Buried nonmetal pipeline depth positioning method based on acoustic mutual information

Abstract

The invention discloses a buried non-metal pipeline depth positioning method based on acoustic mutual information, and relates to the field of buried pipeline detection, and the method comprises the following steps: firstly, testing the wave velocity in a pipeline-free region, and calculating the wave velocity of longitudinal waves according to the ratio of the distance between longitudinal wave detectors to the delay time of a received signal, and the Poisson's ratio of soil; sound wave signals are collected in a pipeline area; at the same time, de-noising the received signal by using ensemble empirical mode decomposition combined with a wavelet threshold algorithm; a reflected longitudinal wave signal is extracted; establishing a rectangular coordinate system, dividing a measuring surface and calculating longitudinal wave propagation time; in addition, calculating a frequency domain mutual information superposition value to obtain a superposition value matrix; and finally, imaging by using the superposition value matrix to determine the depth of the pipeline. According to the method, the effect of positioning the depth of the buried pipeline is good, the method is not limited by geological conditions, and a reliable technical means is provided for depth positioning of the buried non-metal pipeline.

Description

Technical Field

[0001] The invention relates to the field of buried pipeline detection, and in particular to a buried non-metallic pipeline depth positioning method based on acoustic mutual information. Background Art

[0002] With the acceleration of urbanization and the expansion of underground infrastructure, non-metallic pipes are widely used in important pipe networks such as gas and water supply due to their advantages of light weight, corrosion resistance, and convenient construction. However, due to the long-term use of non-metallic pipes, defects of varying degrees may occur due to various factors. In particular, due to the uncertainty of pipeline location information, problems such as mis-digging and damage frequently occur during construction, which may cause safety accidents such as gas leakage and explosion in serious cases. Traditional positioning methods such as electromagnetic induction, radar ground penetration, and radio frequency identification are greatly affected by the material of the pipeline and the dryness and humidity of the soil. They perform poorly in the positioning of buried pipelines and it is difficult to accurately locate the position of non-metallic pipelines. For this reason, it is crucial to develop more efficient positioning technologies. In recent years, the positioning research of buried non-metallic pipelines has continued to develop. Among them, the positioning method based on acoustic wave reflection is more flexible and low-cost. There are roughly two strategies for this method: one is the time domain signal positioning method, and the other is the frequency domain cross-correlation function positioning method. The principles of the two methods are roughly the same, and the difference is that the selected source signal and the reflected longitudinal wave signal are extracted differently. Although these two methods can achieve pipeline positioning, the calculated pipeline burial depth still has a large deviation from the actual situation.

[0003] Reference 1 (GE Liang, HE Jian, GAO Yan, XIAO Xiaoting, WU Jiaye, WANGTian. Acoustic imaging detection using time domain stacking for buried polyethylene pipelines [J]. ACTA ACUSTICA, 2024, 49 (3): 569-576.) proposed an optimized time domain stacking acoustic imaging method. The positioning depth of non-metallic pipelines with a tested burial depth of 0.4m and 1.0m was not accurate enough, and there was a lack of signal denoising. Reference 2 (Cui X, Gao Y, Muggleton J, et al. Superimposed imaging of acoustic wave reflections for the detection of underground nonmetallic pipelines [J]. Mechanical Systems and Signal Processing, 2024, 209: 111127.) used the strategy of "multi-point emission, multi-point reception and mutual correlation coefficient superposition" to draw the acoustic field map of underground non-metallic pipelines to locate buried non-metallic pipelines. However, this work requires a lot of experimental equipment, complex wiring, and lacks quantitative evaluation of the positioning results. Summary of the invention

[0004] In order to solve the problems existing in the above-mentioned buried non-metallic pipeline positioning method, the present invention proposes a buried non-metallic pipeline depth positioning method based on acoustic mutual information. Firstly, a test device is arranged on the ground in a pipeline-free area, and an acoustic wave signal is generated by source excitation. The acoustic wave propagates through the soil to the underground pipeline, is reflected back to the surface by the pipeline surface and is received by multiple longitudinal wave detectors. The longitudinal wave velocity is calculated according to the ratio of the distance between the longitudinal wave detectors to the delay time of the received signal and the Poisson's ratio of the soil. Then, the acoustic wave signal is collected in the pipeline area. The received signal is denoised by using the ensemble empirical mode decomposition combined with the wavelet threshold denoising method. Then, the reflected longitudinal wave is extracted according to the energy intensity of various sound waves in the received signal and the time sequence of arriving at the longitudinal wave detector. A rectangular coordinate system is established, the measurement surface is divided and the longitudinal wave propagation time is calculated. In addition, the mutual information function between the source signal and each reflected longitudinal wave signal is calculated. By calculating the frequency domain mutual information superposition value of the signal, a superposition value matrix is ​​obtained for each reflection unit based on the calculated longitudinal wave propagation time. Finally, the superposition value matrix is ​​used for imaging, and the pipeline position is obtained according to the maximum frequency domain mutual information superposition value.

