Quality detection system for metal products

Abstract

The invention relates to the technical field of quality detection, and particularly discloses a metal product quality detection system which is provided with a three-dimensional scanning module, a data capturing module, a data self-encoding module, a destructive testing module and an image analysis module. The three-dimensional shape of the metal product is scanned and reproduced based on a laser scanning imaging technology to obtain a three-dimensional model of the metal product, a light signal value is captured and analyzed to obtain surface parameters of the metal product, a surface quality evaluation value is generated, and further sampling is carried out for destructive testing. And analyzing the surface quality evaluation equipartition and anti-damage parameter weighted qualified rate of each metal product sample set to obtain a comprehensive quality score of each metal product sample set, and drawing, so that the reduction of errors of metal product quality detection evaluation is facilitated, and the quality detection and logicality and the metal product quality detection efficiency are improved. Visual presentation is carried out based on the comprehensive quality score, so that a decision maker can read a quality detection result and make a further decision.

Description

Technical Field

[0001] This invention relates to the field of quality inspection technology, specifically to a quality inspection system for metal products. Background Technology

[0002] Metal products play a crucial role in the daily operation of machinery and equipment, affecting the normal operation of the entire mechanical system. The quality of metal products directly impacts the performance and lifespan of the entire machinery. High-quality processed metal products ensure the stability and reliability of machinery. Metal product quality inspection is of great significance for production safety in many manufacturing fields. Using quality-inspected metal products ensures that products meet actual production standards and regulations, guaranteeing stable and reliable output quality. Furthermore, quality inspection allows for monitoring and timely detection and resolution of quality problems during the metal product manufacturing process, reducing scrap and rework caused by substandard products, improving production efficiency, and saving costs. In addition, research and testing of metal product quality helps designers select appropriate manufacturing processes and optimize the design of metal product quality characteristics based on the needs of machinery, thereby improving the compatibility between metal products and machinery, and enhancing the overall production efficiency and performance of the machinery in actual operation.

[0003] At present, there are still some shortcomings in the quality inspection system of metal products, which are specifically reflected in the following aspects: (1) In the past, the quality inspection of metal products was mostly based on the metal products themselves. On the one hand, the production of metal products is usually large, and the workload of measurement and testing and parameter acquisition based on the metal products themselves is huge. Often, due to manual operation and subjective judgment in the actual large-scale measurement work, data errors are caused, which are difficult to be detected and considered. On the other hand, the inspection behavior based on the metal products themselves may affect the quality of the metal products. Bumps, squeezing, loss, etc. may affect their own quality level, interfere with the quality inspection of metal products, and reduce the accuracy of the quality inspection of metal products.

[0004] (2) The destructive testing of metal products is conducted completely randomly and independently, without considering the logical relationship between the surface quality of metal products and the destructive testing. The three-dimensional morphological feature points of metal products, such as the size of the metal products, surface roughness, circumference and parallel error rate, may have different degrees of defects in the three-dimensional morphological feature points of metal products. These defects may exhibit different damage characteristics and test results in the destructive testing, making it difficult to accurately reflect the different effects of the surface quality of metal products in the destructive testing. It may even produce errors in the quality inspection and evaluation of metal products, thus improving the coherence and logic of quality inspection and evaluation.

[0005] (3) The traditional perspective of metal product quality inspection is only to provide the quality inspection results of metal products. The quantitative expression of the quality inspection results lacks visualization, which is not conducive to decision-makers or designers reading the results of metal product quality inspection and making further decisions. In addition, it has the shortcoming of quality inspection that pursues the improvement of metal product quality while ignoring the quality fluctuation. It lacks the practical significance of metal product quality inspection in actual work and ignores the potential value of metal product quality inspection. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a quality inspection system for metal products, which can effectively solve the problems mentioned in the background section.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a quality inspection system for metal products, comprising a three-dimensional scanning module, a data capture module, a data self-encoding module, a destructive testing module, and an image analysis module. The three-dimensional scanning module is used to scan and reproduce the three-dimensional morphology of the metal product based on laser scanning imaging technology, calculate the laser scanning effect evaluation value, and mark the three-dimensional morphological feature points of the metal product. The data capture module is used to acquire the light signal values ​​captured based on the three-dimensional morphological feature points of the metal product, forming a light signal feedback set. The data self-encoding module is used to base the light signal values ​​on the light signal feedback set and the feature point light signals set in the database. The feedback correction factor analysis obtains the surface parameters of the metal products, and generates surface quality assessment values ​​based on these parameters. The destructive testing module generates several metal product sample sets based on randomly sampled metal products and performs tensile, compression, and fatigue tests to obtain the metal products' resistance to damage. A comprehensive analysis of these resistance parameters yields a weighted pass rate. The image analysis module obtains a comprehensive quality score for each metal product sample set based on the average surface quality assessment score and the weighted pass rate of the resistance to damage parameters. It then plots the comprehensive quality score image for each metal product sample set and analyzes its quality stability.

