A wind turbine tower crack identification method in a noise scene

Abstract

The application relates to the technical field of signal processing, in particular to a wind turbine tower crack identification method in a noise scene. After amplitude increment processing is performed on the collected tower vibration signals of the wind turbine, a sliding window fitting method is used for signal fitting, the fitted signals are reconstructed into vibration signals, and then the tower of the wind turbine is cracked based on the vibration signals, so that effective identification of the tower crack of the wind turbine in the noise scene is realized. The application aims to solve the problem of effective identification of the tower crack of the wind turbine in the noise scene.

Description

Technical Field

[0001] This application relates to the field of signal processing technology, and in particular to a method for identifying cracks in the tower of a wind turbine in a noisy environment. Background Technology

[0002] Tower vibration is a significant factor affecting the structural stability of wind turbines. Even minor vibrations can cause aerodynamic instability in the entire wind turbine, while larger vibrations can directly lead to tower collapse. Cracks weaken the tower's load-bearing capacity, causing localized stress concentration. If cracks continue to propagate, they can cause the tower to tilt or even break, resulting in equipment damage and personal injury. When cracks appear in the wind turbine tower, changes in stiffness can lead to deviations and frequency drift in its natural frequency.

[0003] Traditional crack fault detection methods involve measuring the change in the circular frequency of the tower using an Euler-Bernoulli beam model to determine if a crack has appeared. However, since wind turbines typically operate in noisy environments, discrepancies often arise between the identified tower faults and the actual situation during testing.

[0004] Against this background, this application proposes a method for identifying wind turbine tower cracks in noisy environments, aiming to achieve effective identification of wind turbine tower cracks in noisy environments. Summary of the Invention

[0005] The main purpose of this application is to provide a method for identifying cracks in wind turbine towers under noisy conditions, aiming to solve the problem of effectively identifying cracks in wind turbine towers under noisy conditions.

[0006] To achieve the above objectives, this application provides a method for identifying cracks in wind turbine towers under noisy conditions, the method comprising: S10, assign a DC component to the collected wind turbine tower vibration signal, and resample the tower vibration signal into an amplitude increment signal based on the DC component; S20, determine the fitting signal corresponding to the amplitude increment signal based on the sliding window fitting method; S30, the fitted signal is reconstructed into a vibration signal to identify cracks in the tower of the wind turbine based on the vibration signal.

[0007] Optionally, in step S20, the expression for the fitted signal is: ; In the formula, For the fitted signal, k is the index of the window after the nth sliding window; W is the window width; ε is the residual; P represents the fitting order; β is the coefficient of the mth polynomial; x c (k)The center point of the window width is represented by x, which is the vibration amplitude increment signal.

[0008] Optionally, the expression for the polynomial coefficients is: ; In the formula, X (k) ∈ R W×(p+1) To design the matrix, y (k) ∈R W For observation vectors; in, ; in: ; ; In the formula, j = k, k+1...k+W-1; , ... This represents the amplitude increment signal within the same window width. , ... This represents the fitted signal.

[0009] Optionally, in step S30, the step of reconstructing the fitted signal into a vibration signal includes: The fitted signal is obtained by interpolation. Reconstructed into vibration signals : ; In the formula, This represents the fitted signal.

[0010] Optionally, in step S30, the step of identifying cracks in the tower of the wind turbine based on the vibration signal includes: For the reconstructed signal Perform a Fourier transform and identify the frequency of crack faults in the wind turbine tower from the spectrum obtained by the Fourier transform.

[0011] Optionally, in step S10, the expression for the amplitude increment signal is: ; In the formula, i = 1, 2, ..., n, x i It is an amplitude increment signal. Let α be the i-th vibration signal, and α be the DC component. Attached Figure Description

[0012] Figure 1 This is a flowchart illustrating the method for identifying wind turbine tower cracks in a noisy environment, as described in an embodiment of this application. Figure 2 This is the original signal diagram of the tower involved in the embodiments of this application; Figure 3 The amplitude increment signal involved in the embodiments of this application Figure 4 This is a vibration signal diagram reconstructed after fitting, as described in an embodiment of this application. Figure 5 This is a spectrum diagram of the reconstructed vibration signal involved in the embodiments of this application; Figure 6 This is a spectrum diagram of the original signal involved in the embodiments of this application.

[0013] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0014] To better understand the above technical solutions, exemplary embodiments of this disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of this disclosure to those skilled in the art. Example

[0015] Reference Figure 1 This embodiment provides a method for identifying wind turbine tower cracks in noisy environments. The method includes the following steps: S10, assign a DC component to the collected wind turbine tower vibration signal, and resample the tower vibration signal into an amplitude increment signal based on the DC component; This example describes a tower crack simulation signal: ; In the formula, A is the signal amplitude; ζ is the damping coefficient; f is the characteristic frequency of the tower crack; t is time; and n is the noise signal.

[0016] For example, the initial tower crack simulation signal is as follows: Figure 2 As shown.

