Method and system for monitoring blade damage of a wind turbine generator

Abstract

The present disclosure provides a wind turbine blade damage monitoring method and system, the method comprising: acquiring acoustic emission signals of acoustic emission sensors on each blade; performing feature analysis on the acoustic emission signals to obtain feature values of each blade on multiple acoustic emission features; for each acoustic emission feature, analyzing the distribution difference between the feature values of the multiple blades to obtain a difference degree corresponding to each acoustic emission feature; in response to the difference degrees of the multiple acoustic emission features satisfying a preset difference condition, determining that at least one of the multiple blades has a damage risk. The present disclosure solves the problem of difficult accurate monitoring of blade damage, and can analyze the damage risk of the blade by considering multiple acoustic emission features and comparing the feature value distribution between blades, thereby providing more accurate damage monitoring results.

Description

Technical Field

[0001] This disclosure relates to the field of wind power generation, and more specifically, to a method and system for monitoring blade damage in wind turbine generator sets. Background Technology

[0002] With the widespread application of wind turbine generators, the units may need to face various harsh and complex external environments, which may accelerate the damage to the turbine blades. Therefore, higher requirements are placed on the monitoring scheme for blade damage.

[0003] In response to this, related technologies have found that, in some cases, tensile tests on glass fiber, epoxy resin, and composite laminate specimens can be used to obtain the characteristic frequency bands of the fracture modes, and blade failure modes can be classified based on these characteristic frequency bands. However, such methods cannot provide a monitoring scheme for the damage modes of blades during actual operation. Furthermore, blind signal separation (BSS) acoustic emission signals can be used to identify the main crack and crack propagation state of the blade. However, this method relies on relatively singular signal characteristics in blade failure analysis and cannot provide accurate blade damage monitoring. Summary of the Invention

[0004] In view of the difficulty in accurately monitoring blade damage in related technologies, this disclosure provides a method and system for monitoring blade damage in wind turbine generators.

[0005] The first aspect of this disclosure provides a method for monitoring blade damage in a wind turbine generator set. The wind turbine generator set includes multiple blades, each blade being equipped with an acoustic emission sensor. The blade damage monitoring method includes: acquiring acoustic emission signals from the acoustic emission sensors on each blade; performing feature analysis on the acoustic emission signals to obtain feature values ​​for each blade on multiple acoustic emission features; analyzing the distribution differences among the feature values ​​of the multiple blades for each acoustic emission feature to obtain the degree of difference corresponding to each acoustic emission feature; and determining that at least one blade among the multiple blades is at risk of damage in response to the degree of difference of the multiple acoustic emission features satisfying a preset difference condition, wherein the preset difference condition indicates that there is an acoustic emission feature among the multiple acoustic emission features whose degree of difference exceeds a preset value.

[0006] Optionally, the blade damage monitoring method further includes: in response to the degree of difference of the plurality of acoustic emission features satisfying the preset difference condition, determining that the blades among the plurality of blades that satisfy the preset damage condition have a damage risk, wherein the preset damage condition characterizes that the feature value of the blade on at least one acoustic emission feature is the largest among all blades.

[0007] Optionally, the plurality of acoustic emission features include a plurality of signal waveform features, the signal waveform features characterizing the waveform feature parameters of the acoustic emission signal, wherein the preset damage condition includes: for more than a predetermined number of signal waveform features, the feature value of the blade is the maximum value among all feature values ​​of the blade.

[0008] Optionally, the plurality of acoustic emission features include a plurality of frequency band energy features, each frequency band energy feature corresponding to a plurality of frequency bands, and each frequency band energy feature characterizes the energy feature of the corresponding frequency band. The plurality of frequency band energy features include a first energy feature and a second energy feature. The minimum frequency of the frequency band corresponding to the first energy feature is greater than or equal to the maximum frequency of the frequency band corresponding to the second energy feature. The preset damage condition includes satisfying that the feature value of the blade on the first energy feature is greater than the feature value of the blade on the second energy feature.

[0009] Optionally, both the first energy feature and the second energy feature are multiple, wherein the preset damage condition further includes: satisfying that the sum of the feature values ​​of the blade on the first energy feature is greater than a first threshold; and satisfying that the sum of the feature values ​​of the blade on the second energy feature is less than a second threshold.

[0010] Optionally, the plurality of frequency band energy features further includes a third energy feature, wherein the minimum frequency of the frequency band corresponding to the third energy feature is greater than or equal to the maximum frequency of the frequency band corresponding to the first energy feature, wherein the preset damage condition further includes: the feature value of the blade on the third energy feature is not a null value.

[0011] Optionally, the plurality of acoustic emission features include a plurality of signal waveform features and a plurality of frequency band energy features. The signal waveform features characterize the waveform feature parameters of the acoustic emission signal. The plurality of frequency band energy features correspond to a plurality of frequency bands, and each frequency band energy feature characterizes the energy feature of the corresponding frequency band. The preset difference condition includes: satisfying that there are signal waveform features among the plurality of signal waveform features whose difference exceeds a first preset value; and satisfying that there are frequency band energy features among the plurality of frequency band energy features whose difference exceeds a second preset value.

[0012] Optionally, the plurality of frequency bands are all included in the full frequency band of the acoustic emission signal, and the frequency band energy characteristics characterize the proportion of energy of the corresponding frequency band in the full frequency band, wherein the degree of difference is the consistency difference between the characteristic values ​​of the plurality of blades, and the consistency difference characterizes the degree of dispersion between each characteristic value.

[0013] Optionally, the blade damage monitoring method further includes: before performing feature analysis on the acoustic emission signal, removing the signal component corresponding to the vibration frequency from the acoustic emission signal based on the vibration frequency of the transmission chain of the wind turbine generator set, to obtain the removed acoustic emission signal.

[0014] A second aspect of this disclosure provides a blade damage monitoring system for a wind turbine generator set, the wind turbine generator set including multiple blades, the blade damage monitoring system including a monitoring platform and an acoustic emission sensor configured on each blade, the monitoring platform including: a processor; and a memory for storing processor-executable instructions, wherein the processor-executable instructions, when executed by the processor, cause the processor to perform a blade damage monitoring method for a wind turbine generator set according to an embodiment of this disclosure.

