Three-dimensional flaw detection method and device

Abstract

The invention provides a three-dimensional flaw detection method and device. The method comprises the steps that target probe operation parameter information corresponding to type information of a target hollow shaft is acquired, and the target probe operation parameter information comprises first operation parameter information corresponding to multiple types of target probes when the multiple types of target probes move in a spiral mode at the same time; controlling multiple types of target probes to move in a spiral mode in the axial direction according to the first operation parameter information, and obtaining first probe axial position information of the multiple types of target probes and first ultrasonic reflection signal information collected by the multiple types of target probes; processing the first ultrasonic reflection signal information in combination with a three-dimensional image processing technology, and extracting defect feature information from a processing result; and based on the axial position information of the first probe and the defect feature information, performing three-dimensional visual display in a three-dimensional model corresponding to the target hollow shaft by adopting a preset early warning display mode. The detection efficiency and the detection precision are improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of flaw detection technology, and in particular, to a three-dimensional flaw detection method and device. Background Art

[0002] As a key component of EMU, the internal quality of hollow shaft directly affects the safety and reliability of the entire system. Therefore, non-destructive testing of hollow shaft has become one of the important means to ensure product quality.

[0003] Currently, commonly used flaw detection methods include X-ray testing, magnetic particle testing, eddy current testing, and ultrasonic testing. Among them, ultrasonic testing is widely used due to its high sensitivity and wide applicability. Traditional ultrasonic testing methods usually use linear scanning or fixed-point scanning. Although ultrasonic testing methods can detect some obvious defects, when encountering hollow shafts with complex shapes, it is often necessary to repeatedly scan all areas inside the hollow shaft, which increases the detection time and cost. In addition, for small or deeply buried defects, it is difficult for traditional methods to determine the location of the defects, which is easy to cause missed detection. Summary of the invention

[0004] The embodiments of the present application provide a three-dimensional flaw detection method and device to solve the problems of low detection efficiency and low detection accuracy in the prior art.

[0005] In a first aspect, an embodiment of the present application provides a three-dimensional flaw detection method, comprising: Acquire target probe operation parameter information corresponding to the type information of the target hollow shaft, wherein the target probe operation parameter information includes first operation parameter information respectively corresponding to when multiple types of target probes move in a spiral manner at the same time; Controlling the multi-type target probe to move in a spiral manner in the axial direction according to the first operating parameter information, and acquiring the first probe axial position information of the multi-type target probe and the first ultrasonic reflection signal information collected by the multi-type target probe; Processing the first ultrasonic reflection signal information in combination with three-dimensional image processing technology, and extracting defect feature information from the processing result; Based on the axial position information of the first probe and the defect feature information, a preset warning display method is used in a three-dimensional model corresponding to the target hollow shaft for three-dimensional visualization display; the warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display.

[0006] Optionally, the processing of the first ultrasonic reflection signal information in combination with a three-dimensional image processing technology includes: Arranging the first ultrasonic reflection signal information in time sequence to form a one-dimensional time series signal; Preprocessing the one-dimensional time series signal to obtain a preprocessed one-dimensional time series signal; wherein the preprocessing includes noise removal and normalization processing; The preprocessed one-dimensional time series signal is converted into a frequency domain signal by Fourier transform, and the frequency domain signal is converted into a plurality of three-dimensional images of different resolutions by wavelet transform; each three-dimensional image represents features of different scales; Combining multi-scale feature fusion technology and attention mechanism, respectively performing feature enhancement processing on the multiple three-dimensional images with different resolutions; All the three-dimensional images after feature enhancement are stitched together to form a three-dimensional image signal; Perform image processing on the three-dimensional image signal to obtain a processing result.

[0007] Optionally, performing image processing on the three-dimensional image signal includes: Using a histogram equalization technique to perform contrast enhancement on the three-dimensional image signal, and using a Gaussian filter to perform smoothing on the enhanced three-dimensional image; The target edge detection algorithm is used to perform edge detection on the smoothed 3D image to obtain the outline of the suspected defect; The contour is corroded to remove noise points in the contour and obtain a processing result.

[0008] Optionally, the edge detection of the smoothed three-dimensional image using a target edge detection algorithm to obtain the outline of the suspected defect includes: The Sobel operator is used to calculate the gradient amplitude of each three-dimensional pixel in each direction in the smoothed three-dimensional image; Perform non-maximum suppression on the gradient amplitude to obtain the gradient amplitude after non-maximum suppression; The gradient amplitude after non-maximum suppression is compared with the double threshold, and the initial edge and potential edge preliminarily detected are determined according to the comparison result; Edge connection processing is performed based on the initial edge and the potential edge to obtain the outline of the suspected defect.

[0009] Optionally, the target probe operation parameter information further includes second operation parameter information respectively corresponding to when multiple types of target probes move simultaneously in a local reciprocating manner; Based on the axial position information of the first probe and the defect feature information, a preset warning display mode is used in a three-dimensional model corresponding to the target hollow shaft for three-dimensional visualization display, including: When a suspected defect is detected based on the defect characteristic information, the range of the partial axial direction is determined based on the axial position information of the first probe, and the spiral mode is switched to a local reciprocating mode, so as to control the multi-type target probe to move in a local reciprocating mode on the partial axial direction based on the second operating parameter information, and obtain the second probe position information and the second ultrasonic reflection signal information collected by the target probe; the range of the partial axial direction is [Zp-L, Zp+L], Zp is the axial position corresponding to the detection of the suspected defect, and L is the preset axial length; When the suspected defect is determined to be a target defect based on the second ultrasonic reflection signal information, determining the position information of the target defect based on the second probe position information; Based on the location information of the target defect, a preset warning display method is used in a three-dimensional model corresponding to the target hollow shaft to perform a three-dimensional visual display of the target defect; the warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display.

