CN117007681A

CN117007681A - Ultrasonic flaw detection method and system

Info

Publication number: CN117007681A
Application number: CN202311261987.8A
Authority: CN
Inventors: 李世健; 刘�文; 石何飞
Original assignee: Suzhou Youzhi Laser Intelligent Technology Co ltd
Current assignee: Suzhou Youzhi Laser Intelligent Technology Co ltd
Priority date: 2023-09-27
Filing date: 2023-09-27
Publication date: 2023-11-07
Anticipated expiration: 2043-09-27
Also published as: CN117007681B

Abstract

The invention relates to the technical field of ultrasonic detection, in particular to an ultrasonic flaw detection method and an ultrasonic flaw detection system. The method comprises the following steps: transmitting ultrasonic signals to the interior of a material to be detected by utilizing an ultrasonic transmitter, and receiving the signals by utilizing an ultrasonic receiver so as to acquire ultrasonic signal data of the material; extracting dynamic data of the ultrasonic data of the material to obtain the ultrasonic dynamic data of the material; performing significant feature extraction and potential hidden danger feature extraction on the ultrasonic dynamic data of the material to obtain significant feature data and potential hidden danger feature data respectively; performing defect position determination processing on the remarkable characteristic data to obtain material defect position data, and performing defect degree determination processing on the potential hidden danger characteristic data to obtain material defect degree data; and generating material ultrasonic flaw detection report data according to the material defect position data and the material defect degree data. The invention can find finer defects through ultrasonic detection.

Description

Ultrasonic flaw detection method and system

Technical Field

The invention relates to the technical field of ultrasonic detection, in particular to an ultrasonic flaw detection method and an ultrasonic flaw detection system.

Background

The ultrasonic flaw detection method is a nondestructive detection technology for detecting the problems of internal defects, foreign matters, cracks and structural changes of materials by applying an ultrasonic technology. The method comprises the steps of transmitting ultrasonic signals in a material to be detected, and receiving signals reflected by internal defects or interfaces of the material, so that the characteristics of the signals are analyzed to judge the health state and defect condition of the material. Ultrasonic flaw detection methods still rely on experienced operators during data acquisition, signal analysis and result interpretation, which results in subjectivity and inconsistency of results.

Disclosure of Invention

The application provides an ultrasonic flaw detection method and an ultrasonic flaw detection system for solving at least one technical problem.

The application provides an ultrasonic flaw detection method, which comprises the following steps:

step S1: transmitting ultrasonic signals to the interior of a material to be detected by utilizing an ultrasonic transmitter, and receiving the signals by utilizing an ultrasonic receiver so as to acquire ultrasonic signal data of the material;

step S2: extracting dynamic data of the ultrasonic data of the material, thereby obtaining the ultrasonic dynamic data of the material;

step S3: performing significant feature extraction and potential hidden danger feature extraction on the ultrasonic dynamic data of the material, thereby respectively obtaining significant feature data and potential hidden danger feature data;

Step S4: performing defect position determination processing on the remarkable characteristic data to obtain material defect position data, and performing defect degree determination processing on the potential hidden danger characteristic data to obtain material defect degree data;

step S5: and generating material ultrasonic flaw detection report data according to the material defect position data and the material defect degree data.

According to the invention, through dynamic data extraction and significant feature extraction, abnormal signals and potential hidden danger in the material can be more accurately captured, so that the detection accuracy and reliability of defects are enhanced. Not only ultrasonic signal data of the material are extracted, but also deep mining of internal information of the material is realized through analysis of dynamic data, obvious characteristics and potential hidden danger characteristics, so that finer defects can be found. By the defect position determination and the defect degree determination, the position and degree of the defect can be quantitatively evaluated. The invention is finer and deeper in data processing and feature extraction, and can capture more useful information. Meanwhile, the burden of operators is reduced by automatically generating reports. In whole, the invention can improve the accuracy, efficiency and reliability of detection, thereby having remarkable superiority in the field of ultrasonic flaw detection of materials.

Preferably, step S1 is specifically:

step S11: acquiring material parameter data;

step S12: constructing a material model according to the material parameter data, thereby obtaining a three-dimensional model of the detection material;

step S13: acquiring material detection demand data;

step S14: generating test points of the three-dimensional model data of the detection material according to the material detection demand data, thereby obtaining first detection material test point data;

step S15: performing optimized grid division on the three-dimensional model of the detection material according to the material parameter data so as to obtain a gridding model of the detection material;

step S16: performing ultrasonic propagation simulation according to the data of the first detection material test point and the detection material gridding model, so as to obtain ultrasonic propagation simulation data;

step S17: generating material ultrasonic simulation detection data according to the ultrasonic propagation simulation data, and optimizing the first detection material test point data according to the material ultrasonic simulation detection data so as to acquire second detection material test point data;

step S18: and transmitting ultrasonic signals to the interior of the material to be detected through the second material detection point data by utilizing the ultrasonic transmitter, and receiving the signals through the ultrasonic receiver so as to acquire material ultrasonic signal data.

