Nondestructive testing method for prestress of anchor cable

Abstract

The invention relates to the technical field of data processing, in particular to a nondestructive testing method for the prestress of an anchor cable. Obtaining the difference degree according to the difference characteristics of the time-frequency data of the sound wave signals in the adjacent preset time periods and the change characteristics of the sound wave signals; respectively segmenting the sound wave signal and the background signal according to the difference degree; eMD decomposition is carried out on the sound wave signal segment and the background signal segment, and the comprehensive interference degree is obtained according to the difference characteristics of the frequency spectrum data of the sound wave component and all the background components; adjusting the initial covariance matrix of the sound wave component according to the comprehensive interference degree, and denoising through a Kalman filtering algorithm to obtain a denoised component; and performing signal reconstruction according to the de-noising component and the comprehensive interference degree of the sound wave signal segment to obtain a de-noised wave signal segment. Corrected sound wave signals are obtained according to all de-noised wave signal segments; the anchor cable prestress is detected according to the corrected sound wave signal, and the accuracy and reliability of anchor cable detection are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a nondestructive testing method for anchor cable prestress. BACKGROUND

[0002] The anchor cable is a prestressed reinforcement component composed of high-strength steel strand, anchor and grouting body, and is widely used in the fields of rock and soil engineering such as slope support, deep foundation pit engineering and tunnel reinforcement. In long-term use, the prestress of the anchor cable may be gradually lost due to the influence of factors such as stratum creep, corrosion and load fluctuation, thereby reducing the bearing efficiency of the anchoring system. Therefore, it is necessary to nondestructively test the prestress of the anchor cable to evaluate the change rule and risk of the prestress.

[0003] Due to the good penetration ability and non-contact measurement advantage of the acoustic wave method (also known as acoustic wave transmission method or ultrasonic detection method), it is widely used for qualitative and quantitative evaluation of the prestress state. The stress state of the anchor cable is mainly inferred according to the characteristic that the propagation speed of the acoustic wave in the anchor cable changes with the stress. However, in actual application, environmental interference such as external vibration signal and wind noise often overlaps with the detection acoustic wave in the frequency domain, which easily causes signal waveform distortion, unclear propagation path identification and other problems, seriously affecting the accuracy of acoustic velocity interpretation, and further reducing the accuracy and reliability of the prestress evaluation result. SUMMARY

[0004] In order to solve the above technical problems, the purpose of the present application is to provide a nondestructive testing method for anchor cable prestress, and the technical scheme adopted is as follows: Acoustic wave signals of the anchor cable and different background signals of the environment are obtained. The difference degree of adjacent preset time periods is obtained according to the difference characteristics of the time-frequency data of the acoustic wave signals in the adjacent preset time periods and the change characteristics of the acoustic wave signals. The acoustic wave signal and the background signal are segmented according to the difference degree, and different acoustic wave signal segments and background signal segments are obtained. The acoustic wave signal segment is decomposed by EMD to obtain an acoustic wave component. The background signal segment is decomposed by EMD to obtain a background component. The interference degree is obtained according to the difference characteristics of the frequency spectrum data of the acoustic wave component and the background component. The comprehensive interference degree is obtained according to the interference degree between the acoustic wave component and all background components. The adaptive initial covariance matrix is obtained by adjusting the initial covariance matrix of the acoustic wave component in the Kalman filtering algorithm according to the comprehensive interference degree. The denoising component is obtained by denoising the acoustic wave component through the Kalman filtering algorithm according to the adaptive initial covariance matrix. According to all de-noised components of the acoustic wave signal segment and the comprehensive interference degree, a signal is reconstructed to obtain a de-noised acoustic wave signal segment; according to all de-noised acoustic wave signal segments, a corrected acoustic wave signal is obtained; and according to the corrected acoustic wave signal, an anchor cable prestress is detected.

