Metal plate optimization machining method and system based on automatic deviation correction

Abstract

The invention discloses a metal plate optimization machining method and system based on automatic deviation rectification, and relates to the technical field of machining optimizing.The metal plates to be machined are recorded as first plates, and surface images and machining tracks of the first plates are obtained to analyze whether the first plates can be subjected to deviation rectification machining through a current automatic deviation rectification system or not; if yes, an image in the metal plate machining process is obtained, and a deviation correction effect coefficient is calculated according to the image; according to the deviation rectifying effect coefficient, the qualified state of the deviation rectifying effect in the current machining process is evaluated, and machining optimization is conducted according to the different qualified states of the deviation rectifying effect; in this way, the metal plates which can be subjected to deviation correction machining through the current automatic deviation correction system can be screened out, whether the current automatic deviation correction system is still used for machining the metal plates or not can be judged according to the automatic deviation correction effect in the machining process, intervention is conducted in time, the influence on the production efficiency is reduced, and the production efficiency is improved. The product quality is ensured to be qualified; and the production cost waste is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of processing optimization, and particularly relates to an optimized processing method and system for metal sheets based on automatic deviation correction. Background Art

[0002] With the increasing requirements for precision and efficiency in modern manufacturing, automation technology has been widely applied in the field of metal sheet processing; the automatic deviation correction technology can effectively improve the processing precision, reduce manual intervention and production costs by real-time monitoring and adjusting the position deviation of metal sheets during the processing. This technology mainly uses sensors, vision systems and control algorithms to correct the offset of the sheet in real time and ensure that the deviation remains within an acceptable range during the processing.

[0003] However, in actual processing production lines, due to their own reasons, it may be difficult to automatically correct the deviation of some metal sheets during the processing, and for these metal sheets with higher difficulty, the current automatic deviation correction system of the processing production line may not be able to effectively correct the deviation, resulting in difficulty in ensuring the processing precision; in addition, during the processing, if the effect of automatic deviation correction is not good and the current automatic deviation correction system is still used to process the metal sheets without intervention, it will not only affect the production efficiency, but may also lead to unqualified product quality, and even require reprocessing or scrapping, increasing the production cost. Summary of the Invention

[0004] The purpose of the present invention is to solve the above-mentioned problems and provide an optimized processing method and system for metal sheets based on automatic deviation correction.

[0005] In the first aspect of the implementation of the present invention, an optimized processing method for metal sheets based on automatic deviation correction is first proposed. The method includes: 1. An optimized processing method for metal sheets based on automatic deviation correction, characterized by including the following steps: Denote each metal sheet to be processed as the first sheet, obtain the surface image and processing trajectory of the first sheet, calculate the geometric irregularity coefficient, surface defect coefficient and processing complexity coefficient of the first sheet, and evaluate the geometric shape regularity degree, surface defect degree and processing complexity degree of the first sheet; Analyze the first sheet according to the geometric irregularity coefficient, surface defect coefficient and processing complexity coefficient to determine whether the first sheet can be corrected and processed by the current automatic deviation correction system; Denote the first sheet that can be corrected and processed by the current automatic deviation correction system as the second sheet, and obtain the deviation correction effect coefficient by calculating the image during the processing of the second sheet; Evaluate the qualified state of the deviation correction effect during the current processing according to the deviation correction effect coefficient, and perform processing optimization according to different qualified states of the deviation correction effect.

[0006] Optionally, the calculation steps of the geometric irregularity coefficient are as follows: Obtain the surface image of the first sheet, preprocess the image, and divide the preprocessed image into several local regions; For each local region, calculate the curvature of each pixel coordinate in the local region, and calculate the mean value of the curvatures of all pixel coordinates in the region to obtain the geometric irregularity value of the corresponding local region; Multiply the geometric irregularity values of each local region by the corresponding preset weights to obtain the final geometric irregularity values of the corresponding local regions, calculate the mean value of the final geometric irregularity values of all local regions, and normalize the mean value. Take the result of the normalization process as the geometric irregularity coefficient of the first sheet.

[0007] Optionally, the calculation steps of the surface defect coefficient are as follows: Obtain the surface image of the first sheet, preprocess the image, divide the preprocessed image into several windows, and calculate the standard deviation of the pixel values in each window to measure the brightness fluctuation within the local region; A larger standard deviation usually means the presence of defects or irregularities; For each window, compare the standard deviation of the pixel values with a preset standard deviation threshold. If the standard deviation is not less than the preset standard deviation threshold, mark the corresponding window as a defective window; Calculate the total area of the defective windows as the total defect area, and divide the total defect area by the total area of the preprocessed image to obtain the surface defect coefficient of the first sheet.

[0008] Optionally, the calculation steps of the processing complexity coefficient are as follows: On the first sheet, generate a processing trajectory according to the set process requirements and discretize it into sampling points, and the position of each discrete point is a three-dimensional coordinate; Record the trajectory between two adjacent sampling points on the processing trajectory as a trajectory segment, and calculate the total length of the processing trajectory. The calculation formula is: , where is the total length of the processing trajectory, represents the length of the th trajectory segment; that is, the Euclidean distance between the th sampling point and the th sampling point; Calculate the included angle between adjacent trajectory segments. The calculation formula is: , where is the direction vector of the th trajectory segment, and the expression is , is the dot product of vectors, and are respectively the moduli of the and the th trajectory segments; Calculate the local curvature of each trajectory segment, and the calculation formula is: , where in the formula, is the local curvature of the th trajectory segment; Calculate the machining complexity coefficient, and the calculation formula is: , where in the formula, is the machining complexity coefficient.

