A DAS-based underground continuous wall pouring quality detection device and method

Abstract

The present invention discloses a device and method for detecting the pouring quality of underground continuous wall based on distributed acoustic wave sensing DAS. The detection device includes a vibration strain sensing optical cable, a vibration unit, a distributed acoustic wave sensing demodulator, a data processing unit, a result analysis unit and a pouring quality detection unit. The detection method is as follows: the distributed acoustic wave sensing demodulator transmits the vibration signal to the data processing unit, and the data processing unit performs denoising and cross-correlation processing on the vibration data to obtain a correlation spectrum between the signals; the result analysis unit uses the correlation spectrum obtained by the data processing unit to judge the pouring uneven area in the underground continuous wall, and the pouring quality detection unit determines the pouring quality of the underground continuous wall. The present invention adopts a distributed acoustic wave sensing method to obtain a vibration signal, and applies a cross-correlation method to analyze the correlation of the vibration signals between each slot section to determine the uneven pouring position and degree of the underground continuous wall, and finally realizes the pouring quality detection of the underground continuous wall.

Description

Technical Field

[0001] The present invention relates to the technical field of underground continuous wall quality detection, and in particular to an underground continuous wall pouring quality detection device and method based on distributed acoustic wave sensing (DAS). Background Art

[0002] The underground continuous wall is a reinforced concrete wall built below the ground surface during the construction of underground projects. This technology is widely used in foundation pit projects and shield working shafts, and has the functions of water interception, anti-seepage, load-bearing and water-retaining.

[0003] The construction of underground continuous walls includes the steps of grooving, pouring mud, hanging steel cages, and pouring concrete. The process of pouring concrete is extremely critical. During this process, the concrete will gradually replace the mud in the groove. If the mud replacement is not thorough enough, the concrete will be mixed with soil that does not match its density and strength, and even form voids, which to a certain extent affects the strength and waterproof performance of the underground continuous wall.

[0004] At present, the main detection methods for underground continuous wall casting quality include ultrasonic transmission method, resistivity method, high-density electrical method, core drilling method and distributed optical fiber temperature measurement method.

[0005] Among them, the ultrasonic transmission method refers to the use of an acoustic wave device to emit and receive sound waves to the completed underground continuous wall, and judge the casting quality of the underground continuous wall by detecting the changes in acoustic parameters such as sound velocity, frequency and amplitude of the sound waves propagating in the underground continuous wall. Although this method can obtain the casting quality information of underground continuous walls at a certain depth, due to the limited transmission capacity of sound waves and the difficulty in receiving their reflected waves, it is impossible to detect the quality of underground continuous walls at a greater depth. At the same time, this method cannot carry out long-term monitoring of the quality of underground continuous walls.

[0006] The resistivity method is a method of identifying the type of medium in the underground continuous wall by utilizing the resistivity difference between different media in the underground continuous wall. Although this method can effectively identify the type of medium, its identification range is small and it cannot monitor in real time. The principle of the high-density electrical method is the same as that of the resistivity method. The difference is that this method sets a higher density of measuring points in the observation. When measuring on-site, it is only necessary to arrange all electrodes at measuring points at a certain interval and then observe. Although this method can effectively monitor the integrity of the underground continuous wall, the problem of longitudinal resolution when detecting hidden dangers has not been solved. The core drilling method verifies quality accidents such as unqualified wall quality detected by the ultrasonic transmission method by coring and drilling. This method is costly and cannot complete the quality inspection of underground continuous walls alone.

[0007] The distributed fiber optic temperature measurement method uses a distributed fiber optic temperature measurement system based on ROTDR to obtain temperature distribution information on the entire optical fiber, and determines the type of medium the optical fiber passes through by the thermal conductivity of the medium, thereby determining the incomplete parts of the underground continuous wall. Although this method can measure the integrity of the underground continuous wall in a distributed manner, the cost of the temperature measurement optical cable is relatively high and is not suitable for long-distance monitoring.

[0008] How to determine the uneven location and degree of underground continuous wall pouring by analyzing the correlation of vibration signals between the various slot sections of the underground continuous wall is a technical problem that needs to be solved urgently. Summary of the invention

[0009] Purpose of the invention: In view of the technical problems existing in the prior art, the present invention proposes a DAS-based underground continuous wall casting quality detection device and method, in which a vibration strain sensing optical cable is laid in the underground continuous wall, and distributed acoustic wave sensing is used to collect vibration data during the underground continuous wall casting process. After filtering and denoising the vibration data, the correlation of vibration signals between each groove section is analyzed by the cross-correlation method to determine the uneven location and degree of the underground continuous wall casting, and finally realize the detection of the underground continuous wall casting quality.

