A non-destructive flaw detection method and system based on ultrasonic wave modulation eddy current effect

Abstract

The application relates to the technical field of nondestructive testing of metal forgings, in particular to a nondestructive flaw detection method and system based on ultrasonic wave modulation eddy current effect, which comprises S100 double-mode cooperative excitation and collection; S200 dynamic response characteristic vector extraction; S300 four-dimensional topological field construction; and S400 defect quantitative diagnosis. Specifically, microcosmic vibration is generated in a defect area by emitting focused ultrasonic waves, and eddy current signal dynamic changes are synchronously collected, then four-dimensional characteristics of response intensity, phase angle, attenuation time and main frequency are extracted, a defect topological network is constructed in a characteristic space, real defect boundaries are identified based on a sensitivity threshold, three-dimensional dimensions and an activity index are calculated, the breakthrough of eddy current detection depth is realized, false signals are automatically filtered through the four-dimensional topological field, the false alarm rate of defects is reduced, and three-dimensional dimensions of defect length, width and depth and a danger level are synchronously output, the detection accuracy and the detection range are effectively improved, and the operation difficulty of the detection process is reduced.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of nondestructive testing of metal forgings, and particularly relates to a nondestructive testing method and system based on ultrasonic wave modulation eddy current effect. BACKGROUND

[0002] In the processing of metal forgings, due to material or process factors, various defects are prone to occur, such as cracks caused by improper temperature, severe deformation and rapid cooling during forging, internal cavities formed by incomplete closure of shrinkage holes, pores or porosity in the raw material after forging, inclusions formed by the strip-shaped or sheet-shaped distribution of impurities such as oxides and sulfides mixed during smelting in the matrix during forging, and the like. These defects destroy the continuity of the metal matrix, reduce the strength, toughness and fatigue life of the forgings, and therefore often need to be evaluated by defect detection.

[0003] In the traditional process, eddy current detection or ultrasonic detection is often used for defect detection. However, firstly, the skin effect leads to a high missed detection rate for deep defects (> 3mm), such as subsurface inclusions in titanium alloy forgings; secondly, local conductivity variations cause false positives, and actual tests show that the false positive rate in the uneven heat treatment area is more than 30%; thirdly, it is difficult to distinguish the defect type, for example, the similarity of the signals of cracks and hard inclusions is high and difficult to distinguish. Although ultrasonic testing (UT) has depth resolution capability, it has insufficient sensitivity to closed defects (such as fatigue cracks), it is difficult to quantify defect size, and coupling agents (water, oil, gel) need to be used to ensure effective transmission of acoustic energy during detection, and the detection process is complicated. SUMMARY

[0004] The purpose of the present application is to provide a nondestructive testing method and system based on ultrasonic wave modulation eddy current effect, which can improve the detection accuracy and detection range while reducing the operation difficulty of the detection process.

[0005] In a first aspect, the embodiments of the present application provide a nondestructive testing method based on ultrasonic wave modulation eddy current effect, comprising:

[0006] S100, dual-mode cooperative excitation and acquisition, ultrasonic excitation sound waves are emitted during eddy current detection of the target forging, and ultrasonic echo signals and eddy current impedance signals are collected in the dual-time window before and after the ultrasonic wave emission;

[0007] S200, dynamic response feature vector extraction, based on the eddy current impedance signal, a dynamic response difference value is calculated, and a dynamic response feature vector related to four-dimensional topological field construction is extracted; wherein the dynamic response feature vector at least includes a dynamic response modulus, a phase angle, a response main frequency and an attenuation time constant;

[0008] S300, four-dimensional topological field construction, determining the depth coordinates according to the ultrasonic echo signal, combining the spatial coordinates of the scanning points with the depth coordinates, associating the dynamic response feature vectors to form a four-dimensional feature data point cloud, and determining the defect boundary according to the dynamic response feature vectors and the defect sensitivity scalar field model;

[0009] S400, defect quantification and diagnosis, determining the defect size according to the defect boundary point cloud and the acoustic echo signal, and further evaluating the defect activity level.

