Control method and system for product marking

Abstract

The invention relates to the technical field of marking control, and discloses a control method and system for product marking. The method comprises the steps of scanning the surface of a part to obtain material data; performing three-dimensional positioning according to the material data to obtain coordinate data; marking parameters are screened according to the coordinate data; using the parameters to control the marking device to generate process data; analyzing the process data to evaluate the marking quality; and optimizing parameters by using the quality data and storing the parameters in a database to form a control scheme. According to the invention, accurate identification of the material and shape characteristics of the part can be realized, the optimal marking position and process parameters are automatically determined according to the identification result, real-time monitoring and adjustment are carried out in the marking process, the marking quality is comprehensively evaluated and fed back and optimized after the real-time monitoring and adjustment are completed, and the consistency and reliability of the marking quality are improved.

Description

Technical Field

[0001] This application relates to the technical field of marking control, and particularly to a control method and system for product marking. Background Art

[0002] Laser marking technology, as a non-contact and high-precision product identification method, is widely used in fields such as aerospace, medical devices, automotive parts, and electronic products, providing permanent identification and traceability information for products. Traditional laser marking technology usually adopts fixed parameter settings, and operators set process parameters such as laser power and scanning speed for parts with different materials and shapes according to experience. In recent years, with the improvement of industrial automation level, computer vision-assisted marking systems have gradually been applied to production lines. By capturing part images through cameras, simple position detection and marking quality inspection are carried out, realizing preliminary monitoring of the marking process. At the same time, some advanced enterprises have begun to establish material libraries and process parameter libraries, storing recommended marking parameters corresponding to different materials for operators to refer to.

[0003] However, the existing marking control methods still have obvious deficiencies. First, the identification of part material characteristics mainly relies on manual experience judgment, lacking automated and precise material analysis means, resulting in the inability to make fine distinctions for parts of the same material but with different surface characteristics. Second, the determination of the marking position often uses simple two-dimensional positioning, making it difficult to handle the three-dimensional surfaces of complex-shaped parts, especially with lower marking accuracy for non-planar areas such as curved surfaces and inclined surfaces. Third, the setting of marking parameters is mainly based on static empirical values, lacking real-time monitoring and dynamic adjustment mechanisms, and unable to optimize immediately according to the material response changes during the marking process. The most crucial point is that the existing technology lacks a complete closed-loop control method, and the whole process from part detection, parameter optimization to quality evaluation lacks systematicness and coherence, resulting in large fluctuations in marking quality, low efficiency, and difficulty in meeting the increasingly strict marking requirements for high-precision parts. Summary of the Invention

[0004] This application provides a control method and system for product marking, which can accurately identify the material and shape characteristics of parts, automatically determine the best marking position and process parameters according to the identification results, conduct real-time monitoring and adjustment during the marking process, comprehensively evaluate the marking quality and provide feedback for optimization after completion, improving the consistency and reliability of marking quality.

[0005] In a first aspect, the present application provides a control method for product marking. The control method for product marking includes: scanning the surface of a metal part through image acquisition, analyzing the material characteristics of the obtained surface image to obtain product material data and surface feature data; performing three-dimensional coordinate positioning on the part according to the product material data and surface feature data, and correcting the positioning result through micro-displacement calibration to obtain marking position coordinate data; querying and screening process parameters in a material library according to the marking position coordinate data to obtain a marking power value and a speed value corresponding to the current part material; controlling a laser marking device through the marking power value and the speed value to perform marking processing on the part to obtain marking process data; performing imaging detection on the marking result according to the marking process data, and performing edge analysis and depth analysis on the detection image to obtain marking quality data; comparing and adjusting the marking process parameters through the marking quality data, and recording the adjusted parameters into a process database to obtain a marking control scheme.

[0006] In a second aspect, the present application provides a control system for product marking. The control system for product marking includes:

[0007] A scanning module for scanning the surface of a metal part through image acquisition, analyzing the material characteristics of the obtained surface image to obtain product material data and surface feature data;

[0008] A positioning module for performing three-dimensional coordinate positioning on the part according to the product material data and surface feature data, and correcting the positioning result through micro-displacement calibration to obtain marking position coordinate data;

[0009] A screening module for querying and screening process parameters in a material library according to the marking position coordinate data to obtain a marking power value and a speed value corresponding to the current part material;

[0010] A control module for controlling a laser marking device through the marking power value and the speed value to perform marking processing on the part to obtain marking process data;

[0011] A detection module for performing imaging detection on the marking result according to the marking process data, and performing edge analysis and depth analysis on the detection image to obtain marking quality data;

[0012] A comparison module for comparing and adjusting the marking process parameters through the marking quality data, and recording the adjusted parameters into a process database to obtain a marking control scheme.

[0013] A third aspect of the present invention provides a computer device, comprising: a memory and at least one processor, wherein instructions are stored in the memory; the at least one processor invokes the instructions in the memory to cause the computer device to execute the above-mentioned control method for product marking.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium, wherein instructions are stored in the computer-readable storage medium, and when it runs on a computer, it causes the computer to execute the above-mentioned control method for product marking.

[0015] In the technical solution provided by this application, the surface of the hardware components is scanned through image acquisition, and the material characteristics of the obtained surface image are analyzed to obtain the product material data and surface feature data, achieving the accurate identification of the component material and surface characteristics, eliminating the material judgment error caused by relying on manual experience in the traditional method, and providing accurate material basic information for the subsequent marking process; according to the product material data and surface feature data, the three-dimensional coordinate positioning of the components is carried out, and the positioning result is corrected through micro-displacement calibration to obtain the marking position coordinate data, solving the problem that the traditional two-dimensional positioning method cannot handle components with complex shapes, significantly improving the marking position accuracy, especially the positioning accuracy of non-planar areas such as curved surfaces and inclined surfaces has been fundamentally improved; according to the marking position coordinate data, the process parameters in the material library are queried and screened to obtain the marking power value and speed value corresponding to the current component material, and the mapping relationship between the material characteristics and the optimal process parameters is established through intelligent algorithms, breaking through the limitations of the traditional fixed parameter setting, realizing the personalized customization of parameters, and ensuring that components with different materials and shapes can obtain the most suitable marking parameters; the laser marking equipment is controlled by the marking power value and speed value to carry out marking processing on the components to obtain the marking process data, and at the same time, a real-time monitoring and feedback adjustment mechanism is introduced to change the marking process from open-loop control to closed-loop control, significantly improving the stability and adaptability of the marking process; according to the marking process data, the marking result is imaged and detected, and the edge analysis and depth analysis are carried out on the detected image to obtain the marking quality data, and a multi-dimensional quality evaluation system is introduced to comprehensively evaluate the marking effect from two levels of apparent quality and material microscopic changes, providing an objective and quantitative quality judgment basis; through the marking quality data, the marking process parameters are compared and adjusted, and the adjusted parameters are recorded in the process database to obtain the marking control scheme, forming a complete knowledge accumulation and optimization iteration mechanism, and continuously improving the marking quality with the accumulation of data. In addition, various artificial intelligence algorithms are applied in the whole process of this solution, such as the deep learning classification algorithm in material recognition, the machine learning regression algorithm in parameter optimization, and the computer vision algorithm in quality evaluation. The contributions of these algorithm features to the solution are as follows: the accurate identification of complex material characteristics is realized through the deep learning algorithm, breaking through the limitation that it is difficult to distinguish similar materials in the traditional method; the non-linear mapping relationship between the material characteristics and the marking parameters is established through the machine learning regression algorithm, realizing the accurate prediction and dynamic adjustment of the parameters; the automatic evaluation of the marking quality is realized through the computer vision algorithm, eliminating the subjectivity and inconsistency of manual detection. BRIEF DESCRIPTION OF THE DRAWINGS

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

[0017] Figure 1 It is a schematic diagram of an embodiment of the control method for product marking in the embodiments of the present application;

[0018] Figure 2 It is a schematic diagram of an embodiment of the control system for product marking in the embodiments of the present application;