[0005] A method for locating the depth of a buried non-metallic pipeline based on acoustic mutual information specifically comprises the following steps:

[0006] S1. Arrange longitudinal wave detectors and seismic sources, calculate the longitudinal wave velocity according to the ratio of the distance of the longitudinal wave detector to the start time of the corresponding received signal and the Poisson's ratio of the soil, and take the average value after multiple measurements;

[0007] The longitudinal wave detectors and the seismic source should be arranged above the soil area without pipelines, and four longitudinal wave detectors should be placed at equal intervals on the same side of the seismic source;

[0008] The longitudinal wave velocity is calculated based on the ratio of the distance of the longitudinal wave detector to the start time of the corresponding received signal and the Poisson's ratio of the soil, and the average is taken after multiple measurements, specifically:

[0009] The distance of the longitudinal wave geophones refers to the distance between each geophone and the earthquake source;

[0010] Then the ratio of the distance of the longitudinal wave detector to the start time of the corresponding received signal is calculated to obtain the surface wave velocity; the surface wave refers to the wave that the source sound wave signal reaches the detector along the soil surface;

[0011] Then the longitudinal wave velocity is obtained through the relationship between the longitudinal wave velocity and the Poisson's ratio of the soil; the longitudinal wave refers to a wave propagated by the source sound wave signal in the soil;

[0012] Finally, the soil Poisson's ratio is used to obtain the longitudinal wave velocity of the site according to the relationship between the surface wave velocity and the longitudinal wave velocity. The soil longitudinal wave velocity can be accurately measured by taking the average value after multiple measurements.

[0013] S2, rearrange the longitudinal wave geophone and the seismic source, excite the seismic source and collect the acoustic wave signal;

[0014] The rearrangement of longitudinal wave geophones and seismic sources refers to selecting soil near the top of the pipeline, arranging a survey line along a direction roughly perpendicular to the distribution of the pipeline, placing longitudinal wave geophones at equal intervals on the survey line, and placing the seismic source between any two geophones;

[0015] The acquisition of the acoustic wave signal is performed by setting the sampling rate and the number of sampling points by the host computer, and each longitudinal wave detector receives the acoustic wave signal and uploads it to the host computer;

[0016] S3, performing collective empirical mode decomposition and wavelet threshold denoising on the sound wave signal collected in step S2 to obtain a denoised received signal; the specific denoising process is:

[0017] Firstly, the signal to be processed is decomposed by collective empirical mode decomposition to obtain a set of natural frequency components of intrinsic mode functions, which are arranged from high frequency to low frequency, wherein the high-frequency intrinsic mode function mainly contains noise, while the low-frequency intrinsic mode function mainly contains the main components of the signal; then, the intrinsic mode function components to be denoised are selected, and the frequency characteristics of each intrinsic mode function component are analyzed to select some high-frequency intrinsic mode function components to be denoised; then, wavelet transform is applied to the selected high-frequency intrinsic mode function components to decompose the signal into approximate coefficients and detail coefficients of different scales, and threshold processing is performed on the detail coefficients to remove the noise component; the threshold processing is performed using a soft threshold; then, the denoised intrinsic mode function components are reconstructed by inverse wavelet transform; finally, the intrinsic mode function components subjected to wavelet threshold processing are recombined together with the unprocessed low-frequency intrinsic mode function components and the residual term to obtain the denoised signal;

[0018] S4, extracting the reflected longitudinal wave from the received signal after denoising in step S3 according to the energy intensity of various sound waves in the received signal in step S2 and the time sequence of arrival at the longitudinal wave detector;

[0019] The various sound waves in the received signal mainly include surface waves, direct waves and reflected longitudinal waves. Since the reflected longitudinal waves are actually used for pipeline positioning, other interference waves need to be eliminated;

[0020] The extraction of the reflected longitudinal wave from the received signal after denoising in step S3 is performed by establishing a model without pipeline as a reference model, and subtracting the signal received under the model without pipeline from the signal received under the model with pipeline, so as to obtain the reflected longitudinal wave signal of the buried pipeline;

