WO2020248736A1

WO2020248736A1 - Inverse path difference signal-based lamb wave non-reference imaging method for plate structure

Info

Publication number: WO2020248736A1
Application number: PCT/CN2020/087353
Authority: WO
Inventors: 焦敬品; 李海平; 何存富; 吴斌
Original assignee: 北京工业大学
Priority date: 2019-06-12
Filing date: 2020-04-28
Publication date: 2020-12-17
Also published as: CN110208383A

Abstract

The present invention relates to the field of nondestructive testing, and disclosed thereby is an inverse path difference signal-based Lamb wave non-reference imaging method for a plate structure. By means of a Lamb wave sparse array detection system for a plate structure, detection signals received by each sensor exciting another sensor are collected; the signals are filtered and amplitudes are normalized; direct waves of two signals under an inverse path are aligned and the difference is calculated to obtain a difference signal; an envelope of an original signal is taken, and the difference signal is divided by the envelope of the original signal to obtain an inverse path difference signal of each pair of adjacent sensors; and an inverse path difference signal obtained by preprocessing is used to delay superimposed imaging, so that the spatial distribution of a scattered sound field of the entire plate structure may be obtained so as to achieve defect monitoring and positioning. The described method no longer requires a reference signal of a healthy plate, and eliminates the influence of direct waves and boundary reflection echoes on imaging by means of calculating the difference between received signals under an inverse path.

Description

Lamb wave non-reference imaging method for plate structure based on inversion path difference signal

Technical field

The invention relates to a plate structure Lamb wave imaging method based on a reverse path difference signal. The method is suitable for plate structure defect detection and positioning under the condition of unknown reference signals, and belongs to the field of nondestructive testing.

Background technique

Due to the advantages of long transmission distance and high detection efficiency, Lamb wave technology has been widely used in non-destructive testing and health monitoring of structures such as plates and tubes. The use of sensor arrays sparsely distributed on the plate structure can also realize large-scale imaging of the plate structure. In Lamb wave detection, the detection signal waveform is complex and contains rich information, including direct waves and defect echoes, as well as boundary echoes and other characteristic body echoes. Compared with the direct wave and the boundary echo, the defect echo has a smaller amplitude and is easily submerged by other echoes and noise. The dispersion and multi-modal characteristics of the Lamb wave make the analysis and identification of the detection signal more difficult.

In the detection and monitoring of the board structure, some simple and efficient signal processing methods have been developed for the problem of weak defect echo extraction. The baseline subtraction method is one of the most typical and effective methods of defect information extraction [1]. This method subtracts the detection signal from the baseline (reference) signal obtained in advance in a defect-free state to offset the direct wave and boundary echo, and highlight The purpose of defect echo [2]. In actual use, the environmental conditions for acquiring the detection signal and the baseline signal are generally different. When the environmental conditions change greatly, the baseline subtraction method cannot accurately and effectively remove the direct wave and interface echo in the detection signal. The environmental factors that affect the detection mainly include load and boundary conditions, humidity and humidity, etc. Studies have shown that ambient temperature is one of the external factors that has the greatest impact on structural health monitoring methods [3]. Aiming at the problem that the baseline subtraction method is mainly affected by the environmental temperature, domestic and foreign scholars have carried out a lot of research on its compensation and improvement methods. For example, on the basis of the traditional baseline subtraction method, Paul D [4] proposed an optimal baseline subtraction method, and experimentally studied the influence of temperature change rate on the applicability and robustness of the optimal baseline subtraction method. The results show that, compared with the baseline subtraction method, the best baseline subtraction method is more applicable, and its signal-to-noise ratio is nearly 20dB higher than the conventional baseline method.

Aiming at the problem that the traditional baseline subtraction method needs to obtain the baseline reference signal in the structural health state in advance, scholars at home and abroad have improved the method of baseline signal acquisition. H.W.Park et al. [5] analyzed the damage response under time reversal excitation, and used the reconstructed wave source signal and the initial wave source signal to determine the path of the damage, thereby estimating the damage location. This method no longer needs the baseline signal of structural health, but due to the limitation of the monitoring path, its positioning accuracy is low. Jan Hettler[6] proposed a fatigue crack damage identification method based on instantaneous reference. This method realizes the detection of nonlinear damage sources (such as fatigue cracks) in the structure by proportionally subtracting the nonlinear ultrasonic response of the structure under different excitation amplitudes. Based on the second harmonic nonlinear response generated by layered defects, F.Ciampa [7] proposed a dual coherence coefficient imaging method based on the second phase coupling information. This method does not require a baseline reference signal, but requires a structural arrangement Larger number of sensors. Yun-KyuAn [8] proposed an impedance transfer technology, in which two symmetrical pairs of sensors are arranged on both sides of the board under test, and the received signals from different sides are added and subtracted to eliminate the direct wave and retain the defect back. Wave, realizing damage detection without reference.

