An efficient phase-shift imaging method for multi-layer structure hole defects using ultrasonic phased array

Abstract

The present invention provides an ultrasonic phased array high-efficiency phase shift imaging method for hole defects in a multilayer structure, comprising the following steps: (1) converting the full matrix data of the collected multilayer structure into the frequency wave number domain through a three-dimensional fast Fourier transform; (2) for any untraversed layer in the multilayer structure, extrapolating the surface wave field of the multilayer structure to the upper interface of the untraversed layer to obtain the wave field information of the untraversed layer; (3) focusing imaging the untraversed layer in the frequency wave number domain based on the obtained wave field information; (4) repeating steps (2) and (3) until all layers are traversed, obtaining the imaging result of the multilayer structure. The imaging method of the present invention only uses two operators, superposition and two-dimensional Fourier transform, and does not require the cross-correlation operator in the frequency domain phase shift, which greatly reduces the memory requirement and calculation complexity, shortens the calculation time, and improves the imaging efficiency; and the imaging resolution is high.

Description

Technical Field

[0001] The present invention belongs to the technical field of detection methods, and in particular relates to an ultrasonic phased array high-efficiency phase shift imaging method for hole defects in a multi-layer structure. Background Art

[0002] In order to combine the advantages of different materials, such as high strength, light weight, corrosion resistance, durability and flexibility, multilayer structures are used in the fields of construction, electronics and aerospace. However, due to the complex manufacturing process and severe service process, internal void defects are very common in these products, which can greatly damage the mechanical properties of the products and shorten the service life. Common non-destructive detection methods for internal void defects in composite products are infrared (IR) thermal imaging and X-ray tomography. Infrared thermal imaging captures the local high temperature at the void defect to determine the defect location, but cannot determine the defect depth and has limited resolution; X-ray tomography is often used to reconstruct void defects in products with high precision, but complex calculations limit its application in engineering practice. In addition, some material processing technologies will also form multilayer structures during the processing process, among which injection molding is the most typical, and the melt in the mold shell and cavity forms a natural two-layer structure.

[0003] Impurities and bubbles in polymer melts are common defects, usually caused by insufficient process parameters and material contamination, but there is still a lack of effective online detection methods. Therefore, efficient imaging methods for internal defects in multilayer structures have received widespread attention. Ultrasonic phased array defect imaging technology has been well developed in the field of non-destructive testing, with high detection efficiency and accuracy, and has been widely used in the inspection of concrete or metal structures and products.

[0004] At present, the phased array signal processing method has been greatly developed. In order to maximize the flexibility of phased array signal processing, as much information as possible should be extracted from the phased array sensor. With the advancement of electronic technology, full matrix acquisition (FMC) technology has attracted widespread interest in ultrasound imaging, which acquires the complete time domain signal of each pair of transmit-receive pairs. Compared with other data acquisition methods, FMC data contains more information and greatly improves the signal-to-noise ratio of ultrasound imaging.

[0005] Based on the time domain (time-space domain) and frequency domain (frequency-wavenumber domain) propagation models of sound waves in homogeneous media, researchers have proposed various FMC data post-processing methods. The total focusing method (TFM) is a typical time-domain ultrasonic imaging method that focuses all sound beams on the corresponding pixels in the target area. The detection sensitivity and image resolution of the TFM method for FMC data have been effectively improved, and it has become the standard for non-destructive testing. In contrast, research on frequency domain ultrasonic imaging methods for FMC data is relatively limited. Hunter et al. proposed a Stolt mapping wavenumber algorithm for ultrasonic FMC data, which showed high imaging efficiency; Fan used a frequency domain model for time reversal migration. However, these methods are only applicable to single sound velocity media.

[0006] Researchers have proposed several ultrasonic imaging methods for multilayer structures. The time-domain ultrasonic imaging method for multilayer structures is mainly based on the TFM method. The ray-based TFM algorithm uses ray tracing to calculate the propagation trajectory of the sound beam in the multilayer structure to obtain the time delay of the transmission-reception combination at each pixel point, and then produces the result of the same imaging conditions as TFM. However, the complexity of the iterative calculation of the interface refraction point reduces its imaging efficiency, so the root mean square (RMS) velocity is introduced to avoid the iterative calculation of the trajectory, which is called RMS-based TFM. However, this method is only applicable to cases with small velocity changes, otherwise the imaging quality will be seriously deteriorated. For the frequency-domain ultrasonic imaging method, the phase shift (PSM) of B-scan data was first extended to the multilayer case. It reconstructs the ultrasonic image by extrapolating the surface acoustic field to the inside. Skjelvareid et al. introduced Stolt interpolation to improve the computational efficiency of the PSM algorithm for multilayer media. In addition, Wu et al. proposed the extended phase shift migration (EPSM) to extend the multilayer imaging algorithm to FMC data. Compared with RMS-based TFM, the accuracy and resolution of EPSM imaging results are improved, but the efficiency is lower. Therefore, an efficient and robust FMC algorithm for ultrasonic imaging of multi-layer structures is urgently needed. Summary of the invention