[0008] As a further solution, the three-dimensional morphology of metal products is scanned and reproduced based on laser scanning imaging technology, and the laser scanning effect evaluation value is calculated. The specific analysis process includes: obtaining laser scanning parameters, including scanning accuracy, resolution, scanning range, and scanning light source focus; scanning the three-dimensional morphology of the metal product; performing three-dimensional modeling on the obtained three-dimensional morphology image of the metal product to obtain a three-dimensional morphology model of the metal product; obtaining the three-dimensional morphology matching value set in the three-dimensional modeling software database; and calculating the laser scanning effect evaluation value based on the three-dimensional morphology matching value and the laser scanning parameters.

[0009] As a further option, the laser scanning effect evaluation value is calculated using the following formula:

[0010]

[0011] In the formula, γ is the laser scanning effect evaluation value, S is the scanning accuracy, F is the resolution, M is the scanning range, J is the scanning light source focus, k is the three-dimensional shape matching value set in the database of the three-dimensional modeling software, and e is the natural constant.

[0012] As a further solution, the three-dimensional morphological feature points of the metal product are marked. The specific analysis process is as follows: obtain the three-dimensional morphological model of the metal product; mark the three-dimensional morphological feature points of the metal product based on the three-dimensional morphological model of the metal product. The three-dimensional morphological feature points of the metal product include the size of the metal product, surface roughness, circumferential scale and parallelism error rate.

[0013] As a further approach, optical signal values ​​captured based on the three-dimensional morphological feature points of the metal product are obtained to form an optical signal feedback set. The specific analysis process is as follows: simulated laser beam irradiation is performed on the three-dimensional morphological model of the metal product based on the three-dimensional morphological feature points of the metal product; the light reflection sites of the simulated laser beam on the three-dimensional morphological feature points of the metal product are obtained, and the light reflection sites are recorded as optical signal values; based on the optical signal values ​​and the correspondence between the three-dimensional morphological feature points of the metal product and the light reflection sites of the simulated laser beam, an optical signal feedback set is formed.

[0014] As a further approach, the surface parameters of the metal product are obtained by analyzing the optical signal feedback set and the feature point optical signal feedback correction factor set in the database. The specific analysis process is as follows: obtain the optical signal feedback set of the metal product; extract the feature point optical signal feedback correction factor set in the database based on the three-dimensional morphology feature points of the metal product corresponding to the light reflection sites of the simulated laser beam in the optical signal feedback set; and obtain the optical signal values ​​corresponding to each feature point of the three-dimensional morphology of the metal product after correction by the feature point optical signal feedback correction factor set in the database based on the optical signal feedback set. The corrected signal values ​​of each optical feature point are the surface parameters of the metal product.

[0015] As a further solution, surface parameters of metal products are used to generate surface quality assessment values. The specific analysis process is as follows: obtain the surface parameters of the metal products; calculate the surface quality assessment values ​​based on the surface parameters of the metal products and the surface quality assessment model.

[0016] As a further approach, several metal product sample sets are generated from randomly sampled metal products, and tensile, compression, and fatigue tests are performed to obtain the metal products' resistance to damage. The specific analysis process is as follows: the surface quality assessment values ​​of the metal products are sorted, and the surface quality assessment values ​​are divided into twenty equal grades from high to low; 0.5% of the metal products in each grade are randomly selected as a metal product sample set, which serves as the basis for subsequent destructive testing and obtaining the metal products' resistance to damage. Tensile, compression, and fatigue tests are performed on each metal product in any metal product sample set; the metal products' resistance to damage after tensile, compression, and fatigue tests are obtained.