According to the invention, through the step S11 and the step S12, the material parameter data is adopted to construct the material model, so that the characteristics of the material to be detected can be more accurately described, and a more accurate ultrasonic propagation simulation and analysis basis is provided. Step S13 can acquire material detection demand data, and the material detection demand data is customized according to actual demands, so that the material detection demand data is better suitable for different scenes and purposes. Step S14 is to generate test points of the three-dimensional model data of the detection material, so that more optimized point location layout can be realized under the condition of limited number of test points, and more comprehensive data can be obtained. And the steps S15 to S18 realize ultrasonic propagation simulation, fully combine simulation and actual measurement data, optimize through ultrasonic simulation detection data, and improve the efficiency and accuracy of simulation calculation.

Preferably, step S12 is specifically:

matching according to the material parameter data and preset material acoustic characteristic data to obtain material parameter acoustic characteristic data;

and constructing a three-dimensional material model of the material parameter data according to the material parameter acoustic characteristic data, thereby obtaining a three-dimensional model of the detection material.

In the invention, the step S12 fully utilizes the material parameter data and the preset material acoustic characteristic data, and personalized material parameter acoustic characteristic data is obtained through matching. Thus, the constructed three-dimensional material model can reflect the acoustic characteristics of the actual material more accurately, thereby improving the detection accuracy. Through preset acoustic characteristic data, corresponding model construction schemes can be customized for each material according to different material types and performance requirements, so that higher customization is realized. By utilizing the acoustic characteristic data of the material parameters obtained by matching, the three-dimensional material model is constructed, and the propagation of sound waves in the material can be more accurately simulated, so that the model is more accurate. The personalized model construction mode can reduce errors and uncertainty caused by inaccurate acoustic characteristics of materials, and improves the reliability of detection data. The procedure of step S12 is suitable for different types of materials, including various acoustic properties, so that it can be widely applied to detection requirements of different fields and material types.

Preferably, the material detection requirement data includes detection area data and detection precision requirement data, and step S14 specifically includes:

step S141: carrying out region division on the three-dimensional model data of the detection material according to the detection region data so as to obtain detection region division data;

step S142: generating test points of the detection area division data according to the detection precision requirement data and the material parameter data, thereby obtaining detection area test point data;

step S143: screening test nodes on the detection area division data so as to obtain test point candidate data;

step S144: and carrying out batch random selection and maximum relative distance selection on the candidate data of the test points according to the test electric data of the test area, thereby obtaining the data of the test points of the first detection material.

In the first substep of step S14 in the present invention, the three-dimensional model data of the detection material is divided into regions according to the detection region data, so that the complex material model can be divided into smaller regions, and the detection process is more refined. And a second sub-step of step S14, combining the detection precision requirement data and the material parameter data to generate test point data. This means that an adaptive test point layout can be generated according to different detection requirements and material characteristics, and the effectiveness and accuracy of the data are improved. The third substep of step S14 screens candidate data of test points, so that unnecessary points can be eliminated, the test data is more concentrated in a key area, and the importance and reliability of the data are improved. The fourth sub-step of step S14 can ensure more uniform distribution of test points and larger coverage in relative distance by selecting test points randomly in batches and selecting the largest relative distance. The whole flow of the step S14 can effectively select test points, so that valuable data can be acquired to the greatest extent in a detection area, and the detection efficiency is improved.

Preferably, the material parameter data includes material density data, material elastic modulus data and material acoustic wave propagation speed data, and the step of optimizing mesh division in step S15 specifically includes:

step S151: performing initial grid division on the three-dimensional model of the detection material so as to obtain initial grid data;

step S152: grid quality evaluation is carried out on the initial grid data according to the material density data, so that grid quality evaluation data are obtained;

step S153: global grid division optimization is carried out on global grid data according to grid quality evaluation data and material elastic modulus data, so that global optimization grid data are obtained;

step S154: and carrying out minimum refraction area grid optimization on the global optimization grid data by using the material acoustic wave propagation speed data, thereby obtaining a detection material grid model.

In the first substep of step S15 in the present invention, that is, performing initial meshing on the three-dimensional model of the detection material, the model can be decomposed into smaller mesh units, and a foundation is provided for subsequent optimization. And in the second sub-step of the step S15, grid quality evaluation is carried out on the initial grid data through the material density data, so that the quality of the grids can be ensured to meet the requirements, and the influence of bad grids on the detection precision is avoided. And the third sub-step of the step S15 is to combine the grid quality evaluation data and the material elastic modulus data to optimize the global grid data, so that the overall structure of the grid can be optimized while the model detail is kept, and the calculation efficiency is improved. And a fourth sub-step of step S15, performing grid optimization of the minimum refraction area on the global optimization grid data according to the acoustic wave propagation characteristics of the material through the acoustic wave propagation speed data of the material, so that grid layout in the acoustic wave propagation simulation process is more accurate.