[0005] Further, the step of obtaining a difference degree of adjacent preset time periods according to the difference feature of the time-frequency data of the acoustic wave signal in the adjacent preset time periods and the change feature of the acoustic wave signal comprises: An average value of the square of the value of the acoustic wave signal at all time points in a preset time period is calculated to obtain a vibration intensity feature value; a sum of the product of all frequencies and corresponding amplitudes in the time-frequency data of the preset time period is calculated to obtain a first value; a sum of all amplitudes of the time-frequency data in the preset time period is calculated to obtain a second value; a ratio of the first value to the second value is calculated to obtain a concentrated frequency value; an absolute value of the difference of the vibration intensity feature values of adjacent preset time periods is calculated and normalized to obtain a first difference value; an absolute value of the difference of the concentrated frequency values of adjacent preset time periods is calculated and normalized to obtain a second difference value; and an average value of the first difference value and the second difference value is calculated to obtain the difference degree of adjacent preset time periods.

[0006] Further, the step of segmenting the acoustic wave signal and the background signal according to the difference degree to obtain different acoustic wave signal segments and background signal segments comprises: When the difference degree exceeds a preset difference threshold, a signal segmentation point is inserted in adjacent preset time periods of the acoustic wave signal; an overlapping time period of adjacent preset time periods is taken as a selection interval of the signal segmentation point, adjacent preset time periods are segmented at any time point in the selection interval, the difference degree between the two time periods after segmentation is calculated to obtain a segmentation fitness of the any time point; a time point at which a maximum value of the segmentation fitness is located is marked as the signal segmentation point; the acoustic wave signal is segmented at all signal segmentation points to obtain different acoustic wave signal segments; and the background signal is segmented according to the time periods of the acoustic wave signal segments to obtain background signal segments with the same time periods as the acoustic wave signal segments.

[0007] Further, the step of obtaining an interference degree according to the difference feature of the frequency spectrum data of the acoustic wave component and the background component comprises: In the two frequency spectrum data of the acoustic wave component and the background component in the same time period, a mean square error of the amplitudes corresponding to the same frequency is calculated and is negatively correlated to be mapped to obtain a first similarity degree; a ratio of the number of the same frequencies to the number of the frequencies of the acoustic wave component is calculated to obtain a second similarity degree; and a product of the first similarity degree and the second similarity degree is calculated to obtain the interference degree.

[0008] Further, the step of obtaining the comprehensive interference degree according to the interference degree between the sound wave component and all background components comprises: calculating the sum of the interference degrees between the sound wave component and all background components to obtain the comprehensive interference degree of the sound wave component.

[0009] Further, the step of adjusting the initial covariance matrix of the sound wave component in the Kalman filtering algorithm according to the comprehensive interference degree to obtain an adaptive initial covariance matrix comprises: calculating the product of the initial covariance matrix, the comprehensive interference degree and a preset mapping parameter to obtain the adaptive initial covariance matrix of the sound wave component in the Kalman filtering algorithm.

[0010] Further, the step of reconstructing the signal according to all denoised components of the sound wave signal segment and the comprehensive interference degree to obtain a denoised sound wave signal segment comprises: calculating the product of the energy intensity of the corresponding denoised component after the negative correlation mapping of the comprehensive interference degree to obtain the importance degree of the denoised component; calculating the proportion of the importance degree in the importance degrees of all denoised components to obtain the reconstruction weight of the denoised component; and reconstructing the signal according to all denoised components and the corresponding reconstruction weights to obtain the denoised sound wave signal segment.

[0011] Further, the step of obtaining the corrected sound wave signal according to all denoised sound wave signal segments comprises: splicing the sound wave signal segments in time sequence to obtain the corrected sound wave signal.