[0009] Optionally, the steps for analyzing whether the first sheet can be rectified by the current automatic rectification system are as follows: Normalize the geometric irregularity coefficient, surface defect coefficient, and machining complexity coefficient so that they are all within [0, 1]. Represent the normalized geometric irregularity coefficient, surface defect coefficient, and machining complexity coefficient as data points for each first sheet, and the corresponding data points of all first sheets form a data set; cluster the data set to determine the automatic rectification difficulty coefficient for each first sheet; Compare the automatic rectification difficulty coefficient of each first sheet with the preset automatic rectification difficulty coefficient threshold. If the automatic rectification difficulty coefficient is less than the preset automatic rectification difficulty coefficient threshold, it means that the first sheet can be rectified by the current automatic rectification system; if the automatic rectification difficulty coefficient is not less than the preset automatic rectification difficulty coefficient threshold, it means that the first sheet cannot be rectified by the current automatic rectification system.

[0010] Optionally, the steps for clustering the data set to determine the automatic rectification difficulty coefficient for each first sheet are as follows: Cluster the data set by the K-means clustering method to obtain several final clustering clusters, and map each final clustering cluster to three-dimensional coordinates; Calculate the weight coefficient of each final clustering cluster, and its calculation formula is , where represents the weight coefficient of the th final clustering cluster, represents the number of all data points in the data set, represents the th final clustering cluster, and represents the number of data points in the Take the actual coordinate distance of each data point as the automatic rectification difficulty coefficient of the corresponding first sheet.

[0011] Optionally, the steps for calculating the rectification effect coefficient based on the image during the processing are as follows: Obtain the image during the processing of the second sheet, and intercept the processed part of the image from the image during the processing as the actual processed image; compare the actual processed image with the preset processed image corresponding to the processed part, and calculate the structural similarity index between the two images as the rectification effect coefficient.

[0012] Optionally, the steps for processing optimization according to different qualified states of the rectification effect are as follows: Compare the rectification effect coefficient with the preset rectification effect coefficient threshold. If the rectification effect coefficient is not less than the preset rectification effect coefficient threshold, it means that the qualified state of the rectification effect during the current processing is qualified, and continue to process the metal sheet based on the current automatic rectification system; If the rectification effect coefficient is less than the preset rectification effect coefficient threshold, it means that the qualified state of the rectification effect during the current processing is unqualified, and immediately stop the processing of the current metal sheet.

[0013] In the second aspect of the implementation of the present invention, an optimized processing system for metal sheets based on automatic rectification is proposed. The system includes: Calculation module: Denote each metal sheet to be processed as the first sheet, obtain the surface image and processing trajectory of the first sheet, calculate the geometric irregularity coefficient, surface defect coefficient and processing complexity coefficient of the first sheet, and evaluate the geometric shape regularity, surface defect degree and processing complexity of the first sheet; Analysis and judgment module: Analyze the first sheet according to the geometric irregularity coefficient, surface defect coefficient and processing complexity coefficient, and analyze whether the first sheet can be rectified and processed by the current automatic rectification system; Rectification effect module: Denote the first sheet that can be rectified and processed by the current automatic rectification system as the second sheet, and obtain the image during the processing of the second sheet to calculate the rectification effect coefficient; Processing optimization module: Evaluate the qualified state of the rectification effect during the current processing according to the rectification effect coefficient, and perform processing optimization according to different qualified states of the rectification effect.

[0014] Advantages of the present invention: The present invention provides an optimized processing method and system for metal sheets based on automatic rectification. Each metal sheet to be processed is denoted as the first sheet. The surface image and processing trajectory of the first sheet are obtained to analyze whether the first sheet can be rectified and processed by the current automatic rectification system. If it can, the image during the processing of the metal sheet is obtained, and the rectification effect coefficient is calculated according to the image. The qualified state of the rectification effect during the current processing is evaluated according to the rectification effect coefficient, and processing optimization is performed according to different qualified states of the rectification effect. In this way, metal sheets that can be rectified and processed by the current automatic rectification system can be screened out, and it can be determined whether to continue using the current automatic rectification system to process the metal sheets according to the effect of automatic rectification during the processing, and timely intervention can be carried out to reduce the impact on production efficiency and ensure that the product quality is qualified and reduce the waste of production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The present invention will be further described below with reference to the accompanying drawings.