Preferably, in step S16, the ultrasonic propagation simulation performs simulation calculation by using an ultrasonic propagation simulation calculation formula, where the ultrasonic propagation simulation calculation formula is specifically:

；

a (t, x) is the ultrasonic amplitude data at time t and position x, P ₀ The ultrasonic initial pressure data comprise rho, c, x, gamma, omega, t, k, and k, wherein rho is density data of a material to be detected, c is ultrasonic propagation speed, x is ultrasonic position data, gamma is bulk modulus data of the material, omega is ultrasonic angular frequency data, t is ultrasonic time dataAcoustic wave number data.

The invention constructs an ultrasonic wave propagation simulation calculation formula which can simulate the propagation process of ultrasonic waves in the material to be detected and provides ultrasonic wave amplitude data at different times and positions, thereby realizing the simulation of the propagation process of the ultrasonic waves. By calculating the ultrasonic amplitude data in the formula, whether the defect exists in the material to be detected can be judged, and the defect can interfere the propagation of ultrasonic waves, so that the amplitude change is influenced. Representing the sound wave amplitude at different times and locations. The a (t, x) calculated will reveal the amplitude variation of the ultrasonic wave as it propagates in the material to be examined. P (P) ₀ Representing the initial pressure of the sound wave as it is emitted. It is the initial energy of the sound wave. By adjusting P ₀ The propagation of sound waves under different initial conditions can be simulated. The material density data ρ to be detected affects the propagation speed and the energy dissipation of the sound wave in the material. The density differences of the different materials will produce different acoustic responses in the simulation. The ultrasonic wave propagation velocity c represents the propagation velocity of an acoustic wave in a material. The physical properties of materials can affect the speed of sound, and the speed of sound propagation can vary from material to material and from frequency to frequency. The ultrasonic position data x indicates a certain position in the material. By simulating acoustic propagation at different locations, acoustic responses at different locations are obtained for defects detected. Bulk modulus data gamma for a material, representing the stiffness and elasticity of the material, describes the response of the material to external stresses, gamma affecting the frequency and amplitude of sound waves. Ultrasonic angular frequency data ω represents the frequency of the acoustic wave. The frequency determines the vibration period of the sound waves, which will interact in different ways with structures and defects in the material. The invention provides a detailed understanding of the propagation of ultrasonic waves in materials through simulation calculations.

Preferably, step S2 is specifically:

step S21: signal segmentation is carried out on ultrasonic data of the material, so that ultrasonic signal segmentation data are obtained;

step S22: calculating the instantaneous frequency of the ultrasonic signal segment data so as to obtain instantaneous frequency data;

step S23: extracting pulse characteristics from the ultrasonic signal segment data so as to obtain pulse characteristic data;

step S24: extracting time-frequency characteristic data from the ultrasonic signal segment data so as to obtain time-frequency characteristic representation data;

step S25: and constructing a data graph by using the instantaneous frequency data, the pulse characteristic data, the time frequency characteristic representation data and the corresponding ultrasonic signal segmentation data, so as to obtain the ultrasonic dynamic data of the material.

According to the invention, by carrying out signal segmentation on the ultrasonic data of the material, the complex ultrasonic signal can be segmented into a series of relatively short paragraphs, so that finer analysis can be carried out on different paragraphs. By calculating the instantaneous frequency of the ultrasound signal segment data, the frequency variation of the signal over time can be known, revealing structural features and variations within the material. The pulse characteristics are extracted from the ultrasonic signal segmentation data, so that the characteristics of the intensity and the width of the pulse in the ultrasonic signal can be captured, and the analysis of the defect condition inside the material is facilitated. The ultrasonic signal segmentation data is subjected to time-frequency characteristic data extraction, so that information of the signal in a frequency domain and a time domain can be obtained, and the structure and the characteristics of the inside of the material can be better understood.

Preferably, step S3 is specifically:

step S31: performing frequency spectrum conversion on the ultrasonic dynamic data of the material so as to obtain frequency representation data;

step S32: performing wavelet packet decomposition on the frequency representation data to obtain wavelet packet decomposition sub-spectrum data;

step S33: sub-spectrum energy calculation is carried out on sub-spectrum data decomposed by the wavelet packet, so that sub-spectrum energy data are obtained;

step S34: extracting energy characteristics of the sub-spectrum energy data so as to obtain energy characteristic data;

step S35: carrying out statistical feature extraction and nonlinear feature extraction on the energy feature data so as to obtain statistical feature data and nonlinear feature data respectively;

step S36: performing salient data integration according to the statistical characteristic data and the nonlinear characteristic data, thereby obtaining salient characteristic data;

step S37: extracting extremum data from the energy characteristic data so as to obtain potential hidden danger characteristic data;

the sub-spectrum energy calculation is calculated through a sub-spectrum energy calculation formula, and the sub-spectrum energy calculation formula specifically comprises:

；

E _sub (f, t) is the frequency range [ f, f+Δf]In the time range [ t, t+Δt ]]The sub-spectrum energy in the frequency spectrum, f is initial frequency variable data, delta f is frequency range width data, t is initial time variable data, delta t is time range width data, j is imaginary unit data, Time domain signal data which is ultrasonic dynamic data of material, < >>For the frequency range integration data,for the time-range integral data, e is the natural exponential direction, pi is the periodic data, ++>D is a differential sign, which is a material impulse response function.