[0012] The present application has the following advantages: In the present application, since the noise in the detection environment has the characteristic of randomness, and the interference of the sound wave signal also has randomness, the difference feature can be obtained to judge the similarity of the signal features of adjacent preset time periods in the sound wave signal; the sound wave signal and the background signal are segmented according to the difference degree, so that different degrees of denoising are performed on different sound wave signal segments, and the denoising accuracy is initially improved. The EMD decomposition of the sound wave signal is helpful for noise separation and frequency feature analysis, and further improves the denoising accuracy. The comprehensive interference degree can accurately represent the noise interference degree of the sound wave component, so as to determine the denoising degree of each sound wave component. The adaptive initial covariance matrix can determine the denoising degree of the sound wave component by the Kalman filtering algorithm, and improve the denoising accuracy. The denoised component can effectively reduce the noise interference in the sound wave component; the denoised sound wave signal segment can further reduce the noise information in the sound wave signal segment, and improve the accuracy of the sound wave signal in the detection of prestress. Finally, the anchor cable prestress is detected according to the corrected sound wave signal, and the detection accuracy and reliability are improved. BRIEF DESCRIPTION OF DRAWINGS

[0013] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0014] Figure 1 A flow chart of a non-destructive testing method for anchor cable prestress provided by an embodiment of the present application. DETAILED DESCRIPTION

[0015] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined purpose, the specific implementation, structure, features and effects of the non-destructive testing method for anchor cable prestress according to the present application will be described in detail below in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0016] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0017] The specific scheme of the non-destructive testing method for anchor cable prestress provided by the present application will be described in detail below in combination with the drawings.

[0018] Please refer to Figure 1 which shows a flow chart of a non-destructive testing method for anchor cable prestress provided by an embodiment of the present application. The method comprises the following steps: Step S1, obtaining the sound wave signal of the anchor cable and the different background signals of the environment.

[0019] In the embodiment of the present application, the implementation scenario is to perform non-destructive testing on anchor cable prestress and improve the accuracy of detection. In the process of detecting anchor cable prestress by using sound wave method, joint collection of sound wave signal and multi-source background noise needs to be carried out at the same time. While the excitation device sends excitation sound wave signal to the anchor cable, a plurality of groups of sensors arranged in the detection area are used to collect anchor cable response signal and background noise from the environment, including wind noise, ground vibration, construction noise and structure resonance, etc. Thus, synchronous sampling of sound wave channel and noise channel is realized, and finally the sound wave signal of the anchor cable and the different background signals of the environment are obtained. This parallel collection strategy establishes a data basis for subsequent noise separation analysis, thereby improving the accuracy and reliability of prestress evaluation.

[0020] In step S2, a difference degree of adjacent preset time periods is obtained according to a difference feature of time-frequency data of the acoustic wave signals in the adjacent preset time periods and a change feature of the acoustic wave signals; and the acoustic wave signals and the background signals are respectively segmented according to the difference degree, so as to obtain different acoustic wave signal segments and background signal segments.

[0021] The acoustic wave method uses the propagation characteristics of acoustic waves in the anchor cable to judge the stress change of the anchor cable, and has the advantages of non-contact and strong penetration. However, in actual engineering environments such as subways, tunnels or construction sites, there are usually background signals such as mechanical vibration and wind noise. The noise interference may overlap with the acoustic wave signals in frequency, causing distortion of the acoustic wave waveform and ambiguity of the propagation path, thereby reducing the signal interpretation accuracy and affecting the accuracy and reliability of the prestress evaluation. Since the background signals such as wind noise usually have the characteristics of randomness, the interference caused by the acoustic wave signals for the anchor cable prestress detection may also have a certain randomness, and thus the collected acoustic wave signals may be subjected to different degrees of noise interference in different time periods. Therefore, in order to more accurately remove the noise interference existing in the acoustic wave signals, the acoustic wave signals need to be segmented for different degrees of noise removal. First, a difference degree of adjacent preset time periods is obtained according to a difference feature of time-frequency data of the acoustic wave signals in the adjacent preset time periods and a change feature of the acoustic wave signals.

