A hot-rolled aluminum coil intelligent and standardized welding position identification method and system

Abstract

The application discloses a kind of hot-rolled aluminum roll intelligent and standardized weld position identification method and system, belong to weld position identification technical field, the present application aims at realizing the accurate planning of large aluminum roll welding path.First, the accurate identification of welding area is carried out, including the identification of three weld point areas of outer ring end face, one weld point area of inner ring end face and one weld point area of band head.This method can automatically exclude the abnormal areas in the aluminum roll that are not suitable for welding, such as warping and misalignment, thereby achieving high-quality welding area identification. Then, using the accurate identification result, a standardized welding path planning is carried out, providing accurate guidance information for subsequent welding work.

Description

Technical Field

[0001] The present invention belongs to the technical field of welding position recognition, and in particular relates to an intelligent and standardized welding position recognition method and system for hot-rolled aluminum coils. Background Art

[0002] In the current aluminum coil production process, manual welding is required after cold rolling to enhance the coil's stability and prevent cracking during subsequent transportation. However, manual welding currently presents two major challenges: first, standardization. Manual welding is typically performed based on experience, resulting in an inconsistency in the number and placement of weld points. Furthermore, height issues often lead to weld points being concentrated in the lower half, creating a potential risk of cracking. Second, safety concerns arise. The high-temperature and toxic working environment poses a significant threat to worker health. Therefore, intelligent welding systems that utilize robotic arms and deep learning methods for automated, intelligent welding are necessary to replace manual welding.

[0003] However, due to the inherent metallic properties of aluminum, the unique structure of aluminum coils, and the weak light conditions prevalent in actual production environments, this poses a series of challenges to intelligent welding systems. Specifically: 1) Structural challenges. The coiled structure of aluminum causes misalignment between layers, resulting in welds not being on the same plane. This is significantly different from the flat seam welding of steel plates, making the seam recognition module designed for steel plates ineffective. 2) Lighting challenges. Aluminum production often involves a low-light environment, which makes it difficult to identify welds within aluminum coils during this period. In this case, this increases the complexity of weld identification. 3) Material sensitivity. As a soft metal, artificial intelligence is very sensitive to force. The welding point affects the internal stress distribution of the entire aluminum coil, and improper handling may lead to uneven stress and internal collapse. Therefore, the selection of the welding path is crucial. Summary of the Invention

[0004] To address these challenges, this invention proposes a novel intelligent identification method for aluminum coil welding areas. This method comprehensively considers the characteristics of aluminum coils and the various anomalies that may arise during the production process. By sampling the aluminum coil from multiple angles and conducting data analysis, it can accurately identify ten specific weldable areas on both sides of the coil head and end face. Furthermore, this technology utilizes a standardized identification process, combined with advanced image processing techniques and deep learning algorithms, to achieve intelligent identification of aluminum coil welding areas. This not only significantly reduces operator involvement in high-risk environments but also improves aluminum coil production efficiency and welding quality. This innovative invention not only ensures the health and safety of operators, but also brings about a dual improvement in efficiency and safety in the aluminum coil hot rolling industry.

[0005] The technical solution provided by the present invention is: an intelligent and standardized welding position identification method for hot-rolled aluminum coils, comprising:

[0006] The 2D image and 3D point cloud data of the aluminum coil are collected. The brightness of the 2D image is enhanced to address low-light issues, and the noise of the 3D point cloud is removed. The shooting area includes an end face lead area and three outer end face peripheral areas. The three outer end face peripheral areas are located at the vertices of the equilateral triangle inscribed in the end face circle.

[0007] A target instance segmentation network was constructed for key feature detection of aluminum coils. This network consists of two branches: a 2D branch based on the YOLOv8 model, which extracts a single weld point region on the inner wall of the coil from the 2D image. The 3D branch, based on the PointNet++ architecture, extracts the circular end faces from the 3D point cloud, including three weld points on the outer end face and one weld point on the inner end face. These regions are then classified into non-weldable and weldable regions.

[0008] The welding area is further divided within the weldable area, and accurate path planning is performed on the weld point area of ​​the inner wall and the welding area based on the actual size of the welding gun.

[0009] Furthermore, a two-stage exposure method is used to collect 2D images and 3D point cloud data, with the first stage being long exposure and the second stage being short exposure.