[0010] Technical solution: The DAS-based underground continuous wall pouring quality detection device of the present invention includes a vibration strain sensing optical cable, a vibration unit, a distributed acoustic wave sensor demodulator, a data processing unit, a result analysis unit, a pouring quality detection unit, a steel cage and a fixing rod;

[0011] The vibration unit includes a vibration controller and a vibration exciter;

[0012] The vibration strain sensing optical cable is laid on the steel cage and connected to the distributed acoustic wave sensor demodulator; the distributed acoustic wave sensor demodulator transmits the collected vibration data to the data processing unit;

[0013] The data processing unit transmits the correlation map between the vibration signals obtained by performing noise reduction processing on the vibration data to the result analysis unit.

[0014] The DAS-based underground continuous wall pouring quality detection method of the present invention is implemented by a DAS-based underground continuous wall pouring quality detection device, and the method comprises the following steps:

[0015] Step (1), drilling a hole beside the underground continuous wall to be cast, and setting a fixing rod in the hole; before the steel cage is lowered, the vibration strain sensor optical cable is fixedly arranged on the main reinforcement of the steel cage; after the steel cage is lowered to the bottom of the underground continuous wall, starting from the bottom of the underground continuous wall, an exciter is arranged on the fixing rod at a set interval;

[0016] Step (2), connecting the vibration strain sensing optical cable to the distributed acoustic wave sensor demodulator through an optical fiber jumper;

[0017] Step (3), starting from the bottom of the underground continuous wall, each time a set depth of concrete is poured, the vibration controller controls the vibrator at the corresponding depth to generate a pulse vibration with a sine wave waveform at a time interval; pouring is continued to excite the vibrator on the fixed rod from the bottom to the top in sequence;

[0018] Step (4), the distributed acoustic wave sensor demodulator collects vibration data of the vibration strain sensor optical cable within the time interval when the exciter generates vibration;

[0019] Step (5), the distributed acoustic wave sensor demodulator transmits the collected vibration data to the data processing unit. After the data processing unit performs noise reduction processing on the vibration data, multiple parallel channels are set at multiple depths within the control area of ​​an exciter, and the vibration signals after noise reduction processing in each channel at the same depth are cross-correlated to obtain a correlation spectrum between the vibration signals. The process of cross-correlation processing is as follows:

[0020] Select the signals f(x) and g(x) in the two channels at the same depth for cross-correlation analysis. The cross-correlation function R of the signals f(x) and g(x) is fg Defined as:

[0021]

[0022] Where m is the time difference between f(x) and g(x); the above formula means that after the signal g(x) is shifted to the left by m times, the sum of the products of the function values ​​of f and g after the shift at any time in the entire time domain is calculated; the cross-correlation function R fg represents the distribution of the correlation between f and g after sliding with m;

[0023] The xcorr function in Matlab is used to process the vibration signals in each channel at the same depth. The process is as follows: the collected vibration signals are loaded into Matlab, and the data of each channel is stored in a matrix, with each row representing the signal of a channel; then the xcorr function is used to calculate the correlation coefficient of the vibration signals between different channels; the xcorr function code is as follows:

[0024] [r, lags] = xcorr(x, y)

[0025] Where [r, lags] are the two return values ​​of the xcorr function, r is the correlation coefficient, and its value range is [0, 1]; lags is the time difference between the two signals, x is the vibration signal of any channel, and y is the vibration signal of the remaining channels;

[0026] The correlation coefficient r tends to 1, indicating that the concrete backfill between the two channels is uniform; if the correlation coefficient fluctuates, it indicates that there are impurities or voids in the concrete between the two channels;

[0027] Step (6), the result analysis unit detects the abnormal protrusion in the signal correlation map after the cross-correlation processing, and then determines the depth interval where the underground continuous wall has the uneven casting area;

[0028] In step (7), the result analysis unit transmits the depth interval of the uneven casting area of ​​the underground continuous wall to the casting quality detection unit. The casting quality detection unit sets a correlation threshold according to the mix ratio and strength of the concrete. When the correlation coefficient r is lower than the correlation threshold, the casting of the underground continuous wall is uneven.

[0029] In step (5), at the same depth, the vibration signals between every two channels are subjected to a cross-correlation operation. If there are n channels, the number of cross-correlation operations S is:

[0030]

[0031] The S processing results are averaged to obtain a correlation map of the correlation coefficient at the depth and the length of the underground continuous wall.

[0032] In step (1), the vibration strain sensing optical cable is laid in a U-shape on the main reinforcement of the underground continuous wall steel cage in a fixed-point pasting manner.

[0033] In step (2), the vibration strain sensing optical cable is connected to the distributed acoustic wave sensor demodulator through an optical fiber jumper, and the sampling frequency, channel spacing and number of channels of the demodulator are set.

[0034] In step (3), starting from the bottom of the underground continuous wall, each time a set depth of concrete is poured, the vibration controller controls the exciter at the corresponding depth to generate a transverse pulse vibration with a time interval and a sinusoidal waveform.