[0022] Preferably, in the embodiments of the present application, the step of obtaining the difference degree of adjacent preset time periods comprises: calculating the average value of the square of the value of the sound wave signal at all times within the preset time period to obtain a vibration intensity feature value; the greater the vibration intensity feature value, the greater the energy of the sound wave signal within the preset time period. In the embodiments of the present application, a preset sliding window is constructed, the sliding window length is the length of the signal with a time length of 1 second, the sliding step is half of the sliding window length, and one preset time period can be obtained from the starting position of the sound wave signal by sliding the preset sliding window once, and the time length of the preset time period can be determined by the implementation scene. The sum of the product of all frequencies and corresponding amplitudes in the time-frequency data of the preset time period is calculated to obtain a first value; the time-frequency data can be obtained by performing short-time Fourier transform on the sound wave signal, and it should be noted that this algorithm is prior art and the specific steps will not be described again. The sum of all amplitudes in the time-frequency data within the preset time period is calculated to obtain a second value; the ratio of the first value to the second value is calculated to obtain a concentrated frequency value; the concentrated frequency value can reflect the frequency position at which the frequency amplitude is mainly concentrated within the preset time period, and the greater the amplitude of a certain frequency, the closer the concentrated frequency value to the frequency. The absolute value of the difference of the vibration intensity feature values of adjacent preset time periods is calculated and normalized to obtain a first difference value; the greater the first difference value, the greater the energy difference of the sound wave signal between adjacent preset time periods. In the embodiments of the present application, the normalization method is linear normalization. The absolute value of the difference of the concentrated frequency values of adjacent preset time periods is calculated and normalized to obtain a second difference value; the greater the second difference value, the greater the difference in data change frequency between adjacent preset time periods. The average value of the first difference value and the second difference value is calculated to obtain the difference degree of adjacent preset time periods; the greater the difference degree, the greater the feature difference of the sound wave signal between adjacent preset time periods, and the more likely the noise interference received is different, and the more the adjacent preset time period needs to be divided. The formula for obtaining the difference degree comprises:

[0023] In the formula, represents the difference degree between adjacent preset time periods a and a+1, represents normalization, represents the vibration intensity feature value of the preset time period a, represents the vibration intensity feature value of the preset time period a+1, represents the concentrated frequency value of the preset time period a, represents the concentrated frequency value of the preset time period a+1, represents the first difference value, represents the second difference value.

[0024] Further, after the difference degree between all adjacent preset time periods is obtained, the sound wave signal and the background signal can be segmented according to the difference degree, respectively, to obtain different sound wave signal segments and background signal segments; preferably, in the embodiment of the present application, the step of obtaining different sound wave signal segments and background signal segments comprises: when the difference degree exceeds a preset difference threshold, a signal segmentation point is inserted in the adjacent preset time periods of the sound wave signal; in the embodiment of the present application, the preset difference threshold is 0.4, which can be determined by the implementer according to the implementation scene; the difference degree exceeding the preset difference threshold means that the sound wave signal feature difference is large, and the adjacent preset time periods are different in the degree of noise interference, so segmentation is needed in the time period. Since there is an overlapping time period in the adjacent preset time periods, in order to more accurately select the position of the signal segmentation point, the overlapping time period of the adjacent preset time periods is needed as the selection interval of the signal segmentation point, and the adjacent preset time periods are segmented at any time in the selection interval, the difference degree between the two time periods after segmentation is calculated to obtain the segmentation fitness of the any time; the larger the segmentation fitness is, the larger the sound wave signal feature difference before and after the time is. The time at which the maximum value of the segmentation fitness is located is marked as the signal segmentation point; the sound wave signal is segmented at all signal segmentation points to obtain different sound wave signal segments; the sound wave signal features in each sound wave signal segment are relatively similar. Since the background signal and the sound wave signal are collected in parallel, the background signal is segmented according to the time period of the sound wave signal segment to obtain a background signal segment with the same time period as the sound wave signal segment.