[0015] Optionally, the wind turbine generator set further includes a main control system, and the monitoring platform sends the execution result of the processor to the main control system so that the main control system executes control actions corresponding to the execution result. The control actions include controlling the wind turbine generator set to retract its propellers and / or shut down.

[0016] A third aspect of this disclosure provides a computer-readable storage medium that, when instructions in the computer-readable storage medium are executed by a processor of an electronic device, enables the electronic device to perform a blade damage monitoring method for a wind turbine generator according to embodiments of this disclosure.

[0017] A fourth aspect of this disclosure provides a computer program product including computer-executable instructions that, when executed by at least one processor, implement the blade damage monitoring method for a wind turbine generator according to embodiments of this disclosure.

[0018] According to the wind turbine blade damage monitoring method and system disclosed herein, acoustic emission signals on the blades can be acquired, and feature analysis can be performed on the acoustic emission signals to obtain feature values ​​of the blades on multiple acoustic emission features. Furthermore, by analyzing the distribution differences between the feature values ​​of each blade, the degree of difference corresponding to each acoustic emission feature can be obtained. Thus, if the degree of difference of the acoustic emission features meets the preset difference conditions, it can be determined that at least one blade among multiple blades is at risk of damage. In this way, by considering multiple acoustic emission features and comparing the feature value distributions between each blade, the damage risk of the blade can be comprehensively analyzed, thereby providing more accurate damage monitoring results and enabling timely blade maintenance. Attached Figure Description

[0019] Figure 1 This is a schematic diagram showing the configuration location of an acoustic emission sensor in a wind turbine blade damage monitoring method according to an exemplary embodiment of the present disclosure.

[0020] Figure 2 This is a schematic flowchart illustrating a method for monitoring blade damage of a wind turbine generator according to an exemplary embodiment of the present disclosure.

[0021] Figure 3 This is a schematic flowchart illustrating the wavelet packet denoising steps in a wind turbine blade damage monitoring method according to an exemplary embodiment of the present disclosure.

[0022] Figure 4 This is a comparison diagram showing the acoustic emission signals before and after wavelet packet denoising in a wind turbine blade damage monitoring method according to an exemplary embodiment of the present disclosure.

[0023] Figure 5 This is a comparison diagram showing the acoustic emission signals before and after filtering in a wind turbine blade damage monitoring method according to an exemplary embodiment of the present disclosure.

[0024] Figure 6 This is a schematic flowchart illustrating the acoustic emission feature processing method in a wind turbine blade damage monitoring method according to an exemplary embodiment of the present disclosure.

[0025] Figure 7 This is a schematic diagram showing the waveform of acoustic emission signals during blade damage in a wind turbine blade damage monitoring method according to an exemplary embodiment of the present disclosure.

[0026] Figure 8 This is a schematic flowchart illustrating an example of a blade damage monitoring method for a wind turbine generator according to an exemplary embodiment of the present disclosure.

[0027] Figure 9 This is a schematic block diagram illustrating a blade damage monitoring system for a wind turbine generator according to an exemplary embodiment of the present disclosure. Detailed Implementation

[0028] The following detailed embodiments are provided to aid the reader in gaining a comprehensive understanding of the methods, apparatus, and / or systems described herein. However, various changes, modifications, and equivalents of the methods, apparatus, and / or systems described herein will become apparent upon understanding this disclosure. For example, the order of operations described herein is merely illustrative and is not limited to those orders set forth herein, but may be changed as will become clear upon understanding this disclosure, except for operations that must occur in a specific order. Furthermore, for clarity and conciseness, descriptions of features known in the art may be omitted.

[0029] The features described herein may be implemented in different forms and should not be construed as limited to the examples described herein. Rather, the examples described herein are provided only to illustrate some of the many feasible ways of implementing the methods, apparatus, and / or systems described herein, which will become clear upon understanding the disclosure of this application.

[0030] As used herein, the term “and / or” includes any one of the associated listed items and any combination of any two or more.

[0031] Although terms such as “first,” “second,” and “third” may be used herein to describe various components, assemblies, regions, layers, or parts, these components, assemblies, regions, layers, or parts should not be limited by these terms. Rather, these terms are used only to distinguish one component, assembly, region, layer, or part from another. Thus, without departing from the teaching of the examples described herein, the first component, first assembly, first region, first layer, or first part referred to as the first component, first assembly, first region, first layer, or first part may also be referred to as the second component, second assembly, second region, second layer, or second part.

[0032] In the specification, when an element (such as a layer, region, or substrate) is described as being "on" another element, "connected to," or "bonded to" another element, the element may be directly "on" another element, directly "connected to," or "bonded to" the other element, or one or more other elements may be present in between. Conversely, when an element is described as being "directly on" another element, "directly connected to," or "directly bonded to" another element, no other elements may be present in between.

[0033] The terminology used herein is for the purpose of describing various examples only and is not intended to limit disclosure. Unless the context clearly indicates otherwise, the singular form is intended to include the plural form as well. The terms “comprising,” “including,” and “having” indicate the presence of the described features, quantities, operations, components, elements, and / or combinations thereof, but do not preclude the presence or addition of one or more other features, quantities, operations, components, elements, and / or combinations thereof.

[0034] Unless otherwise defined, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains upon understanding this disclosure. Unless expressly defined herein, terms (such as those defined in a general dictionary) shall be interpreted as having a meaning consistent with their meaning in the context of the relevant field and in this disclosure, and shall not be interpreted in an idealized or overly formalistic manner.

[0035] Furthermore, in the description of the examples, detailed descriptions of well-known related structures or functions will be omitted when it is believed that such detailed descriptions would lead to a vague interpretation of this disclosure.

[0036] Acoustic emission monitoring (AE) is a non-destructive testing method that monitors and records acoustic signals in real time and assesses the health of a structure through signal analysis. For example, when wind turbine blades are subjected to external influences or suffer internal damage, specific stress wave signals are generated. By analyzing and comparing these signals, potential problems with the blades can be identified.

