A phased array ultrasonic testing method for deep delamination defects of a multilayer curved surface component

Abstract

The application provides a phased array ultrasonic detection method for deep debonding defects of a multilayer curved surface component, and belongs to the technical field of nondestructive testing.The steps of the application are as follows: obtaining acoustic parameters of the component and collecting initial ultrasonic data, constructing and verifying a longitudinal and transverse bidirectional heterogeneous medium coupling sound velocity model; matching phased array probe center frequency and array element number parameters in the balance of component acoustic characteristics and imaging quality and efficiency; collecting data based on the optimal matching parameters of the phased array probe, integrating a curvature compensation model and cooperating with attenuation-dispersion compensation to reconstruct ultrasonic data, performing wave field extrapolation processing and defect imaging; establishing an error feedback mechanism based on imaging deviation of simulation and experiment, and iteratively correcting model parameters and signal compensation coefficients.The method of the application optimizes precision through simulation-experiment feedback calibration, has the advantages of complete theory and technology closed loop, breaks through the limitations of traditional detection, can realize intelligent and accurate detection, and improves the efficiency and reliability of debonding defect detection of multilayer curved surface components.

Description

Technical Field

[0001] This invention relates to a phased array ultrasonic testing method for deep debonding defects in multilayer curved surface components, belonging to the field of nondestructive testing technology. Background Technology

[0002] Multi-layer curved surface components, due to their advantages in both lightweight structure and mechanical performance, are widely used in key fields such as aerospace, high-end equipment manufacturing, and new energy. The bonding integrity between the layers of materials directly determines the overall load-bearing capacity and service safety of the component. Deep debonding, a common hidden defect in such components, gradually expands during service, leading to a decrease in structural strength, functional failure, and even safety accidents. Therefore, accurate detection of deep debonding defects in multi-layer curved surface components is a core link in ensuring the reliable operation of related equipment. Ultrasonic testing technology, with its strong penetration ability, moderate testing cost, and convenient operation, has become one of the mainstream technologies for detecting internal defects in components and is widely used in the identification of defects in layered structures. Its core principle is to locate and characterize debonding defects by analyzing the changes in the propagation characteristics of ultrasonic waves inside the component.

[0003] Traditional ultrasonic testing methods have significant limitations when dealing with multi-layered curved components, making it difficult to meet the accuracy requirements for deep debonding detection. On the one hand, traditional testing techniques are mostly based on planar or single-layered structures, failing to fully consider the interlayer acoustic impedance coupling effect and the influence of surface curvature on ultrasonic wave propagation in multi-layered curved components. This results in large deviations in the predicted sound wave propagation path, making deep defect signals easily interfered with. On the other hand, traditional methods do not quantify the impact of transducer parameters on imaging effects, making it impossible to adjust them according to the component's material properties and defect distribution. This makes it difficult to balance penetration and resolution, and lacks a targeted signal compensation mechanism. When dealing with materials with high acoustic attenuation, the problems of defect signal amplitude attenuation and phase distortion are prominent, leading to a decrease in the signal-to-noise ratio. This, in turn, results in missed detections, false detections, or positioning errors of deep debonding defects. Furthermore, traditional testing procedures lack a systematic error calibration mechanism, and the disconnect between simulation and experimental data further restricts the improvement of testing accuracy. Summary of the Invention

[0004] The purpose of this invention is to solve the problems existing in the prior art and to provide a phased array ultrasonic detection method for deep debonding defects in multilayer curved surface components.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] A phased array ultrasonic testing method for deep debonding defects in multilayer curved surface components includes the following steps:

[0007] S1, Acoustic Parameter Measurement and Geometric Model Construction: Obtain the basic acoustic parameters of each layer of the material in the multi-layer curved surface component to be tested, including: natural sound velocity. Acoustic impedance and attenuation parameters And collect the geometric contour data and initial ultrasonic propagation data of the component;

[0008] S2, Vertical and horizontal bidirectional heterogeneous medium coupling sound velocity model: Based on the basic acoustic parameters and geometric contour data obtained in step S1, construct a vertical and horizontal bidirectional heterogeneous medium coupling sound velocity model that integrates the interlayer acoustic impedance coupling effect and the surface curvature characteristics, and complete the parameterization assignment of the model in combination with the attenuation parameters.

[0009] S3, Phased array probe parameter matching: Construct a numerical simulation environment for the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model described in step S2. Simulate the ultrasonic wave propagation process under different combinations of center frequencies and array element parameters in the numerical simulation environment. Use penetration capability and imaging resolution as evaluation indicators to select the optimal phased array detection parameters for the current component.