[0017] As a further approach, a comprehensive quality score for each metal product sample set is obtained based on the average surface quality assessment score and the weighted pass rate of the anti-breakage parameter. A comprehensive quality score image is plotted for each metal product sample set, and its quality stability is analyzed. The specific analysis process is as follows: Obtain the surface quality assessment values ​​of each basic test metal product in any metal product sample set; calculate the average of these values ​​to obtain the average surface quality assessment score; obtain the weighted pass rate of the anti-breakage parameter; calculate the comprehensive quality score for each metal product sample set based on the average surface quality assessment score and the weighted pass rate; plot a line graph based on the comprehensive quality scores; obtain the comprehensive quality score reference value set in the database and mark it on the line graph; count the number of data points whose comprehensive quality scores exceed the reference value set in the database, and use the proportion of this number to the total comprehensive quality scores of each metal product sample set as the basis for analyzing the quality stability of the metal products. The analysis then examines the quality stability reflected by this proportion.

[0018] As a further step, a comprehensive quality score was calculated for each metal product sample set. The specific analysis process is as follows:

[0019]

[0020] In the formula, ω j The overall quality score of the metal product sample set is given, j = 1, 2, 3, ..., m, where m is the total number of metal product samples, φ ji Let be the surface quality assessment value of the i-th basic test metal product in the j-th metal product sample set, where i = 1, 2, 3, ..., n, n is the total number of basic test metal products in the metal product sample set, Kp is the weighted pass rate of the anti-damage parameter of the metal product sample set, and e is a natural constant.

[0021] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects:

[0022] (1) This invention uses laser scanning imaging technology to scan and reproduce the three-dimensional morphology of metal products. It marks the three-dimensional morphology feature points of metal products in the form of a three-dimensional morphology model of the metal products, reducing the impact on the metal products themselves during the measurement process, reducing the wear and tear on the metal products themselves during quality inspection, and avoiding unnecessary cost increases. At the same time, it can record the data and images of the three-dimensional morphology of metal products with constant standards, accurately extract the feature point data and images of any three-dimensional morphology of metal products, reduce the errors caused by subjective judgment in manual measurement and data reading, and quantify the scanning error through the laser scanning effect evaluation value. In addition, the processing can process a large number of metal products in batches, which meets the production characteristics of large quantities of metal products in actual work, helps to improve the efficiency of metal product quality inspection, and enhances the stability of metal product quality inspection.

[0023] (2) The surface quality assessment values ​​of metal products are sorted and graded, and 0.5% of the metal products in each grade are randomly selected as a metal product sample set as the basis for subsequent destructive testing. The results of destructive testing of metal products under different surface quality levels are fully considered to improve the integrity of metal product quality inspection. On the basis of achieving horizontal coverage of metal product quality inspection, the reliability of quality inspection is improved, and the quality status of metal products can be more comprehensively evaluated, reducing the possibility of missed detection and false detection.

[0024] (3) This invention presents the quality test results of different metal product sample sets by drawing a line graph based on the comprehensive quality score of each metal product sample set. This breaks through the previous shortcomings of quality testing that only pursued the improvement of metal product quality while ignoring the quality fluctuation. The comprehensive quality score reference value set in the database is marked on the line graph, which can more intuitively observe the changes in the comprehensive quality score of different sampling sets of metal products, and provide decision-makers with a visual expression of quality stability. Attached Figure Description

[0025] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0026] Figure 1 This is a schematic diagram of the system module connections of the present invention;

[0027] Figure 2 A flowchart illustrating the steps for forming an optical signal feedback set based on optical signal values ​​captured from three-dimensional morphological feature points of a metal product. Detailed Implementation

[0028] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0029] Please see Figure 1 The present invention provides a technical solution: a quality inspection system for metal products, including a three-dimensional scanning module, a data capture module, a data self-encoding module, a destructive testing module, and an image analysis module.

[0030] The 3D scanning module is used to scan and reproduce the 3D morphology of metal products based on laser scanning imaging technology, calculate the laser scanning effect evaluation value, and mark the 3D morphology feature points of the metal products.

[0031] Specifically, the process involves scanning and reproducing the three-dimensional morphology of metal products using laser scanning imaging technology, and calculating the laser scanning effect evaluation value. The specific analysis includes: acquiring laser scanning parameters, including scanning accuracy, resolution, scanning range, and scanning light source focus; scanning the three-dimensional morphology of the metal product; performing three-dimensional modeling on the obtained three-dimensional morphology image of the metal product to obtain a three-dimensional morphology model of the metal product; acquiring the three-dimensional morphology matching value set in the three-dimensional modeling software database; and calculating the laser scanning effect evaluation value based on the three-dimensional morphology matching value and the laser scanning parameters.

