Multimodal complex coherence factor full focusing imaging method for ultrasonic guided wave inspection of pipes

Abstract

The application discloses a kind of multimodal complex coherence factor full focusing imaging methods for pipeline ultrasonic guided wave detection.The method is first to the full matrix capture data pre-processing, constructs complex analytical signal matrix;Second, with torsional mode as benchmark, calculate the shortest time of flight in cylindrical coordinate system, generate benchmark complex focusing amplitude image;Subsequently extract the group velocity of the first M order guided wave mode, construct local complex coherence factor for each imaging pixel point respectively, finally with each order mode wave source amplitude factor as weight, each mode local coherence factor is adaptively weighted and fused, generate global multimodal complex coherence factor matrix, act on benchmark amplitude image, finally output pipeline defect reconstruction image.The application effectively overcomes the artifact interference caused by pipeline guided wave multimodal and dispersion characteristics, can still realize accurate defect positioning and high-quality imaging under the condition of sparse array.

Description

Technical Field

[0001] This invention belongs to the field of ultrasonic guided wave nondestructive testing and structural health monitoring, specifically relating to a multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave testing of pipelines. Background Technology

[0002] As a critical infrastructure for energy transportation, the structural integrity of pipelines is essential for their safe operation. Ultrasonic guided wave testing technology, due to its long propagation distance and low attenuation, is widely used for rapid screening of long-distance pipelines. To address the challenges of traditional single-channel testing in achieving circumferential defect localization and low sensitivity to minute defects, multi-channel array total focusing imaging (TFM) has become a current research hotspot.

[0003] However, directly applying traditional full-focusing methods to ultrasonic guided wave detection of pipelines has the following significant limitations: 1. Multimodal and dispersion interference. Under local load excitation, multiple high-order guided wave modes with different group velocities are simultaneously excited in the pipeline. Traditional TFM algorithms are usually based on the "single wave velocity" assumption, using a single velocity for delay compensation, which results in the inability of different mode signals to achieve phase alignment at the virtual focal point, producing severe multimodal artifacts. 2. Multipath propagation effect. Guided waves propagate in a spiral pattern along the cylindrical surface of the pipeline in both clockwise and counterclockwise directions, causing the same defect to generate multiple geometric diffraction echoes with different arrival times, easily forming far-point artifacts in areas far from the real defect. 3. Isochronous trajectory artifacts. Under sparse array conditions, the geometric isochronous trajectories of non-defect areas cannot be effectively suppressed, forming arc-shaped artifact bands in the image background, significantly reducing the signal-to-noise ratio and contrast of the image.

[0004] Existing full-focusing algorithms based on phase coherence factor (PCF) or symbol coherence factor (SCF) discard signal amplitude for weighted optimization during calculation, which can easily lead to misjudgment or missed detection of defects. Therefore, there is an urgent need for a full-focusing imaging optimization method that can adapt to the multimodal propagation characteristics of pipelines. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention proposes a multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave detection of pipelines. This method incorporates the multimodal propagation characteristics of guided waves and the complex coherence factor into the full-focusing framework, aiming to suppress artifacts caused by multimodal and multipath interference, and significantly improve the imaging resolution, signal-to-noise ratio, and positioning accuracy of pipeline defects.

[0006] The technical solution adopted in this invention is as follows: I. A multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave testing of pipelines Includes the following steps: Step S1: Use a ring transducer array consisting of multiple transducer elements to acquire the original guided wave echo signal of the detection pipeline; transmit the original guided wave echo signal to a computer for processing and construct a complex analytical signal matrix; Step S2: Obtain the group velocity under the reference guided wave mode, simulate the detection pipeline in the computer, and obtain the shortest total flight time between each position of the detection pipeline and the transducer array element under the group velocity of the reference guided wave mode; process the complex analytical signal matrix according to the shortest total flight time to generate a complex focused amplitude image. Step S3: Obtain the group velocity of each of the multi-order guided wave modes. In the computer, use the same processing method as the reference guided wave mode in step S2 to obtain the shortest total flight time corresponding to the group velocity of each of the multi-order guided wave modes, and then obtain the local complex coherence factor of each position of the detection pipe. Step S4: Obtain the source amplitude factors of each of the multi-order guided wave modes and use them as weighting coefficients to perform weighted fusion of the local complex coherence factors at various locations of the detected pipeline to construct a global multi-mode complex coherence factor matrix; apply the global multi-mode complex coherence factor matrix to the complex focusing amplitude image to finally output the pipeline defect reconstruction image.

[0007] Step S1 specifically involves: A ring transducer array consisting of multiple transducer elements is arranged on the outer periphery of the detection pipe for traversal excitation. Each excitation method involves sequentially selecting each transducer element as the transmitting element and using all transducer elements as the receiving elements for each transmission. The signals obtained from each excitation are collected to form the original guided wave echo signal. The original guided wave echo signal is then filtered by applying a zero-phase digital bandpass filter in the computer to obtain the filtered signal. Subsequently, the Hilbert transform is used to perform analytical reconstruction on the filtered signal to extract the envelope amplitude and phase information and construct a complex analytic signal matrix.