[0016] Figure 1 is a flowchart of an optimized processing method for metal sheets based on automatic rectification; Figure 2 is a framework diagram of an optimized processing system for metal sheets based on automatic rectification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0019] The embodiments of the present invention provide an optimized processing method for metal sheets based on automatic rectification. Refer to Figure 1 , Figure 1 is a flowchart of an optimized processing method for metal sheets based on automatic rectification provided by the embodiments of the present invention. The method includes the following steps: Each metal sheet to be processed is denoted as the first sheet. The surface image and processing trajectory of the first sheet are obtained, the geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient of the first sheet are calculated, and the geometric shape regularity degree, surface defect degree, and processing complexity degree of the first sheet are evaluated; Analyze the first sheet according to the geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient to determine whether the first sheet can be rectified and processed by the current automatic rectification system; Mark the first sheet that can be rectified and processed by the current automatic rectification system as the second sheet, and obtain the rectification effect coefficient by calculating the image during the processing of the second sheet; Evaluate the qualified state of the rectification effect during the current processing according to the rectification effect coefficient, and perform processing optimization according to different qualified states of the rectification effect.

[0020] Based on an optimized processing method for metal sheets based on automatic rectification provided by an embodiment of the present invention, in the above manner, metal sheets that can be rectified and processed by the current automatic rectification system can be screened out, and it can be determined whether to continue using the current automatic rectification system to process the metal sheets according to the effect of automatic rectification during the processing, and intervene in a timely manner to reduce the impact on production efficiency and ensure that the product quality is qualified and reduce the waste of production costs.

[0021] In one embodiment, each metal sheet to be processed is marked as the first sheet, the surface image and processing trajectory of the first sheet are obtained, the geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient of the first sheet are calculated, and the geometric shape regularity degree, surface defect degree, and processing complexity degree of the first sheet are evaluated; Among them, the geometric irregularity coefficient is an index used to quantify the complexity and irregularity of the geometric shape of a metal sheet. It is usually calculated based on the shape characteristics of the sheet, including the curvature of the edge, surface undulation, local curvature change, etc. Specifically, the larger the geometric irregularity coefficient, the more irregular the shape of the sheet, and it may contain more local unevenness, deformation, or complex curved surfaces. For an automatic rectification system, these irregular geometric features will increase the difficulty of its rectification because most traditional automatic rectification systems usually rely on simplified geometric models (such as planes or regular geometric shapes) for correction. When the geometric shape of the metal sheet is more complex, with uneven edges and irregular curved surfaces, the existing automatic rectification system may not be able to accurately identify and adjust these complex forms, resulting in the system being difficult to precisely control the deviation during the processing, thereby affecting the rectification effect. Therefore, an increase in the geometric irregularity coefficient means that higher-precision sensors, more complex calculation models, or more manual intervention are required to effectively perform rectification; Specifically, the calculation steps of the geometric irregularity coefficient are as follows: Obtain the surface image of the first sheet, preprocess the image, and divide the preprocessed image into several local regions; For each local region, calculate the curvature of each pixel coordinate in the local region, and calculate the average value of the curvatures of all pixel coordinates in the region to obtain the geometric irregularity value of the corresponding local region; Multiply the geometric irregularity values of each local region by the corresponding preset weights to obtain the final geometric irregularity values of the corresponding local regions. Calculate the mean value of the final geometric irregularity values of all local regions, and normalize the mean value. Take the result of the normalization process as the geometric irregularity coefficient of the first sheet.

[0022] It should be noted that, first, obtain the surface images of the metal sheet through a high-precision imaging device (such as a camera or a laser scanner); these images should have sufficient resolution to accurately capture the details of the metal sheet. Before further analysis, preprocess the obtained surface images. Common preprocessing steps include: Grayscale conversion: Convert the color image to a grayscale image to simplify the processing. Denoising: Use techniques such as median filtering and Gaussian filtering to remove the noise points in the image and smooth the image. Edge detection: Apply edge detection algorithms (such as the Canny algorithm and the Sobel operator) to extract the edge information on the surface of the metal sheet. Edges are important features describing the surface morphology changes and are crucial for geometric irregularity calculation; Divide into local regions: To reduce the computational complexity, divide the surface image into multiple small regions. The geometric morphology of each small region can be analyzed separately. There are two common division methods: Uniform division: Divide the entire surface into rectangular or square regions of equal size, which is suitable for regular surfaces; Adaptive division: Intelligently divide the regions according to the changes in the surface texture or morphology, which is suitable for irregular surfaces. In this way, the geometric morphology of the sheet surface can be simplified into multiple manageable local regions, facilitating the analysis of the geometric irregularity of each local region.

[0023] It should be noted that the greater the curvature, the greater the degree of surface bending of the local region, which also means that the geometric irregularity of the region is more significant and the geometric irregularity degree of the surface of the first sheet is greater; in addition, the preset weights corresponding to each local region are set by professionals according to the relative importance of the local region. For example, the edge region may require a greater weight.

[0024] In one implementation, the benefit of analyzing the geometric irregularity coefficient of the first sheet for determining whether the current automatic rectification system can accurately rectify its processing process is as follows: The larger the geometric irregularity coefficient, the greater the bending, protrusion or unevenness on the surface of the sheet, which makes the automatic rectification system face higher difficulties during the processing; if this coefficient is too large, it may cause the existing automatic rectification system to be unable to accurately correct the errors, and even may require frequent adjustments and manual interventions. On the contrary, when the geometric irregularity coefficient is small and the surface is relatively flat and regular, the system can more easily perform effective rectification according to the preset models and algorithms; therefore, understanding the geometric irregularity coefficient can not only help to identify possible processing problems in advance, but also avoid blind processing and reduce cost waste.

