A method, system, device and medium for detecting an anchorage end of a cable

Abstract

The application discloses a cable anchorage end detection method, system, device and medium, and relates to the field of structural health monitoring and nondestructive testing.The method comprises the following steps: decomposing guided wave reflection signals by using an adaptive noise complete set empirical mode decomposition method; filtering high-frequency parts in the decomposition results by using a denoising method, and reconstructing the filtered signals and low-frequency parts; determining characteristic data of the reconstructed signals; comparing the characteristic data of all anchorage ends, and dividing all the anchorage ends into healthy anchorage ends and anchorage ends with damage; extracting the center frequencies of the guided wave reflection signals of each anchorage end, and dividing all the anchorage ends into healthy anchorage ends and anchorage ends with damage according to the degree of deviation of the center frequencies from the excitation frequency; and if the two classification results are consistent, the final classification result is obtained.The application can process and analyze low signal-to-noise ratio anchorage end guided wave reflection signals under the condition of no reference, and realize the detection of the health status of the anchorage end.

Description

Technical Field

[0001] This invention relates to the fields of structural health monitoring and non-destructive testing, and in particular to a method, system, equipment, and medium for testing cable anchorage ends. Background Technology

[0002] In long-span cable-stayed bridges, the cable anchorage is a critical load-bearing component. When defects appear at the cable anchorage, they can severely impact the bridge's normal service life and significantly threaten its safety. With increasing service life and the influence of external environmental factors, further deterioration of defects such as anchorage corrosion can potentially lead to catastrophic structural damage. Therefore, inspection of cable anchorages must be conducted in their early stages to detect and address serious defects, preventing severe safety accidents and economic losses.

[0003] Due to the concealed location and unique internal structure of the anchorage end, it is often impossible to conduct specific visual inspections. Conventional non-destructive testing techniques, including eddy current testing, ultrasonic testing, magnetic particle testing, radiographic testing, and penetrant testing, are also difficult to effectively inspect the anchorage end.

[0004] In recent years, with the research and development of inspection technologies for structures such as anchor ends and cables, non-destructive testing methods suitable for anchor ends and cables, such as vibration methods, weak magnetic field testing methods, acoustic emission passive monitoring methods, and ultrasonic guided wave methods, have gradually demonstrated their advantages. Among them, magnetostrictive guided waves have been extensively and deeply studied due to their advantage of enabling long-distance detection with single-point excitation. In cable damage detection, they have advantages such as high detection efficiency, non-contact operation, ability to detect areas that are difficult to reach with traditional testing techniques, and convenient installation, and are gradually becoming a widely used new non-destructive testing method for cables.

[0005] However, in the actual field testing process using magnetostrictive guided wave technology, significant noise interference is often encountered. Furthermore, due to environmental factors, this noise often includes colored noise components, which are difficult to remove using conventional filtering and denoising methods to obtain a clean guided wave reflection signal. Simultaneously, in field testing, it is often difficult to obtain guided wave signals from intact structures as reference signals for comparative analysis. Therefore, when analyzing guided wave detection signals at anchor ends under conditions without a reference, a signal processing method is needed to analyze guided wave reflection signals from anchor ends with low signal-to-noise ratios. Summary of the Invention

[0006] Based on this, embodiments of the present invention provide a method, system, device and medium for detecting the anchorage end of a cable, which can process and analyze the guided wave reflection signal of the low signal-to-noise ratio anchorage end under no-reference conditions, thereby realizing the detection of the health status of the anchorage end.

[0007] To achieve the above objectives, embodiments of the present invention provide the following solutions:

[0008] A method for detecting the anchorage end of a cable includes:

[0009] Acquire guided wave reflection signals at each anchorage end of the target cable;

[0010] The guided wave reflection signal is decomposed using an adaptive noise complete set empirical mode decomposition method to obtain the decomposition result, which includes multiple modal components and corresponding residuals.

[0011] Calculate the autocorrelation function of the decomposition result, and determine the high-frequency and low-frequency components of the decomposition result based on the autocorrelation function and mutual information theory;

[0012] The high-frequency component is filtered using a denoising method to obtain a filtered signal, and the filtered signal and the low-frequency component are reconstructed to obtain a reconstructed signal.