[0010] S4, Wavefield Extrapolation Processing and Defect Imaging: Configure the optimal phased array detection parameters described in step S3, control the phased array probe to scan the multi-layer curved surface component under test and acquire full matrix capture data; use the wavefield extrapolation algorithm to convert the acquired time-domain data to the frequency domain, combine the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model to perform interlayer sound field extension, and introduce attenuation-dispersion compensation factor and curvature phase correction factor as signal compensation coefficients to enhance the signal and reconstruct the internal defect image of the multi-layer curved surface component under test;

[0011] S5, Simulation-Experiment Closed-Loop Calibration: Compare the internal defect image described in step S4 with the imaging results corresponding to the optimal phased array detection parameter combination selected in step S3 in the simulation environment, and calculate the imaging deviation; if the imaging deviation exceeds a preset threshold, iteratively correct the parameters of the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model described in step S2 and the signal compensation coefficient in step S4 according to the deviation characteristics, and repeat imaging steps S2 to S5 based on the corrected model parameters and signal compensation coefficients until the deviation meets the preset requirements, and output the final detection result.

[0012] Preferably, in step S1, the specific steps for acquiring the geometric contour data and initial ultrasonic propagation data of the component include: using a laser contour scanner to acquire three-dimensional point cloud data of the component surface and establishing a surface curvature mapping map of the component; controlling the phased array probe to perform point measurements along the normal direction of the component surface to acquire echo time and amplitude information of interfaces at different depths; and spatially registering the three-dimensional point cloud data and echo time information to generate an initial geometric model containing thickness distribution and interlayer interface position.

[0013] Preferably, the mathematical expression for the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model in step S2 is:

[0014]

[0015] in, Coordinates The speed of ultrasonic wave propagation at that location For the number of component layers, For the first The inherent sound velocity of the layer, For the first Layer acoustic impedance deviation value, Horizontal position radius of curvature at that point For the first Layer attenuation coefficient, The coupling coefficient is... Curvature influencing factor For dynamic correction items, For depth coordinates, For the first The initial depth of the layer material, For the first The thickness of the layer material, This is the reference acoustic impedance.

[0016] Preferably, the evaluation indicators and screening strategies described in step S3 are as follows:

[0017] Construct multiple sets of alternative parameter combinations containing different center frequencies, number of array elements, and array element spacing;

[0018] Define penetration capability index The calculation formula is:

[0019]

[0020] in The amplitude of the reflected signal from the deepest debonding defect. The amplitude of the reflected signal from the surface defect;

[0021] Define imaging resolution index The calculation formula is:

[0022]

[0023] Wherein, FWHM is the full width at half maximum (FWHM) of the defect echo signal;

[0024] Define data processing efficiency evaluation indicators The calculation formula is:

[0025]

[0026] in, The total time taken from acquisition to image reconstruction for a single full matrix capture dataset;

[0027] Define comprehensive score The calculation formula is:

[0028]

[0029] in, , , The weights of each indicator are, and The parameter combination with the highest total score S is selected as the optimal phased array probe detection parameter group.

[0030] Preferably, the construction features of the numerical simulation environment in step S3 are: using a non-uniform mesh generation method, setting a high-density mesh at the interface of multi-layer media and the predetermined defect area, and the mesh size meets the following requirements. ,in The minimum wavelength within the signal bandwidth. The grid size of the simulation model is used; a perfect matching layer is set at the boundary of the simulation area to reduce the interference of boundary reflection on the internal sound field; the sound velocity distribution matrix calculated by the bidirectional heterogeneous medium coupling sound velocity model is mapped to the grid nodes as the medium parameters for solving the wave equation, ensuring that the model is consistent with the actual acoustic characteristics of the component.

[0031] Preferably, the mathematical expression for the attenuation-dispersion compensation factor model in step S4 is:

[0032] Establish frequency domain compensation function :

[0033]

[0034] in, Angular frequency, This is the index of the material layer to which the current extrapolation depth belongs, where j is the imaginary unit. The pre-power-law attenuation factor of the material at the current depth layer. To decay the power-law exponent, The depth step size; the model is calculated recursively with depth, where the first term... The second term is used to compensate for depth attenuation of signal amplitude in two-dimensional imaging. Used to correct phase dispersion distortion caused by the viscoelastic properties of materials.

[0035] Preferably, the mathematical expression for the curvature phase correction factor in step S4 is:

[0036]

[0037] in, For wave number, The horizontal axis is... is the equivalent radius of curvature when the wavefront reaches the current depth layer, and j is an imaginary unit; during the compensation process, the calculated curvature compensation amount is superimposed on the phase correction amount of the wavefield extrapolation to correct the wavefront distortion caused by the bending of the component surface and interlayer interface, and to ensure that the sound field after the wavefield extrapolation process is consistent with the actual sound field.