[0025] In one embodiment, each metal sheet to be processed is denoted as the first sheet. The surface image and processing trajectory of the first sheet are obtained, and the geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient of the first sheet are calculated to evaluate the geometric shape regularity degree, surface defect degree, and processing complexity degree of the first sheet. Among them, the surface defect coefficient is a key indicator to measure the surface defect degree of the metal sheet, and it is usually calculated by analyzing the surface image of the sheet. It reflects the situation of defects existing on the surface of the metal sheet. The larger the surface defect coefficient, the more serious the surface defects of the metal sheet, and there may be more complex irregular structures. Such defects will not only affect the processing accuracy but also may increase the stress concentration of the material during processing. The design of the automatic rectification system is usually optimized for relatively regular and smooth surfaces. When there are many surface defects, the system may not be able to accurately identify and process these defects during rectification, resulting in an unsatisfactory rectification effect. For example, if there are large scratches or cracks, the sensors of the automatic rectification system may not be able to accurately capture their impact on the overall surface shape, thus unable to perform precise correction. Surface defects will affect the sensor's perception of the actual shape of the sheet, increasing the error in the real-time detection and adjustment process of the system. Therefore, the larger the surface defect coefficient, the greater the challenge to the current automatic rectification system's effect, and it is difficult to guarantee the accuracy.

[0026] Specifically, the calculation steps of the surface defect coefficient are as follows: Obtain the surface image of the first sheet, and preprocess the image. Divide the preprocessed image into several windows, and calculate the standard deviation of the pixel values in each window to measure the brightness fluctuation in the local area. A larger standard deviation usually means the existence of defects or irregularities. For each window, compare the standard deviation of the pixel values with a preset standard deviation threshold. If the standard deviation is not less than the preset standard deviation threshold, mark the corresponding window as a defect window. Calculate the total area of the defect windows as the total defect area, and divide the total defect area by the total area of the preprocessed image to obtain the surface defect coefficient of the first sheet.

[0027] It should be noted that the preset standard deviation threshold is set by professionals according to the actual situation, and specific details are not limited and will not be elaborated. In addition, the preprocessing includes denoising, grayscale conversion, etc.; and dividing the preprocessed image into several windows is usually a square window.

[0028] In one implementation, analyzing the surface defect coefficient of the first sheet has the following benefits for determining whether the current automatic rectification system can accurately rectify its processing process: it can accurately evaluate the degree of surface defects of the sheet, thereby avoiding blind processing and unnecessary precision requirements. If the surface defect coefficient is high, it indicates that there are relatively serious defects on the sheet surface, which may lead to unstable or unpredictable deviations during the processing process. The automatic rectification system may not be able to effectively correct these defects, thus affecting the processing quality. In this case, over-relying on the automatic rectification system may not only fail to solve the problem but may also increase costs and waste of time. On the contrary, the quantification of the surface defect coefficient can help the system make more reasonable processing decisions, such as selecting different processing strategies or determining whether manual intervention is required for correction. Through such analysis, the efficiency and accuracy of the processing process can be ensured, resource waste can be reduced, the overall processing quality can be improved, costs can be effectively controlled, and unnecessary losses can be avoided.

[0029] In one embodiment, each metal sheet to be processed is denoted as the first sheet. The surface image and processing trajectory of the first sheet are obtained, and the geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient of the first sheet are calculated to evaluate the geometric shape regularity, surface defect degree, and processing complexity degree of the first sheet. Among them, the processing complexity coefficient is used to measure the complexity of the preset processing trajectory of the first sheet in space; the larger the processing complexity coefficient, the more complex the trajectory; when the processing complexity coefficient is large, the automatic rectification system needs to adjust the movement trajectory of the tool more frequently during the processing process to ensure the processing accuracy. However, due to certain limitations in the response speed, compensation accuracy of the rectification system and the mechanical characteristics of the processing equipment, a more complex processing trajectory may lead to a large deviation between the actual processing path and the theoretical trajectory, thereby affecting the final processing accuracy. In addition, the higher the trajectory complexity, the more obvious the error accumulation effect that may be introduced during the processing process, further reducing the reliability of automatic rectification. Therefore, when the processing complexity coefficient is large, the current automatic rectification system may be difficult to achieve high-precision processing, and even may need to optimize the processing strategy or adjust the trajectory to ensure the final processing quality.

[0030] Specifically, the calculation steps of the processing complexity coefficient are as follows: On the first sheet, a processing trajectory is generated according to the set process requirements and discretized into sampling points, and the position of each discrete point is a three-dimensional coordinate; the trajectory may include straight line segments, curve segments, and changes in different processing depths.

[0031] The trajectory between two adjacent sampling points on the processing trajectory is denoted as a trajectory segment, and the total length of the processing trajectory is calculated. The calculation formula is: , where is the total length of the machining trajectory, represents the length of the th trajectory segment; that is, the Euclidean distance between the th sampling point and the th sampling point; Calculate the included angle between adjacent trajectory segments , and the calculation formula is: , where is the direction vector of the th trajectory segment, and the expression is , is the dot product of vectors, and are the magnitudes of the th and the th trajectory segments respectively; Calculate the local curvature of each trajectory segment, and the calculation formula is: , where is the local curvature of the th trajectory segment; Calculate the machining complexity coefficient, and the calculation formula is: , where is the machining complexity coefficient.