[0032] It should be noted that acquiring laser scanning parameters, including accuracy, resolution, scanning range, and scanning light source focus, directly affects the quality and accuracy of the scanning effect in terms of the range and implementation precision. To achieve accurate evaluation of the laser scanning effect and the quantitative quality of the evaluation value, the 3D shape matching value set in the 3D modeling software database is also considered here. The 3D shape matching value refers to the degree of restoration and matching between the 3D shape image of the metal product and the 3D shape of the actual metal product, which can be achieved by the 3D modeling software based on actual working limitations and computing power. This operation can accurately reproduce the 3D shape of the metal product, which helps to understand its surface features more in detail and comprehensively. By calculating the laser scanning effect evaluation value, potential problems in the scanning process can be identified in time, and scanning parameters can be optimized to improve the quality and stability of the scanning effect, thereby improving the accuracy of the overall quality inspection of metal products.

[0033] Specifically, the formula for calculating the laser scanning effect evaluation value is as follows:

[0034]

[0035] In the formula, γ is the laser scanning effect evaluation value, S is the scanning accuracy, F is the resolution, M is the scanning range, J is the scanning light source focus, k is the three-dimensional shape matching value set in the database of the three-dimensional modeling software, and e is the natural constant.

[0036] It should be noted that scanning accuracy refers to the precision that a laser scanning device can achieve during scanning. Higher scanning accuracy means that the scanned result is closer to the shape of the actual metal product. Resolution refers to the ability to distinguish different details in a 3D morphological scanned image of a metal product. Higher resolution provides clearer and more detailed scan results. Scanning range is the spatial range that the laser scanning device can cover. Laser scanning devices with a larger scanning range are better at capturing the feature details of different metal products, improving the completeness of the 3D morphological reproduction. Scanning light source focus refers to the degree of focus of the beam generated by the laser scanning device, affecting the beam focus and dimensional accuracy during scanning. There are interrelationships among the optical scanning parameters. Scanning accuracy and resolution directly affect the fineness of the scanning results, while scanning range and the focus of the scanning light source affect the scanning range and clarity. When establishing a 3D morphology model, parameters need to be adjusted according to the actual 3D morphology of the metal product and the performance of the equipment to ensure the accuracy and reliability of the scanning results. At the same time, the 3D morphology matching value can be used as a reference standard during the modeling process to help evaluate the accuracy of the model and optimize the modeling process. By comprehensively considering and reasonably adjusting the laser scanning parameters, the scanning effect and modeling quality can be improved. This can be effectively applied to the scanning and reproduction process of the 3D morphology of metal products and the accurate calculation of laser scanning effect evaluation values.

[0037] Specifically, the process of marking three-dimensional morphological feature points of metal products involves: obtaining a three-dimensional morphological model of the metal product; marking three-dimensional morphological feature points of the metal product based on the three-dimensional morphological model, wherein the three-dimensional morphological feature points of the metal product include the size of the metal product, surface roughness, circumferential dimensions, and parallelism error rate.

[0038] It should be noted that marking the three-dimensional morphological feature points of metal products based on their three-dimensional morphological models can reduce the wear and tear on the metal products themselves. This approach enables the large-scale collection of three-dimensional morphological feature point data of metal products and achieves low-cost and efficient data acquisition. Here, the dimensions of the metal product refer to its actual dimensions in terms of length, width, and height; surface roughness refers to the roughness of the metal product's surface, described by flatness; circumferential dimensions refer to the size of the circumferential or annular portions of the metal product, measured using a dedicated roundness measuring instrument and image measurement system; and parallelism error rate refers to the error rate that occurs during the manufacturing and processing of the metal product, describing the geometric deviation in the circumferential dimension, which affects the assembly accuracy and rotational stability of the metal product. These three-dimensional morphological feature points comprehensively showcase the surface characteristics of the metal product from multiple geometric dimensions, facilitating a comprehensive evaluation and analysis of its quality and ensuring that the metal product meets standard quality requirements and possesses good working performance.

[0039] It should be noted that the scanning and reproduction of the three-dimensional morphology of metal products based on laser scanning imaging technology marks the three-dimensional morphological feature points of the metal products in the form of a three-dimensional morphological model, reducing the impact on the metal products themselves during the measurement process, reducing the wear and tear on the metal products during quality inspection, and avoiding unnecessary cost increases. At the same time, it can record the data and images of the three-dimensional morphology of metal products with constant standards, accurately extract the feature point data and images of any metal product's three-dimensional morphology, reduce the errors caused by subjective judgment in manual measurement and data reading, and quantify the scanning error through laser scanning effect evaluation values. In addition, the processing can batch process a large number of metal products, meeting the characteristics of large-volume production in actual work, which helps to improve the efficiency and stability of metal product quality inspection.

