Wind power bearing fault diagnosis method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a wind power bearing fault diagnosis method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a cycle frequency corresponding to an original vibration signal of a to-be-diagnosed wind power bearing; channelizing the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain a sub-channel signal, and determining a to-be-filtered signal based on the sub-channel signal; taking the original vibration signal as a reference signal, optimizing a filter coefficient of a frequency shift filter based on a maximum Versoria criterion to obtain an optimal filter corresponding to the original vibration signal, and performing frequency shift filtering on the to-be-filtered signal through the optimal filter to obtain a bearing fault feature signal; and analyzing the bearing fault characteristic signal to obtain a fault diagnosis result of the wind power bearing. According to the embodiment of the invention, fault diagnosis can be accurately carried out on the wind power bearing.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment, and in particular, to a method, device, electronic device, and storage medium for diagnosing faults of wind power bearings. Background Art

[0002] With the increasing demand for green clean energy in various countries around the world, wind power generation, as a mature technology in this field, has been widely applied. Wind power generation equipment mainly consists of blades, generators, bearings, gearboxes, high-speed rotating shafts, etc. Among them, the bearings that play a supporting role are often in extreme working conditions of high rotational speed and large load, and are extremely prone to failure. Therefore, a simple and effective bearing fault diagnosis method is crucial for improving the reliability of wind turbine units. Existing research shows that the fault vibration signals of bearings often exhibit second-order cyclostationary characteristics. Therefore, the waveform of the bearing fault characteristic signal can be estimated based on this characteristic, and fault diagnosis can be carried out based on the estimated waveform. When the adaptive filter parameter update algorithm in the prior art extracts the fault characteristic signal of the bearing, it performs well on the measurement signal containing Gaussian noise. However, when the signal contains impulse noise, the filter with the criterion of minimizing the cumulative mean square error is prone to divergence and failure during the parameter update process. Therefore, the fault characteristic signal of the bearing cannot be accurately extracted, and thus the bearing fault cannot be accurately diagnosed. Summary of the Invention

[0003] Embodiments of the present invention provide a method, device, electronic device, and storage medium for diagnosing faults of wind power bearings, which can accurately diagnose faults of wind power bearings.

[0004] In a first aspect, an embodiment of the present invention provides a method for diagnosing faults of wind power bearings, including:

[0005] Obtaining the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed;

[0006] Performing channelization processing on the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain sub-channel signals, and determining a signal to be filtered based on the sub-channel signals;

[0007] Taking the original vibration signal as a reference signal to optimize the filter coefficients of the frequency shift filter based on the maximum Versoria criterion to obtain an optimal filter corresponding to the original vibration signal, and performing frequency shift filtering on the signal to be filtered through the optimal filter to obtain a bearing fault characteristic signal; and

[0008] Analyzing the bearing fault characteristic signal to obtain a fault diagnosis result of the wind power bearing.

[0009] In a second aspect, an embodiment of the present invention provides a device for diagnosing faults of wind power bearings, including:

[0010] A cyclic frequency acquisition module, configured to acquire the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed;

[0011] A signal-to-be-filtered acquisition module, configured to perform channelization processing on the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain a sub-channel signal, and determine a signal to be filtered based on the sub-channel signal;

[0012] An optimization and filtering module, configured to optimize the filter coefficients of the frequency shift filter based on the maximum Versoria criterion by using the original vibration signal as a reference signal to obtain an optimal filter corresponding to the original vibration signal, and perform frequency shift filtering on the signal to be filtered through the optimal filter to obtain a bearing fault feature signal; and

[0013] A fault diagnosis result acquisition module, configured to analyze the bearing fault feature signal to obtain a fault diagnosis result of the wind power bearing.

[0014] In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the program, the wind power bearing fault diagnosis method according to any one of the embodiments of the present invention is implemented.

[0015] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the wind power bearing fault diagnosis method according to any one of the embodiments of the present invention is implemented.

[0016] A wind power bearing fault diagnosis method, device, electronic device, and storage medium provided by an embodiment of the present invention can suppress spectral aliasing interference and Gaussian and non-Gaussian noises in the original vibration signal and improve the accuracy of the extracted bearing fault feature signal of the wind power bearing by acquiring the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed, then performing channelization processing on the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain a sub-channel signal, and determining a signal to be filtered based on the sub-channel signal; further, by using the original vibration signal as a reference signal and optimizing the filter coefficients of the frequency shift filter based on the maximum Versoria (variable) criterion, the method can accurately extract the fault feature signal of the generator bearing of the wind turbine from a signal containing band aliasing interference and noise without prior information of the reference signal, directly perform waveform estimation on the fault feature signal from the time domain perspective, which is beneficial to improving the accuracy and reliability of the diagnosis result, enabling the extraction of the bearing fault feature signal of the wind power bearing to be convenient and efficient, and further enabling efficient, convenient, and accurate fault diagnosis of the wind power bearing and improving the reliability of the wind turbine. Description of the Drawings

[0017] To more clearly illustrate the technical solution of the present invention, the accompanying drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0018] Figure 1 It is a schematic flowchart of a wind power bearing fault diagnosis method provided by an embodiment of the present invention;

