1003 are connected through bus 1010 and complete communication with each other.

[0242] The communication interface 1003 is mainly used to realize communication between various modules, devices, units and / or devices in the embodiments of the present invention.

[0243] Bus 1010 includes hardware, software, or both, that couples components of the main controller together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 1010 may include one or more buses. While specific buses are described and illustrated in embodiments of the invention, the invention contemplates any suitable bus or interconnect.

[0244] The controller can execute the wind turbine bearing fault detection method in this embodiment of the invention, thereby achieving... Figures 1 to 9 The method and apparatus for detecting bearing faults in wind turbine generators are described.

[0245] Furthermore, in conjunction with the wind turbine bearing fault detection method in the above embodiments, this invention can be implemented using a computer-readable storage medium. This computer-readable storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the wind turbine bearing fault detection methods described in the above embodiments.

[0246] It should be understood that the present invention is not limited to the specific configurations and processes described above and illustrated in the figures. For the sake of brevity, a detailed description of known methods is omitted. In the above embodiments, several specific steps are described and illustrated as examples. However, the method of the present invention is not limited to the specific steps described and illustrated. Those skilled in the art may make various changes, modifications, and additions, or change the order of the steps after understanding the spirit of the present invention.

[0247] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the required tasks. The programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0248] It should also be noted that the exemplary embodiments described herein describe methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the steps described above. In other words, the steps may be performed in the order described in the embodiments, or in a different order, or several steps may be performed simultaneously.

[0249] The above description is merely a specific embodiment of the present invention. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A method for detecting bearing faults in wind turbine generators, characterized in that, include: The operating parameters of the wind turbine are obtained, including the nacelle acceleration and the rotational speed of the wind turbine. When the rotational speed is stable, the nacelle acceleration is high-pass filtered to obtain the target acceleration within a preset frequency range; Extract the frequency domain features of the target acceleration; Based on the frequency domain characteristics and the rotational speed, it is detected whether the bearing of the wind turbine unit has failed; The frequency domain feature includes the amplitude and the frequency corresponding to the amplitude; The step of detecting whether the bearing of the wind turbine has failed based on the frequency domain characteristics and the rotational speed includes: Among the amplitude values ​​of the frequency domain features, the M largest first target amplitude values ​​are obtained, where M is a positive integer; For the M first target amplitudes, determine the first target frequency corresponding to each first target amplitude to obtain the M first target frequencies; The characteristic frequency and harmonic frequency n of the bearing are calculated based on the rotational speed, where n is a positive integer. Based on the M first target frequencies, the characteristic frequencies, and harmonics, detect whether the bearings of the wind turbine unit have failed; The nacelle acceleration includes a spectrum of a first nacelle acceleration in a first direction and a spectrum of a second nacelle acceleration in a second direction, wherein the first direction is perpendicular to the second direction; The method of detecting whether the bearing of the wind turbine has failed based on the M first target frequencies, the characteristic frequency, and the harmonics includes: The first nth harmonic is calculated based on the characteristic frequency and harmonics. Based on the first n-fold frequency, determine the first target n-fold frequency corresponding to the maximum amplitude in the first frequency interval of the spectrum diagram of the first cabin acceleration, and the second target n-fold frequency corresponding to the maximum amplitude in the second frequency interval of the spectrum diagram of the second cabin acceleration; If, among the M first target frequencies, there exists a first target frequency that is the same as the nth harmonic of the first target frequency or the nth harmonic of the second target frequency, it is determined that the bearing of the wind turbine has failed.

2. The method according to claim 1, characterized in that, Before determining that the bearing of the wind turbine has failed, if among the M first target frequencies there exists a first target frequency that is the same as the nth harmonic of the first target frequency or the nth harmonic of the second target frequency, the method further includes: Determine whether a first interference frequency exists in the third frequency range and the fourth frequency range respectively. The third frequency range is the frequency range in which the frequency deviation from the first target n-fold frequency is within the third deviation range. The fourth frequency range is the frequency range in which the frequency deviation from the second target n-fold frequency is within the fourth deviation range. If, among the M first target frequencies, there exists a first target frequency that is the same as the nth harmonic of the first target frequency or the nth harmonic of the second target frequency, determining that the bearing of the wind turbine has failed includes: If the first interference frequency is not present in either the third or fourth frequency range, and if among the M first target frequencies there is a first target frequency that is the same as the nth harmonic of the first target or the nth harmonic of the second target, then it is determined that the bearing of the wind turbine has failed.