[0040] The data capture module is used to acquire optical signal values ​​based on the three-dimensional morphological feature points of the metal product and form an optical signal feedback set.

[0041] Specifically, the optical signal values ​​captured based on the three-dimensional morphological feature points of the metal product are obtained to form an optical signal feedback set. The specific analysis process is as follows: simulated laser beam irradiation is performed on the three-dimensional morphological model of the metal product based on the three-dimensional morphological feature points of the metal product; the light reflection sites of the simulated laser beam on the three-dimensional morphological feature points of the metal product are obtained, and the light reflection sites are recorded as optical signal values; based on the optical signal values ​​and the correspondence between the three-dimensional morphological feature points of the metal product and the light reflection sites of the simulated laser beam, an optical signal feedback set is formed.

[0042] In one specific embodiment, a light reflection point refers to a spot formed when a laser beam irradiates the surface of a metal product. Due to the optical properties of the surface, the light is reflected and forms a light spot. These light reflection points can be captured and recorded by measuring instruments to generate the light signal values ​​corresponding to the three-dimensional morphological feature points on the surface of the metal product. By acquiring these light reflection points, the three-dimensional morphological feature point data of the metal product can be accurately displayed using a unified standard. By analyzing the position and distribution of the light reflection points, the quality and manufacturing precision of the metal product can be evaluated, and problems such as defects, deformations, or foreign objects can be detected. In addition, the summarized light signal feedback set can reflect the light signal values ​​and the correspondence between the three-dimensional morphological feature points of the metal product and the light reflection points of the simulated laser beam based on the recorded data. This provides a data foundation for optimizing the manufacturing process of metal products, improving product design, and controlling quality, helping manufacturers improve the efficiency of subsequent quality inspection of metal products.

[0043] The data self-encoding module is used to analyze the surface parameters of metal products based on the optical signal feedback set and the feature point optical signal feedback correction factor set in the database, and to generate a surface quality assessment value based on the surface parameters of the metal products.

[0044] Specifically, the surface parameters of the metal product are obtained by analyzing the optical signal feedback set and the feature point optical signal feedback correction factor set in the database. The specific analysis process is as follows: obtain the optical signal feedback set of the metal product; extract the feature point optical signal feedback correction factor set in the database based on the three-dimensional morphology feature points of the metal product corresponding to the light reflection sites of the simulated laser beam in the optical signal feedback set; and obtain the optical signal values ​​corresponding to each feature point of the three-dimensional morphology of the metal product after correction by the feature point optical signal feedback correction factor set in the database. The corrected signal values ​​of each optical feature point are the surface parameters of the metal product.

[0045] It should be noted that by obtaining surface parameters of metal products based on optical signal feedback sets and combining them with feature point optical signal feedback correction factors set in the database, the corrected optical feature point signal values ​​can better reflect the actual situation of the metal product surface, reducing data inaccuracies caused by errors and deviations. This allows for more accurate extraction of optical signal values ​​from feature points on the metal product surface. The corrected optical signal values ​​more accurately reflect the three-dimensional morphological features of the metal product, thereby improving data accuracy and reliability. Furthermore, using data from the optical signal feedback set and the feature point optical signal feedback correction factors set in the database, surface parameters of metal products can be extracted quickly and accurately. This reduces human intervention and subjective errors, improves data processing efficiency and operability, and enhances the processing efficiency of metal product surface parameters. This helps manufacturers better control product quality and improve the efficiency of metal product quality assessment.

[0046] Specifically, the surface parameters of metal products generate surface quality assessment values. The specific analysis process is as follows: obtain the surface parameters of the metal products; calculate the surface quality assessment values ​​based on the surface parameters of the metal products and the surface quality assessment model.

[0047] In one specific embodiment, if the dimensions of a metal product are deviated or unstable, it will directly affect the assembly capability and functionality of the metal product. Inaccurate dimensions may lead to difficulties in product assembly, abnormal function, or performance degradation, thereby affecting surface quality. Surface roughness affects the appearance, friction performance, and coating adhesion of metal products. High roughness will affect the appearance quality, making the product look rough and reducing its aesthetic appeal. It may also reduce the adhesion performance of the coating and affect the protective performance, thus causing the metal product to fail to meet the quality level of testing and evaluation in the actual working environment. Circumferential dimensions and parallel error rate mainly affect the rotational stability and geometric accuracy of metal products in actual assembly, thereby affecting the quality level of the metal product.