The invention constructs a sub-spectrum energy calculation formula whose integration operation is such that the energy calculation is directed to a specific frequency range [ f, f+Δf ]]And a time range [ t, t+Δt ]]So that the frequency characteristics of the signal in this local range can be captured more accurately. The formula provides information on the energy distribution of the signal in a specific frequency band by performing an integration operation in the time domain and frequency domain transforming the time domain signal to calculate the energy in the specified frequency range. In the formulaRepresenting the material impulse response function, which takes into account the influence of the material response during propagation of the signal, making the calculation closer to reality. By combining integration and frequency domain transformation, the formula links the time and frequency characteristics together, and can capture the characteristic change of the signal at different time and frequency. The spectral energy of the signal within the local time window is analyzed so that local features of the signal, such as variations in instantaneous frequency, can be more finely analyzed. The sub-spectrum energy calculation formula fully considers the characteristics of frequency and time dimensions, can more accurately capture the characteristics of signals in different frequency and time windows through local analysis, provides a powerful tool for acoustic signal analysis of materials, and is helpful for detecting anomalies or potential defects in the materials.

According to the invention, through carrying out frequency spectrum conversion on the ultrasonic dynamic data of the material, the signal can be converted from a time domain to a frequency domain, so that the propagation conditions of different frequency components in the material can be better understood. The wavelet packet decomposition can decompose the spectrum data into sub-spectrum data with different scales and frequencies, helps capture the characteristics of different frequency components in the signal, and helps analyze the internal structure of the material. The energy calculation is carried out on the wavelet packet decomposition sub-spectrum data, so that the energy distribution situation of different frequency components in the signal can be quantified, and the recognition of abnormal or specific signal modes is facilitated. Energy characteristics are extracted from the sub-spectrum energy data, so that the energy distribution condition of signals can be captured, and further richer characteristic information is provided for subsequent analysis. The extraction of the statistical features and the nonlinear features of the energy feature data can more fully describe the characteristics of the signals, so that different material states or defect conditions can be distinguished more accurately. The statistical characteristic data and the nonlinear characteristic data are integrated, so that characteristics in different aspects can be comprehensively considered, and the distinguishing capability of the characteristics is improved. Extremum data is extracted from the energy characteristic data, and potential hidden danger or abnormal situations are found.

Preferably, step S4 is specifically:

step S41: acquiring image data of a material to be detected;

step S42: performing anomaly identification on the salient feature data so as to obtain salient anomaly identification data;

step S43: coordinate labeling is carried out on the image data of the material to be detected according to the detection point data corresponding to the obvious abnormal identification data, so that defect position coordinate data are obtained;

step S44: performing defect influence area assessment according to the significant anomaly identification data so as to obtain defect influence area assessment data;

step S45: performing defect position data integration according to the defect position coordinate data and the defect influence area evaluation data, so as to obtain material defect position data;

step S46: performing defect type determination processing according to the significant characteristic data and the potential hidden trouble characteristic data, thereby obtaining defect type data;

step S47: performing defect degree evaluation on the potential defect characteristic data so as to obtain defect degree evaluation data;

step S48: and integrating the defect degree data according to the defect type data and the defect degree evaluation data, so as to obtain the material defect degree data.

By combining the significant feature data with the anomaly identification, the method can effectively identify and locate significant anomalies in the material, thereby determining the defects. By coordinate-labeling the material image data with respect to the detection point data corresponding to the significant anomaly identification data, the defect position in the material can be accurately located. By evaluating the defect impact area, the impact degree of the defect on the surrounding area can be analyzed, and the severity of the defect can be further determined in an auxiliary manner. By integrating defect location data and defect impact area assessment data, the location and impact of defects can be described and quantified in multiple dimensions, providing detailed information for further analysis. Based on the significant characteristic data and the potential hidden trouble characteristic data, the defect types in the material can be accurately judged, and further processing and analysis are facilitated. By combining the defect type data and the defect degree evaluation data, the severity degree of the defect can be comprehensively evaluated, and a basis is provided for decision making. By introducing the significant features and the objective data of the abnormal identification, the influence of the artificial subjective judgment on the result can be reduced, and the objectivity and accuracy of the judgment can be improved.