[0038] Preferably, the reconstructing of the internal defect image of the multi-layer curved surface component to be tested in step S4 employs a phase-shifting transfer algorithm, and the specific calculation steps are as follows:

[0039] S41, Extract the pulse echo signal from the acquired full matrix capture data. Perform a two-dimensional fast Fourier transform on it to convert it to the frequency-wavenumber domain, and obtain the initial two-dimensional wave field. ;

[0040] S42: Based on the explosion reflector model, according to depth step size The two-dimensional wave field obtained in step S41 is recursively extrapolated layer by layer, and the recursive formula is as follows:

[0041]

[0042] in, For the frequency domain wavefield data at the current depth, For the phase shift operator of sound wave propagation, This is the attenuation-dispersion compensation factor. Curvature phase correction factor;

[0043] The acoustic wave propagation phase shift operator The two-dimensional expression is:

[0044]

[0045] in, The current sound velocity of the material at the current depth layer. Angular frequency, The transverse wavenumber is along the arrangement direction of the phased array probes;

[0046] S43: Push outwards to each depth layer The wavefield data is subjected to full-band integral summation and inverse Fourier transform, based on the imaging conditions. Extract pixel values ​​to generate a two-dimensional defect image. : .

[0047] Preferably, the specific method for iterative correction based on deviation characteristics in step S5 is as follows:

[0048] Define defect location error Dimensional measurement relative error and image signal-to-noise ratio To calibrate the evaluation indicators, three levels of deviation threshold ranges were set;

[0049] When the imaging deviation is in the first interval, it is determined to be an initial geometric model mismatch. The main approach is to iteratively correct the layer thickness parameters in the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model. With natural speed of sound ;

[0050] When the imaging deviation is in the second interval, it is determined to be the attenuation-dispersion compensation factor. Model inaccuracies are mainly corrected by iteratively adjusting the pre-power-law factor in the attenuation-dispersion compensation factor. Curvature Influence Factor ;

[0051] When the imaging deviation is in the third interval, it is determined that the dispersion phase correction is insufficient, and the attenuation power-law exponent in the attenuation-dispersion compensation factor is mainly fine-tuned. and the dynamic correction term in the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model This process is used to sharpen defect edges and eliminate phase distortion until all metrics meet the preset convergence requirements.

[0052] Compared with the prior art, the phased array ultrasonic detection method for deep debonding defects in multilayer curved surface components of the present invention has the following beneficial effects:

[0053] I. This invention constructs a bidirectional coupled acoustic velocity model that integrates the effects of interlayer acoustic impedance coupling and surface curvature. By optimizing the model parameters using material attenuation parameters, it effectively addresses the problem of sound wave propagation in multi-layered curved components being affected by interlayer interactions and surface structure interference. Through a phased array probe parameter matching strategy, it selects the optimal parameter combination that balances penetration capability, resolution, and data processing efficiency, ensuring effective detection of deep debonding defects. In the imaging processing stage, a wavefield extrapolation phase shift migration algorithm is employed, innovatively incorporating an exponential attenuation-dispersion compensation and curvature phase correction model. This method can simultaneously perform layer-by-layer recursive compensation for amplitude loss caused by high-attenuation materials, phase dispersion caused by viscoelastic properties, and wavefront distortion caused by curved structures. This significantly improves the imaging resolution and signal-to-noise ratio of deep debonding defects, providing reliable technical support for the detection of deep debonding in multi-layered curved components and greatly reducing safety hazards caused by inaccurate detection.

[0054] Second, this invention utilizes a simulation-experiment closed-loop calibration mechanism and a three-level error feedback system to iteratively adjust model parameters and signal compensation coefficients to address multiple echo interference and high acoustic attenuation characteristics at multi-layer interfaces, continuously optimizing detection accuracy. The entire detection process, from basic acoustic parameter acquisition to final defect imaging, forms a complete technical closed loop. It considers both the differences in acoustic properties of the materials themselves and fully adapts to the structural characteristics of curved components, overcoming the limitations of traditional ultrasonic testing in multi-layer curved structures, such as susceptibility to interference, limited detection depth, and insufficient accuracy. Through systematic model construction, parameter optimization, signal processing, and error calibration, it achieves intelligent and precise detection, improving the detection efficiency and reliability of deep debonding defects in multi-layer curved components, and meeting the stringent requirements for component quality inspection in complex engineering scenarios.

[0055] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0056] Figure 1 This is a flowchart of a phased array ultrasonic testing method for deep debonding defects in multilayer curved surface components according to the present invention.

[0057] Figure 2 This is a schematic diagram of the wavefield coating treatment and defect imaging steps in the phased array ultrasonic detection method for deep debonding defects of multi-layer curved surface components according to the present invention. Detailed Implementation

[0058] The present invention will be further described in detail below with reference to the accompanying drawings: This embodiment is implemented under the premise of the technical solution of the present invention, and detailed implementation methods are given, but the protection scope of the present invention is not limited to the following embodiments.

