An optimized polishing method and system for nonferrous metal alloy die castings

Abstract

The present invention discloses an optimized polishing method and system for non-ferrous metal alloy die-castings, which relates to the technical field of optimized polishing of alloy die-castings, and includes the following steps: The laser polishing equipment first performs surface treatment on the zinc-aluminum alloy die-casting at an initial scanning speed, while the vision detection system real-time obtains the defect data on the surface of the casting, dynamically tracks the surface changes of the casting, and provides accurate input for intelligent evaluation. By combining the laser polishing equipment and the vision detection system, the present invention can real-time identify the surface defects of the casting and dynamically adjust the laser scanning speed and power, solving the problem of uneven surface caused by a fixed scanning speed. The system intelligently evaluates the severity of the defects, ensures that the laser energy acts uniformly on the surface, avoids overheating or insufficient treatment, improves the surface finish and appearance quality of the casting, reduces surface damage and material waste, thereby improving production efficiency and product performance.

Description

Technical Field

[0001] The invention relates to the technical field of optimized polishing of alloy die castings, and in particular to an optimized polishing method and system for nonferrous metal alloy die castings. Background Art

[0002] The optimized polishing of non-ferrous metal alloy die castings refers to the process of finely processing the surface of metal alloy parts formed during the die casting process to improve their surface quality, finish and appearance performance. Since there are often casting defects, burrs, scales and other problems on the surface of die castings, the optimized polishing not only needs to remove these surface defects, but also needs to consider improving the hardness, wear resistance and corrosion resistance of the metal surface. By selecting appropriate polishing processes, materials and equipment, combined with mechanical polishing, chemical polishing and other technical means, the surface quality of alloy parts can be effectively improved, making them smoother and more uniform, and improving the gloss of their appearance. This process has an important impact on the application performance, durability and aesthetics of non-ferrous metal alloy parts, especially in high-end consumer products, electronic products, automobiles and other industries, optimized polishing is the key link to improve product quality and market competitiveness.

[0003] The existing technology has the following deficiencies: When the existing technology uses laser polishing to polish zinc-aluminum alloy die castings, a fixed scanning speed is usually used for polishing operations. The fixed scanning speed is suitable for the case where the surface is relatively uniform, but when the surface defects of the die casting (such as protrusions, depressions, cracks or pores, etc.) are more complex, the effect of the laser in different areas may be very different. For example, in the concave area of ​​the surface, the energy of the laser may not fully act on the bottom, while in the convex area, the laser may overheat, resulting in local melting or material loss. This will cause uneven surface treatment, and ultimately affect the appearance and performance of the die casting.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0005] The purpose of the present invention is to provide an optimized polishing method and system for nonferrous metal alloy die castings, which can identify surface defects of castings in real time and dynamically adjust the laser scanning speed and power by combining laser polishing equipment and visual inspection systems, thereby solving the problem of surface unevenness caused by fixed scanning speed. The system intelligently evaluates the severity of defects, ensures that the laser energy acts evenly on the surface, avoids overheating or insufficient treatment, improves the surface finish and appearance quality of castings, reduces surface damage and material waste, thereby improving production efficiency and product performance, and solving the problems in the above-mentioned background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solution: an optimized polishing method for nonferrous metal alloy die castings, comprising the following steps:

[0007] The laser polishing equipment first performs surface treatment on the zinc-aluminum alloy die casting at the initial scanning speed, while the visual inspection system acquires defect data on the casting surface in real time, dynamically tracks changes in the casting surface, and provides accurate input for intelligent evaluation;

[0008] Establish a data set based on the defect data obtained by the visual inspection system. By establishing the data set, the surface defect data can be systematically stored and managed;

[0009] After preprocessing the casting surface defect data in the data set, the key features reflecting the current casting surface with serious defects are extracted from the preprocessed casting surface defect data, and the extracted key features are analyzed and processed under the detection window to further quantify the current casting surface defect degree;

[0010] The key features after analysis and processing are input into the pre-trained deep learning model for intelligent evaluation, in-depth analysis of defects on the casting surface, and determination of the nature of the defects;

[0011] When the evaluation results show that there are serious defects on the casting surface, the laser scanning speed is dynamically reduced according to the evaluation results to ensure that the laser stays in the defective area for a longer time and fully removes surface defects. At the same time, the laser power is increased to provide more energy to process the protruding parts and remove defects.

[0012] Preferably, the laser polishing equipment first performs surface treatment on the zinc-aluminum alloy die casting at an initial scanning speed in the following specific steps:

[0013] The laser polishing equipment first starts polishing the surface of the zinc-aluminum alloy die casting at a preset initial scanning speed, with the aim of evenly removing the surface roughness and defects, and preparing the foundation for subsequent fine adjustments;

[0014] At the same time, the visual inspection system monitors the casting surface in real time, capturing and recording surface defect data;

[0015] Visual inspection systems accurately identify defects on the surface and convert this data into digital information for subsequent processing and analysis.

[0016] Preferably, key features reflecting that the current casting surface is in serious defects are extracted from the pre-processed casting surface defect data, including the smoothness and regularity of the casting surface contour and the extension degree and direction of the casting surface cracks. The smoothness and regularity of the casting surface contour and the extension degree and direction of the casting surface cracks are analyzed to generate defect contour irregularity reference values ​​and surface crack extension reference values, respectively. The severity of the casting surface defects is comprehensively quantified through the defect contour irregularity reference values ​​and the surface crack extension reference values, and an accurate basis is provided for subsequent dynamic adjustment and polishing processing.