[0008] The zero-phase digital bandpass filter is implemented by passing the original guided wave echo signal forward through the digital bandpass filter to obtain a forward output; then, the forward output is inverted in the time domain and passed through the digital bandpass filter again to obtain an inverse filtering result; finally, the inverse filtering result is inverted in time, and then the filtered signal is analytically reconstructed using Hilbert transform to construct a complex analytic signal matrix, which is set according to the following formula: H ij (t)=S ij (t)+kS' ij (t)=A(t)·e kφ(t) ; Where · represents the dot product, i is the transmitter element number, j is the receiver element number, A(t) is the envelope amplitude, φ(t) is the signal phase, and Sij (t) is the original guided wave echo signal of transmitting element i and receiving element j at time t after the zero-phase digital bandpass filtering process, S' ij (t) is the signal obtained by performing a Hilbert transform on the filtered signals of transmitting element i and receiving element j at time t. ij (t) is the complex analytic signal generated by the reconstruction of transmitting element i and receiving element j at time t.

[0009] Step S2 specifically involves: The group velocity under the reference guided wave mode is obtained, and a simulation three-dimensional model of each transducer unit and the detection pipe is established in the computer. Then, the simulation three-dimensional model is unfolded along the axis into an imaging plane and an imaging plane coordinate system is established to obtain the coordinates of each position of the detection pipe and the coordinates of each transducer unit under the imaging plane coordinate system. Based on the coordinates of each transducer unit and each position of the detection pipe in the imaging plane coordinate system, the shortest total flight time t from the coordinates (θ, z) of each position of the detection pipe to the coordinates of each transducer unit is obtained under the reference guided wave mode group velocity. ij (θ,z) is represented as: Δθ i =min(|θ-θ i |,2π-|θ-θ i |); Δθ j =min(|θ-θ j |,2π-|θ-θ j |); d ij (θ,z)=((R·Δθ i ) 2 +(zz T ) 2 ) 0.5 +((R·Δθ j ) 2 +(zz T ) 2 ) 0.5 ; t ij (θ,z)=d ij (θ,z) / c base +t delay ; Where · represents the dot product, Δθ i Let Δθ be the shortest angular distance in the circumferential direction from each position (θ, z) of the detection pipe to the coordinate of the transmitting element i in the imaging plane coordinate system. jTo detect the shortest circumferential angular distance from each position of the pipe to the j-coordinate of the receiving array element in the imaging plane coordinate system, θ is the circumferential angular coordinate of each position of the pipe in the imaging plane coordinate system. i Let θ be the circumferential angular coordinate of the transmitting element i in the imaging plane coordinate system. j Let d be the circumferential angular coordinate of the receiving array element j in the imaging plane coordinate system, π be the mathematical constant pi, min() be the minimum value, and d be the minimum value. ij (θ,z) represents the shortest total propagation distance from the coordinates (θ,z) of each position of the detected pipe in the imaging plane coordinate system to the coordinates of the transmitting element i and the receiving element j, respectively; R is the outer radius of the detected pipe; and (θ,z) represents the coordinates of each position of the detected pipe in the imaging plane coordinate system. T The annular transducer array is arranged at equal intervals along the circumference of the detection pipe along its axial position, where z is the axial position coordinate of each location on the detection pipe in the imaging plane coordinate system, and t... delay Due to the inherent delay of the system, c base The reference guided wave mode group velocity; Based on the shortest total flight time obtained in real time, from the complex analytical signal matrix H ij Indexing is performed in (t) to obtain the corresponding complex amplitude value. Delayed summation is then performed in the complex domain to obtain the complex focused amplitude image. The formula for the complex focused amplitude image is: I base (θ,z)=Σ N i=1 Σ N j=1 H ij (t ij(θ,z) ); Among them, I base (θ,z) represents the complex amplitude of the detected pipe at each position (θ,z) on the complex focused amplitude image, H ij (t ij(θ,z) ) is the amplitude of the complex analytical signal corresponding to the shortest total flight time from each position (θ, z) in the pipeline to the transmitting array element i and the receiving array element j.

[0010] The reference waveguide mode in step S2 is the zero-order torsional mode.

[0011] Step S3 specifically involves: Using the semi-analytical finite element method, the group velocities of the first M guided wave modes are obtained by solving the pipe dispersion equation. The first M guided wave modes are processed in the same way as the reference guided wave mode in step S2 to obtain the shortest total flight time t corresponding to the coordinates of each transducer unit at each position of the pipe under the group velocities of the first M guided wave modes. ij,m (θ,z); M is the total order of the extracted guided wave modes, and m is the index, i.e., m=1,2,…,M; The local complex coherence factors at various locations of the detection pipeline under the m-th guided wave mode are obtained based on the shortest total flight time and are expressed as follows: CCF m (θ,z)=|Σ N i=1 Σ N j=1 H ij (t ij,m (θ,z))| 2 / (N 2 ·Σ N i=1 Σ N j=1 |H ij (t ij,m (θ,z))| 2 +ε); CCF m (θ,z) is the local complex coherence factor at each position (θ,z) of the detection pipe under the m-th order guided wave mode group velocity, H ij (t ij,m (θ,z) represents the complex analytic signal amplitude corresponding to the shortest total flight time of transmitting element i and receiving element j at each position (θ,z) in the m-th guided wave mode, where N is the total number of transducer elements. 2 The total number of transmitting and receiving array elements is given by ε, which is a constant to prevent the denominator from being zero. This leads to the local complex coherence factor of each position (θ, z) of the detection pipe under the first M-order guided wave modes.