[0032] It should be noted that on each unprocessed first plate, the machining trajectory can be generated according to the set process requirements through CAD / CAM software or a numerical control programming system to generate the machining trajectory of the first plate and perform discretization processing on it, extracting the key sampling points on the trajectory and their three-dimensional coordinate information; it can also be other methods, which are not specifically limited and elaborated here.

[0033] In one implementation, analyzing the machining complexity coefficient of the first plate for determining whether the current automatic rectification system can precisely rectify its machining process has the following advantages: It can predict the complexity of the machining path before machining, thereby evaluating the adaptability and compensation ability of the automatic rectification system and identifying in advance the key areas that may cause excessive machining errors. This helps to optimize machining parameters, such as adjusting the feed rate, optimizing the tool path, or appropriately simplifying the trajectory, to reduce machining errors. In addition, if the trajectory complexity coefficient exceeds the capacity range of the rectification system, the machining strategy can be adjusted in a timely manner, such as using higher-precision equipment, improving the control algorithm, or machining in steps, thereby improving the overall machining accuracy, reducing rework and material waste, and enhancing production efficiency and quality stability.

[0034] In one embodiment, analyze the first plate according to the geometric irregularity coefficient, surface defect coefficient, and machining complexity coefficient to determine whether the first plate can be rectified and machined by the current automatic rectification system; Among them, the steps of analyzing whether the first sheet can be rectified by the current automatic rectification system are as follows: Normalize the geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient so that they are all within [0, 1]. Represent the normalized geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient as data points for each first sheet. The corresponding data points of all first sheets form a data set; perform clustering on the data set to determine the automatic rectification difficulty coefficient of each first sheet; Compare the automatic rectification difficulty coefficient of each first sheet with the preset automatic rectification difficulty coefficient threshold. If the automatic rectification difficulty coefficient is less than the preset automatic rectification difficulty coefficient threshold, it means that the first sheet can be rectified by the current automatic rectification system; if the automatic rectification difficulty coefficient is not less than the preset automatic rectification difficulty coefficient threshold, it means that the first sheet cannot be rectified by the current automatic rectification system.

[0035] It should be noted that the preset automatic rectification difficulty coefficient threshold is set by professionals according to the actual situation and is not specifically limited or elaborated here.

[0036] It should be noted that before processing the metal sheet, comparing the automatic rectification difficulty coefficient of each first sheet with the preset automatic rectification difficulty coefficient threshold is an important step in judging whether the current automatic rectification system can accurately process the sheet; if the automatic rectification difficulty coefficient of a certain first sheet is less than the preset threshold, it means that the geometric irregularity degree, surface defect degree, and processing trajectory complexity of the sheet are all within an acceptable range, and the current automatic rectification system can effectively identify and perform high-precision rectification processing, and processing can be carried out; on the contrary, if the automatic rectification difficulty coefficient of the first sheet is greater than or equal to the preset threshold, it means that the shape deviation of the sheet is too large, there are many surface defects, or the processing path complexity exceeds the capabilities of the current automatic rectification system, which may lead to a decrease in rectification accuracy, error accumulation, and even processing failure during actual processing; therefore, processing cannot be carried out, unnecessary processing attempts are reduced, material waste of the metal sheet is reduced, costs are lowered, and processing efficiency is optimized; at the same time, employees are reminded to find a suitable production line for processing the metal sheet, and the automatic rectification system of the production line can meet the rectification requirements during the processing of the metal sheet; through this warning mechanism, high-risk sheets can be pre-screened before processing, ensuring the stability of the automatic rectification system, and improving processing efficiency and quality.

[0037] In one embodiment, represent the geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient as data points for each first sheet. The corresponding data points of all first sheets form a data set; perform clustering on the data set to determine the automatic rectification difficulty coefficient of each first sheet; Among them, the normalized geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient are expressed as data points for each first sheet, and the corresponding data points of all first sheets form a data set; specifically: The normalized geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient of the first sheet are expressed as data points for each first sheet. The data points of each first sheet can be represented as a three-dimensional vector, that is, [geometric irregularity coefficient, surface defect coefficient, processing complexity coefficient]. The corresponding data points of all first sheets form a data set; the data set is clustered to determine the automatic rectification difficulty coefficient of each first sheet; In one embodiment, the steps of clustering the data set to determine the automatic rectification difficulty coefficient of each first sheet are as follows: The steps of clustering the data set by the K-means clustering method are as follows: S1: Use the elbow method to determine the optimal number of clusters K of the data set; S2: Randomly select K data points in the data set as the initial cluster centers. For each data point, calculate its Euclidean distance to each initial cluster center, traverse the K initial cluster centers, and assign it to the cluster corresponding to the nearest initial cluster center; S3: After traversing all data points, obtain the initial clusters. For each initial cluster, calculate the mean value of the data points within it to obtain a new cluster center; S4: Repeat S2 and S3 until the cluster centers no longer change, obtaining several final clusters and several corresponding final cluster centers; Map the several final clusters to three-dimensional coordinates to obtain the automatic rectification difficulty coefficient of each first sheet.