[0037] Acoustic emission sensors can be used to monitor damage in the early and middle stages of wind turbine blades. Acoustic emission monitoring technology determines the presence of potential structural damage by detecting acoustic signals generated within the material. For wind turbine blades, this technology can help monitor the structural integrity of the blades and detect any possible defects or damage in a timely manner.

[0038] In view of the problems described above, this disclosure provides a method, system, computer-readable storage medium, and computer program product for monitoring blade damage of wind turbine generators, in order to solve or at least alleviate the problem of inaccurate monitoring of blade damage in related technologies.

[0039] According to a first aspect of an exemplary embodiment of the present disclosure, a method for monitoring blade damage of a wind turbine generator set is provided, wherein the wind turbine generator set includes a plurality of blades, each blade being equipped with an acoustic emission sensor.

[0040] Figure 1 An example configuration location of an acoustic emission sensor on a blade according to an exemplary embodiment of this disclosure is shown. Figure 1 As shown, a wind turbine generator may include three blades. Acoustic emission sensors can be configured at the location with the highest risk of damage at the trailing edge of each blade. For example, the acoustic emission sensors can be configured within a range of 15 to 30 meters from the blade tip, such as at a position of 22 meters from the blade tip. Acoustic emission sensors can be configured within the aforementioned range on the leading and / or trailing edges of the blade. The acoustic emission sensors can be attached to the leeward (SS) surface of the blade for measuring P-waves, S-waves, and mixed waves. These acoustic emission sensors can form the blade's acoustic emission system. The sampling frequency of each acoustic emission sensor can be, for example, 1.67 MHz, where 2 seconds of acoustic emission data are stored every 10 seconds. As an example, the configuration positions of the acoustic emission sensors on each blade can be the same, but this is not a limitation. The configuration positions of the acoustic emission sensors on different blades can also be different, and one or more acoustic emission sensors can be configured on each blade.

[0041] Although the location of the acoustic emission sensor is described here as an example at the trailing edge SS surface skin 22 meters from the blade tip, the embodiments of this disclosure are not limited to this. For example, the location of the acoustic emission sensor could also be at the leading edge windward surface (PS surface) 20 meters from the blade root. Furthermore, in the embodiments of this disclosure, various combinations of acoustic emission sensor locations and distances, the number of sensors, and the sensor mounting process can all be changed.

[0042] Based on the configured acoustic emission sensor, the blade damage monitoring method may include the following steps:

[0043] like Figure 2 As shown, in step S210, the acoustic emission signals of the acoustic emission sensors on each blade can be acquired.

[0044] Specifically, acoustic emission signals can be collected by acoustic emission sensors configured on each blade during the operation of the wind turbine generator set.

[0045] As an example, after acquiring the acoustic emission signal, it can be preprocessed to obtain a processed signal for subsequent feature analysis.

[0046] In one example, preprocessing of the acoustic emission signal may include wavelet packet denoising.

[0047] Specifically, when monitoring blades in the field, external factors such as the field environment have a significant impact on the acquired acoustic emission signals, making the acquired signals more complex and resulting in a large amount of noise mixed in with the acoustic emission signals. The presence of this noise significantly affects signal processing, thus reducing the signal-to-noise ratio (SNR) of the acoustic emission signals. Furthermore, noise reduction processing of the acoustic emission signals can improve the accuracy of subsequent feature signal extraction and qualitative analysis, thereby improving the SNR of the acoustic emission signals.

[0048] Noise reduction by wavelet packet transformation (NREM) is a commonly used signal processing method suitable for removing noise from signals. This algorithm, developed based on wavelet analysis, effectively reduces noise while preserving important signal information. NREM decomposes the signal into multiple frequency bands, selects appropriate bands for noise reduction, and then reconstructs the processed signal. This method can suppress noise to varying degrees in different frequency bands, thereby improving the noise reduction effect.

[0049] As an example, wavelet packet denoising can be performed in the following way:

[0050] like Figure 3 As shown, in step S310, wavelet basis functions and decomposition levels can be selected.

[0051] Specifically, wavelet packet denoising can be a processing method based on wavelet analysis, in which wavelet transform can be used to denoise acoustic emission signals.

[0052] Here, the object of wavelet packet denoising is the acoustic emission signal of the blade composite material during the degradation process. For example, the dB5 wavelet packet basis function can be selected to perform 7-level wavelet packet decomposition on the acoustic emission signal. The denoising program completed by Python language and its wavelet analysis module package wavelet can be used to perform wavelet denoising on the acoustic emission signal.

[0053] In step S320, wavelet packet decomposition can be performed on the acoustic emission signal.

[0054] Specifically, the acoustic emission signal can be projected into the space spanned by the wavelet packet basis functions mentioned above. The detailed parts of the input signal can be analyzed through multiple iterations of wavelet transformation. The wavelet packet can be represented by an analysis tree to achieve wavelet packet decomposition of the signal.

[0055] In step S330, wavelet coefficients of different layers can be processed.

[0056] In wavelet thresholding denoising, noise can be considered to exist in high-frequency components. For each layer of wavelet coefficients, the noise corresponds to a relatively small value. Therefore, an appropriate threshold can be selected to set wavelet coefficients with absolute values ​​less than a preset threshold to 0, while coefficients with larger absolute values ​​can be retained or reduced. In this way, a noise processing method that compromises between soft and hard thresholds can be achieved.

[0057] In step S340, the denoised signal can be reconstructed by inverse wavelet transform.

[0058] In this step, the inverse wavelet transform can be used for reconstruction, thus obtaining the denoised signal.

[0059] The relevant parameters selected in the wavelet threshold denoising method mentioned above can be used to denoise the acoustic emission data segments collected by a certain unit. The effect of wavelet packet denoising before and after is compared as follows: Figure 4 As shown, from Figure 4 It can be seen that the wavelet thresholding denoising method can effectively remove the interference of noise signals in the experiment and retain the effective components in the signal.

[0060] Although the wavelet packet decomposition denoising method is described herein as an example, the embodiments of this disclosure are not limited thereto. Alternatively or additionally, other denoising methods for acoustic emission signals may be selected, such as denoising methods based on the principles of Empirical Mode Decomposition (EMD) and correlation coefficients.

[0061] In another example, preprocessing of the acoustic emission signal may include filtering.