[0019] Figure 2 It is another schematic flowchart of a wind power bearing fault diagnosis method provided by an embodiment of the present invention;

[0020] Figure 3 It is another schematic flowchart of a wind power bearing fault diagnosis method provided by an embodiment of the present invention;

[0021] Figure 4 It is another schematic flowchart of a wind power bearing fault diagnosis method provided by an embodiment of the present invention;

[0022] Figure 5 It is a convergence schematic diagram of the frequency shift filter in the sense of mean square error in the wind power bearing fault diagnosis method provided by an embodiment of the present invention;

[0023] Figure 6 It is a schematic structural diagram of a wind power bearing fault diagnosis device provided by an embodiment of the present invention;

[0024] Figure 7 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0025] In order to enable those skilled in the art of the present technology to better understand the present invention solution, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0026] It should be noted that the terms "first", "second", etc. in the specification, claims and above-mentioned drawings of the present invention are used to distinguish similar objects and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0027] Figure 1 FIG. 0 is a schematic flowchart of a wind power bearing fault diagnosis method provided by an embodiment of the present invention. This method can be executed by a wind power bearing fault diagnosis device provided by an embodiment of the present invention, and the device can be implemented in a software and / or hardware manner. In a specific embodiment, the device can be integrated in an electronic device, such as a computer, a server, etc. The following embodiments will be described by taking the integration of the device in an electronic device as an example. Refer to Figure 1 , and the method specifically may include the following steps:

[0028] Step 101, obtain the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed. This step is conducive to channelizing the original vibration signal based on the cyclic frequency and performing subsequent frequency shift filtering processing.

[0029] Specifically, a real-time vibration analog signal can be collected by installing a vibration sensor on the wind power bearing and transmitted to an analog-to-digital conversion device for analog-to-digital conversion to obtain the above-mentioned original vibration signal.

[0030] Specifically, the process of obtaining the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed can be carried out by any one or more of the following methods: the direct estimation method based on the cyclic autocorrelation function, the spectral analysis method based on the fast Fourier transform (FFT), the time-frequency analysis method based on wavelet transform, and the strip spectrum correlation estimator method.

[0031] Optionally, the above-mentioned original vibration signal can be denoted as x[n], which consists of three parts, specifically:

[0032] x[n]=d[n]+i[n]+v[n]

[0033] In the formula, d[n] represents the fault feature signal; i[n] represents the frequency band aliasing interference; v[n] represents the noise.

[0034] The desired signal and the band aliasing interference are both rectangular pulse binary phase shift keying signals with known cyclic statistical characteristics.

[0035] Step 102: Channelize the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain sub-channel signals, and determine the signal to be filtered based on the sub-channel signals. This step can help suppress the spectral aliasing interference and Gaussian and non-Gaussian noises in the original vibration signal, so as to improve the accuracy of the extracted fault feature signals of the wind power bearing.

[0036] Optionally, the cyclic frequency includes multiple cyclic frequencies, and the frequency shift filter is a multi-channel frequency shift filter. The number of channels of the multi-channel frequency shift filter can be equal to or different from the number of cyclic frequencies.

[0037] Optionally, the process of channelizing the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain sub-channel signals and determining the signal to be filtered based on the sub-channel signals includes: Channelize the current input signal segment through an optimized frequency shift filter based on the cyclic frequency to obtain multiple sub-channel signal segments corresponding to the current input signal segment, and splice the corresponding multiple sub-channel signal segments to obtain the signal segment to be filtered corresponding to the current input signal segment.

[0038] Specifically, before channelizing the current input signal segment through an optimized frequency shift filter based on the cyclic frequency to obtain multiple sub-channel signal segments corresponding to the current input signal segment, the normalized frequency offset parameter of the frequency shift filter can be determined based on the cyclic frequency.

[0039] Specifically, the cyclic frequency can be determined as the normalized frequency offset parameter, or after calibrating the cyclic frequency, the calibrated cyclic frequency can be determined as the normalized frequency offset parameter.

[0040] Specifically, the process of determining the signal to be filtered based on the sub-channel signals can include: Splicing the sub-channel signals or calibrating the sub-channel signals with a preset weight coefficient and then splicing them to obtain the signal to be filtered.

[0041] Step 103: Use the original vibration signal as a reference signal to optimize the filter coefficients of the frequency shift filter based on the maximum Versoria criterion to obtain the optimal filter corresponding to the original vibration signal, and perform frequency shift filtering on the signal to be filtered through the optimal filter to obtain the bearing fault feature signal. This step can accurately extract the fault feature signals of the generator bearings of the wind turbine from the signal containing band aliasing interference and noise without prior information of the reference signal, directly estimate the waveform of the fault feature signal from the time domain perspective, which is beneficial to improving the accuracy and reliability of the diagnosis results, and enables the extracted fault feature signals of the wind power bearing to be carried out conveniently and efficiently.

[0042] Optionally, the process of using the original vibration signal as a reference signal to optimize the filter coefficients of the frequency shift filter based on the maximum Versoria criterion to obtain the optimal filter corresponding to the original vibration signal includes:

[0043] Performing multiple rounds of iteration on the filter coefficients based on the current input signal segment and the pre-established maximum cost function until the value of the corresponding maximum cost function meets the preset requirements, and obtaining the corresponding filter coefficients as the optimal filter coefficients.