[0010] Optionally, the second operating parameter information includes adaptive amplitude and dynamic frequency; The calculation formula of the adaptive amplitude is: ; in, is the adaptive amplitude that varies with the local round trip time t, is the initial amplitude, is the attenuation factor, Indicates the location information of the suspected defect, is the gradient of the suspected defect location; The calculation formula of the dynamic frequency is: ; in, is the dynamic frequency that varies with the local reciprocating time t, is the initial frequency, is the adjustment factor, is the gradient of the initial position, which is the position of the probe when switching to the local reciprocating mode.

[0011] Optionally, obtaining target probe operation parameter information corresponding to the type information of the target hollow shaft includes: Get the type information of the target hollow shaft; According to the preset mapping relationship between the hollow shaft type information, the probe type information and the probe operation parameter information, the target probe type information and the target probe operation parameter information corresponding to the target hollow shaft type information are determined.

[0012] Optionally, a mapping relationship between preset hollow shaft type information, probe type information, and probe operation parameter information is established, including: Obtaining flaw detection test samples of different types of hollow shafts under different flaw detection conditions, wherein the flaw detection conditions include: probe type information and probe operation parameter information, and the flaw detection test samples include: ultrasonic reflection signal samples collected by the probe, estimated defect location information, and actual defect location information; For each type of hollow shaft, the parameters of the genetic algorithm are initialized, and the hollow shaft test samples under different flaw detection conditions are used as individuals to generate the initial population; the parameters include population size, crossover probability, mutation probability and maximum number of iterations; after the initial population is generated, the following operations are repeated: Calculate the fitness value of individuals in the population; the corresponding population in the first iteration process is the initial population; the calculation formula of the fitness value is ,in, Is an individual The fitness value of is the actual location information of the defect, is the estimated location information of the defect, is the standard deviation, is the Euclidean distance between the actual position and the estimated position; Perform a selection operation, and based on the crossover probability and the mutation probability, respectively perform a crossover operation and a mutation operation to generate a new generation of population; Until a preset stop condition is met, the preset stop condition is reaching a maximum number of iterations or fitness value convergence; wherein the individual with the highest fitness value is the mapping relationship between the hollow shaft type information, the probe type information and the probe operation parameter information.

[0013] Optionally, the preset warning display mode includes a warning display mode corresponding to the severity; Based on the position information of the target defect, a preset warning display mode is used in a three-dimensional model corresponding to the target hollow shaft to perform three-dimensional visualization of the target defect, including: The cross-sectional area of ​​the target defect is calculated, and based on the cross-sectional area of ​​the target defect, the influence coefficient of the target defect on the load-bearing capacity of the hollow shaft is determined using a coefficient calculation formula, the coefficient calculation formula being: K = a × (A d / A0); Wherein, K is the influence coefficient, a is the correction factor, and A d is the cross-sectional area of ​​the target defect, and A0 is the original cross-sectional area of ​​the hollow shaft without defects; Determining the severity of the target defect according to the influence coefficient of the target defect on the load-bearing capacity of the hollow shaft; Based on the position information of the target defect, the target defect is three-dimensionally visualized in a three-dimensional model corresponding to the target hollow shaft according to a warning display method corresponding to the severity.

[0014] In a second aspect, an embodiment of the present application provides a three-dimensional flaw detection device, comprising: An acquisition module, used for acquiring target probe operation parameter information corresponding to the type information of the target hollow shaft, wherein the target probe operation parameter information includes first operation parameter information corresponding to when multiple types of target probes move in a spiral manner at the same time; a control acquisition module, configured to control the multi-type target probes to move in a spiral manner in the axial direction according to the first operating parameter information, and to acquire the first probe axial position information of the multi-type target probes, and the first ultrasonic reflection signal information respectively collected by the multi-type target probes; A processing and extraction module, used to process the first ultrasonic reflection signal information in combination with a three-dimensional image processing technology, and extract defect feature information from the processing result; A display module is used to perform three-dimensional visual display in a three-dimensional model corresponding to the target hollow shaft using a preset early warning display method based on the axial position information of the first probe and the defect feature information; the early warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display.

[0015] In an embodiment of the present application, a three-dimensional flaw detection method is provided, including: obtaining target probe operating parameter information corresponding to type information of a target hollow shaft, the target probe operating parameter information including first operating parameter information corresponding to when multiple types of target probes move in a spiral manner at the same time; controlling the multiple types of target probes to move in a spiral manner in the axial direction according to the first operating parameter information, and obtaining first probe axial position information of the multiple types of target probes, and first ultrasonic reflection signal information respectively collected by the multiple types of target probes; processing the first ultrasonic reflection signal information in combination with three-dimensional image processing technology, and extracting defect feature information from the processing result; based on the first probe axial position information and the defect feature information, using a preset warning display method to perform three-dimensional visualization display in a three-dimensional model corresponding to the target hollow shaft; the warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display. The embodiment of the present application can cover all areas inside the target hollow shaft by spirally moving multiple types of target probes, and can deeply process the first ultrasonic reflection signal information using three-dimensional image processing technology. It can extract tiny or deeply buried defect features from the complex first ultrasonic reflection signal information, thereby improving detection accuracy. It can use an early warning display method in the three-dimensional model for three-dimensional visualization, thereby improving detection accuracy and reliability.