[0021] S5, establish a rectangular coordinate system, divide the measurement surface and calculate the propagation time of the longitudinal wave of each reflection unit;

[0022] The rectangular coordinate system is established with the earthquake source as the coordinate origin O, the horizontal rightward direction is the positive direction of the x-axis, the vertical upward direction is the positive direction of the y-axis, and the measurement surface is divided into the third and fourth quadrants;

[0023] The division of the measuring surface refers to the unit division of the measuring surface located in the third and fourth quadrants. The measuring surface is a rectangular area symmetrical about the y-axis, and its height is approximately 3 times the buried depth of the pipeline, and its width is 5 times the rectangular area of ​​the longitudinal wave detector array length. The unit division usually takes the length less than the diameter of the pipeline as the side length, and discretely divides the measuring surface into a number of square units;

[0024] The longitudinal wave propagation time refers to the propagation time of the sound wave from the earthquake source through the soil to the surface of the pipe, then reflected by the surface of the pipe, and transmitted to the surface to be received by the longitudinal wave detector; each reflection unit refers to the assumption that the center point of each unit is the reflection point, that is, each unit has a pipe; the specific method of calculating the propagation time is:

[0025] The height of the measuring surface is H, the width is W, and the side length of the dividing unit is dx. Then the measuring surface has a total of H / dx rows and W / dx columns. The four longitudinal wave geophones are arranged symmetrically about the source, with an adjacent spacing of l. Their coordinates can be recorded as (x k ,0), where x k =-1.5·l+l·(k-1), k=1,...,4, the coordinate of the earthquake source is (0,0), that is, it is located between longitudinal wave detector 2 and longitudinal wave detector 3. Assuming that the buried depth of the pipeline is h, that is, the distance from the upper surface of the pipeline to the ground, and the horizontal distance between the center of the pipeline and the earthquake source is d, the coordinate of the pipeline is (d,-h);

[0026] For each reflection unit (i, j), i = 1, ..., W / dx, j = 1, ..., H / dx, the geometric center coordinates of the unit need to be calculated, cx is the horizontal coordinate of the reflection unit center, cy is the vertical coordinate of the reflection unit center, and the calculation formulas are:

[0027]

[0028] The distance that the sound wave travels from the earthquake source to the surface of the buried pipeline and then reflected by the pipeline surface back to the kth longitudinal wave geophone is:

[0029]

[0030] Let the longitudinal wave velocity be c, then the propagation time of the sound wave from the source to the pipe surface, and then reflected by the pipe surface back to the kth longitudinal wave geophone is:

[0031] t k =S k / c

[0032] Based on this, the longitudinal wave propagation time of each reflection unit can be obtained;

[0033] S6, calculating the mutual information function between the source signal and each reflected longitudinal wave signal; further calculating the frequency domain mutual information superposition value of the signal, for each reflection unit, based on the longitudinal wave propagation time calculated in step S5, obtaining a superposition value matrix;

[0034] The mutual information function between the source signal and each reflected longitudinal wave signal is calculated as follows:

[0035] The source signal is represented by x(t), and the signal received by the longitudinal wave detector is represented by y(t). First, the continuous time signals x(t) and y(t) are sampled to obtain discrete signals x(n) and y(n); then x(n) and y(n) are regarded as random variables, and the joint probability distribution P(x, y) and marginal probability distribution P(x) and P(y) between the x(n) and y(n) signals under different time delays are estimated by statistical histogram data; the mutual information between the x(t) and y(t) signals is calculated based on the two distribution results obtained, and the formula is as follows:

[0036]

[0037] The maximum mutual information corresponds to the best estimate of the time delay of signal y(t) relative to x(t);

[0038] The frequency domain mutual information superposition value of the reflection unit is calculated based on the determination of the maximum mutual information value, and the maximum mutual information value is used as the demarcation point to intercept a mutual information function with a period length backward; the one period length is the signal length of one period of the source signal; then for each reflection unit, according to the propagation time of different longitudinal wave detectors corresponding to each unit obtained in step S5, the intercepted mutual information functions of each path are shifted to the left by the corresponding longitudinal wave propagation time; finally, the intercepted mutual information functions of each path are added to obtain the frequency domain mutual information superposition value of a reflection unit; the remaining reflection units are calculated in this way to obtain a superposition value matrix;

[0039] S7, using the superposition value matrix obtained in step S6 to perform color imaging, the buried depth information of the pipeline can be obtained;

[0040] The two-dimensional imaging of the pipeline position is performed by color imaging of the superposition value matrix obtained in step S6, so as to obtain an imaging map of the detection area. The larger the superposition value, the brighter the color of the corresponding reflection unit. Therefore, the brightest area in the image is the position of the pipeline.