[0019] Figure 3 It is a schematic block diagram of the structure of the computer device in the embodiments of the present invention. Specific embodiments

[0020] The embodiments of the present application provide a control method and system for product marking. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described here can be implemented in an order other than that illustrated or described here. In addition, the term "including" or "having" and any variation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device including a series of steps or units does not necessarily have to be limited to those clearly listed steps or units, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the following describes the specific process of the embodiments of the present application. Please refer to Figure 1 An embodiment of the control method for product marking in the embodiments of the present application includes:

[0022] Step S101: Scan the surface of the hardware parts through image acquisition, analyze the material characteristics of the obtained surface image, and obtain product material data and surface feature data;

[0023] Step S102: According to the product material data and surface feature data, perform three-dimensional coordinate positioning on the parts, and correct the positioning result through micro-displacement calibration to obtain the marking position coordinate data;

[0024] Step S103: According to the marking position coordinate data, query and filter the process parameters in the material library to obtain the marking power value and speed value corresponding to the current part material;

[0025] Step S104: Control the laser marking device according to the marking power value and speed value to perform marking processing on the component, and obtain the marking process data;

[0026] Step S105: According to the marking process data, perform imaging detection on the marking result, and perform edge analysis and depth analysis on the detected image to obtain the marking quality data;

[0027] Step S106: Compare and adjust the marking process parameters based on the marking quality data, and record the adjusted parameters into the process database to obtain the marking control scheme.

[0028] It can be understood that the execution subject of this application can be a control system for product marking, or a terminal or a server. Specifically, it is not limited here. In this embodiment of the application, the server is used as the execution subject for illustration.

[0029] Specifically, the surface of the hardware component is scanned through image acquisition, and the component is imaged and acquired from multiple angles by a high-resolution camera system. In practical applications, for a precision part of an automotive engine, the camera system will obtain the original image data from different lighting angles, and these data contain the light reflection characteristics of the component surface. Subsequently, material characteristic analysis is carried out. The polarization filtering technology is applied to the highly reflective metal surface area to eliminate the reflection interference, and the contrast is enhanced for the dark area to obtain an optimized surface image. Pixel analysis is performed on these images through feature extraction to identify the edge contour and surface texture of the component, and the morphological feature data is obtained. For parts made of stainless steel, the wavelength characteristics of the surface reflected light are measured through spectral analysis, and these data are compared with the data in the material parameter library to determine that it is made of 304 stainless steel, and its hardness and surface flatness parameters are recorded to form the product material data and surface feature data.

[0030] Three-dimensional coordinate positioning of the component is performed according to the product material data and surface feature data. Specifically, the surface of the part is scanned by a laser triangulation system to generate preliminary point cloud data, and a set of feature points, such as edge and corner regions, are extracted from it. A three-dimensional model of the component is constructed by combining the texture information in the surface feature data, and coordinate positioning is performed on the surface to select the area position suitable for marking. Taking a part of an automotive drive shaft as an example, the system identifies a flat area on its cylindrical surface as a marking candidate point, and performs micro-contact scanning through a micro-displacement actuator to obtain feedback data on the surface hardness and micro-topography. When it is found that there are small unevennesses on the surface, the system will perform coordinate correction to generate accurate marking position coordinate data.

[0031] Then, based on the marking position coordinate data, the process parameters in the material library are queried and filtered. The system extracts the part material type identifier from the product material data, queries the material library to obtain the basic process parameter table for this material. For aluminum alloy parts, the system calculates the correction factor according to the flatness information in the surface feature data, conducts a thermal field distribution analysis, predicts the response of the material at different powers, and generates a power response curve. Combining the curvature and thickness characteristics of the marking position, the geometric shape factor is calculated to determine the power range that meets the quality requirements. For 2-mm-thick aluminum alloy parts, the system determines that a marking power of 25 W and a speed value of 80 mm / s are the most suitable.

[0032] Then, the marking power value and speed value are used to control the laser marking equipment for marking processing. The five-axis linkage precision positioning platform moves the part to the working area. After precise positioning, a micro pre-test is carried out in the non-critical area, and a dot matrix test is carried out at 10% power to obtain the initial material response data. The parameters are fine-tuned based on these data to form real-time marking parameters. The laser control system starts the marking process, and at the same time, the spectral changes on the material surface are recorded through spectral monitoring, and the acoustic wave frequency data during the marking process are collected through an acoustic sensor. For precision bearing parts, the system records that the spectral change on the material surface drops from a reflectivity of 98% to 92%, and the acoustic wave frequency stabilizes around 18 kHz, indicating that the marking depth and width are controlled within the ideal range. These data are integrated with the execution record to form the marking process data.

[0033] Subsequently, based on the marking process data, the marking result is subjected to imaging inspection. The focus stacking imaging device performs multi-focal plane scanning on the marked part, obtains an image sequence with different depths of field and fuses them to form a high-depth-of-field marking result image. The marking contour is pixel-level depicted through edge extraction, and the edge sharpness coefficient and contour consistency value are calculated. The three-dimensional optical interferometer performs a depth scan to obtain the marking depth distribution map, and combines spectral analysis to detect the molecular structure changes in the marking area. For the QR code marking on medical device components, the system measures that the edge sharpness coefficient reaches 0.92, the contour consistency value is 0.95, the average marking depth is 0.08 mm, and the coefficient of variation is less than 3%, indicating good marking quality. These indicators are comprehensively formed into the marking quality data.

[0034] Compare and adjust the marking process parameters based on the marking quality data. Compare the quality data with the preset threshold values to generate a quality deviation matrix, analyze the correlation between the marking power and speed and the quality indicators, and establish a parameter-quality correlation model. Calculate the direction and amplitude of parameter adjustment to form parameter correction values and generate an optimized combination of process parameters. Conduct a correlation analysis with the historical successful case data to verify the rationality of the adjustment. For the marking of aviation components, when it is found that the edge sharpness is insufficient, the system reduces the marking speed from 100 mm / s to 85 mm / s while keeping the power unchanged, and stores this adjustment together with the product material data and surface feature data in the process database to form a marking control plan.

[0035] In a specific embodiment, the process of performing step S101 may specifically include the following steps:

[0036] (1) Collect the image data of the component to be marked under multi-angle illumination through a high-resolution camera system to obtain the original image data of the component surface;

[0037] (2) According to the original image data, perform polarization filtering on the reflective area of the component surface and contrast enhancement on the dark area to obtain an optimized surface image;

[0038] (3) Perform pixel analysis on the optimized surface image through feature extraction to identify the edge contour and surface texture of the component and obtain the morphological feature data of the component;

[0039] (4) Classify the material of the component surface according to the morphological feature data and match and compare it with the data in the material parameter library to obtain the material type data of the component;

[0040] (5) Conduct wavelength analysis on the surface reflected light of the component through spectral analysis to distinguish the spectral characteristics of the three materials of metal, plastic, and ceramic and obtain the material characteristic parameters;

[0041] (6) Integrate the morphological feature data, material type data, and material characteristic parameters to construct product material data and surface feature data including the surface flatness, material type, and surface texture of the component.

[0042] Specifically, a high-resolution camera system is used to collect images of the parts to be marked under multi-angle illumination, obtaining the complete information on the surface of the parts. The high-resolution camera system refers to an industrial camera equipped with at least 12 million pixels, combined with an adjustable light source array, which can illuminate the surface of the parts from different angles. Multi-angle illumination means illuminating the parts from at least three different directions, usually including top direct light, side 45-degree angle light source, and annular diffused light source, to ensure that all details on the surface of the parts are captured. The original image data contains basic information such as the reflection situation, texture features, and color distribution on the surface of the parts, and these data are stored as a high-bit-depth image matrix. After collecting the original image data, it enters the image preprocessing stage. Polarization filtering is performed on the reflective areas on the surface of the parts. Polarization filtering means using a polarizer to screen the reflected light, allowing only light with a specific vibration direction to pass through, thereby eliminating the overexposed areas caused by strong reflections. Specifically, during the processing, the brightness of the original image is analyzed, and the areas with brightness values exceeding the threshold are identified and marked as reflective areas, and then the polarization filtering algorithm is applied to these areas to adjust the light intensity. At the same time, contrast enhancement is performed on the dark areas. The gray range of the dark areas is stretched through histogram equalization technology to increase the image contrast. Dark areas refer to the image areas with brightness values lower than a certain threshold, and contrast enhancement is to broaden the narrow gray interval through a mapping function to improve the image recognition. The combination of these two steps results in an optimized surface image. The optimized surface image refers to an image that can clearly display the surface features of the parts after the reflection and dark problems are corrected.