[0007] In order to solve the problems existing in the prior art, the present invention provides a method for efficiently imaging hole defects inside a multi-layer structure using full matrix data of an ultrasonic phased array. The method greatly reduces the computational complexity of ultrasonic full matrix imaging and can perform fast focused imaging of hole defects in a multi-layer structure.

[0008] An ultrasonic phased array high-efficiency phase shift imaging method for multilayer structure hole defects comprises the following steps:

[0009] (1) The full matrix data of the multilayer structure with hole defects is converted into the frequency wavenumber domain by three-dimensional fast Fourier transform;

[0010] (2) for any untraversed layer in the multi-layer structure, the surface wave field of the multi-layer structure is extrapolated to the upper interface of the untraversed layer to obtain the wave field information of the untraversed layer;

[0011] (3) focusing imaging the untraversed layer in the frequency wavenumber domain based on the obtained wave field information;

[0012] (4) Repeat steps (2) and (3) until all layers are traversed to obtain the imaging result (defect image) of the multilayer structure.

[0013] In the above imaging method, full matrix data (FMC data) of a multilayer structure with hole defects and initialization parameters are first input, and then the full matrix data is converted by three-dimensional fast Fourier transform (3D-FFT). The full matrix data of the multilayer structure with hole defects is measured by a phased array sensor.

[0014] The initialization parameters include the number of layers N of the multilayer structure, the speed of sound c of each layer n , the thickness of each layer d n , imaging resolution of each layer Δz n And select the frequency band range, where n∈[1,N]. The full matrix data is used as three-dimensional data of the receiving unit position, the transmitting unit position and time.

[0015] Preferably, in step (2), the multi-layer structure is traversed from top to bottom.

[0016] Preferably, in step (2), the double square root vertical wave number is used to extrapolate (phase shift) the surface wave field of the multilayer structure.

[0017] The double square root vertical wave number of any layer (set as m layers) in the multilayer structure is calculated by the following formula:

[0018]

[0019] Where ω represents the frequency; c m represents the speed of sound in the mth layer; k rx is the horizontal wave number of the receiving oscillator; k sx is the horizontal wave number of the transmitting oscillator.

[0020] The wave field calculation formula at depth z in a multi-layer medium (multi-layer structure) is:

[0021]

[0022] Among them, Φ(k rx ,k sx ,z,ω) is the corrected phase offset; D(k rx ,ksx ,ω) represents the wave field at the top of the nth layer; k z,n represents the offset wave number of the nth layer in the z direction; z n-1 Indicates the lower interface depth of the n-1th layer.

[0023] Preferably, in step (3), the untraversed layer is focused and imaged in the frequency wavenumber domain with time t=0 as the imaging condition.

[0024] Preferably, the imaging result I(x,z) with time t=0 as the imaging condition is expressed as follows:

[0025]

[0026] Among them, k rx is the horizontal wave number of the receiving oscillator; i is the imaginary unit; k sx is the horizontal wave number of the transmitting oscillator; x represents the arrangement direction of the phased array sensor; ω represents the frequency; k z,n represents the offset wave number of the nth layer in the z direction; z represents the extrapolated depth; z n-1 represents the depth of the lower interface of the n-1th layer; n∈[1,N], N represents the number of layers in the multilayer structure; k z,m represents the offset wave number of the mth layer in the z direction; d m Represents the thickness of the mth layer.