[0041] Due to the adoption of the above technical solution, the present invention has the following advantages:

[0042] 1. The present invention proposes a method for locating the depth of buried non-metallic pipelines based on acoustic mutual information. Compared with the traditional positioning method, the mutual information method is based on statistical analysis and calculates the amount of information by analyzing the frequency domain distribution characteristics of the acoustic wave signal. It is not affected by the dryness and humidity of the soil, can better adapt to the variability of the underground environment, and is applicable to more types of geological conditions;

[0043] 2. The ensemble empirical mode decomposition combined with wavelet threshold denoising algorithm proposed in the present invention has the ability of adaptive decomposition and suppression of mode aliasing, and provides multi-scale denoising, which can effectively separate and remove noise while maintaining the detailed characteristics of the signal, and can significantly improve the depth positioning effect of the pipeline;

[0044] 3. The method of establishing a rectangular coordinate system to calculate the propagation time of a signal proposed by the present invention can simplify the problem, clearly define the position of a point and systematically calculate the distance between any two points, which is convenient for calculation and solution;

[0045] 4. The present invention proposes a method for calculating signal delay using the mutual information function. For nonlinear and non-stationary signals collected by longitudinal wave geophones, it can effectively detect the time delay of the geophone output signal compared to the source signal, thereby achieving high-precision buried pipeline positioning. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 This is a flow chart of a method for locating the depth of a buried pipeline based on acoustic mutual information according to an embodiment of the present invention;

[0047] Figure 2 This is a layout diagram of a longitudinal wave geophone according to an embodiment of the present invention;

[0048] Figure 3 The original signal collected by the longitudinal wave geophone in the embodiment of the present invention;

[0049] Figure 4 The received signal is denoised by using the collective empirical mode decomposition combined with the wavelet threshold algorithm in the embodiment of the present invention;

[0050] Figure 5 This is a time delay estimation curve of the mutual information method according to an embodiment of the present invention;

[0051] Figure 6 This is a pipeline positioning diagram of the denoising mutual information function method according to an embodiment of the present invention;

[0052] Figure 7 This is the pipeline positioning map without denoising mutual information function method;

[0053] Figure 8 This is the pipeline positioning diagram using the denoising cross-correlation function method. DETAILED DESCRIPTION

[0054] In order to further describe the buried non-metallic pipeline depth positioning method based on acoustic mutual information proposed by the present invention, an embodiment of the present invention is described in conjunction with the accompanying drawings to enhance the understanding of the purpose, method structure and application effect of the present invention. The embodiment is only a further explanation of the present invention, and researchers in the field can make further non-fundamental technical improvements on the basis of the present invention, which are also within the scope of protection of the present invention.

[0055] A method for locating the depth of a buried non-metallic pipeline based on acoustic mutual information specifically comprises the following steps:

[0056] S1. Arrange longitudinal wave detectors and seismic sources, calculate the longitudinal wave velocity according to the ratio of the distance of the longitudinal wave detector to the start time of the corresponding received signal and the Poisson's ratio of the soil, and take the average value after multiple measurements;

[0057] The longitudinal wave detectors and the seismic source should be arranged above the soil area without pipelines, and four longitudinal wave detectors should be placed at equal intervals on the same side of the seismic source;

[0058] The longitudinal wave velocity is calculated based on the ratio of the distance of the longitudinal wave detector to the start time of the corresponding received signal and the Poisson's ratio of the soil, and the average is taken after multiple measurements, specifically:

[0059] The distance of the longitudinal wave geophones refers to the distance between each geophone and the earthquake source;

[0060] Then the ratio of the distance of the longitudinal wave detector to the start time of the corresponding received signal is calculated to obtain the surface wave velocity; the surface wave refers to the wave that the source sound wave signal reaches the detector along the soil surface;

[0061] Then the longitudinal wave velocity is obtained through the relationship between the longitudinal wave velocity and the Poisson's ratio of the soil; the longitudinal wave refers to a wave propagated by the source sound wave signal in the soil;

[0062] Finally, the soil Poisson's ratio is used to obtain the longitudinal wave velocity of the site according to the relationship between the surface wave velocity and the longitudinal wave velocity. The soil longitudinal wave velocity can be accurately measured by taking the average value after multiple measurements.