[0043] Pixel analysis is performed on the optimized surface image through feature extraction. Feature extraction refers to the process of extracting feature information such as the contour and texture of an object from an image using edge detection algorithms. Pixel analysis refers to the calculation and analysis of the gray value, color value of each pixel point in the image, and its relationship with neighboring pixels. In specific implementation, the Canny edge detector is used to process the image to extract the edge contour of the component; then the gray-level co-occurrence matrix is used to quantitatively analyze the surface texture, and texture parameters such as energy, contrast, and entropy are calculated; these information are integrated to form the morphological feature data of the component. The morphological feature data is a data set describing the shape and surface characteristics of the component, including contour information, edge sharpness, surface roughness parameters, etc. According to the morphological feature data, the material classification of the component surface is carried out. Material classification refers to the process of classifying components into a specific material type according to surface characteristics. In specific implementation, first, a feature vector library containing the texture characteristics of common industrial materials is established, then the morphological feature data of the component to be tested is converted into a feature vector, and through similarity calculation, it is matched and compared with the known materials in the material parameter library. The material parameter library is a pre-established database containing the characteristic parameters of various common industrial materials. During the comparison process, the Euclidean distance between the morphological feature vector and each material in the library is calculated, and the material type with the smallest distance is selected as the matching result to obtain the material type data of the component. The material type data is a data label indicating the material type of the component, usually the code or name of the material.

[0044] Wavelength analysis is performed on the reflected light from the component surface through spectral analysis to verify the material type. Spectral analysis is a method of measuring the reflection spectrum to determine the material type by using the characteristic that different materials have different reflectivities for light of each wavelength. Wavelength analysis refers to the intensity analysis of different wavelength components of the reflected light to obtain the reflection spectrum curve. In actual operation, a spectral analyzer is used to scan the surface of the component, record the reflectivity data from the visible light to the near-infrared region, form the reflection spectrum curve, and distinguish the three materials of metal, plastic, and ceramic by analyzing the positions of the spectral characteristic peaks and valleys. Metal materials usually have a higher reflectivity in the near-infrared region, plastic materials have characteristic absorption bands in specific visible light bands, and ceramic materials show different spectral characteristics. Through these feature differentiations, the material characteristic parameters are obtained. The material characteristic parameters refer to the specific values of the optical characteristics such as the reflectivity and absorptivity of the material, and the material property indexes derived therefrom.

[0045] Fuse the morphological feature data, material type data, and material property parameters to construct product material data and surface feature data. Data fusion is the process of comprehensively processing multi-source data to form more complete information. During the fusion process, standardize the three types of data to make their dimensions consistent; then use the weighted average method to integrate the standardized data, and the weights are determined according to the reliability and importance of each data; form a comprehensive data packet containing the material and surface characteristics of the parts. This data packet includes key information such as surface flatness, material type, and surface texture, providing a data basis for subsequent marking position determination and parameter selection.

[0046] For example, mark a stainless steel precision joint. Use an industrial camera equipped with a ring-shaped LED light source to collect original images from three different angles, and it is found that there are severely reflective polished areas and darker threaded areas on the surface of the joint. Solve the highlight problem of the polished area through polarization filtering, reducing the pixel value of the reflective area from the overexposed 255 to about 180 that can be recognized; enhance the contrast of the threaded dark area to improve the visibility of details in the dark area. Then perform edge detection and texture analysis on the processed image to extract the precise contour and surface texture parameters of the joint, forming morphological feature data. Based on the comparison of these data with the material parameter library, it is initially judged to be made of stainless steel. Measure the surface reflection spectrum through a spectral analyzer and record the reflectivity curve in the wavelength range of 450 - 900 nm, which is confirmed to be the characteristic spectrum of 304 stainless steel. Integrate these data to construct a comprehensive data packet containing surface flatness, material confirmation information, and surface texture parameters, providing an accurate basis for determining the subsequent laser marking power and speed parameters, and ensuring that marking processing is carried out at the appropriate position with appropriate parameters.

[0047] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0048] (1) Perform surface scanning on the parts through a laser triangulation system to generate preliminary point cloud data of the parts;

[0049] (2) According to the preliminary point cloud data, extract feature points from the edge and corner regions of the parts to obtain a set of feature points of the parts;

[0050] (3) Perform image processing on the set of feature points through a machine vision system, and combine the texture information in the surface feature data to construct a three-dimensional model of the parts;

[0051] (4) According to the three-dimensional model, perform coordinate positioning on the surface of the parts, select marking candidate points according to the marking required area, and obtain a preliminary coordinate positioning result;

[0052] (5) Perform micro-contact scanning on the surface of the parts through a micro-displacement actuator to obtain feedback data on surface hardness and micro-topography;

[0053] (6) Compare and calibrate the feedback data with the preliminary coordinate positioning result to eliminate the coordinate deviation caused by the part shape, and generate the marking position coordinate data.

[0054] Specifically, the surface of the part is scanned by a laser triangulation system to generate preliminary point cloud data. The laser triangulation system is a non-contact measurement device, which consists of a laser emitter, a CCD sensor and a processing unit. The working principle is that after the laser beam irradiates the surface of the part, it is reflected onto the CCD sensor, and the surface undulation is measured by calculating the position change of the laser line on the sensor. During the scanning process, the laser line moves along the surface of the part according to a predetermined path, forming a set of high-density sampling points. Each point contains its three-dimensional spatial coordinate (x, y, z) information, constituting the preliminary point cloud data. The preliminary point cloud data is a set of three-dimensional coordinate points describing the surface shape of the part, usually containing thousands to millions of discrete points, and each point accurately records the position information of the part surface. According to the obtained preliminary point cloud data, the feature points of the edge and corner regions of the part are extracted next. Feature point extraction is the process of screening representative key points from a large amount of point cloud data. The edge refers to the contour line where the surface of the part changes significantly, and the corner region is the position where the surface normal vector changes violently. In the extraction process, the local curvature value of each point is calculated to identify the region with significant curvature change; then the region growing algorithm is applied to cluster adjacent high-curvature points to form the edge and corner regions; representative points are selected from these regions to obtain the set of feature points of the part. The set of feature points refers to the set of points selected from the preliminary point cloud data that can represent the key structural features of the part. The number is much less than the original point cloud, but it contains the main geometric feature information of the part.

[0055] Image processing is performed on the set of feature points through a machine vision system, and combined with the texture information in the surface feature data, a three-dimensional model of the part is constructed. The machine vision system refers to a visual information acquisition and analysis system composed of an optical sensor, image processing hardware and algorithm software. Image processing includes operations such as filtering and denoising the feature points, spatial transformation and geometric relationship analysis. Specifically, Delaunay triangulation is performed on the set of feature points to establish the connection relationship between points; then combined with the texture information in the surface feature data, the surface is subdivided and supplemented to enhance the model details; the surface fitting algorithm is applied to generate a continuous surface model to form the three-dimensional model of the part. The three-dimensional model refers to a digital representation containing the complete geometric shape and surface characteristics of the part, which is more structured and complete than the point cloud data.