[0017] Preferably, the specific steps of analyzing the smoothness and regularity of the casting surface profile under the detection window to generate the reference value of the defect profile irregularity are as follows:

[0018] First, the height map of the casting surface is obtained through the visual inspection system, and the contour features of the casting surface are extracted from the data using the local contour analysis algorithm. The morphological changes of each area of ​​the surface are accurately captured based on the local morphological transformation algorithm. The local contour morphological feature extraction formula is as follows: , where is in position The local surface height change reflects the irregularity of the contour of the current location relative to its neighboring area. Is the casting surface in position The height value on It is the horizontal offset, which controls the horizontal distance between adjacent pixels during the calculation process. is the vertical offset, which controls the vertical scanning distance of the analysis window. is in position The maximum height value in the surrounding area;

[0019] Next, the local irregularity of each point on the casting surface is calculated to quantify the degree of defects in the area. The local curvature is used to characterize the surface change of each point. According to the curvature calculation formula, the local irregularity index of each point is obtained. The local irregularity index quantifies the degree of defects in each local area. The calculation expression is as follows: , where is in position The local irregularity index, i.e. the local curvature, Indicates along The change in curvature in the horizontal direction, that is, the degree of curvature of the surface in the horizontal direction, Indicates along The change in curvature in the direction, that is, the degree to which the surface is curved in the vertical direction;

[0020] Finally, the local irregularity index is integrated to obtain the reference value of the defect contour irregularity. The calculation expression is as follows: , where is the reference value of defect contour irregularity, is the weighting factor.

[0021] Preferably, the specific steps of analyzing the extension degree and direction of the casting surface cracks in the detection window to generate the surface crack extension reference value are as follows:

[0022] First, the crack boundary is extracted through image processing algorithm, and the crack extension degree is calculated. The calculation expression is as follows: , where is the length of the crack in the transverse position, is the crack depth in the lateral position, is the maximum length of the crack, is the maximum depth of the crack, is the weight parameter that controls the contribution of crack length to the extent of extension, is the weight parameter that controls the contribution of crack depth to the extent of crack extension, is the extent of crack extension;

[0023] The directional information of crack propagation is extracted through geometric analysis, and the angular deviation between the main propagation direction of the crack and the expected stress distribution is used to quantify the degree of abnormality in the propagation direction. The calculation expression of the crack propagation direction deviation is as follows: , where is the crack growth direction deviation, is the crack propagation direction in the The angle of the sampling point, It is in The expected stress direction at each sampling point is is the total number of sampling points, is the weighting coefficient that controls the influence of the deviation in the expansion direction;

[0024] Finally, the comprehensive crack extension degree and crack propagation direction deviation , generate the surface crack extension reference value, the generation formula is as follows: , where is the reference value of surface crack extension, and The crack extension degree and crack propagation direction deviation The weight coefficient of .

[0025] Preferably, the analyzed defect contour irregularity reference value and surface crack extension reference value are input into a pre-learned deep learning model, a surface defect coefficient is generated by the deep learning model, and the defects on the casting surface are deeply analyzed by the surface defect coefficient to determine the nature of the defects.

[0026] Preferably, the surface defect coefficient generated when the defects on the surface of the casting are deeply analyzed by the pre-learned deep learning model is compared and analyzed with the pre-set surface defect coefficient reference threshold to determine whether there are serious defects on the current casting surface. The specific steps are as follows;

[0027] If the surface defect coefficient is greater than a preset surface defect coefficient reference threshold, the current casting is classified as having severe surface defects; if the surface defect coefficient is less than or equal to the preset surface defect coefficient reference threshold, the current casting is classified as having no severe surface defects.

[0028] Preferably, when the evaluation results show that there are serious defects on the casting surface, the specific steps of dynamically reducing the laser scanning speed according to the evaluation results and increasing the laser power are as follows:

[0029] When the evaluation results show that there are serious defects on the casting surface, the scanning speed is dynamically reduced to extend the laser's residence time in the defective area to effectively remove the defects. The adjustment formula is as follows: , where is the adjusted scan speed, is the initial scan speed, and are adjustment coefficients, Used to adjust the impact of defect severity on scanning speed. Used to control the degree of nonlinearity between defect severity and scan speed, is the surface defect coefficient, is the reference threshold of surface defect coefficient;

[0030] At the same time, in order to ensure that the laser effectively removes surface defects and processes raised parts, the laser power is increased, and the calculation expression is as follows: , where is the adjusted laser power, is the initial laser power, is the power adjustment coefficient, which is used to adjust the influence of surface defects on laser power. is the index adjustment coefficient, which is used to adjust the relationship curve between the defect coefficient and the laser power adjustment;

[0031] Dynamically reduce the scanning speed and increase laser power At the same time, the relationship between scanning speed and power is comprehensively considered to ensure that the treatment of the defect area achieves the best effect. The adjustment formula is as follows: , where It's in time At this moment, the total energy of the laser applied to the defect area is is the area of ​​the defect region, It is the area element when integrating within the defect area.

[0032] An optimized polishing system for nonferrous metal alloy die castings, comprising an initial laser scanning and real-time defect detection module, a defect data storage and management module, a defect data preprocessing and feature extraction module, a deep learning intelligent evaluation module and a dynamic laser control module:

[0033] The initial laser scanning and real-time defect detection module and laser polishing equipment first perform surface treatment on the zinc-aluminum alloy die casting at the initial scanning speed. At the same time, the visual inspection system obtains the defect data of the casting surface in real time, dynamically tracks the changes of the casting surface, and provides accurate input for intelligent evaluation;

[0034] Defect data storage and management module, which establishes a data set based on the defect data obtained by the visual inspection system, and systematically stores and manages surface defect data by establishing a data set;

[0035] The defect data preprocessing and feature extraction module preprocesses the casting surface defect data in the data set, extracts the key features reflecting the current casting surface is in serious defects from the preprocessed casting surface defect data, analyzes and processes the extracted key features under the detection window, and further quantifies the current casting surface defect degree;

[0036] The deep learning intelligent evaluation module inputs the key features after analysis and processing into the pre-trained deep learning model for intelligent evaluation, conducts in-depth analysis of defects on the casting surface, and determines the nature of the defects;

[0037] Dynamic laser control module: When the evaluation results show that there are serious defects on the casting surface, the laser scanning speed is dynamically reduced according to the evaluation results to ensure that the laser stays in the defective area for a longer time and fully removes surface defects. At the same time, the laser power is increased to provide more energy to process the protruding parts and remove defects.