[0027] Preferably, in step (3), the specific operation of performing focused imaging on the untraversed layer in the frequency wavenumber domain according to the obtained wave field information comprises the following steps:

[0028] 1) In the imaging area, the wave field is recursively deduced layer by layer according to the resolution in the depth direction to obtain a discrete wave field in the frequency wavenumber domain;

[0029] 2) Superimpose the obtained discrete wave fields in the frequency dimension to obtain the wave field at time 0;

[0030] 3) The wave field at time 0 is transformed from the frequency wavenumber domain back to the spatial domain through two-dimensional fast Fourier transform;

[0031] 4) Extract x in the spatial domain r =x s = The wave field corresponding to x is taken as the imaging result of the untraversed layer; wherein x represents the arrangement direction of the phased array sensor array; x s Indicates the position of the transmitting oscillator; x r Indicates the receiving oscillator position.

[0032] The imaging method of the present invention introduces double square root vertical wavenumbers to perform phase shift (wave field extrapolation), reconstructs the acoustic field of the measurement area (multi-layer structure) in the frequency wavenumber (fk) domain, and directly applies the explosion reflection condition in the frequency wavenumber domain to perform fast high-quality focused imaging of the defect to obtain a defect image of the multi-layer structure. The explosion reflection model assumes that the received reflection signal is regarded as actively emitted at the reflection position in the measurement area at t=0.

[0033] The present invention proposes an ultrasonic phased array efficient phase shift imaging method for hole defects in a multilayer structure. The FMC data is regarded as three-dimensional data of the receiving unit position, the transmitting unit position and the time, and is converted to the frequency wave number (fk) domain through a three-dimensional fast Fourier transform (3D-FFT). At the same time, the double square root vertical wave number is introduced to extrapolate the wave field from the surface to an arbitrary depth z, and the imaging condition t=0 is directly applied in the fk domain to generate a defect image. The imaging method only uses two operators, superposition and two-dimensional Fourier transform, and does not require the cross-correlation operator in the frequency domain (frequency wave number domain) phase shift, which greatly reduces the memory requirement and computational complexity. The imaging resolution of this method is the same as that of EPSM, but the calculation time is only 1 / 6 of the latter.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] The ultrasonic phased array high-efficiency phase shift offset imaging method of multi-layer structure hole defects of the present invention only uses two operators, superposition and two-dimensional Fourier transform, and does not require the cross-correlation operator in the frequency wavenumber domain phase shift, which greatly reduces the memory requirement and calculation complexity, shortens the calculation time, and improves the imaging efficiency; and the imaging resolution is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of a method for efficient phase-shift imaging of multi-layer structure hole defects using an ultrasonic phased array according to an embodiment of the present invention;

[0037] Where: D represents the full matrix data; P represents the wave field sound pressure; i represents the imaginary unit; n and m represent a layer of the multi-layer structure; N represents the number of layers of the medium (multi-layer structure); c m represents the sound speed of the mth layer; k z,m represents the offset wave number of the mth layer in the z direction; d m represents the thickness of the mth layer; k sx represents the horizontal wave number of the transmitting oscillator; k rx Indicates the horizontal wave number of the receiving oscillator; I n It represents the ultrasonic imaging result corresponding to the nth layer of medium, and the corresponding pixel point is M n ×N x ; I(x,z) represents the final imaging result, and the corresponding pixel is Mz ×N x ,in, Δz n represents the resolution in the z direction; t represents the sampling time; ω represents the frequency; M z Indicates the number of pixels of the imaging result in the z direction; M n represents the number of discretized layers of the nth layer of medium in the z direction; Indicates the number of transmitting elements; Indicates the number of received oscillators; for FMC data, N t Indicates the number of time domain sampling points; N ω Indicates the frequency points of the selected frequency band;

[0038] Figure 2 In the figure: (a) is a schematic diagram of full matrix acquisition of single-layer ultrasonic phased array; (b) is a schematic diagram of full matrix acquisition of multi-layer ultrasonic phased array; z represents the depth direction; x represents the arrangement direction of the phased array sensor array; d N Indicates the material thickness of the Nth layer; z N Indicates the position of the lower interface of the Nth layer; x s Indicates the position of the transmitting oscillator; x r Indicates the receiving oscillator position;

[0039] Figure 3 Figure 1 is a simulation experiment diagram; where: (a) is a simulation model; (b) is a schematic diagram of the FMC data interface; (c) is the imaging result of the Raybased-TFM algorithm; (d) is the imaging result of the RMS-TFM algorithm; (e) is the imaging result of the EPSM algorithm; (f) is the imaging result of the method of this embodiment;

