Method and device for detecting tensile property of fastener

Abstract

The invention discloses a method and device for detecting the tensile property of a fastener, and the method comprises the steps: obtaining the attribute parameters of a to-be-detected fastener, and installing the to-be-detected fastener in a clamping mechanism of the tensile property detection device; starting the tensile property detection device, carrying out stress stretching on the fastener to be tested according to a preset stretching rate, carrying out synchronous processing on collected stretching stress data and deformation image data, then carrying out noise removal and data smoothing processing on the stress data, and carrying out image preprocessing on the deformation image data; the method comprises the following steps of: analyzing deformation image data, extracting key deformation characteristics, constructing a stress-deformation model, analyzing to obtain a critical deformation point, judging an abnormal deformation point in a tensile test process, calculating to obtain a maximum stress value through the critical deformation point, and correcting the maximum stress value by adopting the abnormal deformation point, so as to obtain a tensile test result. The tensile property index of the fastener is obtained; and the accuracy and the reliability of the tensile property detection of the fastener are greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of performance detection, and particularly relates to a method and device for detecting the tensile resistance performance of fasteners. Background Art

[0002] In the aerospace field, the requirements for structural components and connecting parts are extremely strict. Among them, fasteners, as important elements for connecting and fixing key components, their tensile resistance performance is directly related to the safety and reliability of the overall structure. In aerospace vehicles such as airplanes, satellites, and rockets, fasteners not only need to withstand high-intensity tensile loads but also must adapt to complex environments such as extreme temperatures, vibrations, and impacts. Therefore, the aerospace field has put forward higher requirements for the materials, processes, and performance detection standards of fasteners, and there is an urgent need to develop a high-precision detection technology that can not only accurately capture the stress state of fasteners but also monitor the deformation details in real time.

[0003] Currently, the commonly used fastener tensile resistance testing technologies mainly rely on traditional mechanical sensors or strain gauges. By applying a tensile load and recording the curve of force and deformation, the performance of fasteners is evaluated. Although this method can reflect the tensile resistance ability of fasteners macroscopically, relying solely on single mechanical data, it is often unable to accurately capture the micro-deformations and local damages on the surface and inside of fasteners during the tensile process. Especially in application scenarios with extremely high precision requirements in the aerospace field, traditional technologies have obvious deficiencies in identifying key critical deformation points and abnormal deformation signals.

[0004] In view of this, it is necessary to improve the fastener tensile resistance testing technology in the existing technology to solve the technical problem that it is difficult to meet the high-precision testing requirements. Summary of the Invention

[0005] The purpose of the present invention is to provide a method and device for detecting the tensile resistance performance of fasteners to solve the above technical problems.

[0006] To achieve this purpose, the present invention adopts the following technical solutions: A device for detecting the tensile resistance performance of fasteners, comprising: Obtain the attribute parameters of the fastener to be tested, and install the fastener to be tested in the clamping mechanism of the tensile resistance performance detection device; the tensile resistance performance detection device is configured with a depth camera and a mechanical sensor; Start the tensile resistance performance detection device, apply a force to stretch the fastener to be tested at a preset stretching rate. During the stretching process, the mechanical sensor continuously records the force data of the sample, and at the same time, the depth camera captures the deformation image data of the sample surface; During the test, the collected tensile force data and deformation image data will be processed synchronously. After that, noise removal and data smoothing will be performed on the force data, and image preprocessing will be performed on the deformation image data; By analyzing the deformation image data, key deformation features are extracted, and combined with the force data during the stretching process, a force-deformation model is constructed; Based on the force-deformation model, the critical deformation point is analyzed, and the abnormal deformation points during the tensile test are judged. The maximum force value is obtained through the operation of the critical deformation point, and the abnormal deformation points are used to correct the maximum force value to obtain the corrected maximum force value; Based on the corrected maximum force value, the anti-tensile performance index of the fastener is obtained.