[0063] S2, rearrange the longitudinal wave geophone and the seismic source, excite the seismic source and collect the acoustic wave signal;

[0064] The rearrangement of longitudinal wave geophones and seismic sources refers to selecting soil near the top of the pipeline, arranging a survey line in a direction roughly perpendicular to the distribution of the pipeline, placing longitudinal wave geophones at equal intervals on the survey line, and placing the seismic source between any two geophones. Figure 2 This is a diagram of the arrangement of P-wave geophones on a survey line. The spacing between adjacent P-wave geophones is l = 0.4 m. The earthquake source coordinates are (0, 0), located between P-wave geophones 2 and 3.

[0065] The acquisition of the acoustic wave signal is performed by setting the sampling rate and sampling time of the host computer, which are respectively set to f = 10000 Hz and t = 0.045 s. Each longitudinal wave detector receives the acoustic wave signal and uploads it to the host computer. The original signal collected by the longitudinal wave detector is as follows: Figure 3 As shown;

[0066] S3, performing collective empirical mode decomposition and wavelet threshold denoising on the sound wave signal collected in step S2 to obtain a denoised received signal; the specific denoising process is:

[0067] First, the signal to be processed is decomposed by ensemble empirical mode decomposition to obtain a set of natural frequency components of intrinsic mode functions, which are arranged from high frequency to low frequency, wherein high-frequency intrinsic mode functions may contain noise, while low-frequency intrinsic mode functions mainly contain the main components of the signal; then, the intrinsic mode function components that need to be denoised are selected, and by analyzing the frequency characteristics of each intrinsic mode function component, some high-frequency intrinsic mode function components that need to be denoised are selected; then, wavelet transform is applied to the selected high-frequency intrinsic mode function components to decompose the signal into approximate coefficients and detail coefficients of different scales, and threshold processing is performed on the detail coefficients to remove the noise component; the threshold processing is performed using a soft threshold; then, the denoised intrinsic mode function components are reconstructed by inverse wavelet transform; finally, the intrinsic mode function components that have been processed by wavelet thresholding are recombined together with the unprocessed low-frequency intrinsic mode function components and the residual term to obtain the denoised signal, such as Figure 4 As shown;

[0068] S4, extracting the reflected longitudinal wave from the received signal after denoising in step S3 according to the energy intensity of various sound waves in the received signal in step S2 and the time sequence of arrival at the longitudinal wave detector;

[0069] The various sound waves in the received signal mainly include surface waves, direct waves and reflected longitudinal waves. Since the reflected longitudinal waves are actually used for pipeline positioning, other interference waves need to be eliminated;

[0070] The extraction of reflected longitudinal waves from the received signal after denoising in step S3 is based on the different energy intensities of various sound waves in the received signal and the different arrival times at the longitudinal wave detector. By establishing a model without pipelines as a reference model, the reflected longitudinal wave signal of the buried pipeline can be obtained by subtracting the signal received under the model with pipelines from the signal received under the model without pipelines.

[0071] S5, establish a rectangular coordinate system, divide the measurement surface and calculate the propagation time of the longitudinal wave of each reflection unit;

[0072] The rectangular coordinate system is established with the earthquake source as the coordinate origin O, the horizontal rightward direction is the positive direction of the x-axis, the vertical upward direction is the positive direction of the y-axis, and the measurement surface is divided into the third and fourth quadrants;

[0073] The division of the measuring surface refers to the unit division of the measuring surface located in the third and fourth quadrants. The measuring surface is a rectangular area symmetrical about the y-axis. The height of the measuring surface is approximately 3 times the buried depth of the pipeline. The approximate depth of the pipeline is 0.9m, so the height of the measuring surface is 3m. The width of the measuring surface is approximately 5 times the rectangular area of ​​the length of the longitudinal wave detector array. The length of the longitudinal wave detector array is 1.2m, so the width of the measuring surface is 6m. The unit division is usually based on a length less than the diameter of the pipeline as the side length. The diameter of the pipeline is 0.1m, so the measuring surface is discretely divided into a number of square units with a side length of 0.01m.