The above defect signal extraction methods mostly focus on the reflection information of the defect, and do not consider the scattered signal generated by the ultrasonic wave at the defect. In fact, when the ultrasonic wave propagates to the defect, it will produce scattering phenomenon, and the scattering field contains more abundant defect information. Therefore, if the sensor array is used to receive the scattered field signal, more accurate defect status information, such as the type, size, and shape of the defect, can be extracted from it. For example, Zhang J et al. [9] performed all-focus imaging on the ultrasonic signal received by the linear phased array probe, and used the scattering coefficient matrix extracted from the defect position to well realize the defect type (crack, hole) and direction characterization. Zheng Yang et al. [10] used a ring-distributed electromagnetic acoustic sensor array to carry out a large-scale defect detection study of the plate structure, and realized the characterization of the crack length and angle in the plate structure through the extracted scattering coefficient matrix.

The present invention uses the difference of the ultrasonic scattering field at the defect under the reverse path to make the difference of the received signal under the reverse path, and proposes a Lamb wave sparse array imaging method based on the reverse path difference signal, and realizes the plate structure The reference-free Lamb wave imaging.

Summary of the invention

The purpose of the present invention is to provide a lamb wave imaging method for a plate structure without a reference signal, by which the position of the defect can be determined more accurately. Under the condition that the excitation and receiving performance of each sensor is consistent and the defect is located in the asymmetric position of the sensor pair, this method makes the difference between the two received signals of the sensor under each pair of reversal paths to eliminate the direct waves and boundary reflection echoes that affect defect imaging. Using the obtained defect echo difference signal, the non-reference detection and imaging of the defects in the board are realized.

The present invention proposes a plate structure Lamb wave non-reference imaging method based on inverted path difference signals, and its basic principle is as follows:

In the infinitely large isotropic thin plate, two circular piezoelectric sensors are used for the excitation and reception of Lamb waves, and the center distance is d ₀ . Assuming that the performance of the two piezoelectric sensors is the same, they can excite and receive ultrasonic waves in all directions in space in a circumferential direction. The radius of the circular defect is r and it is located in the far field. If the size effect of the defect is considered, the multiple acoustic waves emitted by the excitation sensor will propagate to the defect and produce reflection and scattering at the defect. These reflected waves and a part of the scattered wave will be received by the receiving sensor, as shown in Figure 1. Show. Considering that the defect echoes received by the two sensors are mainly scattered waves in the direction of the sensor’s main sound beam (when the No. 1 piezoelectric sensor is excited and the No. 2 piezoelectric sensor is received, the propagation paths are d ₁ and d ₂ ; When the piezoelectric sensor is excited and the No. 3 piezoelectric sensor receives, the propagation paths are d ₃ and d ₄ ), then one of the piezoelectric sensors is used as the excitation, and the frequency domain expression of the signal received by the other piezoelectric sensor is simplified as:

X ₁₂ (ω) and X ₂₁ (ω) are the frequency spectrums of the two piezoelectric sensors as excitation sensors, and the signal received by the other sensor;

Is the displacement response amplitude; T(ω), R(ω) are the excitation and reception transfer functions of the piezoelectric sensor; k is the wave number of the Lamb wave; D(α,ω) is the time after the Lamb wave interacts with the defect The scattering coefficient in the α direction. In the formula, the first term is the defect scattered echo, and the second term is the direct wave. If

The formula (1) is expressed as:

Make the difference between the two expressions in equation (2) and call it the reverse path difference signal:

It can be seen from equation (3) that the direct wave in the reverse path difference signal is eliminated, and only the difference of the defect scattered echo that can reflect the defect status information is retained. At the same time, it is observed that the difference signal of the defect scattered echo under the reverse excitation is related to the scattering coefficient D and the propagation distance. It is precisely because of these two factors that the amplitude and phase of the defect scattered echo under inversion excitation are different.