[0040] Figure 4 These are experimental pictures of three-layer composite structures; among them: (a) three-layer composite structure test block; (b) schematic diagram of FMC data interface; (c) imaging result of Raybased-TFM algorithm; (d) imaging result of RMS-TFM algorithm; (e) imaging result of EPSM algorithm; (f) imaging result of the method of this embodiment. DETAILED DESCRIPTION

[0041] An ultrasonic phased array high-efficiency phase shift imaging method for multilayer structure hole defects comprises the following steps:

[0042] (1) The full matrix data of the multilayer structure with hole defects is converted into the frequency wavenumber domain by three-dimensional fast Fourier transform;

[0043] (2) for any untraversed layer in the multi-layer structure, the surface wave field of the multi-layer structure is extrapolated downward to the upper interface of the untraversed layer using the double square root vertical wave number to obtain the wave field information of the untraversed layer;

[0044] (3) focusing imaging of the untraversed layer is performed in the frequency wavenumber domain with t=0 as the imaging condition according to the obtained wave field information;

[0045] The expression with time t = 0 as the imaging condition is as follows:

[0046]

[0047] Among them, k rx is the horizontal wave number of the receiving oscillator; i is the imaginary unit; k sx is the horizontal wave number of the transmitting oscillator; x represents the arrangement direction of the phased array sensor; ω represents the frequency; k z,n represents the offset wave number of the nth layer in the z direction; z represents the depth; z n-1 represents the depth of the lower interface of the n-1th layer; n∈[1,N], N represents the number of layers in the multilayer structure; k z,m represents the offset wave number of the mth layer in the z direction; d m Represents the thickness of the mth layer.

[0048] (4) Repeat steps (2) and (3) until all layers are traversed and a defect image of the multi-layer structure is obtained.

[0049] The detailed flow chart of the ultrasonic phased array high-efficiency phase shift imaging method for the above-mentioned multi-layer structure hole defects is as follows: Figure 1 The initialization input includes the collected FMC data of the multi-layer structure, the number of layers N of the multi-layer structure, and the sound velocity c of each layer. n , thickness of each layer d n , imaging resolution of each layer Δz n And select the frequency band range. First, the FMC data is converted to the frequency wave number (fk) domain by three-dimensional fast Fourier transform. Then, the final image I (defect image) of the multilayer structure is obtained by cyclic imaging of each layer of the medium. The wave field at the top of the nth layer is D(k rx ,k sx ,ω) by phase shift Extrapolate, then image the internal defects of the layer, and output the imaging result of the layer I n .

[0050] Efficient imaging workflow in a single layer Figure 1 As shown on the right. In the imaging area, the wave field is recursively deduced layer by layer according to the resolution in the z direction, and the discrete wave field P(k) in the frequency wave number (fk) domain of the calculation area is obtained. rx ,z n +jΔz n ,k sx ,z n +jΔz n,ω); after superimposing the obtained discrete wave fields in the frequency dimension, the wave field at t = 0 is obtained. The wave field at t = 0 is transformed from the fk domain back to the spatial domain through a two-dimensional fast Fourier transform; then the spatial domain extracts x r =x s = The wave field corresponding to x is taken as the imaging result of the layer, where x represents the arrangement direction of the phased array sensor array; x s Indicates the position of the transmitting oscillator; x r Indicates the receiving oscillator position. Figure 1 The right side of each step in lists the computational complexity of that step.

[0051] Theoretical formula

[0052] 1. Full matrix data

[0053] Full Matrix Capture (FMC) collects the time domain signals between all transmit-receive pairs in the phased array probe. For an n-element phased array probe, the FMC data contains n 2 A scan signal. D(x r ,x s ,t) represents the signal transmitted from the sth oscillator and received by the rth oscillator at time t, such as Figure 2 As shown, the ultrasonic wave moves from the oscillator position (x s ,0), and is reflected by the scattering point at (x,z) and then by the r ,0) is received by the sensor. The wave field in the time and space domain is represented by P(x r ,z r ,x s ,z s ,t), then the wave field at the measurement surface is:

[0054] P(x r ,0,x s ,0,t)=D(x r ,x s ,t) (1)

[0055] After three-dimensional fast Fourier transform, it is converted to the frequency wavenumber domain and expressed as:

[0056]

[0057] Among them, k sx is the horizontal wave number of the transmitting oscillator; k rx is the horizontal wave number of the receiving oscillator; i represents the imaginary unit; ω represents the frequency; t represents the time; x s Indicates the position of the transmitting oscillator; x r The Fourier transform separates the waves from the transmitting element to the scattering point and from the scattering point to the receiving element into a superposition of individual plane wave components.