[0012] Step S4 specifically involves: The source amplitude factors of the first M guided wave modes are obtained according to the normal mode expansion method. The source amplitude factors of the first M guided wave modes are used as weighting coefficients to adaptively weight and fuse the local complex coherence factors of each position of the detected pipeline under the first M guided wave modes obtained in step S3, thereby obtaining the global multimode complex coherence factors. The global multimode complex coherence factors of each position of the detected pipeline under the first M guided wave modes constitute the global multimode complex coherence factor matrix. All global multimode complex coherence factors in the global multimode complex coherence factor matrix are used as masks and applied to the reference mode amplitude image generated in step S2 for contrast adjustment and suppression. Then, through normalization and dB conversion, the final output pipeline defect reconstruction image is obtained.

[0013] The global multimodal complex coherence factor is expressed as: CCF MM (θ,z)=Σ M m=1 (ω m ·CCF m(θ,z) / (Σ) M m=1 ω m ); Among them, CCF MM (θ,z) represents the global multimodal complex coherence factor at various locations (θ,z) in the pipeline, CCF m (θ,z) is the local complex coherence factor for detecting various positions (θ,z) of the pipeline under the m-th guided wave mode, ω m is the source amplitude factor of the m-th guided wave mode, and M is the total number of extracted guided wave modes; Using all global multimodal complex coherence factors as masks, the following formula is applied to the reference mode amplitude image generated in step S2 for contrast adjustment and suppression: I MCCF (θ,z)=[CCF MM (θ,z)] γ ·I base (θ,z); Where γ is the contrast adjustment coefficient, I MCCF (θ,z) represents the complex amplitude values ​​of (θ,z) at various positions in the detection pipe after contrast adjustment. base (θ,z) represents the complex amplitude corresponding to each position (θ,z) in the detection pipeline; After normalization and dB conversion, the final output pipeline defect reconstruction image is obtained, represented as follows: 20·log 10 [I MCCF (θ,z) / Max(I MCCF (θ,z))]; CCF MM (θ,z) is the global multimodal complex coherence factor, γ is the contrast adjustment coefficient, and Max() is the maximum value.

[0014] II. A computer device The system includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method.

[0015] 3. A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention effectively overcomes the phase misalignment and positioning drift encountered by traditional full-focusing methods when applied to pipelines by independently calculating the theoretical wave velocities of multiple guided wave modes. This invention also effectively overcomes artifact interference caused by the multi-mode and dispersion characteristics of guided waves in pipelines, achieving accurate defect localization and high-quality imaging even under sparse array conditions.

[0017] 2. The proposed complex coherence factor retains the envelope amplitude of each channel as a weight, and performs multimodal weighted fusion in conjunction with the source amplitude of normal mode theory, which greatly eliminates the severe background artifact interference caused by multipath propagation and isochronous trajectory.

[0018] 3. This method effectively compensates for the imaging quality loss of sparse arrays, and can still achieve extremely high signal-to-noise ratio and positioning accuracy with only a few array elements. It significantly reduces the hardware development difficulty and engineering application cost of multi-channel ultrasonic guided wave detection equipment, and fully meets the industrial real-time detection needs when combined with GPU acceleration. Attached Figure Description

[0019] Figure 1 This is a flowchart of the Multimodal Complex Coherence Factor Total Focusing Imaging Method (MCCF-TFM).

[0020] Figure 2 This is a schematic diagram of zero-phase bandpass filtering and Hilbert analytical reconstruction.

[0021] Figure 3 This is a schematic diagram of a multi-defect pipeline model unfolded along its axis.

[0022] Figure 4 These are comparison images of the focusing effects of different imaging algorithms (traditional TFM, PCF-TFM, SCF-TFM, and the MCCF-TFM of this invention) on pipe defects. Among them, (a) is a comparison image of the focusing effect of the traditional multi-channel array full-focusing imaging method (TFM) on pipe defects, (b) is a comparison image of the focusing effect of the phase coherence factor-based multi-channel array full-focusing imaging method (PCF-TFM) on pipe defects, (c) is a comparison image of the focusing effect of the sign coherence factor-based multi-channel array full-focusing imaging method (SCF-TFM) on pipe defects, and (d) is a comparison image of the focusing effect of the multi-modal complex coherence factor-based multi-channel array full-focusing imaging method (MCCF-TFM) of this invention on pipe defects. Detailed Implementation

[0023] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the scope of protection of the present invention.

[0025] Step S1: Use a ring transducer array consisting of multiple transducer elements to acquire the original guided wave echo signal of the detection pipeline; process the original guided wave echo signal to construct a complex analytical signal matrix; Step S1 is as follows: A ring transducer array, consisting of N ultrasonic guided wave transducer elements, is arranged on the outer periphery of a detection pipe with length L, outer radius R, wall thickness d, density ρ, Young's modulus E, and Poisson's ratio μ, and then sequentially excites transducers with a center frequency f. c The ultrasonic guided wave is used for traversal excitation. A ring transducer array is formed by arranging N transducer elements at equal intervals along the circumference of the outer wall of the detection pipe. Each excitation method employs a full matrix capture (FMC) mode, sequentially selecting each transducer element as the transmitting element, with the center frequency of each transmission being f. c In ultrasonic guided wave generation, all transducer array elements are used as receiving array elements, and the signals obtained from each excitation are collected to form the original guided wave echo signal; specifically, the echo signals of all N array elements are collected simultaneously, resulting in a total of N×N channels of original time-domain signal data S. ij (t) (i is the number of the transmitting array element, j is the number of the receiving array element), forming the complete matrix dataset.