[0035] In step (5), the noise reduction process is as follows: setting the filter frequency band and performing bandpass filtering on the vibration data; then superimposing the vibration data collected after multiple vibrations to obtain a vibration signal with a distinct waveform.

[0036] In step (5), if the correlation coefficient fluctuates, it indicates that there are impurities or voids in the concrete between the two channels, and the impurities and voids are determined by the degree of reduction of the correlation coefficient.

[0037] In step (5), when the correlation coefficient decreases slightly, impurities exist in the concrete.

[0038] In step (5), when the correlation coefficient decreases significantly, there are voids in the concrete.

[0039] Working principle: Distributed acoustic sensing technology DAS is an advanced sensing technology that uses coherent backscattered Rayleigh light in optical fiber to detect and analyze long-distance propagating sound waves or vibration signals. It has the advantages of long-distance monitoring, high sensitivity, high spatial resolution, low cost, passivity, anti-electromagnetic interference and easy installation. Although this technology is used to detect the integrity of bored piles, the vertical excitation vibration wave makes this technology limited to small engineering scenes such as drilling, and the data processing and analysis methods based on wave velocity inversion have large computational complexity, which is not conducive to engineering applications.

[0040] The vibration strain sensing optical cable of the present invention is connected to a distributed acoustic wave sensor demodulator; after the underground continuous wall is poured, a vibration unit is used to generate vibration; the distributed acoustic wave sensor demodulator collects vibration data and transmits the vibration data to a data processing unit; the data processing unit is connected to a result analysis unit; and the result analysis unit is connected to a pouring quality detection unit.

[0041] Among them, the vibration strain sensing optical cable is fixed on the main reinforcement of the underground continuous wall steel cage by fixed-point pasting, and during the laying process, the optical cable and optical fiber jumper are reserved for welding; after the optical cable is laid, the channel point of signal collection is determined by the knocking positioning method.

[0042] When the present invention performs signal cross-correlation processing, when the vibration signal propagates in the medium, only the signal amplitude is attenuated without changing the waveform of the vibration signal. The formula is as follows:

[0043] A(x, t) = A0e -αx cos(wt-kx+φ)

[0044] Where: A(x, t) is the instantaneous amplitude of the sound wave, x is the horizontal distance from the exciter, t is the interval time, A0 is the initial amplitude of the sound wave, α is the attenuation coefficient, ω is the angular frequency of the vibration signal, k is the wave number of the sound wave, which is related to the refractive index of the medium, and φ is the initial phase of the sound wave.

[0045] Since x is the horizontal distance between the signal receiving point and the exciter, the vibration signals at the same depth are selected for cross-correlation processing; k is related to the refractive index of the medium, so when the type of medium changes, the waveform of the entire vibration signal will change, and this can be used to judge whether the medium is uniform during signal propagation.

[0046] In step (5), the cross-correlation processing operation is as follows:

[0047] Select the signals f(x) and g(x) from the two channels at the same depth for cross-correlation analysis. Define the signal of channel 1 as f(x) and the signal of channel 2 as g(x). The cross-correlation function R of the two signals is fgDefined as:

[0048]

[0049] Wherein, m is the time slip. In the present invention, m is defined as the time difference between f(x) and g(x).

[0050] The above formula means that after the signal g(x) is shifted to the left by m time periods, the sum of the products of the function values ​​of f and g after the shift at any time in the entire time domain is calculated. There is one sum corresponding to each m, so the cross-correlation function represents the distribution of the correlation between f and g after the shift with m.

[0051] The xcorr function in Matlab is used to process the vibration signals in each channel at the same depth. The process is as follows: the collected vibration signals are loaded into Matlab, and the data of each channel is stored in a matrix, with each row representing the signal of a channel; then the xcorr function is used to calculate the correlation coefficient of the vibration signals between different channels; the xcorr function code is as follows:

[0052] [r, lags] = xcorr(x, y)

[0053] Among them, [r, lags] are the two return values ​​of the xcorr function; r is the correlation coefficient, lags is the time difference between the two signals, x is the vibration signal of any channel, and y is the vibration signal of the remaining channels. The value range of the correlation coefficient r is [0, 1]. The correlation represents the similarity of the waveforms between the two signals. Combined with the propagation law of vibration waves in the medium, it reflects whether the medium between the channels is uniform. The closer the correlation coefficient is to 1, the more uniform the medium between the channels is, that is, the better the correlation between the two signals, and the more uniform the corresponding underground continuous wall casting is.