Based on the above analysis of the inversion path difference signal, the inversion path difference signal of the piezoelectric array sparsely distributed in the plate structure can be further used to perform Lamb wave imaging on the plate structure. If the number of elements contained in the sparse array is N, N(N-1)/2 piezoelectric sensor pairs are formed, and N(N-1) groups of detection signals s _{ij are obtained} . Among them, the subscripts i and j respectively represent the serial numbers of the excitation and receiving piezoelectric sensors, and i≠j. Correspondingly, N(N-1) groups of reverse path difference signals Δs _ij =s _ij -s _ji can also be obtained. Use these inverted path difference signals Δs _{ij to} obtain the intensity of the scattered sound field at any point (x, y) in the plate:

Among them, t _ij (x, y) is that the Lamb wave propagates from the excited piezoelectric sensor (coordinates (x _i , y _i )) to this point (coordinates (x, y)), and then propagates to the receiving piezoelectric sensor (The coordinates are (x _j , y _j )) used time, the expression is:

Among them, c _g is the group velocity of Lamb wave propagation at the detection frequency.

Obviously, the spatial distribution of the scattered field shown in formula (4) can realize damage detection and imaging of the plate structure. It is worth noting that in the above imaging, the difference between the excitation and reception signals of the inverted piezoelectric sensor is achieved to eliminate the influence of direct waves and highlight the effect of defect scattered waves. At the same time, this method does not require a reference signal in the absence of defects, and can well avoid the influence of factors such as ambient temperature caused by this.

The technical scheme of the present invention is as follows:

The device used in the present invention is shown in Fig. 2, and includes an arbitrary function generator 1, a voltage amplifier 2, an oscilloscope 3, a piezoelectric sensor 4 and an aluminum plate test piece 5. The output of arbitrary function generator 1 is connected to the input port of voltage amplifier 2, the output of voltage amplifier 2 is connected to PZT5H piezoelectric ceramic sensor 4, the output of arbitrary function generator 1 is connected to channel 1 of oscilloscope 3, and the output of oscilloscope 3 is connected. , 3 and 4 channels are connected to other piezoelectric sensors 4 respectively. Through the arbitrary function generator 1, each piezoelectric sensor 4 is controlled in turn for excitation.

Channels

2, 3, and 4 of the oscilloscope 3 receive signals from other piezoelectric ceramic sensors 4, and the signals are received and stored repeatedly until the entire pressure is excited. The electrical sensor 4 and other piezoelectric sensors 4 are all received.

The Lamb wave reference-free imaging method of plate structure based on the inverted path difference signal proposed by the present invention is realized through the following steps:

Step 1: Build a plate structure Lamb wave experimental system according to the system diagram of the detection device shown in Figure 2. The system includes an arbitrary function generator 1, a voltage amplifier 2, an oscilloscope 3, a piezoelectric sensor 4 and an aluminum plate test piece 5. Connect the output of arbitrary function generator 1 to the input port of voltage amplifier 2, the output of voltage amplifier 2 to piezoelectric sensor 4, the output of arbitrary function generator 1 to channel 1 of oscilloscope 3, the 2, 3, The 4 channels are connected to other piezoelectric sensors 4 respectively. The test piece selects a thin aluminum plate containing defects, that is, the aluminum plate test piece 5, and N sensors are respectively arranged at any position in the aluminum plate test piece 5, and the sensor's inner surrounding area is the monitoring area.

Step 2: Set the frequency, period and voltage of the excitation signal through the function generator 1, and use the voltage amplifier 2 to amplify the excitation signal. Each piezoelectric sensor 4 is controlled in turn for excitation, and

channels

2, 3, and 4 of the oscilloscope 3 receive signals from other (N-1) piezoelectric ceramic sensors 4 respectively. The signal is received and stored repeatedly until all the piezoelectric sensors 4 are excited and all other piezoelectric sensors 4 are received. N sensors can form N(N-1)/2 pairs of sensors, and a total of N(N-1) groups of received signals s _{ij are obtained} , 1<i<N, 1<j<N, i≠j.

Step 3: Use a band-pass filter to filter the signal; after filtering, normalize the signal based on the maximum amplitude of the direct wave, and reset the direct wave phase difference of the received signal under the inversion path to zero, making the direct wave The waves are aligned. The difference between the received signals of each pair of sensors under the reversal path is obtained to obtain N(N-1) groups of reversal path difference signals Δs _ij =s _ij -s _ji . Divide the differenced signal Δs _ij by the envelope s _ij of the original signal to amplify the defect echo difference signal:

Among them, a=0.1. The setting of a is to prevent the original signal envelope

The existence of certain minima makes

Some maxima appear.