[0044] Specifically, after the value of the maximum cost function meets the preset requirements, the iteration can also be terminated after continuing to iterate for the first preset number of times, or the iteration can be directly terminated after iterating for the second preset number of times.

[0045] Optionally, the process of performing frequency shift filtering on the signal to be filtered through the optimal filter to obtain the bearing fault feature signal includes:

[0046] Determining the complex exponential sequence of the frequency shift filter based on the cyclic frequency, where the length of the complex exponential sequence is equal to the length of the signal to be filtered; and multiplying each element of the complex exponential sequence by the signal to be filtered, convolving the multiplication result with the optimal filter coefficients, and determining the bearing fault feature signal based on the convolution result.

[0047] Step 104, analyzing the bearing fault feature signal to obtain the fault diagnosis result of the wind power bearing. Based on Steps 101 to 103, this step can suppress the spectral aliasing interference and Gaussian and non-Gaussian noises in the original vibration signal by obtaining the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed, then performing channelization processing on the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain sub-channel signals, and determining the signal to be filtered based on the sub-channel signals, improving the accuracy of the extracted bearing fault feature signal of the wind power bearing; further, in the embodiment of the present invention, by using the original vibration signal as a reference signal and optimizing the filter coefficients of the frequency shift filter based on the maximum Versoria criterion, the fault feature signal of the generator bearing of the wind turbine can be accurately extracted from the signal containing frequency band aliasing interference and noise without prior information of the reference signal, directly performing waveform estimation on the fault feature signal from the time domain perspective, which is beneficial to improving the accuracy and reliability of the diagnosis result, enabling the extraction of the bearing fault feature signal of the wind power bearing to be carried out conveniently and efficiently, and thus being able to diagnose the fault of the wind power bearing efficiently, conveniently and accurately, and improving the reliability of the wind turbine.

[0048] Optionally, the process of analyzing the bearing fault feature signal to obtain the fault diagnosis result of the wind power bearing includes: performing cyclic stationary analysis on the above-mentioned bearing fault feature signal to obtain the fault diagnosis result of the above-mentioned wind power bearing.

[0049] Specifically, the process of performing cyclic stationary analysis on the above-mentioned bearing fault characteristic signals to obtain the fault diagnosis result of the above-mentioned wind power bearing may include: successfully detecting the characteristic frequency components of the bearing fault characteristic signals by applying cyclic stationary analysis, and determining the above-mentioned fault diagnosis result based on the characteristic frequencies.

[0050] Specifically, the characteristic frequency components can be matched with the theoretical characteristic frequencies of different fault types, and the fault diagnosis result can be determined based on the matching result.

[0051] Specifically, statistical parameter analysis, waveform analysis, etc. can also be performed on the above-mentioned bearing fault characteristic signals to obtain the fault diagnosis result of the above-mentioned wind power bearing.

[0052] The following further introduces the wind power bearing fault diagnosis method provided by the embodiments of the present invention. As Figure 2 shown, that is Figure 1 step 101 in

[0053] can include the following steps:

[0054] In a specific example, the length of the input signal x[n] is N orig , the length of the sliding window for performing sliding window processing is N′, the overlapping area between the sliding windows is N′ - 1, and the number of times of performing sliding window processing is P.

[0055] Step 1012, arranging the column vectors corresponding to the windowed signals in time order to obtain a first signal matrix, and performing a fast Fourier transform on each column of the first signal matrix to obtain a second signal matrix.

[0056] In a specific example, both the above-mentioned first signal matrix and the second signal matrix are matrices of N′×P.

[0057] Step 1013, repeatedly expanding the second signal matrix to have the same number of columns as the length of the original vibration signal to obtain a third signal matrix, and multiplying the third signal matrix conjugate by the original vibration signal to obtain a fourth signal matrix.

[0058] Step 1014, performing an inverse fast Fourier transform on each row of the fourth signal matrix to obtain a fifth signal matrix.

[0059] Specifically, the above-mentioned fifth signal matrix is the spectral correlation matrix of the original signal x[n].

[0060] In a specific example, the above-mentioned third signal matrix, fourth signal matrix, and fifth signal matrix are all matrices of N′×N orig of.

[0061] Step 1015: Take the modulus square of each element of the fifth signal matrix to obtain the sixth signal matrix, and sum the elements of each column of the sixth signal matrix to obtain the signal cyclic information vector.

[0062] In a specific example, for the fifth signal matrix of N′×N orig take the modulus square of each element to obtain the sixth signal matrix of N′×N orig and sum the elements of each column thereof, then an N orig -dimensional signal cyclic information vector can be obtained.

[0063] Step 1016: Determine the cyclic frequency based on the signal cyclic information vector.

[0064] Optionally, the process of determining the cyclic frequency based on the signal cyclic information vector includes: obtaining the target frequencies corresponding to one or more maximum vector elements in the signal cyclic information vector, and determining the cyclic frequency based on the target frequencies.