[0056] According to the constructed three-dimensional model, the surface of the component is coordinate located and candidate points for marking are selected. Coordinate positioning is the process of determining the precise position and posture of the component in the workspace. In the specific operation, a working coordinate system is established, usually based on the working platform of the laser marking equipment; then the three-dimensional model is aligned with the working coordinate system to eliminate rotation and translation deviations; then, according to the marking requirement area, positions that meet the marking conditions are selected on the model surface as candidate points, and factors to be considered include surface flatness, accessibility, and visibility of the marking effect. Candidate points for marking refer to positions on the surface of components that are suitable for marking operations. Each candidate point contains its three-dimensional coordinate value and surface normal vector information in the working coordinate system to form a preliminary coordinate positioning result.

[0057] Micro-contact scanning of the surface of the component is performed through a micro-displacement actuator to obtain more accurate surface information. A micro-displacement actuator is a mechanical device that can achieve micron-level precise movement, usually composed of a precision motor, a guide rail, and a sensor. Micro-contact scanning refers to a detection method in which a probe touches the surface of the component and records the contact feedback. During the scanning process, the probe contacts the surface of the component with extremely small pressure and moves along a predetermined path, while measuring the surface hardness and microscopic morphology. The surface hardness is calculated by measuring the relationship between the probe pressure and the sinking depth, and the microscopic morphology is obtained by recording the change in the probe displacement. This contact scanning can capture tiny features that cannot be accurately obtained by laser measurement, and the feedback data obtained includes the surface hardness value and the micron-level surface morphology description.

[0058] Compare and calibrate the feedback data with the preliminary coordinate positioning results to generate the marking position coordinate data. Comparison and calibration is the process of comparing and analyzing the data obtained by the two measurement methods to eliminate errors. In specific implementation, the feedback data obtained by micro-contact scanning is aligned with the corresponding area in the three-dimensional model to calculate the position deviation; then the preliminary coordinate positioning results are corrected according to the deviation value to eliminate the coordinate deviation caused by factors such as surface curvature and uneven material of the parts; and the corrected marking position coordinate data is generated, including accurate marking position coordinates, marking angle and depth parameters.

[0059] Taking a precision gear of an automotive transmission as an example, this gear needs to be marked with a serial number and production batch information at specific positions. The laser triangulation system scans the gear surface at a spacing of 0.1 mm, generating preliminary point cloud data containing approximately 100,000 points. Then, the curvature distribution of the point cloud data is analyzed, and 800 feature points at the gear contour edges and tooth groove corners are extracted to form a set of feature points. The machine vision system processes these feature points, combines the previously obtained gear surface texture information, and constructs an accurate three-dimensional gear model through triangulation and surface fitting. According to the three-dimensional model, the planar area on the gear side is identified as the ideal marking area, and the center point coordinates are selected as the preliminary coordinate positioning result. To ensure the marking accuracy, the micro-displacement actuator performs a micro-contact scan on this area and finds that there are wavy undulations of approximately 20 microns and slight hardness variations on the surface. Based on this feedback data, the preliminary coordinates are corrected, the marking position is offset by 1.5 mm towards the area with more uniform hardness, and the marking angle is adjusted by 3 degrees to adapt to the surface micro-inclination, generating marking position coordinate data containing accurate coordinates, angles, and depth parameters to ensure that subsequent laser marking can be carried out at the best position and in the best posture, improving the marking quality and consistency.

[0060] In a specific embodiment, the process of performing step S103 may specifically include the following steps:

[0061] (1) Extract the material type identifier of the component from the product material data, index and query the material library, and obtain the basic process parameter table;

[0062] (2) Calculate the correction factor for the basic process parameter table according to the surface flatness information in the surface feature data to obtain the surface adaptation parameter;

[0063] (3) Analyze the thermal field distribution of the component through heat conduction calculation, combine the material characteristic parameters, predict the material response under different powers, and generate a power response curve;

[0064] (4) Analyze the curvature and thickness of the marking area according to the geometric features in the marking position coordinate data, and calculate the geometric shape factor;

[0065] (5) Combine the geometric shape factor with the power response curve, calculate the power range that meets the marking quality requirements, and determine the marking power value corresponding to the current component material;

[0066] (6) Calculate the optimal heat input rate according to the marking power value and the material thermal conductivity, convert it into the speed value of the marking head movement, and obtain the marking power value and speed value corresponding to the current component material.

[0067] Specifically, extract the material type identifier of the component from the product material data and perform an indexed query on the material library. The material type identifier is a code that uniquely identifies the material type and usually contains information such as material category, grade, and characteristics. For example, for metal materials, the identifier might be "M-304SS-H", where M represents the metal category, 304SS represents 304 stainless steel, and H represents hardening treatment. The extraction process is to separate the material identification part from the structured information of the product material data and then use this identifier to perform an exact match query in the material library. The material library is a data warehouse containing the marking process parameters of various industrial materials, and the indexed query is a process of quickly locating the corresponding material parameters through the material identifier. The basic process parameter table returned by the query contains the laser marking parameters of this material under standard conditions, such as basic parameters like power range, pulse frequency, and scanning speed.

[0068] Calculate the correction factor for the basic process parameter table based on the surface flatness information in the surface feature data. The surface flatness information describes the roughness of the component surface and includes indicators such as root mean square roughness value and peak-to-valley value. The correction factor calculation is a process of adjusting the standard parameters according to the actual surface condition. The calculation formula is as follows:

[0069]

[0070] Where C sf represents the surface correction factor, α, β, γ are weight coefficients, R actual is the actual surface roughness, R standard is the roughness corresponding to the standard parameter, W surface is the surface waviness, W baseline is the reference waviness, P reflectivity is the surface reflectivity, P base is the reference reflectivity. After calculating the correction factor, multiply the basic process parameters by the correction factor to obtain the surface adaptation parameters, which have considered the influence of the surface characteristics of a specific component.

[0071] Perform a thermal field distribution analysis on the component through heat conduction calculation, and combine with the material characteristic parameters to predict the material response under different powers. Heat conduction calculation is a mathematical method to simulate the propagation law of heat energy in materials, and thermal field distribution analysis is a process of calculating the temperature distribution inside the material. The heat conduction partial differential equation is used in the calculation:

[0072]

[0073] Where ρ is the material density, c p is the specific heat capacity, T is the temperature, t is the time, k is the thermal conductivity, Q laseris the laser heat source term. The material property parameters include physical properties such as the reflectivity, absorptivity, and thermal conductivity of the material. By solving the equation, the temperature field evolution of the material under different laser powers can be predicted, and then the relationship curve of the material surface temperature varying with the power can be obtained, which is the power response curve. The power response curve describes the relationship between the laser power and parameters such as the material surface temperature and melting depth, and is an important basis for subsequent parameter optimization.

[0074] According to the geometric features in the marking position coordinate data, analyze the curvature and thickness of the marking area, and calculate the geometric shape factor. Geometric features refer to the shape characteristics of the marking position, such as plane, curved surface, edge, etc. Curvature describes the degree of surface bending, and thickness represents the thickness value of the material at the marking position. The formula for calculating the geometric shape factor is:

[0075]

[0076] where, G f represents the geometric shape factor, δ is the reference coefficient, η is the curvature influence coefficient, K surface is the principal curvature of the surface, ζ is the thickness influence coefficient, D min is the minimum thickness of the marking position, D ref is the reference thickness. The geometric shape factor reflects the influence degree of the part shape on the marking parameters. The larger the value, the greater the required parameter deviation.

[0077] Combine the geometric shape factor with the power response curve to calculate the power range that meets the marking quality requirements. The combination process uses the following formula:

[0078]

[0079] where, P optimal is the calculated optimized power value, P base is the base power value, C sf is the surface correction factor, G f is the geometric shape factor, λ is the temperature adjustment coefficient, T required is the material surface temperature required for marking, T base is the reference temperature. By substituting different target temperatures, calculate the corresponding power values, select the power interval that meets the marking quality requirements from them, and determine the marking power value corresponding to the current part material.