[0026] The original guided wave echo signal is filtered by a zero-phase digital bandpass filter within the computer to eliminate the phase distortion caused by the filter, resulting in a filtered signal. Then, the Hilbert transform is used to reconstruct the filtered signal analytically, extracting the distortion-free instantaneous envelope amplitude and phase information, constructing a complex analytic signal matrix, and storing it in the computer's memory.

[0027] The zero-phase digital bandpass filtering is achieved by passing the original guided wave echo signal x[n] in the positive direction through the digital bandpass transfer function H(e kω The filter is used to obtain the positive output Y1(e) kω )=H(e kω )·X(e kω The forward output y1[n] is then inverted in the time domain and passed through a digital bandpass filter again to obtain the inverse filtered result Y2(e). kω )=|H(e kω )| 2·X*(e^kω); where lowercase x and y represent the real domain, uppercase X and Y represent the frequency domain, * represents multiplication, and ^ represents exponentiation; finally, the time of the inverse filtering result is reversed to obtain the equivalent transfer function, and then the filtered signal is analytically reconstructed through Hilbert transform to construct a complex analytic signal matrix, which is set according to the following formula: H ij (t)=S ij (t)+kS' ij (t)=A(t)·e kφ(t) ; Where · represents the dot product, i is the transmitter element number, j is the receiver element number, A(t) is the envelope amplitude, representing the signal energy; φ(t) is the instantaneous phase of the signal, and S ij (t) is the filtered signal of transmitting element i and receiving element j at time t after the zero-phase digital bandpass filtering process, S' ij (t) is the orthogonal signal obtained by performing a Hilbert transform on the filtered signals of transmitting element i and receiving element j at time t. H ij (t) is the complex analytic signal reconstructed by transmitting element i and receiving element j at time t. The equivalent transfer function, obtained by zero-phase digital bandpass filtering, is set according to the following formula: H eq (e kω )=|H(e kω )| 2 ·e k·0 ; Where · represents the dot product, H eq (e kω ) is the equivalent transfer function of the zero-phase digital bandpass filtering process, e k·0 ω is the complex phase term of the equivalent transfer function, k is the imaginary unit, k·0 is the phase angle of the equivalent transfer function, where the superscript 0 indicates that the phase offset of the filtered signal is always zero, representing the zero-phase filtering characteristic; ω is the angular frequency, and e is the natural constant; that is, the amplitude frequency response is the square of the amplitude of the original filter, and the phase response is strictly zero, which holds true for any stable FIR or IIR filter, thereby ensuring the phase accuracy required for subsequent time-delay focusing while effectively removing noise.

[0028] The core mechanism of the Hilbert Transform (HT) is to apply a uniform 90° phase delay to all frequency components of the input signal, thus mapping the real signal to the complex analytic domain.

[0029] For any real signal x(t), the Hilbert transform is defined as x'(t) = (1 / π)∫x(τ) / (t-τ)dτ, and its frequency response is H(kω) = -k·sgn(ω). From this, a complex analytic signal H can be constructed. ij (t)=S ij (t)+kS' ij (t)=A(t)·e kφ(t) , where A(t)=(S ij 2 (t)+S' ij 2 (t)) 0.5 Let be the envelope amplitude, representing the signal energy; φ(t) be the instantaneous phase of the signal. Through, as shown in... Figure 2 The analytical reconstruction decoupled the signal energy envelope from the phase, laying the foundation for subsequent delay superposition and coherence assessment. Figure 2 The process of extracting the envelope from the original signal, the filtered signal, and the signal after Hilbert transform is demonstrated.

[0030] Step S2: Obtain the reference guided wave mode formed by the ring transducer array in the detection pipeline at f c The group velocity at a given frequency is simulated in the computer to obtain the shortest total flight time between each position of the detected pipeline and the transducer array element under the reference guided wave mode group velocity. The complex analytical signal matrix is ​​then processed based on the shortest total flight time to generate a reference complex focused amplitude image, which is stored in the computer's memory. Step S2 is as follows: The strongest energy mode, i.e., the reference guided wave mode f, formed by the ring transducer array in the detection pipeline was obtained through simulation. c The group velocity at the specified frequency is based on the zero-order torsional mode as the reference guided wave mode. Since multiple guided wave modes are excited simultaneously in the pipe under local excitation, the zero-order torsional mode T(0,1) exhibits non-dispersion characteristics across the entire frequency band and often dominates the energy distribution under local excitation, resulting in the most stable group velocity. Therefore, the T(0,1) mode is selected as the reference mode, and its group velocity is used to construct a reference focusing image. The order indicates the spiral propagation pattern of the guided wave mode around the pipe wall.