[0007] Optionally, both the mechanical sensor and the depth camera are attached with timestamps generated by a unified system clock; during the test, the synchronous processing of the collected tensile force data and deformation image data specifically includes: According to the timestamps attached to each data point, the mechanical data and the image data are initially matched according to a preset sampling period; For the case where there are differences in the sampling frequencies, linear interpolation or spline interpolation algorithms are used to adjust the data sequences so that one-to-one corresponding data pairs are formed on the same time scale; The formed synchronous data pairs are stored in the cache module to provide unified time-sequenced cached data.

[0008] Optionally, the subsequent noise removal and data smoothing processing of the force data and the image preprocessing of the deformation image data specifically include: A low-pass filter is used to initially reduce the noise of the original force data and suppress the high-frequency noise components; The moving average filtering or median filtering algorithm is applied to perform secondary smoothing processing on the force data to eliminate instantaneous abnormal fluctuations; The internal and external parameters obtained during the pre-calibration of the camera are used to correct the geometric distortion of the deformation image data; For the original image noise in the deformation image data, Gaussian filtering is used with a filter kernel size of 5×5 and a standard deviation set to 1.0 for noise suppression; The image is subjected to brightness normalization through histogram equalization technology to enhance the contrast and detail performance of the deformation area in the image; According to the preset region of interest degree for the test requirements, the image is cropped to retain the image data of the key surface area of the fastener.

[0009] Optionally, after the image is cropped according to the preset region of interest degree for the test requirements to retain the image data of the key surface area of the fastener, the following further includes: Uniformly store the obtained force data and deformation image data after processing according to the time series, and construct a corresponding data index database.

[0010] Optionally, by analyzing the deformation image data, extract key deformation features, and combine the force data during the stretching process to construct a force-deformation model, specifically including: Perform morphological operations including erosion and dilation on the preprocessed deformation image data, and the operation size can be set as a convolution kernel of 3×3 or 5×5; Apply an image segmentation algorithm, such as threshold-based segmentation, to segment the image into multiple regions to highlight the regions where deformation occurs on the sample surface; Extract key deformation features from the segmented deformation regions, and the key deformation features include the area, deformation rate, and deformation mode of the deformation region; Through multi-scale analysis, track the deformation at different stretching stages to form deformation sequence data.

[0011] Optionally, after tracking the deformation at different stretching stages through multi-scale analysis to form deformation sequence data, it further includes: Match the force data after synchronous processing with the key deformation features one by one according to the time stamp, so that each force data point corresponds to a deformation state; Adopt regression analysis method to analyze the quantitative relationship between force and deformation features, and identify the non-linear relationship between force and deformation; Extract key points during the stretching process, including initial force, elastic stage, yield point, and maximum force point, and combine the turning points of deformation to construct a mechanical model describing the tensile resistance performance; Adopt the finite element analysis method to establish a stress-deformation model to simulate the deformation mode and stress distribution of the fastener under different force conditions; Verify and correct the stress-deformation model through actual test data and the constructed mechanical model.

[0012] Optionally, based on the force-deformation model, analyze and obtain the critical deformation point, and judge the abnormal deformation point during the tensile test, specifically including: According to the force-deformation model, first calculate the deformation rate and force change of the fastener sample at each stage during the entire stretching process; Smooth the force-deformation curve through the local smoothing algorithm; Use the force-deformation curve and combine the second derivative method to identify the critical point, that is, by calculating the second derivative of the curve to find the inflection point of the curve; Compare all the calculated deformation points in the force-deformation model with the actual test data, and identify the abnormal deformation points that are significantly different from most of the data by calculating the standard deviation of the deformation or based on the maximum difference method; For the identified abnormal deformation points, use an outlier detection algorithm to further verify whether the point belongs to abnormal data caused by external interference or equipment failure during the test; If the abnormal deformation point is valid data, keep it; otherwise, the abnormal point will be marked as invalid data and excluded from the calculation.