[0054] In step (5), at the same depth, the vibration signals between every two channels are subjected to a cross-correlation operation. If there are n channels, the number of cross-correlation operations required is S:

[0055]

[0056] The S processing results are averaged. The averaging process is as follows: by using the moving average method, the window size is selected according to the size of S, starting from the beginning of the data sequence, the average value of all data points in the first window is calculated. Then, the window is slid forward by one data point, and the average value of the data points in the next window is calculated. This process is repeated until the entire data sequence is covered. After averaging, the correlation map of the correlation coefficient at this depth and the length of the underground continuous wall is obtained.

[0057] In step (6), the result analysis unit detects the abnormal protrusions in the correlation map and preliminarily determines the scope of the uneven casting area of ​​the underground continuous wall.

[0058] In step (7), the casting quality detection unit sets two thresholds r1 and r2 in the correlation map, and divides the casting quality of the underground continuous wall of each trench section into three levels. The part with a correlation coefficient r higher than r1 has good casting quality, and when it is between r1 and r2, the casting quality is average; when the correlation coefficient is lower than r2, the casting quality is poor, wherein the values ​​of r1 and r2 are determined by the mix ratio and strength of concrete.

[0059] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0060] (1) Compared with the prior art, the distributed acoustic wave sensing method adopted in the present invention has a test frequency as wide as 0.1 Hz to 40 kHz, which meets the accuracy requirements of various quality problems in the process of pouring underground continuous walls, thereby solving the problem of blind spots in the acoustic wave transmission method.

[0061] (2) The distributed acoustic wave sensing method adopted by the present invention has a long detection distance, realizes the deep detection of the casting quality of underground continuous walls, and solves the problem of the narrow detection range of the acoustic wave transmission method.

[0062] (3) The present invention uses a method of exciting transverse vibration waves, utilizes the propagation law of vibration waves in the medium, and combines a cross-correlation signal processing method to determine the integrity of the underground continuous wall in the middle area of ​​the two optical cables, thereby achieving distributed detection.

[0063] (4) The distributed acoustic wave sensing method adopted by the present invention has strong anti-electromagnetic interference capability and can adapt to the harsh underground environment. At the same time, the strong anti-corrosion ability of optical fiber also enhances the environmental adaptability of the technology, improves the detection depth of the underground continuous wall casting quality, reduces the detection cost, and realizes long-term detection of the underground continuous wall quality.

[0064] (5) After the underground continuous wall is put into use, the distributed acoustic wave sensing method is also used to detect seismic waves. Therefore, the vibration strain sensing optical cable laid in the underground continuous wall is also used as a seismic monitoring sensor.

[0065] (6) The present invention utilizes the cross-correlation method to process the vibration signal, effectively judges the casting quality of each part in the underground continuous wall, and accurately locates the uneven casting parts of the underground continuous wall, and uses this as a basis to judge whether the part needs to take remedial measures such as grouting reinforcement. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1 It is a structural schematic diagram of the underground continuous wall pouring quality detection device based on DAS of the present invention;

[0067] Figure 2 This is a schematic diagram of the layout of the uplink optical cable on a steel cage of a certain slot section of the present invention;

[0068] Figure 3 This is a schematic diagram of the correlation spectrum obtained by cross-correlation processing in the present invention. DETAILED DESCRIPTION

[0069] like Figure 1 As shown, the DAS-based underground continuous wall pouring quality detection device of the present invention includes a vibration strain sensing optical cable 1, a vibration unit 2, a distributed acoustic wave sensor demodulator 3, a data processing unit 4, a result analysis unit 5, a pouring quality detection unit 6, a steel cage 7 and a fixing rod 8. The vibration unit 2 includes a vibration controller 2-1 and a vibration exciter 2-2.

[0070] The vibration strain sensing optical cable 1 is pre-laid on the steel cage 7 of the underground continuous wall; one end of the distributed optical fiber acoustic wave sensor demodulator 3 is connected to the vibration strain sensing optical cable 1 through a jumper to collect data, and the other end is connected to the data processing unit 4. The exciter 2-2 is fixed on the fixed rod 8 in the borehole next to the underground continuous wall at a certain interval, and the vibration waveform and frequency of the exciter 2-2 are controlled by the vibration controller 2-1. The distributed acoustic wave sensor demodulator 3 collects the vibration data generated by the vibration of the exciter 2-2 in the vibration strain sensing optical cable 1; the distributed acoustic wave sensor demodulator 3 transmits the vibration data to the data processing unit 4; the data processing unit 4 performs noise reduction processing on the vibration data to obtain a vibration signal with a high signal-to-noise ratio, and then uses the cross-correlation method to obtain the correlation map between the vibration signals of each channel and transmits it to the result analysis unit 5; the result analysis unit 5 uses the vibration signal correlation map to determine the uneven backfill area of ​​the borehole; the casting quality detection unit 6 detects the uneven casting degree of the underground continuous wall and finally determines the casting quality of the underground continuous wall.