[0031] A simulated 3D model of each transducer unit and the detection pipe is created in the computer. Unlike the flat plate structure, when the ultrasonic guided wave propagates on the surface of the cylindrical pipe, its spatial path needs to be unfolded along the cylindrical surface. Then, the simulated 3D model is unfolded along the axis into a 2D imaging plane, and a 2D imaging plane coordinate system θOz is established with the circumferential angle of the detection pipe as the abscissa and the axial position of the detection pipe as the ordinate. Based on the positional relationship of each transducer unit and the detection pipe before unfolding, the coordinates of each transducer unit and the detection pipe in the imaging plane coordinate system are determined. The coordinates of each position of the detection pipe and the coordinates of each transducer unit in the imaging plane coordinate system are obtained, and the coordinates of each position of the detection pipe correspond to the coordinates of each position in the actual detection pipe area. Specifically, since the transducer unit has physical dimensions, the actual spatial coordinates of the center point of the transducer unit are used to represent the actual spatial coordinates of the transducer unit. Based on the positional relationship of the actual spatial coordinates of each transducer unit on the detection pipeline, the coordinates of each transducer unit in the plane coordinate system are then determined. Specifically, the two-dimensional region to be imaged is discretized by meshing. The two-dimensional imaging plane has a value range of [0, 2π) in the circumferential angle θ direction (corresponding to a circumferential unfolded length of 2πR on the outer wall of the pipe) and a value range of [0, L] in the axial position z direction. The two-dimensional imaging plane is divided into meshes, and the number of meshes can be freely adjusted. The θ direction is divided into a rows and the z direction is divided into b columns, thereby obtaining a total of axb discrete points, which are pixels. The pixels correspond to the various positions of the detected pipe.

[0032] Based on the coordinates of each transducer unit and the various positions of the detection pipe in the imaging plane coordinate system, the reference guided wave mode f is obtained. c At group velocity at a given frequency, the shortest total flight time t corresponding to the coordinates of each position on the detection pipeline and the coordinates of each transducer element in the transmit / receive array is determined. ij (θ,z) is represented as: Considering that higher-order helical guided waves can propagate clockwise or counterclockwise along the circumference, the time-of-flight calculation of the reference guided wave mode only considers the shortest direct path along the surface. The shortest circumferential angular distance between the pixel coordinates P and the array element i at each position of the detection pipe in the imaging plane coordinate system is: Δθ i =min(|θ-θ i |,2π-|θ-θ i |); Δθ j =min(|θ-θ j |,2π-|θ-θ j |); The shortest total propagation distance of the reference guided wave mode from coordinate i of the transmitting array element to coordinate P of the pixel point and then reflected back to coordinate j of the receiving array element is: d ij (θ,z)=((R·Δθ i ) 2 +(zz T ) 2 ) 0.5 +((R·Δθ j ) 2 +(zz T ) 2 ) 0.5 ; By combining the baseline mode group velocity and the system's inherent delay, the shortest total flight time t is obtained. ij (θ,z): t ij (θ,z)= d ij (θ,z) / c base +t delay ; Where · represents the dot product, Δθ i To detect the shortest angular distance in the circumferential direction from each position (θ, z) of the pipeline to the coordinates of the transmitting array element i, Δθ j To detect the shortest circumferential angular distance from the coordinates of each position on the pipe to the coordinates of the receiving array element j in the imaging plane coordinate system, θ is the circumferential angular coordinate of each position on the pipe in the imaging plane coordinate system. i Let θ be the circumferential angular coordinate of the transmitting element i in the imaging plane coordinate system. j Let d be the circumferential angular coordinate of the receiving array element j in the imaging plane coordinate system, π be the mathematical constant pi, min() be the minimum value, and d be the minimum value. ij (θ,z) represents the shortest total propagation distance from the coordinates of each position in the detected pipe to the coordinates of the transmitting element i and the receiving element j in the imaging plane coordinate system, R is the outer radius of the detected pipe, and (θ,z) represents the coordinates of pixel point P at each position in the detected pipe in the imaging plane coordinate system. T The annular transducer array is arranged at equal intervals along the circumference of the detection pipe along its axial position, where z is the axial position coordinate of each location on the detection pipe in the imaging plane coordinate system, and t... delay Due to the inherent delay of the system, c base The reference guided wave mode group velocity; Based on the shortest total flight time obtained in real time, from the complex analytical signal matrix H ij Indexing is performed in (t) to obtain the corresponding complex amplitude. Delayed summation is then performed in the complex domain, and the summation is iterated through all N×N transmit / receive channel combinations to obtain the reference complex focused amplitude image. The formula for generating the reference complex focused amplitude image is: I base (θ,z)=ΣN i=1 Σ N j=1 H ij (t ij(θ,z) ); Among them, I base (θ,z) represents the complex amplitude of (θ,z) at various positions of the detected pipe on the complex focused amplitude image, H ij (t ij(θ,z) This image represents the complex analytical signal amplitude corresponding to the shortest total flight time from (θ, z) to transmitting element i and receiving element j at various locations along the pipeline. It fully preserves the energy distribution of the signals from each channel in the complex domain under the reference mode, serving as the basis for subsequent coherence factor weighting.