[0013] Determine the characteristic data of the reconstructed signal; the characteristic data includes: covariance coefficients, eigenvalues, time-domain characteristic indicators, and frequency-domain characteristic indicators;

[0014] By comparing the feature data of all anchor ends, all anchor ends are divided into healthy anchor ends and damaged anchor ends, and the first classification result is obtained.

[0015] The center frequency of the guided wave reflection signal of each anchorage end is extracted, and all anchorage ends are divided into healthy anchorage ends and damaged anchorage ends according to the degree of deviation of the center frequency from the excitation frequency, thus obtaining the second classification result.

[0016] For any anchorage end, if the first classification result and the second classification result are consistent, then the first classification result or the second classification result is determined as the final classification result of the anchorage end.

[0017] Optionally, the cable anchorage end detection method further includes:

[0018] Nonlinear guided wave analysis is performed on the guided wave reflection signals of each anchorage end to determine whether there is any minor damage at the corresponding anchorage end; the minor damage refers to damage with a damage degree less than a set damage value.

[0019] Optionally, nonlinear guided wave analysis is performed on the guided wave reflection signals at each anchorage end to determine whether there is any minor damage at the corresponding anchorage end, specifically including:

[0020] Nonlinear guided wave analysis was performed on the guided wave reflection signals at each anchorage end to calculate the nonlinear parameters of each anchorage end.

[0021] For any anchorage end, if the nonlinear parameter is greater than the set parameter value, it is determined that there is minor damage to the corresponding anchorage end; otherwise, there is no minor damage.

[0022] Optionally, the formula for calculating the nonlinear parameter is:

[0023] β=A2 / A1 2 ;

[0024] Where β is a nonlinear parameter; A2 is the amplitude of the second harmonic in the guided wave reflected signal; and A1 is the amplitude of the fundamental wave in the guided wave reflected signal.

[0025] Optionally, all anchorage ends are classified into healthy anchorage ends and damaged anchorage ends based on the degree to which the center frequency deviates from the excitation frequency, specifically including:

[0026] If the center frequency deviates from the excitation frequency by more than 5%, the corresponding anchor end is determined to be a damaged anchor end; otherwise, the corresponding anchor end is a healthy anchor end.

[0027] Optionally, a denoising method is used to filter the high-frequency component to obtain a filtered signal, specifically including:

[0028] The high-frequency component is filtered using a wavelet soft thresholding denoising method to obtain the filtered signal.

[0029] Optionally, the time-domain characteristic indicators include: maximum value, minimum value, peak-to-peak value, mean, variance, standard deviation, mean square value, root mean square value, kurtosis, skewness, waveform factor, peak factor, impulse factor, and margin factor;

[0030] The frequency domain characteristic indicators include: centroid frequency, mean square frequency, root mean square frequency, frequency variance, and frequency standard deviation.

[0031] The present invention also provides a cable anchorage end detection system, comprising:

[0032] The signal acquisition module is used to acquire the guided wave reflection signals at each anchorage end of the target cable;

[0033] The signal decomposition module is used to decompose the guided wave reflection signal using an adaptive noise complete set empirical mode decomposition method to obtain the decomposition result; the decomposition result includes: multiple modal components and corresponding residuals;

[0034] The high- and low-frequency partitioning module is used to calculate the autocorrelation function of the decomposition result and determine the high-frequency and low-frequency components in the decomposition result based on the autocorrelation function and mutual information theory.

[0035] A high-frequency denoising module is used to filter the high-frequency component using a denoising method to obtain a filtered signal, and to reconstruct the filtered signal and the low-frequency component to obtain a reconstructed signal.

[0036] The feature extraction module is used to determine the feature data of the reconstructed signal; the feature data includes: covariance coefficients, eigenvalues, time-domain feature indices, and frequency-domain feature indices.

[0037] The first classification module is used to compare the feature data of all anchor ends and classify all anchor ends into healthy anchor ends and damaged anchor ends to obtain the first classification result.

[0038] The second classification module is used to extract the center frequency of the guided wave reflection signal of each anchorage end, and classify all anchorage ends into healthy anchorage ends and damaged anchorage ends according to the degree of deviation of the center frequency from the excitation frequency, thus obtaining the second classification result.

[0039] The health status determination module is used to determine the first classification result or the second classification result as the final classification result of the anchorage end if the first classification result and the second classification result are consistent for any anchorage end.