[0071] The DAS-based underground continuous wall pouring quality detection method of the present invention comprises the following steps:

[0072] (1) Drill a hole beside the underground continuous wall to be cast, and set a fixing rod 8 in the hole; before the steel cage is lowered, the vibration strain sensing optical cable 1 is fixedly arranged on the main reinforcement of the steel cage; after the steel cage is lowered to the bottom of the underground continuous wall, starting from the bottom of the underground continuous wall, an exciter 2-2 is arranged on the fixing rod 8 at a set interval; and concrete is used to cast each groove section of the underground continuous wall.

[0073] (2) connecting the vibration strain sensing optical cable 1 to the distributed acoustic wave sensing demodulator 3 through an optical fiber jumper;

[0074] (3) Starting from the bottom of the underground continuous wall, each time a set depth of concrete is poured, the vibration controller 2-1 controls the vibrator 2-2 at the corresponding depth to generate a pulse vibration with a time interval and a waveform of a sine wave; pouring is continued, so that the vibrator on the fixed rod is excited in sequence from the bottom to the top;

[0075] (4) the distributed acoustic wave sensor demodulator 3 collects vibration data of the vibration strain sensing optical cable 1 within the time interval when the vibration exciter 2-2 generates vibration;

[0076] (5) The distributed acoustic wave sensor demodulator 3 transmits the collected vibration data to the data processing unit 4. After the data processing unit 4 performs noise reduction processing on the vibration data, multiple parallel channels are set at multiple depths within the control area of ​​an exciter 2-2, and the vibration signals after noise reduction processing in each channel at the same depth are cross-correlated to obtain a correlation spectrum between the vibration signals;

[0077] (6) The result analysis unit 5 detects the abnormal protrusion in the signal correlation map after the cross-correlation processing, and then determines the depth interval where the underground continuous wall has the uneven casting area;

[0078] (7) The result analysis unit 5 transmits the depth interval of the uneven casting area of ​​the underground continuous wall to the casting quality detection unit 6. The casting quality detection unit 6 sets a correlation threshold according to the mix ratio and strength of the concrete. When the correlation coefficient r is lower than the correlation threshold, the casting of the underground continuous wall is uneven.

[0079] Among them, in step (1), the vibration strain sensing optical cable 1 is fixed on the main reinforcement of the underground continuous wall steel cage by fixed-point pasting, and is symmetrically arranged in a "U" shape in each groove section. During the laying process, the reserved optical cable is fused with the optical fiber jumper.

[0080] In step (1), the drilling depth is greater than or equal to the height of the underground continuous wall. The fixed rod arranged in the drill hole is used to fix the exciter. Starting from the bottom of the underground continuous wall, an exciter is arranged at a certain interval. The vibration unit is used to generate transverse vibration waves in sequence, and the vibration waveform and frequency of the exciter are controlled in sequence by the vibration controller.

[0081] In step (3), the vibrator is excited sequentially from the bottom to the top as the pouring process proceeds.

[0082] In step (5), the noise reduction process includes the following steps: first, setting the filter frequency band and performing bandpass filtering on the vibration data; second, in order to improve the signal-to-noise ratio of the data, superimposing the vibration data collected after multiple vibrations to obtain a vibration signal with a distinct waveform.

[0083] In step (5), when the signal is cross-correlated, when the vibration signal propagates in the medium, only the signal amplitude is attenuated without changing the waveform of the vibration signal. The formula is as follows:

[0084] A(x, t) = A0e -αx cos(wt-kx+φ)

[0085] Where: A(x, t) is the instantaneous amplitude of the sound wave, x is the horizontal distance from the exciter, t is the interval time, A0 is the initial amplitude of the sound wave, α is the attenuation coefficient, ω is the angular frequency of the vibration signal, k is the wave number of the sound wave, which is related to the refractive index of the medium, and φ is the initial phase of the sound wave.

[0086] Since x is the horizontal distance between the signal receiving point and the exciter, the vibration signals at the same depth are selected for cross-correlation processing; k is related to the refractive index of the medium. When the type of medium changes, the waveform of the entire vibration signal will change, which can be used to judge whether the medium is uniform during signal propagation.

[0087] In step (5), the cross-correlation processing process is as follows:

[0088] Select the signals f(x) and g(x) in the two channels at the same depth for cross-correlation analysis. The cross-correlation function R of the signals f(x) and g(x) is fg Defined as:

[0089]

[0090] Where m is the time slip, and in the present invention, m is the time difference between f(x) and g(x); the above formula means that after the signal g(x) is shifted to the left by m time periods, the sum of the product of the function values ​​of f and g after the shift at any time in the entire time domain is calculated. There is one sum corresponding to one m, so the function represents the distribution of the correlation between f and g after the shift with m. The cross-correlation function R fg Represents the distribution of the correlation between f and g after sliding with m.