[0038] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0039] The present invention can accurately identify defects on the surface of castings and dynamically adjust them by combining the real-time feedback of laser polishing equipment and visual inspection systems, effectively solving the problem of uneven surface treatment caused by fixed scanning speed in traditional laser polishing. By intelligently evaluating the severity of defects, the system can dynamically adjust the laser scanning speed and power according to the specific defects of the castings, ensuring that the laser energy fully acts on the surface, evenly removing defects, avoiding overheating or insufficient treatment, thereby greatly improving the surface finish and appearance quality of the castings, while reducing surface damage and material waste, and improving production efficiency and product performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0041] Figure 1 The present invention is a method flow chart of an optimized polishing method for nonferrous metal alloy die castings.

[0042] Figure 2 The module schematic diagram of the optimized polishing system for nonferrous metal alloy die castings of the present invention. DETAILED DESCRIPTION

[0043] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these example embodiments are provided so that the description of the present disclosure will be more comprehensive and complete, and the concept of the example embodiments will be fully conveyed to those skilled in the art.

[0044] The present invention provides Figure 1 The optimized polishing method of a nonferrous metal alloy die casting shown comprises the following steps:

[0045] The laser polishing equipment first performs surface treatment on the zinc-aluminum alloy die casting at the initial scanning speed, while the visual inspection system acquires defect data on the casting surface in real time, dynamically tracks changes in the casting surface, and provides accurate input for intelligent evaluation;

[0046] The purpose of this stage is to lay the foundation for the entire polishing process, preliminarily remove the rough defects on the surface, and ensure the overall flatness of the surface. In this process, the fixed scanning speed helps to control the consistency of the preliminary polishing and provide a standard benchmark for subsequent dynamic adjustments. This operation ensures the consistency of the polishing process and creates conditions for subsequent defect detection and fine adjustment.

[0047] The visual inspection system acquires defect data on the surface of castings in real time through high-resolution cameras or laser scanners. The core purpose is to monitor the defect types and distribution on the surface of castings in real time, accurately capture every tiny defect, and ensure the real-time and accuracy of the data. The visual inspection system can effectively identify and locate surface defects such as dents, protrusions, cracks or pores, and generate high-quality surface defect data. This step provides comprehensive raw data support for subsequent defect analysis, feature extraction and decision-making.

[0048] At this stage, the laser polishing equipment first starts polishing the surface of the zinc-aluminum alloy die casting at a preset initial scanning speed. The purpose of this preliminary polishing is to evenly remove the surface roughness and large defects, and lay the foundation for subsequent fine adjustments. At the same time, the visual inspection system (such as a high-resolution camera, laser scanner or 3D scanner) monitors the surface of the casting in real time, captures and records surface defect data. The visual inspection system accurately identifies defects on the surface, such as cracks, depressions, protrusions and pores, and converts this data into digital information for subsequent processing and analysis.

[0049] Establish a data set based on the defect data obtained by the visual inspection system. By establishing the data set, the surface defect data can be systematically stored and managed;

[0050] This data set contains all relevant information about casting surface defects, such as defect type, location, size, depth, etc. By establishing a data set, the system can systematically store and manage surface defect data, providing a standardized basis for subsequent data preprocessing, feature extraction and analysis. The establishment of a data set is crucial to improving the efficiency and accuracy of subsequent algorithms.

[0051] After preprocessing the casting surface defect data in the data set, the key features reflecting the current casting surface with serious defects are extracted from the preprocessed casting surface defect data, and the extracted key features are analyzed and processed under the detection window to further quantify the current casting surface defect degree;

[0052] The purpose of preprocessing a data set is to improve data quality and reduce noise and redundant information. The preprocessing process includes steps such as data cleaning, denoising, and smoothing to ensure the accuracy and validity of the data. The preprocessed data is more standardized and consistent, making subsequent analysis and feature extraction more accurate and efficient. For example, after removing duplicate or unnecessary data, the system can focus more on the key parts of the defect, thereby providing a clearer and more accurate data basis for the extraction of key features.

[0053] The key features reflecting that the current casting surface is in serious defects are extracted from the pre-processed casting surface defect data, including the smoothness and regularity of the casting surface contour and the extension degree and direction of the casting surface cracks. The smoothness and regularity of the casting surface contour and the extension degree and direction of the casting surface cracks are analyzed to generate defect contour irregularity reference values ​​and surface crack extension reference values ​​respectively. The severity of casting surface defects is comprehensively quantified through the defect contour irregularity reference values ​​and surface crack extension reference values, and an accurate basis is provided for subsequent dynamic adjustment and polishing processing.