[0040] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to perform the above-described cable anchorage end detection method.

[0041] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described cable anchorage end detection method.

[0042] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0043] This invention employs an adaptive noise-complete ensemble empirical mode decomposition method to decompose guided wave reflection signals. A denoising method is used to filter the high-frequency components of the decomposition results, and the filtered signal and low-frequency components are reconstructed, improving the signal-to-noise ratio (SNR) of the reconstructed signal. Based on the characteristic data of the reconstructed signal and the center frequencies of the guided wave reflection signals at each anchorage end, the health status of the anchorage end is classified. The two classification results correspond to each other, improving classification accuracy. The entire process requires no reference signal. Therefore, this invention can process and analyze low SNR guided wave reflection signals at anchorage ends under reference-free conditions, enabling the detection of anchorage end health status. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 A flowchart of the cable anchorage end detection method provided in an embodiment of the present invention;

[0046] Figure 2 A diagram illustrating the more detailed implementation process of the cable anchorage end detection method in practical applications;

[0047] Figure 3 This is a diagram of the signal measured on-site;

[0048] Figure 4 This is a schematic diagram of the original anchorage end signal;

[0049] Figure 5 A schematic diagram of the decomposition of ICEEMDAN;

[0050] Figure 6 This is a schematic diagram of the reconstructed signal;

[0051] Figure 7 A schematic diagram showing the changes in frequency domain characteristic indicators for different anchoring ends;

[0052] Figure 8 A schematic diagram showing the changes in time-domain characteristic indicators for different anchoring ends;

[0053] Figure 9 A schematic diagram showing the characteristic values ​​of different anchoring ends;

[0054] Figure 10 A schematic diagram showing the covariance coefficients for different anchoring ends;

[0055] Figure 11 A schematic diagram showing the center frequencies of waveguide reflection signals at different anchoring ends;

[0056] Figure 12 A schematic diagram showing the deviation of the center frequency of the waveguide reflection signal at the anchorage end;

[0057] Figure 13 A schematic diagram of the second harmonic phenomenon of the cable;

[0058] Figure 14 This is a structural diagram of the cable anchorage end detection system provided in an embodiment of the present invention. Detailed Implementation

[0059] The technical solutions of the present invention will now be described with reference to the accompanying drawings in the embodiments of the present invention.

[0060] The purpose of this invention is to provide a method, system, device, and medium for detecting anchorage ends of cables. By combining an adaptive noise complete set empirical mode decomposition method and a denoising method, and by using a method in which the two classification results are mutually verified, it is possible to process and analyze guided wave reflection signals at low signal-to-noise ratio anchorage ends under reference conditions, and to detect the health status of anchorage ends.

[0061] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0062] Example 1

[0063] See Figure 1 The cable anchorage end detection method of this embodiment includes:

[0064] Step 101: Obtain the guided wave reflection signals at each anchorage end of the target cable.

[0065] Each guided wave reflection signal is then processed according to steps 102-105.

[0066] Step 102: Decompose the guided wave reflection signal using the adaptive noise complete set empirical mode decomposition method to obtain the decomposition result; the decomposition result includes: multiple mode components and corresponding residuals.

[0067] Specifically, the guided wave reflection signal is decomposed using the improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) method, and the decomposition results are obtained as follows:

[0068] (1) Denoise the signal using the ICEEMDAN method, defining x as the signal to be decomposed, E k (.) represents the k-th order mode component generated by EMD decomposition, N(.) represents the local mean of the generated signal, and w (i) This represents Gaussian white noise.

[0069] (2) Add Gaussian white noise w to the original signal (i) Construct sequence x (i) =x+β0E(w) (i) ), thus obtaining the first set of residuals R1 = (N(x (i) )), E(w (i) ) represents the Gaussian white noise generated in the i-th processing, β0 represents the Gaussian white noise weight coefficients, and x (i)This represents the signal to be decomposed with i-th Gaussian white noise added; the first modal component d1 = x - R1 is calculated; more white noise is added and the second set of residuals R2 = (N(R1 + β1E(w)) is calculated using local mean decomposition. (i) Define the second modal classification d2, d2 = R1 - R2; calculate the residual R of the k-th group. k =(N(R) k-1 +β k-1 E(w (i) ))) and the kth modal component d k =R k-1 -R k The calculation continues until all modal components and residuals are obtained.