[0091] The xcorr function in Matlab is used to process the vibration signals in each channel at the same depth. The process is as follows: the collected vibration signals are loaded into Matlab, and the data of each channel is stored in a matrix, with each row representing the signal of a channel; then the xcorr function is used to calculate the correlation coefficient of the vibration signals between different channels; the xcorr function code is as follows:

[0092] [r, lags] = xcorr(x, y)

[0093] Among them, [r, lags] are the two return values ​​of the xcorr function; r is the correlation coefficient; lags is the time difference between the two signals, x is the vibration signal of any channel, and y is the vibration signal of the remaining channels. The value range of the correlation coefficient r is [0, 1]. The correlation represents the similarity of the waveforms between the two signals. Combined with the propagation law of vibration waves in the medium, it reflects whether the medium between the channels is uniform. The correlation coefficient r tends to 1, indicating that the more uniform the medium between the channels, the more uniform the medium between the channels. Therefore, the change of the correlation coefficient r is used to detect the difference in the medium in the concrete and judge the pouring quality.

[0094] In step (5), at the same depth, the vibration signals between every two channels must undergo a cross-correlation operation. If there are n channels, the number of cross-correlation operations S is

[0095]

[0096] The S processing results are averaged. The averaging process is as follows: by using the moving average method, the window size is selected according to the size of S, starting from the beginning of the data sequence, the average value of all data points in the first window is calculated. Then, the window is slid forward by one data point, and the average value of the data points in the next window is calculated. This process is repeated until the entire data sequence is covered. After averaging, the correlation map of the correlation coefficient at this depth and the length of the underground continuous wall is obtained.

[0097] In step (6), the result analysis unit detects the abnormal protrusion in the signal correlation map after the cross-correlation processing, and then determines the depth interval where the underground continuous wall has an uneven casting area.

[0098] In step (7), the casting quality detection unit sets two thresholds r1 and r2 in the correlation map, and divides the casting quality of the underground continuous wall of each trench section into three levels. The part with a correlation coefficient r higher than r1 has good casting quality, and when it is between r1 and r2, the casting quality is average; when the correlation coefficient is lower than r2, the casting quality is poor, wherein the values ​​of r1 and r2 are determined by the mix ratio and strength of concrete.

[0099] Example

[0100] The DAS-based underground continuous wall casting quality detection device and method of the present invention are used to detect the casting quality of three slot underground continuous walls of a certain project. The length of each slot underground continuous wall is 5m, the wall thickness is 1m, and the wall height is 10m. During the construction process, sodium-based bentonite slurry is used for wall protection, and concrete with a strength grade of C40 is used for underground continuous wall casting.

[0101] The DAS-based underground continuous wall pouring quality detection method of the present invention comprises the following steps:

[0102] (1) Figure 2 As shown, the vibration strain sensing optical cable 1 is 100m long in total. Before the steel cage is lowered, it is fixedly attached to the main reinforcement in the center of the steel cage and arranged symmetrically in a "U" shape. After the steel cage is lowered to the bottom of the groove, the vibrator 2-2 is fixed to the fixing rod 8 drilled in the hole next to the underground continuous wall by a clamp. In this embodiment, a total of 5 vibrators 2-2 are set, the initial vibrator is 1m away from the bottom of the underground continuous wall, and the interval between adjacent vibrators is set to 2m. Figure 1 shown.

[0103] (2) The optical fiber in the vibration strain sensing optical cable 1 is fused with the optical fiber jumper, and the jumper is connected to the distributed acoustic wave sensor demodulator 3. After the connection is completed, the optical path connectivity is checked and the demodulator sampling frequency is set to 1000 Hz, the channel spacing is 5 m, and the number of channels is 600.

[0104] (3) After completing the above operations, concrete pouring is carried out. Every time two meters of concrete is poured, the vibration controller 2-1 controls the exciter 2-2 to generate a pulse vibration with a time interval of 1 minute and a waveform of a sine wave. The number of vibrations is 10. Figure 1 As shown, when pouring reaches 4m, the second vibrator 2-2 is excited.

[0105] (4) The distributed acoustic wave sensor demodulator 3 performs line readings within the time interval when the exciter 2-2 generates vibrations, thereby collecting vibration data.

[0106] (5) The distributed acoustic wave sensor demodulator 3 transmits the collected vibration data to the data processing unit 4, and the data processing unit 4 performs noise reduction processing on the vibration data in each channel. The noise reduction steps are as follows: first, the filter frequency band is set to perform bandpass filtering on the vibration data; second, in order to improve the signal-to-noise ratio of the data, the vibration data collected after multiple vibrations are superimposed to obtain a vibration signal with a clear waveform.