[0058] 2. Downward wavefield extrapolation

[0059] For the full matrix data, in order to extrapolate the wavefield from the surface to depth z, it is necessary to simultaneously migrate the transmitted wavefield and the received wavefield to depth z. Therefore, the wavefield at depth z can be obtained by the following formula:

[0060]

[0061] where k sz is the wavenumber in the vertical direction of the excitation element; k rz is the wavenumber in the vertical direction of the receiving element; x s represents the position of the excitation element; x r represents the position of the receiving element; ω represents the frequency; t represents the time.

[0062] For a homogeneous medium with a sound speed of c, as shown in (a) of Figure 2 k sz and k rz can be calculated by the following formula:

[0063]

[0064]

[0065] Therefore, the wavefield at depth z in the frequency-wavenumber domain can be obtained by applying a phase shift exp(i(k rz - k sz )z) to the surface sound field, as follows:

[0066]

[0067] Then, the total wavenumber in the vertical direction during downward extrapolation is:

[0068]

[0069] where k z is called the double square root vertical wavenumber. When the horizontal wavenumber satisfies -ω / c < k rx < ω / c and -ω / c < k sx < ω / c, k z is real and k z ∈[-ω / c, 0]. However, if the horizontal wavenumber exceeds the aforementioned limit, k z is imaginary, representing an evanescent wave, which should be filtered out.

[0070] In engineering practice, the measurement area is usually not homogeneous and can be simplified to a multi-layer structure model, as shown in Figure 2As shown in (b), the ultrasonic wave passes through N layers of material to reach the defect location and is received by the sensor after reflection. In the extrapolation process, the amplitude change of the wave when passing through the interface should be considered. Its effect can generally be modeled by the transmission coefficient, which is a function of the material impedance and the wave incident angle. Under the narrow beam assumption, the change of the transmission coefficient with the incident angle can be ignored. Therefore, there is a proportional relationship between the wave field amplitudes above and below the interface, as shown in equation (8):

[0071]

[0072] in, and Respectively represent the upper and lower sides of the lower interface of the nth layer. Therefore, the wave field calculation formula at depth z in a multi-layer medium (multi-layer structure) can be modified to:

[0073]

[0074] Among them, Φ(k rx ,k sx ,z,ω) is the corrected phase offset; D(k rx ,k sx ,ω) represents the wave field at the top of the nth layer; k z,n represents the offset wave number of the nth layer in the z direction; z n-1 Indicates the depth of the lower interface of the n-1th layer;

[0075] The offset wave number k of the mth layer in the z direction z,m Calculated by the following formula:

[0076]

[0077] In fact, it is the relative amplitude in each layer that reflects the defects, because the amplitude scaling effect introduced by the interface has little influence on the image construction and can be ignored.

[0078] 3. Imaging Conditions

[0079] The explosion reflection model assumes that the received reflection signal is actively emitted at the reflection position in the measurement area at t = 0. For B-scan data, the received signal can be directly pushed back to time zero, but the extra dimension of FMC data complicates it. In order to ensure that the downward extrapolated transmitting oscillator and the receiving oscillator converge to the same position (x, z) to simulate the coincidence of the upgoing and downgoing waves, the imaging condition of the full matrix data is defined as:

[0080] I(x,z)=∫P(x,z,x,z,ω)dω (11)

[0081] Among them, I(x,z) is the imaging result, and P(x,z,x,z,ω) is obtained by the following formula:

[0082]

[0083] Combining equations (9), (11) and (12), the imaging condition of full matrix data in a multi-layer medium (multi-layer structure) is defined as:

[0084]

[0085] Among them, k rx is the horizontal wave number of the receiving oscillator; i is the imaginary unit; k sx is the horizontal wave number of the transmitting oscillator; x represents the arrangement direction of the phased array sensor; ω represents the frequency; k z,n represents the offset wave number of the nth layer in the z direction; z represents the depth; z n-1 represents the depth of the lower interface of the n-1th layer; n∈[1,N], N represents the number of layers in the multilayer structure; k z,m represents the offset wave number of the mth layer in the z direction; d m Represents the thickness of the mth layer.