[0010] In some embodiments, step S100 comprises:

[0011] S110, eddy current detection of the target forging, focused ultrasonic pulses are emitted to the interior of the forging during the detection process to excite micro-mechanical effects in the defect area;

[0012] S120, collecting ultrasonic echo signals, extracting echo peak amplitude and time of flight;

[0013] S130, collecting eddy current impedance signals in the first time period before ultrasonic emission to obtain static reference values, and collecting eddy current impedance signals in the time window of the second time period after emission to obtain dynamic response values.

[0014] In some embodiments, the dynamic response feature vector related to the four-dimensional topological field construction comprises:

[0015] The dynamic response difference value is calculated, and then the dynamic response modulus, phase angle, response main frequency and decay time constant are extracted.

[0016] In some embodiments, step 300 comprises:

[0017] S310, determining the depth coordinates according to the ultrasonic echo signal, and determining the three-dimensional physical model in combination with the spatial coordinates of the scanning points;

[0018] S320, fusing the dynamic response modulus, phase angle, response main frequency and decay time constant with the three-dimensional physical model to form a four-dimensional feature data point cloud;

[0019] S330, through the Delaunay triangulation algorithm, taking each feature vector as the vertex coordinate, establishing the tetrahedral unit connection structure between the spatial adjacent vertices, and generating an unstructured grid;

[0020] S340, substituting the dynamic response feature vector into the defect sensitivity scalar field model to determine the defect sensitivity;

[0021] S350, extracting the isosurface reaching the preset defect sensitivity to determine the defect boundary.

[0022] In some embodiments, step S350 further comprises:

[0023] S351: Morphological filtering of the defect boundary removes outliers using an anisotropic diffusion algorithm model.

[0024] In some embodiments, step 400 comprises:

[0025] S410, in the defect boundary point cloud, the defect length is calculated by projecting along the maximum principal inertia axis, the defect width is fitted as the minor axis of an ellipse on the cross section in the vertical length direction, and the defect depth is determined based on the range of depth coordinates;

[0026] S420, the defect activity level is evaluated by a risk coefficient model.

[0027] In some embodiments, the defect length is calculated by projecting along the maximum principal inertia axis, comprising:

[0028] S411, principal component analysis (PCA) is performed on the defect boundary point cloud, and the eigenvector corresponding to the maximum eigenvalue is selected as the length direction;

[0029] S412, the defect boundary point cloud is projected to the direction, and the projection extreme difference is taken as the defect length.

[0030] The second aspect of the application provides a non-destructive testing system based on ultrasonic wave modulation eddy current effect, comprising:

[0031] An ultrasonic wave-eddy current composite integrated detection module is used to emit ultrasonic excitation sound waves when performing eddy current detection on a target forging, and to collect ultrasonic echo signals and eddy current impedance signals in double time windows before and after the ultrasonic wave emission;

[0032] An extraction module is used to calculate a dynamic response difference value based on the eddy current impedance signals, and to extract a dynamic response feature vector related to four-dimensional topological field construction; wherein the dynamic response feature vector at least includes a dynamic response modulus, a phase angle, a response main frequency, and an attenuation time constant;

[0033] A four-dimensional topological field construction module is used to determine a depth coordinate according to the ultrasonic echo signals, to combine the scanning point space coordinates and the depth coordinates, to associate the dynamic response feature vector to form a four-dimensional feature data point cloud, and to determine a defect boundary according to the dynamic response feature vector and a defect sensitivity scalar field model;

[0034] A defect quantification diagnosis module is used to determine a defect size according to the defect boundary point cloud and the sound wave echo signals, and to evaluate a defect activity level.

[0035] In some embodiments, the ultrasonic wave-eddy current composite integrated detection module comprises a composite probe sensor module and a synchronous control unit, and the synchronous control unit is used to control the composite probe sensor module to perform excitation according to a preset frequency, a preset energy level, and a preset time length.

[0036] The third aspect of the present application provides a non-volatile storage medium storing a computer program, wherein the program is executed by a processor to implement the steps of the above method.