[0062] Specifically, the blade damage monitoring method may further include: before performing feature analysis on the acoustic emission signal, removing the signal component corresponding to the vibration frequency from the acoustic emission signal based on the vibration frequency of the wind turbine's drive chain, thereby obtaining the removed acoustic emission signal.

[0063] Filtering can be performed after wavelet packet denoising. Here, the acoustic emission signal after wavelet packet denoising may contain vibration frequency components corresponding to the unit's drivetrain, such as low-frequency components at 900Hz and its harmonics. These low-frequency components can be preliminarily identified as frequency components belonging to the drivetrain. Therefore, high-pass filtering is required on the acoustic emission signal to prevent sound frequency components caused by other components in the unit from affecting the blade monitoring method of this disclosure. As an example, a Butterworth high-pass filter (with a cutoff frequency of 200kHz) can be used to perform the above filtering process. The filtered time-domain signal is as follows: Figure 5 As shown, the filtered acoustic emission signal retains only the high-frequency components of the signal. Figure 5 A comparison diagram of acoustic emission signals before and after filtering is shown in a wind turbine blade damage monitoring method according to an exemplary embodiment of the present disclosure.

[0064] In step S220, the acoustic emission signal can be analyzed to obtain the characteristic values ​​of each blade in multiple acoustic emission features.

[0065] In this step, such as Figure 6 As shown, for the acquired acoustic emission signal or the signal after the above preprocessing (e.g., the signal after wavelet packet denoising S610 and filtering S620), the signal can be processed for features, the features of the signal can be analyzed, and the feature values ​​of each blade on each feature can be obtained for subsequent monitoring and diagnosis.

[0066] As an example, feature analysis may include signal waveform feature analysis S630 and / or frequency band energy feature analysis S640. Accordingly, multiple acoustic emission features may include multiple signal waveform features and multiple frequency band energy features.

[0067] Specifically, for signal waveform characteristic analysis, the process of blade damage can be divided into three types: damage initiation, damage propagation, and fracture. Different stages of damage occurrence are accompanied by significant changes in acoustic emission signal parameters, thus exhibiting different signal waveform characteristics. For example, Figure 7 The waveforms of acoustic emission signals at different stages of blade damage are shown, starting when the blade damage begins ( Figure 7 In stage 1), the signal amplitude and energy are low, the duration is short, and the rise is slow; when the blade damage expands ( Figure 7 In stage 2), the signal amplitude and energy are high, the duration is long, and the rise is rapid; when the blade material is close to fracture ( Figure 7 In stage 3), the signal amplitude and energy increase dramatically, last for a longer period, and rise the fastest.

[0068] Using signal waveform characteristics (or also known as "acoustic emission signal characterization") treats the defect signals of wind turbine blade composite materials as burst-type signals. Signal waveform characteristics may include, but are not limited to, the energy, amplitude, cumulative number of events, number of impacts, number of rings, rise time and duration of acoustic emission. In addition, signal waveform characteristics that can characterize the continuity of acoustic emission signals may include the number of rings, signal level and energy, and the effective value of voltage.

[0069] In addition, signal waveform characteristics may also include peak count, average frequency, inverse frequency, initial frequency, signal strength, absolute frequency, local power spectrum, centroid frequency, peak frequency, absolute energy, time of arrival, etc. The embodiments of this disclosure do not impose particular limitations on specific signal waveform characteristics and the frequency band energy characteristics described below. In the embodiments of this disclosure, the combination of various characteristics, the various characteristic quantization methods, and the different characteristic calculation durations can all be changed.

[0070] Through the above signal waveform feature analysis, signal waveform features (also known as "waveform parameter features") can be obtained. Features exceeding a preset threshold value can be extracted. Signal waveform features can characterize the waveform feature parameters of acoustic emission signals. Signal waveform features may include, for example, ring count, rise time, duration, energy, peak amplitude, and root mean square (RMS). Based on such signal waveform features, the level of blade degradation can be determined by comparing the magnitudes of acoustic emission parameter features between test bench blades and in-service units.

[0071] For frequency band energy characteristic analysis, the wavelet energy spectrum coefficient method can be used to calculate the energy of different frequency bands, and the energy of different frequency bands can be spread out in time order to determine the blade deterioration trend.

[0072] Multiple frequency band energy characteristics can be associated with multiple frequency bands, with each frequency band energy characteristic representing the energy characteristics of its corresponding frequency band. For example, the entire frequency band of acoustic emission signals can be divided into multiple frequency bands, which can be further divided into multiple sub-frequency bands. In this way, energy characteristic analysis can be performed on different frequency bands to obtain the energy intensity of different frequency bands. These energy intensity values ​​reflect the distribution of acoustic emission signals in different frequency bands. Therefore, based on the frequency band distribution of acoustic emission signals, it is possible to assess whether blade damage has occurred and the extent of the damage.

[0073] As an example, multiple frequency bands can be included within the full frequency band of the acoustic emission signal, and the frequency band energy characteristics can characterize the proportion of energy of the corresponding frequency band in the full frequency band. For example, the full frequency band can be divided into high-frequency bands, mid-frequency bands, and low-frequency bands, and each of these bands can be further divided into multiple sub-bands, each of which can have corresponding frequency band energy characteristics.

[0074] Multiple frequency band energy characteristics may include a first energy characteristic, a second energy characteristic, and a third energy characteristic. The minimum frequency of the frequency band corresponding to the first energy characteristic (e.g., corresponding to the mid-frequency band) may be greater than or equal to the maximum frequency of the frequency band corresponding to the second energy characteristic (e.g., corresponding to the low-frequency band). The minimum frequency of the frequency band corresponding to the third energy characteristic (e.g., corresponding to the high-frequency band) may be greater than or equal to the maximum frequency of the frequency band corresponding to the first energy characteristic.

[0075] By processing the acoustic emission signal through the above-described signal waveform feature analysis and frequency band energy feature analysis, multiple acoustic emission features can be obtained, including, for example, signal waveform features and / or frequency band energy features. Table 1 below shows 13 example features obtained through processing; however, the embodiments of this disclosure are not limited to these, and other acoustic emission features can also be selected.