[0086] For discrete data, the integration over the frequency ω can be achieved by a superposition operation, which consumes very little time, and then a 2D inverse Fourier transform is performed in the frequency wavenumber domain to obtain the image I(x,z), as shown below:

[0087]

[0088] This imaging method only uses two operators, superposition and fast Fourier transform, and does not require the cross-correlation operation in existing imaging methods, thus greatly reducing memory usage and computational complexity.

[0089] Detection experiment

[0090] 1. Simulation experiment

[0091] In order to verify the performance of the method of this embodiment, ultrasonic FMC data were obtained in a simulation model of a multi-layer structure. Figure 3 As shown in (a), the first layer simulates the coupling layer or wedge commonly used in phased array measurement, and the next two layers are the measurement area with holes inside. The simulation simulates a 64-element phased array sensor with a spacing of 0.6mm and a center frequency of 5MHz, and collects FMC data at a sampling frequency of 50MHz and a time range of 60us. The simulation was performed in k-wave Matlab2018b, and all calculations were performed on an Intel Core TM The experiment was carried out on a workstation with i7-8700K CPU @ 3.70GHz, and all algorithms were implemented using MATLAB language.

[0092] FMC data D(x r ,x s ,t) is a three-dimensional matrix, and the five cross sections are as follows Figure 3 As shown in (b), they correspond to the signals received by all the oscillators at the 1st, 16th, 32nd, 48th and 64th excitations respectively.

[0093] Figure 3 (c) to (f) are the imaging results obtained by four imaging algorithms. The four imaging algorithms are Raybased-TFM, RMS-TFM, EPSM and the method of this embodiment, all of which are used to process FMC data to reconstruct defect images. The frequency band selected by EPSM and the method of this embodiment is from 2MHz to 8MHz. For fair comparison, before Raybased-TFM and RMS-TFM processing, the FMC data is filtered by the butter bandpass filter in MATLAB to perform a frequency selection process similar to the first two frequency domain algorithms.

[0094] Ray-based TFM uses Fermat's principle to calculate the refraction point at each interface and accurately calculate the wave propagation time. Figure 3 (c) shows the imaging result of ray-based TFM, where there are serious multiple echo artifacts in the third layer. In addition, ray tracing calculations are very time-consuming, and its computational complexity increases exponentially with the increase in the number of layers. In contrast, RMS-TFM improves computational efficiency by using RMS to approximate the speed of sound. However, its imaging quality is not as good as that of Ray-based TFM. Figure 3 As shown in (d), multiple echo artifacts still exist. The results of EPSM are as follows Figure 3 As shown in (e), it clearly images all defects and suppresses multiple echo artifacts. The imaging results of the method of this embodiment are shown in Figure 3 As shown in (f), its imaging quality is similar to that of EPSM, and multiple echo artifacts are suppressed.

[0095] In order to quantitatively analyze the imaging results, a dimensionless parameter API (array performance indicator) is introduced as follows:

[0096] API = A -6dB / λ 2 (15)

[0097] Among them, A -6dBis the area whose ratio to the peak value of the defect area image exceeds -6dB, and λ is the wave number corresponding to the center frequency. Obviously, the smaller the API, the greater the resolution. The corresponding average API values ​​at the defect points obtained by different imaging methods are listed in Table 1. The imaging resolution of the method in this embodiment is the same as that of EPSM, and the resolution of imaging the second layer defect is higher than that of other time domain methods.

[0098] Table 1 Average API of defect imaging results obtained by different methods in simulation experiments

[0099]

[0100] The four imaging methods processed the simulated FMC data six times, and their respective average calculation time and standard deviation are shown in Table 2. The calculation time of the method in this embodiment is about 1 / 2 of RMS-TFM, 1 / 6 of EPSM, and 1 / 287 of Ray-Based TFM, indicating that the method in this embodiment effectively improves the calculation efficiency of defect imaging.

[0101] Table 2 Computation time of imaging by different methods in simulation experiments

[0102]

[0103]

[0104] 2. Three-layer composite structure experiment

[0105] In order to evaluate the effect of the method of this embodiment, a three-layer composite structure was measured. Figure 4 As shown in (a), the materials of the three-layer composite structure are acrylonitrile butadiene styrene (ABS), carbon fiber reinforced polyether ketone (PEEK / CF) and aluminum. A 64-element phased array probe (Shantou Ultrasound) with a center frequency of 2.5MHz and an element spacing of 0.75mm was used to acquire FMC data in conjunction with a data acquisition card 64 / 64OEM-PA (AOS. Ltd, America). The sampling frequency was 50MHz and the time range was 60μs. Several representative slices of the measured FMC data are shown in Figure 1. Figure 4 As shown in (b).