[0037] The beneficial effects of the present application are: by emitting focused ultrasonic waves to excite micro-vibration in the defect area, synchronously collecting the dynamic changes of eddy current signals, then extracting four-dimensional features of response intensity, phase angle, decay time and main frequency, and constructing a defect topology network in the feature space, then identifying the real defect boundary based on the sensitivity threshold, and calculating the three-dimensional size and activity index, thereby realizing the breakthrough of eddy current detection depth, and automatically filtering false signals through the four-dimensional topology field, reducing the false positive rate of defects, and synchronously outputting the three-dimensional size of defect length / width / depth and the danger level, through the fusion detection of the excitation state of ultrasonic waves and eddy current, the high-depth effect of ultrasonic detection is realized on the basis of the advantages of eddy current detection, and without the use of coupling agent, therefore, while effectively improving the detection accuracy and detection range, the operation difficulty of the detection process is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 The flowchart of the non-destructive flaw detection method based on the ultrasonic wave modulated eddy current effect in the present application;

[0039] Figure 2 The flowchart of the dual-mode collaborative excitation and collection in the present application;

[0040] Figure 3 The flowchart of the four-dimensional topology field construction in the present application;

[0041] Figure 4 The module diagram of the non-destructive flaw detection system based on the ultrasonic wave modulated eddy current effect in the present application. DETAILED DESCRIPTION

[0042] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme of the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0043] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present application.

[0044] In the embodiments of the present application, it should be understood that the disclosed apparatus and method can also be implemented in other manners. The embodiments described above are merely exemplary. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architectures, functions and operation of the apparatus, method and computer program product according to the embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment or a part of code, which contains one or more executable instructions for implementing the specified logic function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks can actually be executed in parallel, and sometimes they can be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and the combination of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0045] In addition, the functional modules in the various embodiments of the present application can be integrated together to form a separate part, or each module can exist independently, or two or more modules can be integrated to form a separate part.

[0046] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product is stored in a storage medium, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0047] Traditional ECT faces three technical bottlenecks: first, the skin effect leads to a detection rate of more than 60% for deep defects (>3mm), such as subsurface inclusions in titanium alloy forgings; second, local conductivity variations cause false positives, with a false positive rate of 28% in the unevenly heat-treated area; third, it is difficult to distinguish between defect types, with a similarity of 85% between crack and hard inclusion signals. Although UT has depth resolution capability, it lacks sensitivity to closed defects (such as fatigue cracks) and is difficult to quantify defect size. Existing ultrasonic- eddy current combined technology only performs simple data superposition and does not solve the core problem of multi-modal signal fusion and dynamic response analysis.

[0048] Therefore, the present application provides a non-destructive testing method and system based on ultrasonic wave modulation eddy current effect, which cooperatively excites the micro-mechanical response and electromagnetic property change of the defect area, and realizes accurate determination, size quantification and activity evaluation of deep defects by combining high-dimensional topological field construction technology.

[0049] Embodiment one

[0050] With reference to Figure 1 The first aspect of the present application provides a non-destructive testing method based on ultrasonic wave modulation eddy current effect, comprising the following steps.

[0051] S100, dual-mode cooperative excitation and collection, ultrasonic excitation sound waves are emitted during eddy current detection of the target forging, and ultrasonic echo signals and eddy current impedance signals are collected in the dual time windows before and after the ultrasonic wave emission;

[0052] With reference to Figure 2 In step S100, the following steps are included:

[0053] S110, eddy current detection of the target forging, focused ultrasonic pulses are emitted into the forging during the detection process to excite micro-mechanical effects in the defect area;

[0054] First, an ultrasonic wave-eddy current composite integrated detection module is constructed, the composite probe sensor is driven by a high-precision mechanical scanning platform to move along the surface of the forging according to a predetermined path, the relative spatial coordinates and scanning time stamps at each scanning point spatial coordinates are recorded, the ultrasonic wave emission unit is controlled by the synchronous control unit to emit high-frequency pulses into the forging, the sound wave vibration causes micro-friction heating at the defect, at this time the electromagnetic signal will abnormally fluctuate, so that the eddy current detection can detect deeper defects. The pulse frequency is controlled in the range of 5-10MHz, and the pulse energy is accurately adjusted in the range of 50-200mJ, so as to ensure that the defect area generates effective micro-mechanical effects without causing material damage.