[0025] Step S3, the sound wave signal segment is decomposed by EMD to obtain a sound wave component; the background signal segment is decomposed by EMD to obtain a background component; the interference degree is obtained according to the difference features of the spectrum data of the sound wave component and the background component; the comprehensive interference degree is obtained according to the interference degrees between the sound wave component and all background components; the adaptive initial covariance matrix is obtained by adjusting the initial covariance matrix of the sound wave component in the Kalman filtering algorithm according to the comprehensive interference degree; the denoising component is obtained by denoising the sound wave component through the Kalman filtering algorithm according to the adaptive initial covariance matrix.

[0026] In order to more accurately denoise the sound wave signal segment, the sound wave signal segment is first decomposed by EMD to obtain a sound wave component; it should be noted that EMD, empirical mode decomposition, is prior art, and the specific decomposition steps will not be repeated; non-stationary data can be decomposed into multiple stationary component signals through decomposition, which is helpful for noise separation and frequency feature analysis. Similarly, the background signal segment is decomposed by EMD to obtain a background component. After different sound wave components and background components are obtained, the interference degree can be obtained according to the difference features of the spectrum data of the sound wave component and the background component.

[0027] Preferably, in the embodiments of the present application, the step of obtaining the interference degree comprises: calculating the mean square error of the amplitudes corresponding to the same frequency in the two frequency spectrum data of the sound wave component and the background component in the same time period and performing a negative correlation mapping to obtain a first similarity degree; the same frequency is a frequency whose amplitude is not 0 in the sound wave component and the background component. The greater the first similarity degree is, the smaller the mean square error is, and the more similar the amplitudes of the same frequency are; the sound wave component is more likely to contain information of the background noise and is more likely to be interfered by the noise. Calculating the ratio of the number of the same frequencies to the number of the frequencies of the sound wave component to obtain a second similarity degree; the number of the frequencies of the sound wave component is the number of the frequencies whose amplitudes are not 0. The greater the second similarity degree is, the more similar the frequency components of the sound wave component and the background component are, and the more likely the sound wave component is interfered by the noise. Calculating the product of the first similarity degree and the second similarity degree to obtain the interference degree; the greater the interference degree is, the more likely the sound wave component is interfered by the background component, and the more the sound wave component needs to be denoised.

[0028] Further, since multiple background signals are collected in the environment, multiple background signal segments and multiple background components exist in the same time period, and therefore the comprehensive interference degree is obtained according to the interference degrees between the sound wave component and all the background components; preferably, in the embodiments of the present application, the step of obtaining the comprehensive interference degree comprises: calculating the sum of the interference degrees between the sound wave component and all the background components to obtain the comprehensive interference degree of the sound wave component. The greater the comprehensive interference degree is, the greater the degree of the noise interference on the sound wave component is. The formula for obtaining the comprehensive interference degree comprises:

[0029] In the formula, W represents the comprehensive interference degree of the sound wave component, N represents the number of the background components corresponding to the time period in which the sound wave component is located, represents an exponential function with a natural constant as the base, represents the mean square error of the amplitudes corresponding to the same frequency between the sound wave component and the nth background component, represents the first similarity degree, represents the number of the same frequencies between the sound wave component and the nth background component, and H represents the number of the frequencies corresponding to the sound wave component, represents the second similarity degree, represents the interference degree of the nth background component on the sound wave component.