[0076] Table 1

[0077]

[0078] In Table 1 above, f i_j This represents the j-th acoustic emission feature of the i-th blade of a wind turbine generator. The value of i can range from 1 to the total number of blades, and the value of j can range from 1 to the total number of acoustic emission features, for example, from 1 to 13 in Table 1 above. Here, the j-th acoustic emission feature can be a signal waveform feature or a frequency band energy feature.

[0079] In step S230, for each acoustic emission feature, the distribution difference between the feature values ​​of multiple blades can be analyzed to obtain the degree of difference corresponding to each acoustic emission feature.

[0080] As mentioned above, the characteristic value of each acoustic emission feature can be calculated for each blade, and the distribution of characteristic values ​​among blades can be statistically analyzed along the dimension of each acoustic emission feature. The differences in the distribution of characteristic values ​​can be analyzed. In this way, characteristic values ​​that deviate from the group can be found, and the blades corresponding to such characteristic values ​​have a higher risk of damage.

[0081] The degree of difference can be represented by the statistical value of the distributional differences among the characteristic values ​​of the leaves. As an example, the degree of difference can be the consistency difference among the characteristic values ​​of multiple leaves; consistency difference characterizes the degree of dispersion among the characteristic values. Methods for measuring consistency difference can include calculations such as standard deviation and variance; correspondingly, the degree of difference can be represented by statistics such as standard deviation and variance.

[0082] In addition to the aforementioned statistics such as standard deviation and variance, examples for calculating consistency differences are provided in the embodiments of this disclosure.

[0083] Specifically, the degree of difference for each acoustic emission characteristic can be determined based on the maximum and minimum values ​​of the characteristic values ​​of each blade. For example, referring to the representation in Table 1, taking a three-bladed unit as an example, assuming that for the same measuring point, the acoustic emission signal characteristic of the j-th blade is represented as f i_j The consistency difference ε of the eigenvalues ​​of the three blades on the j-th acoustic emission feature can be calculated using the following formula (1). j :

[0084]

[0085] The value of j can range from 1 to the total number of acoustic emission features. For example, taking Table 1 as an example, j can take the values ​​1, 2, ..., 13.

[0086] In step S240, in response to the degree of difference of multiple acoustic emission characteristics satisfying a preset difference condition, it can be determined that at least one of the multiple blades is at risk of damage.

[0087] Here, the preset difference condition can characterize the acoustic emission features among multiple acoustic emission features whose degree of difference exceeds a preset value.

[0088] In this step, in one example, the preset difference condition may include: (1) satisfying that there are signal waveform features among multiple signal waveform features with a difference degree exceeding a first preset value; and (2) satisfying that there are frequency band energy features among multiple frequency band energy features with a difference degree exceeding a second preset value.

[0089] Here, the first and second preset values ​​can be set according to actual needs, for example, both can be 0.5.

[0090] As an example, for each signal waveform feature, it can be determined whether the maximum value among the feature values ​​of each blade exceeds a first preset value, and for each frequency band energy feature, it can be determined whether the maximum value among the feature values ​​of each blade exceeds a second preset value. For example, the above preset difference condition can be represented by the following formula (2):

[0091]

[0092] Where [1,n] represents the sequence number of the signal waveform feature, and [n+1,m] represents the sequence number of the frequency band energy feature. Taking Table 1 above as an example, the preset difference condition can be: max(ε j )>0.5, j takes values ​​1, 2, ..., 5; and max(ε) j )>0.5, j takes values ​​of 6, 7, ..., 13.

[0093] However, the preset difference conditions are not limited to this. In another example, the judgment can also be based solely on signal waveform characteristics or frequency band energy characteristics. Specifically, the preset difference conditions may include the above conditions (1) or conditions (2).

[0094] If the degree of difference of each acoustic emission feature meets the above-mentioned preset difference condition, it can be determined that at least one of the multiple blades is at risk of damage; if the degree of difference of each acoustic emission feature does not meet the above-mentioned preset difference condition, it can be considered that the risk of blade damage is relatively small.

[0095] Specifically, by analyzing the distribution of characteristic values ​​among the blades, the degree of dispersion of the characteristic values ​​can be determined. Since the characteristic values ​​of damaged blades may deviate from, or even severely deviate from, the characteristic values ​​of normal blades, the greater the degree of dispersion, the greater the possibility that damaged blades are included among these blades. Thus, by comparing the characteristic values ​​between blades, it is possible to quickly determine whether there are currently damaged blades in the unit, so that they can be maintained or dealt with in a timely manner.

[0096] The above describes the process of determining whether any blades are damaged, thus enabling early warning or maintenance. However, the embodiments of this disclosure are not limited thereto; embodiments of this disclosure can also specifically locate damaged blades.

[0097] As an example, the blade damage monitoring method may further include: in response to the degree of difference of multiple acoustic emission features satisfying a preset difference condition, determining that the blades among the multiple blades that meet the preset damage condition are at risk of damage. Here, the preset damage condition may characterize the blade having the largest eigenvalue of at least one acoustic emission feature among all blades.

[0098] Specifically, compared to normal or undamaged blades, damaged blades may exhibit more pronounced acoustic emission characteristics, such as higher energy, more ringing frequency, and longer duration. These acoustic emission characteristics can be positively correlated with the degree of blade damage; therefore, damaged blades will exhibit the highest characteristic value in at least one acoustic emission feature, significantly deviating from that of normal blades, resulting in a substantial difference. In this way, based on the initial assessment of blade damage, the analyzed acoustic emission characteristics of each blade can be used to further pinpoint the specific damaged blade without introducing new monitoring or calculation processes, achieving a progressive monitoring scheme that integrates initial warning and precise location.

[0099] For the preset damage condition, in one example, as described above, the acoustic emission characteristics may include multiple signal waveform characteristics, which can characterize the waveform characteristic parameters of the acoustic emission signal. In this example, the preset damage condition may include: for more than a predetermined number of signal waveform characteristics, the characteristic value of the blade is the maximum value among all the characteristic values ​​of the blade.