Preferably, the present application also provides an ultrasonic flaw detection system for performing the ultrasonic flaw detection method as described above, the ultrasonic flaw detection system comprising:

the material ultrasonic signal data acquisition module is used for transmitting ultrasonic signals to the interior of a material to be detected by utilizing the ultrasonic transmitter and receiving the signals by utilizing the ultrasonic receiver so as to acquire material ultrasonic signal data;

the dynamic data extraction module is used for carrying out dynamic data extraction on the material ultrasonic data so as to obtain the material ultrasonic dynamic data;

the feature data extraction module is used for carrying out significant feature extraction and potential hidden danger feature extraction on the ultrasonic dynamic data of the material so as to obtain significant feature data and potential hidden danger feature data;

the defect determining module is used for performing defect position determining processing on the remarkable characteristic data so as to obtain material defect position data, and performing defect degree determining processing on the potential hidden danger characteristic data so as to obtain material defect degree data;

and the material ultrasonic flaw detection report generation module is used for generating material ultrasonic flaw detection report data according to the material defect position data and the material defect degree data.

The invention has the beneficial effects that: the invention can acquire ultrasonic signal data inside the material and provide source data, which enables the method to detect the structure and defect condition inside the material. Dynamic data extraction is carried out on the ultrasonic data, so that the change trend and fluctuation condition in the material can be captured, and the analysis accuracy is enhanced. By combining the significant features and the potential hidden danger features, the multi-dimensional features are extracted from the ultrasonic dynamic data of the material, so that the significant features can be distinguished, the potential hidden danger features can be found, and the sensitivity and the comprehensiveness of defect detection are improved. The defect position determination and defect degree determination processes, the method can accurately locate and evaluate defects in materials. This helps engineers to more finely analyze the impact of defects and the degree of urgency. Ultrasonic flaw detection report data are generated, and information of positions and degrees of defects is comprehensively presented, so that quantitative basis is provided for decision making. The condition of the material can be rapidly and accurately analyzed, thereby providing reliable support for decision making. This helps to reduce subjectivity of human judgment and improve reliability of decision making. The invention combines ultrasonic nondestructive testing and material characteristic analysis, so that the evaluation of the internal structure and performance of the material is more comprehensive.

Drawings

Other features, objects and advantages of the application will become more apparent upon reading of the detailed description of a non-limiting implementation, made with reference to the accompanying drawings in which:

FIG. 1 is a flow chart showing the steps of an ultrasonic inspection method of an embodiment;

FIG. 2 shows a step flow diagram of step S1 of an embodiment;

FIG. 3 shows a step flow diagram of step S14 of an embodiment;

FIG. 4 shows a step flow diagram of step S15 of an embodiment;

FIG. 5 shows a step flow diagram of step S2 of an embodiment;

FIG. 6 shows a step flow diagram of step S3 of an embodiment;

fig. 7 shows a step flow diagram of step S4 of an embodiment.

Detailed Description

The following is a clear and complete description of the technical method of the present patent in conjunction with the accompanying drawings, and it is evident that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, are intended to fall within the scope of the present application.

Furthermore, the drawings are merely schematic illustrations of the present application and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus a repetitive description thereof will be omitted. Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. The functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor methods and/or microcontroller methods.

It will be understood that, although the terms "first," "second," and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of example embodiments. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

Referring to fig. 1 to 7, the application provides an ultrasonic flaw detection method, which comprises the following steps:

specifically, for example, a piezoelectric ultrasonic transmitter and a receiver are used, the transmitter is made to generate an ultrasonic signal by applying an electric field, and the receiver receives the reflected signal. The ultrasonic transmitters and receivers may be positioned at different locations of the material as needed to obtain more comprehensive signal data.

specifically, for example, data within a specific time window is extracted from the acquired ultrasonic signals to form a dynamic data segment. The data segment can be subjected to filtering and denoising operations through a data processing algorithm so as to improve the signal quality.

specifically, features of the signal in the time and frequency domains are extracted, for example, using a time-frequency analysis method such as short-time fourier transform (STFT). The application of wavelet transforms extracts information from different frequency ranges, helping to detect defects of different depths. The data is automatically feature extracted using a machine learning method, such as Convolutional Neural Network (CNN) or Recurrent Neural Network (RNN).

Specifically, the position of the defect reflection in the signal is determined, for example, using a signal processing method such as phase analysis. The propagation time of the signal is used in conjunction with the ultrasound propagation model to estimate the depth of the defect. The extent and size of the defect is assessed based on the signal strength and the characteristic variation. Phase analysis: by analyzing the phase change of the signal, the position of the defect reflection in the signal is determined. When an ultrasonic signal encounters a defect in the material, a portion of the signal is reflected back and a change in the phase of the reflected signal reveals the location of the defect. Propagation time estimation: in conjunction with the ultrasound propagation model, the propagation time of the signal is used to estimate the depth of the defect. The speed at which ultrasonic waves propagate in a material is known, and therefore the propagation distance of a signal is deduced from the round-trip propagation time of the signal, thereby estimating the depth of a defect. Signal intensity analysis: the detected signal strength may provide information about the defect. Larger defects generally result in stronger signal reflections, so the signal strength can be used to estimate the size of the defect. And (3) characteristic change analysis: different types of defects may cause characteristics of the signal to change, such as amplitude, frequency. By analyzing the variations of these features, the properties of the defect, such as the direction, shape of the crack, are obtained.