[0016] These and other aspects of the present application will become more clearly understood in the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 A flow chart of a three-dimensional flaw detection method provided in an embodiment of the present application; Figure 2 A flow chart of another three-dimensional flaw detection method provided in an embodiment of the present application; Figure 3 This is an example of the rendered three-dimensional flaw detection data corresponding to the one-dimensional flaw detection data collected by the oblique probe (+); Figure 4 A schematic diagram of the structure of a three-dimensional flaw detection device provided in an embodiment of the present application; Figure 5 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0020] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as 11, 12, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0022] Figure 1 A flowchart of a three-dimensional flaw detection method provided in an embodiment of the present application is shown in FIG. Figure 1 As shown, the method includes: S11. Obtain target probe operation parameter information corresponding to the type information of the target hollow shaft, where the target probe operation parameter information includes first operation parameter information corresponding to when multiple types of target probes move in a spiral manner at the same time.

[0023] It should be understood that the target hollow shaft can be referred to as the hollow shaft, and the target probe can be referred to as the probe. The type information of the target hollow shaft includes: the model, material, size and other information of the hollow shaft. The target probe operation parameter information includes: the moving speed, frequency, amplitude and other parameters of each target probe. Among them, the frequency can be a dynamic frequency or a fixed frequency, and the amplitude can be an adaptive amplitude or a fixed amplitude. The first operation parameter information is used to control the movement and detection of the target probe during the spiral movement.

[0024] The multi-type target probe includes more than two target probes. Exemplarily, the multi-type target probe includes a dual-chip focusing combination probe, an oblique probe (+) and an oblique probe (-) arranged on the probe rod. Among them, the dual-chip focusing combination probe, the oblique probe (+) and the oblique probe (-) are three types of probes in ultrasonic flaw detection. The dual-chip focusing combination probe includes two independent crystals, one for transmitting ultrasonic waves and the other for receiving ultrasonic waves. The probe design includes a focusing lens, which can focus ultrasonic waves to a specific depth to improve the detection resolution. Due to the dual-chip design, the blind area near the inner surface of the hollow shaft can be reduced, which is suitable for detecting defects close to the inner surface of the hollow shaft. The oblique probe (+) has an oblique angle design, and the incident angle is a positive oblique angle, which is usually used to detect defects perpendicular to the inner surface of the hollow shaft. After entering the material, the ultrasonic wave will change direction due to the refraction effect, so that defects in different directions can be detected. The incident angle of the oblique probe (-) is a negative oblique angle, which is usually used to detect defects parallel to the inner surface of the hollow shaft. The refraction effect is also used, but the direction is opposite, and defects parallel to the inner surface of the hollow shaft can be detected.

[0025] S12. Control the multi-type target probes to move in a spiral manner in the axial direction according to the first operating parameter information, and obtain first probe axial position information of the multi-type target probes and first ultrasonic reflection signal information respectively collected by the multi-type target probes.

[0026] The first ultrasonic reflection signal information refers to the ultrasonic reflection signal collected by the target probe during the spiral movement. Optionally, while acquiring the axial position information of the first probe, this embodiment can also acquire axial angle information to provide data support for subsequent defect position calculation.

[0027] S13. Process the first ultrasonic reflection signal information in combination with three-dimensional image processing technology, and extract defect feature information from the processing result.

[0028] The defect characteristic information refers to the characteristic information about the defect extracted from the first ultrasonic reflection signal, such as the defect position, defect size, defect shape, etc.

[0029] S14. Based on the axial position information of the first probe and the defect feature information, a preset early warning display method is used to perform three-dimensional visualization display in a three-dimensional model corresponding to the target hollow shaft.

[0030] Among them, the warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display. The preset warning display method can be stored in the visualization library. The three-dimensional model corresponding to the target hollow shaft can be a three-dimensional image space. The target defect can be three-dimensionally visualized in the three-dimensional model corresponding to the target hollow shaft. At the same time, this embodiment can also indicate the position, size and shape of the defect in the form of a label. For flaw detection employees, without professional training, by observing the three-dimensional model displayed in three-dimensional visualization, they can quickly view the defects displayed in the preset warning display method, which improves the accuracy of defect recognition and reduces the difficulty of defect recognition.

[0031] In the three-dimensional model, this embodiment can use color coding, arrow indication, transparency level, text label or animation display to intuitively display the location, size and shape of the defect, which can improve the visualization effect of the detection results.