[0074] The longitudinal wave propagation time refers to the propagation time of the sound wave from the earthquake source through the soil to the surface of the pipe, then reflected by the surface of the pipe, and transmitted to the surface to be received by the longitudinal wave detector; each reflection unit refers to the assumption that the center point of each unit is the reflection point, that is, each unit has a pipe; the specific method of calculating the propagation time is:

[0075] It is known that the height of the measuring surface is 3m, the width is 6m, and the side length of the division unit is 0.01m. Then the measuring surface has 300 rows and 600 columns. The four longitudinal wave geophones are arranged symmetrically about the source, and the adjacent spacing is l = 0.4m. Their coordinates can be recorded as (x k ,0), where x k =-1.5·l+l·(k-1), k=1,...,4, the coordinates of the earthquake source are (0,0), and the buried depth of the pipeline is 0.9m (the distance from the upper surface of the pipeline to the ground);

[0076] For each reflection unit (i, j), i = 1, ..., 600, j = 1, ..., 300, the geometric center coordinates of the unit need to be calculated, cx is the horizontal coordinate of the reflection unit center, cy is the vertical coordinate of the reflection unit center, and the calculation formulas are:

[0077]

[0078] The distance that the sound wave travels from the earthquake source to the surface of the buried pipeline and then reflected by the pipeline surface back to the kth longitudinal wave geophone is:

[0079]

[0080] Let the longitudinal wave velocity be c, then the propagation time of the sound wave from the source to the pipe surface, and then reflected by the pipe surface back to the kth longitudinal wave geophone is:

[0081] t k =S k / c

[0082] According to this formula, the longitudinal wave propagation time of each reflection unit can be obtained;

[0083] S6, calculating the mutual information function between the source signal and each reflected longitudinal wave signal; further calculating the frequency domain mutual information superposition value of the signal, for each reflection unit, based on the longitudinal wave propagation time calculated in step S5, obtaining a superposition value matrix;

[0084] The mutual information function between the source signal and each reflected longitudinal wave signal is calculated as follows:

[0085] The source signal is represented by x(t), and the signal received by the longitudinal wave detector is represented by y(t). First, the continuous time signals x(t) and y(t) are sampled to obtain discrete signals x(n) and y(n); then x(n) and y(n) are regarded as random variables, and the joint probability distribution P(x, y) and marginal probability distribution P(x) and P(y) between the x(n) and y(n) signals under different time delays are estimated by statistical histogram data; the mutual information between the x(t) and y(t) signals is calculated based on the two distribution results obtained, and the formula is as follows:

[0086]

[0087] The maximum mutual information value corresponds to the best time delay estimate of signal y(t) relative to x(t). The calculation result is as follows: Figure 5 As shown;

[0088] The frequency domain mutual information superposition value of the reflection unit is calculated based on the determination of the maximum mutual information value, and the maximum mutual information value is used as the demarcation point to intercept a mutual information function with a period length backward; the one period length is the length of one period of the source signal; then for each reflection unit, according to the longitudinal wave propagation time obtained in step S5, the intercepted mutual information functions of each path are shifted to the left by the corresponding longitudinal wave propagation time; finally, the intercepted mutual information functions of each path are added to obtain the frequency domain mutual information superposition value of a reflection unit; the remaining reflection units are calculated in this way to obtain a superposition value matrix;

[0089] S7, using the superposition value matrix obtained in step S6 to perform color imaging, the buried depth information of the pipeline can be obtained;

[0090] The two-dimensional imaging of the pipeline position is performed by color imaging of the superposition value matrix obtained in step S6, so as to obtain an imaging map of the detection area. The larger the superposition value, the brighter the color of the corresponding reflection unit. Therefore, the brightest area in the image is the position of the pipeline.

[0091] In order to verify the effectiveness of the proposed collective empirical mode decomposition combined with the wavelet threshold denoising algorithm and the mutual information function method for calculating the signal delay, different positioning methods were selected for the non-metallic pipeline with a buried depth of 0.9m to compare the positioning effect. In order to verify the effectiveness of the denoising process for pipeline depth positioning, the longitudinal wave geophone received signal was processed by the denoised mutual information function method and the non-denoised mutual information function method; in order to verify the effectiveness of different delay calculation methods for pipeline depth positioning, the longitudinal wave geophone received signal was processed by the denoised mutual information function method and the denoised mutual correlation function method. All the methods to be compared in the experiment are based on the experimental environment constructed above. The depth of pipeline positioning is an important indicator to measure the quality of the positioning method. Therefore, the present invention selects the positioning depth as the evaluation indicator of the comparative experiment.