[0038] It should be noted that the K-means clustering method is a commonly used unsupervised learning algorithm for dividing a data set into a predefined number K of clusters; the goal of this algorithm is to divide the data points into K clusters such that each data point belongs to the cluster represented by its nearest cluster center, and the data points within the cluster are as similar as possible, while the data points between different clusters may not be similar; The K-means clustering method classifies these devices into a cluster, and based on the center point of this cluster, the automatic rectification difficulty coefficient can be obtained for each first sheet; and the geometric irregularity coefficient, surface defect coefficient, processing complexity coefficient, and automatic rectification difficulty coefficient of each first sheet show a common linear relationship. When the geometric irregularity coefficient is larger, the surface defect coefficient is larger, and the processing complexity coefficient is larger, the corresponding automatic rectification difficulty coefficient of the first sheet is larger.

[0039] Specifically, the step of assigning a data point to the cluster to which the nearest clustering center belongs is to sort the distances between the data point and each clustering center from largest to smallest, and select the smallest distance to assign the data point to the cluster it belongs to.

[0040] In one embodiment, mapping a number of final clustering clusters and final clustering centers to three-dimensional coordinates, the obtained automatic rectification difficulty coefficient of each first sheet includes: Calculate the weight coefficient of each final clustering cluster, and its calculation formula is , where represents the weight coefficient of the th final clustering cluster, represents the number of all data points in the dataset, represents the th final clustering cluster, and represents the number of data points in the th final clustering cluster;

[0041] In one implementation manner, through clustering analysis and weighted distance calculation, the automatic rectification difficulty degree of the first sheet can be comprehensively considered.

[0042] It should be noted that since the geometric irregularity coefficient, surface defect coefficient, and processing complexity coefficient after normalization are selected as data points for clustering, now these three data points can be used as coordinates and mapped to the three-dimensional coordinate axes. The automatic rectification difficulty coefficient of each first sheet will be mapped to a point on the three-dimensional coordinate axes; after clustering is completed, all data points will be distributed in different final clustering clusters on the three-dimensional coordinate axes. The closer the final clustering cluster is to the origin and the data points it includes, the greater the automatic rectification difficulty coefficient of the corresponding first sheet, and vice versa; at the same time, calculate the weight coefficient of each final clustering cluster to correct the distance from the data point to the origin. In this way, the different importance of each final clustering cluster can be ensured, and at the same time, the influence of noise and outliers on the clustering result can be reduced to a certain extent, making the finally calculated automatic rectification viscosity coefficient of the first sheet more accurate.

[0043] In one embodiment, the first sheet that can be rectified by the current automatic rectification system is denoted as the second sheet, and the rectification effect coefficient is calculated by obtaining the image during the processing of the second sheet; Specifically, the steps of calculating the rectification effect coefficient according to the image during the processing are: Obtain the image during the processing of the second sheet metal, and intercept the processed part of the image from the image during the processing as the actual processed image; compare the actual processed image with the preset processed image corresponding to the processed part, and calculate the structural similarity index between the two images as the deviation correction effect coefficient. It should be noted that during the calculation process, the deviation correction effect coefficient needs to be obtained by acquiring the processing image in real time and comparing it with the preset processing image. Image acquisition is usually carried out by a camera or sensor installed on the processing equipment for real-time capture. The processed part is extracted through image processing algorithms, and the similarity with the preset image is calculated.

[0044] It should be noted that during the processing of the second sheet metal, the larger the deviation correction effect coefficient, the better the deviation correction effect of the automatic deviation correction system. Because the larger the deviation correction effect coefficient, it means that the automatic deviation correction system can adjust the processing path and control errors more precisely, so that the actual processing state of the sheet metal is closer to the ideal processing state. This precise adjustment and deviation correction can effectively reduce the deviation generated during the processing, improve the processing quality, reduce the impact of errors on the quality of the final product, and ensure that the processing process is more stable and efficient.

[0045] In one implementation, the advantage of using the structural similarity index to measure the deviation correction effect of the automatic deviation correction system is that the structural similarity index can effectively capture the subtle differences between images, especially the changes in texture, shape, and local structure. This method is more sensitive than traditional pixel-level comparison because it does not simply focus on the color or brightness value of each pixel, but evaluates the similarity of the image from the perspectives of overall and local by analyzing the structural information of the image. Therefore, the structural similarity index can truly reflect the small errors and deviations in details during the processing, making the evaluation of the deviation correction effect more accurate and reliable. This has important practical significance for optimizing the automatic deviation correction system and improving its adaptability and adjustment ability to complex processing paths.

[0046] In one embodiment, the steps of evaluating the qualified state of the deviation correction effect during the current processing according to the deviation correction effect coefficient and performing processing optimization according to different qualified states of the deviation correction effect are as follows: Compare the deviation correction effect coefficient with the preset deviation correction effect coefficient threshold. If the deviation correction effect coefficient is not less than the preset deviation correction effect coefficient threshold, it means that the qualified state of the deviation correction effect during the current processing is qualified, and continue to process the metal sheet based on the current automatic deviation correction system; If the deviation correction effect coefficient is less than the preset deviation correction effect coefficient threshold, it means that the qualified state of the deviation correction effect during the current processing is unqualified, and immediately stop the processing of the current metal sheet.