[0033] Step S3: Obtain the group velocity of the ring transducer array under the multi-order guided wave modes formed by the detection pipe. In the computer, use the same processing method as the reference guided wave mode in step S2 to obtain the shortest total flight time corresponding to the group velocity under the multi-order guided wave modes. Then, calculate the local complex coherence factor of each position of the detection pipe under the multi-order guided wave modes based on the shortest total flight time and store it in the computer's memory. Step S3 is as follows: In a computer, the semi-analytical finite element method (SAFE) is used to solve the pipe dispersion equation for the dimensional parameters (outer radius R and inner radius Rd) and mechanical property parameters (density ρ, Young's modulus E, and Poisson's ratio μ) of the tested pipe. From the solution of the pipe dispersion equation, the first M main guided wave modes at f are obtained. c The group velocity at frequency f is obtained by applying the same processing method to the first M principal guided wave modes as to the reference guided wave mode in step S2. c At group velocity at a given frequency, the shortest total flight time t corresponding to the coordinates of each position on the detection pipeline and the coordinates of each transducer element in the transmit / receive array is determined. ij,m (θ,z) is represented as: t ij,m (θ,z)= d ij (θ,z) / c g,m + t delay ; Where, d ij (θ,z) represents the shortest total propagation distance from each position (θ,z) in the detection pipeline to the coordinates of the transmitting element i and the receiving element j, respectively; t ij,m (θ,z) represents the shortest total flight time of the m-th order guided wave mode from each position (θ,z) in the detection channel to the coordinates of the transmitting element i and the receiving element j; c g,m The group velocity of the m-th dominant guided wave mode is given by (θ, z); (θ, z) represents the coordinates of each position in the detection pipe; tdelay The system inherent delay is M; the total number of extracted guided wave modes is M, which is generally taken as 5; m is the index, i.e., m = 1, 2, ..., M; Based on the shortest total flight time, the ratio of the total energy of the coherent superposition of complex analytical signals to the scalar sum of the incoherent energies of each channel is obtained. This leads to the local complex coherence factor at each location in the detection pipeline under the m-th guided wave mode, expressed as: CCF m (θ,z)=|Σ N i=1 Σ N j=1 H ij (t ij,m (θ,z))| 2 / (N 2 ·Σ N i=1 Σ N j=1 |H ij (t ij,m (θ,z))| 2 +ε); CCF m (θ,z) is the local complex coherence factor for the m-th guided wave mode, H ij (t ij,m (θ,z) represents the complex analytic signal amplitude corresponding to the shortest total flight time of transmitting element i and receiving element j at the m-th guided wave mode group velocity, and N is the total number of transducer elements. 2 Let ε be the total number of transmit and receive array elements in the full matrix, and let ε be a minimal constant to prevent the denominator from being zero. This allows us to obtain the local complex coherence factors at various locations in the pipeline under the first M guided wave modes. The complex analytic signal amplitudes between channels are retained as adaptive weights. Echo signals caused by real defects occupy a larger weight in the coherence assessment; even if other interfering modes are accidentally in phase, their contribution to the overall coherence is negligible, thus effectively removing noise artifacts without sacrificing the detection sensitivity of the sparse array.

[0034] Step S4: Obtain the source amplitude factors of each of the multi-order guided wave modes and use them as weighting coefficients to perform weighted fusion of the local complex coherence factors at various locations of the detected pipeline under the multi-order guided wave modes, construct a global multi-mode complex coherence factor matrix, and store it in the computer's memory; apply the global multi-mode complex coherence factor matrix to the reference complex focusing amplitude image, and finally output the pipeline defect reconstruction image; Step S4 is as follows: To quantify the contribution of each higher-order guided wave mode to full-focus imaging, this invention uses the normal mode expansion method to derive the source amplitude factor ω of different circumferential order modes under local load excitation. mThis factor reflects the proportion of energy carried by each guided wave mode under a given excitation condition. It can be used as a weighting coefficient for the local coherence factor of each mode, achieving adaptive fusion with clear physical meaning.

[0035] In a computer simulation, the source amplitude factors of the first M guided wave modes formed by the ring transducer array in the detection pipeline are obtained according to the normal mode expansion method. These factors reflect the energy proportion carried by each guided wave mode under a given excitation condition and can be used as weighting coefficients for the local coherence factors of each mode, achieving physically meaningful adaptive fusion. Using the source amplitude factors of the first M guided wave modes as weighting coefficients, the local complex coherence factors at various locations in the detection pipeline under the first M guided wave modes obtained in step S3 are adaptively weighted and fused to obtain the global multi-mode complex coherence factor calculation formula, expressed as: CCF MM (θ,z)=Σ M m=1 (ω m ·CCF m (θ,z) / (Σ) M m=1 ω m ); Among them, CCF MM (θ,z) represents the global multimodal complex coherence factor at various locations (θ,z) in the pipeline, CCF m (θ,z) is the local complex coherence factor for detecting various positions (θ,z) of the pipeline under the m-th guided wave mode, ω m is the source amplitude factor of the m-th guided wave mode, and M is the total order of the extracted guided wave modes, typically taken as 5. The global multimode complex coherence factors corresponding to the first M guided wave modes constitute the global multimode complex coherence factor matrix.

[0036] Due to the local complex coherence factor (CCF) at various locations in the pipeline under each mode, m The values ​​of all values ​​are in the range [0,1]. The global multimodal complex coherence factor (CCF) obtained after weighted averaging is... MM Similarly, the values ​​are between [0,1]. A higher value indicates that the waveguide echo corresponding to that pixel is more likely to come from a real scatterer; a lower value indicates that the signal coherence at that pixel is poor, corresponding to background noise or artifact areas.