[0030] After obtaining the comprehensive interference degree of all acoustic components of the acoustic signal segment, different degrees of denoising can be performed according to the comprehensive interference degree. Therefore, the initial covariance matrix of the acoustic component in the Kalman filtering algorithm is adjusted according to the comprehensive interference degree to obtain an adaptive initial covariance matrix. Preferably, in the embodiment of the present application, the step of obtaining the adaptive initial covariance matrix includes: calculating the product of the initial covariance matrix, the comprehensive interference degree and the preset mapping parameter to obtain the adaptive initial covariance matrix of the acoustic component in the Kalman filtering algorithm. It should be noted that the Kalman filtering algorithm is a prior art, and the step of obtaining the initial covariance matrix will not be repeated. Since the Kalman filtering algorithm realizes state estimation through the weighted fusion between the observation value and the system model, and the weight is determined by the Kalman gain, which is inversely proportional to the covariance matrix, when the covariance matrix increases, it means that the observation noise is strong, and the filter will reduce the degree of trust in the observation value and rely more on model prediction; when the covariance matrix decreases, it means that the observation data is reliable, and the filter will enhance the response to the observation value; therefore, the comprehensive interference degree can be used as an adjustment factor of the initial covariance matrix, so that the filter can flexibly adjust the weight distribution of observation and prediction in different noise environments, thereby improving the denoising effect and enhancing the robustness to abnormal disturbances. Therefore, the larger the comprehensive interference degree is, the larger the adaptive initial covariance matrix is, and the stronger the denoising effect is. The preset mapping parameter is used to adjust the size of the comprehensive interference degree, and the specific obtaining step includes: collecting a pure acoustic signal from an experimental anchor cable in the laboratory, collecting an acoustic signal in a simulated noise environment and a background signal of the environment, initially selecting a mapping parameter to apply to the denoising process of the acoustic component, for example, 0.5; calculating the mean square error of the denoised acoustic signal and the pure acoustic signal, the smaller the mean square error is, the better the denoising effect is, and the more accurate the mapping parameter is; traversing all mapping parameters, and taking the mapping parameter corresponding to the minimum value of the mean square error as the preset mapping parameter in the actual detection scene.

[0031] Further, after obtaining the adaptive initial covariance matrix of the acoustic component in the Kalman filtering algorithm, the acoustic component can be denoised by the Kalman filtering algorithm according to the adaptive initial covariance matrix to obtain a denoised component; it should be noted that the Kalman filtering algorithm is a prior art, and the specific denoising steps will not be repeated. The adaptive initial covariance matrix can improve the denoising accuracy of the acoustic component, and further improve the accuracy and reliability of the final acoustic method for detecting the prestress of the anchor cable.

[0032] In step S4, the signal is reconstructed according to all denoised components of the acoustic signal segment and the comprehensive interference degree to obtain a denoised acoustic signal segment; the corrected acoustic signal is obtained according to all denoised acoustic signal segments; and the prestress of the anchor cable is detected according to the corrected acoustic signal.

[0033] After obtaining the de-noised components of all acoustic wave components of the acoustic wave signal segment, all de-noised components can be reconstructed to obtain a de-noised acoustic wave signal segment; since the proportions of signal information and noise information contained in different acoustic wave components are different, the importance and contribution degree of different acoustic wave components in the acoustic wave signal segment also differ, and if all de-noised components are directly superimposed with equal weight, certain noise may still remain after signal reconstruction, affecting the de-noising effect. Therefore, signal reconstruction is performed according to all de-noised components and the comprehensive interference degree of the acoustic wave signal segment to obtain a de-noised acoustic wave signal segment; preferably, in the embodiments of the present application, the step of obtaining the de-noised acoustic wave signal segment comprises: calculating the product of the energy intensity of the corresponding de-noised component after negative correlation mapping of the comprehensive interference degree to obtain the importance degree of the de-noised component; it should be noted that the energy intensity of the component signal after empirical mode decomposition belongs to the prior art, and the specific obtaining steps will not be described here; the greater the energy intensity, the greater the energy proportion of the de-noised component in the acoustic wave signal segment, the greater the importance, and the greater the reconstruction weight. The smaller the comprehensive interference degree, the smaller the degree of noise interference, and the greater the corresponding reconstruction weight. The proportion of the importance degree in the importance degrees of all de-noised components is calculated to obtain the reconstruction weight of the de-noised component; the sum of the reconstruction weights of all de-noised components is 1. Signal reconstruction is performed according to all de-noised components and the corresponding reconstruction weights to obtain a de-noised acoustic wave signal segment; the de-noised acoustic wave signal segment effectively removes the noise in the detection environment. The formula for obtaining the reconstruction weight comprises:

[0034] wherein, represents the reconstruction weight of the bth de-noised component, B represents the number of de-noised components corresponding to the acoustic wave signal segment, represents the energy intensity of the bth de-noised component, represents the comprehensive interference degree corresponding to the bth de-noised component, represents negative correlation mapping of the comprehensive interference degree, represents the importance degree.