[0070] Step 103: Calculate the autocorrelation function of the decomposition result, and determine the high-frequency and low-frequency components of the decomposition result based on the autocorrelation function and mutual information theory.

[0071] Specifically, the formula for calculating the autocorrelation function is R. x (t1,t2)=E[x(t1)x(t2)]. Where t1 and t2 represent two arbitrarily different times of the same modal component or residual x(t); R x (t1,t2) represents the correlation between the modal component or residual x(t) at times t1 and t2; x(t1) and x(t2) represent the values ​​of the modal component or residual x(t) at times t1 and t2, respectively; E[x(t1)x(t2)] represents the expected value of the product of x(t1) and x(t2) at times t1 and t2; R x (t1,t2) represents the autocorrelation function of the modal component or residual x(t) at times t1 and t2.

[0072] For noise, the weak correlation at various time points determines that the autocorrelation function is maximum at zero, while the autocorrelation function at other points rapidly decays to a very small value. This allows us to obtain the boundary between the high-frequency and low-frequency components that distinguish the concentrated distribution of noise. Calculating the autocorrelation function is how we distinguish between the high-frequency component, which contains the majority of noise, and the low-frequency component, which contains information.

[0073] Step 104: Filter the high-frequency part using a noise reduction method to obtain a filtered signal, and reconstruct the filtered signal and the low-frequency part to obtain a reconstructed signal.

[0074] Since noise is typically concentrated in the high-frequency range, filtering of this high-frequency component is necessary. Specifically, wavelet soft-threshold denoising is used to filter the high-frequency portion, resulting in a filtered signal. The expression for the wavelet soft-threshold denoising method is as follows:

[0075]

[0076] Where Wx represents wavelet coefficients; t represents a given threshold; and sgn represents the sign function.

[0077] The high-frequency components are filtered and denoised to reconstruct the signal, resulting in a signal with a high signal-to-noise ratio for subsequent analysis.

[0078] Step 105: Determine the feature data of the reconstructed signal; the feature data includes: covariance coefficients, eigenvalues, time-domain feature indices, and frequency-domain feature indices.

[0079] The time-domain characteristic indicators include: maximum value, minimum value, peak-to-peak value, mean, variance, standard deviation, mean square value, root mean square value, kurtosis, skewness, waveform factor, peak factor, impulse factor, and margin factor.

[0080] The frequency domain characteristic indicators include: centroid frequency, mean square frequency, root mean square frequency, frequency variance, and frequency standard deviation.

[0081] The formula for calculating the covariance coefficient is: Cov(X,Y)=E[XY]-E[X]E[Y], where X represents the reconstructed signal X; Y represents the reconstructed signal Y; E[X] represents the expectation of the reconstructed signal X; E[Y] represents the expectation of the reconstructed signal Y; and E[XY] represents the expectation of the product of the reconstructed signal X and the reconstructed signal Y.

[0082] The formula for solving the eigenvalues ​​is: Ax = λx, where A represents the covariance matrix composed of different reconstructed signals; λ represents the eigenvalues; and x represents the eigenvectors corresponding to the eigenvalues.

[0083] Step 106: Compare the feature data of all anchor ends and classify all anchor ends into healthy anchor ends and damaged anchor ends to obtain the first classification result.

[0084] Specifically, characteristic data that are significantly different from other anchor ends are differentiated and screened to distinguish between healthy anchor ends and anchor ends with damage.

[0085] Step 107: Extract the center frequency of the guided wave reflection signal of each anchorage end, and classify all anchorage ends into healthy anchorage ends and damaged anchorage ends according to the degree of deviation of the center frequency from the excitation frequency, thus obtaining the second classification result.

[0086] After different anchoring ends are categorized in step 106, the center frequency of the guided wave reflection signal of the anchoring end is extracted. Since the center frequency of the guided wave propagation signal will shift with the change of the structural cross section, the health status of the anchoring end is divided into two categories according to the degree of deviation of the center frequency of the guided wave reflection signal of the anchoring end from the excitation frequency. Specifically, if the degree of deviation of the center frequency from the excitation frequency is greater than 5%, the corresponding anchoring end is determined to be a damaged anchoring end; otherwise, the corresponding anchoring end is a healthy anchoring end.

[0087] Step 108: For any anchoring end, if the first classification result and the second classification result are consistent, then the first classification result or the second classification result is determined as the final classification result of the anchoring end.