[0047] It should be noted that the preset threshold of the deviation correction effect coefficient is set by professionals according to the actual situation, and specific details are not limited and elaborated here.

[0048] In one implementation, when the deviation correction effect coefficient is not less than the preset threshold, it indicates that the automatic deviation correction system can maintain a high processing accuracy, ensuring the quality and consistency of metal sheet processing. At this time, the current system can continue to be used for processing to ensure that excessive errors or deviations will not occur during the processing. On the contrary, if the deviation correction effect coefficient is lower than the preset threshold, it means that the system fails to effectively correct the deviation, which may lead to a large deviation in the processing quality. At this time, the processing must be stopped immediately to avoid the continuous production of unqualified metal sheets. Through this mechanism, the automatic deviation correction system can monitor the processing quality in real time, detect problems in a timely manner and take measures, thereby avoiding the accumulation of errors and waste of resources in the production process. This optimized processing solution for metal sheets based on automatic deviation correction significantly improves the accuracy and reliability of the processing process. It not only optimizes the operation efficiency of the automatic deviation correction system but also can dynamically monitor and adjust the processing process. By evaluating the deviation correction effect in real time, it ensures that the processing quality is always maintained within the qualified range, reducing rework or waste caused by errors. At the same time, the system can stop the machine in time, preventing the continuous processing of unqualified products, avoiding resource waste and cost increase caused by unqualified products, and thus improving production efficiency and economic benefits.

[0049] Based on the same inventive concept, an embodiment of the present invention further provides an optimized processing system for metal sheets based on automatic deviation correction. Refer to Figure 2 , Figure 2 which is a framework diagram of an optimized processing system for metal sheets based on automatic deviation correction provided by an embodiment of the present invention. The system includes: Calculation module: Denote each metal sheet to be processed as the first sheet, obtain the surface image and processing trajectory of the first sheet, calculate the geometric irregularity coefficient, surface defect coefficient and processing complexity coefficient of the first sheet, and evaluate the geometric shape regularity degree, surface defect degree and processing complexity degree of the first sheet; Analysis and judgment module: Analyze the first sheet according to the geometric irregularity coefficient, surface defect coefficient and processing complexity coefficient to analyze whether the first sheet can be corrected and processed by the current automatic deviation correction system; Deviation correction effect module: Denote the first sheet that can be corrected and processed by the current automatic deviation correction system as the second sheet, and obtain the deviation correction effect coefficient by calculating the image during the processing of the second sheet; Processing optimization module: Evaluate the qualified state of the deviation correction effect during the current processing according to the deviation correction effect coefficient, and perform processing optimization according to different qualified states of the deviation correction effect.

[0050] Based on an optimized metal sheet processing system with automatic deviation correction provided by an embodiment of the present invention, through the above method, it is possible to screen out metal sheets that can be processed with deviation correction by the current automatic deviation correction system, and based on the effect of automatic deviation correction during the processing, determine whether to continue using the current automatic deviation correction system to process the metal sheets, and intervene in a timely manner to reduce the impact on production efficiency, ensure that the product quality is qualified, and reduce the waste of production costs.

[0051] The above has described in detail an embodiment of the present invention, but the content described is only the preferred embodiment of the present invention and cannot be used to artificially limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. A metal sheet optimization processing method based on automatic deviation correction, characterized in that: The following steps are involved: Each metal plate to be processed is recorded as a first plate, a surface image and a processing trajectory of the first plate are obtained, a geometric irregularity coefficient, a surface defect coefficient, and a processing complexity coefficient of the first plate are calculated, and a geometric shape regularity, a surface defect degree, and a processing complexity degree of the first plate are evaluated; Analyze the first plate according to the geometric irregularity coefficient, the surface defect coefficient and the processing complexity coefficient to analyze whether the first plate can be corrected by the current automatic correction system; The first plate that can be corrected by the current automatic correction system is recorded as the second plate, and the image of the second plate during the processing is obtained to calculate the correction effect coefficient; The qualified status of the correction effect in the current processing is evaluated according to the correction effect coefficient, and the processing is optimized according to different qualified status of the correction effect.

2. The metal sheet optimization processing method based on automatic deviation correction according to claim 1 is characterized in that: The calculation steps of the geometric irregularity coefficient are: Acquire a surface image of the first plate, preprocess the image, and divide the preprocessed image into a plurality of local areas; For each local area, the curvature of each pixel coordinate in the local area is calculated, and the mean value of the curvature of all pixel coordinates in the area is calculated to obtain the geometric irregularity value of the corresponding local area; The geometric irregularity value of each local area is multiplied by the corresponding preset weight to obtain the final geometric irregularity value of the corresponding local area, the average of the final geometric irregularity values ​​of all local areas is calculated, and the average is normalized, and the result of the normalization is used as the geometric irregularity coefficient of the first plate.

3. The metal sheet optimization processing method based on automatic deviation correction according to claim 1 is characterized in that: The calculation steps of surface defect coefficient are: Obtaining a surface image of the first plate, and preprocessing the image, dividing the preprocessed image into a number of windows, and calculating the standard deviation of the pixel values ​​in each window to measure the brightness fluctuation in the local area; a larger standard deviation usually means the presence of defects or irregularities; For each window, the standard deviation of the pixel value is compared with the preset standard deviation threshold. If the standard deviation is not less than the preset standard deviation threshold, the corresponding window is recorded as a defective window. The total area of ​​the defect windows is calculated as the total defect area, and the total defect area is divided by the total area of ​​the preprocessed image to obtain the surface defect coefficient of the first plate.