[0010] Furthermore, adaptive block-wise γ-correction is used to enhance the brightness of the 2D image. The specific implementation is as follows:

[0011] First, the image is divided into blocks, and the γ value is adaptively adjusted according to the average brightness of each block:

[0012]

[0013] Where LU(p ij ) represents the brightness of the pixel, M and N are the length and width of the image after the block is divided; calculate the average brightness, using 128 as the threshold: areas with brightness above 128 are considered darker, and a γ value less than 1 is applied to enhance the brightness. The smaller the γ value, the stronger the enhancement; areas below 128 are considered brighter, and a γ value greater than 1 is used to compress the brightness. The larger the γ value, the stronger the compression. Then use a lookup table to accelerate the γ correction in the image block:

[0014]

[0015] k represents the current pixel, Indicates the enhanced pixel brightness value, and performs inverse γ enhancement on the pixel.

[0016] Furthermore, a statistical method based on the z-axis is used to denoise the noise points. The intervals are divided according to the length of the z-axis to obtain the interval length h:

[0017]

[0018] Among them, h is the length of the interval, l is the length of the point cloud in the z-axis interval, and z max is the maximum value of the z-axis data in the point cloud data, z min is the minimum value of the z-axis data in the point cloud data, and then record the number of points in each interval, using the interval with the most points as the reference:

[0019]

[0020] P j is the set of all points in the jth interval, and all points p in the interval i Satisfy the z-axis value z i In a size of The limits are set successively by extending to adjacent zero intervals:

[0021]

[0022] j lower is the lower limit of the z-axis size of the final selected valid point, j upper The upper limit of the z-axis size of the final selected valid points, and the points outside these limits are regarded as noise and removed.

[0023] Furthermore, the input of the two-dimensional branch is a 2D image after brightness enhancement, and it outputs a feature map for predicting the box, a feature map for predicting the classification, and a mask coefficient for segmentation. At the same time, the output of the P5 layer in the Head is upsampled to obtain a mask prototype; the model output is filtered into two prediction boxes by NMS, and the mask prototype and mask coefficient are used to generate the target prediction mask. The prediction box is cropped and thresholded to obtain the segmentation result.

[0024] Furthermore, the input of the three-dimensional branch is a set of three-dimensional coordinates. First, a clustering algorithm is used in the sampling layer to select representative points from the input point cloud to form a smaller subset. Then, for each sampling point, a group of points are selected in its neighborhood in the grouping layer, and a spherical or K-nearest neighbor search is used to define the neighborhood. Finally, the feature extraction layer applies a small PointNet to the points in each neighborhood to extract local features. The local features are processed by MLP to generate a feature vector of fixed length. Through layer-by-layer aggregation, the local features are gradually summarized to a higher level to obtain a global feature vector. The fully connected layer further processes and classifies the extracted global feature vector, and finally outputs the classification label to divide the 3D point cloud into non-weldable areas and weldable areas.

[0025] Furthermore, for a welding point area of ​​the inner wall head, the center line is the actual weld path, and the actual weld path is an arc.

[0026] Furthermore, the welding area is further divided within the weldable area according to the thickness n of the plate. When n is less than a certain threshold, the welding area moves radially inward and downward by k1 mm from the boundary line of the weldable area. When n is greater than a certain threshold, the welding area moves radially inward and downward by k2 mm from the boundary line of the weldable area. The outer edge of the welding area is parallel to the outer edge of the weldable area. Wherein k1 and k2 are constants.

[0027] Furthermore, the thickness of the plate and strip is obtained by transferring parameters during the rolling process of the aluminum coil, and the welding area is a rectangular arc.

[0028] The present invention also provides an intelligent and standardized welding position identification system for hot-rolled aluminum coils, comprising:

[0029] A processor and a memory, the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute an intelligent and standardized welding position identification method for hot-rolled aluminum coils as described in the above technical solution.

[0030] Compared with the prior art, the beneficial effects and advantages of the present invention are as follows:

[0031] ① Improved production efficiency: Traditionally, identification of welding zones relies on experienced workers working at close range, which, especially in high-temperature environments, requires frequent breaks and adjustments. The proposed system can operate continuously for 8 to 10 hours at 50°C, significantly improving production efficiency.

[0032] ② Improve production standardization: During manual welding, the selection of welding positions and weld points usually relies on the worker's experience and is highly subjective. This system can perform scientific and accurate weld point selection based on accurate feature recognition, ensuring standardized weld point selection to meet engineering requirements and providing a standardized solution for automated weld point area identification across the industry.