[0054] Obvious irregularities in the surface profile of a casting usually indicate that there are serious defects on the current casting surface. The irregularity of the surface profile of a casting refers to the unevenness, fluctuation or deformation of the surface shape, which is usually caused by defects such as pores, sand holes, cracks, shrinkage holes, cold shuts, etc. that appear during the casting process. When these irregularities exceed a certain degree, it indicates that the surface quality of the casting has been seriously damaged, which may affect its subsequent mechanical properties and service life. For example, surface irregularities may lead to uneven contact surfaces, affecting the sealing, strength and wear resistance of the casting; in addition, irregular surface structures may also lead to stress concentration, increasing the risk of crack propagation or rupture of the casting during use. Therefore, obvious irregularities are not only an intuitive manifestation of surface defects, but also a potential signal that may lead to performance degradation. When this happens on the surface of a casting, it indicates that the defects have penetrated into the structural level of the material, seriously affecting the appearance, structural integrity and reliability of the casting. At this time, these defects must be effectively treated by sophisticated surface repair techniques (such as laser polishing) to avoid unnecessary risks during use.

[0055] The specific steps for analyzing the smoothness and regularity of the casting surface profile under the detection window to generate the reference value of the defect profile irregularity are as follows:

[0056] First, the height map of the casting surface is obtained through the visual inspection system. The image data is obtained through laser scanning, 3D scanning or image acquisition equipment, showing the height information of each point on the casting surface. The contour features of the casting surface are extracted from the data using the local contour analysis algorithm. Specifically, the surface point cloud is sliced, partitioned and edge detected to determine the irregular parts such as small undulations, cracks, and protrusions on the casting surface. The morphological changes of each area of ​​the surface are accurately captured based on the local morphological transformation algorithm. The local contour morphological feature extraction formula is as follows: , where is in position The local surface height change reflects the irregularity of the contour of the current location relative to its neighboring area. Is the casting surface in position The height value on is the horizontal direction ( The offset in the axis direction controls the horizontal distance between adjacent pixels during the calculation process. In the vertical direction ( The offset in the vertical direction controls the scanning distance of the analysis window. is in position The maximum height value in the surrounding area, within the specified neighborhood window Within the range, calculate the height values ​​of all neighboring points and take the maximum value;

[0057] This step is to detect the local area of ​​the casting surface (in pixels) Centered on ) performs morphological processing to extract the local maximum height change to determine the surface convexity and concavity features around the point and generate height data based on local changes.

[0058] Next, the degree of defects in the area is quantified by calculating the local irregularity of each point on the casting surface. The local irregularity can be expressed as a highly variable nonlinear measure using a surface morphology function. The local curvature is used to characterize the surface changes at each point. Areas with higher curvature usually indicate that there are large fluctuations or defects on the surface. According to the curvature calculation formula, the local irregularity index of each point is obtained. The local irregularity index quantifies the degree of defects in each local area. The calculation expression is as follows: , where , where is in position The local irregularity index, i.e. the local curvature, Indicates along The change in curvature in the horizontal direction, that is, the degree of curvature of the surface in the horizontal direction, Indicates along The change in curvature of the direction, that is, the degree to which the surface is curved in the vertical direction;

[0059] The above steps calculate the surface height map exist and The second-order derivative in the direction is used to obtain the local irregularity index, that is, the local curvature. The larger the curvature value, the more drastic the surface change in the area, and there may be more serious defects (such as depressions, protrusions, etc.). The local curvature can effectively quantify the complexity and severity of surface defects.

[0060] Finally, the local irregularity index is combined to obtain the reference value of the irregularity of the defect profile. This reference value is calculated by weighting the irregularity index of each point on the surface, and ultimately reflects the severity of the overall surface defect. In this process, considering that some areas (such as bulges or cracks) have a greater influence, they are given a higher weight. The calculation expression is as follows: , where is the reference value of defect contour irregularity, is a weighting factor that adjusts the weight according to the severity of the surface defect or the importance of the area to the overall defect level.

[0061] Through weighted summation, the global irregularity reference value is finally obtained. The larger the value, the more serious the defects on the casting surface.

[0062] The smoothness and regularity of the casting surface profile are analyzed under the detection window to generate the defect profile irregularity reference value. The larger the irregularity reference value, the more serious the defects on the current casting surface. This reference value is obtained by analyzing the smoothness and regularity of the casting surface profile. Specifically, it reflects the degree of surface irregularity. If there are obvious convex and concave, undulating or uneven shapes on the casting surface, the smoothness and regularity of the surface profile will decrease significantly, resulting in an increase in the value of the defect profile irregularity reference value. Conversely, when the casting surface is relatively smooth and regular, the value of the irregularity reference value is lower, indicating that there are fewer surface defects and better overall quality.

[0063] The extension degree and direction of the cracks on the casting surface do not conform to the normal stress distribution, which does indicate that there are serious defects on the current casting surface. Under normal circumstances, the surface stress distribution of the casting is uniform, and the expansion of the cracks usually follows the mechanical properties of the material and the law of stress concentration areas. However, when the extension degree and direction of the surface cracks are abnormal, it often means that there are internal stress unevenness, material defects, or stress concentration that has not been completely eliminated during the casting process. These cracks usually extend in irregular directions and may be caused by pores, inclusions, cold shuts or other structural defects in the casting. These defects not only affect the appearance of the casting, but also significantly reduce the overall strength and durability of its structure. If the cracks extend in a direction that does not conform to the normal stress distribution, it may cause stress concentration in the area around the cracks, further aggravate the expansion of the cracks, and may even cause fractures during subsequent use. Therefore, the abnormal extension direction and degree of the cracks indicate that there are quality problems in the manufacturing process of the casting, which seriously affects the reliability and service life of the casting, and further repair or treatment is required. Therefore, the expansion behavior of the cracks can be used as an important basis for judging whether there are serious defects on the surface of the casting.