[0106] The experimental data were post-processed using four imaging methods: Raybased-TFM, RMS-TFM, EPSM, and the method of this embodiment. The selected frequency band was 0.5MHz to 5MHz. The imaging results obtained by the four imaging methods are shown in Figure 1. Figure 4As shown in (c)-(f), the average API and time consumption of defect imaging using different methods are listed in Table 3 and Table 4 respectively. The API mean of the Raybased-TFM imaging result is the lowest, indicating that its imaging resolution is higher, but the cost of ray path calculation increases exponentially with the increase in the number of layers, reaching 207.11s in the experiment. RMS-TFM reduces the calculation time, but its bottom-layer defect imaging resolution is the worst. EPSM and the method of this embodiment have the same resolution of imaging results. The method of this embodiment is more efficient than the other three methods. The imaging time of the method of this embodiment is approximately 1 / 3 of RMS-TFM, 1 / 6 of EPSM, and 1 / 363 of Ray-Based TFM.

[0107] Table 3 Average API of defect imaging results obtained by different methods in actual experiments

[0108]

[0109] Table 4 Computation time of imaging by different methods in actual experiments

[0110]

[0111] The above is only an application example of the present invention, and does not limit the scope of the applicable samples and the types of measuring sensors. It is not necessary and impossible to list all the materials that can be measured by the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An ultrasonic phased array high-efficiency phase shift imaging method for multi-layer structure hole defects, characterized in that: The following steps are involved: (1) converting the collected full matrix data of the multilayer structure into the frequency wavenumber domain through three-dimensional fast Fourier transform; (2) for any untraversed layer in the multi-layer structure, the surface wave field of the multi-layer structure is extrapolated to the upper interface of the untraversed layer to obtain the wave field information of the untraversed layer; (3) focusing imaging the untraversed layer in the frequency wavenumber domain based on the obtained wave field information; (4) Repeating steps (2) and (3) until all layers are traversed, and obtaining an imaging result of the multilayer structure; In step (2), a double square root vertical wave number is used to extrapolate the surface wave field of the multilayer structure; The double square root vertical wave number k of any layer m in the multilayer structure z,m Calculated by the following formula: Where ω represents the frequency; c m represents the speed of sound in the mth layer; k rx is the horizontal wave number of the receiving oscillator; k sx is the horizontal wave number of the transmitting oscillator; In step (3), focusing imaging is performed on the untraversed layer in the frequency wavenumber domain with time t=0 as the imaging condition; The expression of the imaging result I(x,z) with time t=0 as the imaging condition is as follows: Among them, k rx is the horizontal wave number of the receiving oscillator; i is the imaginary unit; k sx is the horizontal wave number of the transmitting oscillator; x represents the arrangement direction of the phased array sensor; ω represents the frequency; k z,n represents the offset wave number of the nth layer in the z direction; z represents the extrapolated depth; z n-1 represents the depth of the lower interface of the n-1th layer; n∈[1,N], N represents the number of layers in the multilayer structure; k z,m represents the offset wave number of the mth layer in the z direction; d m Represents the thickness of the mth layer.

2. The ultrasonic phased array high-efficiency phase shift imaging method for multi-layer structure hole defects according to claim 1 is characterized in that: In step (2), the multi-layer structure is traversed from top to bottom.

3. The ultrasonic phased array high-efficiency phase shift imaging method for multi-layer structure hole defects according to claim 1 is characterized in that: In step (3), the specific operation of performing focused imaging on the untraversed layer in the frequency wavenumber domain according to the obtained wave field information includes the following steps: 1) In the imaging area, the wave field is recursively deduced layer by layer according to the resolution in the depth direction to obtain a discrete wave field in the frequency wavenumber domain; 2) Superimpose the obtained discrete wave fields in the frequency dimension to obtain the wave field at time 0; 3) The wave field at time 0 is transformed from the frequency wavenumber domain back to the spatial domain through two-dimensional fast Fourier transform; 4) Extract x in the spatial domain r =x s = The wave field corresponding to x is taken as the imaging result of the untraversed layer; wherein x represents the arrangement direction of the phased array sensor array; x s Indicates the position of the transmitting oscillator; x r Indicates the receiving oscillator position.

Images