[0065] Specifically, the above target frequency can be determined as the cyclic frequency, or the above target frequency and the integer multiple frequencies of the target frequency can be determined as the cyclic frequency.

[0066] Specifically, the set of obtained cyclic frequencies can be expressed as A={α1,α2,…,α M}.

[0067] In a specific example, obtain the target frequencies 1 / 5, 1 / 6 and 1 / 7 corresponding to the three maximum vector elements in the signal cyclic information vector, and determine the three target frequencies as the cyclic frequency.

[0068] The embodiments of the present invention can accurately obtain the cyclic frequency corresponding to the original vibration signal.

[0069] Next, the wind power bearing fault diagnosis method provided by the embodiments of the present invention will be further introduced.

[0070] Optionally, the number of cyclic frequencies of multiple cyclic frequencies is equal to the number of channels of the multi-channel frequency shift filter.

[0071] In a specific example, both the number of cyclic frequencies and the number of channels are M, and the length of each channel filter of the multi-channel frequency shift filter is L, and the filter coefficients can be expressed as:

[0072] h n,m =[h n,m [1],h n,m [2],…,h n,m [L]] T ,m=1,…,M

[0073] Such asFigure 3 As shown in Figure 3 , the wind power bearing fault diagnosis method provided by the embodiments of the present invention may include the following steps:

[0074] Step 301: Obtain the cyclic frequencies corresponding to the original vibration signals of the wind power bearing to be diagnosed.

[0075] Step 302: Set each cyclic frequency as the normalized frequency offset parameter of one channel of the multi-channel frequency shift filter to obtain the frequency shift filter to be optimized.

[0076] Step 303: Based on the cyclic frequencies, perform channelization processing on the current input signal segment through the frequency shift filter to be optimized to obtain multiple sub-channel signal segments corresponding to the current input signal segment, and splice the corresponding multiple sub-channel signal segments to obtain the filtered signal segment corresponding to the current input signal segment.

[0077] In a specific example, each of the above sub-channel signal segments can be expressed as:

[0078]

[0079] The above filtered signal segment has a total length of: N in = M × L, and can be expressed as:

[0080]

[0081] Step 304: Perform frequency shift filtering on the filtered signal segment corresponding to the current input signal segment through the frequency shift filter to be optimized to obtain the fault feature signal segment of the current input signal segment.

[0082] Optionally, the process of performing frequency shift filtering on the filtered signal segment corresponding to the current input signal segment through the frequency shift filter to be optimized to obtain the fault feature signal segment of the current input signal segment includes:

[0083] Determine the channel complex exponential sequences of each channel of the frequency shift filter to be optimized based on one of the cyclic frequencies respectively.

[0084] Splice the channel complex exponential sequences of each channel to obtain the composite frequency shift complex exponential sequence; and

[0085] Multiply each element of the composite frequency shift complex exponential sequence by the filtered segment corresponding to the current input signal segment, and perform convolution on the multiplication result with the optimal filter coefficients to obtain the fault feature signal corresponding to the current input signal segment.

[0086] In a specific example, the channel complex exponential sequence of one channel of the multi-channel filter, based on the cyclic frequency α m is determined as:

[0087]

[0088] After splicing the complex exponential sequences of each channel, a composite frequency-shifted complex exponential sequence with a length of N in = M×L can be obtained, which can be expressed as:

[0089]

[0090] It can be understood that the optimal filter coefficients of the above frequency-shift filter are composed of the optimal filter coefficients h n,m of the single filters of each channel, and can be specifically expressed as:

[0091]

[0092] Specifically, the above process of multiplying the composite frequency-shifted complex exponential sequence element by element with the to-be-filtered segment corresponding to the current input signal segment, and convolving the multiplication result with the optimal filter coefficients to obtain the fault feature signal corresponding to the current input signal segment can be carried out based on the following formula:

[0093]

[0094] In the formula, represents the fault feature signal, h n,m [n] represents the filter coefficients, x[n] represents the input signal, represents the composite frequency-shifted complex exponential sequence, H represents the conjugate transpose; ⊙ represents the element-by-element product.

[0095] Step 305: Take the current input signal segment as the corresponding reference signal, determine the optimization filter coefficients of the to-be-optimized frequency-shift filter corresponding to the current input signal segment based on the maximum Versoria criterion, and determine the optimal filter coefficients of the optimal filter based on the optimization filter coefficients corresponding to the current input signal segment based on the maximum Versoria criterion.

[0096] Optionally, the process of determining the optimal filter coefficients of the optimal filter based on the optimization filter coefficients corresponding to the current input signal segment based on the maximum Versoria criterion includes:

[0097] Based on the function value of the pre-established first maximum cost function, judge whether the optimization filter coefficients of the to-be-optimized frequency-shift filter corresponding to the current input signal segment are the optimal filter coefficients, and when the judgment result is no, start to execute step 305 for the next input signal segment, where the function value of the first maximum cost function is established based on the generalized Versoria function.

[0098] Specifically, when the function value of the first maximum cost function falls within the acceptable range, it can be determined that the optimized filter coefficient of the frequency shift filter corresponding to the current input signal segment is the optimal filter coefficient; otherwise, it is determined that the optimized filter coefficient of the frequency shift filter corresponding to the current input signal segment is not the optimal filter coefficient.