[0037] Using all global multimodal complex coherence factors in the global multimodal complex coherence factor matrix generated in step S4 as a mask, the following formula is applied to the reference mode amplitude image generated in step S2 for contrast adjustment and suppression: I MCCF (θ,z)=[CCF MM (θ,z)]γ ·I base (θ,z); CCF MM (θ,z) represents the global multimodal complex coherence factor at each position (θ,z) on the pipeline, and γ is the contrast adjustment coefficient γ≥1, used to further suppress background artifacts. The default value is γ=1. MCCF (θ,z) represents the complex amplitude values ​​(θ,z) at various positions on the detected pipe in the fully focused image after contrast adjustment using the multimodal complex coherence factor. base (θ,z) represents the complex amplitude corresponding to each position (θ,z) of the detection pipe in the reference modal amplitude image; Then, through normalization and dB conversion, the amplitude image is converted to a logarithmic decibel scale, which effectively compresses the dynamic range of the fully focused image, enhances the contrast between weak defect echoes and background noise in the image, and improves the detection rate of small defects, resulting in a final output high signal-to-noise ratio, high-resolution pipe defect reconstruction image, represented as: 20·log 10 [I MCCF (θ,z) / Max(I MCCF (θ,z))]; CCF MM (θ,z) is the global multimodal complex coherence factor, and γ is the contrast adjustment coefficient γ≥1, which is used to further suppress background artifacts. The default value is γ=1, and Max() is to take the maximum value.

[0038] The verification of the full matrix dataset of the Abaqus 3D pipeline finite element simulation model is illustrated with a specific example: an industrial pipeline model with an outer radius R=201mm, wall thickness d=10mm, and length L=2m, and its mechanical performance parameters are: density ρ=7800kg / m³. 3 The elastic modulus E = 210 MPa, Poisson's ratio μ = 0.28, and 16 transducer elements are arranged at equal intervals on the same axial section; the excitation signal is the center frequency f. c A 5-cycle Hanning window modulated sinusoidal pulse at 64kHz was used. Four circular defects with dimensions Φ1, Φ2, Φ3, and Φ4 were added to the pipe surface, with axial distances L1, L1+L2, L1+L2+L3, and L1+L2+L3+L4 from the transducer array, respectively. These defects were arranged at 90° intervals circumferentially to simulate the defect distribution in an actual pipe. The pipe model, unfolded along the axis, is shown below. Figure 3 The locations of transducer array elements and the distribution of defects are marked.

[0039] After obtaining the full matrix data using the above simulation model, the multimodal complex coherence factor full-focusing imaging method proposed in this invention generates the final defect reconstruction image through the aforementioned steps of analytical reconstruction, reference mode focusing, multimodal local coherence factor extraction and global adaptive weighted fusion.

[0040] To verify the imaging quality and anti-interference capability of the present invention, Figure 4 A comparison of the focusing effects of different imaging algorithms (traditional TFM, PCF-TFM, SCF-TFM and the MCCF-TFM of this invention) on the pipeline defect is presented. Figure 4 (a) is a comparison chart of the focusing effect of traditional TFM on pipeline defects. Figure 4 (b) is a comparison chart of the focusing effect of traditional PCF-TFM on pipeline defects. Figure 4 (c) is a comparison chart of the focusing effect of traditional SCF-TFM on pipeline defects. Figure 4 (d) is a comparison diagram of the focusing effect of the MCCF-TFM of the present invention on pipeline defects. It can be seen that, under sparse array conditions, the present invention, while maintaining defect focusing, greatly eliminates background noise, achieves accurate circumferential and axial positioning, and exhibits extremely high signal-to-noise ratio and resolution.

[0041] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave testing of pipelines, characterized in that, Includes the following steps: Step S1: Use a ring transducer array to acquire the original guided wave echo signal of the detection pipeline and transmit it to the computer to construct a complex analytical signal matrix; Step S2: Simulate the detection pipeline in the computer to obtain the shortest total flight time from the detection pipeline to the transducer array under the group velocity of the reference guided wave mode, and generate a complex focusing amplitude image by combining the complex analytical signal matrix; Step S3: In the computer, using the same method as in step S2, obtain the shortest total flight time from the detection pipe to the transducer array under the group velocity of each of the multi-order guided wave modes, and obtain the local complex coherence factor corresponding to the detection pipe. Step S4: Obtain and weight the local complex coherence factors according to the respective source amplitude factors of the multi-mode guided wave, construct the global multi-mode complex coherence factor matrix, and output the pipeline defect reconstruction image by combining the complex focusing amplitude image.

2. The multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave detection of pipelines according to claim 1, characterized in that, Step S1 specifically involves: A ring transducer array consisting of multiple transducer elements is arranged on the outer periphery of the detection pipe for traversal excitation. Each excitation method involves sequentially selecting each transducer element as the transmitting element and using all transducer elements as the receiving elements for each transmission. The signals obtained from each excitation are collected to form the original guided wave echo signal. The original guided wave echo signal is then filtered by applying a zero-phase digital bandpass filter in the computer to obtain the filtered signal. Subsequently, the Hilbert transform is used to perform analytical reconstruction on the filtered signal to extract the envelope amplitude and phase information and construct a complex analytic signal matrix.

3. The multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave detection of pipelines according to claim 2, characterized in that: The zero-phase digital bandpass filter is implemented by passing the original guided wave echo signal forward through the digital bandpass filter to obtain a forward output; then, the forward output is inverted in the time domain and passed through the digital bandpass filter again to obtain an inverse filtering result; finally, the inverse filtering result is inverted in time, and then the filtered signal is analyzed and reconstructed through Hilbert transform to construct a complex analytic signal matrix.