[0017] A DC component is assigned to the collected tower vibration signal of the wind turbine, and the tower vibration signal is resampled into an amplitude increment signal based on the DC component.

[0018] It should be noted that the amplitude increment signal is the cumulative value of the vibration signal amplitude.

[0019] Optionally, the expression for the amplitude increment signal is: ; In the formula, i = 1, 2, ..., n, x i It is an amplitude increment signal. Let α be the i-th vibration signal, and α be the DC component.

[0020] For example, the obtained amplitude increment signal is as follows Figure 3 As shown.

[0021] S20, determine the fitting signal corresponding to the amplitude increment signal based on the sliding window fitting method; In this embodiment, after obtaining the amplitude increment signal, the amplitude increment signal is converted into a corresponding fitted signal based on the sliding window fitting method. The conversion steps are as follows: S21, within window Dk, establish a p-order polynomial model: ; In the formula, j = k, k+1, ..., k+W-1; k is the index value of the window after the nth sliding window; W is the window width; ε is the residual; p represents the fitting order; β is the coefficient of the mth order polynomial; xc(k) represents the center point of the window width; n = 1, 2, ..., M; S is the sliding step size. Specifically, the window width is set to W = 51, S = 3, and p = 1. S22, Solving using the least squares method, constructing the design matrix X (k) ∈ R W×(p+1) and observation vector y (k) ∈R W : ; In the formula, j = k, k+1...k+W-1; , ... This represents the magnitude increment within the j-th window. , ... This represents the fitted value within the j-th window.

[0022] The coefficient estimates are obtained by minimizing the sum of squared residuals. : ; S23, Calculate the center point x of the window. c (k) The value at that point is used as the fitted signal. : ; It should be noted that the core principle of sliding window width local fitting is to fit and denoise the high-frequency noise of the vibration signal. However, directly fitting the vibration signal may result in the failure impact being fitted out, so it needs to be converted into amplitude increment.

[0023] S30, the fitted signal is reconstructed into a vibration signal to identify cracks in the tower of the wind turbine based on the vibration signal.

[0024] Optionally, interpolation can be used to fit the signal. Reconstructed into vibration signals : ; In the formula, This represents the amplitude increment signal after interpolation.

[0025] Optionally, this embodiment also provides a method for identifying cracks in the tower of the wind turbine based on vibration signals, including: For the reconstructed signal Perform a Fourier transform and identify the frequency of tower crack faults that matches the unit from the spectrum obtained by the Fourier transform.

[0026] In some specific implementations, the reconstructed vibration signal is as follows: Figure 4 As shown, Figure 5 This is the spectrum of the reconstructed vibration signal.

[0027] Traditionally, Fourier transform of the original vibration signal is directly performed to obtain... Figure 6 As shown. (Through) Figure 6 and Figure 5 The comparison reveals that the spectrum diagram obtained using the method described in this embodiment can clearly identify the fault characteristic frequencies, while the spectrum of the original signal exhibits sideband modulation, which can easily lead to misjudgment of tower crack faults.

[0028] In the technical solution provided in this embodiment, after the amplitude increment processing of the collected wind turbine tower vibration signal is performed, the sliding window fitting method is used to fit the signal. After the fitted signal is reconstructed into a vibration signal, the wind turbine tower is cracked based on the vibration signal, thereby achieving effective identification of wind turbine tower cracks in noisy scenarios.

[0029] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0030] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for identifying cracks in wind turbine towers under noisy conditions, characterized in that, The method includes the following steps: S10, assign a DC component to the collected tower vibration signal of the wind turbine, and resample the tower vibration signal into an amplitude increment signal based on the DC component; The expression for the amplitude increment signal is: ； In the formula, i = 1, 2, ..., n, x i It is an amplitude increment signal. Let α be the i-th vibration signal, and α be the DC component. S20, Determine the fitting signal corresponding to the amplitude increment signal based on the sliding window fitting method; The expression for the fitted signal is: ； In the formula, For the fitted signal, k is the index of the window after the nth sliding window; W is the window width; ε is the residual; P represents the fitting order; β is the coefficient of the mth polynomial; x c (k) The center point of the window width is represented by x, which is the amplitude increment signal. The expression for the polynomial coefficients is: ； In the formula, X (k) ∈ R W×(p+1) To design the matrix, y (k) ∈R W For observation vectors; in, ； in: ； ； In the formula, j = k, k+1...k+W-1; , ... This represents the amplitude increment signal within the same window width. , ... This represents the fitted signal; S30, reconstruct the fitted signal into a vibration signal, and identify cracks in the tower of the wind turbine based on the vibration signal; In step S30, the step of reconstructing the fitted signal into a vibration signal includes: The fitted signal is obtained by interpolation. Reconstructed into vibration signals : ； In the formula, This represents the fitted signal; In step S30, the step of identifying cracks in the tower of the wind turbine based on the vibration signal includes: For the reconstructed signal Perform a Fourier transform and identify the frequency of tower crack faults that matches the unit from the spectrum obtained by the Fourier transform.

Images