[0080] According to the marking power value and the material thermal conductivity, calculate the optimal heat input rate and convert it into the speed value of the marking head movement. The heat input rate is the thermal energy input to the material per unit time, and the material thermal conductivity describes the ability of the material to conduct heat. The calculation uses the following formula:

[0081]

[0082] wherein, v optimal is the optimal moving speed of the marking head, P optimal is the optimized power value, τ absorption is the light energy absorption efficiency of the material, E required is the energy required for marking per unit length, L spot is the laser spot diameter, θ is the thermal conduction influence coefficient, k material is the thermal conductivity of the material, k ref is the reference thermal conductivity.

[0083] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0084] (1) Move the component to the marking working area through a five-axis linkage precision positioning platform, and perform precise positioning according to the marking position coordinate data to obtain the actual position information of the component in the working area;

[0085] (2) Conduct a micro pre-test on the non-critical area of the component, and perform a dot matrix test using 10% of the marking power value to obtain the initial response data of the material;

[0086] (3) Fine-tune the marking power value and speed value according to the initial response data of the material to establish the real-time marking parameters of the component;

[0087] (4) Through the laser control system, start the laser marking process according to the real-time marking parameters, and simultaneously monitor the spectrum of the marking area to record the spectral change data of the material surface;

[0088] (5) Collect the sound wave frequency generated during the marking process through an acoustic sensor, perform signal analysis on the sound wave frequency, and extract the acoustic characteristics during the marking process;

[0089] (6) Integrate the spectral change data and acoustic characteristics with the execution records of the marking power value and speed value to form the marking process data.

[0090] Specifically, the component is moved to the marking work area by a five-axis linkage precision positioning platform. The five-axis linkage precision positioning platform is a high-precision mechanical device with three linear motion axes of X, Y, and Z and two rotary axes of A and B, capable of adjusting the position and attitude of the component at any position in space. The platform receives the marking position coordinate data as input and converts it into motion commands for each axis. The conversion process includes coordinate system conversion and inverse kinematic calculation to determine the displacement and rotation angle of each axis. Precise positioning means moving the component to the predetermined marking position and adjusting the attitude simultaneously to keep the marking surface at the best incident angle with the laser beam. After positioning, the actual position of the component in the work area is measured and confirmed by a laser displacement sensor and mechanical contacts to form actual position information. This information includes the coordinates of the component center point, the surface normal vector, and the deviation value from the theoretical position, which is used for precise adjustment of subsequent marking parameters.

[0091] Before formal marking, conducting a micro pre-test on the non-critical area of the component is a crucial step to ensure marking quality. The non-critical area refers to the secondary positions on the component that do not affect its function and appearance, usually selected at the edges, bottom, or areas that will be blocked. The micro pre-test means conducting a trial marking with a low-power laser in a small area. During specific operation, 10% of the marking power value is taken as the test power, which can not only observe the response characteristics of the material but also cause no obvious damage to the component. The dot matrix test is to mark multiple points arranged in a matrix, and each point uses a slightly different parameter combination. After the test, the test area is photographed by an image acquisition device to analyze the interaction effect between the laser and the material and obtain the initial response data of the material. These data include characteristic indexes reflecting the response of the material to the laser, such as the degree of color change, ablation depth, and edge sharpness of the material.

[0092] Based on the initial response data of the material, the marking power value and speed value are finely adjusted to establish real-time marking parameters suitable for the actual situation of the current component. The fine adjustment process analyzes the marking effects of different parameter combinations in the dot matrix test and identifies the parameter combination closest to the ideal effect. Then, parameter correction is carried out according to the surface characteristic differences between the test points and the target marking area. The correction considerations include surface reflectivity differences, curvature changes, and minor material changes, etc. Linear or non-linear adjustments are made to the marking power value and speed value during calculation to form the compensated parameters. Real-time marking parameters refer to the parameter combinations such as power, speed, and pulse frequency after fine adjustment, and these parameters will be dynamically adjusted according to real-time monitoring data during the marking process.

[0093] Through the laser control system, the laser marking process is started according to the real-time marking parameters. The laser control system is a collection of hardware and software that controls the laser output characteristics, including the laser source, optical path control, galvanometer system, and control software. Starting the marking process includes four stages: laser preheating, parameter loading, galvanometer calibration, and marking execution. At the same time, spectral monitoring of the marking area is a key link to achieve closed-loop control. Spectral monitoring refers to using a spectral analyzer to capture the light signals emitted and reflected by the marking area in real time, and analyzing their wavelength distribution and intensity changes. The monitoring device is usually installed at a position coaxial with the laser, and can capture the spectral changes on the material surface during the marking process. After the spectral data is processed by background noise filtering, peak identification, and feature extraction, it is converted into spectral change data representing the change of the material state. These data reflect the phase change, temperature change, and chemical composition change of the material during the marking process, and are an important basis for judging whether the marking process is within the control range.

[0094] The acoustic frequency generated during the marking process is collected by an acoustic sensor to monitor the marking quality from another dimension. An acoustic sensor is a device that converts acoustic signals into electrical signals, usually using a high-sensitivity microphone array installed around the marking area. The acoustic frequency refers to the speed of sound vibration, and the acoustic frequency characteristics generated by different materials under the action of laser are different. The collected original sound signal is processed by amplification and filtering, and then converted to the frequency domain using the fast Fourier transform algorithm to form a spectrogram. Peak detection and pattern recognition are performed on the spectrogram to extract the frequency components and intensity distributions representing specific marking states, which are the acoustic characteristics. There is a clear corresponding relationship between the acoustic characteristics and the material type, thickness, and marking depth, and they can indirectly reflect the marking depth and quality.

[0095] The spectral change data and acoustic characteristics are integrated with the execution records of the marking power value and speed value to form the marking process data. Data integration is the process of organizing multi-source heterogeneous data into a unified structure. During integration, the time synchronization of each data is performed to ensure that the data collected at different times can be correctly corresponded; then data cleaning is performed to remove outliers and noise; the data is organized according to a predetermined format to form a structured marking process data set. This data set includes timestamps, marking positions, actual executed power and speed values, spectral data and acoustic data at corresponding times, etc., and completely records the parameter changes and material responses during the marking process, providing a data basis for subsequent quality analysis and parameter optimization.

[0096] For example, regarding the precision parts of automobile fuel injectors, the parts need to be marked with batch numbers and traceability codes. The five-axis linkage precision positioning platform moves this aluminum alloy part to the work area, and adjusts the part position and posture according to the marking position coordinate data determined in the early stage, so that the cylindrical surface of the injector housing forms an 80-degree incident angle with the laser beam. The laser displacement sensor measures the actual position and the preset position deviation is less than 0.02mm, and the actual position information is recorded. Then select a 5mm×5mm non-critical area at the bottom edge of the part, use 2.5W power (10% of the original 25W) for 3×3 dot matrix testing, and use slightly different parameter combinations for each point. The image analysis test results show that the edge clarity and depth of the test point are most ideal at a power of 2.5W and a speed of 65mm / s. Taking into account the difference in surface polish between the test area and the target marking area, the power is increased by 1.5% and the speed is reduced by 2% to establish real-time marking parameters. Then, the formal marking was started according to the adjusted parameters. At the same time, the spectrum analyzer monitored the marking area and recorded the transformation process of the material surface from the normal aluminum alloy spectrum to the characteristic peak of aluminum oxide, indicating that the expected oxide layer was formed by marking. The acoustic sensor also captured the characteristic sound in the 15-20kHz frequency band, and extracted the stable main frequency component through spectrum analysis, which matched the known good marking acoustic characteristics. The time-synchronized spectral changes, acoustic characteristics and actual power speed data were integrated to form the marking process data. These data clearly recorded the parameter execution and material response of the entire marking process, ensuring the high quality and traceability of the key automotive parts identification.