[0048] It should be explained that the calculation formula for the surface quality assessment model is as follows:

[0049]

[0050] In the formula, Pg is the surface quality assessment value, t is the feature point optical signal feedback correction factor set in the database, a is the optical signal value corresponding to the size of the metal product, b is the optical signal value corresponding to the surface roughness, c is the optical signal value corresponding to the circumferential scale, d is the optical signal value corresponding to the parallel error rate, and e is the natural constant.

[0051] The destructive testing module is used to generate several metal product sample sets based on random sampling of metal products and to perform tensile, compression and fatigue tests to obtain the metal products' resistance to destruction parameters. The weighted pass rate of the metal products' resistance to destruction parameters is obtained by comprehensively analyzing the resistance parameters.

[0052] It should be noted that, generally speaking, if the dimensions of metal products are unstable or deviate, the inaccuracy of the dimensions may lead to significant deformation during processing and surface treatment, potentially increasing surface roughness and affecting the smoothness of the surface. It may also cause the diameter or circumference of circular structures to not meet design requirements, affecting the accuracy of circumferential dimensions. Surface roughness usually affects the measurement accuracy of circumferential dimensions because it affects the contact of measuring instruments. In addition, the non-uniformity of surface roughness may also lead to an increase in the parallelism error rate. These factors need to be comprehensively considered during production and quality control, and reasonable process control and testing methods should be used to ensure the quality inspection level of metal products.

[0053] Specifically, several metal product sample sets are generated from randomly sampled metal products, and tensile, compression, and fatigue tests are performed to obtain the metal products' resistance to damage. The specific analysis process is as follows: the surface quality assessment values ​​of the metal products are sorted and divided into twenty equal grades from high to low; 0.5% of the metal products in each grade are randomly selected as a metal product sample set, which serves as the basis for subsequent destructive testing and obtaining the metal products' resistance to damage. Tensile, compression, and fatigue tests are performed on each metal product in any metal product sample set; the metal products' resistance to damage after tensile, compression, and fatigue tests are obtained.

[0054] It should be noted that the tensile and compression tests refer to the metal product sample being mounted on a tensile testing machine and a compression testing machine, the tensile testing machine being started to perform two-dimensional tensile and compression tests on the metal product in both the transverse and longitudinal directions, and the load and displacement being recorded. After the above tensile and compression tests, fatigue tests are then performed, in which repeated alternating loads are applied to the metal product, and the frequency of the cyclic load is adjusted so that the metal product will gradually develop cracks under continuous cyclic loading and eventually break, which is considered the end of the test.

[0055] In one specific embodiment, the surface quality assessment values ​​of metal products are sorted and graded, and 0.5% of the metal products in each grade are randomly selected as a metal product sample set, which serves as the basis for subsequent destructive testing. This fully considers the results of destructive testing of metal products under different surface quality levels, improves the completeness of metal product quality inspection, enhances the reliability of quality inspection on the basis of achieving horizontal coverage of metal product quality inspection, and enables a more comprehensive assessment of the quality status of metal products, reducing the possibility of missed detections and false detections.

[0056] The image analysis module is used to obtain the comprehensive quality score of each metal product sample set based on the average surface quality assessment score and the weighted pass rate of the anti-damage parameter, to draw the comprehensive quality score image of each metal product sample set and to analyze its quality stability.

[0057] Specifically, the comprehensive quality score of each metal product sample set is obtained based on the average surface quality assessment score and the weighted pass rate of the anti-breakage parameter. A comprehensive quality score image of each metal product sample set is plotted, and its quality stability is analyzed. The specific analysis process is as follows: Obtain the surface quality assessment values ​​of each basic test metal product in any metal product sample set; calculate the average of these values ​​to obtain the average surface quality assessment score of the metal product sample set; obtain the weighted pass rate of the anti-breakage parameter for the metal product sample set; calculate the comprehensive quality score of each metal product sample set based on the average surface quality assessment score and the weighted pass rate of the anti-breakage parameter; plot a line graph based on the comprehensive quality score of each metal product sample set; obtain the comprehensive quality score reference value set in the database and mark it on the line graph; count the number of data points whose comprehensive quality score exceeds the reference value set in the database, and use the proportion of this number to the total comprehensive quality score of each metal product sample set as the basis for analyzing the quality stability of the metal products; analyze the quality stability of the metal products reflected by this proportion.