[0059] like Figure 1 As shown in the figure, the phased array ultrasonic testing method for deep debonding defects in multilayer curved surface components involved in this embodiment includes the following steps:

[0060] S1, Acoustic Parameter Measurement and Geometric Model Construction: The basic acoustic parameters of each layer of the multi-layered curved surface component were measured using ultrasonic testing equipment. These basic acoustic parameters include: natural sound velocity. Acoustic impedance and attenuation parameters The component's geometric contour data was obtained using a laser contour scanner, and initial ultrasonic propagation data was acquired.

[0061] The acquisition of geometric contour data and initial ultrasonic propagation data of the component in step S1 specifically includes: acquiring three-dimensional point cloud data of the component surface using a laser contour scanner, establishing a surface curvature mapping map of the component, and accurately extracting the local curvature radius. The phased array probe is controlled to perform point measurements along the normal direction of the component surface to acquire echo time and amplitude information at interfaces at different depths. The three-dimensional point cloud data and echo time information are spatially registered to generate an initial geometric model containing thickness distribution and interlayer interface positions. During the initial acquisition of ultrasonic propagation data, a linear array probe and external synchronous triggering method are used, and the number of sampling points shall not be less than 1024 points / group.

[0062] S2, Vertical and horizontal bidirectional heterogeneous medium coupled sound velocity modeling: Construct a vertical and horizontal bidirectional coupled sound velocity model that integrates the interlayer acoustic impedance coupling effect and the surface curvature characteristics, and complete the parameterization assignment of the model by combining the attenuation parameters.

[0063] The mathematical expression for the longitudinal and transverse bidirectional coupled sound velocity model is:

[0064]

[0065] in, Coordinates The speed of ultrasonic wave propagation at that location For the number of component layers, For the first The inherent sound velocity of the layer, For the first Layer acoustic impedance deviation value, Horizontal position radius of curvature at that point For the first Layer attenuation coefficient, The coupling coefficient is... Curvature influencing factor For dynamic correction items, For depth coordinates, For the first The initial depth of the layer material, For the first The thickness of the layer material, This is the reference acoustic impedance.

[0066] S3, Phased array probe parameter matching: Construct a numerical simulation environment for the coupled sound velocity model, simulate the ultrasonic propagation process under different parameter combinations, and use penetration capability and imaging resolution as evaluation indicators to screen the optimal phased array detection parameters;

[0067] The phased array probe has four characteristic frequency points: 2MHz, 5MHz, 7.5MHz, and 10MHz. The number of phased array elements is selected from four specifications: 16 elements, 32 elements, 48 ​​elements, and 64 elements. They are arranged in a full permutation to form 16 complete parameter combinations. In the basic parameters of the probe for each parameter combination, the element spacing is determined according to the center frequency and the number of elements, the probe bandwidth is set to ±50% of the center frequency, and the transmit pulse width is set to five cycles.

[0068] Penetration capability evaluation indicators The calculation formula is:

[0069]

[0070] in, The amplitude of the reflected signal from the deepest debonding defect. The amplitude of the reflected signal from the surface defect;

[0071] Imaging resolution evaluation metrics The calculation formula is:

[0072]

[0073] Where FWHM is the half-width at half-maximum of the defect edge;

[0074] Data processing efficiency evaluation indicators The calculation formula is:

[0075]

[0076] in, The total time taken from acquisition to image reconstruction for a single full matrix capture dataset;

[0077] Overall score of parameter combination The calculation formula is:

[0078]

[0079] in, , , The weights of each indicator are, and The parameter combination with the highest total score S is selected as the optimal parameter set.

[0080] The numerical simulation environment is characterized by the following features: a non-uniform mesh generation method is used, the mesh is refined in the interlayer interface region, and the mesh size meets the requirements. ,in The minimum wavelength within the signal bandwidth. This refers to the mesh size of the simulation model; a Perfect Match Layer (PML) is set at the model boundary to reduce boundary reflection interference; the excitation signal type and amplitude are set according to the center frequency in the parameter combination list, and the corresponding number of virtual receiving array elements are matched. The sound velocity and attenuation distribution data output by the longitudinal and transverse bidirectional coupled sound velocity-attenuation joint model are imported according to spatial location association to ensure that the model and the actual acoustic characteristics of the components are consistent.