[0097] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0098] (1) Perform multi-focal plane scanning on the marked parts using a high-resolution focus stacking imaging device to obtain image sequences of the marked area with different depths of field;

[0099] (2) The image sequence is fused through image synthesis technology to form a high depth of field marking result image;

[0100] (3) Based on the high depth of field marking result image, the marking contour is depicted at the pixel level through the edge extraction algorithm, and the sharpness coefficient and contour consistency value of the marking edge are calculated;

[0101] (4) Performing a depth scan of the marking area using a three-dimensional optical interferometer to obtain a marking depth distribution map and calculate the marking depth mean and coefficient of variation;

[0102] (5) Based on the spectral change data in the marking process data, combined with Raman spectroscopy analysis, the molecular structure changes in the marking area are detected to obtain the material state change index;

[0103] (6)Comprehensively analyze the sharpness coefficient, contour consistency value, average marking depth, coefficient of variation, and material state change index to form marking quality data.

[0104] Specifically, use a high-resolution focus stacking imaging device to perform multi-focus plane scanning on the marked parts to obtain multi-layer images of the marking area. The high-resolution focus stacking imaging device is a special optical imaging device that can collect images at different focal lengths to solve the problem of shallow depth of field in conventional microscopic imaging. Multi-focus plane scanning refers to the process of moving the focal plane in the vertical direction with a small step size and taking a clear image at each focal plane position. In actual operation, the device starts from above the marking surface and moves the focal plane downward at intervals of 5-10 microns to obtain 10-30 images with different depths of field. Each image only contains the part located at the current focal plane that is clear, while the other parts are blurred. This series of images constitutes the image sequence of the marking area.

[0105] Fuse the obtained image sequence through image synthesis technology to form a high-depth-of-field marking result image with all areas clear. Image synthesis technology is a processing method that combines multiple images with complementary information into a new image. In the application of focus stacking, a fusion strategy based on sharpness evaluation is mainly adopted. In the specific implementation process, each image is divided into blocks, usually the image is divided into small pieces; then the sharpness evaluation value is calculated for each small piece, and common evaluation functions include gradient operators, Laplace operators, or wavelet transform coefficients; then the blocks with the highest sharpness are selected from each layer for splicing; the edge of the splicing result is smoothed to eliminate the discontinuity of the transition between blocks. After this processing, the obtained high-depth-of-field marking result image contains clear details of all depth positions in the marking area, providing high-quality visual data for subsequent analysis.

[0106] Based on the high-depth-of-field marking result image, pixel-level description of the marking contour is performed through an edge extraction algorithm. The edge extraction algorithm is a calculation method for locating object boundaries in digital images. In marking quality assessment, the Canny edge detector is mainly used. The Canny algorithm includes four steps: Gaussian filtering, gradient calculation, non-maximum suppression, and double-threshold linking. Apply Gaussian filtering to the image to reduce noise; then calculate the gradient magnitude and direction of the image; then perform non-maximum suppression to retain the local maximum points in the gradient direction; apply the double-threshold method to determine the edge points and connect them. Through this processing, the position and shape of the marking contour are accurately extracted from the high-depth-of-field image. Based on the extracted contour information, two important indicators are calculated: the sharpness coefficient and the contour consistency value. The sharpness coefficient measures the clarity of the marking edge, and the calculation method is to statistically calculate the gradient intensity of the pixel gray values on both sides of the contour line; the contour consistency value evaluates the degree of coincidence between the marking contour and the ideal contour, and is quantified by calculating the deviation between the actual contour and the ideal contour.

[0107] The marking area is scanned in depth by a three-dimensional optical interferometer to obtain three-dimensional topography data of the marking. The three-dimensional optical interferometer is a precision instrument that measures minute height changes on the surface using the coherence of light and can achieve three-dimensional measurements with nanometer-level resolution. During the depth scanning process, the beam emitted by the interferometer is divided into a reference beam and a measurement beam. After the measurement beam is reflected, it coincides with the reference beam to generate interference fringes, and the surface height information is obtained by analyzing the distribution of the interference fringes. After the scanning is completed, a marking depth distribution map is generated, which is a pseudo-color image representing the depth change in the marking area, where the color of each pixel corresponds to a depth value. Based on the depth distribution map, two statistical parameters are calculated: the mean marking depth and the coefficient of variation. The depth mean is the average of the depth values of all measurement points and reflects the overall depth of the marking; the coefficient of variation is the ratio of the standard deviation to the mean and reflects the uniformity of the depth distribution. The smaller the value, the more uniform the marking depth.

[0108] According to the spectral change data in the marking process data, combined with Raman spectroscopy analysis, the material structure of the marking area is detected. Raman spectroscopy analysis is a technique that analyzes the molecular structure of materials using the scattering spectrum generated by the interaction between incident light and molecular vibrations of materials. In marking quality assessment, a Raman spectrometer is used to perform comparative measurements on the marked area and the unmarked area to obtain Raman spectra of the two areas. By comparing the changes in the position, intensity, and shape of the spectral peaks, the changes in the molecular structure of the material during the marking process are identified. For example, oxidation on the metal surface will produce characteristic Raman peaks of oxides, and thermal decomposition of organic materials will cause the appearance of characteristic peaks of carbonized structures. These spectral changes are quantified as material state change indicators, reflecting the degree of influence of marking on material properties.

[0109] The sharpness coefficient, contour consistency value, mean marking depth, coefficient of variation, and material state change indicator are comprehensively analyzed to form marking quality data. In the comprehensive analysis process, each index is normalized so that its range is unified to between 0 and 1; then, according to the requirements of different products and application scenarios, weight coefficients are assigned to each index; the weighted comprehensive score is calculated and compared with the preset threshold to obtain a quantitative evaluation result of the marking quality. The marking quality data is a structured data set containing the original values, normalized values, weight values, and comprehensive scores of each index, which not only reflects the apparent quality of the marking but also contains information on changes in the internal structure of the material.

[0110] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0111] (1) Compare the sharpness coefficient, contour consistency value, mean marking depth, coefficient of variation, and material state change indicator in the marking quality data with the preset quality threshold to generate a quality deviation matrix;

[0112] (2) According to the quality deviation matrix, perform a correlation analysis on the marking power value and the speed value, and establish a parameter-quality correlation model;

[0113] (3) Calculate the adjustment direction and amplitude of the marking process parameters through the parameter-quality correlation model to form a parameter correction value;

[0114] (4) Combine the marking power value and the speed value with the parameter correction value to generate an optimized process parameter combination;

[0115] (5) Perform a correlation analysis on the optimized process parameter combination and the historical successful case data to verify the rationality of the parameter adjustment and form a verified parameter plan;

[0116] (6) Associate and store the verified parameter plan with the product material data, surface feature data, and marking position coordinate data in the process database to form a marking control plan.

[0117] Specifically, the optimization and iteration process of the product marking control method compares each index in the marking quality data with the preset quality threshold to generate a quality deviation matrix. The quality deviation matrix is a structured data set that describes the degree of deviation of each quality index from the target value. During specific calculation, first subtract the measured data such as the sharpness coefficient, contour consistency value, average marking depth, coefficient of variation, and material state change index from the corresponding preset thresholds to obtain the deviation values in each dimension. A positive deviation indicates better than expected, and a negative deviation indicates lower than expected. Subsequently, perform a normalization process on each deviation value to unify the deviation magnitudes of each index within a comparable range to form a standardized quality deviation matrix. This matrix intuitively reflects the performance of the current marking result in each quality dimension and provides a data basis for subsequent parameter adjustment.

[0118] According to the generated quality deviation matrix, perform a correlation analysis on the marking power value and the speed value, and establish a parameter-quality correlation model. Correlation analysis is a statistical method for exploring the strength of the relationship between variables, and the parameter-quality correlation model is a mathematical expression that describes the influence law of process parameter changes on quality indicators. During the establishment process, collect historical marking data, including the quality evaluation results under different power and speed combinations; then calculate the correlation coefficients between the process parameters and each quality index to quantify the influence direction and degree of parameter changes on quality; construct a multiple regression model, taking parameters such as power and speed as independent variables and quality indicators as dependent variables, and fitting out the mathematical relationship between parameters and quality. Through this model, the influence effect of parameter adjustment on each quality index can be predicted, providing a theoretical basis for the optimization of marking process parameters.