[0064] The specific steps for analyzing the extension degree and direction of the casting surface cracks in the detection window to generate the reference value of the surface crack extension are as follows:

[0065] First, the crack boundary is extracted through image processing algorithm, and the crack extension degree is calculated. The calculation expression is as follows: , where The crack is in the horizontal direction ( The length of the axis, The crack is in the horizontal direction ( axis) position, is the maximum length of the crack, is the maximum depth of the crack, is the weight parameter that controls the contribution of crack length to the extent of extension, is the weight parameter that controls the contribution of crack depth to the extent of crack extension. is the extent of crack extension;

[0066] This step calculates the integral of crack extension, with different weights for crack length and depth at different locations, thereby precisely measuring the extent of crack extension.

[0067] The directional information of crack propagation is extracted through geometric analysis, and the angular deviation between the main propagation direction of the crack and the expected stress distribution is used to quantify the abnormality of the propagation direction. The calculation expression of the crack propagation direction deviation is as follows: , where is the crack growth direction deviation, is the crack propagation direction in the The angle of the sampling point, It is in The expected stress direction at each sampling point is usually obtained through mechanical analysis. is the total number of sampling points, is the weighting coefficient that controls the influence of the deviation in the expansion direction;

[0068] By calculating the directional deviation, the abnormal degree of crack growth can be quantified and the directionality of the crack can be risk assessed. This step helps to reveal the potential danger area of ​​the crack, especially when the crack growth direction does not conform to the stress distribution, indicating that the casting may have serious defects.

[0069] Finally, the comprehensive crack extension degree and crack propagation direction deviation , generate the surface crack extension reference value, the generation formula is as follows: , where is the reference value of surface crack extension, and The crack extension degree and crack propagation direction deviation The weight coefficient of .

[0070] The larger the surface crack extension reference value generated after analyzing the extension degree and direction of the surface crack of the casting under the detection window, the more serious defects there are on the current casting surface. Conversely, it means that the surface defects are lighter or there are no serious problems. When the extension degree and direction of the crack on the surface of the casting are abnormal, the crack extension reference value will be higher, which means that the crack extends to a greater extent on the surface and its extension direction deviates from the normal stress distribution. This is usually caused by uneven materials, stress concentration or other defects (such as pores, inclusions, etc.) during the casting process. Abnormal crack extension will lead to stress concentration, which may further cause material damage or fracture, seriously affecting the mechanical properties and structural stability of the casting. Therefore, the larger the surface crack extension reference value, the more serious the surface defects are, and corresponding treatment measures need to be taken. On the contrary, if the crack extension reference value is smaller, it means that the surface defects of the casting are fewer, the crack extension is normal, and the structural integrity of the casting is better.

[0071] The key features after analysis and processing are input into the pre-trained deep learning model for intelligent evaluation, in-depth analysis of defects on the casting surface, and determination of the nature of the defects;

[0072] The analyzed defect contour irregularity reference value and surface crack extension reference value are input into the pre-learned deep learning model, and the surface defect coefficient is generated by the deep learning model. The surface defects of the casting are deeply analyzed through the surface defect coefficient to determine the nature of the defects.

[0073] A pre-learned deep learning model is a deep learning model trained with a large amount of labeled data. The model has the ability to automatically extract features from the input data and perform inference, classification, or regression. In the scenario of surface defect detection, deep learning models are usually trained using large-scale casting surface defect datasets. These datasets contain different types of defect samples, such as cracks, dents, pores, scratches, etc., and each sample is labeled with the corresponding defect type or defect severity. Through training, the model learns how to extract key information that helps to judge surface defects from the input feature data, including the shape, size, location, depth, etc. of the defects. After training, the deep learning model can identify similar defects in unknown casting samples and accurately classify or evaluate them.

[0074] The core advantage of deep learning models is that they can automatically learn complex nonlinear relationships from data without the need for manual feature extraction. Compared with traditional machine learning methods, deep learning models do not rely on rules designed by experts or preset features. They automatically extract multi-level features through a multi-layer neural network architecture (such as convolutional neural networks, CNNs), and optimize their parameters through training to achieve high-precision prediction capabilities. In the application of casting surface defect detection, the deep learning model can generate a comprehensive "surface defect coefficient" based on the input defect contour irregularity reference value and crack extension reference value. This coefficient not only reflects the current defect level of the casting surface, but also further evaluates the nature of the defect, such as whether it is a crack, dent or pore, or whether there is a certain pattern of distribution trend. With this deep learning model, the system can analyze and diagnose casting surface defects in real time and automatically, providing accurate evaluation results, thereby providing a scientific basis for subsequent polishing, repair or quality control decisions.

[0075] The deep learning model is not limited here and can realize the reference value of the irregularity of the defect contour and surface crack extension reference value Perform comprehensive analysis to generate surface defect coefficients In order to realize the technical solution of the present invention, the present invention provides a specific implementation method;

[0076] Surface defect coefficient The generation formula is as follows: , where , are the reference values ​​of defect contour irregularity and surface crack extension reference value The preset scaling factor of , Both are greater than 0.

[0077] The preset scale factor refers to the parameter used in the deep learning model to weightedly fuse features from different sources or different types (such as reference values ​​of surface defects). The model needs to consider multiple input variables, which may include the reference value of defect contour irregularity. and surface crack extension reference value . Preset scale factor ( and ) is used to define how these input features are combined in proportion to produce the final defect coefficient.

[0078] Specifically, and Respectively represent the weight or importance of the crack extension reference value and the surface crack extension reference value in calculating the surface defect coefficient. The role of the preset proportional coefficient is to balance the impact of different features on the final output. and The value of is usually large, which means that the reference values ​​of crack extension and surface crack have a significant impact on the generation of the final defect coefficient. By adjusting these proportional coefficients, the model can more flexibly reflect the characteristics of different types of defects in practical applications, thereby improving the accuracy and reliability of defect assessment.