[0055] S120, collecting ultrasonic echo signals, extracting echo peak amplitude and flight time;

[0056] During the scanning process, the synchronous trigger synchronous control unit performs signal acquisition, and records the ultrasonic scanning waveform and the scanning time through the ultrasonic receiving channel, and extracts the echo peak amplitude parameter and the ultrasonic time-of-flight parameter .

[0057] S130, in the first time period before the ultrasonic wave is emitted, the eddy current impedance signal is collected to obtain a static reference value, and in the time window of the second time period after the emission, the eddy current impedance signal is collected to obtain a dynamic response value.

[0058] In this embodiment, the first time period is 100 microseconds, the second time period is 500 microseconds, and the eddy current detection channel continuously collects complex impedance signals in the time window of 100 microseconds before the ultrasonic pulse is emitted to 500 microseconds after the emission at a sampling rate of not less than 10MS / s, thereby obtaining the electromagnetic signal before the sound wave impact and the electromagnetic signal after the impact, and determining the static reference value and the dynamic response value .

[0059] S200, dynamic response feature vector extraction, based on the eddy current impedance signal, calculating the dynamic response difference value, and then extracting the dynamic response feature vector related to the four-dimensional topological field construction; wherein the dynamic response feature vector at least includes the dynamic response modulus, the phase angle, the response main frequency and the decay time constant;

[0060] First, according to the static reference value and the dynamic response value, the dynamic response difference vector components ΔR and ΔX are calculated, wherein:

[0061]

[0062] Then the dynamic response modulus is derived by Euclidean norm operation:

[0063]

[0064] According to the phase angle θ of ΔR and ΔX:

[0065]

[0066] The time-domain decay curve of the dynamic response modulus ‖ΔZ‖ is nonlinear least squares fitting, which adopts an exponential decay model to determine the decay time constant τ:

[0067]

[0068] Where A is the amplitude coefficient, then the fast Fourier transform is performed to analyze the dynamic response spectrum, and the dominant frequency component is extracted, this process converts the physical response into a vector containing four feature dimensions [‖ΔZ‖, θ, τ, wherein the eigenvalue range of phase angle θ is mapped to the type of defect.

[0069] S300, four-dimensional topological field construction, determining the depth coordinate according to the ultrasonic echo signal, combining the spatial coordinates of the scanning points with the depth coordinates, associating the dynamic response feature vector to form a four-dimensional feature data point cloud, and determining the defect boundary according to the dynamic response feature vector and the defect sensitivity scalar field model;

[0070] Referring to Figure 3 , specifically, step 300 includes:

[0071] S310, determining the depth coordinate according to the ultrasonic echo signal, and determining the three-dimensional physical model in combination with the spatial coordinates of the scanning points;

[0072] Combining the two-dimensional plane coordinates (x, y) of the scanning points with the depth coordinate z, wherein the depth coordinate is obtained by converting the ultrasonic time of flight, and the conversion relationship is , C represents the ultrasonic propagation speed in the material.

[0073] S320, fusing the dynamic response modulus, phase angle, response main frequency and decay time constant with the three-dimensional physical model to form a four-dimensional feature data point cloud;

[0074] According to the time stamp of data acquisition, the obtained three-dimensional physical coordinates of each scanning point are fitted with the four-dimensional feature vector, and each scanning point generates a data record: {x, y, z, ‖ΔZ‖, θ, τ, }, for example, when a point is detected:

[0075] Coordinate positioning: (x, y) = (35.2, 48.7) mm

[0076] Ultrasonic depth measurement:

[0077] Eddy current feature extraction: ΔR = 0.15Ω, ΔX = 0.12Ω

[0078] Feature vector extraction: ‖ΔZ‖ = 0.192Ω, θ = 0.78 rad, τ = 0.8 ms, = 85 kHz

[0079] Generating data points: DP k ={35.2, 48.7, 8.28, 0.192, 0.78, 0.8, 85}

[0080] When scanning a 100×100 grid, 10000 data points are generated, thereby forming a four-dimensional feature data point cloud.