[0099] In a specific example, the expression of the above-mentioned generalized Versoria function is as follows:

[0100]

[0101] Specifically, the above-mentioned first maximum cost function can be expressed as:

[0102]

[0103] where J GMVC (h n-1 ) represents the function value of the first maximum cost function corresponding to the filter coefficient after the (n - 1)-th round of iterative optimization, and τ and p represent the amplitude adjustment parameter and the shape control parameter of the first maximum cost function.

[0104] Step 306: Perform frequency shift filtering on other signals to be filtered corresponding to the remaining signal segments of the original vibration signal through the optimal filter to obtain the bearing fault feature signal.

[0105] Step 307: Analyze the bearing fault feature signal to obtain the fault diagnosis result of the wind power bearing.

[0106] This step can further help improve the accuracy of extracting the fault feature signal of the wind power bearing.

[0107] The following further describes the wind power bearing fault diagnosis method provided by the embodiments of the present invention.

[0108] Optionally, as Figure 4 shown, that is, Figure 3 Step 304 in

[0109] can include: Based on the amplitude difference between the current input signal segment and the corresponding fault feature signal segment, through the pre-established second maximum cost function, iteratively optimize the filter coefficient of the frequency shift filter to obtain the optimized filter coefficient of the frequency shift filter to be optimized corresponding to the current input signal segment; where the second maximum cost function is established based on the generalized Versoria function.

[0110]

[0111] Among them, e[n] represents the amplitude difference, and x[n] represents the input signal. represents the fault feature signal.

[0112] Optionally, when performing the nth round of iterative optimization on the filter coefficients, the optimization steps are as Figure 4 shown, that is, step 304 can specifically include:

[0113] Step 3041: Based on the step size parameter in the (n - 1)th round of iterative optimization and the function value of the second maximum cost function corresponding to the filter coefficients after the (n - 1)th iterative optimization, determine the step size parameter for the nth round of iterative optimization.

[0114] In a specific example, the gradient of the aforementioned first maximum cost function can be expressed as:

[0115]

[0116] Then the adaptive iteration formula of the filter coefficients can be expressed as:

[0117]

[0118] In the formula, represents the filter coefficients after the nth round of iteration, represents the filter coefficients after the (n - 1)th round of iteration, and μ n represents the step size parameter.

[0119] In a specific example, the above-mentioned second maximum cost function can be expressed as:

[0120]

[0121] In the formula, J' GMVC (h n-1 ) represents the function value of the second maximum cost function corresponding to the filter coefficients after the (n - 1)th round of iterative optimization; γ and s respectively represent the amplitude adjustment parameter and shape control parameter of the second maximum cost function.

[0122] In a specific example, the process of performing the nth round of optimization on the filter coefficients based on the step size parameter of the nth round of iterative optimization includes determining the step size parameter for the nth round of iterative optimization based on the following formula:

[0123]

[0124] In the formula, μ n ′ = αμ n-1 + β(J' GMVC (h n-1 )) 2 , 0 < α < 1, β > 0,, μ nRepresents the step-size parameter for the n-th round of iterative optimization, μ n-1 Represents the step-size constant for the (n - 1)-th round of iterative optimization, J' GMVC (h n-1 ) represents the function value of the second-largest cost function corresponding to the filter coefficients after the (n - 1)-th round of iterative optimization.

[0125] Step 3042: Based on the step-size parameter for the n-th round of iterative optimization, perform the n-th round of optimization on the filter coefficients, where n is a natural number not less than 2.

[0126] In a specific example, such as Figure 5 shown, regardless of whether the noise shown in Figure 5 a is Gaussian noise or non-Gaussian impulse noise, at different signal-to-noise ratio levels, the designed frequency-shift filter achieves fast convergence in terms of the mean square error MSE[n], where the mean square error MSE[n] can be calculated by the following formula:

[0127]

[0128] where d n represents the reference signal, which can specifically be the original vibration signal in the embodiments of the present invention. represents the fault feature signal.

[0129] In the embodiments of the present invention, by adaptively iteratively optimizing the filter coefficients of the frequency-shift filter with a variable step size, the filter coefficients of the frequency-shift filter can be made to converge quickly, and it can be applied to the scenario of real-time fault diagnosis.

[0130] Figure 6 is a structural diagram of a wind power bearing fault diagnosis device provided by the embodiments of the present invention. This device is applicable to performing the wind power bearing fault diagnosis provided by the embodiments of the present invention. Such as Figure 6 shown, this device may specifically include:

[0131] A cyclic frequency acquisition module 601, configured to acquire the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed. This can facilitate channelizing the original vibration signal based on the cyclic frequency and performing subsequent frequency-shift filtering processing.