[0032] In practical applications, the target probe can be carried on the probe rod, and for the flaw detection of a hollow shaft, the corresponding probes can be multiple. Taking the number of probes as an example, combined with Figure 2 , the three-dimensional flaw detection method is explained as follows: In this embodiment, the spiral advancement of the probe rod is controlled by a computer control device, and the probe rod is provided with a dual-chip focusing combination probe, an oblique probe (+) and an oblique probe (-). The dual-chip focusing combination probe, the oblique probe (+) and the oblique probe (-) respectively collect data, and in this embodiment, different probe data buffers can be set to store the one-dimensional flaw detection data (i.e., the ultrasonic reflection signal information) collected by the corresponding probes. For example, the one-dimensional flaw detection data collected by the dual-chip focusing combination probe is stored in buffer 1, the one-dimensional flaw detection data collected by the oblique probe (+) is stored in buffer 2, and the one-dimensional flaw detection data collected by the oblique probe (-) is stored in buffer 3. After performing three-dimensional image processing based on one-dimensional flaw detection data, this embodiment can store three-dimensional flaw detection data (i.e., the following processing results) corresponding to each one-dimensional flaw detection data in different probe data buffers. For example, three-dimensional flaw detection data corresponding to one-dimensional flaw detection data collected by the dual-chip focusing combination probe is stored in buffer 4, three-dimensional flaw detection data corresponding to one-dimensional flaw detection data collected by the oblique probe (+) is stored in buffer 5, and three-dimensional flaw detection data corresponding to one-dimensional flaw detection data collected by the oblique probe (-) is stored in buffer 6. Afterwards, this embodiment can call the visualization library to present a three-dimensional image space with three-dimensional flaw detection data added.

[0033] Figure 3 This is an example of the rendering of the 3D flaw detection data corresponding to the 1D flaw detection data collected by the oblique probe (+). The highlighted dotted areas, line areas, and ring areas are all defects.

[0034] For each probe, the distance modeling method in the three-dimensional model uses the following formula: , , .

[0035] in, Represents the position of the coordinates in the three-dimensional image space, The range is [0, full axis length of hollow shaft]; Indicates the displacement value of the probe spiral forward direction, The range is [0, full axis length of hollow shaft]; Indicates the angle of the probe spiral advance, The range of is [0, 360]; Indicates the pitch of the probe spiral advance; Indicates the probe acquisition depth (or sound path). Indicates the wall thickness of the hollow shaft, Indicates the inner radius of the hollow shaft.

[0036] For example, this embodiment can combine the results collected by the probe in real time (such as amplitude, ranging from 0 to 100%) with It is written into the three-dimensional image space to obtain the entire three-dimensional image. When the amplitude exceeds a preset threshold (such as 40%), it is highlighted (for example, displayed in red).

[0037] The embodiment of the present application can quickly cover all areas inside the target hollow shaft by combining the spiral movement and local reciprocating movement of the target probe, reduce unnecessary repeated scanning, and improve the detection speed. By using three-dimensional image processing technology to deeply process the first ultrasonic reflection signal information, tiny or deeply buried defect features can be extracted from the complex first ultrasonic reflection signal information to improve the detection accuracy. When the target defect is detected, a warning display method can be used in the three-dimensional model for three-dimensional visualization to improve the detection accuracy and reliability.

[0038] In the prior art, since different types of hollow shafts have different structures and materials, traditional flaw detection methods often require manual adjustment of the operating parameters of the probe, which is not only time-consuming and labor-intensive, but also prone to detection errors due to human factors. Therefore, this embodiment can automatically obtain the target probe operating parameter information based on the type information of the target hollow shaft to solve this problem. In some optional embodiments, S11, obtaining the target probe operating parameter information corresponding to the type information of the target hollow shaft, includes: Step 111: Obtain type information of the target hollow shaft.

[0039] Step 112: Determine the target probe type information and the target probe operating parameter information corresponding to the target hollow shaft type information according to the preset mapping relationship among the hollow shaft type information, the probe type information and the probe operating parameter information.

[0040] Specifically, in this embodiment, the model, material, size and other information of the target hollow shaft can be input or read through an input device (such as a keyboard, scanner, etc.), and a pre-established mapping relationship can be searched in the database, and the corresponding probe type and operating parameters can be automatically matched according to the input hollow shaft type information.

[0041] For example, the hollow shaft model is A123, made of stainless steel, and has a diameter of 100 mm. After inputting this information through the input device, the system automatically searches in the database and matches three corresponding probe types, P1, P2 and P3, with the operating parameters of: moving speed 5 mm / s, fixed frequency 1 MHz, fixed amplitude 0.5 mm.

[0042] This embodiment automatically obtains the probe operating parameters, reduces the time for manual adjustment, and improves detection efficiency. In addition, through automatic matching, parameter setting errors caused by human factors are avoided, and the accuracy of detection is improved.

[0043] In some optional embodiments, a large number of parameter combinations can be set for each type of hollow shaft. In order to find the optimal solution from a large number of possible parameter combinations, this embodiment can introduce a genetic algorithm. Therefore, before executing step 112, this embodiment can also include the following steps: establishing a mapping relationship corresponding to preset hollow shaft type information, probe type information, and probe operation parameter information. The process of establishing the mapping relationship includes: Step a1, obtaining flaw detection test samples of different types of hollow shafts under different flaw detection conditions, the flaw detection conditions including: probe type information and probe operation parameter information, the flaw detection test samples including: ultrasonic reflection signal samples collected by the probe, estimated defect location information, and actual defect location information.

[0044] Step a2: For each type of hollow shaft, initialize the parameters of the genetic algorithm, and use the hollow shaft test samples under different flaw detection conditions as individuals to generate an initial population; the parameters include population size, crossover probability, mutation probability and maximum number of iterations; after the initial population is generated, repeat the following steps a3 to a4: Step a3: Calculate the fitness value of individuals in the population; the corresponding population in the first iteration is the initial population; the fitness value calculation formula is: ,in, Is an individual The fitness value of is the actual location information of the defect, is the estimated location information of the defect, is the standard deviation, is the Euclidean distance between the actual location and the estimated location Step a4: performing a selection operation, and performing a crossover operation and a mutation operation respectively based on the crossover probability and the mutation probability to generate a new generation population. Optionally, the crossover probability may be an adaptive crossover probability, and the mutation probability may be an adaptive mutation probability.