[0081] S330, by Delaunay triangulation algorithm, with each feature vector as the vertex coordinates, the tetrahedral unit connection structure between the adjacent vertices in space is established, and the unstructured grid is generated;

[0082] In the four-dimensional feature subspace [‖ΔZ‖, θ, τ, ]Delaunay triangulation algorithm is executed, each feature vector is taken as vertex coordinates, the nearest feature points are automatically found and connected into tetrahedral grid, that is, an adaptive defect correlation network is constructed, and the connection rule follows the principle of "minimum containing sphere": the circumscribed sphere of any tetrahedron does not contain other feature points. Through the subdivision algorithm, the originally discrete weak signal points are integrated and connected, so that the defects that are easy to be ignored as noise are detected, and the isolated noise points that cannot form effective tetrahedrons with other points are automatically removed. At the same time, the tetrahedral units on the edge of the defect can present high gradient characteristics (such as θ value from 0.78 to 0.1), so as to accurately outline the crack tip, and the image is more accurate.

[0083] S340, the dynamic response feature vector is substituted into the defect sensitivity scalar field model to determine the defect sensitivity, and the defect sensitivity scalar field model is:

[0084]

[0085] Where S is the defect sensitivity, ‖ΔZ‖ is the dynamic response modulus, θ is the phase angle, τ is the decay time constant, is the preset crack feature phase angle;

[0086] The four-dimensional feature is compressed into a single-valued metric that can be fissile by the defect sensitivity scalar field model. ‖ΔZ‖ represents the defect vibration intensity, and the larger the value is, the more significant the crack is. e⁻τ represents that the faster τ decays, the larger the value is, so it is used to represent the capture crack fast closing feature, such as open crack: τ≈0.8ms, e⁻τ≈0.45; inclusion τ≈5ms, e⁻τ≈0.006; is the crack reference phase angle, represents the phase deviation degree, so the numerator strengthens the crack transient response, and the denominator suppresses the non-target signal.

[0087] S350, the isosurface of S=0.35 is extracted to determine the defect boundary.

[0088] According to the following Table 1, the feature vector of a certain defect is shown in the following table 1:

[0089] Table 1: Feature vector example table

[0090]

[0091] By detecting a plurality of crack data, S≥0.35 is a significant feature of the crack, so 12 edges of each tetrahedron are detected, if the two endpoints of an edge satisfy: , the intersection coordinates are calculated by linear interpolation method, and then the intersection points are connected as triangular facets, 0-4 triangular facets are generated for each tetrahedron, and finally the continuous curved surface in the three-dimensional space is spliced to determine the defect boundary.

[0092] In some embodiments, step S350 further comprises:

[0093] S351: The morphological filtering of the defect boundary adopts an anisotropic diffusion algorithm model to remove outlier noise points.

[0094] In the traditional algorithm, algorithms such as Gaussian blur will blur the defect edge, resulting in the crack tip being smoothed, so the diffusion equation is used for feature gradient detection in the embodiments of the application:

[0095]

[0096] Then the diffusion judgment coefficient is calculated:

[0097]

[0098] Where, when |∇S| is small, g≈1, representing the edge is a smooth area, then strong diffusion is performed to smooth the noise; when |∇S| is large, g≈0, representing the edge is an edge area, then the diffusion is inhibited to retain the sharp edge.

[0099] S is the main basis for judging the crack, and in other embodiments, the judgment of bubbles and inclusions is also included, so a inclusion sensitivity field G is constructed:

[0100]

[0101] Where, is a preset inclusion phase reference, such as ceramic inclusion =1.2 radian.

[0102] When judging bubbles, since the bubble edge is smooth, the value of is much smaller than that of other defects or even close to 0 radian, and the dominant frequency component is low, so when the defect point cloud satisfies: , and , the defect type can be determined as a pore; and when G>0.2, the defect can be determined as an inclusion.