Specifically, for example, defect location, extent, and other relevant information are integrated into the report content. Graphical interface tools, such as Matplotlib or Plotly, can be used to draw charts, images, and annotations to generate an easily understood report.

Preferably, step S1 is specifically:

step S11: acquiring material parameter data;

specifically, for example, parameter data of the density and the elastic modulus of the material to be detected are acquired by using laboratory test equipment such as a material tester and an acoustic velocimeter. Parameter data of the acoustic wave propagation speed and the bulk modulus of the material are obtained from a material database or literature.

specifically, a three-dimensional model of the material is drawn from the material parameter data, for example, using Computer Aided Design (CAD) software. Using Finite Element Analysis (FEA) software, an accurate three-dimensional model of the material is constructed based on the material parameter data.

Step S13: acquiring material detection demand data;

specifically, information of the detection area, the detection depth, and the defect size is acquired, for example, through a software input interface.

specifically, coordinates of a series of test points are generated on or within the material model, for example, according to the detection requirements. Coordinate data for the test points is generated using a programming tool of MATLAB, python.

specifically, the initial meshing is performed according to the material model, for example, using mesh generation software. And optimizing the grid according to the material parameter data by utilizing a self-adaptive grid algorithm so as to improve the simulation precision.

specifically, ultrasonic propagation simulation is performed based on the material model and test point data, for example, using finite element simulation software. The propagation of ultrasound within the material is simulated using numerical solutions, such as the finite difference time domain method (FDTD) or the Finite Element Method (FEM).

specifically, the propagation analog data is converted into an analog detection signal using, for example, an analog detection data generation algorithm. And optimizing the position of the test point according to the analog detection signal and the actual test point data so as to improve the detection precision.

Specifically, a pulsed ultrasonic signal is transmitted, for example using an ultrasonic transmitter, which propagates inside the material and is captured by an ultrasonic receiver. After the signal is received, the signal is converted into a digital signal through an analog-to-digital converter so as to acquire ultrasonic signal data.

Preferably, step S12 is specifically:

Specifically, for example, acoustic characteristic data of material parameters are acquired, and acoustic characteristic data of sound wave propagation speed and density of a material are acquired from a material database, an experimental test or a literature. And matching and checking are carried out according to the known material parameter data and the acquired acoustic characteristic data, so that the accuracy and consistency of the acoustic characteristic data are ensured. Material acoustic wave propagation velocity data are obtained from an acoustic velocity tester: c=3000 m/s. Material density data were obtained from a density meter: ρ=800 kg/m. And according to the acoustic characteristic matching, confirming that the acoustic wave propagation speed and the density matching are correct.

Specifically, for example, area division: based on the detection area data, a three-dimensional model of the detection material is divided into different detection areas by using a three-dimensional model dividing algorithm. Generating a test point: and generating test points meeting the requirements in each area according to the detection precision requirement data and the material parameter data and combining the area division data. Screening test nodes: and carrying out preliminary screening on the generated test points, and removing redundant or repeated points to optimize the subsequent calculation and analysis process. Batch random selection and maximum relative distance selection: according to the test point data of the test area, the test points are divided into different batches, and random selection is carried out in each batch. Then, in each batch, a point with the largest relative distance from other selected points is selected as a first detection material test point.

Specifically, for example, the detection area data provides a rectangular parallelepiped area with a length, width, and height of 2m, 1m, and 3m, respectively. Accuracy requires data of one test point per square meter, and a total of 2×1×3=6 test points need to be generated. In the test node screening stage, it is assumed that 10 candidate points are generated, and then 6 test points are screened out.

Specifically, for example, initial meshing: and (3) carrying out initial grid division on the three-dimensional model of the detection material by using grid generation software or algorithm to generate initial grid data. Grid quality assessment: using a grid quality assessment algorithm, the quality of the initial grid is assessed based on the material density data, identifying potential grid problems. Global meshing optimization: and according to the material elastic modulus data and the grid quality evaluation data, an optimization algorithm is used for adjusting the initial grid, and global grid division is optimized. Minimum refractive region grid optimization: according to the material sound wave propagation speed data, on the basis of a global optimization grid, grid division is optimized by using a mathematical model and a numerical calculation method so as to adapt to ultrasonic wave propagation, in particular to a minimum refraction area.

Specifically, for example, assume that the initial meshing generates 10000 mesh cells. The mesh quality assessment uses a quantization index, such as mesh shape, ranging from 0 to 1. In global grid division optimization, it is assumed that the grid quality is successfully improved to 0.8 through an iterative optimization algorithm. The grid optimization of the minimum refraction area relates to a complex acoustic wave propagation mathematical model, numerical simulation and calculation are required to be carried out according to specific conditions, and the optimal grid division is obtained.