[0033] ③ Reduce the impact of the production environment on human health: The temperature in aluminum coil production workshops is usually above 40°C. In addition, the temperature during aluminum coil welding is also very high. When workers are working at close range, they are exposed to high temperatures for a long time, which may have adverse effects on their health. This system significantly reduces workers' exposure time to high temperatures and reduces the harm to their health caused by high-temperature operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is the algorithm workflow diagram.

[0035] Figure 2 Schematic diagram of data collection and processing.

[0036] Figure 3 Schematic diagram of normalized welding area. DETAILED DESCRIPTION

[0037] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0038] The embodiment of the present invention proposes an intelligent planning method for the welding area path of large aluminum coils. This method aims to solve the problem of intelligence in the aluminum coil welding process, thereby achieving efficient and standardized welding operations. It includes three parts: data acquisition, feature extraction and path planning.

[0039] The data acquisition component comprehensively considers issues such as temperature, shooting accuracy, and receptive field in the actual production workshop to optimize the optimal shooting sequence. Image preprocessing is then performed based on the actual captured images: brightness enhancement is performed to address low-light issues in 2D images, and noise removal is performed to address noise issues in 3D point clouds. This ensures accurate model recognition. The feature extraction component uses image segmentation to segment the lead portion in the 2D image and the staggered portions in the 3D point cloud. This is crucial for accurate welding path planning. Based on the feature extraction results, the path planning component rationally plans the number and location of weld points, including the weld area where the outer ring end face forms an equilateral triangle, the weld area where the inner ring end face extends beyond the lead position, and the weld area at the outer end of the lead. Precise path planning is also performed based on the actual size of the welding gun, providing a reliable and accurate welding path for the gun.

[0040] The specific working process of the present invention is as follows:

[0041] This invention aims to achieve precise planning of welding paths for large aluminum coils. First, the welding area is precisely identified, including three weld points on the outer coil end face, one weld point on the inner coil end face, and one weld point on the lead. This method automatically eliminates abnormal areas in the aluminum coil that are unsuitable for welding, such as warpage and misalignment, thereby achieving high-quality weld area identification. The precise identification results are then used to perform standardized welding path planning, providing precise guidance for subsequent welding operations. The specific workflow is as follows:

[0042] Lead welding path planning: The robotic arm, carrying a camera, moves to a position 100-150 cm from the center of the aluminum coil (the distance depends on the coil's size) to collect data. The collected data is fed into a pre-trained neural network model to segment the weld area at the coil's head. The segmentation results are used to generate subsequent welding guidance information. Workers can monitor and supervise the process in the background to prevent occasional anomalies from affecting segmentation accuracy and causing recognition errors. Path planning is performed based on the recognition results, generating discrete point coordinates for the path. This path is then converted from the image coordinate system to the robotic arm coordinate system to guide the welding torch.

[0043] End face welding path planning: Based on the height and position of the aluminum coil, the robotic arm carrying the camera first moves to the highest point in the center of the aluminum coil for data collection. The collected data is input into the network model to eliminate abnormal positions such as staggered layers and warping. Finally, five three-dimensional coordinate points based on the robotic arm are output at the outermost circle to calibrate the welding arc area at the highest point. The arc area is 50mm long and has a fixed width of 5mm. In addition, the robotic arm needs to move to two designated positions below to collect information. After similar processing, the coordinate information of the rectangular welding area along the diameter is returned. In this process, since the point cloud coordinates of the MechProS camera are based on the camera as the center of the circle, additional coordinate conversion is required using the coordinates of the center of the aluminum coil to obtain the true rectangular welding area along the diameter.

[0044] The present invention is a method for realizing intelligent identification of welding area of ​​large aluminum coils, which includes three parts: data collection, feature extraction and path planning. The overall flow chart is as follows: Figure 1 The detailed design is as follows:

[0045] 1. Data Collection

[0046] The shooting sequence and receiving field cover four areas, including an end face leading area and three end face peripheral areas. The shooting of the end face leading area will simultaneously capture the leading area and the inner circle of the end face, and obtain two data at a time. The three end face peripheral areas are the vertices of the equilateral triangle inscribed in the end face circle. These areas are selected for feature-rich data collection. A two-stage exposure method is adopted: long exposure 50 milliseconds, short exposure 7 milliseconds. The two-stage exposure method avoids the problems of dark light and incomplete point cloud data in a single exposure information collection. The two-stage exposure acquisition ensures that the details of the light and dark areas are retained in high-contrast environments, reduces noise, and improves imaging quality and recognition accuracy. In order to optimize the point cloud accuracy, the shooting distance is set to 1 meter. In order to completely cover the inner circle of the center, the distance is increased to 1.5 meters, slightly sacrificing accuracy. Data preprocessing is carried out based on the shooting characteristics of the two parts of the data to ensure the accuracy of feature recognition. The flow chart is as follows Figure 2 shown.