[0103] S400, defect quantification diagnosis, the defect size is determined according to the defect boundary point cloud and the sound wave echo signal, wherein step 400 comprises:

[0104] S410, in the defect boundary point cloud, the defect length is calculated by projecting along the maximum principal inertia axis, the short axis of the ellipse fitted on the cross section in the vertical length direction is taken as the defect width, and the defect depth is determined based on the range of depth coordinates;

[0105] Specifically, when calculating the defect length by projecting along the maximum principal inertia axis:

[0106] S411, principal component analysis (PCA) is performed on the defect boundary point cloud, and the eigenvector corresponding to the maximum eigenvalue is selected as the length direction;

[0107] First, input each defect boundary point cloud data:

[0108]

[0109] Then, centering processing is performed:

[0110]

[0111] The eigenvalues are solved by the characteristic equation:

[0112]

[0113] Among them:

[0114]

[0115] Thus, each eigenvector corresponding to the eigenvalue ..., wherein the eigenvector V corresponding to the maximum λ points to the most dispersed direction of the point cloud, which is taken as the length direction.

[0116] S412, project the point cloud to the direction, and take the projection extreme difference as the defect length.

[0117] Project all points to the principal axis V:

[0118]

[0119] Thus, the projection coordinate set is determined, and then the range calculation is performed:

[0120]

[0121] Thus, the defect length is determined.

[0122] Then, the short axis of the ellipse fitted on the cross section in the vertical length direction is taken as the defect width, first, the cutting plane is constructed, and the points within ±δ from the plane (δ≈0.1mm) are selected, and then the ellipse fitting is performed:

[0123]

[0124] wherein (x', y') is the local coordinate of the cutting plane, phi is the inclination angle of the ellipse, a and b are the major and minor axes, and the minor axis of the ellipse is obtained as the defect width.

[0125] Finally, the defect depth is determined based on the range of the depth coordinate:

[0126]

[0127] S420, evaluating the defect activity level through a risk coefficient model:

[0128]

[0129] wherein lambda is the defect activity level, is the dynamic response modulus, is the response main frequency, is the echo peak amplitude of the ultrasonic wave, and theta is the phase angle, is the preset crack characteristic phase angle.

[0130] When the activity index lambda is greater than 0.5, it is determined as a high-risk defect, 0.1 < lambda <= 0.5 is a medium-risk defect, and lambda <= 0.1 is an acceptable defect.

[0131] Through size determination and risk coefficient determination, multi-dimensional defect determination is realized, further improving the accuracy of the judgment.

[0132] By emitting focused ultrasonic waves to excite micro-vibration in the defect area, the dynamic changes of the eddy current signal are synchronously collected, and then the response intensity, phase angle, decay time and main frequency four-dimensional features are extracted, and a defect topology network is constructed in the feature space. Then, based on the sensitivity threshold, the real defect boundary is identified, and the three-dimensional size and activity index are calculated, thereby realizing the breakthrough of eddy current detection depth. At the same time, through the four-dimensional topology field, false signals are automatically filtered, reducing the false positive rate of defects, and synchronously outputting the three-dimensional size of defect length / width / depth and risk level. Through the fusion detection of the excitation state of ultrasonic waves and eddy current, the high-depth effect of ultrasonic detection is realized on the basis of the advantages of eddy current detection.

[0133] Embodiment Two

[0134] Reference Figure 4 The second aspect of the present application provides a non-destructive flaw detection system based on ultrasonic wave modulated eddy current effect, comprising:

[0135] An ultrasonic wave-eddy current composite integrated detection module is used to emit ultrasonic excitation sound waves when performing eddy current detection on the target forging, and to collect ultrasonic echo signals and eddy current impedance signals in the double time windows before and after the ultrasonic wave emission.

[0136] In the embodiment, the ultrasonic transmitting wafer and the eddy current detection coil are integrated in a coaxial package structure, the ultrasonic wafer has a center frequency of 5-10 MHz and an adjustable focal length, the eddy current coil is configured in a differential array mode, and the minimum coil spacing is not greater than 0.5 mm. A high-temperature-resistant ceramic protective layer is covered on the surface of the probe to adapt to a working condition below 400 DEG C.