[0013] Optionally, the maximum force value is obtained by critical deformation point operation, and the maximum force value is corrected by the abnormal deformation point to obtain a corrected maximum force value, which specifically includes: Based on the force-deformation model, first determine the maximum force value at the critical deformation point; According to the identified abnormal deformation points, adjust the maximum force value through a data correction algorithm; Use a weighted average algorithm to combine the critical point and the data after removing the abnormal points to correct the maximum force value calculated at the critical deformation point to obtain a corrected maximum force value.

[0014] The present invention also provides a tensile resistance detection device for fasteners, which is characterized in that it is applied to implement the tensile resistance detection method for fasteners as described above, and the tensile resistance detection device includes: A clamping mechanism for precisely fixing the fastener to be tested; A tensile testing machine for providing a tensile force; A mechanical sensor for real-time recording of the tensile force and strain data of the fastener during the tensile process; A depth camera for real-time capturing of the deformation image on the surface of the fastener to provide deformation data.

[0015] Compared with the prior art, the present invention has the following beneficial effects: First, the attribute parameters of the fastener to be measured are collected first, and it is installed in a detection device equipped with a depth camera and a mechanical sensor. The detection device is started at a preset stretching rate to perform tensile loading on the sample, and the force data recorded by the mechanical sensor and the surface deformation image of the sample captured by the depth camera are collected in real time. The collected force data is subjected to noise removal and data smoothing processing, and at the same time, the deformation image is preprocessed. By analyzing the preprocessed image data, key deformation features are extracted, and a force-deformation model is constructed in combination with the force data to reveal the physical behavior of the sample during the stretching process; the constructed model is used to determine the critical deformation point and identify abnormal deformations, calculate the maximum force value of the sample through this critical point, and further correct the abnormality of this value; the anti-tensile performance index of the fastener is determined based on the corrected maximum force value to complete the entire detection process; this solution combines a depth camera and a mechanical sensor to achieve real-time synchronous collection and processing of force data and deformation images, and through steps such as data preprocessing, noise filtering, force-deformation modeling, and critical point identification, effectively solves the problems of incomplete data capture, difficult identification of abnormal deformations, and inaccurate calculation of the maximum force value in the prior art, thereby greatly improving the accuracy and reliability of the anti-tensile performance detection of fasteners. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0017] The structures, proportions, sizes, etc. shown in the drawings of this specification are only used to cooperate with the content disclosed in the specification for those who are familiar with this technology to understand and read, and are not used to limit the limiting conditions under which the present invention can be implemented. Therefore, they do not have technical essence. Any modification of the structure, change of the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope that can be covered by the technical content disclosed in the present invention.

[0018] Figure 1 It is a flowchart showing the method for detecting the anti-tensile performance of the fastener in the first embodiment; Figure 2 It is a schematic structural layout diagram of the device for detecting the anti-tensile performance of the fastener in the second embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the objectives, features, and advantages of the present invention more obvious and understandable, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0020] In the description of the present invention, it should be understood that the orientation or positional relationships indicated by the terms "upper", "lower", "top", "bottom", "inner", "outer", etc. are based on the orientation or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. It should be noted that when a component is considered to be "connected" to another component, it can be directly connected to the other component or there may be intermediate components present at the same time.

[0021] The technical solutions of the present invention will be further described below with reference to the accompanying drawings and through specific embodiments.

[0022] Embodiment 1: Combined Figure 1 As shown, the embodiment of the present invention provides a method for detecting the tensile resistance performance of a fastener, including: S1, obtaining the attribute parameters of the fastener to be tested, and installing the fastener to be tested in the clamping mechanism of the tensile resistance performance detection device; the tensile resistance performance detection device is configured with a depth camera and a mechanical sensor.

[0023] The attribute parameters include material, size, shape, and use environment. The attribute parameters will help determine the applicable tensile test parameters, such as the maximum force, test loading rate, etc. And ensure that the clamping mechanism can adapt to the specific size and shape of the sample.