[0058] It should be noted that the reference value for the overall quality score set in the database is marked as a dashed line in the line graph. It does not affect the trend of the overall quality score image of each metal product sample set. It only provides decision-makers with a reference to measure the fluctuation and stability of the overall product score of metal products in different metal product sample sets.

[0059] In one specific embodiment, the quality test results of different metal product sample sets are displayed by plotting a line graph based on the comprehensive quality score of each metal product sample set. This breaks through the previous shortcomings of quality testing that focused solely on improving the quality of metal products while ignoring quality fluctuations. By marking the comprehensive quality score reference value set in the database on the line graph, the changes in different sampling sets of metal products can be observed more intuitively, providing decision-makers with a visualized expression of quality stability.

[0060] Specifically, the comprehensive quality score of each metal product sample set is analyzed as follows:

[0061]

[0062] In the formula, ω j The overall quality score for the metal product sample set is given, where j = 1, 2, 3, ..., m, and m is the total number of metal product samples. Let be the surface quality assessment value of the i-th basic test metal product in the j-th metal product sample set, where i = 1, 2, 3, ..., n, n is the total number of basic test metal products in the metal product sample set, Kp is the weighted pass rate of the anti-damage parameter of the metal product sample set, and e is a natural constant.

[0063] It should be noted that the surface quality assessment average score and the weighted pass rate of the anti-destruction parameter in the analysis and evaluation of the metal product sample set both fully consider the surface quality of each basic tested metal product in each metal product sample set. The weighted pass rate of the anti-destruction parameter reduces the extreme data and deviations caused by unexpected situations in the destructive testing of metal products, and can more accurately assess the quality status of metal products. This helps to strengthen the logic of metal product quality inspection, comprehensively consider the surface quality level and anti-destruction ability of metal products, and improve the overall quality inspection level and accuracy of metal products.

[0064] The above description is merely an example and illustration of the structure of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the structure of the invention or exceed the scope defined in the claims, all of which should fall within the protection scope of the present invention.

Claims

1. A quality inspection system for metal products, characterized in that, include: The 3D scanning module is used to scan and reproduce the 3D morphology of metal products based on laser scanning imaging technology, calculate the laser scanning effect evaluation value, and mark the 3D morphology feature points of the metal products. The data capture module is used to acquire optical signal values ​​based on the three-dimensional morphological feature points of metal products and form an optical signal feedback set. The data self-encoding module is used to analyze the surface parameters of metal products based on the optical signal feedback set and the feature point optical signal feedback correction factor set in the database, and to generate a surface quality assessment value based on the surface parameters of the metal products. The destructive testing module is used to generate several metal product sample sets based on random sampling of metal products and to conduct tensile, compression and fatigue tests to obtain the metal products' resistance to destruction parameters. The weighted pass rate of the metal products' resistance to destruction parameters is obtained by comprehensively analyzing the resistance parameters. The image analysis module is used to obtain the comprehensive quality score of each metal product sample set based on the average surface quality assessment score and the weighted pass rate of the anti-damage parameter, to draw the comprehensive quality score image of each metal product sample set and to analyze its quality stability.

2. The quality inspection system for metal products according to claim 1, characterized in that: The process of scanning and reproducing the three-dimensional morphology of metal products based on laser scanning imaging technology, and calculating the laser scanning effect evaluation value, includes the following specific analysis steps: Acquire laser scanning parameters, including scanning accuracy, resolution, scanning range, and scanning light source focus. Scan the three-dimensional shape of the metal product, and then perform three-dimensional modeling on the scanned three-dimensional shape image of the metal product to obtain a three-dimensional shape model of the metal product. Obtain the 3D shape matching value set in the 3D modeling software database, and calculate the laser scanning effect evaluation value based on the 3D shape matching value and laser scanning parameters.

3. The quality inspection system for metal products according to claim 1, characterized in that: The laser scanning effect evaluation value is calculated using the following formula: ； In the formula, S is the laser scanning effect evaluation value, F is the scanning accuracy, M is the scanning range, J is the scanning light source focus, k is the 3D shape matching value set in the 3D modeling software database, and e is the natural constant.