[0088] The two classification results are compared and analyzed to ensure that they correspond to each other.

[0089] The above embodiments can process and analyze guided wave reflection signals at low signal-to-noise ratio anchorages without reference conditions, and obtain the health status classification of different anchorages and the relative damage degree analysis through a combination of multiple analysis indicators.

[0090] In one example, the cable anchorage end detection method further includes:

[0091] Step 109: Perform nonlinear guided wave analysis on the guided wave reflection signals of each anchorage end to determine whether there is any minor damage at the corresponding anchorage end; the minor damage refers to damage less than a set damage value. This step enables the assessment of minor damage at the anchorage end.

[0092] Step 109 specifically includes:

[0093] (1) Perform nonlinear guided wave analysis on the guided wave reflection signals at each anchorage end, and calculate the nonlinear parameters of each anchorage end; the calculation formula for the nonlinear parameters is as follows:

[0094] β=A2 / A1 2 ;

[0095] Where β is a nonlinear parameter; A2 is the amplitude of the second harmonic in the guided wave reflected signal; and A1 is the amplitude of the fundamental wave in the guided wave reflected signal.

[0096] (2) For any anchorage end, if the nonlinear parameter is greater than the set parameter value β0, then it is determined that the corresponding anchorage end has minor damage (or the minor damage is relatively obvious); otherwise, there is no minor damage (or the minor damage is not obvious). Specifically, the set parameter value β0 = μ + σ, that is, the set parameter value β0 is the sum of the mean and standard deviation of the nonlinear parameters of the guided wave reflection signal of each anchorage segment.

[0097] This example illustrates an anchorage end with noticeable minor damage. Such damage is prone to further development and poses a potential safety hazard. It serves to supplement the description of the anchorage end's health condition and acts as an early warning system.

[0098] The method described in the above embodiments is a signal processing method suitable for detecting the health status of anchorage ends. Under no-reference conditions, it classifies and analyzes the condition of anchorage end signals to assess the relative degree of damage. This method uses an improved adaptive noise complete set empirical mode decomposition method for signal denoising, and employs weak signal extraction and eigenvalue decomposition methods to extract eigenvalues, covariance coefficients, and other time-domain and frequency-domain characteristic signal indicators for classification analysis. Based on the classification, and considering the anchorage end structure, it uses a center frequency deviation standard to analyze the relative degree of damage of the anchorage end signals, and utilizes a nonlinear guided wave method for auxiliary analysis.

[0099] The following is combined Figure 2 A more specific implementation process of the above-mentioned cable anchorage end detection method in practical applications will be explained:

[0100] Taking the measured results of a bridge cable as an example, this paper analyzes and explains the signal processing method. The anchorage segments in the measured signal are divided into Class A and Class B cables. This signal processing method can distinguish between the relatively healthy anchorage ends and the damaged anchorage ends within both Class A and Class B cables. (See also...) Figure 2 The details are as follows:

[0101] 1. First obtain, such as Figure 3 The on-site measured signal is shown, and the original signal at the anchorage end is extracted to obtain the original anchorage end signal, as shown. Figure 4 As shown.

[0102] 2. Perform ICEEMDAN decomposition on the obtained signal to obtain IMFs and Res. The ICEEMDAN decomposition process is as follows: Figure 5 As shown.

[0103] 3. Calculate the autocorrelation function of IMFs, and based on mutual information theory, distinguish between the high-frequency and low-frequency components of the IMFs. Apply wavelet soft-threshold denoising to the high-frequency component of the IMFs and reconstruct the signal. The reconstructed signal is as follows: Figure 6 As shown.

[0104] 4. For the reconstructed signal after ICEEMDAN decomposition, weak signal extraction and eigenvalue decomposition are used for further analysis, including covariance coefficients, eigenvalues, time-domain characteristic indicators, and frequency-domain characteristic indicators. The characteristic indicators are shown in Table 1.

[0105] Table 1 Feature Indicators

[0106]

[0107] The changes in frequency domain characteristic indicators at different anchoring ends are as follows: Figure 7 As shown; the changes in time-domain characteristic indicators of different anchoring ends are as follows. Figure 8 As shown; the distribution of characteristic values ​​at different anchoring ends is as follows Figure 9 As shown, Figure 9 Part (a) corresponds to Category A. Figure 9 Part (b) corresponds to type B cable; the covariance coefficients of different anchorage ends are as follows: Figure 10 As shown, Figure 10 Part (a) corresponds to Category A. Figure 10 Part (b) corresponds to Category B.