[0081] S4, Wavefield extrapolation processing and defect imaging: Configure the optimal phased array detection parameters, control the phased array probe to scan the multi-layer curved surface component under test and acquire full matrix capture data; use the wavefield extrapolation algorithm to convert the acquired time domain data to the frequency domain, combine the longitudinal and transverse bidirectional coupled sound velocity model to perform interlayer sound field extension, and introduce attenuation-dispersion compensation factor and curvature phase correction factor to enhance the signal and reconstruct the internal defect image of the component under test;

[0082] like Figure 2 As shown, the wavefield extrapolation defect image reconstruction in step S4 employs a phase shift migration algorithm, and the specific calculation steps are as follows:

[0083] Frequency domain conversion: Extracting pulse echo signals from the acquired full matrix acquisition data. Perform a two-dimensional fast Fourier transform on it to obtain the initial two-dimensional wave field. Recursive extrapolation: Based on the explosion reflector model, according to the depth step size The recursive formula for the layer-by-layer recursive extrapolation of the two-dimensional wave field is as follows:

[0084]

[0085] in, For the phase shift operator of sound wave propagation, This is the attenuation-dispersion compensation factor. This is the curvature phase correction factor.

[0086] The acoustic wave propagation phase shift operator The expression is:

[0087]

[0088] in, The current sound velocity of the material at the current depth layer. Angular frequency, The transverse wavenumber is along the arrangement direction of the phased array probes;

[0089] The attenuation-dispersion compensation factor The mathematical expression is:

[0090]

[0091] in, Angular frequency, This is the index of the material layer to which the current extrapolation depth belongs, where j is the imaginary unit. The pre-power-law attenuation factor of the material at the current depth layer. To decay the power-law exponent, For depth step size;

[0092] The model is calculated recursively with depth, where the first term The second term is used to compensate for depth attenuation of signal amplitude in two-dimensional imaging. Used to correct phase dispersion distortion caused by the viscoelastic properties of materials.

[0093] The curvature phase correction factor The mathematical expression is:

[0094]

[0095] in, For wave number, The horizontal axis is... This is the equivalent radius of curvature when the wavefront reaches the current depth layer. During the compensation process, the calculated curvature compensation amount is superimposed on the phase correction amount extrapolated from the wavefield to correct the wavefront distortion caused by the bending of the component surface and interlayer interfaces.

[0096] Finally, push it outwards to each depth layer. The wavefield data is integrated across the entire frequency band and subjected to inverse Fourier transform, depending on the imaging conditions. Extract pixel values ​​to generate a two-dimensional defect image. :

[0097]

[0098] S5, Simulation-Experiment Closed-Loop Calibration: Compare simulation and experimental imaging results, establish a three-level error feedback mechanism, and adjust model parameters and signal compensation coefficients until the accuracy requirements are met.

[0099] The specific settings of the three-level error feedback mechanism are as follows: Define the defect location error. Dimensional measurement relative error and image signal-to-noise ratio To calibrate the evaluation indicators.

[0100] First interval (geometric model mismatch): when the defect location error... Dimensional error When the problem is identified as a geometric model mismatch, the main approach is to iteratively correct the layer thickness parameters in the longitudinally and transversely coupled sound velocity model. With natural speed of sound ;

[0101] Second interval (energy attenuation misalignment): When the defect location error is within... Dimensional error within When the energy decay model is deemed inaccurate, the primary correction involves iteratively refining the pre-power-law factor in the decay-dispersion compensation factor. Curvature Influence Factor ;

[0102] Third interval (insufficient dissipative phase correction): when defect location error... Dimensional error However, if the signal-to-noise ratio is not optimal, it is determined that the dispersion phase correction is insufficient, and the attenuation power-law exponent is mainly fine-tuned. and the dynamic correction term in the sound speed model This process is used to sharpen defect edges and eliminate phase distortion until all metrics meet the preset convergence requirements.

[0103] Example 1

[0104] This embodiment uses a composite material fan casing of a certain type of aero-engine as the test object. This component is a typical multi-layer curved structure, including an outer carbon fiber composite material layer, an intermediate adhesive layer, and an inner titanium alloy lining, and the surface has varying curvature.

[0105] S1, Acoustic parameter measurement and geometric model construction:

[0106] First, the baseline acoustic parameters of each layer of the casing material were measured using a high-precision thickness gauge. First layer (carbon fiber composite layer): average thickness. Longitudinal wave speed Sound attenuation coefficient Second layer (adhesive layer): average thickness speed of sound Third layer (titanium alloy lining): thickness speed of sound Reference acoustic impedance Set as the acoustic impedance of water .

[0107] Next, a laser contour scanner (scanning precision set to...) was used. The three-dimensional point cloud data of the component surface is acquired, and a curvature mapping map is generated. In the detection area of ​​this embodiment, the local lateral position is measured. radius of curvature at exist to The data changes between these parameters. Simultaneously, the phased array probe is controlled to acquire initial A-scan data along the normal direction, with the sampling frequency set to [value missing]. The number of sampling points is By extracting the interface echo time, a preliminary geometric model containing the interlayer interface is constructed for each point / group.

