Ultrasonic phased array imaging-based non-destructive testing method for nuclear power metal welded joints and application

Abstract

The present application relates to the technical field of nondestructive testing of nuclear power equipment, and provides a nuclear power metal welded joint nondestructive testing method based on ultrasonic phased array imaging and application, comprising calculating the ultrasonic longitudinal wave velocity and transverse wave velocity of the nuclear power metal material; calculating the defect acoustic scattering response distribution under different angles of the organic glass wedge; using an ultrasonic array detection system, identifying bright spot abnormal points in an ultrasonic phased array scanning image and full matrix capture data corresponding to the positions; using a time reversal operator decomposition algorithm to perform signal denoising processing, obtaining denoised ultrasonic array signals; calculating the propagation time of the sound waves emitted by each element of the ultrasonic phased array after refraction through the glass wedge to each grid point, obtaining the Hilbert transform value of each grid point corresponding to the propagation time, and generating an ultrasonic imaging graph of the region; performing delay multiplication superposition and vector coherence factor denoising processing on the imaging data, and obtaining a welded joint nondestructive testing imaging result.

Description

Technical Field

[0001] This invention relates to the technical field of non-destructive testing of nuclear power equipment, and in particular to a non-destructive testing method for nuclear power metal welded joints based on ultrasonic phased array imaging and its application. Background Technology

[0002] Nuclear power equipment plays a crucial role in nuclear power generation systems, providing pressure resistance, sealing, and structural support. Its operational safety directly impacts the stable operation of nuclear power plants and public safety. Nuclear power plant welded joints, as the most stress-concentrated and harshest-environmental components within the equipment, are subject to long-term exposure to thermal cycling, radiation, corrosion, and mechanical vibration, making them highly susceptible to structural defects such as cracks, inclusions, and porosity. Failure to detect these defects in a timely manner can lead to leaks, structural failures, or even major safety accidents. Therefore, research into high-precision, reliable non-destructive testing technologies for nuclear power plant welded joints is of significant engineering and safety importance.

[0003] While commonly used methods such as X-ray inspection and CT imaging offer high resolution, they suffer from drawbacks including high radiation risk, significant limitations imposed by on-site space, high cost, and difficulty in achieving online inspection, making them unsuitable for the rapid inspection needs of nuclear power equipment in complex environments. In contrast, ultrasonic phased array technology possesses advantages such as strong penetration capability, high positioning accuracy, and the ability to achieve multi-angle imaging, making it an important means of inspecting welds in nuclear power plants. However, the internal microstructure of nuclear power plant welded joints is complex, with weld zones generally exhibiting coarse-grained structures, anisotropy, and non-uniform microstructure. This leads to strong scattering and attenuation of ultrasonic waves during propagation, causing traditional imaging methods to suffer from noise enhancement, blurred defect boundaries, and prominent artifacts, severely impacting the reliability of defect identification.

[0004] Traditional ultrasonic imaging methods, such as total focusing (TFM), are widely used in nuclear power plant weld inspection, but their imaging quality is highly dependent on the signal-to-noise ratio. When the weld zone contains coarse-grained structures and grain sizes significantly larger than the base material, ultrasound undergoes strong scattering, attenuation, and mode conversion at grain boundaries, resulting in significantly enhanced background noise and causing defect echoes to be submerged or distorted. Especially in the high-residue welds commonly found in nuclear power equipment, the surface undulations and geometric irregularities further increase the roughness of the incident interface, reducing the coupling efficiency of normally incident ultrasonic energy. Traditional methods are therefore unable to effectively detect minute defects in the fusion region between the weld root and sidewall.

[0005] Although oblique incidence ultrasonic technology can improve the incidence efficiency of welds with high reinforcement, allowing the ultrasonic beam to achieve a longer propagation distance and a better defect irradiation angle after entering the weld, the problem of scattering noise in coarse-grained welds has not been effectively solved. A large amount of random scattering signals will accumulate in the full matrix capture data, causing conventional focusing algorithms, even if they can locate defects, to exhibit problems such as blurred boundaries, increased artifacts, and insufficient defect contrast due to noise interference, which seriously affects the accuracy of detection and the reliability of defect identification.

[0006] Therefore, significantly suppressing scattering noise caused by coarse-grained structures while maintaining the geometric adaptability of oblique-incidence ultrasound to weld seams with high backing is a key technical bottleneck in improving the imaging quality of non-destructive testing of nuclear power plant welds. To address this issue, there is an urgent need to develop an ultrasonic phased array imaging method that combines the characteristics of full-matrix capture data with strong noise suppression capabilities. This method would enhance signal energy focusing and noise separation capabilities at the algorithmic level, thereby obtaining clearer and more reliable defect imaging results and providing a reliable basis for the safety assessment of nuclear power equipment. Summary of the Invention

[0007] To overcome the shortcomings of existing technologies, this invention proposes a non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging. Addressing the issues of strong scattering noise caused by coarse-grained microstructure in the weld region and the sensitivity of traditional imaging algorithms to noise and blurred defect boundaries, this invention significantly improves the imaging contrast and reliability of internal weld defects by performing signal decomposition and noise suppression on the full-matrix capture data and combining the advantages of oblique incidence acoustic beams. This invention focuses on reducing scattering noise and improving imaging clarity. It fully utilizes the information redundancy of multi-channel phased array data, introduces time-reversal operator decomposition technology to achieve signal denoising, and then uses delay multiplication superposition and vector coherence factor to enhance the true reflected echo, thereby obtaining high signal-to-noise ratio and high reliability imaging results of the internal weld structure.

[0008] To achieve the above objectives, the present invention provides a non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging, the testing method comprising:

[0009] S1. Determine the density of the nuclear power metal material based on the properties of the tested material, and use a single ultrasonic longitudinal and transverse wave probe to excite ultrasonic longitudinal and transverse waves on the surface of the metal material. Calculate the ultrasonic longitudinal wave velocity and transverse wave velocity of the nuclear power metal material based on the thickness of the material and the bottom echo time.

[0010] S2, assuming that the ultrasonic single probe is excited on the plexiglass wedge and incident on the weld area in an oblique manner, calculate the defect acoustic scattering response distribution under different plexiglass wedge angles based on the metal material density, longitudinal wave velocity and transverse wave velocity, and determine the optimal wedge angle based on the spatial distribution of ultrasonic energy, that is, at this angle, the sound wave energy completely covers the area to be detected and the energy is as large as possible.

[0011] S3 utilizes an ultrasonic array inspection system to perform oblique incidence phased array scanning of the nuclear power plant metal welded joint at the optimal angle using an plexiglass wedge, acquiring ultrasonic phased array scan images. The accompanying software of the ultrasonic array inspection system identifies abnormal bright spots in the ultrasonic phased array scan images, preliminarily determining that the abnormal bright spots are suspected defect areas, and saving the full matrix capture data (FMC data) corresponding to the abnormal bright spots, including key parameters such as the amplitude, propagation time, and phase of the ultrasonic signal, ensuring the integrity and traceability of the data, and providing raw data support for subsequent signal processing and defect identification.

[0012] S4. For the full matrix capture data, the time reversal operator decomposition algorithm is used to perform signal denoising processing to obtain the denoised ultrasonic array signal.

[0013] S5. The nuclear power metal welded joint area is divided into grids with an accuracy of 0.1mm. Combining the longitudinal and transverse wave velocities of the nuclear power metal, the ultrasonic propagation distance, and the influence of bottom surface reflection, the propagation time of the sound wave emitted by each element of the ultrasonic phased array after being refracted by the glass wedge to reach each grid point is calculated. Then, based on the full matrix capture data, the Hilbert transform value of the propagation time corresponding to each grid point is obtained and used as the pixel value of the grid point to generate an ultrasonic imaging map of the area.

[0014] S6 performs Delayed Multiplication and Stacking (DMAS) and Vector Coherence Factor (VCF) noise reduction on the imaging data to obtain clearer and more reliable non-destructive testing imaging results of welded joints.

[0015] Furthermore, in step S1, the ultrasonic longitudinal wave velocity of the nuclear power metal material... transverse wave velocity They are respectively:

[0016]

[0017] In the formula, The thickness of the nuclear power metal material; and These are the bottom echo times received during ultrasonic longitudinal / transverse wave excitation.

[0018] Furthermore, the formula for calculating the acoustic scattering response distribution in step S2 is:

[0019]

[0020] In the formula, These represent the ultrasonic excitation and reception processes, respectively. These represent the refractive indices during the ultrasonic excitation and reception processes, respectively. These represent path-based ultrasonic attenuation during ultrasonic excitation and reception processes, respectively. These represent the phased array aperture coefficients for the ultrasonic excitation and reception processes, respectively. This represents the scattering coefficient when the ultrasonic wave encounters a scattering body during excitation and reception; the response at each point within the region is then calculated. The acoustic scattering response distribution map can then be obtained.

[0021] Furthermore, the abnormal bright spot locations mentioned in step S3 are the positions where high-energy reflections are observed in the B-scan or S-scan images during phased array scanning by computer software.

[0022] Furthermore, the specific steps of step S4 are as follows:

[0023] The full-matrix captured data is transformed into frequency domain data using Fourier transform to obtain the frequency domain matrix. ;

[0024] For frequency domain matrix Perform singular value decomposition;

[0025]

[0026] in, Represents a left singular vector. Let represent a real diagonal matrix composed of positive singular values. Represents a right singular vector;

[0027] Further decomposition of the frequency domain matrix It is decomposed into the defect signal corresponding part and the noise signal corresponding part;

[0028]

[0029] in, For eigenvalues, For frequency The number of singular values ​​corresponding to the lower signal varies depending on the defect. The number of singular values ​​corresponding to the noise signal; The number of singular values ​​in the denoised signal; This is the number of singular values ​​in the decomposition, typically 64; It is a left singular vector; It is a right singular vector;

[0030] By extracting the defect signal corresponding to the left of the plus sign in the frequency domain matrix decomposition formula, and performing an inverse Fourier transform on this part to obtain the reconstructed time-domain full matrix capture number (FMC data), the denoised ultrasonic array signal can be obtained. This denoised signal can effectively improve the signal-to-noise ratio of the defect signal, clearly present the location, size and morphological characteristics of the defect, and solve the problems of fuzzy defect identification and misjudgment caused by noise interference in the original data, laying the foundation for subsequent qualitative and quantitative analysis of defects.

[0031] Furthermore, the specific steps of step S5 are as follows:

[0032] The area of ​​the nuclear power plant metal welded joint was divided into grids with a precision of 0.1 mm, and a grid spatial coordinate model was established, with each grid point corresponding to a unique spatial coordinate. , This ensures that the grid completely covers the detection area without omissions or overlaps, providing a discretized spatial reference for subsequent point-by-point imaging.

[0033] Based on the propagation path of the ultrasonic transmitter and receiver, the propagation time of each element of the ultrasonic phased array after refraction by the glass wedge to each grid point is calculated. The calculation formula is as follows:

[0034]

[0035] In the formula, and These are the spatial coordinates of the excitation array element and the receiving array element, respectively. and These are the abscissas of the refraction points at the wedge-sample interface during ultrasonic excitation and reception, respectively. The speed of the ultrasonic wave within the plexiglass wedge. It is the speed of the ultrasonic wave in the weld base material; , () represents the coordinates of the p-th grid point within the imaging region; , These represent the one-way propagation time of the ultrasonic wave from the excitation array element to the grid point and from the grid point to the receiving array element, respectively.

[0036] Based on the full matrix capture data, the Hilbert transform value at the corresponding propagation time for each grid point is obtained and used as the pixel value for that grid point. Pixel value at the corresponding propagation time for:

[0037]

[0038] In the formula, () represents the Hilbert transform of time-domain data. This represents the total number of array elements in the phased array probe. , These are the serial numbers of the transmitting and receiving elements in the phased array, respectively.

[0039] Each pixel value is arranged according to the spatial coordinate order of the grid points. The pixel gray values ​​of all grid points are normalized to eliminate the imaging deviation caused by the difference in signal amplitude. Finally, the discrete grid pixels are stitched together and integrated by the imaging software to generate an ultrasonic image of the nuclear power metal welded joint area. The image can clearly show the location, shape and size of the defect, providing visualization support for subsequent qualitative and quantitative analysis of the defect.

[0040] Furthermore, the propagation time includes: propagation time within the wedge, propagation time after refraction within the metal material, and propagation time after reflection from the bottom surface.

[0041] Furthermore, the specific steps for the delay multiplication and superposition process in step S6 are as follows:

[0042] For each transmitting element, considering the propagation time of each element after refraction by the glass wedge to each grid point, calculate the... DAS beamforming signal corresponding to each transmitting element The calculation formula is:

[0043]

[0044] For the The amplitude of the DAS beamforming signal of the first transmitting element is corrected to obtain the corrected signal of the second transmitting element. DAS beamforming signal of each transmitting element The formula for its calculation is:

[0045]

[0046] The DAS beamforming signals of all the corrected transmit elements are delayed, multiplied, and superimposed to obtain the DMAS processing result, calculated as follows:

[0047]

[0048] Furthermore, the formula for calculating the vector coherence factor in step S6 is as follows:

[0049]

[0050] In the formula, , Let represent the real and imaginary parts of the signal after the Hilbert transform, respectively. In equation (6) abbreviation;

[0051] The obtained DMAS processing results and Multiplying the values ​​yields the final image grayscale value for each grid point. The formula for its calculation is:

[0052]

[0053] The final image grayscale values ​​of all grid points are arranged according to their spatial coordinates. After optimization by imaging software, clearer and more reliable non-destructive testing imaging results of nuclear power metal welded joints are obtained. The location, shape and size of defects can be presented more accurately, providing more reliable visualization support for qualitative and quantitative analysis of defects.

[0054] On the other hand, the non-destructive testing method of the present invention is applied to the non-destructive testing of cracks, inclusions, and porosity defects inside the metal welded joints of nuclear power plants.

[0055] The present invention has the following beneficial effects:

[0056] (1) This invention makes full use of the multi-channel characteristics of full matrix capture data and effectively reduces the noise of the signal by time reversal operator decomposition technology. It can significantly suppress the strong scattering noise caused by the coarse grain structure of the weld, solve the problem that the defect echo is easily submerged by noise in the traditional imaging method, and improve the clarity of imaging and the reliability of defect identification.

[0057] (2) The present invention adopts an oblique incident ultrasonic detection strategy, which enables ultrasonic waves to penetrate the weld structure more effectively and irradiate the weld root and sidewall area, thereby improving the sensitivity to defects in key parts. This is an important supplement to the traditional normal incident detection method.

[0058] (3) This invention combines delayed multiplication and superposition with vector coherence factor for post-processing, which further enhances the defect reflection signal, suppresses artifacts, and obtains higher imaging contrast and more accurate defect boundary description, providing a highly reliable image basis for qualitative and quantitative analysis of weld defects in nuclear power equipment.

[0059] (4) This invention can provide an effective solution to the high noise characteristics of coarse-grained structures in nuclear power welds and has good engineering applicability. It is of great significance for improving the non-destructive testing level of key welded structures in nuclear power equipment. Attached Figure Description

[0060] Figure 1 This is a flowchart of the non-destructive testing process of the present invention.

[0061] Figure 2 This is a numerical calculation diagram of the acoustic scattering response of the defect in Example 1.

[0062] Figure 3 The signal diagram after using the time-reversal operator decomposition technique of the present invention in Example 1.

[0063] Figure 4 The ultrasonic array signal map is obtained using a conventional method in Example 1.

[0064] Figure 5 This is a two-dimensional imaging image of the defects of the porosity of the nuclear power metal welded joint obtained by using the present invention in Example 1.

[0065] Figure 6 This is a two-dimensional imaging image of a nuclear power plant metal welded joint crack obtained using the present invention in Example 1.

[0066] Figure 7 This is a two-dimensional imaging image of the defects of slag inclusions in the nuclear power metal welding joint obtained by using the present invention in Example 1.

[0067] Figure 8 This is a two-dimensional imaging image of the undamaged area in Example 1.

[0068] Figure 9 This is a two-dimensional imaging image of the defects of the nuclear power metal welded joint obtained by using the traditional full-focus imaging method in Example 1.

[0069] Figure 10 This is a two-dimensional imaging image of a crack in a nuclear power plant metal welded joint obtained using the traditional full-focus imaging method in Example 1.

[0070] Figure 11 This is a two-dimensional imaging image of the defects of slag inclusions in the nuclear power metal welded joint obtained by using the traditional full-focus imaging method in Example 1.

[0071] Figure 12 This is a radiographic scan image of the porosity defect obtained by radiographic testing in Example 1.

[0072] Figure 13 This is a radiographic scan image of a crack defect obtained using radiographic testing in Example 1.

[0073] Figure 14 This is a radiographic scan image of the inclusion defect obtained by radiographic testing in Example 1. Detailed Implementation

[0074] The technical solution of the present invention will be further described in detail below with reference to specific embodiments. However, these embodiments are not intended to limit the present invention. Any similar structures and similar variations of the present invention should be included in the protection scope of the present invention. The commas in the present invention all indicate the relationship between and. The English letters in the present invention are case-sensitive.

[0075] like Figure 1 As shown, this invention provides a non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging, the testing method comprising:

[0076] S1. Determine the density of the nuclear power metal material based on the properties of the tested material, and use a single ultrasonic longitudinal and transverse wave probe to excite ultrasonic longitudinal and transverse waves on the surface of the metal material. Calculate the ultrasonic longitudinal wave velocity and transverse wave velocity of the nuclear power metal material based on the thickness of the material and the bottom echo time.

[0077] Ultrasonic longitudinal wave velocity of nuclear power metal materials transverse wave velocity They are respectively:

[0078] In the formula, The thickness of the nuclear power metal material; and These are the bottom echo times received during ultrasonic longitudinal / transverse wave excitation.

[0079] S2. Assuming that an ultrasonic single probe is excited on an acrylic wedge and incident on the weld area at an oblique angle, the acoustic scattering response distribution of defects under different acrylic wedge angles is calculated based on the metal material density, longitudinal wave velocity, and transverse wave velocity. The optimal wedge angle is determined based on the spatial distribution of ultrasonic energy, i.e., the sound wave energy completely covers the area to be tested at this angle, and the energy is maximized as much as possible. Numerical simulation is used to construct the sound field distribution map under oblique incidence conditions to evaluate the sound beam coverage effect of different wedge angles on the weld root, sidewall fusion area, and excess height structure.

[0080] The formula for calculating the acoustic scattering response distribution is:

[0081]

[0082] In the formula, These represent the ultrasonic excitation and reception processes, respectively. These represent the refractive indices during the ultrasonic excitation and reception processes, respectively. These represent path-based ultrasonic attenuation during ultrasonic excitation and reception processes, respectively. These represent the phased array aperture coefficients for the ultrasonic excitation and reception processes, respectively. This represents the scattering coefficient when the ultrasonic wave encounters a scattering body during excitation and reception; the response at each point within the region is then calculated. The acoustic scattering response distribution map can then be obtained.

[0083] The calculation process considered multiple factors, including the refraction, reflection, and attenuation of ultrasonic waves in the weld area, as well as the influence of the array element directivity on ultrasonic energy propagation. This process calculated the sound velocity distribution and propagation path at different incident angles to obtain the acoustic response corresponding to each angle. Through these precise modeling and calculations, a more accurate acoustic scattering response of the defect can be obtained, providing a foundation for subsequent wedge design and phased array scanning.

[0084] The experimental setup of this invention employs a 64-element ultrasonic phased array system. This system has an excitation frequency of 5 MHz and a sampling rate of 19.2308 MHz, enabling precise detection of welded joints in nuclear power plants.

[0085] S3 utilizes an ultrasonic array inspection system to perform oblique-incident phased-array scanning of the nuclear power plant's metal welded joints at an optimal angle using an plexiglass wedge. This acquires ultrasonic phased-array scan images, with the ultrasonic signal obliquely incident onto the weld area through the glass wedge, thus addressing the inspection of nuclear power equipment under conditions of excess height. The accompanying software of the ultrasonic array inspection system identifies anomalous bright spots in the ultrasonic phased-array scan images, preliminarily determining these spots as suspected defect areas. Full-matrix capture data (FMC data) corresponding to these anomalous bright spots is saved. Full-matrix capture is a data acquisition method that controls each element in the phased-array probe as both a transmitter and receiver, thereby obtaining the most complete dataset containing information on all sound wave propagation paths. The full-matrix capture data includes key parameters such as the amplitude, propagation time, and phase of the ultrasonic signal, ensuring data integrity and traceability, and providing raw data support for subsequent signal processing and defect identification. The anomalous bright spots are locations where the computer software displays high-energy reflections in the B-scan or S-scan images during the phased-array scan.

[0086] S4. For the full-matrix capture data, a time-reversal operator decomposition algorithm is used for signal denoising to separate the characteristic echo from the scattered noise, obtaining the denoised ultrasonic array signal. The ultrasonic array signal is the set of time-domain signals from the transmitting and receiving channels of each element of the phased array. The time-reversal operator decomposition separates the noise signal from the effective signal through processing the full-matrix capture data, thereby improving signal quality. The time-reversal operator decomposition technique solves for the characteristic structure of the propagation matrix, achieving the concentration of signal energy and the dispersion of noise energy, thus effectively eliminating scattering interference caused by coarse-grained structures. The denoising process includes:

[0087] Signal decomposition: The original ultrasonic signal is subjected to time reversal transformation to obtain the back propagation characteristics of the target defect signal; through multiple iterations, the focus of the signal is enhanced and the influence of noise is reduced.

[0088] Noise reduction effect: Compared with traditional methods, the signal after decomposition using the time-reversal operator can significantly improve the clarity of the defect echo signal and significantly reduce the noise level.

[0089] The specific steps of step S4 are as follows:

[0090] The full-matrix captured data is transformed into frequency domain data using Fourier transform to obtain the frequency domain matrix. ;

[0091] For frequency domain matrix Perform singular value decomposition;

[0092]

[0093] in, Represents a left singular vector. Let represent a real diagonal matrix composed of positive singular values. Represents a right singular vector;

[0094] Further decomposition of the frequency domain matrix It is decomposed into the defect signal corresponding part and the noise signal corresponding part;

[0095]

[0096] in, For eigenvalues, and These are the singular vectors corresponding to the eigenvalues. For frequency The number of singular values ​​corresponding to the lower signal varies depending on the defect. The number of singular values ​​corresponding to the noise signal; The number of singular values ​​in the denoised signal; This is the number of singular values ​​in the decomposition, typically 64; It is a left singular vector; It is a right singular vector;

[0097] By extracting the defect signal corresponding to the left of the plus sign in the frequency domain matrix decomposition formula, and performing an inverse Fourier transform on this part to obtain the reconstructed time-domain full matrix capture number (FMC data), the denoised ultrasonic array signal can be obtained. This denoised signal can effectively improve the signal-to-noise ratio of the defect signal, clearly present the location, size and morphological characteristics of the defect, and solve the problems of fuzzy defect identification and misjudgment caused by noise interference in the original data, laying the foundation for subsequent qualitative and quantitative analysis of defects.

[0098] S5. The nuclear power plant metal welding joint area is divided into grids with a precision of 0.1mm. Taking into account the longitudinal and transverse wave velocities of the nuclear power plant metal, the ultrasonic propagation distance, and the influence of bottom surface reflection, the propagation time of the sound wave emitted by each element of the ultrasonic phased array after refraction by the glass wedge to each grid point is calculated. The propagation time includes: the propagation time within the wedge, the propagation time after refraction within the metal material, and the propagation time after bottom surface reflection. Then, based on the full matrix capture data, the Hilbert transform value of the propagation time corresponding to each grid point is obtained and used as the pixel value of that grid point to generate an ultrasonic imaging map of the area.

[0099] The specific steps are as follows:

[0100] The area of ​​the nuclear power plant metal welded joint was divided into grids with a precision of 0.1 mm, and a grid spatial coordinate model was established, with each grid point corresponding to a unique spatial coordinate. , This ensures that the grid completely covers the detection area without omissions or overlaps, providing a discretized spatial reference for subsequent point-by-point imaging.

[0101] Based on the propagation path of the ultrasonic transmitter and receiver, the propagation time of each element of the ultrasonic phased array after refraction by the glass wedge to each grid point is calculated. The calculation formula is as follows:

[0102]

[0103] In the formula, and These are the spatial coordinates of the excitation array element and the receiving array element, respectively. and These are the abscissas of the refraction points at the wedge-sample interface during ultrasonic excitation and reception, respectively. The speed of the ultrasonic wave within the plexiglass wedge. It is the speed of the ultrasonic wave in the weld base material; , () represents the coordinates of the p-th grid point within the imaging region; , These represent the one-way propagation time of the ultrasonic wave from the excitation array element to the grid point and from the grid point to the receiving array element, respectively.

[0104] Since the probe excites the ultrasonic beam at an oblique incidence through the glass wedge, this invention utilizes Snell's law to determine the refraction point of the ultrasonic beam at the interface between the glass wedge and the metal material in order to accurately calculate the propagation time of the ultrasound to each grid point. Based on this refraction point, the length of the refraction path is calculated. The refraction path consists of a propagation segment within the wedge and a refraction segment within the material, while also considering the bottom reflection path to cover critical structures such as the weld root and the sidewall fusion area. After obtaining the refraction path, the theoretical propagation time from the array element to each grid point is calculated using the propagation distance and wave velocity.

[0105] Based on the full matrix capture data, the Hilbert transform value at the corresponding propagation time for each grid point is obtained and used as the pixel value for that grid point. Pixel value at the corresponding propagation time for:

[0106]

[0107] In the formula, () represents the Hilbert transform of time-domain data. This represents the total number of array elements in the phased array probe. , These are the serial numbers of the transmitting and receiving elements in the phased array, respectively.

[0108] Each pixel value is arranged according to the spatial coordinate order of the grid points. The pixel gray values ​​of all grid points are normalized to eliminate the imaging deviation caused by the difference in signal amplitude. Finally, the discrete grid pixels are stitched together and integrated by the imaging software to generate an ultrasonic image of the nuclear power metal welded joint area. The image can clearly show the location, shape and size of the defect, providing visualization support for subsequent qualitative and quantitative analysis of the defect.

[0109] S6. Delayed Multiplication and Superposition (DMAS) and vector coherence factor denoising are applied to the imaging data to obtain clearer and more reliable non-destructive testing imaging results of the weld joint. The delayed multiplication and superposition method is used to enhance the coherent echo, so that the real defect signal is accumulated and strengthened in terms of energy. At the same time, the vector coherence factor is used to effectively suppress random noise and sidelobe interference from the coarse grain structure, making the defect boundary in the imaging area clearer and significantly reducing background noise, thereby obtaining a highly reliable two-dimensional weld imaging result.

[0110] The specific steps for performing delayed multiplication and superposition are as follows:

[0111] For each transmitting element, considering the propagation time of each element after refraction by the glass wedge to each grid point, calculate the... DAS beamforming signal corresponding to each transmitting element The calculation formula is:

[0112]

[0113] For the The amplitude of the DAS beamforming signal of the first transmitting element is corrected to obtain the corrected signal of the second transmitting element. DAS beamforming signal of each transmitting element The formula for its calculation is:

[0114]

[0115] The DAS beamforming signals of all the corrected transmit elements are delayed, multiplied, and superimposed to obtain the DMAS processing result, calculated as follows:

[0116]

[0117] The formula for calculating the vector coherence factor is as follows:

[0118]

[0119] In the formula, , Let represent the real and imaginary parts of the signal after the Hilbert transform, respectively. In equation (6) abbreviation;

[0120] The obtained DMAS processing results and Multiplying the values ​​yields the final image grayscale value for each grid point. The formula for its calculation is:

[0121]

[0122] The final image grayscale values ​​of all grid points are arranged according to their spatial coordinates. After optimization by imaging software, clearer and more reliable non-destructive testing imaging results of nuclear power metal welded joints are obtained. The location, shape and size of defects can be presented more accurately, providing more reliable visualization support for qualitative and quantitative analysis of defects.

[0123] This invention addresses the issues of strong scattering noise caused by coarse-grained microstructure in nuclear power plant weld seams, as well as the sensitivity of traditional imaging algorithms to noise and blurred defect boundaries. By performing signal decomposition and noise suppression on the full-matrix capture data, and combining the advantages of oblique incidence sound beams, it significantly improves the imaging contrast and reliability of internal weld defects. The core of this invention is reducing scattering noise and improving imaging clarity. It fully utilizes the information redundancy of phased array multi-channel data, introduces time-reversal operator decomposition technology to achieve signal denoising, and then uses delay multiplication superposition and vector coherence factor to enhance the true reflected echo, thereby obtaining high signal-to-noise ratio and high reliability imaging results of the internal weld structure.

[0124] Example 1

[0125] To verify the effectiveness of the method of this invention, this embodiment conducted experiments on the welded joints of nuclear power plant metal pipelines according to the detection method of this application. In the experiments, various typical defects were present inside the welded joints, such as porosity, cracks, and slag inclusions. The nuclear power plant metal pipeline material is 304 stainless steel. The specific detection method includes:

[0126] S1. Determine the density of the nuclear power metal pipe based on its properties, and use a single ultrasonic longitudinal and transverse wave probe to excite ultrasonic longitudinal and transverse waves on the surface of the metal material. Calculate the ultrasonic longitudinal wave velocity and transverse wave velocity of the nuclear power metal material based on the thickness of the 304 stainless steel and the bottom echo time.

[0127] Among them, the thickness of nuclear power metal materials The time of bottom echo received during ultrasonic longitudinal wave excitation The time of bottom echo received during ultrasonic transverse wave excitation Ultrasonic longitudinal wave velocity of 304 stainless steel transverse wave velocity They are respectively:

[0128]

[0129] Calculated using formula (1) , .

[0130] S2, an ultrasonic single probe is used to excite an ultrasonic wave on an acrylic wedge and incident obliquely into the weld area. The acoustic scattering response distribution of the defect is calculated based on the metal material density, longitudinal wave velocity, and transverse wave velocity at different acrylic wedge angles. The formula for calculating the acoustic scattering response distribution is:

[0131]

[0132] In the formula, These represent the ultrasonic excitation and reception processes, respectively. These represent the refractive indices during the ultrasonic excitation and reception processes, respectively. These represent path-based ultrasonic attenuation during ultrasonic excitation and reception processes, respectively. These represent the phased array aperture coefficients for the ultrasonic excitation and reception processes, respectively. This represents the scattering coefficient when the ultrasonic wave encounters a scattering body during excitation and reception; the response at each point within the region is then calculated. The acoustic scattering response distribution map can then be obtained, such as Figure 2 As shown.

[0133] The optimal wedge angle was determined to be 50 degrees based on the spatial distribution of ultrasonic energy.

[0134] During the experiment, a 5MHz 128-element ultrasonic probe (Imasonic, France) was used for excitation, and a Vantage 128-channel phased array system (Verasonics, USA) was used for data acquisition.

[0135] S3. Using an ultrasonic array inspection system, an oblique-incident phased array scan is performed on the nuclear power plant's metal welded joint at an optimal angle using an plexiglass wedge. Ultrasonic phased array scan images are acquired, with the ultrasonic signal obliquely incident onto the weld area through the glass wedge. This method is used to inspect nuclear power equipment under conditions of excess height. The accompanying software of the ultrasonic array inspection system identifies abnormal bright spots in the ultrasonic phased array scan images, preliminarily determining these abnormal bright spots as suspected defect areas. Full matrix capture data (FMC data) of the corresponding locations of these abnormal bright spots is saved. These abnormal bright spots are locations where the computer software displays high-energy reflections in the B-scan or S-scan images during the phased array scan.

[0136] S4, for the full matrix capture data, a time-reversal operator decomposition algorithm is used for signal denoising, resulting in the denoised ultrasonic array signal as follows: Figure 3 As shown, ultrasonic array signals are obtained using traditional methods, such as... Figure 4 As shown, through Figure 4 It can be seen that the amplitude range is large (approximately -150 to 150 relative units), the background noise is widely distributed and has a high amplitude, and a large amount of high-frequency clutter is mixed with the defect echo signal; the effective defect echo signal is submerged in noise, the boundaries between multiple peaks are blurred, and it is difficult to clearly distinguish the real defect reflection peak from the noise interference; a large amount of diffuse noise still exists at the signal tail (>1000 sampling points), which continues to interfere with subsequent imaging and defect identification; while Figure 3 The mid-amplitude range is effectively compressed (approximately -80 to 80 relative units), background noise is significantly suppressed, high-frequency clutter is greatly reduced, and the overall signal is smoother and cleaner; the defect echo peaks are more prominent and clear, with distinct boundaries between each peak, and the contrast between the effective defect signal and noise is significantly improved; the noise at the signal tail is greatly attenuated, retaining only a weak matrix scattering signal; the comparative results show that the time-reversal operator decomposition technique used in this invention can significantly improve the clarity of the defect echo signal, effectively suppressing noise while completely preserving defect features, laying the foundation for improving the sensitivity and reliability of defect detection in nuclear power metal welded joints.

[0137] The specific steps of step S4 are as follows:

[0138] The full-matrix captured data is transformed into frequency domain data using Fourier transform to obtain the frequency domain matrix. ;

[0139] For frequency domain matrix Perform singular value decomposition;

[0140]

[0141] in, Represents a left singular vector. Let represent a real diagonal matrix composed of positive singular values. Represents a right singular vector;

[0142] Further decomposition of the frequency domain matrix It is decomposed into the defect signal corresponding part and the noise signal corresponding part;

[0143]

[0144] in, For eigenvalues, and These are the singular vectors corresponding to the eigenvalues. For frequency The number of singular values ​​corresponding to the lower signal varies depending on the defect. The number of singular values ​​corresponding to the noise signal; The number of singular values ​​in the denoised signal; This is the number of singular values ​​in the decomposition, typically 64; It is a left singular vector; It is a right singular vector;

[0145] Extract the defect signal corresponding to the left of the plus sign in the frequency domain matrix decomposition formula, and perform an inverse Fourier transform on this part to obtain the reconstructed time-domain full matrix capture number (FMC data). The denoised ultrasonic array signal can then be obtained. Some of the denoised signals are shown in Table 1.

[0146] Table 1

[0147]

[0148] S5. The nuclear power metal welded joint area is divided into grids with an accuracy of 0.1mm. Combining the longitudinal and transverse wave velocities of the nuclear power metal, the ultrasonic propagation distance, and the influence of bottom surface reflection, the propagation time of the sound wave emitted by each element of the ultrasonic phased array after being refracted by the glass wedge to reach each grid point is calculated. Then, based on the full matrix capture data, the Hilbert transform value of the propagation time corresponding to each grid point is obtained and used as the pixel value of the grid point to generate an ultrasonic imaging map of the area.

[0149] The specific steps are as follows:

[0150] The area of ​​the nuclear power plant metal welded joint was divided into grids with a precision of 0.1 mm, and a grid spatial coordinate model was established, with each grid point corresponding to a unique spatial coordinate. , This ensures that the grid completely covers the detection area without omissions or overlaps, providing a discretized spatial reference for subsequent point-by-point imaging.

[0151] Based on the propagation path of the ultrasonic transmitter and receiver, the propagation time of each element of the ultrasonic phased array after refraction by the glass wedge to each grid point is calculated. The calculation formula is as follows:

[0152]

[0153] In the formula, and These are the spatial coordinates of the excitation array element and the receiving array element, respectively. and These are the abscissas of the refraction points at the wedge-sample interface during ultrasonic excitation and reception, respectively. The speed of the ultrasonic wave within the plexiglass wedge. It is the speed of the ultrasonic wave in the weld base material; , () represents the coordinates of the p-th grid point within the imaging region; , These represent the one-way propagation time of the ultrasonic wave from the excitation array element to the grid point and from the grid point to the receiving array element, respectively.

[0154] Since the probe excites the ultrasonic beam at an oblique incidence through the glass wedge, in order to accurately determine the propagation time of the ultrasonic beam to each grid point, this invention uses Snell's law to determine the refraction point of the ultrasonic beam at the interface between the glass wedge and the metal material, and calculates the refraction path length based on the refraction point.

[0155] Based on the full matrix capture data, the Hilbert transform value at the corresponding propagation time for each grid point is obtained and used as the pixel value for that grid point. Pixel value at the corresponding propagation time for:

[0156]

[0157] In the formula, () represents the Hilbert transform of time-domain data. This represents the total number of array elements in the phased array probe. , These are the serial numbers of the transmitting and receiving elements in the phased array, respectively.

[0158] Each pixel value is arranged according to the spatial coordinate order of the grid points. The pixel gray values ​​of all grid points are normalized to eliminate the imaging deviation caused by the difference in signal amplitude. Finally, Matlab2021a is used to stitch together the discrete grid pixels to generate an ultrasonic image of the nuclear power metal welded joint area. The image can clearly show the location, shape and size of the defect, providing visualization support for subsequent qualitative and quantitative analysis of the defect.

[0159] S6. Delayed Multiplication and Stacking (DMAS) and vector coherence factor denoising are performed on the imaging data to obtain clearer and more reliable non-destructive testing imaging results of the weld joint, such as... Figures 5-7 As shown;

[0160] The specific steps for performing delayed multiplication and superposition are as follows:

[0161] For each transmitting element, considering the propagation time of each element after refraction by the glass wedge to each grid point, calculate the... DAS beamforming signal corresponding to each transmitting element The calculation formula is:

[0162]

[0163] For the The amplitude of the DAS beamforming signal of the first transmitting element is corrected to obtain the corrected signal of the second transmitting element. DAS beamforming signal of each transmitting element The formula for its calculation is:

[0164]

[0165] The DAS beamforming signals of all the corrected transmit elements are delayed, multiplied, and superimposed to obtain the DMAS processing result, calculated as follows:

[0166]

[0167] The formula for calculating the vector coherence factor is as follows:

[0168]

[0169] In the formula, , Let represent the real and imaginary parts of the signal after the Hilbert transform, respectively. In equation (6) abbreviation;

[0170] The obtained DMAS processing results and Multiplying the values ​​yields the final image grayscale value for each grid point. The formula for its calculation is:

[0171]

[0172] The final image grayscale values ​​of all grid points are arranged according to their spatial coordinates. After optimization by imaging software, the non-destructive testing imaging results of the nuclear power plant metal welded joint are obtained.

[0173] To demonstrate that this method can effectively detect three defects—porosity, cracks, and slag inclusions—in welded joints of nuclear power plant metal pipelines, and... Figure 8 Compare the ultrasonic images of the defect-free areas shown, from Figure 8 The image shows a low background noise level, dominated by low-amplitude blue areas, with only a small amount of weak scattering signals (light blue / cyan textures) caused by material structure or system noise. There are no obvious high-amplitude abnormally bright areas, and the overall imaging is uniform and stable. Figure 5 As can be seen from the coordinates, approximately , At this location, an isolated, approximately elliptical, high-amplitude bright region (yellow / orange) appears, with an amplitude significantly higher than the background noise. Its compact shape and clear boundaries are consistent with the acoustic scattering characteristics of spherical defects with porosity. From Figure 6 As can be seen from the coordinates, approximately , At this location, a long, extended, high-amplitude anomaly region is observed, with the signal extending along a certain direction and exhibiting significant amplitude gradient changes, consistent with the acoustic reflection characteristics of crack-like surface defects; from Figure 7 As can be seen from the coordinates, approximately , At this location, there are irregularly shaped, concentrated bright areas with high signal intensity and relatively diffuse distribution, consistent with the acoustic scattering characteristics of non-metallic inclusions such as slag inclusions. Comparative results show that this invention can effectively suppress background noise, clearly identify three typical defects in nuclear power plant metal welded joints: porosity, cracks, and slag inclusions, and achieve preliminary defect type identification through differences in imaging morphology, demonstrating high detection sensitivity and reliability.

[0174] like Figures 9-11 As shown, the imaging results of pores, cracks, and inclusions obtained using traditional total focusing imaging methods are presented respectively. Figure 9 and Figure 5 The comparison shows that, Figure 5 median coordinate , A clearly defined, isolated elliptical area with high amplitude and high brightness was detected, indicating a concentrated defect signal; Figure 9 The stomata signal at the same location was obscured by noise, resulting in blurred boundaries and poor recognizability; through Figure 10 and Figure 6 The comparison shows that, Figure 6 median coordinate , A long, elongated, high-amplitude anomaly region was detected, which can visually reflect the surface distribution characteristics of the crack; Figure 10 Crack signals at the same location are distorted in shape and attenuated in amplitude, making them difficult to identify; through Figure 11 and Figure 7 A comparison shows that, Figure 7median coordinate , Irregularly shaped but concentrated bright areas were detected, reflecting the volumetric scattering characteristics of inclusions; Figure 11 The slag inclusion signals at the same location are diffuse and mixed, resulting in insufficient defect identification. Quantitative tests show that the method of this invention improves the defect echo signal by about 20dB compared with the traditional full-focusing method, significantly improving the detection signal-to-noise ratio. It can achieve accurate location and type differentiation of typical defects in nuclear power metal welded joints, with better detection sensitivity and reliability.

[0175] In addition, X-ray inspection was used to verify three typical defects—porosity, cracks, and slag inclusions—in the welded joints of the nuclear power plant's metal pipes. Figure 12 As can be seen from the X-ray inspection images, there are no obvious local anomalies in the weld area, only uniform grayscale changes, which are consistent with the location and characteristics of the tiny volumetric pores detected by the ultrasonic phased array; from Figure 13 As can be seen from the X-ray inspection image, a thin, elongated black line-like abnormal image appears in the middle of the weld, the morphology of which is completely consistent with the elongated crack detected by the ultrasonic phased array, verifying the surface distribution characteristics of the crack; from Figure 14 As can be seen from the X-ray inspection image, there are irregular dark areas in the weld area, which highly match the location and range of irregular volumetric slag inclusions detected by the ultrasonic phased array. The comparison results show that the detection results of the ultrasonic phased array method of the present invention are completely consistent with the gold standard of X-ray inspection, proving the effectiveness of the noise reduction method and oblique incidence imaging strategy proposed in the present invention in improving the detection accuracy of weld joint defects. It also meets the rapid detection requirements in the complex environment of nuclear power equipment, can accurately locate and identify typical defects in nuclear power metal weld joints, and has the advantages of being radiation-free and capable of depth imaging. It is suitable for efficient and safe non-destructive testing at nuclear power sites.

[0176] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

Claims

1. A non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging, characterized in that, The detection method includes: S1. Determine the density of the nuclear power metal material based on the properties of the tested material, and use a single ultrasonic longitudinal and transverse wave probe to excite ultrasonic longitudinal and transverse waves on the surface of the metal material. Calculate the ultrasonic longitudinal wave velocity and transverse wave velocity of the nuclear power metal material based on the thickness of the material and the bottom echo time. S2, assuming that an ultrasonic single probe is excited on the plexiglass wedge and incident on the weld area in an oblique manner, calculate the defect acoustic scattering response distribution under different plexiglass wedge angles based on the metal material density, longitudinal wave velocity and transverse wave velocity, and determine the optimal wedge angle based on the spatial distribution of ultrasonic energy. S3. Using an ultrasonic array detection system, an oblique incidence phased array scan is performed on the nuclear power metal welding joint at the optimal angle using an plexiglass wedge to obtain ultrasonic phased array scan images. The software supporting the ultrasonic array detection system identifies the abnormal bright spots in the ultrasonic phased array scan images and saves the full matrix capture data corresponding to the abnormal bright spots. S4. For the full matrix capture data, the time reversal operator decomposition algorithm is used to perform signal denoising processing to obtain the denoised ultrasonic array signal. S5. Divide the nuclear power metal welded joint area into grids. Combine the longitudinal and transverse wave velocities of the nuclear power metal, the ultrasonic propagation distance, and the influence of bottom surface reflection. Calculate the propagation time of the sound waves emitted by each element of the ultrasonic phased array after refraction by the glass wedge to each grid point. Then, based on the full matrix capture data, obtain the Hilbert transform value of the propagation time corresponding to each grid point and use it as the pixel value of that grid point to generate an ultrasonic imaging map of the area. S6 performs delay multiplication and superposition on the imaging data and vector coherence factor noise reduction to obtain clearer and more reliable non-destructive testing imaging results of welded joints.

2. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging according to claim 1, characterized in that, In step S1, the ultrasonic longitudinal wave velocity of the nuclear power metal material transverse wave velocity They are respectively: In the formula, The thickness of the nuclear power metal material; and These are the bottom echo times received during ultrasonic longitudinal / transverse wave excitation.

3. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging according to claim 1, characterized in that, The formula for calculating the acoustic scattering response distribution in step S2 is: In the formula, These represent the ultrasonic excitation and reception processes, respectively. These represent the refractive indices during the ultrasonic excitation and reception processes, respectively. These represent path-based ultrasonic attenuation during ultrasonic excitation and reception processes, respectively. These represent the phased array aperture coefficients for the ultrasonic excitation and reception processes, respectively. This represents the scattering coefficient when the ultrasonic wave encounters a scattering body during excitation and reception; the response at each point within the region is then calculated. The acoustic scattering response distribution map can then be obtained.

4. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging according to claim 1, characterized in that, The abnormal bright spot locations mentioned in step S3 are the positions where high-energy reflections are observed in the B-scan or S-scan images during phased array scanning by computer software.

5. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging according to claim 1, characterized in that, The specific steps of step S4 are as follows: The full-matrix captured data is transformed into frequency domain data using Fourier transform to obtain the frequency domain matrix. ; For frequency domain matrix Perform singular value decomposition; in, Represents a left singular vector. Let represent a real diagonal matrix composed of positive singular values. Represents a right singular vector; Further decomposition of the frequency domain matrix It is decomposed into the defect signal corresponding part and the noise signal corresponding part; in, For eigenvalues, For frequency The number of singular values ​​corresponding to the lower signal. The number of singular values ​​corresponding to the noise signal; The number of singular values ​​in the denoised signal; The number of singular values ​​in the decomposition; It is a left singular vector; It is a right singular vector; Extract the defect signal corresponding to the left of the plus sign in the frequency domain matrix decomposition formula, and perform an inverse Fourier transform on this part to obtain the reconstructed time-domain full matrix capture number, thus obtaining the denoised ultrasonic array signal.

6. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging according to claim 1, characterized in that, The specific steps of step S5 are as follows: The area of ​​the nuclear power plant metal welded joint was divided into grids with a precision of 0.1 mm, and a grid spatial coordinate model was established, with each grid point corresponding to a unique spatial coordinate. , ); Based on the propagation path of the ultrasonic transmitter and receiver, the propagation time of each element of the ultrasonic phased array after refraction by the glass wedge to each grid point is calculated. The calculation formula is as follows: In the formula, and These are the spatial coordinates of the excitation array element and the receiving array element, respectively. and These are the abscissas of the refraction points at the wedge-sample interface during ultrasonic excitation and reception, respectively. The speed of the ultrasonic wave within the plexiglass wedge. It is the speed of the ultrasonic wave in the weld base material; , () represents the coordinates of the p-th grid point within the imaging region; , These represent the one-way propagation time of the ultrasonic wave from the excitation array element to the grid point and from the grid point to the receiving array element, respectively. Based on the full matrix capture data, the Hilbert transform value at the corresponding propagation time for each grid point is obtained and used as the pixel value for that grid point. Pixel value at the corresponding propagation time for: In the formula, () represents the Hilbert transform of time-domain data. This represents the total number of array elements in the phased array probe. , These are the serial numbers of the transmitting and receiving elements in the phased array, respectively. Each pixel value is arranged according to the spatial coordinate order of the grid points. The pixel gray values ​​of all grid points are normalized to eliminate imaging deviations caused by differences in signal amplitude. Finally, the discrete grid pixels are stitched together and integrated by imaging software to generate an ultrasonic image of the nuclear power plant metal welded joint area.

7. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging according to claim 6, characterized in that, The propagation time includes: propagation time within the wedge, propagation time after refraction within the metal material, and propagation time after reflection from the bottom surface.

8. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging according to claim 6, characterized in that, The specific steps for the delay multiplication and superposition process in step S6 are as follows: For each transmitting element, considering the propagation time of each element after refraction by the glass wedge to each grid point, calculate the... DAS beamforming signal corresponding to each transmitting element The calculation formula is: For the The amplitude of the DAS beamforming signal of the first transmitting element is corrected to obtain the corrected signal of the second transmitting element. DAS beamforming signal of each transmitting element The formula for its calculation is: The DAS beamforming signals of all the corrected transmit elements are delayed, multiplied, and superimposed to obtain the DMAS processing result, calculated as follows:

9. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging according to claim 8, characterized in that, The formula for calculating the vector coherence factor in step S6 is: In the formula, , Let these represent the real and imaginary parts of the signal after the Hilbert transform, respectively. In equation (6) abbreviation; The obtained DMAS processing results and Multiplying the values ​​yields the final image grayscale value for each grid point. The formula for its calculation is:

10. The non-destructive testing method for nuclear power plant metal welded joints based on ultrasonic phased array imaging as described in any one of claims 1-9 is applied to the non-destructive testing of cracks, inclusions, and porosity defects inside the metal welded joints of nuclear power plants.

Images