[0108] Regarding the variance, standard deviation, root mean square deviation, peak factor, impulse factor, and margin factor in the time-domain characteristic indicators, it is quite evident that there is a significant difference in the magnitude of the indicators for Class A cables numbered 1, 2, and 3 compared to those numbered 4, 5, and 6. Similarly, Class B cables numbered 5 and 6 (i.e., cables numbered 11 and 12 in total) show a significant difference compared to other anchorage end signal characteristic indicators. In the frequency-domain characteristic indicators, the distinction between these two types of indicators is even more pronounced.

[0109] From the eigenvalues ​​of the covariance matrix of the anchorage end signal and its own covariance coefficient, it can be seen that in Class A cables, the relative values ​​of cables numbered 1, 2, and 3 are relatively small. Similarly, in Class B cables, the relative values ​​of cables numbered 5 and 6 (i.e., the total numbers 11 and 12) are relatively small. Their distribution corresponds to the time-frequency domain index.

[0110] 5. Extract the center frequency of the guided wave reflection signal from different cable anchorage ends, such as... Figure 11 As shown, the center frequency deviations of the waveguided signal reflected from the anchorage end are classified according to their deviation from the center frequency. The deviations of the center frequency of the waveguided signal reflected from the anchorage end are as follows: Figure 12 As shown, Figure 12 Part (a) corresponds to Category A. Figure 12 Part (b) corresponds to Category B.

[0111] In Class A cables, the center frequencies of the guided wave reflection signals at the anchor ends of anchors numbered 4, 5, and 6 are closer to the excitation frequency. In Class B cables, the center frequencies of the guided wave reflection signals at the anchor ends of anchors numbered 5 and 11 are more deviating from the excitation frequency.

[0112] 6. The second harmonic phenomenon of the guided wave signal is used to analyze minor damage to the cable structure, serving as an auxiliary analysis method. The second harmonic phenomenon of the cable is as follows: Figure 13 As shown.

[0113] 7. By combining time-frequency domain indices, eigenvalues, covariance coefficients, and the center frequency deviation of the guided wave reflection signal at the anchorage end, it can be analyzed that anchorage ends 1, 2, and 3 in Class A cable anchorage end signals, and anchorage ends 5 and 11 in Class B cable anchorage end signals are the anchorage ends most likely to be damaged.

[0114] The above method is suitable for classifying and analyzing the health status of anchorage ends in the absence of reference. The signal with low signal-to-noise ratio obtained from the field measurement of anchorage ends is subjected to noise reduction analysis and classified into anchorage ends in healthy condition and damaged condition. Indicators are selected to describe the classification of anchorage ends in healthy condition and to evaluate the relative damage degree of anchorage ends in damaged condition.

[0115] Example 2

[0116] In order to implement the method corresponding to Embodiment 1 above and achieve the corresponding functions and technical effects, a cable anchorage end detection system is provided below.

[0117] See Figure 14 The system includes:

[0118] The signal acquisition module 201 is used to acquire the guided wave reflection signals at each anchorage end of the target cable.

[0119] The signal decomposition module 202 is used to decompose the guided wave reflection signal using an adaptive noise complete set empirical mode decomposition method to obtain the decomposition result; the decomposition result includes: multiple mode components and corresponding residuals.

[0120] The high- and low-frequency division module 203 is used to calculate the autocorrelation function of the decomposition result and determine the high-frequency part and low-frequency part of the decomposition result based on the autocorrelation function and mutual information theory.

[0121] The high-frequency denoising module 204 is used to filter the high-frequency part using a denoising method to obtain a filtered signal, and to reconstruct the filtered signal and the low-frequency part to obtain a reconstructed signal.

[0122] The feature extraction module 205 is used to determine the feature data of the reconstructed signal; the feature data includes: covariance coefficients, eigenvalues, time-domain feature indices and frequency-domain feature indices.

[0123] The first classification module 206 is used to compare the feature data of all anchor ends, classify all anchor ends into healthy anchor ends and damaged anchor ends, and obtain the first classification result.