[0024] S2, starting the tensile resistance performance detection device, applying a tensile force to the fastener to be tested at a preset tensile rate. During the tensile process, the mechanical sensor continuously records the force data of the sample, and at the same time, the depth camera captures the deformation image data on the surface of the sample.

[0025] The deformation image data in this process will be the key to subsequent analysis, ensuring that the minute changes of the sample during the force application process can be comprehensively monitored. The real-time monitored data includes tensile force, deformation, initial signs of cracks or fractures on the surface of the sample, etc.

[0026] S3. During the testing process, the collected tensile force data and deformation image data will be processed synchronously. After that, noise removal and data smoothing will be performed on the force data, and image preprocessing will be carried out on the deformation image data.

[0027] Image preprocessing includes removing background noise, correcting image distortion, standardizing image size, etc. The preprocessed data will serve as the basis for subsequent analysis to ensure the accuracy and consistency of the data.

[0028] S4. By analyzing the deformation image data, key deformation features are extracted, and combined with the force data during the stretching process, a force-deformation model is constructed.

[0029] The force-deformation model can reveal the deformation law of the fastener under specific force conditions, especially the key critical deformation points. Through these data, it is determined whether the sample has abnormal deformation, whether it has reached the critical force point, and its ultimate tensile resistance performance (such as tensile fracture point, maximum deformation amount, etc.) is predicted.

[0030] S5. Based on the force-deformation model analysis, the critical deformation point is obtained, and the abnormal deformation points during the tensile test are judged. The maximum force value is obtained through the critical deformation point operation, and the abnormal deformation points are used to correct the maximum force value to obtain the corrected maximum force value.

[0031] S6. Based on the corrected maximum force value, the tensile resistance performance index of the fastener is obtained.

[0032] The working principle of the present invention is as follows: First, the attribute parameters of the fastener to be tested are collected and installed in a detection device equipped with a depth camera and a mechanical sensor. The detection device is started at a preset stretching rate to perform tensile loading on the sample, and the force data recorded by the mechanical sensor and the surface deformation image of the sample captured by the depth camera are collected in real time. Noise removal and data smoothing are performed on the collected force data, and at the same time, preprocessing is carried out on the deformation image. By analyzing the preprocessed image data, key deformation features are extracted, and combined with the force data, a force-deformation model is constructed to reveal the physical behavior of the sample during the stretching process; the constructed model is used to determine the critical deformation point and identify abnormal deformations, and the maximum force value of the sample is calculated through this critical point and further corrected for abnormalities; based on the corrected maximum force value, the tensile resistance performance index of the fastener is determined to complete the entire detection process; this solution combines a depth camera and a mechanical sensor to realize the real-time synchronous acquisition and processing of force data and deformation images, and through steps such as data preprocessing, noise filtering, force-deformation modeling, and critical point identification, effectively solves the problems of incomplete data capture, difficult identification of abnormal deformations, and inaccurate calculation of the maximum force value in the prior art, thereby greatly improving the accuracy and reliability of the tensile resistance performance detection of fasteners.

[0033] In this embodiment, the mechanical sensor and the depth camera are both attached with timestamps generated by a unified system clock; specifically, step S3 includes: S301, according to the timestamps attached to each data point, preliminarily match the mechanical data and the image data according to a preset sampling period (e.g., 10 ms).

[0034] S302, for the case where there are differences in sampling frequencies, use linear interpolation or spline interpolation algorithms to adjust the data sequences so that one-to-one corresponding data pairs are formed on the same time scale.

[0035] S303, the formed synchronized data pairs are stored in the cache module to provide unified time-sequenced cached data.

[0036] S304, use a low-pass filter (e.g., the cut-off frequency can be set to 50 Hz and fine-tuned according to the characteristics of the actual detection instrument and test requirements) to preliminarily denoise the original force data and suppress high-frequency noise components.