Step 4: In the imaging area, reversal path difference signals of adjacent sensor pairs after preprocessing

Substituting formula (4), the time-lapse superimposed imaging result is obtained, and defect detection and positioning are realized.

Compared with the existing detection method, the present invention has the following advantages: (1) The present invention eliminates the direct wave by making the difference between the two received signals under the inverted path in the sparse array of the plate structure, while retaining the echo difference reflecting the defect position. , The reference signal when the structure is healthy is no longer needed, and it is not affected by the temperature and other factors in the environment; (2) For each detection, only N piezoelectric sensors need to be excited, and the other (N-1) piezoelectric sensors are simultaneously Receiving, there is no need to obtain baseline signals through experiments, and the detection process is simple and easy to operate; (3) This method can not only eliminate direct waves, but also eliminate the influence of boundary reflection echoes, and has less requirements on the placement of sensors.

Description of the drawings

Figure 1 Schematic diagram of the Lamb wave propagation model in an infinite thin plate.

Figure 2 System diagram of detection device.

In the figure: 1. Arbitrary function generator, 2. Voltage amplifier, 3. Oscilloscope, 4. Piezoelectric sensor, 5. Aluminum plate test piece.

Figure 3 Typical received signal and partial enlarged view.

Figure 4 Signal and partial enlarged view after preprocessing.

Figure 5 Imaging results.

Figure 6 is a flow chart of the implementation of the method.

Detailed ways

The present invention will be further explained below in conjunction with specific experiments:

The implementation process of this experiment includes the following steps:

1. Experimental system: Build an experimental system according to the system diagram of the detection device shown in Figure 2. The system includes an arbitrary function generator 1, a voltage amplifier 2, an oscilloscope 3, a piezoelectric sensor 4, and a 1mm thick aluminum plate test piece 5. The output of arbitrary function generator 1 is connected to the input port of voltage amplifier 2, the output of voltage amplifier 2 is connected to piezoelectric sensor 4, the output of arbitrary function generator 1 is connected to channel 1 of oscilloscope 3, and the 2, 3, The 4 channels are connected to other piezoelectric sensors 4 respectively. The test piece is an aluminum plate with a size of 800mm*800mm and a thickness of 1mm. There is a transparent hole defect at the position of the aluminum plate (350, 370). The four sensors are arranged at any position in the plate, and the coordinates are the No. 1 sensor (300, 500). , 2nd sensor (500,480), 3rd sensor (250,300), 4th sensor (500,300). The size of the piezoelectric sensor 4 is 8 mm in diameter and 1 mm in thickness.

2. Plate structure Lamb wave detection experiment: use arbitrary function generator 1 to generate a 270kHz Hanning window modulated 5-period single audio excitation signal, amplify the signal voltage amplitude to 100Vpp by voltage amplifier 2, and apply the amplified voltage to Piezoelectric sensor electrodes. Each piezoelectric sensor 4 is controlled in turn for excitation, and

channels

2, 3, and 4 of the oscilloscope 3 receive and save the received signals of the other three piezoelectric ceramic sensors 4, and the sampling frequency is 50 MHz. The signal is received and stored repeatedly until all the piezoelectric sensors 4 are excited and the other three piezoelectric sensors 4 are all received. Four sensors can form N(N-1)/2=6 pairs of sensors, a total of N(N-1)=12 sets of received signals s _ij , where i represents the excitation sensor (1<i<4,1<j< 4, i≠j), Figure 3 shows a group of sensor pairs of typical received signals and their reverse path difference signals.

3. Signal preprocessing: use a bandpass filter with a passband width of 200kHz and a passband attenuation and stopband attenuation of 1dB and 6dB respectively to filter the signal; after filtering, the signal is returned to the standard with the maximum amplitude of the direct wave Unified processing; reset the direct wave phase difference of the received signal under the inversion path to zero, so that the direct wave is aligned. The signal difference of each pair of sensors under the reversal path is made to obtain N(N-1) groups of reversal path difference signals Δs _ij =s _ij -s _ji . Since the amplitude of the defect reflection echo is small, in order to maximize the defect reflection echo difference, divide the difference signal Δs _ij by the original signal envelope

To amplify the defect echo difference signal:

Among them, a=0.1. The setting of a is to prevent the original signal envelope

The existence of certain minima makes

Some maxima appear. The typical signal after preprocessing and the partial enlarged picture are shown in Fig. 4.