[0047] 2D image brightness enhancement: During the data collection process, changes in workshop conditions and environment brought some problems: in 2D images, low brightness caused by changes in natural light and production activities reduces image clarity, and brightness needs to be enhanced. We use adaptive block-wise γ-correction to enhance brightness and improve segmentation performance. Traditional γ-correction applies a uniform transformation across the entire image, which may over-enhance or compress details in certain areas, thereby reducing image quality. To solve this problem, we divide the image into blocks and adaptively adjust the γ value based on the average brightness of each block:

[0048]

[0049] Where LU(p ij ) represents the brightness of the pixel, and M and N are the length and width of the image after it is divided. Average brightness is calculated, using 128 as the threshold: areas with brightness above 128 are considered darker, and a gamma value less than 1 is applied to enhance the brightness. Smaller gamma values ​​increase the enhancement. Areas below 128 are considered brighter, and gamma values ​​greater than 1 are used to compress the brightness. Larger gamma values ​​increase the compression. A lookup table is then used on the image blocks to accelerate gamma correction:

[0050]

[0051] k represents the current pixel, Indicates the enhanced pixel brightness value, and performs inverse γ enhancement on the pixel.

[0052] 3D point cloud noise removal: The collected point cloud data has a high density, with the main object density reaching 200 points / cm 3 , the noise distribution is characterized by a small number of points clustered together. In order to reduce the noise of the point cloud, we use a statistical method based on the z-axis to reduce the noise. The interval is divided according to the length of the z-axis, and the interval length h is obtained:

[0053]

[0054] Where l is the length of the z-axis interval occupied by the point cloud, z max is the maximum value of the z-axis data in the point cloud data, z min is the minimum value of the z-axis data in the point cloud data, and then record the number of points in each interval, using the interval with the most points as the reference:

[0055]

[0056] P j is the set of all points in the jth interval, P is the set of all points, and all points p in the interval i Satisfy the z-axis value z i In a size of The limits are set successively by extending to adjacent zero intervals:

[0057]

[0058] j lower is the lower limit of the z-axis size of the final selected valid point, j upper The upper limit of the z-axis size of the final selected valid points, and the points outside these limits are regarded as noise and removed.

[0059] 2. Feature Extraction

[0060] Accurately extracting the features of the aluminum coil is crucial during the initial localization process. We propose an object instance segmentation network for detecting key features of the aluminum coil. This network consists of two branches: a 2D branch and a 3D branch. The 2D branch extracts the inner wall's lead region, while the 3D branch extracts the ring end face, including the inner and outer rings, and further classifies them into non-weldable and weldable areas.

[0061] Head Region Recognition: For the 2D image input to the model, brightness enhancement is performed on low-brightness pixels using a block-wise γ-correction method to double the brightness of low-brightness pixels. The image is then resized to 3×640×640. The network model is based on the YOLOv8 model and outputs feature maps for predicted boxes, feature maps for predicted classification, and mask coefficients for segmentation. Simultaneously, the output of the P5 layer in the head is upsampled to produce a mask prototype. The network prediction output is filtered into two predicted boxes using NMS. The mask prototype and mask coefficients are used to generate the target prediction mask. The predicted boxes are cropped and thresholded to produce the segmentation result.