4. The metal sheet optimization processing method based on automatic deviation correction according to claim 1 is characterized in that: The calculation steps of the processing complexity coefficient are as follows: On the first plate, the machining trajectory is generated according to the set process requirements and discretized into sampling points, and each discrete point is a three-dimensional coordinate; The trajectory between two adjacent sampling points on the processing trajectory is recorded as a trajectory segment, and the total length of the processing trajectory is calculated. The calculation formula is: , where is the total length of the machining trajectory, Indicates The length of the trajectory segment; sampling points and The Euclidean distance between the sampling points; Calculate the angle between adjacent trajectory segments , the calculation formula is: , where For the The direction vector of the trajectory segment is expressed as , is the vector dot product, and Respectively and The modulus of each trajectory segment; Calculate the local curvature of each trajectory segment using the following formula: , where For the The local curvature of each trajectory segment; Calculate the processing complexity coefficient, the calculation formula is: , where is the processing complexity coefficient.

5. The metal sheet optimization processing method based on automatic deviation correction according to claim 1 is characterized in that: The steps for analyzing whether the first sheet can be corrected by the current automatic correction system are as follows: The geometric irregularity coefficient, the surface defect coefficient and the processing complexity coefficient are all normalized so that they are all between [0,1], and the normalized geometric irregularity coefficient, the surface defect coefficient and the processing complexity coefficient are represented as data points of each first plate, and the corresponding data points of all the first plates constitute a data set; the data set is clustered to determine the automatic correction difficulty coefficient of each first plate; The automatic correction difficulty coefficient of each first plate is compared with the preset automatic correction difficulty coefficient threshold. If the automatic correction difficulty coefficient is less than the preset automatic correction difficulty coefficient threshold, it means that the first plate can be corrected by the current automatic correction system; if the automatic correction difficulty coefficient is not less than the preset automatic correction difficulty coefficient threshold, it means that the first plate cannot be corrected by the current automatic correction system.

6. The metal sheet optimization processing method based on automatic deviation correction according to claim 5 is characterized in that: The steps of clustering the data set to determine the automatic deviation correction difficulty coefficient of each first plate are: The data set is clustered by K-means clustering method to obtain several final clusters, and each final cluster is mapped to a three-dimensional coordinate; Calculate the weight coefficient of each final cluster, the calculation formula is: ,in Expressed as The weight coefficients of the final clusters, Represented as the number of all data points in the dataset, Indicates The number of data points in the final clusters; Calculate the distance of each data point in each final cluster to the coordinate origin in three-dimensional coordinates to obtain the preliminary coordinate distance of the data point, and multiply the preliminary coordinate distance of the data point by the weight coefficient corresponding to the final cluster in which the data point is located to obtain the actual coordinate distance of the data point; The actual coordinate distance of each data point is used as the automatic deviation correction difficulty coefficient of the corresponding first plate.

7. The metal sheet optimization processing method based on automatic deviation correction according to claim 1 is characterized in that: The steps for calculating the correction effect coefficient based on the image during processing are: The image of the second plate during processing is obtained, and the processed part of the image is intercepted from the image during processing as the actual processing image; the actual processing image is compared with the preset processing image corresponding to the processed part, and the structural similarity index between the two images is calculated as the correction effect coefficient.

8. The metal sheet optimization processing method based on automatic deviation correction according to claim 1 is characterized in that: The steps for processing optimization according to different qualified states of correction effect are as follows: The correction effect coefficient is compared with the preset correction effect coefficient threshold. If the correction effect coefficient is not less than the preset correction effect coefficient threshold, it means that the correction effect in the current processing process is qualified, and the metal sheet is processed based on the current automatic correction system. If the correction effect coefficient is less than the preset correction effect coefficient threshold, it means that the qualified state of the correction effect in the current processing process is unqualified, and the processing of the current metal sheet is stopped immediately.

9. A metal sheet optimization processing system based on automatic deviation correction, used to implement a metal sheet optimization processing method based on automatic deviation correction as claimed in any one of claims 1 to 8, characterized in that: The system comprises: Calculation module: record each metal plate to be processed as a first plate, obtain the surface image and processing trajectory of the first plate, calculate the geometric irregularity coefficient, surface defect coefficient and processing complexity coefficient of the first plate, and evaluate the geometric shape regularity, surface defect degree and processing complexity of the first plate; Analysis and judgment module: analyzing the first plate according to the geometric irregularity coefficient, the surface defect coefficient and the processing complexity coefficient, and analyzing whether the first plate can be corrected by the current automatic correction system; Correction effect module: the first plate that can be corrected by the current automatic correction system is recorded as the second plate, and the image during the processing of the second plate is obtained to calculate the correction effect coefficient; Processing optimization module: Evaluate the qualified status of the correction effect in the current processing process according to the correction effect coefficient, and optimize the processing according to different qualified status of the correction effect.

Images