[0079] It can be seen from the surface defect coefficient that the larger the defect contour irregularity reference value generated by analyzing the smoothness and regularity of the casting surface contour under the detection window, the larger the surface crack extension reference value generated after analyzing the extension degree and direction of the casting surface crack under the detection window. This indicates that the larger the surface defect coefficient generated when the defects on the casting surface are deeply analyzed by the pre-learned deep learning model, the greater the surface defect exists on the current casting surface. Otherwise, it indicates that there are no serious defects on the current casting surface.

[0080] The surface defect coefficient generated by the deep analysis of the defects on the casting surface by the pre-learned deep learning model is compared and analyzed with the pre-set reference threshold of the surface defect coefficient to determine whether there are serious defects on the current casting surface. The specific steps are as follows;

[0081] If the surface defect coefficient is greater than a preset surface defect coefficient reference threshold, the current casting is classified as having severe surface defects; if the surface defect coefficient is less than or equal to the preset surface defect coefficient reference threshold, the current casting is classified as having no severe surface defects.

[0082] When the evaluation results show that there are serious defects on the casting surface, the laser scanning speed is dynamically reduced according to the evaluation results to ensure that the laser stays in the defect area for a longer time and fully removes the surface defects. At the same time, the laser power is increased to provide more energy to process the protruding parts and remove the defects.

[0083] When the evaluation results show that there are serious defects on the casting surface, the specific steps for dynamically reducing the laser scanning speed and increasing the laser power according to the evaluation results are as follows:

[0084] When the evaluation results show that there are serious defects on the casting surface, the scanning speed is dynamically reduced to extend the laser's residence time in the defective area to effectively remove the defects. The adjustment formula is as follows: , where is the adjusted scan speed, is the initial scan speed, and are adjustment coefficients, Used to adjust the impact of defect severity on scanning speed. Used to control the degree of nonlinearity between defect severity and scan speed, is the surface defect coefficient, is the reference threshold of surface defect coefficient;

[0085] With surface defect coefficient The increase in scanning speed The laser will decrease, thereby increasing the time the laser stays in the defect area. This adjustment helps ensure that deep defects and larger defects can receive sufficient thermal energy for repair.

[0086] At the same time, in order to ensure that the laser effectively removes surface defects and processes raised parts, the laser power is increased, and the calculation expression is as follows: , where is the adjusted laser power, is the initial laser power, is the power adjustment coefficient, which is used to adjust the influence of surface defects on laser power. is the index adjustment coefficient, which is used to adjust the relationship curve between the defect coefficient and the laser power adjustment;

[0087] With the surface defect coefficient As the laser power increases, it provides more energy to the severely defective areas. This effectively melts the metal in the raised areas and removes the defects, avoiding overheating or inadequate repair.

[0088] Dynamically reduce the scanning speed and increase laser power At the same time, the relationship between scanning speed and power is comprehensively considered to ensure that the treatment of the defect area achieves the best effect. The adjustment formula is as follows: , where It's in time At this moment, the total energy of the laser applied to the defect area is is the area of ​​the defect region, It is the area element when integrating within the defect area.

[0089] This step ensures that sufficient energy is applied to the defective area, especially in deep concave and high convex parts, by comprehensively considering the power and speed adjustment, so as to ensure that the laser can penetrate deeply and effectively remove the defects. The system monitors and calculates in real time, and adjusts the total energy to ensure that the laser energy is not wasted, while maximizing the removal effect and avoiding surface damage caused by excessive heating.

[0090] When the evaluation results show that there are serious defects on the surface of the casting, it is crucial to dynamically adjust the laser scanning speed and power according to the evaluation results. The core purpose of this step is to ensure that the laser can fully act on the surface defect area, thereby effectively improving the quality of the casting surface. Specifically, by reducing the scanning speed of the laser, the laser's residence time in the defect area can be extended, which is especially important for surface depressions, cracks and other irregular defects. In the defect area, the extended focusing time of the laser energy means that the defective part can be fully heated and melted, especially in the depression area, which can ensure that the laser can penetrate to a depth and achieve a more thorough removal effect. This measure avoids the omission of surface treatment, ensures that defects can be completely removed, and the surface is evenly polished.

[0091] At the same time, increasing the laser power is to increase the energy output of each laser pulse, thereby providing more energy for processing those more severe protrusions. For surface protrusions, increasing the laser power helps to melt excessive metal materials more effectively and prevent thermal damage or local material loss caused by excessive heating. Through these two adjustments, the laser polishing process can be optimized in real time according to the surface state of the casting, ensuring that when removing defects, not only uniformity and efficiency are maintained, but also other problems caused by excessive heating are reduced. Finally, dynamically adjusting the laser scanning speed and power can greatly improve the polishing quality of the casting surface, ensure that the surface is flat and smooth, and achieve a high standard of surface treatment effect.

[0092] The present invention can accurately identify defects on the surface of castings and dynamically adjust them by combining the real-time feedback of laser polishing equipment and visual inspection systems, effectively solving the problem of uneven surface treatment caused by fixed scanning speed in traditional laser polishing. By intelligently evaluating the severity of defects, the system can dynamically adjust the laser scanning speed and power according to the specific defects of the castings, ensuring that the laser energy fully acts on the surface, evenly removing defects, avoiding overheating or insufficient treatment, thereby greatly improving the surface finish and appearance quality of the castings, while reducing surface damage and material waste, and improving production efficiency and product performance.