3. The method according to any one of claims 1-2, characterized in that, Before performing high-pass filtering on the nacelle acceleration to obtain the target acceleration within a preset frequency range when the rotational speed is stable, the method further includes: Remove the DC component of the nacelle acceleration to obtain the AC component of the nacelle acceleration; When the rotational speed is stable, the nacelle acceleration is high-pass filtered to obtain the target acceleration within a preset frequency range, including: When the rotational speed is stable, the AC component of the nacelle acceleration is high-pass filtered to obtain the target AC component of the AC component of the nacelle acceleration that belongs to the preset frequency range. The target AC component is determined as the target acceleration.

4. The method according to any one of claims 1-2, characterized in that, Before performing high-pass filtering on the nacelle acceleration to obtain the target acceleration within a preset frequency range when the rotational speed is stable, the method further includes: Calculate the fluctuation coefficient of the rotational speed, which is used to indicate the magnitude of the change in the rotational speed; If the fluctuation coefficient meets the preset fluctuation conditions, the rotational speed is determined to be in the stable state.

5. A method for detecting bearing faults in wind turbine generators, characterized in that, include: The operating parameters of the wind turbine are obtained, including the nacelle acceleration and the rotational speed of the wind turbine. When the rotational speed is stable, the nacelle acceleration is high-pass filtered to obtain the target acceleration within a preset frequency range; Extract the frequency domain features of the target acceleration; Based on the frequency domain characteristics and the rotational speed, it is detected whether the bearing of the wind turbine unit has failed; The frequency domain features include the spectrum of the first cabin acceleration in a first direction and the spectrum of the second cabin acceleration in a second direction, wherein the first direction is perpendicular to the second direction; The step of detecting whether the bearing of the wind turbine has failed based on the frequency domain characteristics and the rotational speed includes: The characteristic frequency and harmonic frequency n of the bearing are calculated based on the rotational speed, where n is a positive integer. Based on the characteristic frequency and harmonics, the third target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the first cabin acceleration, and the fourth target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the second cabin acceleration are determined. The bearings of the wind turbine are tested for failure based on n third target n-fold frequencies and n fourth target n-fold frequencies.

6. The method according to claim 5, characterized in that, The step of detecting whether the bearings of the wind turbine have failed based on n third target n-fold harmonics and n fourth target n-fold harmonics includes: Determine whether a second interference frequency exists in the i-th fifth frequency interval and the i-th sixth frequency interval, respectively, where i is a positive integer less than or equal to n; If the second interference frequency exists in the i-th fifth frequency interval, the i-th third target n-fold frequency is removed from the n third target n-fold frequencies. If the second interference frequency exists in the i-th sixth frequency interval, the i-th fourth target n-fold frequency is removed from the n fourth target n-fold frequencies to obtain the first frequency set. Based on the first frequency set, detect whether the bearing of the wind turbine unit has failed.

7. The method according to claim 6, characterized in that, The step of detecting whether the bearing of the wind turbine has failed based on the first frequency set includes: Obtain the preset threshold and the amplitude corresponding to each frequency in the first frequency set; Determine a second target amplitude in the first frequency set whose amplitude is less than the preset threshold; Remove the second target frequency corresponding to the second target amplitude from the first frequency set to obtain the second frequency set; The second frequency set is deduplicated to obtain the third frequency set. Based on the third frequency set, the bearings of the wind turbine are detected to be faulty.

8. The method according to claim 7, characterized in that, The step of detecting whether the bearing of the wind turbine has failed based on the third frequency set includes: Obtain a preset set of ratios and any two frequencies from the third set of frequencies; Calculate the ratio of any two frequencies; If a target ratio belonging to the preset ratio set exists among all the ratios, it is determined that the bearing of the wind turbine unit has failed.

9. The method according to any one of claims 5-8, characterized in that, Before performing high-pass filtering on the nacelle acceleration to obtain the target acceleration within a preset frequency range when the rotational speed is stable, the method further includes: Remove the DC component of the nacelle acceleration to obtain the AC component of the nacelle acceleration; When the rotational speed is stable, the nacelle acceleration is high-pass filtered to obtain the target acceleration within a preset frequency range, including: When the rotational speed is stable, the AC component of the nacelle acceleration is high-pass filtered to obtain the target AC component of the AC component of the nacelle acceleration that belongs to the preset frequency range. The target AC component is determined as the target acceleration.