[0119] Calculate the adjustment direction and amplitude of the marking process parameters through the parameter-quality correlation model to form a parameter correction value. The parameter correction value is calculated using the following formula:

[0120]

[0121] where, ΔP i represents the correction value of the i-th process parameter (such as power, speed), represents the parameter-quality sensitivity coefficient, reflecting the influence degree of parameter change on the quality index, D j is the standardized deviation value of the j-th quality index, W j is the weight coefficient of the j-th quality index, representing its importance, S pi is the adjustment step coefficient of the i-th parameter, controlling the adjustment amplitude, Q j represents the value of the j-th quality index, P i represents the value of the i-th process parameter. In practical applications, the partial derivatives of each parameter with respect to the quality index are extracted from the parameter-quality correlation model as the sensitivity coefficients; then the standardized deviation values in the quality deviation matrix are substituted; the importance weights of the quality indexes in different application scenarios are considered; the adjustment amplitude is controlled by the step coefficient to ensure that the parameter change can effectively improve the quality without causing violent fluctuations in the marking state. The parameter correction value calculated by this method not only considers the directionality of quality improvement but also takes into account the balance among multiple indexes, avoiding the adjustment risk of neglecting one thing while attending to another.

[0122] Combine the marking power value and speed value with the parameter correction value to generate an optimized process parameter combination. The parameter combination is the process of performing mathematical operations on the original parameters and the correction values to obtain new parameters. Specifically, when operating, obtain the original marking power value and speed value; then add the power correction value to the original power value and add the speed correction value to the original speed value to obtain the adjusted power value and speed value; subsequently, check whether the adjusted parameters exceed the parameter range allowed by the equipment, and if so, truncate them to the boundary values; together with other unadjusted parameters (such as pulse frequency, focal length, etc.), form a process parameter combination. Conduct a correlation analysis on the optimized process parameter combination and the historical successful case data to verify the rationality of the parameter adjustment. The correlation analysis is the process of comparing the newly generated parameter scheme with the existing empirical data to evaluate its feasibility. Specifically, when implementing, retrieve historical successful cases similar to the current part material, shape, and application requirements from the process database; then compare the similarity of the optimized parameters with the parameters of these cases, including the absolute difference and relative difference rate of each parameter; if the similarity is higher than the preset threshold, it is considered that the adjustment scheme is supported by historical experience; if the similarity is low, analyze the reasons for the difference and judge the rationality of the adjustment in combination with expert knowledge. Through this verification process, filter out the parameter combinations that are theoretically deduced but may have poor actual effects, enhance the reliability of parameter optimization, and form a parameter scheme with double verification of experience and theory.

[0123] The verified parameter scheme is associated with the product material data, surface feature data and marking position coordinate data and stored in the process database to form a marking control scheme. Associative storage is the process of saving multiple interrelated data in a structured manner. The marking control scheme is a complete data set containing the information required for the entire marking process. In the specific operation, a data association structure is created, the parameter scheme is set as the core, and a reference relationship is established with the product material data, surface feature data and position coordinate data; then a unique identifier is assigned to the scheme to facilitate subsequent retrieval; then metadata such as the scope of application, generation time and verification status of the scheme are added; the complete structure is written into the process database, and the index is updated to ensure data consistency and accessibility. The marking control scheme formed in this way contains a complete information chain from workpiece characteristics to processing parameters, providing a directly available reference for subsequent marking of similar parts, reducing repeated experiments and improving production efficiency.

[0124] The control method for marking a product in the embodiment of the present application is described above. The control system for marking a product in the embodiment of the present application is described below. Figure 2 In the embodiment of the present application, an embodiment of a control system for product marking includes:

[0125] The scanning module is used to scan the surface of hardware parts through image acquisition, analyze the material characteristics of the acquired surface images, and obtain product material data and surface feature data;

[0126] The positioning module is used to locate the parts in three dimensions according to the product material data and surface feature data, and to correct the positioning results through micro-displacement calibration to obtain the marking position coordinate data;

[0127] A screening module is used to query and screen the process parameters in the material library according to the marking position coordinate data to obtain the marking power value and speed value corresponding to the current component material;

[0128] A control module, used to control the laser marking device through the marking power value and the speed value, to mark the parts and obtain marking process data;

[0129] A detection module, used to perform imaging detection on the marking result according to the marking process data, and perform edge analysis and depth analysis on the detection image to obtain marking quality data;

[0130] The comparison module is used to compare and adjust the marking process parameters through the marking quality data, and record the adjusted parameters into the process database to obtain the marking control plan.

[0131] Through the collaborative cooperation of the above-mentioned various components, the surface of the hardware parts is scanned through image acquisition, and the material characteristics of the obtained surface image are analyzed to obtain product material data and surface feature data, realizing the accurate identification of the material and surface characteristics of the parts, eliminating the material judgment errors caused by relying on manual experience in the traditional method, and providing accurate material basic information for the subsequent marking process; according to the product material data and surface feature data, the three-dimensional coordinate positioning of the parts is carried out, and the positioning result is corrected through micro-displacement calibration to obtain the marking position coordinate data, solving the problem that the traditional two-dimensional positioning method cannot cope with parts with complex shapes, significantly improving the marking position accuracy, especially the positioning accuracy of non-planar areas such as curved surfaces and inclined surfaces is fundamentally improved; according to the marking position coordinate data, the process parameters in the material library are queried and screened to obtain the marking power value and speed value corresponding to the current part material, and the mapping relationship between the material characteristics and the optimal process parameters is established through intelligent algorithms, breaking through the limitations of the traditional fixed parameter setting, realizing the personalized customization of parameters, and ensuring that parts with different materials and shapes can obtain the most suitable marking parameters; the laser marking equipment is controlled by the marking power value and speed value to carry out the marking process on the parts to obtain the marking process data, and at the same time, a real-time monitoring and feedback adjustment mechanism is introduced to change the marking process from open-loop control to closed-loop control, significantly improving the stability and adaptability of the marking process; according to the marking process data, the marking result is imaged and detected, and the edge analysis and depth analysis are carried out on the detected image to obtain the marking quality data, and a multi-dimensional quality evaluation system is introduced to comprehensively evaluate the marking effect from two levels of apparent quality and material microscopic changes, providing an objective and quantitative quality judgment basis; through the marking quality data, the marking process parameters are compared and adjusted, and the adjusted parameters are recorded in the process database to obtain the marking control scheme, forming a complete knowledge accumulation and optimization iteration mechanism, and continuously improving the marking quality with the accumulation of data. In addition, this solution applies a variety of artificial intelligence algorithms throughout the process, such as the deep learning classification algorithm in material recognition, the machine learning regression algorithm in parameter optimization, and the computer vision algorithm in quality evaluation. The contributions of these algorithm features to the solution are as follows: the accurate identification of complex material characteristics is realized through the deep learning algorithm, breaking through the limitation that it is difficult to distinguish similar materials in the traditional method; the non-linear mapping relationship between the material characteristics and the marking parameters is established through the machine learning regression algorithm, realizing the accurate prediction and dynamic adjustment of the parameters; the automatic evaluation of the marking quality is realized through the computer vision algorithm, eliminating the subjectivity and inconsistency of manual detection.

[0132] Referring to Figure 3 , an embodiment of the present invention further provides a computer device, which may be a server, and its internal structure may be as Figure 3As shown in the figure. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.

[0133] Those skilled in the art can understand that Figure 3 the structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.

[0134] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above method is implemented. It can be understood that the computer-readable storage medium in this embodiment can be a volatile readable storage medium or a non-volatile readable storage medium.

[0135] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.

[0136] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0137] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0138] The above is the case. The above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application 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 for some of the technical features. However, 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 various embodiments of the present application.