[0062] Endface region recognition: For input 3D point cloud data, noise is removed from noisy images. Since the number of points in the original image can typically reach 2 million, 100,000 points are randomly sampled from the original image as input to the CNN. This network model is based on the PointNet++ architecture, which features a hierarchical feature learning mechanism tailored for irregular point cloud data. The input is a set of 3D coordinates. The sampling layer first uses a clustering algorithm to select representative points from the input point cloud, forming a small subset. The grouping layer then selects a set of points within the neighborhood of each sampled point, using a spherical or K-nearest neighbor search to define the neighborhood. Finally, the feature extraction layer applies a small PointNet to the points within each neighborhood to extract local features. These features are processed by an MLP (Multi-Layer Perceptron) to generate a fixed-length feature vector. Through layer-by-layer aggregation, local features are gradually aggregated to higher levels, similar to the local-to-global feature extraction process in a CNN, resulting in a global feature vector. The fully connected layer further processes and classifies the extracted global feature vector, ultimately outputting a classification label, categorizing it as a non-weldable area or a weldable area. In practice, the point cloud data of three welds around an aluminum coil have similar shapes and can be treated as a single class by rotating them differently. The input point cloud data is randomly downsampled and then passed through the PointNet++ network to generate segmentation predictions for each point.

[0063] 3. Path Planning

[0064] According to the result of feature extraction, the target area is further divided into paths, such as Figure 3As shown. According to actual production experience and demand, five welding areas are obtained on one side: one for the head part, i.e. Figure 3 The welding area 1 in the figure is obtained by the 2D branch; the inner circle part of the end face is one, that is, Figure 3 The welding area 2 in the end face has three outer ring parts, namely Figure 3 The welding areas 3-5 in the figure are all obtained by 3D branch.

[0065] Leading area path planning: Based on the recognition results, such as Figure 3 The blue area at the head of the middle strip is used as the target area. The approximate center line is the actual weld path. Due to the limitations of the welding width and penetration angle of the welding gun, the path is set to 100 mm from the boundary and divided into arc areas with a length of 50 mm and a width of 5 mm.

[0066] End face area path planning: Based on the recognition results, such as Figure 3 Taking the lower right portion of the middle outer ring end face as an example, the blue area is the target area. The selection of this area is limited by the flatness and compactness required for welding, which requires that the weld area must not have any misalignment or warping.

[0067] To prevent welding gun collisions, the weld zone is offset inward. The green area represents the non-weldable zone (including abnormalities such as dents, protrusions, and warping), while the blue area represents the weldable zone. Both zones have distinct boundaries. In practical applications, the weld zone is closely aligned with the boundary between the weldable and non-weldable zones, and the weld zone is divided within the weldable zone. Considering the 5mm welding gun width and the maximum thickness of the aluminum coil is 6mm, with a minimum weld seam coverage requirement, this zone is divided based on the coil thickness n (derived from the parameters transferred during the aluminum coil rolling process). When n < 4, the weld zone is radially shifted inward by 1mm from the boundary. When n > 4, the weld zone is shifted downward (i.e., radially shifted inward) by 3mm from the boundary. The weld zone is approximately a 5mm x 50mm rectangular arc. The weld zone is shifted downward based on the coil thickness. The outer edge of the weld zone is parallel to the outer edge of the weldable zone, resulting in an arc shape, while the inner edge is a straight line. Since the 50mm length is extremely short compared to the entire coil, it can be approximated as an arc.

[0068] In summary, the present invention provides an advanced intelligent automated planning solution for aluminum coil welding paths. Through meticulous technological innovation and practical verification, it ensures the accuracy, reliability, and safety of welding operations while improving production efficiency.

[0069] On the other hand, an embodiment of the present invention further provides an intelligent and standardized welding position identification system for hot-rolled aluminum coils, comprising:

[0070] A processor and a memory, the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute an intelligent and standardized welding position identification method for hot-rolled aluminum coils as described in the above technical solution.

[0071] The specific embodiments described herein are merely illustrative of the spirit of the present invention. Persons skilled in the art may make various modifications, additions, or substitutions to the described specific embodiments without departing from the spirit of the present invention or exceeding the scope of the appended claims.