[0037] S305, apply moving average filtering (e.g., using a 10-point window) or median filtering algorithm to perform secondary smoothing processing on the force data to eliminate instantaneous abnormal fluctuations; while maintaining the force change trend and key features.

[0038] S306, use the internal and external parameters obtained during the pre-calibration of the camera to perform geometric distortion correction on the deformed image data; ensure the accuracy of the image size and proportion.

[0039] S307, for the original image noise in the deformed image data, use Gaussian filtering with a filter kernel size of 5×5 and a standard deviation set to 1.0 to suppress the noise; improve the image quality.

[0040] S308, perform brightness normalization processing on the image through histogram equalization technology to enhance the contrast and detail performance of the deformed area in the image.

[0041] S309, preset the region of interest (ROI) according to the test requirements, crop the image, and retain the image data of the key surface area of the fastener; for subsequent extraction and analysis of deformation characteristics.

[0042] S310, uniformly store the processed force data and deformed image data in time series and construct a corresponding data index database.

[0043] In this embodiment, specifically, step S4 includes: S41. Perform morphological operations including erosion and dilation on the preprocessed deformed image data, and the operation size can be set as a convolution kernel of 3×3 or 5×5; used to extract edges and details in the image to ensure accurate identification of the deformed area.

[0044] S42. Apply an image segmentation algorithm, such as threshold-based segmentation, to segment the image into multiple regions to highlight the regions where deformation occurs on the sample surface.

[0045] S43. Extract key deformation features from the segmented deformed regions. The key deformation features include the area of the deformed region, the deformation rate (the degree of regional deformation), and the deformation mode (such as expansion, compression, bending, etc.).

[0046] S44. Through multi-scale analysis, track the deformation in different stretching stages to form deformation sequence data; ensure that the whole process from small deformation to large-scale deformation can be accurately captured.

[0047] S45. Match the synchronously processed force data with the key deformation features one by one according to the time stamp, so that each force data point corresponds to a deformation state.

[0048] S46. Use regression analysis to analyze the quantitative relationship between force and deformation features and identify the non-linear relationship between force and deformation.

[0049] S47. Extract key points in the stretching process. The key points include the initial force, the elastic stage, the yield point, and the maximum force point. Combine the turning points of the deformation to construct a mechanical model describing the tensile resistance performance.

[0050] S48. Use the finite element analysis method to establish a stress-deformation model to simulate the deformation mode and stress distribution of the fastener under different force conditions.

[0051] S49. Verify and correct the stress-deformation model through actual test data and the constructed mechanical model; ensure the accuracy of the model and predict the reliability and durability of the fastener under different working conditions.

[0052] In this embodiment, specifically, step S5 specifically includes: According to the force-deformation model, first calculate the deformation rate and force change of the fastener sample in each stage of the entire stretching process.

[0053] Pay special attention to the deformation acceleration stage. In the deformation acceleration stage, the stress and deformation of the sample change sharply, and this is the candidate interval for the critical deformation point.

[0054] Smooth the force-deformation curve through a local smoothing algorithm; To improve accuracy, during the identification process of the critical point, the original curve can be smoothed through a local smoothing algorithm (such as the Savitzky–Golay filter) to reduce the influence of noise and ensure more accurate positioning of the critical point. Use the force-deformation curve and combine the second derivative method (acceleration method) to identify the critical point, that is, by calculating the second derivative of the curve to find the inflection point of the curve; this usually corresponds to the yield point of the sample or the turning point of the maximum deformation rate.

[0055] Compare all the calculated deformation points in the force-deformation model with the actual test data, and identify the abnormal deformation points that are significantly different from most of the data by calculating the standard deviation of the deformation or based on the maximum difference method (such as the maximum cosine distance method); For the identified abnormal deformation points, use an outlier detection algorithm (such as the DBSCAN or LOF algorithm) to further verify whether the point belongs to abnormal data caused by external interference or equipment failure during the test process; If the abnormal deformation point is valid data, it is retained; otherwise, the abnormal point will be marked as invalid data and excluded from the calculation.