[0035] Further, after obtaining all de-noised acoustic wave signal segments, a corrected acoustic wave signal can be obtained according to all de-noised acoustic wave signal segments; specifically comprising: splicing the acoustic wave signal segments in time sequence to obtain a corrected acoustic wave signal. The corrected acoustic wave signal can effectively remove the noise interference in the environment compared with the original acoustic wave signal, so that the corrected acoustic wave signal can more accurately reflect the anchor cable prestress characteristics, improving the accuracy and reliability of the anchor cable prestress detection; finally, the anchor cable prestress is detected according to the corrected acoustic wave signal, and the implementer can self-detect and evaluate the prestress state of the anchor cable according to the corrected acoustic wave signal.

[0036] In summary, the embodiment of the present application provides a nondestructive testing method for anchor cable prestress; the difference degree is obtained according to the difference features of time-frequency data of the acoustic wave signals in adjacent preset time periods and the change features of the acoustic wave signals; the acoustic wave signals and the background signals are respectively segmented according to the difference degree; the acoustic wave signal segments and the background signal segments are decomposed by EMD, and the comprehensive interference degree is obtained according to the difference features of the spectrum data of the acoustic wave components and all background components; the initial covariance matrix of the acoustic wave components is adjusted according to the comprehensive interference degree, and the denoising is performed through the Kalman filtering algorithm to obtain the denoising components; the signal reconstruction is performed according to the denoising components of the acoustic wave signal segments and the comprehensive interference degree to obtain the denoising acoustic wave signal segments. The corrected acoustic wave signals are obtained according to all the denoising acoustic wave signal segments; the anchor cable prestress is detected according to the corrected acoustic wave signals, and the accuracy and reliability of the anchor cable detection are improved.

[0037] It should be noted that the above-mentioned embodiment sequence of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or may be advantageous.

[0038] Each of the embodiments in the specification is described in a progressive manner, and the same and similar parts between the embodiments can be referred to each other. Each embodiment mainly describes the differences from other embodiments.

Claims

1. A non-destructive testing method for anchor cable prestress, characterized in that, The method includes the following steps: Acquire the acoustic signals of the anchor cable and different background signals of the environment; The degree of difference between adjacent preset time periods is obtained based on the difference characteristics of the time-frequency data of the acoustic signal and the change characteristics of the acoustic signal; the acoustic signal and the background signal are segmented according to the degree of difference to obtain different acoustic signal segments and background signal segments. The acoustic signal segment is decomposed using EMD to obtain acoustic components; the background signal segment is also decomposed using EMD to obtain background components; the interference level is obtained based on the difference characteristics of the spectral data of the acoustic components and the background components; the overall interference level is obtained based on the interference level between the acoustic components and all background components; the initial covariance matrix of the acoustic components in the Kalman filtering algorithm is adjusted according to the overall interference level to obtain an adaptive initial covariance matrix; the acoustic components are then denoised using the Kalman filtering algorithm based on the adaptive initial covariance matrix to obtain denoised components. The signal is reconstructed based on all the denoised components of the acoustic signal segment and the overall interference level to obtain a denoised acoustic signal segment; a corrected acoustic signal is obtained based on all the denoised acoustic signal segments; and the anchor cable prestress is detected based on the corrected acoustic signal.