Claims

1. An intelligent and standardized method for identifying welding positions of hot-rolled aluminum coils, characterized in that: include: The 2D image and 3D point cloud data of the aluminum coil are collected. The brightness of the 2D image is enhanced to address low-light issues, and the noise of the 3D point cloud is removed. The shooting area includes an end face lead area and three outer end face peripheral areas. The three outer end face peripheral areas are located at the vertices of the equilateral triangle inscribed in the end face circle. Adaptive block-wise γ-correction is used to enhance the brightness of 2D images. The specific implementation is as follows: First, the image is divided into blocks, and the γ value is adaptively adjusted according to the average brightness of each block: In the formula Represents the brightness of the pixel, M and N are the length and width of the image after segmentation respectively; Calculate the average brightness and use 128 as the threshold: areas with brightness above 128 are considered darker, and a γ value less than 1 is applied to enhance the brightness. The smaller the γ value, the stronger the enhancement. Areas below 128 are considered brighter, and a γ value greater than 1 is used to compress the brightness. The larger the γ value, the stronger the compression. Then use a lookup table on the image block to speed up the γ correction: k Represents the current pixel, Represents the enhanced pixel brightness value, and performs inverse γ enhancement on the pixel; A target instance segmentation network was constructed for key feature detection of aluminum coils. This network consists of two branches: a 2D branch based on the YOLOv8 model, which extracts a single weld point region on the inner wall of the coil from the 2D image. The 3D branch, based on the PointNet++ architecture, extracts the circular end faces from the 3D point cloud, including three weld points on the outer end face and one weld point on the inner end face. These regions are then classified into non-weldable and weldable regions. The input of the two-dimensional branch is the 2D image after brightness enhancement, and it outputs the feature map for predicting the box, the feature map for predicting the classification, and the mask coefficient for segmentation. At the same time, the output of the P5 layer in the Head is upsampled to obtain the mask prototype; the model output is filtered into two prediction boxes by NMS, and the mask prototype and mask coefficient are used to generate the target prediction mask. The prediction box is cropped and thresholded to obtain the segmentation result; The input of the 3D branch is a set of 3D coordinates. First, a clustering algorithm is used in the sampling layer to select representative points from the input point cloud to form a smaller subset. Then, for each sample point, a group of points is selected in its neighborhood in the grouping layer, and a spherical or K-nearest neighbor search is used to define the neighborhood. Finally, the feature extraction layer applies a small PointNet to the points in each neighborhood to extract local features. The local features are processed by MLP to generate a fixed-length feature vector. Through layer-by-layer aggregation, the local features are gradually aggregated to a higher level to obtain a global feature vector. The fully connected layer further processes and classifies the extracted global feature vector, and finally outputs a classification label to divide the 3D point cloud into non-weldable areas and weldable areas. The welding area is further divided within the weldable area, and accurate path planning is performed on the weld point area of ​​the inner wall and the welding area based on the actual size of the welding gun.

2. The intelligent and standardized weld position identification method for hot-rolled aluminum coil according to claim 1, characterized in that: A two-stage exposure method is used to collect 2D images and 3D point cloud data, with the first stage being long exposure and the second stage being short exposure.

3. The intelligent and standardized weld position identification method for hot-rolled aluminum coil according to claim 1, characterized in that: The noise point denoising is performed using a statistical method based on the z-axis. The interval is divided according to the length of the z-axis to obtain the interval length h: in, h is the length of the interval, l is the length of the z-axis interval occupied by the point cloud, is the maximum value of the z-axis data in the point cloud data, is the minimum value of the z-axis data in the point cloud data, and then record the number of points in each interval, using the interval with the most points as the reference: It is The set of all points in the interval, all points in the interval Satisfy the z-axis value In a size of The limits are set successively by extending to adjacent zero intervals: It is the lower limit of the z-axis size of the final selected valid point. The upper limit of the z-axis size of the final selected valid points, and the points outside these limits are regarded as noise and removed.

4. The intelligent and standardized weld position identification method for hot-rolled aluminum coil according to claim 1, characterized in that: For a welding point area of ​​the inner wall, the center line is the actual weld path, and the actual weld path is an arc.

5. The intelligent and standardized welding position identification method for hot-rolled aluminum coil according to claim 1, characterized in that: The welding area is further divided within the weldable area according to the thickness n of the plate. When n is less than a certain threshold, the welding area moves radially inward and downward by k1 mm from the boundary line of the weldable area. When n is greater than a certain threshold, the welding area moves radially inward and downward by k2 mm from the boundary line of the weldable area. The outer edge of the welding area is parallel to the outer edge of the weldable area. k1 and k2 are constants.

6. The intelligent and standardized welding position identification method for hot-rolled aluminum coil according to claim 5, characterized in that: The thickness of the plate and strip is obtained by transferring parameters during the rolling process of the aluminum coil, and the welding area is a rectangular arc.

7. An intelligent and standardized welding position recognition system for hot-rolled aluminum coils, characterized in that: include: A processor and a memory, the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute an intelligent and standardized welding position identification method for hot-rolled aluminum coils as described in any one of claims 1 to 6.

Images