[0056] Based on the force-deformation model, first determine the maximum force value at the critical deformation point. This value usually corresponds to the point where the sample shows obvious deformation or yield, and the force reaches the peak at this time; According to the identified abnormal deformation points, adjust the maximum force value through a data correction algorithm (such as using the median or weighted average method); remove the interference of outliers on the test results; Through the weighted average algorithm, combine the critical point and the data after removing the abnormal points to correct the maximum force value calculated at the critical deformation point to obtain the corrected maximum force value. Use the abnormal deformation points to correct this value, and the correction process ensures that the corrected maximum force value is more accurate and stable.

[0057] Embodiment 2: Combined Figure 2 As shown, the present invention also provides a tensile resistance detection device for fasteners, which is applied to implement the tensile resistance detection method for fasteners as in Embodiment 1. The tensile resistance detection device includes: A clamping mechanism 10 for precisely fixing the fastener to be tested; A tensile testing machine 20 for providing a tensile force; A mechanical sensor 30 for real-time recording of the tensile force and strain data of the fastener during the tensile process; A depth camera 40 for real-time capturing of the deformation image on the surface of the fastener to provide deformation data.

[0058] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the tensile resistance performance of a fastener, characterized in that, Including: Obtain the attribute parameters of the fastener to be tested, and install the fastener to be tested in the clamping mechanism of the tensile property detection device; The tensile property detection device is configured with a depth camera and a mechanical sensor; Start the tensile property detection device, apply a tensile force to the fastener to be tested at a preset tensile rate. During the tensile process, the mechanical sensor continuously records the force data of the sample. At the same time, the depth camera captures the deformation image data on the surface of the sample; During the test, the collected tensile force data and deformation image data will be synchronously processed. Then, noise removal and data smoothing processing will be performed on the force data, and image preprocessing will be performed on the deformation image data; By analyzing the deformation image data, extract the key deformation features, and combine with the force data during the tensile process to construct a force-deformation model; Based on the force-deformation model, analyze and obtain the critical deformation point, and judge the abnormal deformation point during the tensile test process. Obtain the maximum force value through the critical deformation point operation, and use the abnormal deformation point to correct the maximum force value to obtain the corrected maximum force value; Based on the corrected maximum force value, obtain the tensile property index of the fastener.

2. The method for detecting the tensile resistance performance of the fastener according to claim 1, wherein Both the mechanical sensor and the depth camera are attached with timestamps generated by a unified system clock; During the test, the collected tensile force data and deformation image data will be synchronously processed, specifically including: According to the timestamps attached to each data point, preliminarily match the mechanical data and the image data according to a preset sampling period; For the case where there are differences in the sampling frequencies, use linear interpolation or spline interpolation algorithms to adjust the data sequences so that one-to-one corresponding data pairs are formed on the same time scale; The formed synchronous data pairs are stored in the cache module to provide unified time-series cached data.

3. The method for detecting the tensile resistance performance of the fastener according to claim 2, characterized in that, Then, noise removal and data smoothing processing will be performed on the force data, and image preprocessing will be performed on the deformation image data, specifically including: Use a low-pass filter to perform preliminary noise reduction on the original force data to suppress high-frequency noise components; Apply a moving average filter or a median filter algorithm to perform secondary smoothing processing on the force data to eliminate instantaneous abnormal fluctuations; Use the internal and external parameters obtained during the pre-calibration of the camera to perform geometric distortion correction on the deformation image data; For the original image noise in the deformation image data, use Gaussian filtering with a filter kernel size of 5×5 and a standard deviation set to 1.0 to suppress the noise; Perform brightness normalization processing on the image through histogram equalization technology to enhance the contrast and detail performance of the deformation area in the image; Preset the region of interest according to the test requirements, and crop the image to retain the image data of the key surface area of the fastener.