[0107] According to the optical cable laying method, in this example, within the control range of each vibrator 2-2 (underground continuous wall depth 2 to 4 m), at three different depths (underground continuous wall depth 2.5 m is detection depth (1), underground continuous wall depth 3 m is detection depth (2), underground continuous wall depth 3.5 m is detection depth (3), as shown in FIG. Figure 3 ), 5 parallel channels are set respectively, and the channels are numbered from left to right as channel 1, channel 2, channel 3, channel 4 and channel 5. The vibration signals with clear waveforms after noise reduction in each channel at the same depth are cross-correlated to obtain the correlation between the signals. The process is as follows:

[0108] The xcorr function in Matlab is used to process the vibration signals in each channel at the same depth. The process is as follows: the collected vibration signals are loaded into Matlab, and the data of each channel is stored in a matrix, with each row representing the signal of a channel. The xcorr function is then used to calculate the correlation coefficient of the vibration signals between different channels. The xcorr function code is as follows:

[0109] [r, lags] = xcorr(x, y)

[0110] Among them, [r, lags] are the two return values ​​of the xcorr function, r is the correlation coefficient, and its value range is [0, 1]; lags is the time difference between the two signals, x is the vibration signal of any channel, and y is the vibration signal of the remaining channels.

[0111] The correlation coefficient r tends to 1, indicating that the medium between the channels is uniform, which means that the concrete backfill between the two channels is uniform; if the correlation coefficient fluctuates, it means that there are impurities or voids in the concrete between the two channels, and the impurities and voids are determined by the degree of reduction of the correlation coefficient.

[0112] When x is the signal in channel 1, y is the signal in the other four channels; when x is the signal in channel 2, y is the signal in channel 3, channel 4 and channel 5; because when x is the signal in channel 1, it has been cross-correlated with the signal in channel 2, so when x is the signal in channel 2, it is no longer necessary to cross-correlate the signal in channel 1; and so on, a total of 10 cross-correlation operations are performed on the signals in the five channels at the same depth, and the results of these 10 cross-correlation operations are averaged to obtain the following: Figure 3 The total result of the cross-correlation processing is shown. The correlation coefficient in the figure is an indicator of the similarity between the signals. The closer the value of the correlation coefficient is to 1, the more similar the signals are, that is, the waveforms of the two tend to be consistent.

[0113] In this embodiment, the closer the correlation coefficient is to 1, the smaller the waveform change of the signal after the vibration signal passes through the concrete medium, which further indicates that the concrete backfill between the two channels is uniform. If the correlation coefficient fluctuates significantly, it indicates that there are impurities or voids in the concrete between the two channels, and the impurities and voids are judged according to the degree of reduction of the correlation coefficient, because the waveform change of the vibration wave in the concrete containing impurities is smaller than that in the concrete containing voids. Correspondingly, when the correlation coefficient decreases by a small amount, there are impurities in the concrete, and when the correlation coefficient decreases by a large amount, there are voids in the concrete. Therefore, according to this rule, the threshold value of the correlation coefficient is set in the correlation map.

[0114] (6) The result analysis unit 5 detects the abnormal protrusion in the correlation map and determines that there is an uneven casting area at the detection depth (2), and the uneven casting area of ​​the underground continuous wall is between 6 and 9 meters and 12 and 15 meters. There is no uneven casting area at the detection depth (1) and the detection depth (3).

[0115] (7) The result analysis unit 5 transmits the depth range of the uneven casting area of ​​the underground continuous wall to the casting quality detection unit 6. The casting quality detection unit 6 sets the correlation threshold r to 0.8 and 0.5 according to the concrete strength of C40.

[0116] like Figure 3 As shown, when r is higher than 0.8, the casting quality of the underground continuous wall is good; when r is between 0.5 and 0.8, the casting quality of the underground continuous wall is average, and there is soil mixing; when r is lower than 0.5, the casting quality of the concrete is poor, and there are gaps. In this embodiment, the part where r is between 0.5 and 0.8 is between 7 and 8 m, and the part where r is lower than 0.5 is between 12 and 13 m. Therefore, for the underground continuous wall at the detection depth (2), the casting quality at 7 to 8 m is average, and the casting quality at 12 to 13 m is poor. In view of the casting quality problems of the underground continuous wall at the above two locations, grouting reinforcement is adopted for remediation in the subsequent construction process.

[0117] Repeat the above process and conduct quality inspection on the remaining 4 parts of the underground continuous wall.