[0108] S2, Modeling of sound velocity coupled in a bidirectional heterogeneous medium:

[0109] Based on the above measurement data, a bidirectional coupled sound velocity model was constructed. To simulate the smooth transition of sound velocity between layers, the smoothing factor of the hyperbolic tangent function was set to a value equal to the material thickness. The specific expression of the model is as follows:

[0110]

[0111] In this embodiment, the coupling coefficient Initial settings To correct the perturbation of wave velocity by acoustic impedance mismatch; attenuation influence factor The attenuation coefficient measured by S1 is normalized and assigned a value. Dynamic correction term. Initially set to zero, awaiting calibration in step S5.

[0112] S3, Phased array parameter matching:

[0113] Construct a numerical simulation environment, mesh size Set as (satisfy , (The minimum wavelength within the signal bandwidth). A set of candidate parameters is set: four characteristic frequency points are selected for the center frequency (2MHz, 5MHz, 7.5MHz, and 10MHz), and four specifications for the number of phased array elements (16, 32, 48, and 64 elements). These are arranged into 16 complete parameter combinations to cover different detection scenarios. In the basic probe parameters corresponding to each parameter combination, the element spacing is determined based on the center frequency and the number of elements. The probe bandwidth is set to ±50% of the center frequency, and the transmission pulse width is set to five cycles to ensure stable probe performance. A simulation model embedding a bidirectional coupled sound velocity model is constructed based on the acoustic simulation toolbox to simulate the detection effect under different parameter combinations. The penetration capability evaluation index, imaging resolution evaluation index, and data processing efficiency evaluation index for each parameter combination are calculated. The weight of each index is determined using the analytic hierarchy process (AHP), and the comprehensive score for each parameter combination is calculated using a comprehensive scoring formula. The quantitative expression of the AHP evaluation index is as follows:

[0114] Penetration capability evaluation indicators The calculation formula is:

[0115]

[0116] in, The amplitude of the reflected signal from a 15mm deep debonding defect. The amplitude of the reflected signal from a 5mm deep defect;

[0117] Imaging resolution evaluation metrics The calculation formula is:

[0118]

[0119] Where FWHM is the half-width at half-maximum of the defect edge;

[0120] Data processing efficiency evaluation indicators The calculation formula is:

[0121]

[0122] in, This represents the total time taken from acquisition to image reconstruction for a single FMC dataset.

[0123] Penetration capability evaluation indicators Imaging resolution evaluation indicators and data processing efficiency evaluation indicators Overall score of parameter combination The calculation formula is:

[0124]

[0125] in, , , The weights of each indicator are, and The parameter combination with the highest total score S is selected as the optimal parameter set.

[0126] S4, Wavefield Extrapolation Processing and Defect Imaging:

[0127] The phased array probe is configured according to the optimal parameter set, and the excitation voltage is set to... Gain Collect full matrix capture data.

[0128] Image reconstruction employs a wavefield extrapolation phase-shift transfer algorithm, with the following specific steps:

[0129] 1. Frequency Domain Transformation: Perform a two-dimensional Fourier transform on the full matrix data to convert it to frequency domain. domain.

[0130] 2. Recursive extrapolation: Set the depth extension step size In each step of the deduction:

[0131] (1) Calculate the phase shift operator Substitute the speed of sound at the current depth layer ;

[0132] (2) Calculate the attenuation-dispersion compensation factor The power-law exponent of the decay of the CFRP layer Pick (Experience points);

[0133] (3) Calculate the curvature phase correction factor ,in Updated in real time based on the curvature map in S1, for example in Substitution .

[0134] 3. Imaging: Integrate all frequency components and perform inverse Fourier transform to obtain a two-dimensional image.

[0135] S5, Simulation-Experiment Closed-Loop Calibration:

[0136] The imaging results obtained during the simulation are compared and analyzed with the actual experimental imaging results to accurately identify the differences between the two, and a three-level error feedback is implemented:

[0137] First iteration: The measured defect location deviation is (in the first interval) This was determined to be a geometric mismatch. The adhesive layer thickness in the corrected model was adjusted. Depend on Adjusted to And fine-tune the sound velocity of the carbon fiber composite layer to .

[0138] Second iteration: After re-imaging, the positioning error was reduced to However, the measured dimensions are relative error (in the second interval) The energy decay was determined to be inaccurate, and the pre-power-law factor of the decay was corrected. ,Increase The compensation gain.

[0139] Third iteration: Positioning error reduced to Dimensional error However, the edges are blurry and the signal-to-noise ratio is low. (In the third interval, signal-to-noise ratio) Fine-tuning dynamic correction terms. Sharpen the edges.

[0140] Finally, after the third iteration, the defect location error stabilized at... The following dimensional errors are less than The signal-to-noise ratio reaches The test results are output after the testing requirements are met.