[0093] The present invention provides Figure 2 An optimized polishing system for nonferrous metal alloy die castings shown in the figure includes an initial laser scanning and real-time defect detection module, a defect data storage and management module, a defect data preprocessing and feature extraction module, a deep learning intelligent evaluation module, and a dynamic laser control module:

[0094] The initial laser scanning and real-time defect detection module and laser polishing equipment first perform surface treatment on the zinc-aluminum alloy die casting at the initial scanning speed. At the same time, the visual inspection system obtains the defect data of the casting surface in real time, dynamically tracks the changes of the casting surface, and provides accurate input for intelligent evaluation;

[0095] Defect data storage and management module, which establishes a data set based on the defect data obtained by the vision detection system. By establishing the data set, the surface defect data is systematically stored and managed;

[0096] Defect data preprocessing and feature extraction module, which preprocesses the casting surface defect data in the data set, and then extracts the key features reflecting the serious defects on the current casting surface from the preprocessed casting surface defect data. The extracted key features are analyzed and processed under the detection window to further quantify the current casting surface defect degree;

[0097] Deep learning intelligent evaluation module, which inputs the key features after analysis and processing into a pre-trained deep learning model for intelligent evaluation, deeply analyzes the defects on the casting surface, and judges the nature of the defects;

[0098] Dynamic laser regulation module, when the evaluation result shows that there are serious defects on the casting surface, it dynamically reduces the laser scanning speed according to the evaluation result to ensure that the laser stays in the defect area for a longer time to fully remove the surface defects. At the same time, it increases the laser power to provide more energy to process the raised parts and remove the defects.

[0099] An optimized polishing method for non-ferrous metal alloy die-castings provided by an embodiment of the present invention is realized through the above-mentioned optimized polishing system for non-ferrous metal alloy die-castings. The specific methods and processes of the optimized polishing system for non-ferrous metal alloy die-castings are detailed in the embodiments of the above-mentioned optimized polishing method for non-ferrous metal alloy die-castings, and will not be elaborated here.

[0100] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula that is closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0101] Only some exemplary embodiments of the present invention have been described by way of illustration above. Undoubtedly, for those of ordinary skill in the art, without departing from the spirit and scope of the present invention, the described embodiments can be modified in various different ways. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

[0102] It should be noted that, in this article, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0103] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0104] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0105] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0106] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0107] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0108] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0109] The above description is only by way of illustration of certain exemplary embodiments of the present invention. It is undoubted that those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An optimized polishing method for nonferrous metal alloy die castings, characterized in that: The following steps are involved: The laser polishing equipment first performs surface treatment on the zinc-aluminum alloy die casting at the initial scanning speed, while the visual inspection system acquires defect data on the casting surface in real time, dynamically tracks changes in the casting surface, and provides accurate input for intelligent evaluation; Establish a data set based on the defect data obtained by the visual inspection system. By establishing the data set, the surface defect data can be systematically stored and managed; After preprocessing the casting surface defect data in the data set, the key features reflecting the current casting surface with serious defects are extracted from the preprocessed casting surface defect data, and the extracted key features are analyzed and processed under the detection window to further quantify the current casting surface defect degree; The key features after analysis and processing are input into the pre-trained deep learning model for intelligent evaluation, in-depth analysis of defects on the casting surface, and determination of the nature of the defects; When the evaluation results show that there are serious defects on the casting surface, the laser scanning speed is dynamically reduced according to the evaluation results to ensure that the laser stays in the defect area for a longer time and fully removes the surface defects. At the same time, the laser power is increased to provide more energy to process the protruding parts and remove the defects. The key features reflecting that the current casting surface is in serious defects are extracted from the pre-processed casting surface defect data, including the smoothness and regularity of the casting surface contour and the extension degree and direction of the casting surface cracks. The smoothness and regularity of the casting surface contour and the extension degree and direction of the casting surface cracks are analyzed to generate defect contour irregularity reference values ​​and surface crack extension reference values ​​respectively. The severity of casting surface defects is comprehensively quantified through the defect contour irregularity reference values ​​and surface crack extension reference values, and an accurate basis is provided for subsequent dynamic adjustment and polishing processing.

2. The optimized polishing method for nonferrous metal alloy die casting according to claim 1, characterized in that: The specific steps of the laser polishing equipment to perform surface treatment on zinc-aluminum alloy die castings at the initial scanning speed are as follows: The laser polishing equipment first starts polishing the surface of the zinc-aluminum alloy die casting at a preset initial scanning speed, with the aim of evenly removing the surface roughness and defects, and preparing the foundation for subsequent fine adjustments; At the same time, the visual inspection system monitors the casting surface in real time, capturing and recording surface defect data; Visual inspection systems accurately identify defects on the surface and convert this data into digital information for subsequent processing and analysis.

3. The optimized polishing method for nonferrous metal alloy die casting according to claim 1, characterized in that: The specific steps for analyzing the smoothness and regularity of the casting surface profile under the detection window to generate the reference value of the defect profile irregularity are as follows: First, the height map of the casting surface is obtained through the visual inspection system, and the contour features of the casting surface are extracted from the data using the local contour analysis algorithm. The morphological changes of each area of ​​the surface are accurately captured based on the local morphological transformation algorithm. The local contour morphological feature extraction formula is as follows: , where is in position The local surface height change reflects the irregularity of the contour of the current location relative to its neighboring area. Is the casting surface in position The height value on It is the horizontal offset, which controls the horizontal distance between adjacent pixels during the calculation process. is the vertical offset, which controls the vertical scanning distance of the analysis window. is in position The maximum height value in the surrounding area; Next, the local irregularity of each point on the casting surface is calculated to quantify the degree of defects in the area. The local curvature is used to characterize the surface change of each point. According to the curvature calculation formula, the local irregularity index of each point is obtained. The local irregularity index quantifies the degree of defects in each local area. The calculation expression is as follows: , where is in position The local irregularity index, i.e. the local curvature, Indicates along The change in curvature in the horizontal direction, that is, the degree of curvature of the surface in the horizontal direction, Indicates along The change in curvature of the direction, that is, the degree to which the surface is curved in the vertical direction; Finally, the local irregularity index is integrated to obtain the reference value of the defect contour irregularity. The calculation expression is as follows: , where is the reference value of defect contour irregularity, is the weighting factor.