[0137] The ultrasonic wave-eddy current composite integrated detection module is controlled by an industrial robot arm or a mechanical motion assembly, and is provided with a laser tracker for real-time positioning and compensation to ensure the accuracy during detection scanning.

[0138] The extraction module is configured to calculate a dynamic response difference value based on the eddy current impedance signal, and further extract a dynamic response feature vector related to the four-dimensional topological field construction, wherein the dynamic response feature vector at least includes a dynamic response modulus, a phase angle, a response main frequency and a decay time constant.

[0139] The four-dimensional topological field construction module is configured to determine a depth coordinate according to the ultrasonic wave echo signal, combine the scanning point space coordinate with the depth coordinate, associate the dynamic response feature vector to form a four-dimensional feature data point cloud, and determine a defect boundary according to the dynamic response feature vector and a defect sensitivity scalar field model.

[0140] The built-in dynamic response separation algorithm and the four-dimensional topological field engine realize Underlay subdivision and sensitivity field calculation, the graphics processing unit accelerates the topological field construction, and a three-dimensional defect model is generated in real time and directly expressed on an external display screen.

[0141] The defect quantification diagnosis module is configured to determine a defect size according to the defect boundary point cloud and the acoustic wave echo signal, and further evaluate a defect activity level. The defect quantification diagnosis module integrates principal component analysis, elliptical fitting and activity index evaluation algorithms to generate the defect activity level.

[0142] In some embodiments, the ultrasonic wave-eddy current composite integrated detection module includes a composite probe sensor module and a synchronous control unit, and the synchronous control unit is configured to control the composite probe sensor module to be excited at a preset frequency, a preset energy level and a preset time length.

[0143] The synchronous control unit realizes hardware-level triggering based on a field programmable gate array, and the timing synchronization error of ultrasonic pulse transmission and eddy current signal acquisition is controlled within 10 nanoseconds. The analog-to-digital conversion unit adopts a double-channel 14-bit resolution, and the sampling rate is not less than 100 MS / s.

[0144] Embodiment three

[0145] The third aspect of the present application provides a non-volatile storage medium storing a computer program, and the program is executed by a processor to realize the method steps of the first embodiment.

[0146] The above description is only the preferred embodiment of the present application, and is not intended to limit the present application. The present application can have various modifications and changes for those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0147] It is apparent for those skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, but can be realized in other concrete forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be considered in all aspects as illustrative and not restrictive, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended to embrace all changes falling within the meaning and range of equivalents of the claims. Any reference signs in the claims should not be considered as limiting the claims involved.

Claims

1. A non-destructive inspection method based on the effect of ultrasonic wave modulation vortex, characterized in that, Comprising: S100, dual-mode collaborative excitation and acquisition, emitting ultrasonic excitation sound waves during eddy current detection of the target forging, and collecting ultrasonic echo signals and eddy current impedance signals in the dual time windows before and after the emission of the ultrasonic waves; S200, dynamic response feature vector extraction, calculating a dynamic response difference value based on the eddy current impedance signals, and then extracting a dynamic response feature vector related to four-dimensional topological field construction; wherein the dynamic response feature vector at least includes a dynamic response modulus, a phase angle, a response main frequency, and a decay time constant; S300, four-dimensional topological field construction, determining a depth coordinate according to the ultrasonic echo signals, combining the scan point space coordinates to determine a three-dimensional physical model, then fusing the dynamic response feature vector and the three-dimensional physical model to form a four-dimensional feature data point cloud, establishing a tetrahedral unit connection structure between adjacent vertexes in space with each feature vector as a vertex coordinate, generating an unstructured grid, and substituting the dynamic response feature vector into a defect sensitivity scalar field model; ； wherein S is a defect sensitivity, is a dynamic response modulus, θ is a phase angle, and τ is a decay time constant, is a preset crack characteristic phase angle, an iso-surface reaching a preset defect sensitivity is extracted to determine a defect boundary, and when the defect sensitivity S is greater than or equal to the iso-surface of the preset defect sensitivity, the defect is determined to be a crack; S400, defect quantification and diagnosis, determining a defect size according to the defect boundary point cloud and the acoustic echo signals, and then evaluating a defect activity level.