4. Lamb wave imaging: For each pair of excitation receiving sensors, other sensors arranged on the board can also be regarded as defects. The inverted path difference signal of each sensor also contains the scattered waves of ultrasonic waves on these sensors, which will affect the reflection. The analysis of defect scattered waves in the path difference signal brings adverse effects. In order to minimize the influence of scattered waves generated by other sensors on defect waves, only the difference signals between adjacent sensor pairs are used for time-delay superimposition imaging during imaging. Use the difference signal of the inverted path of adjacent sensor pairs

The obtained imaging result is shown in Figure 5. The small circle in the figure is the sensor position, the center of the large circle is the actual position of the defect, and the point with the largest amplitude in the figure is the imaging positioning result.

references

[1]Salmanpour M S, Sharif K Z, MhfA. Impact Damage Localisation with Piezoelectric Sensors under Operational and Environmental Conditions[J].Sensors,2017,17(5):1178.

[2]Muller A, Robertson-Welsh B, Gaydecki P, et al. Structural Health Monitoring Using Lamb Wave Reflections and Total Focusing Method for Image Reconstruction [J]. Applied Composite Materials, 2017, 24(2): 553-573.

[3]Sohn,H.Effects of environmental and operational variability on structural health monitoring[J].Philosophical Transactions of the Royal Society A:Mathematical,Physical and Engineering Science,2007,365(1851):539-560.

[4]Konstantinidis G, Wilcox P D, Drinkwater B W. An Investigation Into the Temperature Stability ofa Guided Wave Structural Health Monitoring System Using Permanently Attached Sensors[J].IEEE-Sensors5: 905, 2007

[5]Park H W, Sohn H, Law K H, et al. Time reversal active sensing for health monitoring of a composite plate[J]. Journal of Sound&Vibration,2007,302(1–2):50-66.

[6]Hettler J, Tabatabaeipour M, Delrue S, et al. Linear and Nonlinear Guided Wave Imaging of Impact Damage in CFRP Using a Probabilistic Approach. [J].Materials,2016,9(11):901.

[7]Ciampa F, Pickering S G, Scarselli G, et al. Nonlinear imaging of damage in composite structures using sparse ultrasonic sensor arrays[J]. Structural Control and Health Monitoring, 2017, 24(5): e1911.

[8]An Y K, Lim H J, Kim M K, et al. Application of local reference-free damage detection techniques to in situ bridges[J]. Journal of Structural Engineering,2013,140(3):04013069.

[9]Zhang J, Drinkwater B W, Wilcox P D. Defect characterization using an ultrasonic array to measure the scattering coefficient matrix[J]. IEEE Transactions on Ultrasonics Ferroelectrics&Frequency Control, 2008, 55(10): 2254-65.

[10]Zheng Yang, He Cunfu, Zhou Jinjie, et al. Study on the two-dimensional scattering characteristics of ultrasonic Lamb waves at defects[J]. Engineering Mechanics, 2013, 30(8): 236-243.

Claims

A plate structure Lamb wave imaging method based on the inverted path difference signal. In an infinitely large isotropic thin plate, two circular piezoelectric sensors are used to excite and receive the Lamb wave, and the center distance is d0; Assuming that the performance of the two piezoelectric sensors is the same, they can both excite and receive ultrasonic waves in all directions in space in a circumferential direction; the radius of the circular defect is r, located in the far field; if the size effect of the defect is considered, the multiple beams emitted by the sensor are excited The sound wave will propagate to the defect and produce reflection and scattering at the defect. Part of these reflected waves and scattered waves will be received by the receiving sensor. Considering that the defect echoes received by the two sensors are in the direction of the sensor's main sound beam The scattered waves are mainly scattered waves. When the piezoelectric sensor No. 1 is excited and the piezoelectric sensor No. 2 receives, the propagation paths are d 1 and d 2 ; when the piezoelectric sensor No. 2 is excited and the piezoelectric sensor No. 3 receives, the propagation paths are d 3 and d 4 , one of the piezoelectric sensors is used as excitation, and the frequency domain expression of the signal received by the other piezoelectric sensor is simplified as:

X 12 (ω) and X 21 (ω) are the frequency spectrums of the two piezoelectric sensors as excitation sensors, and the signal received by the other sensor;
Is the displacement response amplitude; T(ω), R(ω) are the excitation and reception transfer functions of the piezoelectric sensor; k is the wave number of the Lamb wave; D(α,ω) is the time after the Lamb wave interacts with the defect The scattering coefficient in the α direction; the first term in the formula is the defect scattered echo, and the second term is the direct wave.
The formula (1) is expressed as:

Make the difference between the two expressions in equation (2) and call it the reverse path difference signal:

It can be seen from equation (3) that the direct wave in the reverse path difference signal is eliminated, and only the difference of the defect scattered echo that can reflect the defect status information is retained; at the same time, it is observed that the defect scattered echo under the reverse excitation The difference signal of is related to the scattering coefficient D and the propagation distance; it is precisely because of these two factors that the amplitude and phase of the defect scattered echo under inversion excitation are different;

Based on the analysis of the inversion path difference signal, the inversion path difference signal of the piezoelectric array sparsely distributed in the plate structure can be further used to perform Lamb wave imaging of the plate structure; if the number of elements contained in the sparse array is N, Then N(N-1)/2 piezoelectric sensor pairs are formed, and N(N-1) groups of detection signals s ij are obtained ; among them, the subscripts i and j respectively represent the serial numbers of the piezoelectric sensor for excitation and reception, and i≠j ; Correspondingly, N(N-1) groups of inverted path difference signals Δs ij =s ij -s ji can also be obtained; these inverted path difference signals Δs ij are used to obtain the scattered sound field at any point (x, y) in the plate strength:

Among them, t ij (x, y) is that the Lamb wave propagates from the excited piezoelectric sensor (coordinates (x i , y i )) to this point (coordinates (x, y)), and then propagates to the receiving piezoelectric sensor (The coordinates are (x j , y j )) used time, the expression is:

Among them, c g is the group velocity of Lamb wave propagation at the detection frequency;

Obviously, the spatial distribution of the scattered field shown in formula (4) realizes the detection and imaging of plate structure damage; in the above imaging, the influence of the direct wave is eliminated by making the difference between the excitation and reception signals of the inverted piezoelectric sensor pair, and the defect scattered wave is highlighted Effect;

It is characterized in that the method is realized through the following steps:

Step 1: Set up a plate structure Lamb wave experiment system, the system includes arbitrary function generator (1), voltage amplifier (2), oscilloscope (3), piezoelectric sensor (4) and aluminum plate test piece (5); The output of the generator (1) is connected to the input port of the voltage amplifier (2), the output of the voltage amplifier (2) is connected to the piezoelectric sensor (4), and the output of the arbitrary function generator (1) is connected to 1 channel of the oscilloscope (3) Connect, the 2, 3, 4 channels of the oscilloscope (3) are connected to other piezoelectric sensors (4); the tested piece is selected from the thin aluminum plate containing defects, namely the aluminum plate test piece (5), and the N sensors are respectively arranged on the aluminum plate to test At any position in the part (5), the internal enclosed area of the sensor is the monitoring area;

Step 2: Set the frequency, period and voltage of the excitation signal through the function generator (1), amplify the excitation signal with the voltage amplifier (2); control each piezoelectric sensor (4) in turn to excite, and the oscilloscope (3) Channels 2, 3 and 4 respectively receive the receiving signals of other (N-1) piezoelectric sensors (4); repeat receiving and saving the signals until all piezoelectric sensors (4) are excited, and other piezoelectric sensors (4) ) All received; N sensors can form N(N-1)/2 pairs of sensors, and a total of N(N-1) groups of received signals s ij are obtained , 1<i<N,1<j<N,i≠ j;

Step 3: Use a band-pass filter to filter the signal; after filtering, normalize the signal based on the maximum amplitude of the direct wave, and reset the direct wave phase difference of the received signal under the inversion path to zero, making the direct wave Wave alignment; difference the received signals of each pair of sensors under the inversion path to obtain N(N-1) groups of inversion path difference signals Δs ij = s ij -s ji ; divide the differenced signal Δs ij by the original Signal envelope
To amplify the defect echo difference signal:

Among them, a=0.1; the setting of a is to prevent the original signal envelope
The existence of certain minima makes
Certain maxima appear;

Step 4: In the imaging area, reversal path difference signals of adjacent sensor pairs after preprocessing
Substituting equation (4), the time-lapse superimposed imaging result is obtained, and defect detection and positioning are realized.