[0100] Here, the predetermined quantity can be set according to actual needs. It can be a preset fixed value or it can be determined based on the total number of signal waveform features. For example, it can be a predetermined proportion of the total number of signal waveform features. For example, taking Table 1 above as an example, the total number of signal waveform features is 5 (i.e., features numbered 1 to 5). The predetermined quantity can be an integer greater than 1 / 2 of the total number of signal waveform features, such as 3. Here, the predetermined proportion can be set according to actual needs.

[0101] If a blade is considered damaged if its characteristic value is the maximum value across more than a predetermined number of signal waveform features, then that blade is considered damaged. Taking Table 1 as an example, if a blade has a characteristic value that is the maximum value across all blades in at least three of the signal waveform features numbered 1-5, then that blade is considered to be at risk of damage.

[0102] Since the waveform characteristics of the signal can reflect the overall waveform characteristics of the acoustic emission signal, the above method can locate the damaged blade based solely on the waveform characteristics, making the location faster while ensuring the accuracy of the monitoring.

[0103] In another example, as described above, the acoustic emission characteristics may include multiple frequency band energy characteristics, each corresponding to a specific frequency band, and each frequency band energy characteristic characterizes the energy characteristics of its corresponding frequency band. Here, the multiple frequency band energy characteristics may include a first energy characteristic and a second energy characteristic, where the minimum frequency of the frequency band corresponding to the first energy characteristic is greater than or equal to the maximum frequency of the frequency band corresponding to the second energy characteristic. The frequency band corresponding to the first energy characteristic may, for example, be 200kHz to 600kHz, and the frequency band corresponding to the second energy characteristic may, for example, be 0 to 200kHz.

[0104] Preset damage conditions may include: satisfying that the characteristic value of the leaf on the first energy characteristic is greater than the characteristic value of the leaf on the second energy characteristic.

[0105] For wind turbine blades, when a blade is damaged, the energy of the acoustic emission frequency domain is mainly concentrated in the mid-frequency range (e.g., 200kHz to 600kHz). Therefore, for the first and second energy characteristics, if the first energy characteristic is greater than the second energy characteristic, it can be considered that the energy of the acoustic emission signal is more concentrated in the frequency band corresponding to the first energy characteristic compared to the frequency band corresponding to the second energy characteristic. Therefore, there may be an energy concentration phenomenon in the mid-frequency band. Thus, blades that meet such conditions can be identified as damaged blades.

[0106] Here, when the frequency band corresponding to the first energy feature and the frequency band corresponding to the second energy feature are both the same, the first energy feature and the second energy feature can be directly compared. However, this disclosure is not limited to this. The first energy feature and the second energy feature can also correspond to multiple frequency bands, and both the first energy feature and the second energy feature are multiple.

[0107] For example, multiple first energy features can each correspond to multiple sub-bands within the 200kHz–600kHz range; multiple second energy features can each correspond to multiple sub-bands within the 0kHz–200kHz range. For instance, there can be four first energy features, corresponding to the 200kHz–300kHz, 300kHz–400kHz, 400kHz–500kHz, and 500kHz–600kHz sub-bands respectively; and two second energy features, corresponding to the 0–100kHz and 100kHz–200kHz sub-bands respectively. However, the number of first and second energy features and the method of band division are not limited to this; other bands can be set as needed.

[0108] In this example, specifically, the preset damage condition may further include: satisfying that the sum of the eigenvalues ​​of the leaf on the first energy characteristic is greater than a first threshold; and satisfying that the sum of the eigenvalues ​​of the leaf on the second energy characteristic is less than a second threshold.

[0109] Here, the first and second thresholds can be set according to actual needs. For example, the first threshold can be 0.3 and the second threshold can be 0.6. The second threshold can be greater than the first threshold. Through the above conditions, the characteristic values ​​of the blade on multiple first energy features and multiple second energy features can be defined as a whole. When the blade meets the preset damage conditions, it can be determined that the blade is damaged; otherwise, it can be considered that the blade is not damaged.

[0110] Taking Table 1 above as an example, the first energy characteristic can be f i_8 f i_9 and f i_10 The second energy characteristic can be f i_6 and f i_7 The preset damage condition can be expressed by the following equation (3):

[0111]

[0112] When the i-th blade satisfies the above equation (3), it can be considered that the blade is damaged.

[0113] Here, when there are multiple first and second energy characteristics, the accuracy of locating damaged blades can be further improved.

[0114] As an example, multiple frequency band energy features may also include a third energy feature. The minimum frequency of the frequency band corresponding to the third energy feature may be greater than or equal to the maximum frequency of the frequency band corresponding to the first energy feature. The frequency band corresponding to the third energy feature may be, for example, a high-frequency band. Although, as mentioned above, when the blade is damaged, the energy of the acoustic radio frequency domain features is mainly concentrated in the mid-frequency band, and due to the damage to the blade, there may also be an energy distribution in the high-frequency band. The energy proportion of the high-frequency band will still be less than that of the mid-frequency band, but it is not a null value.

[0115] To this end, preset damage conditions can also include: the blade's eigenvalue on the third energy characteristic is not null. By setting such preset damage conditions, energy proportion analysis can be performed from different frequency bands, thereby more accurately locating the damaged blade.

[0116] Although examples of automatically locating damaged blades according to embodiments of the present disclosure have been described above, the process of locating damaged blades is not limited thereto. In cases where it is determined that there are damaged blades among all blades according to embodiments of the present disclosure, specific damaged blades can also be located by manual inspection or other automated inspection methods.

[0117] Figure 8 An application example of a leaf damage monitoring method according to embodiments of the present disclosure is shown. For example... Figure 8 As shown, in step S801, scheduling can be performed at predetermined intervals (e.g., 10 seconds), with acoustic emission signal data of a predetermined duration (e.g., 2 seconds) input for each scheduling. In step S802, it can be determined whether the unit is in power generation mode. If the unit is in power generation mode, damage monitoring can be performed. If the unit is not in power generation mode, acoustic emission signal data can continue to be acquired; if the unit is in power generation mode, step S803 can be executed.