[0141] In summary, this embodiment, targeting a composite material fan casing for a certain type of aero-engine, accurately achieves deep debonding detection by fully executing five steps: acoustic parameter measurement and geometric model construction, longitudinal and transverse heterogeneous medium coupling sound velocity modeling, phased array parameter matching, wavefield extrapolation processing and defect imaging, and simulation-experiment closed-loop calibration. Each step strictly adheres to predetermined parameter standards, incorporating various professional models and technologies to effectively solve the detection challenges posed by curved surface structures, interlayer coupling, and high acoustic attenuation. Ultimately, it achieves high positioning accuracy, high imaging signal-to-noise ratio, and low measurement error, providing reliable technical support for the safety assessment and maintenance of aero-engine casings.

[0142] The above description is merely a preferred embodiment of the present invention. These specific embodiments are different implementations based on the overall concept of the present invention, and the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A phased array ultrasonic testing method for deep debonding defects in multilayer curved surface components, characterized in that, Includes the following steps: S1, Acoustic Parameter Measurement and Geometric Model Construction: Obtain the basic acoustic parameters of each layer of the material in the multi-layer curved surface component to be tested, including: natural sound velocity. Acoustic impedance and attenuation parameters And collect the geometric contour data and initial ultrasonic propagation data of the component; S2, Vertical and horizontal bidirectional heterogeneous medium coupling sound velocity model: Based on the basic acoustic parameters and geometric contour data obtained in step S1, construct a vertical and horizontal bidirectional heterogeneous medium coupling sound velocity model that integrates the interlayer acoustic impedance coupling effect and the surface curvature characteristics, and complete the parameterization assignment of the model in combination with the attenuation parameters. S3, Phased array probe parameter matching: Construct a numerical simulation environment for the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model described in step S2. Simulate the ultrasonic wave propagation process under different combinations of center frequencies and array element parameters in the numerical simulation environment. Use penetration capability and imaging resolution as evaluation indicators to select the optimal phased array detection parameters for the current component. S4, Wavefield Extrapolation Processing and Defect Imaging: Configure the optimal phased array detection parameters described in step S3, control the phased array probe to scan the multi-layer curved surface component under test and acquire full matrix capture data; use the wavefield extrapolation algorithm to convert the acquired time-domain data to the frequency domain, combine the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model to perform interlayer sound field extension, and introduce attenuation-dispersion compensation factor and curvature phase correction factor as signal compensation coefficients to enhance the signal and reconstruct the internal defect image of the multi-layer curved surface component under test; S5, Simulation-Experiment Closed-Loop Calibration: Compare the internal defect image described in step S4 with the imaging results corresponding to the optimal phased array detection parameter combination selected in step S3 in the simulation environment, and calculate the imaging deviation; if the imaging deviation exceeds a preset threshold, iteratively correct the parameters of the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model described in step S2 and the signal compensation coefficient in step S4 according to the deviation characteristics, and repeat imaging steps S2 to S5 based on the corrected model parameters and signal compensation coefficients until the deviation meets the preset requirements, and output the final detection result.

2. The phased array ultrasonic testing method for deep debonding defects in multi-layer curved surface components according to claim 1, characterized in that, In step S1, the specific steps for collecting the geometric contour data and initial ultrasonic propagation data of the component include: using a laser contour scanner to acquire three-dimensional point cloud data of the component surface and establishing a surface curvature mapping map of the component; controlling the phased array probe to perform point measurements along the normal direction of the component surface to acquire echo time and amplitude information of interfaces at different depths; and spatially registering the three-dimensional point cloud data and echo time information to generate an initial geometric model containing thickness distribution and interlayer interface position.

3. The phased array ultrasonic testing method for deep debonding defects in multi-layer curved surface components according to claim 1, characterized in that, The mathematical expression for the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model described in step S2 is: in, Coordinates The speed of ultrasonic wave propagation at that location For the number of component layers, For the first The inherent sound velocity of the layer, For the first Layer acoustic impedance deviation value, Horizontal position radius of curvature at that point For the first Layer attenuation coefficient, The coupling coefficient is... Curvature influencing factor For dynamic correction items, For depth coordinates, For the first The initial depth of the layer material, For the first The thickness of the layer material, This is the reference acoustic impedance.

4. The phased array ultrasonic testing method for deep debonding defects in multi-layer curved surface components according to claim 3, characterized in that, The specific steps of the evaluation indicators and screening strategy mentioned in step S3 are as follows: Construct multiple sets of alternative parameter combinations containing different center frequencies, number of array elements, and array element spacing; Define penetration capability index The calculation formula is: in The amplitude of the reflected signal from the deepest debonding defect. The amplitude of the reflected signal from the surface defect; Define imaging resolution index The calculation formula is: Wherein, FWHM is the full width at half maximum (FWHM) of the defect echo signal; Define data processing efficiency evaluation indicators The calculation formula is: in, The total time taken from acquisition to image reconstruction for a single full matrix capture dataset; Define comprehensive score The calculation formula is: in, , , The weights of each indicator are, and The parameter combination with the highest total score S is selected as the optimal phased array probe detection parameter group.