[0124] The second classification module 207 is used to extract the center frequency of the guided wave reflection signal of each anchorage end, and classify all anchorage ends into healthy anchorage ends and damaged anchorage ends according to the degree of deviation of the center frequency from the excitation frequency, thus obtaining the second classification result.

[0125] The health status determination module 208 is used to determine the first classification result or the second classification result as the final classification result of the anchorage end if the first classification result and the second classification result are consistent for any anchorage end.

[0126] Example 3

[0127] This embodiment provides an electronic device, including a memory and a processor. The memory is used to store computer programs, and the processor runs the computer programs to enable the electronic device to perform the cable anchorage end detection method of Embodiment 1.

[0128] Alternatively, the aforementioned electronic device may be a server.

[0129] In addition, this embodiment of the invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the cable anchorage end detection method of Embodiment 1.

[0130] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0131] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for detecting the anchorage end of a cable, characterized in that, include: Acquire guided wave reflection signals at each anchorage end of the target cable; The guided wave reflection signal is decomposed using the adaptive noise complete set empirical mode decomposition method to obtain the decomposition result; The decomposition results include: multiple modal components and corresponding residuals; Calculate the autocorrelation function of the decomposition result, and determine the high-frequency and low-frequency components of the decomposition result based on the autocorrelation function and mutual information theory; The high-frequency component is filtered using a denoising method to obtain a filtered signal, and the filtered signal and the low-frequency component are reconstructed to obtain a reconstructed signal. Determine the characteristic data of the reconstructed signal; the characteristic data includes: covariance coefficients, eigenvalues, time-domain characteristic indicators, and frequency-domain characteristic indicators; By comparing the feature data of all anchor ends, all anchor ends are divided into healthy anchor ends and damaged anchor ends, and the first classification result is obtained. The center frequency of the guided wave reflection signal of each anchorage end is extracted, and all anchorage ends are divided into healthy anchorage ends and damaged anchorage ends according to the degree of deviation of the center frequency from the excitation frequency, thus obtaining the second classification result. For any anchorage end, if the first classification result and the second classification result are consistent, then the first classification result or the second classification result is determined as the final classification result of the anchorage end; Calculate the autocorrelation function of the decomposition result, and determine the high-frequency and low-frequency components of the decomposition result based on the autocorrelation function and mutual information theory. Specifically: The formula for calculating the autocorrelation function is as follows: R x ( t 1, t 2)= E [ x ( t 1) x ( t 2)], where, t 1 and t 2 represents the same modal component or residual number. x ( t Two arbitrary different times; x ( t 1) with x ( t 2) Represent modal components or residuals respectively. x ( t )exist t 1 and t The value at time 2; E [ x ( t 1) x ( t 2)] represents the modal components or residuals. x ( t )exist t 1 and t 2 moments x ( t 1) with x ( t 2) The expectation of the product; R x ( t 1, t 2) Represents modal components or residuals. x ( t )exist t 1 and t 2 moments x ( t 1) with x ( t 2) The autocorrelation function, which characterizes the modal components or residuals. x ( t )exist t 1 and t The degree of correlation at time 2; For noise, the weak correlation at each time point determines that the autocorrelation function is the largest at zero point, while the autocorrelation function at other points decays rapidly to a very small value. In this way, the boundary between the high-frequency part and the low-frequency part that distinguishes the concentrated distribution of noise can be obtained, and the autocorrelation function can be calculated to distinguish the high-frequency part containing noise and the low-frequency part containing information.

2. The cable anchorage end detection method according to claim 1, characterized in that, Also includes: Nonlinear guided wave analysis was performed on the guided wave reflection signals at each anchorage end to determine whether there was any minor damage at the corresponding anchorage end. The term "minor damage" refers to damage levels less than a set damage value.

3. The cable anchorage end detection method according to claim 2, characterized in that, Nonlinear guided wave analysis was performed on the guided wave reflection signals at each anchorage end to determine whether there was any minor damage at the corresponding anchorage end. Specifically, this included: Nonlinear guided wave analysis was performed on the guided wave reflection signals at each anchorage end to calculate the nonlinear parameters of each anchorage end. For any anchorage end, if the nonlinear parameter is greater than the set parameter value, it is determined that there is minor damage to the corresponding anchorage end; otherwise, there is no minor damage.

