An ultrasonic-based coating thickness measurement method and system

Abstract

The present invention discloses an ultrasonic-based coating thickness measurement method and system, which includes identifying the structure of the workpiece to be measured, obtaining surface curvature data, generating initial coating thickness detection points in areas with large curvature changes, then using an ultrasonic probe to measure the distance to these points, performing time-domain and frequency-domain joint analysis on the echo signals to obtain a thickness matrix containing coating thickness values, confidence levels, and neighborhood correlation parameters, and based on the thickness matrix, dynamically updating the positions of the detection points with the aid of a clustering algorithm, and repeating such operations until the termination conditions of spatial coverage rate difference or the number of detection points are met. By operations such as determining detection points based on the workpiece structure, ultrasonic measurement and echo data analysis, and dynamically updating detection points, the present invention effectively improves the accuracy and efficiency of coating thickness measurement, and is particularly suitable for coating detection of complex workpieces.

Description

Technical Field

[0001] The present invention relates to the technical field of coating thickness measurement, and particularly to a coating thickness measurement method and system based on ultrasonic waves. Background Art

[0002] In the fields of industrial production and quality inspection, the accurate measurement of coating thickness is crucial for ensuring product quality, performance, and service life. Taking the automobile wheel hub in automobile manufacturing as an example, its surface coating not only affects aesthetics but also relates to key properties such as corrosion prevention and wear resistance.

[0003] Currently, there are many deficiencies in traditional coating thickness measurement methods. Some methods adopt a full-scale measurement method, which can obtain comprehensive data, but the efficiency is extremely low. In large-scale production and inspection scenarios, it is time-consuming and laborious, and it is difficult to meet the production rhythm requirements. Although some other methods introduce the idea of measuring at detection points to improve efficiency, there is no scientific basis for determining the detection points. Most of them are selected empirically or at fixed positions, and they cannot accurately fit the complex structural characteristics of the workpiece and the coating thickness change law, resulting in poor accuracy of the measurement results.

[0004] With the increasing complexity of workpiece structures, such as the emergence of components with irregular curved surfaces and special structural characteristics, and the continuous improvement of coating quality requirements, the defects of existing coating thickness measurement technologies in measurement efficiency and accuracy are becoming more and more prominent. There is an urgent need for a coating thickness measurement method that can balance efficiency and accuracy. Summary of the Invention

[0005] In order to solve at least one of the above-mentioned technical problems, the present invention provides a coating thickness measurement method and system based on ultrasonic waves.

[0006] In a first aspect, the present invention provides a coating thickness measurement method based on ultrasonic waves, and the method includes:

[0007] S100. Based on the structure recognition result of the workpiece to be measured, obtain surface curvature data, where the surface curvature data includes the radius of curvature and the curvature change gradient;

[0008] S200. According to the surface curvature data, generate an initial coating thickness detection point set in the region where the curvature change gradient exceeds a preset threshold;

[0009] S300. Use an ultrasonic probe to measure the distance to the initial coating thickness detection points, and obtain the echo signal data of each detection point;

[0010] S400. Perform a time-domain and frequency-domain joint analysis on the echo signal data to generate a thickness matrix including coating thickness values, measurement confidence levels, and neighborhood correlation parameters;

[0011] S500. Dynamically update the detection point positions according to the measurement confidence and neighborhood correlation parameters in the thickness matrix, and form an optimized set of detection points;

[0012] S600. Repeatedly execute steps S300 - S500 until one of the following termination conditions is met:

[0013] a) The difference in the spatial coverage rate of the detection point sets in two consecutive iterations is less than 5%;

[0014] b) The number of detection points reaches a preset threshold associated with the surface complexity.

[0015] Preferably, the S400 includes the following sub - steps:

[0016] S410. Perform wavelet threshold denoising on the ultrasonic echo signal;

[0017] S420. Use the Teager energy operator to detect the wave peaks of the ultrasonic echo signal to identify the leading - edge time of the direct wave and the leading - edge time of the interface reflection wave; calculate the basic value of the coating thickness according to the leading - edge time of the direct wave, the leading - edge time of the interface reflection wave, and the preset sound - speed parameter;

[0018] S430. Perform FFT transformation on the signal within the reflection - wave time window, calculate the energy - spectrum integral value of the preset characteristic frequency band, and correct the basic value of the coating thickness according to the thickness - value trimming model. The thickness - value trimming model is expressed as:

[0019] ,

[0020] In the formula, is the corrected coating - thickness value, is the basic value of the coating thickness, is the energy - spectrum integral value, is the reference energy value of the standard sample, is the material dispersion coefficient;

[0021] S440. Calculate the confidence of the basic value of the coating thickness through the signal - attenuation coefficient of the ultrasonic echo signal. The signal - attenuation coefficient is positively correlated with the echo amplitude;

[0022] S450. Calculate the neighborhood correlation parameter between the current detection point and adjacent points , in the formula, is the standard deviation of the thickness of the current point, is the standard deviation of the thickness of the adjacent point, and ρ is the spatial correlation coefficient;

[0023] S460. Construct a thickness matrix according to the coating - thickness value, the confidence of the basic value of the coating thickness, and the neighborhood correlation parameter.

[0024] Preferably, the clustering algorithm in the step S500 adopts an improved DBSCAN algorithm, and its density parameter ε is adaptively adjusted according to the thickness gradient of adjacent detection points. The adjustment formula is: , where is the initial density parameter, is the curvature sensitivity factor, is the thickness difference between adjacent points, is the average thickness of the current area.

[0025] In a second aspect, the present invention also provides an ultrasonic-based coating thickness measurement system, which includes:

[0026] A structural curvature analysis module, which is used to obtain surface curvature data based on the structural recognition result of the workpiece to be measured, where the surface curvature data includes the radius of curvature and the curvature change gradient;

[0027] An initial coating thickness detection point generation module, which is used to generate a set of initial coating thickness detection points in the area where the curvature change gradient exceeds a preset threshold according to the surface curvature data;

[0028] An ultrasonic ranging module, which is used to measure the distance to the initial coating thickness detection points by using an ultrasonic probe to obtain the echo signal data of each detection point;

[0029] An echo data processing and matrix construction module, which performs time-domain and frequency-domain joint analysis on the echo signal data to generate a thickness matrix including coating thickness values, measurement confidence levels, and neighborhood correlation parameters;

[0030] A detection point optimization module, which dynamically updates the detection point positions according to the measurement confidence levels and neighborhood correlation parameters in the thickness matrix through a clustering algorithm to form an optimized set of detection points;

[0031] An iterative control module, which is used to repeatedly execute the ultrasonic ranging module, the echo data processing and matrix construction module, and the detection point optimization module until one of the following termination conditions is met: the difference in the spatial coverage rate of the detection point sets in two consecutive iterations is less than 5%; the number of detection points reaches a preset surface complexity correlation threshold.

[0032] Preferably, the echo data processing and matrix construction module includes:

[0033] A signal denoising unit, which is used to perform wavelet threshold denoising on the ultrasonic echo signal;

[0034] A wave peak detection and thickness initial calculation unit, which is used to detect the wave peak of the ultrasonic echo signal by using the Teager energy operator to identify the leading edge time of the direct wave and the leading edge time of the interface reflection wave; calculate the basic value of the coating thickness according to the leading edge time of the direct wave, the leading edge time of the interface reflection wave, and the preset sound speed parameter;

[0035] The thickness value refinement unit is used to perform FFT transformation on the signals within the reflection wave time window, calculate the energy spectrum integral value of a preset characteristic frequency band, and correct the basic coating thickness value according to the thickness value refinement model. The thickness value refinement model is expressed as:

[0036] ,

[0037] In the formula, is the corrected coating thickness value, is the basic coating thickness value, is the energy spectrum integral value, is the reference energy value of the standard sample, is the material dispersion coefficient;

[0038] The confidence level evaluation unit is used to calculate the confidence level of the basic coating thickness value through the signal attenuation coefficient of the ultrasonic echo signal, and the signal attenuation coefficient is positively correlated with the echo amplitude;

[0039] The neighborhood parameter calculation unit is used to calculate the neighborhood correlation parameter between the current detection point and adjacent points , in the formula, is the thickness standard deviation of the current point, is the thickness standard deviation of the adjacent point, and ρ is the spatial correlation coefficient;

[0040] The matrix integration and construction unit is used to construct a thickness matrix according to the coating thickness value, the confidence level of the basic coating thickness value, and the neighborhood correlation parameter.

[0041] Preferably, the clustering algorithm in the detection point optimization module adopts an improved DBSCAN algorithm, and its density parameter ε is adaptively adjusted according to the thickness gradient of adjacent detection points. The adjustment formula is: , in the formula, is the initial density parameter, is the curvature sensitivity factor, is the thickness difference between adjacent points, is the average thickness of the current region.

[0042] In a third aspect, the present invention also provides an electronic device, including a processor and a memory. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes the method according to the first aspect and any one of its possible implementation manners as described above.

[0043] Fourthly, the present invention further provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions that, when executed by a processor of an electronic device, cause the processor to execute the method according to the first aspect and any possible implementation manner thereof as described above.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] 1) Based on workpiece structure recognition, the present invention obtains surface curvature data to determine the initial coating thickness detection points, constructs a thickness matrix by combining ultrasonic ranging and echo data processing, and uses a clustering algorithm to dynamically update the detection points. First, determining the detection points according to the surface curvature can accurately focus on the areas where the coating thickness is likely to change, improving the measurement pertinence and accuracy. Second, multi-dimensional processing of the echo data, including confidence calculation, ensures the reliability of the measurement data. Further, the clustering algorithm can perform clustering analysis on the detection points according to the measurement confidence and neighborhood correlation parameters, screen out the detection points with low reliability and move them to the reliable areas, avoiding invalid measurements, and realizing the optimized layout of the detection points. Furthermore, the mechanism of dynamically updating the detection points optimizes the detection process, significantly improving the coating thickness measurement efficiency, especially suitable for complex workpieces such as automobile wheels, and having good application value and competitive advantages in the field of industrial coating detection.

[0046] 2) The present invention adopts an improved DBSCAN algorithm to adaptively adjust the density parameter according to the thickness gradient of adjacent detection points. On the one hand, it can accurately adapt to the coating thickness distribution characteristics of complex workpieces such as automobile wheels. Where the thickness gradient is large, the density parameter is dynamically adjusted to make the clustering more refined, accurately capturing the thickness change area and improving the measurement accuracy. On the other hand, the adaptive adjustment avoids over-dense or over-sparse clustering under fixed parameters, optimizes the layout of the detection points, reduces the invalid detection points, and improves the measurement efficiency. Especially in the industrial detection scenario of complex curved surfaces with uneven coating thickness, it effectively enhances the measurement reliability and flexibility, and has significant technical advantages and application value.

[0047] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and do not limit the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background art, the following will describe the drawings required to be used in the embodiments of the present invention or the background art.

[0049] The drawings herein are incorporated into the specification and constitute a part of this specification. These drawings show embodiments consistent with the present disclosure and are used together with the specification to explain the technical solutions of the present disclosure.

[0050] Figure 1Schematic flowchart of a coating thickness measurement method based on ultrasonic waves provided by an embodiment of the present invention;

[0051] Figure 2 Schematic flowchart of sub - steps of step S400 of a coating thickness measurement method based on ultrasonic waves provided by an embodiment of the present invention;

[0052] Figure 3 Schematic structural diagram of a coating thickness measurement system based on ultrasonic waves provided by an embodiment of the present invention;

[0053] Figure 4 provided by an embodiment of the present invention Figure 3 Schematic structural diagram of a sub - module of the echo data processing and matrix construction module 400 in Detailed implementation manners

[0054] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0055] The mention of "embodiment" in this article means that the specific features, structures or characteristics described in connection with the embodiment may be included in at least one embodiment of the present invention. The phrase appears in various positions in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment mutually exclusive with other embodiments. Those skilled in the art explicitly and implicitly understand that the embodiments described herein can be combined with other embodiments.

[0056] Traditional coating thickness measurement methods have problems such as low efficiency and unscientific determination of detection points, and it is difficult to meet the measurement requirements of complex workpieces and high - quality coatings. The existing measurement ideas based on workpiece structure recognition also need to be improved.

[0057] Please refer to Figure 1 , Figure 1 Schematic flowchart of a coating thickness measurement method based on ultrasonic waves provided by an embodiment of the present invention. As Figure 1 shown, the method includes:

[0058] S100. Based on the structure recognition result of the workpiece to be measured, obtain surface curvature data, where the surface curvature data includes the radius of curvature and the curvature change gradient;

[0059] Use a laser scanner or a 3D scanner to scan the workpiece to be measured to obtain its surface point cloud data; calculate the normal vector of each point in the point cloud data, and then calculate the curvature radius and the curvature change gradient of the surface. The principal curvature method or the Gaussian curvature method can be used for curvature analysis; store the curvature radius and curvature change gradient data in a database. For example, near the hub edge and bolt holes, due to large shape changes, the curvature change gradient is obvious. Record the surface curvature data of these areas to prepare for subsequent determination of detection points.

[0060] S200. Generate an initial coating thickness detection point set in the area where the curvature change gradient exceeds a preset threshold according to the surface curvature data;

[0061] Set a preset threshold for the curvature change gradient according to the coating characteristics and application requirements of the workpiece; traverse the curvature data, identify the areas where the curvature change gradient exceeds the preset threshold, and mark them as initial coating thickness detection points; summarize all the qualified detection points to form an initial coating thickness detection point set. In a possible embodiment, set the preset threshold for the curvature change gradient, such as 0.8. In the areas on the hub surface where the curvature change gradient exceeds this threshold, use a method based on grid division to generate initial coating thickness detection points. In the hub edge area, determine a detection point every 8 mm to ensure that there are enough detection points in the areas with complex shapes and large curvature changes to improve the measurement accuracy.

[0062] S300. Use an ultrasonic probe to measure the distance to the initial coating thickness detection points to obtain the echo signal data of each detection point;

[0063] Select a suitable ultrasonic probe to ensure that its frequency and sensitivity meet the coating measurement requirements; at the position of each initial coating thickness detection point, emit an ultrasonic signal and record the echo signal; collect the echo signals of each detection point to ensure data integrity. In this embodiment, an ultrasonic probe with a frequency of 5 MHz is selected, and its beam angle is 30°. It is suitable for coating measurement of the curved surface structure such as an automotive hub. Connect the ultrasonic probe to a high-precision ultrasonic thickness gauge, and place the probe perpendicular to the hub surface at each initial coating thickness detection point for measurement. To reduce errors, measure each point 3 times, record the echo signal data obtained each time, and take the average value as the final echo signal data of this detection point.

[0064] S400. Perform a time-domain and frequency-domain joint analysis on the echo signal data to generate a thickness matrix including coating thickness values, measurement confidence levels, and neighborhood correlation parameters;

[0065] Please refer to Figure 2 , Figure 2 which is a schematic flowchart of the sub-steps of step S400 of an ultrasonic-based coating thickness measurement method provided by an embodiment of the present invention. AsFigure 2 As shown in

[0066] Preferably, S400 includes the following sub-steps:

[0067] S410. Perform wavelet threshold denoising on the ultrasonic echo signal;

[0068] Select the sym8 wavelet basis function to perform wavelet transform on the echo signal, and determine the threshold using the soft threshold method according to the signal characteristics. After wavelet threshold denoising, the noise in the signal is removed, and the signal quality is improved, providing a guarantee for subsequent accurate analysis.

[0069] S420. Use the Teager energy operator to detect the wave peaks of the ultrasonic echo signal to identify the leading edge time of the direct wave and the leading edge time of the interface reflection wave; calculate the basic coating thickness value based on the leading edge time of the direct wave, the leading edge time of the interface reflection wave, and the preset sound velocity parameter;

[0070] Use the Teager energy operator to detect the wave peaks of the denoised echo signal and identify the leading edge time of the direct wave and the leading edge time of the interface reflection wave . Given the sound velocity of the ultrasonic wave in the coating material , according to the formula calculate the basic coating thickness value. If , , then .

[0071] S430. Perform FFT transform on the signal within the reflection wave time window, calculate the energy spectrum integral value of the preset characteristic frequency band, and correct the basic coating thickness value according to the thickness value trimming model, where the thickness value trimming model is expressed as:

[0072] ,

[0073] In the formula, is the corrected coating thickness value, is the basic coating thickness value, is the energy spectrum integral value, is the reference energy value of the standard sample, is the material dispersion coefficient;

[0074] Perform fast Fourier transform (FFT) on the signal within the reflection wave time window and calculate the energy spectrum integral value of the preset characteristic frequency band (2 - 4 MHz) . Obtain the reference energy value of the standard sample by measuring the standard sample. Given the dispersion coefficient of the coating material, correct the basic coating thickness value according to the thickness value trimming model . In this embodiment, , , then .

[0075] S440. Calculate the confidence level of the basic value of the coating thickness through the signal attenuation coefficient of the ultrasonic echo signal, where the signal attenuation coefficient is positively correlated with the echo amplitude;

[0076] Use an ultrasonic thickness gauge to collect the direct wave signal amplitude at each initial coating thickness detection point of the automotive wheel hub and the reflected wave signal amplitude . According to the formula calculate the signal attenuation rate , for example, at a certain detection point, it is measured that = 12, = 5, then: . Before measurement, measure a standard sample with the same coating material and the same process as the automotive wheel hub multiple times under the calibration working conditions to determine the ideal attenuation rate . According to the formula calculate the confidence level . Substitute = 12, = 5 into: . If the calculated , correct it to 1; if , correct it to 0. The final confidence level of this detection point takes the value of .

[0077] Evaluating the measurement reliability using the signal attenuation rate and calculating the confidence level can more accurately reflect the reliability degree of the measurement results at the detection points, provide a more accurate basis for subsequent construction of the thickness matrix and dynamic update of the detection points, and improve the accuracy and reliability of the automotive wheel hub coating thickness measurement.

[0078] S450. Calculate the neighborhood correlation parameter between the current detection point and the adjacent point , where in the formula, is the standard deviation of the thickness of the current point, is the standard deviation of the thickness of the adjacent point, and ρ is the spatial correlation coefficient;

[0079] Calculate the neighborhood correlation parameter between the current detection point and the adjacent point. Assume that the standard deviation of the thickness of the current point , the standard deviation of the thickness of the adjacent point , and the spatial correlation coefficient ρ = 0.7. According to the formula calculate the neighborhood correlation parameter matrix as .

[0080] S460. Construct a thickness matrix based on the coating thickness value, the confidence level of the basic coating thickness value, and the neighborhood correlation parameter. ​

[0081] Construct a thickness matrix based on the coating thickness value, confidence level, and neighborhood correlation parameter. For a certain detection point, the coating thickness value is 6.272 mm, and the confidence level is , and the neighborhood correlation parameter matrix is , then the thickness matrix of this detection point is .

[0082] S500. According to the measurement confidence level and neighborhood correlation parameter in the thickness matrix, dynamically update the positions of the detection points through a clustering algorithm to form an optimized set of detection points;

[0083] Apply the DBSCAN clustering algorithm and input multiple thickness matrices into it. The algorithm performs clustering analysis on the detection points according to the measurement confidence level and neighborhood correlation parameter. For the detection points with a confidence level lower than 0.4 and abnormal neighborhood correlation parameters, move them to the area with a high confidence level and reasonable neighborhood correlation parameters, such as moving 3 mm, to form an optimized set of detection points.

[0084] S600. Repeat steps S300 - S500 until one of the following termination conditions is met:

[0085] The difference in the spatial coverage rate of the detection point sets between two consecutive iterations is less than 5%;

[0086] The number of detection points reaches the preset threshold associated with the surface complexity of the curved surface.

[0087] Repeat steps S3 - S5, and calculate the difference in the spatial coverage rate of the detection point sets and the number of detection points after each iteration. The spatial coverage rate of the detection point set in the first iteration is 75%, and in the second iteration is 78%, with a difference of 3%. When the difference in the spatial coverage rate of the detection point sets between two consecutive iterations is less than 5%, or the number of detection points reaches the threshold preset according to the hub surface complexity (such as preset to 50 detection points), stop the iteration and complete the measurement of the coating thickness on the surface of the automotive hub.

[0088] In this embodiment, surface curvature data is obtained based on workpiece structure recognition to determine the initial coating thickness detection points. The thickness matrix is constructed by combining ultrasonic ranging and echo data processing, and the clustering algorithm is used to dynamically update the detection points. First, determining the detection points based on the surface curvature can accurately focus on the areas where the coating thickness is likely to change, improving the measurement pertinence and accuracy; second, multi-dimensional processing of the echo data including confidence level calculation ensures the reliability of the measurement data; further, the clustering algorithm can perform clustering analysis on the detection points according to the measurement confidence level and neighborhood correlation parameter, screen out the detection points with low reliability and move them to the reliable area, avoiding invalid measurements and realizing the optimized layout of the detection points; furthermore, the mechanism of dynamically updating the detection points optimizes the detection process, significantly improving the coating thickness measurement efficiency, especially suitable for complex workpieces such as automotive hubs, and having good application value and competitive advantages in the field of industrial coating detection.

[0089] Preferably, the clustering algorithm in step S500 adopts an improved DBSCAN algorithm, and its density parameter ε is adaptively adjusted according to the thickness gradient of adjacent detection points. The adjustment formula is: , where is the initial density parameter, is the curvature sensitivity factor, is the thickness difference between adjacent points, is the average thickness of the current area.

[0090] After completing the construction of the thickness matrix in step S460, the thickness matrix of each initial coating thickness detection point has been obtained at this time, including information such as coating thickness values, confidence levels, and neighborhood correlation parameters. Taking a group of detection points on an automobile wheel hub as an example, assuming there are 10 detection points in total, the thickness matrix information is as follows:

[0091] Detection point number Coating thickness value (mm) Confidence level Neighborhood correlation parameter matrix

[0092] 1 5.8 0.9 ;

[0093] 2 6.0 0.85 ; ... ... ... ... ;

[0095] 10 5.6 0.92 ;

[0096] Set the initial density parameter , and the curvature sensitivity factor . Calculate the average thickness of the current area . After adding up the coating thickness values of the 10 detection points and dividing by 10, assume that 5.8 mm is obtained.

[0097] For detection point 1 and detection point 2, the coating thickness value of detection point 1 is 5.8 mm, and the coating thickness value of detection point 2 is 6.0 mm, then the thickness difference between adjacent points is 0.2 mm. According to the formula calculate the density parameter corresponding to detection point 1. Similarly, calculate the thickness differences between other adjacent detection points in turn, and calculate their respective corresponding density parameters according to the formula .

[0098] Input the thickness matrix of each detection point and the corresponding density parameter after adaptive adjustment into the improved DBSCAN algorithm. Taking detection point 1 as an example, the algorithm takes detection point 1 as the core point, and in its Search for other points within the neighborhood with a radius. Suppose detection point 2 and detection point 3 are found within this neighborhood, and the confidence levels of these three points are all higher than the set threshold (such as 0.8), and the neighborhood association parameters also meet certain conditions, then they are divided into one cluster. For detection points with a confidence level lower than the threshold or abnormal neighborhood association parameters, such as the confidence level of detection point 4 being 0.75, lower than 0.8, the algorithm marks it as a noise point and moves it towards the area with a high confidence level and reasonable neighborhood association parameters according to the positions of reliable detection points in the surrounding clusters. Suppose the moving distance is 2 mm, forming an optimized set of detection points.

[0099] After completing the clustering analysis and adjustment of the detection points, enter the subsequent steps, such as repeating steps S3 - S5 (that is, performing ultrasonic ranging, echo signal processing, constructing the thickness matrix, and dynamically updating the detection points again), until the termination condition is met (such as the difference in the spatial coverage rate of the detection point sets in two consecutive iterations is less than 5%, or the number of detection points reaches the threshold preset according to the complexity of the hub surface), and complete the measurement of the coating thickness on the surface of the automotive hub.

[0100] In this embodiment, an improved DBSCAN algorithm is adopted, and the density parameter is adaptively adjusted according to the thickness gradient of adjacent detection points. On the one hand, it can accurately adapt to the coating thickness distribution characteristics of complex workpieces such as automotive hubs. Where the thickness gradient is large, the density parameter is dynamically adjusted to make the clustering more refined, accurately capturing the thickness change area and improving the measurement accuracy; on the other hand, the adaptive adjustment avoids over - clustering or under - clustering under fixed parameters, optimizes the layout of detection points, reduces invalid detection points, and improves the measurement efficiency. Especially in industrial detection scenarios with complex surfaces and uneven coating thicknesses, it effectively enhances the measurement reliability and flexibility, and has significant technical advantages and application values.

[0101] In summary, the method provided in this embodiment can at least achieve the following effects:

[0102] 1) Based on workpiece structure recognition, the present invention obtains surface curvature data to determine the initial coating thickness detection points, constructs a thickness matrix by combining ultrasonic ranging and echo data processing, and uses a clustering algorithm to dynamically update the detection points. First, determining the detection points based on the surface curvature can accurately focus on the areas where the coating thickness is likely to change, improving the measurement pertinence and accuracy; second, performing multi - dimensional processing on the echo data including confidence level calculation to ensure the reliability of the measurement data; further, the clustering algorithm can perform clustering analysis on the detection points according to the measurement confidence level and neighborhood association parameters, screen out the detection points with low reliability and move them to the reliable area, avoiding invalid measurements, and realizing the optimized layout of the detection points; furthermore, the mechanism of dynamically updating the detection points optimizes the detection process, significantly improving the coating thickness measurement efficiency, especially suitable for complex workpieces such as automotive hubs, and having good application value and competitive advantages in the field of industrial coating detection.

[0103] 2) The present invention adopts an improved DBSCAN algorithm to adaptively adjust the density parameter according to the thickness gradient of adjacent detection points. On the one hand, it can accurately adapt to the coating thickness distribution characteristics of complex workpieces such as automobile wheels. Where the thickness gradient is large, the density parameter is dynamically adjusted to make the clustering more refined, accurately capture the thickness change area, and improve the measurement accuracy. On the other hand, the adaptive adjustment avoids over-dense or over-sparse clustering under fixed parameters, optimizes the detection point layout, reduces invalid detection points, and improves the measurement efficiency. Especially in industrial detection scenarios with complex curved surfaces and uneven coating thicknesses, it effectively enhances the measurement reliability and flexibility, and has significant technical advantages and application values.

[0104] See Figure 3 , in one embodiment, there is also provided an ultrasonic-based coating thickness measurement system, and the system includes:

[0105] A structural curvature analysis module 100, configured to obtain surface curvature data based on the structural recognition result of the workpiece to be measured, where the surface curvature data includes a radius of curvature and a curvature change gradient;

[0106] An initial coating thickness detection point generation module 200, configured to generate a set of initial coating thickness detection points in an area where the curvature change gradient exceeds a preset threshold according to the surface curvature data;

[0107] An ultrasonic ranging module 300, configured to measure the distance to the initial coating thickness detection points by using an ultrasonic probe to obtain echo signal data of each detection point;

[0108] An echo data processing and matrix construction module 400, configured to perform time-domain and frequency-domain joint analysis on the echo signal data to generate a thickness matrix including coating thickness values, measurement confidence levels, and neighborhood correlation parameters;

[0109] A detection point optimization module 500, configured to dynamically update the detection point positions according to the measurement confidence level and neighborhood correlation parameters in the thickness matrix through a clustering algorithm to form an optimized set of detection points;

[0110] An iteration control module 600, configured to repeatedly execute the ultrasonic ranging module, the echo data processing and matrix construction module, and the detection point optimization module until one of the following termination conditions is met: the difference in the spatial coverage rate of the detection point sets in two consecutive iterations is less than 5%; the number of detection points reaches a preset surface complexity correlation threshold.

[0111] See Figure 4 , Figure 4 provided by the embodiment of the present invention Figure 3 is a schematic structural diagram of a sub-module of the echo data processing and matrix construction module 400 in

[0112] Preferably, the echo data processing and matrix construction module 400 includes:

[0113] The signal noise reduction unit 410 is used to perform wavelet threshold noise reduction on the ultrasonic echo signal;

[0114] The wave peak detection and initial thickness calculation unit 420 is used to detect the wave peak of the ultrasonic echo signal by using the Teager energy operator to identify the arrival wave front time and the interface reflection wave front time; calculate the basic value of the coating thickness according to the arrival wave front time, the interface reflection wave front time and the preset sound velocity parameter;

[0115] The thickness value refinement unit 430 is used to perform FFT transformation on the signal within the reflection wave time window, calculate the energy spectrum integral value of the preset characteristic frequency band, and correct the basic value of the coating thickness according to the thickness value refinement model. The thickness value refinement model is expressed as:

[0116] ,

[0117] In the formula, is the corrected coating thickness value, is the basic value of the coating thickness, is the energy spectrum integral value, is the reference energy value of the standard sample, is the material dispersion coefficient;

[0118] The confidence evaluation unit 440 is used to calculate the confidence of the basic value of the coating thickness through the signal attenuation coefficient of the ultrasonic echo signal, and the signal attenuation coefficient is positively correlated with the echo amplitude;

[0119] The neighborhood parameter calculation unit 450 is used to calculate the neighborhood correlation parameter between the current detection point and the adjacent points , in the formula, is the thickness standard deviation of the current point, is the thickness standard deviation of the adjacent point, and ρ is the spatial correlation coefficient;

[0120] The matrix integration and construction unit 460 is used to construct a thickness matrix according to the coating thickness value, the confidence of the basic value of the coating thickness and the neighborhood correlation parameter.

[0121] Preferably, the clustering algorithm in the detection point optimization module 500 adopts an improved DBSCAN algorithm, and its density parameter ε is adaptively adjusted according to the thickness gradient of adjacent detection points. The adjustment formula is: , in the formula, is the initial density parameter, is the curvature sensitivity factor, is the thickness difference between adjacent points, is the average thickness of the current area.

[0122] It can be understood that the functions or modules included in the system provided in this embodiment can be used to execute the methods described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments. For the sake of brevity, it will not be repeated here.

[0123] The present invention also provides an electronic device, including a processor and a memory. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes the method in any of the above possible implementation manners.

[0124] The present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions. When the program instructions are executed by the processor of an electronic device, the processor is caused to execute the method in any of the above possible implementation manners.

[0125] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0126] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, device, and unit can refer to the corresponding processes in the foregoing method embodiments and will not be repeated here. Those skilled in the art can also clearly understand that each embodiment of the present invention has its own emphasis. For the convenience and brevity of description, the same or similar parts may not be repeated in different embodiments. Therefore, the parts not described or not described in detail in a certain embodiment can be referred to the records of other embodiments.

Claims

1. A coating thickness measurement method based on ultrasound, characterized in that: The method comprises: S100, acquiring surface curvature data based on a structure recognition result of the workpiece to be measured, wherein the surface curvature data includes a curvature radius and a curvature change gradient; S200, generating an initial coating thickness detection point set in an area where a curvature change gradient exceeds a preset threshold value according to the surface curvature data; S300, using an ultrasonic probe to measure the distance of the initial coating thickness detection point to obtain echo signal data of each detection point; S400, performing a time domain-frequency domain joint analysis on the echo signal data to generate a thickness matrix including coating thickness values, measurement confidence, and neighborhood correlation parameters; S500, dynamically updating the detection point positions through a clustering algorithm according to the measurement confidence and neighborhood association parameters in the thickness matrix to form an optimized detection point set; S600, repeatedly execute S300-S500 until one of the following termination conditions is met: a) The difference in spatial coverage of the detection point set between two consecutive iterations is less than 5%; b) The number of detection points reaches the preset surface complexity correlation threshold.

2. The method for measuring coating thickness based on ultrasound according to claim 1, characterized in that: The S400 includes the following sub-steps: S410, performing wavelet threshold noise reduction on the ultrasonic echo signal; S420, using the Teager energy operator to perform wave peak detection on the ultrasonic echo signal to identify the direct wave front time and the interface reflection wave front time; calculating the coating thickness base value according to the direct wave front time, the interface reflection wave front time and the preset sound velocity parameter; S430, perform FFT transformation on the signal in the reflected wave time window, calculate the energy spectrum integral value of the preset characteristic frequency band, and correct the coating thickness base value according to the thickness value correction model. The thickness value correction model is expressed as: ， In the formula, is the corrected coating thickness value, is the basic value of coating thickness, is the energy spectrum integral value, is the reference energy value of the standard sample, is the material dispersion coefficient; S440, calculating the confidence level of the coating thickness base value by the signal attenuation coefficient of the ultrasonic echo signal, wherein the signal attenuation coefficient is positively correlated with the echo amplitude; S450, calculating the neighborhood association parameters between the current detection point and the adjacent points , where is the standard deviation of thickness at the current point, is the standard deviation of thickness at adjacent points, ρ is the spatial correlation coefficient; S460, constructing a thickness matrix according to the coating thickness value, the confidence of the coating thickness base value and the neighborhood association parameter.

3. The method for measuring coating thickness based on ultrasound according to claim 1, characterized in that: The clustering algorithm in S500 adopts an improved DBSCAN algorithm, and its density parameter ε is adaptively adjusted according to the thickness gradient of adjacent detection points. The adjustment formula is: , where is the initial density parameter, is the curvature sensitivity factor, is the thickness difference between adjacent points, is the average thickness of the current area.

4. A coating thickness measurement system based on ultrasound, characterized in that: The system comprises: A structural curvature analysis module, used to obtain surface curvature data based on the structural recognition result of the workpiece to be measured, wherein the surface curvature data includes a curvature radius and a curvature change gradient; An initial coating thickness detection point generation module is used to generate an initial coating thickness detection point set in an area where the curvature change gradient exceeds a preset threshold value according to the surface curvature data; An ultrasonic distance measurement module, used to measure the distance of the initial coating thickness detection points using an ultrasonic probe to obtain echo signal data of each detection point; The echo data processing and matrix construction module performs a time domain-frequency domain joint analysis on the echo signal data to generate a thickness matrix including coating thickness values, measurement confidence and neighborhood correlation parameters; A detection point optimization module dynamically updates the detection point positions through a clustering algorithm according to the measurement confidence and neighborhood association parameters in the thickness matrix to form an optimized detection point set; The iterative control module is used to repeatedly execute the ultrasonic ranging module, the echo data processing and matrix construction module and the detection point optimization module until one of the following termination conditions is met: the difference in the spatial coverage rate of the detection point set of two consecutive iterations is less than 5%; the number of detection points reaches a preset surface complexity association threshold.

5. The ultrasonic-based coating thickness measurement system according to claim 4, characterized in that: The echo data processing and matrix construction module includes: A signal noise reduction unit, used for performing wavelet threshold noise reduction on the ultrasonic echo signal; The peak detection and thickness preliminary calculation unit is used to detect the peak of the ultrasonic echo signal using the Teager energy operator to identify the direct wave front time and the interface reflection wave front time; the basic value of the coating thickness is calculated according to the direct wave front time, the interface reflection wave front time and the preset sound velocity parameters; The thickness value refinement unit is used to perform FFT transformation on the signal in the reflection wave time window, calculate the energy spectrum integral value of the preset characteristic frequency band, and correct the basic value of the coating thickness according to the thickness value correction model. The thickness value correction model is expressed as: ， In the formula, is the corrected coating thickness value, is the basic value of coating thickness, is the energy spectrum integral value, is the reference energy value of the standard sample, is the material dispersion coefficient; A confidence evaluation unit, used to calculate the confidence of the coating thickness base value by means of a signal attenuation coefficient of the ultrasonic echo signal, wherein the signal attenuation coefficient is positively correlated with the echo amplitude; Neighborhood parameter calculation unit, used to calculate the neighborhood association parameters between the current detection point and the adjacent points , where is the standard deviation of thickness at the current point, is the standard deviation of thickness at adjacent points, ρ is the spatial correlation coefficient; The matrix integration construction unit is used to construct a thickness matrix according to the coating thickness value, the confidence of the coating thickness base value and the neighborhood correlation parameter.

6. The ultrasonic-based coating thickness measurement system according to claim 4, characterized in that: The clustering algorithm in the detection point optimization module adopts the improved DBSCAN algorithm, and its density parameter ε is adaptively adjusted according to the thickness gradient of adjacent detection points. The adjustment formula is: , where is the initial density parameter, is the curvature sensitivity factor, is the thickness difference between adjacent points, is the average thickness of the current area.

7. An electronic device, characterized in that: include: A processor and a memory, wherein the memory is used to store computer program codes, wherein the computer program codes include computer instructions. When the processor executes the computer instructions, the electronic device executes the ultrasonic-based coating thickness measurement method according to any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes the ultrasonic-based coating thickness measurement method according to any one of claims 1 to 3.

Images