2. The non-destructive testing method for anchor cable prestress according to claim 1, characterized in that, The step of obtaining the degree of difference between adjacent preset time periods based on the difference characteristics of the time-frequency data of the acoustic signal and the change characteristics of the acoustic signal includes: The vibration intensity characteristic value is obtained by calculating the average of the squares of the values ​​of the sound wave signal at all times within a preset time period; the first value is obtained by calculating the sum of the products of all frequencies and their corresponding amplitudes in the time-frequency data of the preset time period; the second value is obtained by calculating the sum of the amplitudes of all time-frequency data within the preset time period; the ratio of the first value to the second value is obtained by calculating the lumped frequency value; the absolute value of the difference between the vibration intensity characteristic values ​​of adjacent preset time periods is calculated and normalized to obtain the first difference; the absolute value of the difference between the lumped frequency values ​​of adjacent preset time periods is calculated and normalized to obtain the second difference; and the degree of difference between adjacent preset time periods is obtained by calculating the average of the first difference and the second difference.

3. The non-destructive testing method for anchor cable prestress according to claim 1, characterized in that, The step of segmenting the acoustic signal and the background signal according to the degree of difference to obtain different acoustic signal segments and background signal segments includes: When the degree of difference exceeds a preset difference threshold, a signal segmentation point is inserted in the adjacent preset time period of the sound wave signal; the overlapping time period of the adjacent preset time period is used as the selection interval of the signal segmentation point, and the adjacent preset time period is segmented at any time within the selection interval. The degree of difference between the two segmented time periods is calculated to obtain the segmentation suitability at the arbitrary time; the time at which the maximum value of the segmentation suitability is located is marked as the signal segmentation point; the sound wave signal is segmented at all signal segmentation points to obtain different sound wave signal segments; the background signal is segmented according to the time period of the sound wave signal segment to obtain a background signal segment with the same time period as the sound wave signal segment.

4. The non-destructive testing method for anchor cable prestress according to claim 1, characterized in that, The step of obtaining the interference level based on the difference characteristics of the spectral data of the acoustic component and the background component includes: In two spectral data sets of the acoustic component and the background component within the same time period, the mean square error of the amplitude corresponding to the same frequency is calculated and negatively correlated to obtain a first degree of similarity; the ratio of the number of the same frequency to the number of frequencies of the acoustic component is calculated to obtain a second degree of similarity; the product of the first degree of similarity and the second degree of similarity is calculated to obtain the interference level.

5. The non-destructive testing method for anchor cable prestress according to claim 1, characterized in that, The step of obtaining the overall interference level based on the interference level between the acoustic component and all background components includes: The sum of the interference levels between the sound wave component and all background components is calculated to obtain the overall interference level of the sound wave component.

6. The non-destructive testing method for anchor cable prestress according to claim 1, characterized in that, The step of adjusting the initial covariance matrix of the acoustic component in the Kalman filtering algorithm according to the overall interference level to obtain an adaptive initial covariance matrix includes: The product of the initial covariance matrix, the comprehensive interference level, and the preset mapping parameters is calculated to obtain the adaptive initial covariance matrix of the acoustic component in the Kalman filtering algorithm.

7. The non-destructive testing method for anchor cable prestress according to claim 1, characterized in that, The step of reconstructing the signal based on all denoised components of the acoustic signal segment and the overall interference level to obtain the denoised acoustic signal segment includes: The importance of the denoised component is obtained by multiplying the negative correlation mapping of the comprehensive interference level with the energy intensity of the corresponding denoised component; the proportion of the importance of the denoised component in the importance of all denoised components is calculated to obtain the reconstruction weight of the denoised component; the signal is reconstructed according to all denoised components and their corresponding reconstruction weights to obtain the denoised wave signal segment.

8. The non-destructive testing method for anchor cable prestress according to claim 1, characterized in that, The step of obtaining the corrected acoustic signal based on all denoised signal segments includes: The sound wave signal segments are spliced ​​together in chronological order to obtain the corrected sound wave signal.

Images