4. The cable anchorage end detection method according to claim 3, characterized in that, The formula for calculating the nonlinear parameter is as follows: β= A2 / A1 2 ; in, β A1 is a nonlinear parameter; A2 is the amplitude of the second harmonic in the guided wave reflected signal, and A1 is the amplitude of the fundamental wave in the guided wave reflected signal.

5. The cable anchorage end detection method according to claim 1, characterized in that, Based on the degree to which the center frequency deviates from the excitation frequency, all anchorage ends are divided into healthy anchorage ends and damaged anchorage ends, specifically including: If the center frequency deviates from the excitation frequency by more than 5%, the corresponding anchor end is determined to be a damaged anchor end; otherwise, the corresponding anchor end is a healthy anchor end.

6. The method for detecting the anchorage end of a cable according to claim 1, characterized in that, The high-frequency component is filtered using a denoising method to obtain the filtered signal, specifically including: The high-frequency component is filtered using a wavelet soft thresholding denoising method to obtain the filtered signal.

7. The cable anchorage end detection method according to claim 1, characterized in that, The time-domain characteristic indicators include: maximum value, minimum value, peak-to-peak value, mean, variance, standard deviation, mean square value, root mean square value, kurtosis, skewness, waveform factor, peak factor, impulse factor, and margin factor. The frequency domain characteristic indicators include: centroid frequency, mean square frequency, root mean square frequency, frequency variance, and frequency standard deviation.

8. A cable anchorage end detection system, characterized in that, include: The signal acquisition module is used to acquire the guided wave reflection signals at each anchorage end of the target cable; The signal decomposition module is used to decompose the guided wave reflection signal using an adaptive noise complete set empirical mode decomposition method to obtain the decomposition result; The decomposition results include: multiple modal components and corresponding residuals; The high- and low-frequency partitioning module is used to calculate the autocorrelation function of the decomposition result and determine the high-frequency and low-frequency components in the decomposition result based on the autocorrelation function and mutual information theory. A high-frequency denoising module is used to filter the high-frequency component using a denoising method to obtain a filtered signal, and to reconstruct the filtered signal and the low-frequency component to obtain a reconstructed signal. The feature extraction module is used to determine the feature data of the reconstructed signal; the feature data includes: covariance coefficients, eigenvalues, time-domain feature indices, and frequency-domain feature indices. The first classification module is used to compare the feature data of all anchor ends and classify all anchor ends into healthy anchor ends and damaged anchor ends to obtain the first classification result. The second classification module is used to extract the center frequency of the guided wave reflection signal of each anchorage end, and classify all anchorage ends into healthy anchorage ends and damaged anchorage ends according to the degree of deviation of the center frequency from the excitation frequency, thus obtaining the second classification result. The health status determination module is used to determine the first classification result or the second classification result as the final classification result of the anchorage end if the first classification result and the second classification result are consistent for any anchorage end. Calculate the autocorrelation function of the decomposition result, and determine the high-frequency and low-frequency components of the decomposition result based on the autocorrelation function and mutual information theory. Specifically: The formula for calculating the autocorrelation function is as follows: R x ( t 1, t 2)= E [ x ( t 1) x ( t 2)], where, t 1 and t 2 represents the same modal component or residual number. x ( t Two arbitrary different times; x ( t 1) with x ( t 2) Represent modal components or residuals respectively. x ( t )exist t 1 and t The value at time 2; E [ x ( t 1) x ( t 2)] represents the modal components or residuals. x ( t )exist t 1 and t 2 moments x ( t 1) with x ( t 2) The expectation of the product; R x ( t 1, t 2) Represents modal components or residuals. x ( t )exist t 1 and t 2 moments x ( t 1) with x ( t 2) The autocorrelation function, which characterizes the modal components or residuals. x ( t )exist t 1 and t The degree of correlation at time 2; For noise, the weak correlation at each time point determines that the autocorrelation function is the largest at zero point, while the autocorrelation function at other points decays rapidly to a very small value. In this way, the boundary between the high-frequency part and the low-frequency part that distinguishes the concentrated distribution of noise can be obtained, and the autocorrelation function can be calculated to distinguish the high-frequency part containing noise and the low-frequency part containing information.

9. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to cause the electronic device to perform the cable anchorage end detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the cable anchorage end detection method as described in any one of claims 1 to 7.

Images