[0092] The results of the pipeline depth positioning method and the comparative method are shown in Table 1: As can be seen from Table 1, the pipeline depth positioning method of the present invention achieves a more accurate positioning depth than the comparative method. Compared with the positioning method of processing the detector receiving signal by the non-denoising mutual information function method, the positioning error of the present invention method is 0.61m lower than that; compared with the positioning method of processing the detector receiving signal by the denoising mutual correlation function method, the positioning error of the present invention method is 0.19m lower than that. The positioning results of different pipeline positioning methods are shown in Table 1. Figure 6 to Figure 8 As shown in the figure, the dotted circle represents the pipeline positioning position, and the solid circle represents the actual position of the pipeline. Figure 6 to Figure 8 It can be seen that the positioning depth of the method of the present invention is closer to the actual depth of the pipeline, and the positioning depth effect is better.

[0093] Table 1 Performance comparison of different pipeline positioning methods

[0094]

[0095] The above simulation experimental results fully demonstrate that the method of the present invention is more accurate in locating buried pipelines, provides technical support for accurately locating the buried depth of pipelines, and has good application prospects.

[0096] The above is an explanation of the method and system of the present invention, and researchers in the field can reproduce the present invention through these explanations. Based on the content of the present invention, other embodiments and results obtained by researchers without creative work should fall within the scope of protection of the present invention.

Claims

1. A method for deep positioning of buried non-metallic pipelines based on acoustic mutual information, characterized in that: The following steps are involved: S1, arranging seismic sources and longitudinal wave detectors, calculating the longitudinal wave velocity according to the ratio of the distance between the longitudinal wave detectors to the start time of the received signal and the Poisson's ratio of the soil, and taking the average value after multiple measurements; the seismic sources and longitudinal wave detectors are arranged above the soil area without pipelines, and four longitudinal wave detectors with equal spacing are placed on the same side of the seismic sources; the surface wave velocity can be obtained according to the ratio of the distance between the longitudinal wave detectors and the start time of the corresponding received signal; and then the longitudinal wave velocity is obtained through the relationship between the longitudinal wave and the Poisson's ratio of the soil; S2, rearrange the seismic source and longitudinal wave geophone, generate acoustic wave signals by using the seismic source excitation, and collect the acoustic wave signals by the longitudinal wave geophone; the said rearrangement of the longitudinal wave geophone and the seismic source means selecting the soil near the top of the pipeline, arranging the survey line along the direction roughly perpendicular to the distribution of the pipeline, placing longitudinal wave geophones at equal intervals on the survey line, and placing the seismic source between any two geophones; S3, performing collective empirical mode decomposition and wavelet threshold denoising on the sound wave signal collected in step S2 to obtain a denoised received signal; S4, extracting the reflected longitudinal wave from the received signal after denoising in step S3 according to the energy intensity of various sound waves in the received signal in step S2 and the time sequence of arriving at the longitudinal wave detector; by establishing a model without pipeline as a reference model, subtracting the signal received under the model without pipeline from the signal received under the model with pipeline, the reflected longitudinal wave signal of the buried non-metallic pipeline can be obtained; S5. Establish a rectangular coordinate system, divide the measurement surface and calculate the longitudinal wave propagation time of each reflection unit; the measurement surface division refers to the unit division of the measurement surface located in the third and fourth quadrants, the measurement surface is a rectangular area symmetrical about the y-axis, its height is approximately 3 times the buried depth of the pipeline, and its width is 5 times the rectangular area of ​​the longitudinal wave detector array length; the unit division usually takes a length less than the diameter of the pipeline as the side length, and discretely divides the measurement surface into a number of square units; S6, calculating the mutual information function between the source signal and each reflected longitudinal wave signal; further calculating the frequency domain mutual information superposition value of the signal, for each reflection unit, based on the longitudinal wave propagation time calculated in step S5, obtaining a superposition value matrix; S7. Use the superposition value matrix obtained in step S6 to perform two-dimensional color imaging to obtain the buried depth information of the pipeline; in the imaging map of the detection area, the larger the superposition value, the brighter the color of the corresponding reflection unit, and the area with the brightest color in the map is the location of the pipeline.