5. The phased array ultrasonic testing method for deep debonding defects in multi-layer curved surface components according to claim 1, characterized in that, The numerical simulation environment described in step S3 is characterized by the following features: a non-uniform mesh generation method is used, and high-density meshes are set at the interfaces of multi-layer media and in the pre-defined defect areas, with mesh sizes meeting the requirements. ,in The minimum wavelength within the signal bandwidth. This refers to the mesh size of the simulation model; a perfectly matched layer is set at the boundary of the simulation region to reduce the interference of boundary reflections on the internal sound field; The sound velocity distribution matrix calculated by the bidirectional heterogeneous medium coupling sound velocity model is mapped to the grid nodes and used as the medium parameters for solving the wave equation, ensuring that the model is consistent with the actual acoustic characteristics of the component.

6. The phased array ultrasonic testing method for deep debonding defects in multi-layer curved surface components according to claim 4, characterized in that, The mathematical expression for the attenuation-dispersion compensation factor model described in step S4 is: Establish frequency domain compensation function : in, Angular frequency, This is the index of the material layer to which the current extrapolation depth belongs, where j is the imaginary unit. The pre-power-law attenuation factor of the material at the current depth layer. To decay the power-law exponent, The depth step size; the model is calculated recursively with depth, where the first term... The second term is used to compensate for depth attenuation of signal amplitude in two-dimensional imaging. Used to correct phase dispersion distortion caused by the viscoelastic properties of materials.

7. The phased array ultrasonic testing method for deep debonding defects in multi-layer curved surface components according to claim 6, characterized in that, The mathematical expression for the curvature phase correction factor mentioned in step S4 is: in, For wave number, The horizontal axis is... is the equivalent radius of curvature when the wavefront reaches the current depth layer, and j is an imaginary unit; during the compensation process, the calculated curvature compensation amount is superimposed on the phase correction amount of the wavefield extrapolation to correct the wavefront distortion caused by the bending of the component surface and interlayer interface, and to ensure that the sound field after the wavefield extrapolation process is consistent with the actual sound field.

8. The phased array ultrasonic testing method for deep debonding defects in multi-layer curved surface components according to claim 7, characterized in that, The reconstructing of the internal defect image of the multi-layer curved surface component to be tested in step S4 employs a phase-shifting transfer algorithm. The specific calculation steps are as follows: S41, Extract pulse echo signal from the acquired full matrix capture data. Perform a two-dimensional fast Fourier transform on it to convert it to the frequency-wavenumber domain, and obtain the initial two-dimensional wave field. ; S42: Based on the explosion reflector model, according to depth step size The two-dimensional wave field obtained in step S41 is recursively extrapolated layer by layer, and the recursive formula is as follows: in, For the frequency domain wavefield data at the current depth, For the phase shift operator of sound wave propagation, This is the attenuation-dispersion compensation factor. Curvature phase correction factor; The acoustic wave propagation phase shift operator The two-dimensional expression is: in, The current sound velocity of the material at the current depth layer. Angular frequency, The transverse wavenumber is along the arrangement direction of the phased array probes; S43: Push outwards to each depth layer The wavefield data is subjected to full-band integral summation and inverse Fourier transform, based on the imaging conditions. Extract pixel values ​​to generate a two-dimensional defect image. : 。 9. The phased array ultrasonic testing method for deep debonding defects in multi-layer curved surface components according to claim 8, characterized in that, The specific method for iterative correction based on deviation characteristics described in step S5 is as follows: Define defect location error Dimensional measurement relative error and image signal-to-noise ratio To calibrate the evaluation indicators, three levels of deviation threshold ranges were set; When the imaging deviation is in the first interval, it is determined to be an initial geometric model mismatch. The main approach is to iteratively correct the layer thickness parameters in the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model. With natural speed of sound ; When the imaging deviation is in the second interval, it is determined to be the attenuation-dispersion compensation factor. Model inaccuracies are mainly corrected by iteratively adjusting the pre-power-law factor in the attenuation-dispersion compensation factor. Curvature Influence Factor ; When the imaging deviation is in the third interval, it is determined that the dispersion phase correction is insufficient, and the attenuation power law exponent in the attenuation-dispersion compensation factor is mainly fine-tuned. and the dynamic correction term in the longitudinal and transverse bidirectional heterogeneous medium coupled sound velocity model This process is used to sharpen defect edges and eliminate phase distortion until all metrics meet the preset convergence requirements.

Images