4. The quality inspection system for metal products according to claim 1, characterized in that: The specific analysis process for the three-dimensional morphological feature points of the marked metal product is as follows: Obtain a three-dimensional morphological model of a metal product; The three-dimensional morphological feature points of the metal product are marked based on the three-dimensional morphological model of the metal product. The three-dimensional morphological feature points of the metal product include the size of the metal product, surface roughness, circumferential scale and parallelism error rate.

5. The quality inspection system for metal products according to claim 1, characterized in that: The specific analysis process for acquiring optical signal values ​​based on the three-dimensional morphological feature points of metal products and forming an optical signal feedback set is as follows: Simulated laser beam irradiation is performed on the three-dimensional morphological feature points of metal products based on the three-dimensional morphological model of the metal products. The light reflection sites of the simulated laser beam on the three-dimensional morphological feature points of the metal product are obtained, and the light reflection sites are recorded as light signal values. Based on the optical signal values ​​and the correspondence between the three-dimensional morphological feature points of the metal product and the optical reflection sites of the simulated laser beam, an optical signal feedback set is formed.

6. The quality inspection system for metal products according to claim 1, characterized in that: The surface parameters of the metal product are obtained by analyzing the optical signal feedback correction factor based on the optical signal feedback set and the feature point optical signal feedback set in the database. The specific analysis process is as follows: Acquire optical signal feedback set of metal products; Based on the feature point optical signal feedback correction factor set in the database for extracting feature points of three-dimensional morphology of metal products corresponding to the light reflection sites of the laser beam in the optical signal feedback centralized simulation of laser beam. The optical signal values ​​corresponding to each feature point of the three-dimensional morphology of the metal product are corrected by the feature point optical signal feedback correction factor set in the database. The corrected optical feature point signal values ​​are the surface parameters of the metal product.

7. The quality inspection system for metal products according to claim 1, characterized in that: The surface parameters of the metal product are used to generate surface quality assessment values. The specific analysis process is as follows: Obtain surface parameters of metal products; The surface quality assessment value is calculated based on the surface parameters of the metal product and the surface quality assessment model.

8. The quality inspection system for metal products according to claim 1, characterized in that: The randomly sampled metal products generate several metal product sample sets, which are then subjected to tensile, compression, and fatigue tests to obtain the metal products' resistance to failure parameters. The specific analysis process is as follows: The surface quality assessment values ​​of metal products are ranked and divided into twenty equal grades from high to low. 0.5% of the metal products in each level are randomly selected as a metal product sample set. The metal product sample set serves as the basis for subsequent destructive testing and obtaining the metal product's resistance to damage parameters. Tensile, compression, and fatigue tests were performed on each metal product in any given metal product sample set. Obtain the anti-damage parameters of basic metal products after tensile, compression and fatigue tests.

9. A quality inspection system for metal products according to claim 2, characterized in that: The comprehensive quality score of each metal product sample set is obtained by comprehensively analyzing the average surface quality assessment score and the weighted pass rate of the anti-damage parameter. A comprehensive quality score image of each metal product sample set is plotted, and its quality stability is analyzed. The specific analysis process is as follows: Obtain the surface quality assessment values ​​of each basic test metal product in any metal product sample set, and calculate the average value of the surface quality assessment of the metal product sample set to obtain the average score of the surface quality assessment of the metal product sample set. Obtain the weighted pass rate of the anti-damage parameters of the metal product sample set; The comprehensive quality score of each metal product sample set is calculated based on the average surface quality assessment score and the weighted pass rate of the anti-damage parameter of each metal product sample set. A line graph was plotted based on the comprehensive quality score of each metal product sample set. Retrieve the comprehensive quality score reference value set in the database and mark it on the line chart; The number of data points whose comprehensive quality score for each metal product sample set exceeds the reference value for comprehensive quality score set in the database is counted. The proportion of the number of data points to the total comprehensive quality score of each metal product sample set is used as the basis for analyzing the stability of metal product quality. The analysis reflects the stability of metal product quality as indicated by the proportion.

10. A quality inspection system for metal products according to claim 2, characterized in that: The comprehensive quality score of each metal product sample set was analyzed in the following manner: ； In the formula, The overall quality score for the metal product sample set is given, where j = 1, 2, 3, ..., m, and m is the total number of metal product samples. Let be the surface quality assessment value of the i-th basic test metal product in the j-th metal product sample set, i=1,2,3...,n, where n is the total number of basic test metal products in the metal product sample set, Kp is the weighted pass rate of the anti-damage parameter of the metal product sample set, and e is a natural constant.

Images