4. The method for optimizing polishing of nonferrous metal alloy die castings according to claim 1, characterized in that: The specific steps for analyzing the extension degree and direction of the casting surface cracks in the detection window to generate the reference value of the surface crack extension are as follows: First, the crack boundary is extracted through image processing algorithm, and the crack extension degree is calculated. The calculation expression is as follows: , where is the length of the crack in the transverse position, is the crack depth in the lateral position, is the maximum length of the crack, is the maximum crack depth, is the weight parameter that controls the contribution of crack length to the extent of extension, is the weight parameter that controls the contribution of crack depth to the extent of crack extension. is the extent of crack extension; The directional information of crack propagation is extracted through geometric analysis, and the angular deviation between the main propagation direction of the crack and the expected stress distribution is used to quantify the abnormality of the propagation direction. The calculation expression of the crack propagation direction deviation is as follows: , where is the crack growth direction deviation, is the crack propagation direction in the The angle of the sampling point, It is in The expected stress direction at each sampling point is is the total number of sampling points, is the weighting coefficient that controls the influence of the deviation in the expansion direction; Finally, the comprehensive crack extension degree and crack propagation direction deviation , generate the surface crack extension reference value, the generation formula is as follows: , where is the reference value of surface crack extension, and The crack extension degree and crack propagation direction deviation The weight coefficient of .

5. The method for optimizing polishing of nonferrous metal alloy die castings according to claim 1, characterized in that: The analyzed defect contour irregularity reference value and surface crack extension reference value are input into the pre-learned deep learning model, and the surface defect coefficient is generated by the deep learning model. The surface defects of the casting are deeply analyzed through the surface defect coefficient to determine the nature of the defects.

6. The method for optimizing polishing of nonferrous metal alloy die castings according to claim 5, characterized in that: The surface defect coefficient generated by the deep analysis of the defects on the casting surface by the pre-learned deep learning model is compared and analyzed with the pre-set reference threshold of the surface defect coefficient to determine whether there are serious defects on the current casting surface. The specific steps are as follows; If the surface defect coefficient is greater than a preset surface defect coefficient reference threshold, the current casting is classified as having severe surface defects; if the surface defect coefficient is less than or equal to the preset surface defect coefficient reference threshold, the current casting is classified as having no severe surface defects.

7. The method for optimizing polishing of nonferrous metal alloy die castings according to claim 6, characterized in that: When the evaluation results show that there are serious defects on the casting surface, the specific steps for dynamically reducing the laser scanning speed and increasing the laser power according to the evaluation results are as follows: When the evaluation results show that there are serious defects on the casting surface, the scanning speed is dynamically reduced to extend the laser's residence time in the defective area to effectively remove the defects. The adjustment formula is as follows: , where is the adjusted scan speed, is the initial scan speed, and are adjustment coefficients, Used to adjust the impact of defect severity on scanning speed. Used to control the degree of nonlinearity between defect severity and scan speed, is the surface defect coefficient, is the reference threshold of surface defect coefficient; At the same time, in order to ensure that the laser effectively removes surface defects and processes raised parts, the laser power is increased, and the calculation expression is as follows: , where is the adjusted laser power, is the initial laser power, is the power adjustment coefficient, which is used to adjust the influence of surface defects on laser power. is the index adjustment coefficient, which is used to adjust the relationship curve between the defect coefficient and the laser power adjustment; Dynamically reduce the scanning speed and increase laser power At the same time, the relationship between scanning speed and power is comprehensively considered to ensure that the treatment of the defect area achieves the best effect. The adjustment formula is as follows: , where It's in time At this moment, the total energy of the laser applied to the defect area is is the area of ​​the defect region, It is the area element when integrating within the defect area.

8. An optimized polishing system for nonferrous metal alloy die castings, used to implement an optimized polishing method for nonferrous metal alloy die castings as described in any one of claims 1 to 7, characterized in that: It includes initial laser scanning and real-time defect detection module, defect data storage and management module, defect data preprocessing and feature extraction module, deep learning intelligent evaluation module and dynamic laser control module: The initial laser scanning and real-time defect detection module and laser polishing equipment first perform surface treatment on the zinc-aluminum alloy die casting at the initial scanning speed. At the same time, the visual inspection system obtains the defect data of the casting surface in real time, dynamically tracks the changes of the casting surface, and provides accurate input for intelligent evaluation; Defect data storage and management module, which establishes a data set based on the defect data obtained by the visual inspection system, and systematically stores and manages surface defect data by establishing a data set; The defect data preprocessing and feature extraction module preprocesses the casting surface defect data in the data set, extracts the key features reflecting the current casting surface is in serious defects from the preprocessed casting surface defect data, analyzes and processes the extracted key features under the detection window, and further quantifies the current casting surface defect degree; The deep learning intelligent evaluation module inputs the key features after analysis and processing into the pre-trained deep learning model for intelligent evaluation, conducts in-depth analysis of defects on the casting surface, and determines the nature of the defects; Dynamic laser control module: When the evaluation results show that there are serious defects on the casting surface, the laser scanning speed is dynamically reduced according to the evaluation results to ensure that the laser stays in the defective area for a longer time and fully removes surface defects. At the same time, the laser power is increased to provide more energy to process the protruding parts and remove defects.

Images