[0132] Optionally, the above loop frequency acquisition module 601 can specifically be configured to perform a sliding window processing on the original vibration signal, and convert each acquired windowed signal into a column vector; arrange the column vectors corresponding to the windowed signals in chronological order to obtain a first signal matrix, and perform a fast Fourier transform on each column of the first signal matrix to obtain a second signal matrix; repeatedly expand the second signal matrix to have the same number of columns as the length of the original vibration signal to obtain a third signal matrix, and perform a conjugate multiplication of the third signal matrix and the original vibration signal to obtain a fourth signal matrix; perform an inverse fast Fourier transform on each row of the fourth signal matrix to obtain a fifth signal matrix; perform a modulus square operation on each element of the fifth signal matrix to obtain a sixth signal matrix, and sum the elements of each column of the sixth signal matrix to obtain a signal cycle information vector; and determine the loop frequency based on the signal cycle information vector.

[0133] The signal to be filtered acquisition module 602 is configured to perform a channelization processing on the original vibration signal through a frequency shift filter based on the loop frequency to obtain a sub-channel signal, and determine the signal to be filtered based on the sub-channel signal. This can help suppress the spectral aliasing interference and Gaussian and non-Gaussian noises in the original vibration signal, so as to improve the accuracy of the extracted fault feature signal of the wind power bearing.

[0134] Optionally, the loop frequency includes multiple loop frequencies.

[0135] Optionally, the frequency shift filter is a multi-channel frequency shift filter.

[0136] Optionally, the number of loop frequencies of the multiple loop frequencies is equal to the number of channels of the multi-channel frequency shift filter.

[0137] The optimization and filtering module 603 is configured to use the original vibration signal as a reference signal to optimize the filter coefficients of the frequency shift filter based on the maximum Versoria criterion to obtain an optimal filter corresponding to the original vibration signal, and perform a frequency shift filtering on the signal to be filtered through the optimal filter to obtain a bearing fault feature signal. This can accurately extract the fault feature signal of the generator bearing of the wind turbine from the signal containing frequency band aliasing interference and noise without prior information of the reference signal, directly perform waveform estimation on the fault feature signal from the time domain perspective, which is beneficial to improving the accuracy and reliability of the diagnosis result, and enables the extracted fault feature signal of the wind power bearing to be conveniently and efficiently obtained.

[0138] The fault diagnosis result acquisition module 604 is used to analyze the bearing fault characteristic signal to obtain the fault diagnosis result of the wind power bearing. In combination with modules 601 to 603, by obtaining the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed, and then based on the cyclic frequency, channelizing the original vibration signal through a frequency shift filter to obtain sub-channel signals, and determining the signal to be filtered based on the sub-channel signals, it is possible to suppress the spectral aliasing interference and Gaussian and non-Gaussian noises in the original vibration signal, and improve the accuracy of the extracted fault characteristic signal of the wind power bearing; in the embodiment of the present invention, further by using the original vibration signal as a reference signal and optimizing the filter coefficients of the frequency shift filter based on the maximum Versoria criterion, it is possible to accurately extract the fault characteristic signal of the generator bearing of the wind turbine from the signal containing band aliasing interference and noise without prior information of the reference signal, directly estimate the waveform of the fault characteristic signal from the time domain perspective, which is beneficial to improving the accuracy and reliability of the diagnosis result, enabling the extracted fault characteristic signal of the wind power bearing to be carried out conveniently and efficiently, and further enabling the fault diagnosis of the wind power bearing to be carried out efficiently, conveniently and accurately, and improving the reliability of the wind turbine.

[0139] Optionally, the above-mentioned fault diagnosis result acquisition module 604 can specifically be used to determine the complex exponential sequence of the frequency shift filter based on the cyclic frequency, and the length of the complex exponential sequence is equal to the length of the signal to be filtered; and

[0140] multiply the complex exponential sequence and the signal to be filtered element by element, convolve the multiplication result with the optimal filter coefficients, and determine the bearing fault characteristic signal based on the convolution result.

[0141] Optionally, the wind power bearing fault diagnosis device provided by the embodiment of the present invention further includes: a normalized frequency offset parameter setting module, which is used to set each cyclic frequency as the normalized frequency offset parameter of a channel of a multi-channel frequency shift filter to obtain the frequency shift filter to be optimized.

[0142] Optionally, the aforementioned signal to be filtered acquisition module 602 can specifically be used to channelize the current input signal segment through the frequency shift filter to be optimized based on the cyclic frequency to obtain a plurality of sub-channel signal segments corresponding to the current input signal segment, and splice the corresponding plurality of sub-channel signal segments to obtain the signal to be filtered segment corresponding to the current input signal segment.

[0143] Optionally, the aforementioned optimization and filtering module 603 can specifically be used to perform frequency shift filtering on the signal to be filtered segment corresponding to the current input signal segment through the frequency shift filter to be optimized to obtain the fault characteristic signal segment of the current input signal segment;

[0144] Taking the current input signal segment as the corresponding reference signal, determining the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment based on the maximum Versoria criterion, and determining the optimal filter coefficients of the optimal filter based on the maximum Versoria criterion according to the optimized filter coefficients corresponding to the current input signal segment.

[0145] Optionally, the foregoing optimization and filtering module 603 can specifically be used to determine whether the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment are the optimal filter coefficients based on the function value of the pre-established first maximum cost function, and when the judgment result is no, start executing for the next input signal segment by taking the current input signal segment as the corresponding reference signal, determining the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment based on the maximum Versoria criterion, and determining the optimal filter coefficients of the optimal filter based on the maximum Versoria criterion according to the optimized filter coefficients corresponding to the current input signal segment;

[0146] Wherein, the function value of the first maximum cost function is established based on the generalized Versoria function.