Claims

1. A detection method for an underground continuous wall casting quality detection device based on DAS, characterized in that: The method is implemented by a DAS-based underground continuous wall pouring quality detection device, which comprises a vibration strain sensing optical cable (1), a vibration unit (2), a distributed acoustic wave sensor demodulator (3), a data processing unit (4), a result analysis unit (5), a pouring quality detection unit (6), a steel cage (7) and a fixing rod (8); The vibration unit (2) comprises a vibration controller (2-1) and a vibration exciter (2-2); The vibration strain sensing optical cable (1) is arranged on the steel cage (7) and connected to a distributed acoustic wave sensing demodulator (3); the distributed acoustic wave sensing demodulator (3) transmits the collected vibration data to a data processing unit (4); The data processing unit (4) performs noise reduction and cross-correlation processing on the vibration data to obtain a correlation spectrum between vibration signals and transmits it to the result analysis unit (5); The method comprises the following steps: Step (1), drilling a hole beside the underground continuous wall to be cast, and setting a fixing rod (8) in the hole; before the steel cage is lowered, the vibration strain sensing optical cable (1) is fixedly arranged on the main reinforcement of the steel cage; after the steel cage is lowered to the bottom of the underground continuous wall, starting from the bottom of the underground continuous wall, a vibration exciter (2-2) is arranged on the fixing rod (8) at a set interval; Step (2), connecting the vibration strain sensing optical cable (1) to the distributed acoustic wave sensing demodulator (3) via an optical fiber jumper; Step (3), starting from the bottom of the underground continuous wall, each time a set depth of concrete is poured, the vibration controller (2-1) controls the vibration exciter (2-2) at the corresponding depth to generate a pulse vibration with a time interval and a waveform of a sine wave; pouring is continued, so that the vibration exciter (2-2) on the fixed rod (8) is excited in sequence from the bottom to the top; Step (4), the distributed acoustic wave sensor demodulator (3) collects vibration data of the vibration strain sensing optical cable (1) within the time interval when the vibration exciter (2-2) generates vibration; Step (5), the distributed acoustic wave sensor demodulator (3) transmits the collected vibration data to the data processing unit (4), and after the data processing unit (4) performs noise reduction processing on the vibration data, multiple parallel channels are set at multiple depths within the control area of ​​an exciter (2-2), and the vibration signals after noise reduction processing in each channel at the same depth are cross-correlated to obtain the correlation between the signals, and the process is as follows: The xcorr function in Matlab is used to process the vibration signals in each channel at the same depth. The process is as follows: the collected vibration signals are loaded into Matlab, and the data of each channel is stored in a matrix, with each row representing the signal of a channel; then the xcorr function is used to calculate the correlation coefficient of the vibration signals between different channels; the xcorr function code is as follows: [r, lags] = xcorr(x, y) Where [r, lags] are the two return values ​​of the xcorr function, r is the correlation coefficient, and its value range is [0, 1]; lags is the time difference between the two signals, x is the vibration signal of any channel, and y is the vibration signal of the remaining channels; The correlation coefficient r tends to 1, indicating that the concrete backfill between the two channels is uniform; if the correlation coefficient fluctuates, it indicates that there are impurities or voids in the concrete between the two channels; Step (6), the result analysis unit (5) detects the abnormal protrusion in the signal correlation spectrum after the cross-correlation processing, and then determines the depth interval where the underground continuous wall has the uneven casting area; Step (7), the result analysis unit (5) transmits the depth interval of the uneven casting area of ​​the underground continuous wall to the casting quality detection unit (6), and the casting quality detection unit (6) sets a correlation threshold according to the mix ratio and strength of the concrete. When the correlation coefficient r is lower than the correlation threshold, the casting of the underground continuous wall is uneven; Wherein, in step (3), starting from the bottom of the underground continuous wall, each time a set depth of concrete is poured, a vibration controller (2-1) controls a vibration exciter (2-2) at a corresponding depth to generate a transverse pulse vibration with a time interval and a waveform of a sine wave; In step (5), at the same depth, the vibration signals between every two channels are subjected to a cross-correlation operation. If there are n channels, the number of cross-correlation operations S is: The S processing results are averaged to obtain a correlation map of the correlation coefficient at the depth and the length of the underground continuous wall.

2. The detection method of the underground continuous wall pouring quality detection device based on DAS according to claim 1 is characterized in that: In step (1), the vibration strain sensing optical cable (1) is arranged in a U-shape on the main reinforcement of the underground continuous wall steel cage in a fixed-point pasting manner.

3. The detection method of the underground continuous wall pouring quality detection device based on DAS according to claim 1 is characterized in that: In step (2), the vibration strain sensing optical cable (1) is connected to the distributed acoustic wave sensor demodulator (3) via an optical fiber jumper, and the sampling frequency, channel spacing and channel number of the demodulator are set.

4. The detection method of the underground continuous wall pouring quality detection device based on DAS according to claim 1 is characterized in that: In step (5), the noise reduction process is as follows: setting the filter frequency band and performing bandpass filtering on the vibration data; then superimposing the vibration data collected after multiple vibrations to obtain a vibration signal with a distinct waveform.

5. The detection method of the underground continuous wall pouring quality detection device based on DAS according to claim 1 is characterized in that: In step (5), if the correlation coefficient fluctuates, it indicates that there are impurities or voids in the concrete between the two channels, and the impurities and voids are determined by the degree of reduction of the correlation coefficient.

Images