10. The method according to any one of claims 5-8, characterized in that, Before performing high-pass filtering on the nacelle acceleration to obtain the target acceleration within a preset frequency range when the rotational speed is stable, the method further includes: Calculate the fluctuation coefficient of the rotational speed, which is used to indicate the magnitude of the change in the rotational speed; If the fluctuation coefficient meets the preset fluctuation conditions, the rotational speed is determined to be in the stable state.

11. A wind turbine bearing fault detection device, characterized in that, include: The acquisition module is used to acquire the operating parameters of the wind turbine, including the nacelle acceleration and the rotational speed of the wind turbine. The filtering module is used to perform high-pass filtering on the nacelle acceleration when the rotational speed is stable, so as to obtain the target acceleration within a preset frequency range; An extraction module is used to extract the frequency domain features of the target acceleration; The detection module is used to detect whether the bearing of the wind turbine has failed based on the frequency domain characteristics and the rotational speed. The frequency domain feature includes the amplitude and the frequency corresponding to the amplitude; The detection module includes: The first acquisition unit is used to acquire the M first target amplitudes with the largest values ​​among the amplitudes of the frequency domain features, where M is a positive integer; The first determining unit is configured to determine the first target frequency corresponding to each of the M first target amplitudes, thereby obtaining the M first target frequencies; The second determining unit is used to calculate the characteristic frequency and harmonic frequency n of the bearing based on the rotational speed, where n is a positive integer. The first detection unit is used to detect whether the bearing of the wind turbine has failed based on the M first target frequencies, the characteristic frequency and the harmonics; The nacelle acceleration includes a spectrum of a first nacelle acceleration in a first direction and a spectrum of a second nacelle acceleration in a second direction, wherein the first direction is perpendicular to the second direction; The first detection unit includes: The calculation subunit is used to calculate the first nth harmonic based on the characteristic frequency and the harmonic. The first determining subunit is used to determine, based on the first n-fold frequency, a first target n-fold frequency corresponding to the maximum amplitude in a first frequency interval of the spectrum diagram of the first cabin acceleration, and a second target n-fold frequency corresponding to the maximum amplitude in a second frequency interval of the spectrum diagram of the second cabin acceleration; The second determining subunit is used to determine that the bearing of the wind turbine has failed if there is a first target frequency among the M first target frequencies that is the same as the nth harmonic of the first target frequency or the nth harmonic of the second target frequency.

12. A wind turbine bearing fault detection device, characterized in that, include: The acquisition module is used to acquire the operating parameters of the wind turbine, including the nacelle acceleration and the rotational speed of the wind turbine. The filtering module is used to perform high-pass filtering on the nacelle acceleration when the rotational speed is stable, so as to obtain the target acceleration within a preset frequency range; An extraction module is used to extract the frequency domain features of the target acceleration; The detection module is used to detect whether the bearing of the wind turbine has failed based on the frequency domain characteristics and the rotational speed. The frequency domain features include the spectrum of the first cabin acceleration in a first direction and the spectrum of the second cabin acceleration in a second direction, wherein the first direction is perpendicular to the second direction; The detection module includes: The second determining unit is used to calculate the characteristic frequency and harmonic frequency n of the bearing based on the rotational speed, where n is a positive integer. The fourth determining unit is used to determine, based on the characteristic frequency and the harmonics, the third target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the first cabin acceleration, and the fourth target n harmonic corresponding to the maximum amplitude value of each of the n harmonics in the spectrum diagram of the second cabin acceleration; The second detection unit is used to detect whether the bearing of the wind turbine has failed based on n third target n-fold frequencies and n fourth target n-fold frequencies.

13. A controller, characterized in that, The controller includes: a processor, and a memory storing computer program instructions; The processor reads and executes the computer program instructions to implement the wind turbine bearing fault detection method as described in any one of claims 1-10.

14. A readable storage medium, characterized in that, The readable storage medium stores computer program instructions, which, when executed by a processor, implement the wind turbine bearing fault detection method as described in any one of claims 1-10.

Images