Acoustic emission-based asphalt pavement cracking damage source positioning method

Abstract

The application provides a method for locating a cracking damage source of an asphalt pavement based on acoustic emission, which comprises the following steps: 1. arranging a plurality of acoustic emission sensors in a matrix form in a monitoring area; 2. after receiving acoustic emission signals, a processing device processes the signals in the following manner: performing wavelet denoising processing on the detection signals to obtain denoised signals; then finding a time domain interval with the highest similarity degree according to the denoised signals, taking the time domain interval as a time difference to calculate the signal time difference between each denoised signal; and calculating the position of the cracking damage source by using a source positioning calculation method according to the signal time difference, a through wave velocity and sensor coordinates; the method has the beneficial technical effects that: the method for locating a cracking damage source of an asphalt pavement based on acoustic emission can accurately obtain the information about the position and development process of micro-damage, and makes it possible to clearly describe the mutual relationship between the micro-damage and external action.

Description

Technical Field

[0001] This invention relates to an asphalt pavement damage detection technology, and more particularly to a method for locating cracking damage sources in asphalt pavements based on acoustic emission. Background Technology

[0002] As a crucial component of transportation infrastructure, highways are vital for social development through their planning, design, construction, maintenance, and repair. As the structure that directly bears the load, asphalt pavement possesses excellent load-bearing capacity, skid resistance, deformation resistance, economy, and durability. During normal service, asphalt pavement experiences various degrees and types of micro-damage due to vehicle loads, environmental factors, and subgrade settlement. Currently, the relationship between micro-damage and external forces remains relatively unclear, particularly the generation and development processes of micro-damage, which are difficult to describe precisely. Summary of the Invention

[0003] To address the problems in the background art, this invention proposes a method for locating cracking damage sources in asphalt pavements based on acoustic emission, comprising:

[0004] 1) Sensor arrangement: After the asphalt pavement is laid, a monitoring area is set up on the asphalt pavement, and multiple installation holes are drilled within the monitoring area. The multiple installation holes are distributed in a matrix form, and an acoustic emission sensor is installed in each installation hole; each acoustic emission sensor is connected to the processing device.

[0005] 2) After the asphalt pavement is put into use, the treatment device detects the acoustic emission signal in real time through an acoustic emission sensor. Upon receiving the acoustic emission signal, the treatment device processes it in the following manner:

[0006] The detected signal is subjected to wavelet denoising to obtain the denoised signal;

[0007] Then, based on the cross-correlation theory, the similarity of each denoised signal at different times is calculated, the time domain interval with the highest similarity is found, and the time domain interval is used as the time difference to calculate the signal time difference between each denoised signal;

[0008] Based on the signal time difference, the through wave velocity, and the sensor coordinates, the location of the crack damage source is calculated using the source localization calculation method.

[0009] The penetration velocity is determined based on a pre-conducted lead-breaking test; the sensor coordinates are determined based on the positions of each acoustic emission sensor.

[0010] The principle behind the aforementioned scheme is that asphalt concrete will generate acoustic emission activity when subjected to external forces. By capturing the acoustic emission signals through acoustic emission sensors and calculating the location of the wave source, the sound source of the damage point can be located. Combined with the basic characteristic parameters of acoustic emission, it is possible to achieve comprehensive dynamic monitoring of the asphalt pavement damage process. After adopting the aforementioned scheme, we can obtain information on the location and development process of micro-damage. Combined with vehicle load information captured by video monitoring devices, environmental factors, and on-site investigation, we can more clearly describe the relationship between micro-damage and external forces.

[0011] Preferably, in step 2), the wavelet function used for wavelet denoising is the Symlets wavelet basis, and the number of decomposition layers in the denoising process is 5.

[0012] Preferably, the threshold function used in wavelet denoising is:

[0013]

[0014] In the formula, W j,k These are wavelet coefficients. To estimate the wavelet coefficients, λ is the wavelet threshold.

[0015] Preferably, the acoustic emission sensors are distributed in a matrix form, with the distance between two adjacent acoustic emission sensors on the diagonal being less than 30cm.

[0016] Preferably, the burial depths of the three adjacent acoustic emission sensors forming a right angle are different.

[0017] The beneficial technical effects of this invention are: it proposes a method for locating cracking damage sources in asphalt pavement based on acoustic emission. This method can accurately obtain information on the location and development process of micro-damage, making it possible to more clearly describe the relationship between micro-damage and external effects. Detailed Implementation

[0018] A method for locating cracking damage sources in asphalt pavement based on acoustic emission includes:

[0019] 1) Sensor arrangement: After the asphalt pavement is laid, a monitoring area is set up on the asphalt pavement, and multiple installation holes are drilled within the monitoring area. The multiple installation holes are distributed in a matrix form, and an acoustic emission sensor is installed in each installation hole; each acoustic emission sensor is connected to the processing device.

[0020] 2) After the asphalt pavement is put into use, the treatment device detects the acoustic emission signal in real time through an acoustic emission sensor. Upon receiving the acoustic emission signal, the treatment device processes it in the following manner:

[0021] The detected signal is subjected to wavelet denoising to obtain the denoised signal;

[0022] Then, based on the cross-correlation theory, the similarity of each denoised signal at different times is calculated, the time domain interval with the highest similarity is found, and the time domain interval is used as the time difference to calculate the signal time difference between each denoised signal;

[0023] Based on the signal time difference, the through wave velocity, and the sensor coordinates, the location of the crack damage source is calculated using the source localization calculation method.