2. The non-destructive inspection method of claim 1, wherein, Step S100 includes: S110, eddy current detection of the target forging, emitting focused ultrasonic pulses into the interior of the forging during the detection process to excite micro-mechanical effects in the defect area; S120, collecting ultrasonic echo signals, extracting echo peak amplitude and time of flight; S130, collecting eddy current impedance signals in a first time period before the emission of the ultrasonic waves to obtain a static reference value, and collecting eddy current impedance signals in a time window of a second time period after the emission to obtain a dynamic response value.

3. The non-destructive inspection method of claim 2, wherein, Extracting a dynamic response feature vector related to four-dimensional topological field construction includes: Determining a static calculation dynamic response difference value according to the static reference value and the dynamic response value, extracting a dynamic response modulus, a phase angle using the dynamic response difference value according to a preset algorithm, and obtaining a response main frequency and a decay time constant through spectral analysis.

4. The non-destructive inspection method of claim 1 further comprising the step of: Step S300 further includes: S351: The morphology filtering of the defect boundary uses an anisotropic diffusion algorithm model to remove outlier noise points.

5. The non-destructive inspection method of claim 1, wherein, Step 400 includes: S410, in the defect boundary point cloud, projecting along the maximum principal inertia axis to calculate the defect length, fitting an ellipse minor axis on the cross section perpendicular to the length direction as the defect width, and determining the defect depth based on the range of the depth coordinate; S420, evaluating the defect activity level through a risk coefficient model.

6. The non-destructive inspection method of claim 5, wherein, Projecting along the maximum principal inertia axis to calculate the defect length includes: S411, performing principal component analysis (PCA) on the defect boundary point cloud, and selecting the feature vector corresponding to the maximum eigenvalue as the length direction; S412, projecting the defect boundary point cloud to this direction, and taking the projection extreme difference as the defect length.

7. A non-destructive inspection system based on ultrasonic modulation of eddy current effect, characterized in that, Comprising: An ultrasonic- eddy current composite integrated detection module for emitting ultrasonic excitation sound waves during eddy current detection of the target forging, and collecting ultrasonic echo signals and eddy current impedance signals in the dual time windows before and after the emission of the ultrasonic waves; The extraction module is configured to calculate a dynamic response differential value based on the eddy current impedance signal, and further extract a dynamic response feature vector related to four-dimensional topological field construction, wherein the dynamic response feature vector at least includes a dynamic response modulus, a phase angle, a response main frequency, and a decay time constant. The four-dimensional topological field construction module is configured to determine a depth coordinate according to the ultrasonic echo signal, determine a three-dimensional physical model in combination with a scanning point space coordinate, fuse the dynamic response feature vector and the three-dimensional physical model to form a four-dimensional feature data point cloud, establish a tetrahedral unit connection structure between adjacent vertexes in space with each feature vector as a vertex coordinate, generate an unstructured mesh, and substitute the dynamic response feature vector into a defect sensitivity scalar field model. ； wherein S is a defect sensitivity, is a dynamic response modulus, θ is a phase angle, and τ is a decay time constant, is a preset crack characteristic phase angle, an isosurface reaching a preset defect sensitivity is extracted to determine a defect boundary, and when the defect sensitivity S is greater than or equal to the isosurface of the preset defect sensitivity, the defect is determined to be a crack. The defect quantitative diagnosis module is configured to determine a defect size according to a defect boundary point cloud and the ultrasonic echo signal, and further evaluate a defect activity level.

8. The non-destructive inspection system of claim 7, wherein, The ultrasonic- eddy current composite integrated detection module includes a composite probe sensor module and a synchronous control unit, and the synchronous control unit is configured to control the composite probe sensor module to be excited at a preset frequency, a preset energy level, and a preset time length.

9. A non-volatile storage medium storing a computer program, characterized in that, The program is executed by the processor to implement the steps of the nondestructive testing method of any one of claims 1-6.

Images