[0147] Optionally, the foregoing optimization and filtering module 603 can specifically be used to iteratively optimize the filter coefficients of the frequency shift filter through the pre-established second maximum cost function based on the amplitude difference between the current input signal segment and the corresponding fault feature signal segment to obtain the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment; wherein the second maximum cost function is established based on the generalized Versoria function.

[0148] Optionally, the foregoing optimization and filtering module 603 can specifically be used to determine the step size parameter for the nth round of iterative optimization based on the step size parameter during the (n - 1)th round of iterative optimization and the function value of the second maximum cost function corresponding to the filter coefficients after the (n - 1)th iterative optimization; and perform the nth round of optimization on the filter coefficients based on the step size parameter for the nth round of iterative optimization, where n is a natural number not less than 2.

[0149] Those skilled in the art can clearly understand that for the convenience and conciseness of description, only the above-mentioned division of each functional module is used as an example for illustration. In practical applications, the above-mentioned functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the above-described functional modules can refer to the corresponding process in the foregoing method embodiments and will not be elaborated here.

[0150] An embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the method for diagnosing faults of a wind power bearing provided in any of the above embodiments is implemented.

[0151] An embodiment of the present invention further provides a computer-readable medium, on which a computer program is stored. When the program is executed by a processor, the method for diagnosing faults of a wind power bearing provided in any of the above embodiments is implemented.

[0152] An embodiment of the present invention further provides a computer program product, including a computer program, which when executed by a processor, implements the method for diagnosing faults of a wind power bearing as described in any of the embodiments of the present invention.

[0153] Next, refer to Figure 7 , which shows a schematic structural diagram of a computer system 700 of an electronic device suitable for implementing the embodiments of the present invention. Figure 7 The electronic device shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present invention.

[0154] As Figure 7 shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 702 or the program loaded from the storage section 708 into the random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, ROM 702, and RAM 703 are connected to each other via a bus 704. The input / output (I / O) interface 705 is also connected to the bus 704.

[0155] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as required. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as required, so that the computer program read from it can be installed into the storage section 708 as required.

[0156] In particular, according to the embodiments disclosed in the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present invention include a computer program product that includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through the communication section 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-described functions defined in the system of the present invention are executed.

[0157] It should be noted that the computer-readable medium shown in the present invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device. In the present invention, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0158] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, as well as combinations of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0159] The modules and / or units involved in the embodiments of the present invention can be implemented in software or in hardware. The described modules and / or units can also be provided in a processor. For example, it can be described as: a processor includes a cyclic frequency acquisition module, a signal to be filtered acquisition module, an optimization and filtering module, and a fault diagnosis result acquisition module; or it can be described as: a processor includes an acquisition module, a query module, an identification module, and a determination module. Among them, the names of these modules do not, in some cases, constitute a limitation on the module itself.

[0160] As another aspect, the present invention also provides a computer-readable medium. The computer-readable medium can be included in the device described in the above embodiments; or it can exist separately and not be assembled into the device. The above computer-readable medium carries one or more programs. When the one or more programs are executed by the device, the device is caused to: acquire the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed; perform channelization processing on the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain a sub-channel signal, and determine the signal to be filtered based on the sub-channel signal; use the original vibration signal as a reference signal to optimize the filter coefficients of the frequency shift filter based on the maximum Versoria criterion to obtain an optimal filter corresponding to the original vibration signal, and perform frequency shift filtering on the signal to be filtered through the optimal filter to obtain a bearing fault feature signal; and analyze the bearing fault feature signal to obtain a fault diagnosis result of the wind power bearing.

[0161] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A method for diagnosing faults in a wind power bearing, characterized in that, Including: Obtaining the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed; Based on the cyclic frequency, performing channelization processing on the original vibration signal through a frequency shift filter to obtain sub-channel signals, and determining the signal to be filtered based on the sub-channel signals; Using the original vibration signal as a reference signal, optimizing the filter coefficients of the frequency shift filter based on the maximum Versoria criterion to obtain the optimal filter corresponding to the original vibration signal, and performing frequency shift filtering on the signal to be filtered through the optimal filter to obtain the bearing fault feature signal; and Analyzing the bearing fault feature signal to obtain the fault diagnosis result of the wind power bearing.

2. The wind power bearing fault diagnosis method according to claim 1, wherein The obtaining the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed includes: Performing sliding windowing processing on the original vibration signal, and converting each obtained windowed signal into a column vector; Arranging the column vectors corresponding to the windowed signals in time sequence to obtain a first signal matrix, and performing fast Fourier transform on each column of the first signal matrix to obtain a second signal matrix; Repeatedly expanding the second signal matrix to have the same number of columns as the length of the original vibration signal to obtain a third signal matrix, and performing conjugate multiplication on the third signal matrix and the original vibration signal to obtain a fourth signal matrix; Performing inverse fast Fourier transform on each row of the fourth signal matrix to obtain a fifth signal matrix; Taking the square of the modulus of each element of the fifth signal matrix to obtain a sixth signal matrix, and summing the elements of each column of the sixth signal matrix to obtain a signal cyclic information vector; and Determining the cyclic frequency based on the signal cyclic information vector.

3. The wind power bearing fault diagnosis method according to claim 1, characterized in that The cyclic frequency includes multiple cyclic frequencies, the frequency shift filter is a multi-channel frequency shift filter, and the number of cyclic frequencies and the number of channels of the multi-channel frequency shift filter are equal; Before performing channelization processing on the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain sub-channel signals and determining the signal to be filtered based on the sub-channel signals, the method further includes: Setting each cyclic frequency as the normalized frequency offset parameter of a channel of the multi-channel frequency shift filter to obtain the frequency shift filter to be optimized; The performing channelization processing on the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain sub-channel signals and determining the signal to be filtered based on the sub-channel signals includes: Based on the cyclic frequency, performing channelization processing on the current input signal segment through the frequency shift filter to be optimized to obtain multiple sub-channel signal segments corresponding to the current input signal segment, and splicing the corresponding multiple sub-channel signal segments to obtain the signal to be filtered segment corresponding to the current input signal segment; The using the original vibration signal as a reference signal, optimizing the filter coefficients of the frequency shift filter based on the maximum Versoria criterion to obtain the optimal filter corresponding to the original vibration signal includes: Performing frequency shift filtering on the signal to be filtered segment corresponding to the current input signal segment through the frequency shift filter to be optimized to obtain the fault feature signal segment of the current input signal segment; Taking the current input signal segment as the corresponding reference signal, determining the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment based on the maximum Versoria criterion, and determining the optimal filter coefficients of the optimal filter based on the optimized filter coefficients corresponding to the current input signal segment based on the maximum Versoria criterion.

4. The wind power bearing fault diagnosis method according to claim 3, wherein The determining the optimal filter coefficients of the optimal filter based on the optimized filter coefficients corresponding to the current input signal segment based on the maximum Versoria criterion includes: Based on the function value of the pre-established first maximum cost function, determining whether the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment are the optimal filter coefficients, and when the judgment result is no, starting to execute taking the current input signal segment as the corresponding reference signal, determining the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment based on the maximum Versoria criterion, and determining the optimal filter coefficients of the optimal filter based on the optimized filter coefficients corresponding to the current input signal segment based on the maximum Versoria criterion for the next input signal segment; Wherein, the function value of the first maximum cost function is established based on the generalized Versoria function.

5. The wind power bearing fault diagnosis method according to claim 3, characterized in that, The taking the current input signal segment as the corresponding reference signal, determining the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment based on the maximum Versoria criterion includes: Based on the amplitude difference between the current input signal segment and the corresponding fault feature signal segment, iteratively optimizing the filter coefficients of the frequency shift filter through the pre-established second maximum cost function to obtain the optimized filter coefficients of the frequency shift filter to be optimized corresponding to the current input signal segment; Wherein the second maximum cost function is established based on the generalized Versoria function.

6. The wind power bearing fault diagnosis method according to claim 5, wherein, When performing the nth round of iterative optimization on the filter coefficients, the iteratively optimizing the filter coefficients of the frequency shift filter based on the amplitude difference between the current input signal segment and the corresponding fault feature signal segment through the pre-established second maximum cost function includes: Determining the step size parameter of the nth round of iterative optimization based on the step size parameter in the (n - 1)th round of iterative optimization and the function value of the second maximum cost function corresponding to the filter coefficients after the (n - 1)th iterative optimization; and Performing the nth round of optimization on the filter coefficients based on the step size parameter of the nth round of iterative optimization, where n is a natural number not less than 2.

7. The wind power bearing fault diagnosis method according to claim 1, characterized in that The obtaining the bearing fault feature signal by performing frequency shift filtering on the signal to be filtered through the optimal filter includes: Determining the complex exponential sequence of the frequency shift filter based on the cyclic frequency, and the length of the complex exponential sequence is equal to the length of the signal to be filtered; and Performing element-wise multiplication on the complex exponential sequence and the signal to be filtered, convolving the multiplication result with the optimal filter coefficients, and determining the bearing fault feature signal based on the convolution result.

8. A wind power bearing fault diagnosis device, characterized in that, Includes: A cyclic frequency acquisition module, configured to acquire the cyclic frequency corresponding to the original vibration signal of the wind power bearing to be diagnosed; A signal to be filtered acquisition module, configured to perform channelization processing on the original vibration signal through a frequency shift filter based on the cyclic frequency to obtain sub-channel signals, and determine a signal to be filtered based on the sub-channel signals; An optimization and filtering module, configured to use the original vibration signal as a reference signal to optimize the filter coefficients of the frequency shift filter based on the maximum Versoria criterion to obtain an optimal filter corresponding to the original vibration signal, and perform frequency shift filtering on the signal to be filtered through the optimal filter to obtain a bearing fault feature signal; And A fault diagnosis result acquisition module, configured to analyze the bearing fault feature signal to obtain a fault diagnosis result of the wind power bearing.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the wind power bearing fault diagnosis method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the wind power bearing fault diagnosis method according to any one of claims 1 to 7.

Images