[0024] The penetration velocity is determined based on a pre-conducted lead-breaking test; the sensor coordinates are determined based on the positions of each acoustic emission sensor.

[0025] Furthermore, in step 2), the wavelet function used for wavelet denoising is the Symlets wavelet basis, and the decomposition level in the denoising process is 5 levels. There are various wavelet bases available in existing theories, and the following factors are mainly considered when selecting one:

[0026] 1. Meets the requirements of discrete transformation;

[0027] 2. The acoustic emission signal of asphalt pavement materials is a burst-type signal, which requires the wavelet basis to have a strong local signal processing capability. Therefore, a tightly supported wavelet basis is selected.

[0028] 3. In order to decompose the signal into mutually orthogonal subspaces, wavelet bases with orthogonality are selected;

[0029] 4. To accurately perform noise reduction, a wavelet basis of order N is selected;

[0030] 5. In order to avoid or reduce phase distortion as much as possible, wavelet basis with symmetry is selected.

[0031] The wavelet bases that meet the conditions are Daubescies (dbN) and Symlets (symN); then, the information entropy value function is further introduced, given a set of random data X as x1, x2, x3, ..., x n The corresponding probabilities are p1, p2, p3, ..., p n Therefore, the information entropy H(X) of this set of random data X can be defined as follows: From this, we can derive the magnitude of the wavelet basis entropy, among which the Symlets wavelet basis entropy is the smallest, and therefore it is used as the wavelet basis function of the wavelet transform.

[0032] Secondly, it is necessary to determine the number of waveform decomposition layers during the noise reduction process. There has never been a universal theoretical basis for selecting the number of wavelet decomposition layers. This invention uses the SR comprehensive evaluation index to determine the optimal number of decomposition layers. S is the signal-noise ratio (SNR) and R is the root mean square error (RMSE). The number of decomposition layers corresponding to the abrupt change point of the SR index is considered to be the optimal number of decomposition layers. According to the inventor's calculations, the processing effect is best when the number of decomposition layers is 5.

[0033] Furthermore, the threshold function used in wavelet denoising is:

[0034]

[0035] In the formula, W j,k These are wavelet coefficients. To estimate the wavelet coefficients, λ is the wavelet threshold.

[0036] The characteristic of this threshold function is: when |W j,k When |≥λ, the proportion of noise wavelet coefficients is very small and can be ignored; therefore, hard thresholding is used for noise reduction. However, when |W j,k When |<λ, a new exponential function is designed that can retain some wavelet coefficients that are close to λ. Furthermore, due to the rapid decay characteristic of the exponential function, a large number of noisy wavelet coefficients can be quickly filtered out as the wavelet coefficients decrease.

[0037] Furthermore, the acoustic emission sensors are distributed in a matrix, with the distance between two adjacent acoustic emission sensors on the diagonal being less than 30 cm. Since elastic waves attenuate within the material, the distance between the acoustic emission sensors should not be too large. Experiments have verified that the distance between two adjacent acoustic emission sensors should ideally be less than 30 cm.

[0038] Furthermore, the three adjacent acoustic emission sensors forming a right angle are buried at different depths. To achieve three-dimensional positioning, the acoustic emission sensors need to be located on different planes.

Claims

1. A method for locating cracking damage sources in asphalt pavement based on acoustic emission, characterized in that: include: 1) Sensor arrangement: After the asphalt pavement is laid, a monitoring area is set up on the asphalt pavement, and multiple installation holes are drilled within the monitoring area. The multiple installation holes are distributed in a matrix form, and an acoustic emission sensor is installed in each installation hole; each acoustic emission sensor is connected to the processing device. 2) After the asphalt pavement is put into use, the treatment device detects the acoustic emission signal in real time through an acoustic emission sensor. Upon receiving the acoustic emission signal, the treatment device processes it in the following manner: The detected signal is subjected to wavelet denoising to obtain the denoised signal; Then, based on the cross-correlation theory, the similarity of each denoised signal at different times is calculated, the time domain interval with the highest similarity is found, and the time domain interval is used as the time difference to calculate the signal time difference between each denoised signal; Based on the signal time difference, the through wave velocity, and the sensor coordinates, the location of the crack damage source is calculated using the source localization calculation method. The penetration velocity is determined based on a pre-conducted lead-breaking test; the sensor coordinates are determined based on the positions of each acoustic emission sensor. The threshold function used in wavelet denoising is: In the formula, These are wavelet coefficients. To estimate wavelet coefficients, The threshold is the wavelet threshold.

2. The method for locating asphalt pavement cracking damage sources based on acoustic emission according to claim 1, characterized in that: In step 2), the wavelet function used for wavelet denoising is the Symlets wavelet basis, and the number of decomposition layers in the denoising process is 5.

3. The method for locating asphalt pavement cracking damage sources based on acoustic emission according to claim 1, characterized in that: The acoustic emission sensors are distributed in a matrix, with the distance between two adjacent acoustic emission sensors on the diagonal being less than 30cm.

4. The method for locating asphalt pavement cracking damage sources based on acoustic emission according to claim 3, characterized in that: The three adjacent acoustic emission sensors that form a right angle are buried at different depths.

Images