[0045] Until the preset stopping condition is met, the preset stopping condition is reaching the maximum number of iterations or the fitness value converges; wherein the individual with the highest fitness value is the mapping relationship between the hollow shaft type information, the probe type information and the probe operation parameter information.

[0046] Exemplarily, the population size is 100, the crossover probability is 0.8, the mutation probability is 0.1, and the maximum number of iterations is 100. For the hollow shaft of model A123, the individual with the highest fitness value is selected as the optimal solution. The probe type of the individual is P1, and the operating parameters are a moving speed of 5 mm / s, a frequency of 1 MHz, and an amplitude of 0.5 mm. Through the above steps, a mapping relationship between the hollow shaft of model A123 and the probe type P1 and operating parameters can be established to ensure that the most appropriate probe and operating parameters are selected during the detection process, thereby improving the accuracy and efficiency of the detection.

[0047] Traditional ultrasonic reflection signal processing methods usually directly process one-dimensional time series signals. This processing method is difficult to effectively extract complex defect features, especially when dealing with tiny or deeply buried defects, and is prone to missing important information. In order to improve the detection accuracy, this embodiment can use three-dimensional image processing technology. In some optional embodiments, S13, combining three-dimensional image processing technology to process the first ultrasonic reflection signal information, including: Step 131: Arrange the first ultrasonic reflection signal information in chronological order to form a one-dimensional time series signal.

[0048] Step 132: preprocess the one-dimensional time series signal to obtain a preprocessed one-dimensional time series signal; wherein the preprocessing includes noise removal and normalization processing.

[0049] Step 133: convert the preprocessed one-dimensional time series signal into a frequency domain signal through Fourier transform, and convert the frequency domain signal into multiple three-dimensional images of different resolutions through wavelet transform; each three-dimensional image represents features of different scales.

[0050] Step 134: Combining multi-scale feature fusion technology and attention mechanism, perform feature enhancement processing on multiple three-dimensional images with different resolutions.

[0051] Step 135: splice all the three-dimensional images after feature enhancement processing to form a three-dimensional image signal.

[0052] Step 136: Perform image processing on the three-dimensional image signal to obtain a processing result.

[0053] In this embodiment, through multi-scale images, defect features of different scales can be captured, improving the comprehensiveness of feature extraction. The preprocessing step can effectively remove noise and improve the signal-to-noise ratio of the signal. Combining multi-scale feature fusion technology and attention mechanism can effectively enhance defect features and improve detection accuracy.

[0054] As a possible implementation, step 136, performing image processing on the three-dimensional image signal, includes: Step b1: using a histogram equalization technique to perform contrast enhancement on a three-dimensional image signal, and using a Gaussian filter to perform smoothing on the enhanced three-dimensional image.

[0055] Step b2: Use a target edge detection algorithm to perform edge detection on the smoothed three-dimensional image to obtain the outline of the suspected defect.

[0056] Step b3: perform corrosion processing on the contour to remove noise points in the contour and obtain a processing result.

[0057] In this embodiment, the histogram equalization technology can improve the contrast of the image and make the defect features more obvious. The Gaussian filter can effectively remove the noise in the image and improve the clarity of the image. The corrosion process can remove small noise points and make the defect contour clearer and more accurate.

[0058] Optionally, step b2, using a target edge detection algorithm to perform edge detection on the smoothed three-dimensional image to obtain the outline of the suspected defect, includes: Step b21: Use the Sobel operator to calculate the gradient amplitude of each three-dimensional pixel in each direction in the smoothed three-dimensional image.

[0059] Step b22: perform a non-maximum suppression operation on the gradient amplitude to obtain the gradient amplitude after non-maximum suppression.

[0060] Step b23, comparing the gradient amplitude after non-maximum suppression with the double threshold, and determining the initially detected initial edge and potential edge according to the comparison result.

[0061] The dual threshold includes a first threshold and a second threshold, and the first threshold is greater than the second threshold. Points with gradient amplitudes greater than the first threshold are initial edges, and points with gradient amplitudes between the first threshold and the second threshold are potential edges.

[0062] Step b24: perform edge connection processing based on the initial edge and the potential edge to obtain the outline of the suspected defect.

[0063] This embodiment uses the Sobel operator to perform gradient calculation, combined with non-maximum suppression and dual threshold processing, to effectively improve the accuracy and completeness of edge detection. Through non-maximum suppression and edge connection processing, the continuity of the edge can be enhanced, making the defect contour more complete and clear. In addition, through dual threshold processing and edge connection, a large amount of data can be quickly processed, the efficiency of detection can be improved, and the detection time can be reduced.

[0064] In some other embodiments, the target probe operation parameter information further includes second operation parameter information respectively corresponding to when multiple types of target probes move simultaneously in a local reciprocating manner. The second operation parameter information is used to control the movement and detection of the multiple types of target probes during the local reciprocating process.

[0065] Therefore, S14, based on the axial position information of the first probe and the defect feature information, a preset warning display mode is used in a three-dimensional model corresponding to the target hollow shaft for three-dimensional visualization display, including: Step 141, when a suspected defect is detected based on defect characteristic information, the range of the partial axial direction is determined based on the axial position information of the first probe, and the spiral mode is switched to a local reciprocating mode, so as to control the movement of multiple types of target probes in a local reciprocating manner on the partial axial direction based on the second operating parameter information, and obtain the second probe position information and the second ultrasonic reflection signal information collected by the target probe.