[0118] In step S803, feature analysis can be performed on the acoustic emission signal to obtain the feature values ​​of each blade (e.g., blade 1, blade 2, and blade 3) on each acoustic emission feature. For example, the 13 acoustic emission signal features in Table 1 can be processed. In step S804, the consistency difference of the blade on each acoustic emission feature can be calculated to obtain the degree of difference of each acoustic emission feature. In step S805, it can be determined whether the degree of difference of each acoustic emission feature meets the preset difference condition. If the preset difference condition is met, it is determined that there is a damaged blade in the unit's blades; if the preset difference condition is not met, acoustic emission signal data can continue to be acquired for continued monitoring.

[0119] If it is determined that there are damaged blades in the unit's blades, in step S806, it can be determined whether each blade meets the preset damage conditions in order to locate the damaged blade and complete the blade damage monitoring.

[0120] According to the embodiments of the present disclosure, the blade damage monitoring method for wind turbine generators can use acoustic emission monitoring technology to monitor the blade status, monitor and record acoustic signals in real time, and by analyzing and comparing these signals, when blade damage risk is identified, it can perform rapid pitch retraction and shutdown protection, effectively reducing the risk of blade breakage caused by blade damage propagation, and minimizing the risk of extreme events such as blade breakage and tower sweeping.

[0121] Here, acoustic emission monitoring technology offers several advantages: First, it enables continuous monitoring without downtime or blade disassembly. Second, it allows monitoring throughout the blade's entire lifespan, helping to predict potential damage trends, such as in monitoring the fatigue life development of wind turbine blades. Furthermore, this technology can help optimize maintenance plans, reduce repair costs, and improve the reliability and availability of wind turbines.

[0122] Furthermore, the wind turbine blade damage monitoring method according to the embodiments of this disclosure can compare the consistency of acoustic emission characteristics of the three blades based on the processing characteristics of acoustic emission signals. When the consistency difference of the three blades exceeds a threshold, the wind turbine quickly issues an alarm and shuts down, effectively reducing the risk of blade breakage caused by the propagation of blade damage and minimizing the risk of extreme events such as blade breakage and tower sweeping. In this way, real-time online protection against blade damage can be achieved.

[0123] According to a second aspect of an exemplary embodiment of the present disclosure, a blade damage monitoring system for a wind turbine generator set is provided, the wind turbine generator set including a plurality of blades 10, such as... Figure 9 As shown, the blade damage monitoring system may include a monitoring platform 910 and an acoustic emission sensor 920 configured on each blade. The monitoring platform 910 may include: a processor; and a memory for storing processor-executable instructions, wherein the processor-executable instructions, when executed by the processor, cause the processor to perform the blade damage monitoring method for wind turbine generators according to embodiments of the present disclosure.

[0124] As an example, the wind turbine generator set also includes a main control system 20. The monitoring platform 910 can send the execution results of the processor to the main control system 20, so that the main control system 20 can perform control actions corresponding to the execution results. The control actions may include, for example, controlling the wind turbine generator set to retract its paddles and / or shut down. For example, the monitoring platform 910 may include a data acquisition unit for receiving signals from acoustic emission sensors, which can send the data of the received signals to the processor.

[0125] Specifically, the acoustic emission sensor 920 can provide the acoustic emission data required to calculate all indicators; the monitoring platform 910 can perform relevant logical judgments, including acoustic emission feature processing and difference degree calculation, and output corresponding diagnostic results and control commands. The main control system 20 can receive the monitoring results from the monitoring platform 910, and can execute actions such as propeller retraction and shutdown according to the alarm or shutdown command corresponding to the monitoring results, thereby realizing real-time online protection against blade damage.

[0126] In addition, the wind farm where the wind turbine generators are located can also be equipped with an equipment management system, which can uniformly manage information such as the status of acoustic emission sensors, the status of the monitoring platform, and the blade damage status of each unit in the entire wind farm.

[0127] Here, the monitoring platform 910 may include electronic equipment, which may include the processor and memory described above, to perform the blade damage monitoring method for wind turbine generators.

[0128] As an example, an electronic device can be a PC, tablet, personal digital assistant, smartphone, or other device capable of executing the aforementioned set of instructions. Here, the electronic device is not necessarily a single device; it can be any collection of devices or circuits capable of executing the aforementioned instructions (or instruction sets) individually or in combination. The electronic device can also be part of an integrated control system or system manager, or can be configured to interconnect locally or remotely (e.g., via wireless transmission) through an interface.

[0129] In electronic devices, a processor may include a central processing unit (CPU), a graphics processing unit (GPU), a programmable logic device, a dedicated processor system, a microcontroller, or a microprocessor. By way of example and not limitation, a processor may also include analog processors, digital processors, microprocessors, multi-core processors, processor arrays, network processors, etc.

[0130] The processor can execute instructions or code stored in memory, which can also store data. Instructions and data can also be sent and received over a network via a network interface device, which can employ any known transport protocol.

[0131] Memory can be integrated with the processor; for example, RAM or flash memory can be housed within an integrated circuit microprocessor. Alternatively, memory can comprise a separate device, such as an external disk drive, storage array, or other storage device that can be used by any database system. Memory and processor can be operatively coupled, or can communicate with each other, for example, via I / O ports, network connections, etc., enabling the processor to read files stored in the memory.

[0132] In addition, electronic devices may include video displays (such as liquid crystal displays) and user interaction interfaces (such as keyboards, mice, touch input devices, etc.). All components of the electronic device may be interconnected via buses and / or networks.

[0133] According to a third aspect of the embodiments of this disclosure, a computer-readable storage medium is provided. Specifically, a blade damage monitoring method for a wind turbine generator set according to embodiments of this disclosure can be programmed into a computer program and stored on a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by at least one processor, the at least one processor is caused to perform the blade damage monitoring method for a wind turbine generator set according to exemplary embodiments of this disclosure. Examples of computer-readable storage media include: read-only memory (ROM), random access programmable read-only memory (PROM), electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), flash memory, non-volatile memory, CD-ROM, CD-R, CD+R, CD-RW, CD+RW, DVD-ROM, DVD-R, DVD+R, DVD-RW, DVD+RW, DVD-RAM, BD-ROM, BD-R, BD-R LTH, BD-RE, Blu-ray or optical disc storage, hard disk drive (HDD), solid-state drive (SSD), card storage (such as multimedia cards, secure digital (SD) cards, or ultra-fast digital (XD) cards), magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid-state drive, and any other device configured to store a computer program and any associated data, data files, and data structures in a non-transitory manner and to provide the computer program and any associated data, data files, and data structures to a processor or computer so that the processor or computer can execute the computer program. In one example, the computer program and any associated data, data files, and data structures are distributed across a networked computer system, such that the computer program and any associated data, data files, and data structures are stored, accessed, and executed in a distributed manner through one or more processors or computers.