4. The multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave detection of pipelines according to claim 2, characterized in that, Step S2 specifically involves: The group velocity under the reference guided wave mode is obtained. A simulated 3D model of each transducer unit and the detection pipe is built in the computer. Then, the simulated 3D model is unfolded along the axis into an imaging plane and an imaging plane coordinate system is established. Based on the coordinates of each position of the detection pipe and the coordinates of each transducer unit under the imaging plane coordinate system, the shortest total flight time t from each position of the detection pipe to each transducer unit under the reference guided wave mode group velocity is obtained. ij (θ,z) is represented as: Dth i =min(|θ-θ i |,2π-|θ-θ i |); Dth j =min(|θ-θ j |,2π-|θ-θ j |); d ij (θ,z)=((R Δθ i ) 2 +(zz T ) 2 ) 0.5 +((R Δθ j ) 2 +(zz T ) 2 ) 0.5 ; t ij (θ,z)=d ij (θ,z) / c base +t delay ; Where θ, θ i、 θ j、 Δθ i、 Δθ j The following parameters are, in order: the circumferential angular coordinates of each position on the detection pipeline, the circumferential angular coordinates of transmitting element i and receiving element j, and the shortest circumferential angular distance from each position (θ, z) to the coordinates of transmitting element i and transmitting element j. π represents pi, min() is the minimum value, and d... ij (θ,z) represents the shortest total propagation distance from each position (θ,z) in the detection pipe to the transmitting element i and the receiving element j, where R is the outer radius of the detection pipe, and z T Z represents the axial position of the annular transducer, z represents the axial position coordinates of various locations on the detection pipeline, and t represents the axial position of the transducer. delay Due to the inherent delay of the system, c base The reference guided wave mode group velocity; Based on the shortest total flight time obtained in real time, from the complex analytical signal matrix H ij The corresponding complex amplitude is obtained by indexing (t), and a delay summation process is performed to obtain a complex focused amplitude image.

5. The multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave detection of pipelines according to claim 4, characterized in that: The reference waveguide mode in step S2 is the zero-order torsional mode.

6. The multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave detection of pipelines according to claim 1, characterized in that, Step S3 specifically involves: Using the semi-analytical finite element method, the group velocities of the first M guided wave modes are obtained by solving the pipe dispersion equation. The same processing method as in step S2 is applied to all the first M guided wave modes to obtain the shortest total flight time t from each position in the pipe to each transducer unit when using the group velocities of the first M guided wave modes. ij,m (θ,z); M is the total order of the extracted guided wave modes, and m is the index, i.e., m=1,2,…,M; The local complex coherence factors at various locations of the detection pipeline under the m-th guided wave mode are obtained based on the shortest total flight time and are expressed as follows: CCF m (θ,z)=|Σ N i=1 Σ N j=1 H ij (t ij,m (θ,z))| 2 / (N 2 ·Σ N i=1 Σ N j=1 |H ij (t ij,m (θ,z))| 2 +ε); CCF m (θ,z) is the local complex coherence factor at each position (θ,z) of the detection pipe under the m-th order guided wave mode, H ij (t ij,m (θ,z) represents the complex analytic signal amplitude corresponding to the shortest total flight time of transmitting element i and receiving element j at each position (θ,z) in the m-th guided wave mode, where N is the total number of transducer elements. 2 ε represents the total number of transmit and receive array elements combined, and ε is a constant to prevent the denominator from being zero.

7. The multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave detection of pipelines according to claim 1, characterized in that, Step S4 specifically involves: The source amplitude factors of the first M waveguide modes are obtained according to the normal mode expansion method. The source amplitude factors of the first M waveguide modes are used as weighting coefficients to adaptively weight and fuse the local complex coherence factors of the detection pipeline at each position under the first M waveguide modes obtained in step S3, thereby obtaining the global multimode complex coherence factors. The global multimode complex coherence factors of the detection pipeline at each position under the first M waveguide modes constitute the global multimode complex coherence factor matrix. All global multimodal complex coherence factors in the global multimodal complex coherence factor matrix are used as masks and applied to the reference mode amplitude image generated in step S2 for contrast adjustment and suppression. Then, through normalization and dB conversion, the final output pipeline defect reconstruction image is obtained.

8. A multimodal complex coherence factor full-focusing imaging method for ultrasonic guided wave detection of pipelines according to claim 7, characterized in that: Global multimodal complex coherence factors are obtained through adaptive weighted fusion. All global multimodal complex coherence factors are used as masks to perform contrast adjustment and suppression on the baseline modality amplitude image generated in step S2, as follows: CCF MM (θ,z)=Σ M m=1 (oh m ·CCF m (θ,z)) / (Σ M m=1 oh m ); I MCCF (θ,z)=[CCF MM (θ,z)] γ ·I base (θ,z); Among them, CCF MM (θ,z) represents the global multimodal complex coherence factor at various locations (θ,z) in the pipeline, CCF m (θ,z) is the local complex coherence factor for detecting various positions (θ,z) of the pipeline under the m-th guided wave mode, ω m γ is the source amplitude factor of the m-th guided wave mode, M is the total order of the extracted guided wave modes, γ is the contrast adjustment coefficient, and I... MCCF (θ,z) represents the complex amplitude I at each position (θ,z) of the detection pipeline after contrast adjustment. base (θ,z) represents the complex amplitude corresponding to each position (θ,z) in the detection pipeline.

9. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-8.

Images