[0066] Among them, the partial axial range is [Zp-L, Zp+L], Zp is the axial position corresponding to when the suspected defect is detected, and L is the preset axial length.

[0067] Step 142: When the suspected defect is determined to be a target defect based on the second ultrasonic reflection signal information, the position information of the target defect is determined based on the second probe position information. The suspected defect may be a real defect or an area misjudged as a defect, and the probability of the target defect being a real defect is higher than the probability of the suspected defect being a real defect.

[0068] Optionally, this embodiment can use computer graphics technology and three-dimensional image processing technology to three-dimensionally process the first ultrasonic reflection signal information and the second ultrasonic reflection signal information, respectively, so as to determine that the suspected defect is a target defect through the three-dimensionally processed first ultrasonic reflection signal information and the second ultrasonic reflection signal information.

[0069] Step 143: Based on the position information of the target defect, a preset warning display method is used to perform a three-dimensional visualization display of the target defect in a three-dimensional model corresponding to the target hollow shaft.

[0070] It should be noted that in some areas inside the hollow shaft, defects may not need too large an amplitude to be detected, while in other areas, a larger amplitude is required. The adaptive amplitude can adjust the amplitude according to actual needs, thereby reducing unnecessary energy waste. In some optional embodiments, the second operating parameter information includes an adaptive amplitude and a dynamic frequency.

[0071] The calculation formula of the adaptive amplitude is: ; in, is the adaptive amplitude that varies with the local round trip time t, is the initial amplitude, is the attenuation factor, Indicates the location information of the suspected defect, is the gradient of the suspected defect location.

[0072] Therefore, this embodiment can reduce unnecessary energy waste, and by adjusting the amplitude, the probe can locate and detect defects more accurately, especially when the defect is small, a larger amplitude can help the probe better capture the second ultrasonic reflection signal information corresponding to the defect signal. In addition, the adaptive amplitude can be dynamically adjusted according to the change of the defect position, which improves the flexibility of detection.

[0073] It should be noted that using a fixed frequency may not be able to effectively detect all types of defects. The dynamic frequency can be adjusted according to actual conditions to meet different detection requirements. The calculation formula for dynamic frequency is: ; in, is the dynamic frequency that varies with the local reciprocating time t, is the initial frequency, is the adjustment factor, is the gradient of the initial position, which is the position of the probe when switching to the local reciprocating mode.

[0074] This embodiment can adjust the dynamic frequency according to the actual situation to meet different detection requirements. In addition, by dynamically adjusting the frequency, the most suitable frequency can be used in different detection stages, thereby improving the detection efficiency. In some cases, fixed-frequency ultrasound may be interfered by the environment. The dynamic frequency can avoid these interferences by adjusting the frequency and improve the stability of the detection.

[0075] Accordingly, this embodiment introduces adaptive amplitude and dynamic frequency to enable the three-dimensional flaw detection method to detect defects inside the hollow shaft more flexibly, efficiently and accurately. Among them, the adaptive amplitude can dynamically adjust the amplitude size according to the change of the suspected defect position, reduce energy waste and improve detection accuracy. The dynamic frequency can adjust the frequency according to the gradient to improve detection efficiency and stability. These designs make the detection process more intelligent and suitable for various complex detection scenarios.

[0076] Traditional three-dimensional flaw detection methods usually only provide simple defect location and size information, lack of defect severity assessment, making it difficult for technicians to accurately judge the impact of defects on equipment performance. In some optional embodiments, the preset warning display method includes a warning display method corresponding to the severity; therefore, step 143, based on the location information of the target defect, uses the preset warning display method in the three-dimensional model corresponding to the target hollow shaft to perform a three-dimensional visualization of the target defect, including: Step c1, calculate the cross-sectional area of ​​the target defect, and based on the cross-sectional area of ​​the target defect, use the coefficient calculation formula to determine the influence coefficient of the target defect on the load-bearing capacity of the hollow shaft. The coefficient calculation formula is: K = a × (A d / A0); Among them, K is the influence coefficient, a is the correction factor, and A d is the cross-sectional area of ​​the target defect, and A0 is the original cross-sectional area of ​​the hollow shaft without defects, that is, the original cross-sectional area of ​​the hollow shaft at the defect position. It should be understood that the correction factor is used to correct the additional influence of different defect shapes on the load-bearing capacity. For example, for the target defect in the shape of a hole, a=3. If the additional influence of different defect shapes on the load-bearing capacity is ignored, a=1.

[0077] Optionally, this embodiment can calculate the cross-sectional area of ​​the target defect according to its shape. By calculating the cross-sectional area and the influence coefficient, the influence of the defect on the equipment performance can be quantitatively evaluated, thereby improving the accuracy of the evaluation.

[0078] Step c2: determining the severity of the target defect according to the influence coefficient of the target defect on the load-bearing capacity of the hollow shaft.

[0079] Among them, this embodiment can classify defects into different severity levels according to the size of the impact coefficient, such as slight, medium, severe, etc. Through the severity classification, the severity of the defect can be clearly defined, which is convenient for technicians and managers to make quick judgments.

[0080] For example, according to the size of the impact coefficient, the defects are divided into different severity levels. It is assumed that the impact coefficient is less than 0.5 for minor, 0.5 to 0.8 for medium, and greater than 0.8 for severe. If the impact coefficient of the defect is 0.7854, it is of medium severity.

[0081] Step c3: Based on the position information of the target defect, the target defect is three-dimensionally visualized in a three-dimensional model corresponding to the target hollow shaft according to an early warning display method corresponding to the severity.

[0082] This embodiment selects a suitable warning display mode, such as color coding, arrow indication, transparency adjustment, text label or animation display, according to the severity of the defect. In the 3D model, the selected warning display mode is used to intuitively display the location, size and severity of the defect. Through rich visual effects, the severity of the defect is clear at a glance, improving the visualization effect. Technicians and managers can quickly understand the nature and impact of the defect and improve work efficiency.

[0083] Figure 4 A schematic diagram of the structure of a three-dimensional flaw detection device provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, the device comprises: The acquisition module 41 is used to acquire target probe operation parameter information corresponding to the type information of the target hollow shaft, wherein the target probe operation parameter information includes first operation parameter information corresponding to when multiple types of target probes move in a spiral manner at the same time.

[0084] The control acquisition module 42 is used to control the multi-type target probe to move in a spiral manner in the axial direction according to the first operating parameter information, and obtain the first probe axial position information of the multi-type target probe and the first ultrasonic reflection signal information collected by the multi-type target probe.

[0085] The processing and extraction module 43 is used to process the first ultrasonic reflection signal information in combination with the three-dimensional image processing technology, and extract defect feature information from the processing result.

[0086] The display module 46 is used to perform three-dimensional visual display in a three-dimensional model corresponding to the target hollow shaft using a preset warning display method based on the axial position information of the first probe and the defect feature information; the warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display.

[0087] Figure 4 The three-dimensional flaw detection device can perform Figure 1 The implementation principle and technical effect of the three-dimensional flaw detection method described in the embodiment are not described in detail. The specific way in which each module and unit performs operations in the three-dimensional flaw detection device in the above embodiment has been described in detail in the embodiment of the method, and will not be described in detail here.

[0088] In one possible design, Figure 4 The three-dimensional flaw detection device of the embodiment shown can be implemented as a computing device, such as Figure 5 As shown, the computing device may include a storage component 51 and a processing component 52 .

[0089] The storage component 51 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 52 .

[0090] The processing component 52 is used to: obtain target probe operating parameter information corresponding to the type information of the target hollow shaft, the target probe operating parameter information includes first operating parameter information corresponding to when multiple types of target probes move in a spiral manner at the same time; control the multiple types of target probes to move in a spiral manner in the axial direction according to the first operating parameter information, and obtain the first probe axial position information of the multiple types of target probes, and the first ultrasonic reflection signal information respectively collected by the multiple types of target probes; process the first ultrasonic reflection signal information in combination with three-dimensional image processing technology, and extract defect feature information from the processing results; based on the first probe axial position information and the defect feature information, use a preset early warning display method in the three-dimensional model corresponding to the target hollow shaft to perform three-dimensional visualization of the target defect; the early warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display.

[0091] The processing component 52 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0092] The storage component 51 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as random access memory (RAM), static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0093] Of course, the computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0094] The input / output interface provides an interface between the processing component and the peripheral interface module, which may be an output device, an input device, etc.

[0095] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0096] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0097] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The three-dimensional flaw detection method of the illustrated embodiment.

[0098] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0099] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A three-dimensional flaw detection method, characterized in that: include: Acquire target probe operation parameter information corresponding to the type information of the target hollow shaft, wherein the target probe operation parameter information includes first operation parameter information respectively corresponding to when multiple types of target probes move in a spiral manner at the same time; Controlling the multi-type target probes to move in a spiral manner in the axial direction according to the first operating parameter information, and acquiring the first probe axial position information of the multi-type target probes, and the first ultrasonic reflection signal information respectively collected by the multi-type target probes; Processing the first ultrasonic reflection signal information in combination with three-dimensional image processing technology, and extracting defect feature information from the processing result; Based on the axial position information of the first probe and the defect feature information, a preset warning display method is used in a three-dimensional model corresponding to the target hollow shaft for three-dimensional visualization display; the warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display.

2. The method according to claim 1, characterized in that The processing of the first ultrasonic reflection signal information in combination with the three-dimensional image processing technology includes: Arranging the first ultrasonic reflection signal information in time sequence to form a one-dimensional time series signal; Preprocessing the one-dimensional time series signal to obtain a preprocessed one-dimensional time series signal; wherein the preprocessing includes noise removal and normalization processing; The preprocessed one-dimensional time series signal is converted into a frequency domain signal by Fourier transform, and the frequency domain signal is converted into a plurality of three-dimensional images of different resolutions by wavelet transform; each three-dimensional image represents features of different scales; Combining multi-scale feature fusion technology and attention mechanism, respectively performing feature enhancement processing on the multiple three-dimensional images with different resolutions; All the three-dimensional images after feature enhancement are stitched together to form a three-dimensional image signal; Perform image processing on the three-dimensional image signal to obtain a processing result.

3. The three-dimensional flaw detection method according to claim 2, characterized in that: The performing image processing on the three-dimensional image signal comprises: Using a histogram equalization technique to perform contrast enhancement on the three-dimensional image signal, and using a Gaussian filter to perform smoothing on the enhanced three-dimensional image; The target edge detection algorithm is used to perform edge detection on the smoothed 3D image to obtain the outline of the suspected defect; The contour is corroded to remove noise points in the contour and obtain a processing result.

4. The three-dimensional flaw detection method according to claim 3, characterized in that: The method of using a target edge detection algorithm to perform edge detection on the smoothed three-dimensional image to obtain the contour of the suspected defect includes: The Sobel operator is used to calculate the gradient amplitude of each three-dimensional pixel in each direction in the smoothed three-dimensional image; Perform non-maximum suppression on the gradient amplitude to obtain the gradient amplitude after non-maximum suppression; The gradient amplitude after non-maximum suppression is compared with the double threshold, and the initial edge and potential edge preliminarily detected are determined according to the comparison result; Edge connection processing is performed based on the initial edge and the potential edge to obtain the outline of the suspected defect.

5. The method according to claim 1, characterized in that The target probe operation parameter information also includes second operation parameter information respectively corresponding to when multiple types of target probes move simultaneously in a local reciprocating manner; Based on the axial position information of the first probe and the defect feature information, a preset warning display mode is used in a three-dimensional model corresponding to the target hollow shaft for three-dimensional visualization display, including: When a suspected defect is detected based on the defect characteristic information, the range of the partial axial direction is determined based on the axial position information of the first probe, and the spiral mode is switched to a local reciprocating mode, so as to control the multi-type target probe to move in a local reciprocating mode on the partial axial direction based on the second operating parameter information, and obtain the second probe position information and the second ultrasonic reflection signal information collected by the target probe; the range of the partial axial direction is [Zp-L, Zp+L], Zp is the axial position corresponding to the detection of the suspected defect, and L is the preset axial length; When the suspected defect is determined to be a target defect based on the second ultrasonic reflection signal information, determining the position information of the target defect based on the second probe position information; Based on the position information of the target defect, a preset early warning display method is used in a three-dimensional model corresponding to the target hollow shaft to perform a three-dimensional visual display of the target defect.

6. The three-dimensional flaw detection method according to claim 5, characterized in that: The second operating parameter information includes adaptive amplitude and dynamic frequency.

7. The three-dimensional flaw detection method according to claim 1, characterized in that: Obtain the target probe operation parameter information corresponding to the target hollow shaft type information, including: Get the type information of the target hollow shaft; According to the preset mapping relationship between the hollow shaft type information, the probe type information and the probe operation parameter information, the target probe type information and the target probe operation parameter information corresponding to the target hollow shaft type information are determined.

8. The three-dimensional flaw detection method according to claim 7, characterized in that: Establish a mapping relationship between preset hollow shaft type information, probe type information, and probe operation parameter information, including: Obtaining flaw detection test samples of different types of hollow shafts under different flaw detection conditions, wherein the flaw detection conditions include: probe type information and probe operation parameter information, and the flaw detection test samples include: ultrasonic reflection signal samples collected by the probe, estimated defect location information, and actual defect location information; For each type of hollow shaft, the parameters of the genetic algorithm are initialized, and the hollow shaft test samples under different flaw detection conditions are used as individuals to generate the initial population; the parameters include population size, crossover probability, mutation probability and maximum number of iterations; after the initial population is generated, the following operations are repeated: Calculate the fitness values ​​of individuals in the population; the corresponding population in the first iteration process is the initial population; Perform a selection operation, and based on the crossover probability and the mutation probability, respectively perform a crossover operation and a mutation operation to generate a new generation of population; Until a preset stop condition is met, the preset stop condition is reaching a maximum number of iterations or fitness value convergence; wherein the individual with the highest fitness value is the mapping relationship between the hollow shaft type information, the probe type information and the probe operation parameter information.

9. The three-dimensional flaw detection method according to claim 5, characterized in that: The preset warning display mode includes a warning display mode corresponding to the severity; Based on the position information of the target defect, a preset warning display mode is used in a three-dimensional model corresponding to the target hollow shaft to perform three-dimensional visualization of the target defect, including: The cross-sectional area of ​​the target defect is calculated, and based on the cross-sectional area of ​​the target defect, the influence coefficient of the target defect on the load-bearing capacity of the hollow shaft is determined using a coefficient calculation formula, the coefficient calculation formula being: K=a×(A d / A0); Wherein, K is the influence coefficient, a is the correction factor, and A d is the cross-sectional area of ​​the target defect, and A0 is the original cross-sectional area of ​​the hollow shaft without defects; Determining the severity of the target defect according to the influence coefficient of the target defect on the load-bearing capacity of the hollow shaft; Based on the position information of the target defect, the target defect is three-dimensionally visualized in a three-dimensional model corresponding to the target hollow shaft according to a warning display method corresponding to the severity.

10. A three-dimensional flaw detection device, characterized in that: include: An acquisition module, used for acquiring target probe operation parameter information corresponding to the type information of the target hollow shaft, wherein the target probe operation parameter information includes first operation parameter information respectively corresponding to when multiple types of target probes move in a spiral manner at the same time; a control acquisition module, configured to control the multi-type target probes to move in a spiral manner in the axial direction according to the first operating parameter information, and to acquire the first probe axial position information of the multi-type target probes, and the first ultrasonic reflection signal information respectively collected by the multi-type target probes; A processing and extraction module, used to process the first ultrasonic reflection signal information in combination with a three-dimensional image processing technology, and extract defect feature information from the processing result; A display module is used to perform three-dimensional visual display in a three-dimensional model corresponding to the target hollow shaft using a preset early warning display method based on the axial position information of the first probe and the defect feature information; the early warning display method includes at least one of the following: color coding, arrow indication, degree of transparency, text label or animation display.

Images