[0134] According to a fourth aspect of the embodiments of the present disclosure, a computer program product is provided, the computer program product including computer instructions, which, when executed by a processor, can implement a blade damage monitoring method for a wind turbine generator set according to the embodiments of the present disclosure.

[0135] The specific embodiments of this disclosure have been described in detail above. Although some embodiments have been shown and described, those skilled in the art should understand that modifications and variations can be made to these embodiments without departing from the principles and spirit of this disclosure, which are defined by the claims and their equivalents. Such modifications and variations should also be within the protection scope of the claims of this disclosure.

Claims

1. A method for monitoring blade damage in a wind turbine generator set, characterized in that, The wind turbine generator set includes multiple blades, each blade being equipped with an acoustic emission sensor, wherein the blade damage monitoring method includes: Acquire the acoustic emission signals from the acoustic emission sensors on each blade; The acoustic emission signal is subjected to feature analysis to obtain the feature values ​​of each blade in multiple acoustic emission features; For each acoustic emission feature, the distribution differences among the feature values ​​of the multiple blades are analyzed to obtain the degree of difference corresponding to each acoustic emission feature; In response to the fact that the degree of difference among the plurality of acoustic emission features meets a preset difference condition, it is determined that at least one of the plurality of blades is at risk of damage, wherein the preset difference condition indicates that there are acoustic emission features among the plurality of acoustic emission features whose degree of difference exceeds a preset value. The blade damage monitoring method further includes: in response to the degree of difference of the plurality of acoustic emission characteristics satisfying the preset difference condition, determining that the blades among the plurality of blades that meet the preset damage condition have a risk of damage. The plurality of acoustic emission features include multiple frequency band energy features, each corresponding to a specific frequency band, and each frequency band energy feature characterizes the energy characteristics of its corresponding frequency band. The plurality of frequency band energy characteristics include a first energy characteristic and a second energy characteristic, wherein the minimum frequency of the frequency band corresponding to the first energy characteristic is greater than or equal to the maximum frequency of the frequency band corresponding to the second energy characteristic. The preset damage condition includes: satisfying that the characteristic value of the blade on the first energy characteristic is greater than the characteristic value of the blade on the second energy characteristic.

2. The leaf damage monitoring method according to claim 1, characterized in that, The preset damage condition indicates that the blade has the largest eigenvalue in at least one acoustic emission feature among all blades.

3. The leaf damage monitoring method according to claim 2, characterized in that, The plurality of acoustic emission features also include a plurality of signal waveform features, wherein the signal waveform features characterize the waveform feature parameters of the acoustic emission signal. The preset damage condition further includes: for more than a predetermined number of signal waveform features, the feature value of the blade is the maximum value among all the feature values ​​of the blades.

4. The leaf damage monitoring method according to claim 1, characterized in that, Both the first energy feature and the second energy feature are multiple, wherein the preset damage condition further includes: The sum of the eigenvalues ​​of the blade on the first energy characteristic is greater than a first threshold; and, The sum of the eigenvalues ​​of the blade on the second energy characteristic is less than the second threshold.

5. The leaf damage monitoring method according to claim 1 or 4, characterized in that, The plurality of frequency band energy features also includes a third energy feature, wherein the minimum frequency of the frequency band corresponding to the third energy feature is greater than or equal to the maximum frequency of the frequency band corresponding to the first energy feature. The preset damage condition also includes: the characteristic value of the blade on the third energy characteristic is not null.

6. The leaf damage monitoring method according to claim 1, characterized in that, The plurality of acoustic emission features also include a plurality of signal waveform features, wherein the signal waveform features characterize the waveform feature parameters of the acoustic emission signal. The preset difference conditions also include: The signal waveform features that satisfy the condition that the degree of difference among the plurality of signal waveform features exceeds a first preset value; and... The frequency band energy characteristics that satisfy the condition that the degree of difference among the multiple frequency band energy characteristics exceeds a second preset value.

7. The leaf damage monitoring method according to claim 6, characterized in that, The multiple frequency bands are all included within the full frequency band of the acoustic emission signal, and the frequency band energy characteristics represent the proportion of energy of the corresponding frequency band in the full frequency band. The degree of difference refers to the consistency difference among the feature values ​​of the multiple leaves, and the consistency difference characterizes the degree of dispersion among the feature values.

8. The leaf damage monitoring method according to claim 1, characterized in that, The blade damage monitoring method also includes: Before performing feature analysis on the acoustic emission signal, the signal component corresponding to the vibration frequency of the transmission chain of the wind turbine generator is removed from the acoustic emission signal to obtain the removed acoustic emission signal.

9. A blade damage monitoring system for wind turbine generators, characterized in that, The wind turbine generator set includes multiple blades, and the blade damage monitoring system includes a monitoring platform and an acoustic emission sensor configured on each blade. The monitoring platform includes: processor; Memory used to store the processor's executable instructions. Wherein, when the processor executes the executable instructions, it causes the processor to perform the blade damage monitoring method for wind turbine generators according to any one of claims 1 to 8.

10. The blade damage monitoring system according to claim 9, characterized in that, The wind turbine generator set also includes a main control system. The monitoring platform sends the execution result of the processor to the main control system so that the main control system can perform control actions corresponding to the execution result. The control actions include controlling the wind turbine generator set to retract its propellers and / or shut down.

11. A computer-readable storage medium, characterized in that, When the instructions in the computer-readable storage medium are executed by the processor of the electronic device, the electronic device is able to perform the blade damage monitoring method for a wind turbine generator according to any one of claims 1 to 8.

12. A computer program product comprising computer-executable instructions, characterized in that, When the computer-executable instructions are executed by at least one processor, they implement the blade damage monitoring method for wind turbine generators according to any one of claims 1 to 8.

Images