Claims

1. A control method for product marking, characterized in that, The control method for product marking includes: Scanning the surface of the hardware parts through image acquisition, analyzing the material characteristics of the obtained surface image, and obtaining product material data and surface feature data; According to the product material data and surface feature data, performing three-dimensional coordinate positioning on the parts, and correcting the positioning result through micro-displacement calibration to obtain marking position coordinate data; According to the marking position coordinate data, querying and screening the process parameters in the material library to obtain the marking power value and speed value corresponding to the current part material; Controlling the laser marking equipment through the marking power value and speed value to perform marking processing on the parts and obtaining marking process data; According to the marking process data, performing imaging detection on the marking result, and performing edge analysis and depth analysis on the detected image to obtain marking quality data; Through the marking quality data, comparing and adjusting the marking process parameters, and recording the adjusted parameters into the process database to obtain a marking control scheme.

2. The control method for product marking according to claim 1, wherein The scanning of the surface of the hardware parts through image acquisition, analyzing the material characteristics of the obtained surface image, and obtaining product material data and surface feature data includes: Performing image acquisition on the parts to be marked under multi-angle illumination through a high-resolution imaging system to obtain the original image data of the parts surface; According to the original image data, performing polarization filtering on the reflective area of the parts surface and performing contrast enhancement on the dark area to obtain an optimized surface image; Performing pixel analysis on the optimized surface image through feature extraction to identify the edge contour and surface texture of the parts and obtaining the morphological feature data of the parts; According to the morphological feature data, classifying the material of the parts surface and performing matching comparison with the data in the material parameter library to obtain the material type data of the parts; Performing wavelength analysis on the surface reflected light of the parts through spectral analysis to distinguish the spectral characteristics of three materials, namely metal, plastic, and ceramic, and obtaining material characteristic parameters; Fusing the morphological feature data, material type data, and material characteristic parameters to construct product material data and surface feature data including the surface flatness, material type, and surface texture of the parts.

3. The control method for product marking according to claim 2, characterized in that, The performing three-dimensional coordinate positioning on the parts according to the product material data and surface feature data, and correcting the positioning result through micro-displacement calibration to obtain marking position coordinate data includes: Scanning the surface of the parts through a laser triangulation system to generate preliminary point cloud data of the parts; According to the preliminary point cloud data, extracting feature points from the edge and corner areas of the parts to obtain a set of feature points of the parts; Performing image processing on the set of feature points through a machine vision system, and combining the texture information in the surface feature data to construct a three-dimensional model of the parts; According to the three-dimensional model, performing coordinate positioning on the parts surface, and selecting marking candidate points according to the marking required area to obtain a preliminary coordinate positioning result; The surface of the component is scanned in a micro - contact manner by a micro - displacement actuator to obtain feedback data on surface hardness and micro - topography; The feedback data is compared and calibrated with the preliminary coordinate positioning result to eliminate coordinate deviations caused by the shape of the component, and marking position coordinate data is generated.

4. The control method for product marking according to claim 3, characterized in that, According to the marking position coordinate data, process parameters in the material library are queried and filtered to obtain the marking power value and speed value corresponding to the current component material, including: Extract the material type identifier of the component from the product material data, index and query the material library to obtain the basic process parameter table; According to the surface flatness information in the surface feature data, calculate the correction factor for the basic process parameter table to obtain the surface adaptation parameter; Analyze the thermal field distribution of the component through heat conduction calculation, and combine with the material property parameters to predict the material response under different powers, and generate a power response curve; According to the geometric features in the marking position coordinate data, analyze the curvature and thickness of the marking area, and calculate the geometric shape factor; Combine the geometric shape factor with the power response curve to calculate the power range that meets the marking quality requirements, and determine the marking power value corresponding to the current component material; According to the marking power value and the material thermal conductivity, calculate the optimal heat input rate, convert it into the speed value of the marking head movement, and obtain the marking power value and speed value corresponding to the current component material.

5. The control method for product marking according to claim 1, characterized in that, Control the laser marking equipment through the marking power value and speed value to perform marking processing on the component, and obtain marking process data, including: Move the component to the marking work area through a five - axis linkage precision positioning platform, and perform precise positioning according to the marking position coordinate data to obtain the actual position information of the component in the work area; Perform a micro - pre - test in the non - critical area of the component, use 10% of the marking power value for dot matrix testing, and obtain the initial material response data; According to the initial material response data, fine - tune the marking power value and speed value to establish the real - time marking parameters of the component; Through the laser control system, start the laser marking process according to the real - time marking parameters, and at the same time monitor the spectrum of the marking area and record the spectral change data of the material surface; Collect the acoustic wave frequency generated during the marking process through an acoustic sensor, perform signal analysis on the acoustic wave frequency, and extract the acoustic features during the marking process; Integrate the spectral change data and acoustic features with the execution record of the marking power value and speed value to form marking process data.

6. The control method for product marking according to claim 1, wherein According to the marking process data, perform imaging detection on the marking result, and perform edge analysis and depth analysis on the detection image to obtain marking quality data, including: Perform multi - focal plane scanning on the marked component through a high - resolution focus stacking imaging device to obtain an image sequence of the marking area with different depths of field; Fuse and process the image sequence through image synthesis technology to form a high - depth - of - field marking result image; Based on the high-depth-of-field marking result image, the marking contour is pixel-level depicted through an edge extraction algorithm, and the sharpness coefficient and contour consistency value of the marking edge are calculated; The marking area is depth scanned by a three-dimensional optical interferometer to obtain a marking depth distribution map, and the average marking depth and coefficient of variation are calculated; Based on the spectral change data in the marking process data, the molecular structure change in the marking area is detected by combining Raman spectroscopy analysis to obtain a material state change index; The sharpness coefficient, contour consistency value, average marking depth, coefficient of variation, and material state change index are comprehensively analyzed to form marking quality data.

7. The control method for product marking according to claim 6, characterized in that, Based on the marking quality data, the marking process parameters are compared and adjusted, and the adjusted parameters are recorded in the process database to obtain a marking control scheme, including: The sharpness coefficient, contour consistency value, average marking depth, coefficient of variation, and material state change index in the marking quality data are compared with preset quality thresholds to generate a quality deviation matrix; Based on the quality deviation matrix, a correlation analysis is performed on the marking power value and speed value to establish a parameter-quality correlation model; The adjustment direction and amplitude of the marking process parameters are calculated through the parameter-quality correlation model to form a parameter correction value; The marking power value and speed value are combined with the parameter correction value to generate an optimized process parameter combination; An association analysis is performed between the optimized process parameter combination and historical successful case data to verify the rationality of the parameter adjustment and form a verified parameter scheme; The verified parameter scheme is associated and stored in the process database with the product material data, surface feature data, and marking position coordinate data to form a marking control scheme.

8. A control system for product marking, which is used to implement the control method for product marking described in any one of claims 1-7, characterized in that, The control system for product marking includes: A scanning module for scanning the surface of the hardware component through image acquisition, analyzing the material characteristics of the obtained surface image to obtain product material data and surface feature data; A positioning module for performing three-dimensional coordinate positioning on the component according to the product material data and surface feature data, and correcting the positioning result through micro-displacement calibration to obtain marking position coordinate data; A screening module for querying and screening the process parameters in the material library according to the marking position coordinate data to obtain the marking power value and speed value corresponding to the current component material; A control module for controlling the laser marking device through the marking power value and speed value to perform marking processing on the component to obtain marking process data; A detection module for performing imaging detection on the marking result according to the marking process data, and performing edge analysis and depth analysis on the detection image to obtain marking quality data; A comparison module for comparing and adjusting the marking process parameters through the marking quality data, and recording the adjusted parameters in the process database to obtain a marking control scheme.

9. A computer device, characterized in that, It includes a memory and a processor. The memory stores a computer program that can run on the processor. It is characterized in that when the processor executes the computer program, it implements the control method for product marking described in any one of claims 1 to 7.

10. A computer-readable storage medium, on which a computer program is stored. When the computer program is run by a processor, the processor is caused to execute the control method for product marking described in any one of claims 1 to 7.

Images