4. The method for detecting the tensile resistance performance of the fastener according to claim 3, characterized in that, After presetting the region of interest according to the test requirements, cropping the image to retain the image data of the key surface area of the fastener, it also includes: Uniformly store the processed force data and deformation image data in time series, and construct a corresponding data index database.

5. The method for detecting the tensile resistance performance of the fastener according to claim 1, characterized in that, By analyzing the deformation image data, extract the key deformation features, and combine with the force data during the tensile process to construct a force-deformation model, specifically including: Perform morphological operations including erosion and dilation on the preprocessed deformed image data, and the operation size can be set as a convolution kernel of 3×3 or 5×5; Apply an image segmentation algorithm, such as threshold-based segmentation, to segment the image into multiple regions to highlight the regions where deformation occurs on the sample surface; Extract key deformation features from the segmented deformation regions, and the key deformation features include the area, deformation rate, and deformation mode of the deformation region; Through multi-scale analysis, track the deformation in different stretching stages to form deformation sequence data.

6. The method for detecting the tensile resistance performance of the fastener according to claim 5, wherein After the above-mentioned multi-scale analysis is used to track the deformation in different stretching stages to form deformation sequence data, it further includes: Match the synchronously processed force data with the key deformation features one by one according to the time stamp, so that each force data point corresponds to a deformation state; Adopt regression analysis method to analyze the quantitative relationship between force and deformation features, and identify the non-linear relationship between force and deformation; Extract key points in the stretching process, including the initial force, elastic stage, yield point, and maximum force point, and combine the turning points of the deformation to construct a mechanical model describing the anti-tensile performance; Adopt the finite element analysis method to establish a stress-deformation model to simulate the deformation mode and stress distribution of the fastener under different force conditions; Verify and correct the stress-deformation model through actual test data and the constructed mechanical model.

7. The method for detecting the tensile resistance performance of the fastener according to claim 1, characterized in that, Based on the analysis of the force-deformation model, obtain the critical deformation point and judge the abnormal deformation points in the tensile test process, specifically including: According to the force-deformation model, first calculate the deformation rate and force change of the fastener sample in each stage during the entire stretching process; Smooth the force-deformation curve through a local smoothing algorithm; Use the force-deformation curve and combine the second derivative method to identify the critical point, that is, by calculating the second derivative of the curve, find the inflection point of the curve; Compare all the calculated deformation points in the force-deformation model with the actual test data, and identify the abnormal deformation points that are significantly different from most of the data by calculating the standard deviation of the deformation or based on the maximum difference method; For the identified abnormal deformation points, use the outlier detection algorithm to further verify whether the point belongs to the abnormal data caused by external interference or equipment failure and other factors during the test; If the abnormal deformation point is valid data, it is retained, otherwise the abnormal point will be marked as invalid data and excluded from the calculation.

8. The method for detecting the tensile resistance performance of the fastener according to claim 7, characterized in that, The maximum force value is obtained through critical deformation point calculation, and the maximum force value is corrected by the abnormal deformation point to obtain the corrected maximum force value, specifically including: Based on the force-deformation model, first determine the maximum force value at the critical deformation point; According to the identified abnormal deformation points, adjust the maximum force value through a data correction algorithm; Through the weighted average algorithm, combine the critical point and the data after removing the abnormal points to correct the maximum force value calculated by the critical deformation point to obtain the corrected maximum force value.

9. A tensile resistance performance detection device for a fastener, characterized in that, Applied to implement the anti-tensile performance detection method of the fastener as described in any one of claims 1 to 8, the anti-tensile performance detection device includes: A clamping mechanism for precisely fixing the fastener to be tested; A tensile testing machine for providing tensile force; A mechanical sensor for recording in real time the tensile force and strain data of the fastener during the tensile process; A depth camera for capturing in real time the deformation image of the fastener surface and providing deformation data.

Images