2. The method for locating the depth of buried non-metallic pipelines based on acoustic mutual information according to claim 1 is characterized in that: Step S3 performs collective empirical mode decomposition and wavelet threshold denoising on the sound wave signal collected in step S2 to obtain a denoised received signal. The specific denoising process is as follows: First, the signal to be processed is decomposed by a set of empirical mode decomposition to obtain a set of natural frequency components of intrinsic mode functions, wherein these intrinsic mode function components are arranged from high frequency to low frequency, wherein the high-frequency intrinsic mode function mainly contains noise, and the low-frequency intrinsic mode function mainly contains the main components of the signal; then, by analyzing the frequency characteristics of each intrinsic mode function component, some high-frequency intrinsic mode function components that need to be denoised are selected; then, wavelet transform is applied to the selected high-frequency intrinsic mode function components to decompose the signal into approximate coefficients and detail coefficients of different scales, and threshold processing is performed on the detail coefficients to remove the noise component; the threshold processing can be performed by soft threshold processing; Then, the denoised intrinsic mode function components are reconstructed through inverse wavelet transform; finally, the intrinsic mode function components processed by wavelet threshold are recombined with the unprocessed low-frequency intrinsic mode function components and residual terms to obtain the denoised signal.

3. The method for locating the depth of buried non-metallic pipelines based on acoustic mutual information according to claim 1 is characterized in that: The rectangular coordinate system established in step S5 is based on the earthquake source as the coordinate origin O, the horizontal rightward direction is the positive direction of the x-axis, the vertical upward direction is the positive direction of the y-axis, and the measurement surface is divided into the third and fourth quadrants; The longitudinal wave propagation time refers to the propagation time of the sound wave from the earthquake source through the soil to the surface of the pipe, then reflected by the surface of the pipe, and transmitted to the surface to be received by the longitudinal wave detector; each reflection unit refers to the assumption that the center point of each unit is the reflection point, that is, each unit has a pipe; the specific method of calculating the propagation time is: The height of the measuring surface is H, the width is W, and the side length of the dividing unit is dx. Then the measuring surface has a total of H / dx rows and W / dx columns. The four longitudinal wave geophones are arranged symmetrically about the source, with an adjacent spacing of l. Their coordinates can be recorded as (x k ,0), where x k =-1.5·l+l·(k-1), k=1,...,4, the coordinate of the earthquake source is (0,0), that is, it is located between longitudinal wave detector 2 and longitudinal wave detector 3. Assuming that the buried depth of the pipeline is h, that is, the distance from the upper surface of the pipeline to the ground, and the horizontal distance between the center of the pipeline and the earthquake source is d, the coordinate of the pipeline is (d,-h); For each reflection unit (i, j), i = 1, ..., W / dx, j = 1, ..., H / dx, the geometric center coordinates of the unit need to be calculated, cx is the horizontal coordinate of the reflection unit center, cy is the vertical coordinate of the reflection unit center, and the calculation formulas are: The distance that the sound wave travels from the earthquake source to the surface of the buried pipeline and then reflected by the pipeline surface back to the kth longitudinal wave geophone is: Let the longitudinal wave velocity be c, then the propagation time of the sound wave from the source to the pipe surface, and then reflected by the pipe surface back to the kth longitudinal wave geophone is: t k =S k / c Based on this, the longitudinal wave propagation time of each reflection unit can be obtained.

4. The method for locating the depth of buried non-metallic pipelines based on acoustic mutual information according to claim 1 is characterized in that: Step S6 is to calculate the mutual information function between the source signal and each reflected longitudinal wave signal, and the calculation process is as follows: The source signal is represented by x(t), and the signal received by the longitudinal wave detector is represented by y(t). First, the continuous time signals x(t) and y(t) are sampled to obtain discrete signals x(n) and y(n); then x(n) and y(n) are regarded as random variables, and the joint probability distribution P(x, y) and marginal probability distribution P(x) and P(y) between the x(n) and y(n) signals under different time delays are estimated by statistical histogram data; the mutual information function between the x(t) and y(t) signals is calculated based on the two distribution results obtained, and the formula is as follows: The maximum value of the mutual information function corresponds to the best estimate of the time delay of the signal y(t) relative to x(t); The frequency domain mutual information superposition value of the reflection unit is calculated based on the determination of the maximum mutual information value, with the maximum mutual information value as the dividing point, and a mutual information function with a period length is intercepted backward; the one period length is the signal length of one period of the source signal; then for each reflection unit, according to the longitudinal wave propagation time obtained in step S5, the intercepted mutual information functions are shifted to the left by the corresponding longitudinal wave propagation time; finally, the intercepted mutual information functions are added to obtain the frequency domain mutual information superposition value of a reflection unit; the remaining reflection units are calculated in this way to obtain a superposition value matrix.

Images