；

a (t, x) is the ultrasonic amplitude data at time t and position x, P ₀ The ultrasonic wave initial pressure data comprise rho, c, x, gamma, ultrasonic wave position data, ultrasonic wave angular frequency data, t, ultrasonic wave time data and k, wherein rho is the density data of a material to be detected, c is the ultrasonic wave propagation speed, x is the ultrasonic wave position data, gamma is the bulk modulus data of the material, omega is the ultrasonic wave angular frequency data.

The invention constructs an ultrasonic wave propagation simulation calculation formula which can simulate the propagation process of ultrasonic waves in the material to be detected and provides ultrasonic wave amplitude data at different times and positions, thereby realizing the simulation of the propagation process of the ultrasonic waves. By calculating the ultrasonic amplitude data in the formula, whether the defect exists in the material to be detected can be judged, and the defect can interfere the propagation of ultrasonic waves, so that the amplitude change is influenced. Representing the sound wave amplitude at different times and locations. The a (t, x) calculated will reveal the amplitude variation of the ultrasonic wave as it propagates in the material to be examined. P (P) ₀ Representing the initial pressure of the sound wave as it is emitted. It is the initial energy of the sound wave. By adjusting P ₀ The propagation of sound waves under different initial conditions can be simulated. The density data rho of the material to be detected influences the propagation speed and the propagation speed of sound waves in the materialEnergy dissipation. The density differences of the different materials will produce different acoustic responses in the simulation. The ultrasonic wave propagation velocity c represents the propagation velocity of an acoustic wave in a material. The physical properties of materials can affect the speed of sound, and the speed of sound propagation can vary from material to material and from frequency to frequency. The ultrasonic position data x indicates a certain position in the material. By simulating acoustic propagation at different locations, acoustic responses at different locations are obtained for defects detected. Bulk modulus data gamma for a material, representing the stiffness and elasticity of the material, describes the response of the material to external stresses, gamma affecting the frequency and amplitude of sound waves. Ultrasonic angular frequency data ω represents the frequency of the acoustic wave. The frequency determines the vibration period of the sound waves, which will interact in different ways with structures and defects in the material. The invention provides a detailed understanding of the propagation of ultrasonic waves in materials through simulation calculations.

Preferably, step S2 is specifically:

in particular, for example, the signal is divided into smaller segments, implemented using a method of window function.

specifically, for example, the instantaneous frequency is a frequency change of a signal in time, and can be obtained by calculating an instantaneous phase change of the signal, one common method is to obtain an instantaneous phase by performing Hilbert transformation on the signal, and then obtaining a derivative thereof to obtain the instantaneous frequency.

in particular, for example, pulse features may be used to identify abrupt or peak values in a signal, which may be extracted by detecting extreme points of the signal or using amplitude thresholds.

in particular, for example, time-frequency analysis may reveal changes in time and frequency of the signal, which is analyzed into a representation in the time-frequency domain using short-time fourier transform (STFT) or Continuous Wavelet Transform (CWT) methods.

Specifically, for example, integrating the instantaneous frequency data, the pulse characteristic data, the time-frequency characteristic representation data, and the corresponding ultrasonic signal segmentation data, a data map may be created to show dynamic characteristics of the signal, such as an image, a spectrogram, and a waterfall.

Preferably, step S3 is specifically:

specifically, the frequency spectrum conversion is performed using, for example, a fourier transform formula.

Specifically, for example, the time-frequency signal data is subjected to spectrum mapping, thereby obtaining frequency representation data.

specifically, for example, wavelet packet decomposition is a signal analysis method for acquiring frequency characteristics of a signal by decomposing the signal into components of different frequencies and amplitudes. A wavelet packet decomposition algorithm is applied to the previously calculated frequency representation data (e.g., the time-frequency distribution obtained by short-time fourier transform). The signal is decomposed into different frequency subbands and amplitude and phase information is provided for each subband. Frequency characteristics of the signal in different frequency ranges are obtained.

specifically, for example, energy calculation is performed on the wavelet packet decomposed sub-spectrum data to obtain energy distribution conditions of different frequency components. And carrying out energy calculation on the sub-spectrum data obtained by decomposing the wavelet packet. For each sub-band, its energy is calculated, expressed as an integral of the square of the amplitude of the signal within that band. This will provide an energy distribution profile for each frequency subband, identifying the energy concentration and dispersion of the signal.

in particular, energy features are extracted, for example, from sub-spectral energy data, which may reveal the energy distribution and intensity of the acoustic wave signal. Various energy features are extracted from the sub-spectrum energy data. These features include maximum energy, average energy, statistical properties of the energy distribution (e.g., standard deviation, skewness, kurtosis).

specifically, statistical features and nonlinear features are extracted from the energy feature data, for example. The statistical features may include mean, variance, and the non-linear features may reflect the non-linear behavior of the signal. Statistical analysis and nonlinear analysis methods are applied to the energy signature data to obtain statistical and nonlinear signatures. Statistical features may include a series of statistical values such as mean, variance, skewness, and kurtosis, which are the degree of concentration, degree of variation, and distribution morphology of the energy distribution. The nonlinear characteristics then cover some characteristics related to the nonlinear behavior of the signal, such as peak-to-peak, kurtosis, hurst index.

specifically, salient feature data is obtained, for example, by combining a statistical feature and a nonlinear feature.

in particular, extremum data is extracted, for example, from energy signature data, which can be used to capture special features of potential hazards. Extremum data, i.e., maximum and minimum values of the energy distribution, are extracted from the energy characteristic data. These extremum data are relevant to potential hidden individuals because certain material states or defects create anomalies in specific areas of the energy distribution, resulting in significant changes in energy.

；

E _sub (f, t) is the frequency range [ f, f+Δf]In the time range [ t, t+Δt ]]The sub-spectrum energy in the frequency spectrum, f is initial frequency variable data, delta f is frequency range width data, t is initial time variable data, delta t is time range width data, j is imaginary unit data,time domain signal data which is ultrasonic dynamic data of material, < > >For the frequency range integration data,for time range integral data, e is the natural exponential direction, pi is the periodicityData,/->D is a differential sign, which is a material impulse response function.

Preferably, step S4 is specifically:

step S41: acquiring image data of a material to be detected;

in particular, for example, image data of the material to be detected is acquired, which may be a visual representation of the acoustic wave signal, or image data of the material.

specifically, for example, abnormality recognition is performed using the salient feature data extracted before. By comparing the characteristic data with the reference data in the normal state, a remarkable abnormal condition is detected.

specifically, for example, coordinate labeling is performed on the image of the material to be detected according to the detection point data corresponding to the significant anomaly identification data, so as to determine the position of the anomaly point.

specifically, the range of influence of the defect will be evaluated, for example, based on an abnormal situation.

Specifically, position data of defects in the material are integrated according to coordinate labeling and influence region evaluation data, for example.

specifically, the detected defects are classified, for example, based on the previous feature data, and specific types thereof, such as cracks, bubbles, are determined.

specifically, the latent defect feature data is evaluated, for example, to determine the extent of the defect. This may include information of the size, depth of the defect.

Specifically, for example, the kind and degree data of defects are integrated to obtain complete material defect information.

The invention can acquire ultrasonic signal data inside the material and provide source data, which enables the method to detect the structure and defect condition inside the material. Dynamic data extraction is carried out on the ultrasonic data, so that the change trend and fluctuation condition in the material can be captured, and the analysis accuracy is enhanced. By combining the significant features and the potential hidden danger features, the multi-dimensional features are extracted from the ultrasonic dynamic data of the material, so that the significant features can be distinguished, the potential hidden danger features can be found, and the sensitivity and the comprehensiveness of defect detection are improved. The defect position determination and defect degree determination processes, the method can accurately locate and evaluate defects in materials. This helps engineers to more finely analyze the impact of defects and the degree of urgency. Ultrasonic flaw detection report data are generated, and information of positions and degrees of defects is comprehensively presented, so that quantitative basis is provided for decision making. The condition of the material can be rapidly and accurately analyzed, thereby providing reliable support for decision making. This helps to reduce subjectivity of human judgment and improve reliability of decision making. The invention combines ultrasonic nondestructive testing and material characteristic analysis, so that the evaluation of the internal structure and performance of the material is more comprehensive.

The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An ultrasonic flaw detection method is characterized by comprising the following steps:

2. The method according to claim 1, wherein step S1 is specifically:

step S11: acquiring material parameter data;

step S13: acquiring material detection demand data;

3. The method according to claim 2, wherein step S12 is specifically:

4. The method according to claim 2, wherein the material inspection requirement data includes inspection area data and inspection accuracy requirement data, and step S14 is specifically:

5. The method according to claim 2, wherein the material parameter data includes material density data, material elastic modulus data, and material acoustic wave propagation speed data, and the step of optimizing meshing in step S15 is specifically:

6. The method according to claim 2, wherein the ultrasonic propagation simulation in step S16 is simulated by an ultrasonic propagation simulation calculation formula, wherein the ultrasonic propagation simulation calculation formula is specifically:

；

7. The method according to claim 1, wherein step S2 is specifically:

8. The method according to claim 1, wherein step S3 is specifically:

；

E _sub (f, t) is the frequency range [ f, f+Δf]In the time range [ t, t+Δt ]]The sub-spectrum energy in the frequency spectrum, f is initial frequency variable data, delta f is frequency range width data, t is initial time variable data, delta t is time range width data, j is imaginary unit data,time domain signal data which is ultrasonic dynamic data of material, < >>Integrating data for a frequency range +.>For the time-range integral data, e is the natural exponential direction, pi is the periodic data, ++>D is a differential sign, which is a material impulse response function.

9. The method according to claim 1, wherein step S4 is specifically:

step S41: acquiring image data of a material to be detected;

10. An ultrasonic inspection system for performing the ultrasonic inspection method of claim 1, the ultrasonic inspection system comprising: