Steel plate shearing optimization method and device
By acquiring and matching the global contour feature points of the steel plate during the segmented shearing and fixed-length shearing stages, the global contour before fixed-length shearing is reconstructed. Combined with order data, optimization calculations are performed, which solves the problems of low accuracy and material waste in fixed-length shearing and achieves efficient shearing optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CERI DIGITAL TECHNOLOGY (BEIJING) CO LTD
- Filing Date
- 2025-12-19
- Publication Date
- 2026-05-19
AI Technical Summary
In existing technologies, the fixed-length shear cannot obtain the global contour of the steel plate before shearing, which makes it impossible for the high-precision positioning system to optimize the shearing of sub-plates, resulting in low shearing accuracy and serious waste of plate material.
Before segmented shearing, the global contour points of the steel plate are obtained, feature points are extracted, and before fixed-length shearing, the head contour points are matched with the segmented shearing feature points to determine the affine parameters, the global contour points before fixed-length shearing are reconstructed, and shearing optimization calculations are performed in combination with order data to output the shearing optimization results.
This technology improves the accuracy and efficiency of fixed-length cutting and reduces material waste when the overall outline cannot be measured before fixed-length cutting.
Smart Images
Figure CN122066010A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of steel rolling technology, and in particular to a method and apparatus for optimizing steel plate shearing. Background Technology
[0002] Medium and heavy plate shearing optimization technology refers to the shearing optimization system combining the global contour data of medium and heavy plates sent by the contour analyzer and the order data issued by the secondary system to perform shearing optimization calculations. The system calculates the most suitable shearing positions such as the head, tail, segment, and sub-plate and sends them to the shears for shearing, so as to maximize the utilization rate of the plate, reduce waste, and improve the yield and shearing efficiency.
[0003] Medium and heavy plate shearing optimization includes segmented shearing optimization and fixed-length shearing optimization. Segmented shearing is suitable for step-by-step cutting of ultra-long and ultra-wide plates. It achieves the splitting of large-sized blanks through multi-blade segmentation operations and can flexibly adapt to irregularly shaped parts or non-standard size requirements. Fixed-length shearing is designed for standardized batch production scenarios. Relying on high-precision positioning systems such as profilometers and shearing optimization, as well as servo control technology, it can complete precise cutting of fixed lengths and widths in one go. Segmented shearing and fixed-length shearing are complementary processes that can jointly support the flexible and standardized production of medium and heavy plate processing. However, in the existing technology, because fixed-length shearing requires edge alignment before shearing, and the profilometer needs to be installed after the alignment device, the profilometer is too close to the fixed-length shear. At the start of fixed-length shearing, it is impossible to obtain the global profile of the steel plate. As a result, the high-precision positioning system and servo control technology cannot optimize the sub-plate shearing based on the global profile of the steel plate, resulting in low shearing accuracy and material waste. Summary of the Invention
[0004] This invention provides a steel plate shearing optimization method to improve shearing accuracy and reduce plate waste. The method includes:
[0005] The global contour points of the steel plate are obtained by a pre-segmentation contouring instrument; feature points are extracted based on the global contour points, and the feature points of the segmented shearing stage are output; the feature points reflect the local contour changes of the steel plate.
[0006] When the steel plate is transported to the pre-cutting profiler, the head profile points of the steel plate and order data collected by the pre-cutting profiler are obtained; feature points are extracted based on the head profile points of the steel plate, and the feature points of the head profile are output; the order data includes the requirements data for the cutting of the steel plate.
[0007] Feature points of the head contour are matched with feature points of the segmented cutting stage to determine affine parameters;
[0008] Determine the global contour points before length setting based on the affine parameters and the global contour points in the segmented cutting stage;
[0009] Based on the global contour points, affine parameters, and order data before length determination, shearing optimization calculations are performed, and shearing optimization results are output. The shearing optimization results include parameters for controlling the shearing of the head, sub-plate, and tail of the steel plate.
[0010] This invention also provides a steel plate shearing optimization device to improve shearing accuracy and reduce plate waste. The device includes:
[0011] The global contour acquisition and feature extraction module for the segmented shearing stage is used to acquire the global contour points of the steel plate for the segmented shearing stage through the pre-segmented contour instrument; extract feature points based on the global contour points, and output the feature points of the segmented shearing stage; the feature points reflect the local contour changes of the steel plate.
[0012] The pre-cutting head contour acquisition module is used to acquire the head contour points and order data of the steel plate collected by the pre-cutting contour instrument when the steel plate is transported to the pre-cutting contour instrument; extract feature points based on the head contour points of the steel plate and output the feature points of the head contour; the order data includes the requirements data for steel plate cutting.
[0013] The feature point matching module is used to match the feature points of the head contour with the feature points of the segmented cutting stage to determine the affine parameters.
[0014] The global contour reconstruction module before length setting is used to determine the global contour points before length setting based on affine parameters and global contour points in the segmented shearing stage.
[0015] The fixed-length shearing optimization module is used to perform shearing optimization calculations based on the global contour points, affine parameters, and order data before fixed-length cutting, and output the shearing optimization results; the shearing optimization results include parameters for controlling the shearing of the head, sub-plate, and tail of the steel plate.
[0016] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described steel plate shearing optimization method.
[0017] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described steel plate shearing optimization method.
[0018] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described steel plate shearing optimization method.
[0019] In this embodiment of the invention, during the segmented shearing stage, the global contour points of the steel plate are acquired and feature points are extracted. Before the steel plate is transported to the pre-length shearing contour instrument, the head contour points of the steel plate collected by the pre-length shearing contour instrument are acquired first, and feature points are extracted in the same way. After feature point matching, the affine parameters from segmented shearing to pre-length shearing can be estimated. Combined with the global contour points in the segmented shearing stage, the global contour points before pre-length shearing are calculated, and the shearing is further optimized. This realizes the global shearing optimization of pre-length shearing when the pre-length shearing contour instrument cannot measure the global contour, improves the shearing accuracy and efficiency in the pre-length shearing stage, and reduces plate waste. Attached Figure Description
[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:
[0021] Figure 1 This is a schematic flowchart of the steel plate shearing optimization method in an embodiment of the present invention;
[0022] Figure 2 This is a schematic diagram of the global contour feature points of the steel plate in an embodiment of the present invention;
[0023] Figure 3 This is a schematic diagram of the feature points of the head contour of the steel plate before length setting in an embodiment of the present invention;
[0024] Figure 4 This is a schematic diagram of the steel plate shearing optimization device in an embodiment of the present invention;
[0025] Figure 5 This is a schematic diagram of a computer device in an embodiment of the present invention. Detailed Implementation
[0026] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.
[0027] The acquisition, storage, use, and processing of data in this application all comply with the relevant provisions of national laws and regulations.
[0028] In existing technologies, only the cutting of the steel plate head can be optimized before fixed-length shearing. Subsequent cutting of the steel plate generally relies on manual processing, resulting in low cutting efficiency, high error rate, low accuracy, and a lot of waste of sheet metal.
[0029] This invention proposes a steel plate shearing optimization method. Its purpose is to achieve global shearing optimization for fixed-length shearing when the global contour cannot be measured by a contour meter before fixed-length shearing, by using a combination of segmented pre-shearing global contour reconstruction feature extraction and fixed-length shearing global contour reconstruction method.
[0030] Figure 1 This is a flowchart illustrating the steel plate shearing optimization method in an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:
[0031] Step 101: Obtain the global contour points of the steel plate in the segmented shearing stage using a pre-segmented contour instrument; extract feature points based on the global contour points and output the feature points of the segmented shearing stage; the feature points reflect the local contour changes of the steel plate.
[0032] Step 102: When the steel plate is transported to the pre-cutting profiler, acquire the steel plate head profile points and order data collected by the pre-cutting profiler; extract feature points based on the steel plate head profile points and output the feature points of the head profile; the order data includes the requirements data for steel plate cutting.
[0033] Step 103: Match the feature points of the head contour with the feature points of the segmented cutting stage to determine the affine parameters;
[0034] Step 104: Determine the global contour points before length setting based on the affine parameters and the global contour points in the segmented cutting stage;
[0035] Step 105: Perform shearing optimization calculations based on the global contour points, affine parameters, and order data before length determination, and output the shearing optimization results; the shearing optimization results include parameters for controlling the shearing of the head, sub-plate, and tail of the steel plate.
[0036] The steel plate shearing optimization method in the embodiments of the present invention is described in detail below.
[0037] In practice, the shearing optimization system first obtains the global contour points of the steel plate during the segmented shearing stage using a pre-segmented contour gauge.
[0038] Specifically, the shearing optimization system acquires the global contour data of the steel plate before segmented shearing. As shown in Table 1, Table 1 is an example table of global contour data, showing the data that can be obtained in step 101 besides the global contour points. The global contour data of the steel plate before segmented shearing may specifically include the steel plate ID, the length of the head defect, the length of the tail defect, and the coordinates of the global contour points of the steel plate.
[0039] Table 1
[0040]
[0041] In the embodiments, the global contour points include global contour point coordinate data. The coordinate points of the steel plate contour, which are composed of X and Y coordinate pairs, are detected by a profiler (usually calculated by stitching scans of a vision system). As Figure 2 shown, Figure 2 This is a schematic diagram of the global contour feature points of the steel plate in an embodiment of the present invention, showing a simple example of global contour points.
[0042] After that, feature points are extracted. At this time, the output feature points are recorded as the feature points in the segmentation shearing stage. The feature points can also be called key points or feature key points. The feature points reflect the local contour changes of the steel plate, such as the key nodes of shape changes, the corners, inflection points, and extreme points of the contour line, etc.
[0043] Subsequently, before the steel plate is transported to the profiling仪 before the fixed-length shear, the head contour points of the steel plate are obtained through the profiling仪 before the fixed-length shear, and feature points are extracted based on the head contour points of the steel plate, which are recorded as the feature points of the head contour before the fixed-length shear.
[0044] In one embodiment, the extraction of feature points includes:
[0045] Extract feature points by one of the curvature extreme value method, shape context method, and chain code difference method.
[0046] In the embodiments of the present invention, a simplified method example for extracting feature points is given, which can not only improve the processing speed but also effectively extract feature points. The extraction of feature points may include:
[0047] Slide along all contour points with a fixed window size m, sequentially select consecutive m points, calculate the extreme points in each window until all contour points are traversed; among them, calculating the extreme points in each window includes: connecting the first and last points of the m points to form a straight line, calculating the distances from the remaining m - 2 points to this straight line, and marking the point with the maximum distance as the extreme point;
[0048] The points marked as extreme points more than a preset number of times are used as feature points.
[0049] For example, select any point on the contour (global contour point or head contour point of the steel plate) as the starting point, select consecutive m points (m is a positive integer, for example, select consecutive m points clockwise), connect the first and last points of the m points to form a straight line, calculate the distances from the remaining m - 2 points to this straight line, select the point with a distance > 0 and the maximum distance as the extreme point, then increase the starting point index by z steps, z < m, z is a positive integer, select m points again, and repeat the above operations until all points on the contour are traversed. During the traversal, if an extreme point is selected more than the preset number of times (the preset number is a positive integer), then this extreme point is considered a feature key point.
[0050] In the feature point extraction of the head contour before fixed length, since the steel plate rotates during transportation, there will be a deflection angle relative to the global contour data.
[0051] In one embodiment, before extracting feature points based on the global contour points, the method may further include:
[0052] The global contour point coordinate data is filtered; the filtering includes curvature filtering and / or statistical filtering.
[0053] In the embodiments, before extracting key feature points, the contour coordinate data can be filtered to remove noise. Common filtering methods such as three-point cocircle detection, curvature filtering, and statistical filtering can be used.
[0054] In one embodiment, segmentation data of the segmented shearing can also be obtained during the stage when the steel plate undergoes segmented shearing.
[0055] For example, after obtaining the global contour points of the steel plate during the segmented shearing stage using a pre-segmentation contouring instrument, the process also includes:
[0056] Obtain the segmentation data of the segmented shear; the segmentation data includes the number of segments and the segmentation position of the steel plate; the number of segments is either zero or non-zero. If no segments are used, the number of segments and the segmentation position are both zero.
[0057] In step 102, when the steel plate is transported to the pre-cutting profiler, the head profile points of the steel plate and order data collected by the pre-cutting profiler are obtained. The order data includes the requirements for cutting the steel plate.
[0058] During implementation, the steel plate is transported to the pre-cutting profiler. The shearing optimization system obtains the head profile data of the steel plate measured by the pre-cutting profiler and the order data issued by the secondary system L2.
[0059] The steel plate head profile number includes at least the coordinates of the steel plate head profile points measured by the profiler before the sizing cutter when the steel plate reaches the sizing cutter and the steel plate Id.
[0060] Order data must include at least the cut-to-length plan data, such as the number of sub-boards, the minimum length of the sub-boards, the maximum length of the sub-boards, and the width of the sub-boards, as shown in Table 2. Table 2 is an example table of order data.
[0061] Table 2
[0062]
[0063] In step 103, the feature points of the head contour before the fixed length are matched with the feature points of the segmented cutting stage to determine the affine parameters.
[0064] For example, by automatically learning feature representations and matching rules through convolutional neural networks, feature points of the head contour before the sizing stage are matched with feature points of the segmented cutting stage to determine affine parameters.
[0065] To improve matching efficiency, in one embodiment, feature point matching is performed between the feature points of the head contour and the feature points of the segmented slicing stage to determine affine parameters, which may include:
[0066] Based on the starting and ending points of the steel plate head contour points collected by the pre-cutting contour instrument, determine the head contour length collected by the pre-cutting contour instrument.
[0067] Based on the head contour length, feature points located within the head contour length are selected from the feature points in the segmented cutting stage.
[0068] After datafication of feature points within the length of the head contour, the feature points of each head contour are matched one by one with the feature points of the corresponding segmented pruning stage based on the nearest neighbor matching method to obtain matching feature point pairs.
[0069] Affine parameters are determined based on matching feature point pairs.
[0070] Determining affine parameters based on matching feature point pairs can include using one of the following methods: singular value decomposition algorithm, random sample consensus algorithm, or neural network method.
[0071] refer to Figure 2 The global contour feature points are (A1, A2, A3...A6). Figure 3 This is a schematic diagram of the feature points of the steel plate head contour before length setting in an embodiment of the present invention. The key points of the head contour before length setting are (B1, B2, B3...). Figure 2 neutralization Figure 3 Taking the midpoint as an example, feature point matching is performed.
[0072] The matching process is as follows:
[0073] ① Determine the key feature points A1, A2, and A3 of the global contour located on the head before segmentation based on the contour length of the head contour data before the fixed length.
[0074] ② Data is digitized to eliminate coordinate differences. Calculate CA=(A1+A2+……) / n, CB=(B1+B2+……) / n, Cn=An-CA, Dn=Bn-CB.
[0075] ③ Calculate the distance from C1 to Dn respectively, and select Dn with the smallest distance as the feature matching point of C1. Calculate the feature matching points of C2 to Cn in the same way.
[0076] ④ Perform affine parameter estimation based on the matched feature point pairs, including scaling factor S, rotation radians R, and translation T. Use the SVD (Singular Value Decomposition) algorithm, or RANSAC (Random Sample Consensus Algorithm) + SVD, neural network method, etc.
[0077] In step 104, the global contour points before length setting are determined based on the affine parameters and the global contour points in the segmented cutting stage.
[0078] For example, scaling and rotating the global contour before segmentation using affine parameters yields the global contour before scaling. If the global contour coordinates P(X, Y), then the corresponding coordinates Q(X, Y) of the transformed global contour before scaling are as follows:
[0079] Q(X)=(P(X)×cos(R)-P(Y)×sin(R))×S;
[0080] Q(Y)=(P(X)×sin(R)+P(Y)×cos(R))×S.
[0081] Considering whether to segment, based on the affine parameters and the global contour points during the segmentation and shearing stage, the global contour points before the final size are determined, including:
[0082] When the number of segments is non-zero, the global contour points before length setting are determined based on the segment position, the global contour points during the segment cutting stage, and the affine parameters.
[0083] For example, if the plate is segmented, the global profile of the entire steel plate is reconstructed based on the first segment, and then the global profile of each segment is determined sequentially based on the position of the segment and the reconstructed global profile.
[0084] In step 105, shearing optimization calculations are performed based on the global contour points before length determination, affine parameters, and the order data, and the shearing optimization results are output. The shearing optimization results include parameters controlling the shearing of the head, sub-plate, and tail of the steel plate, such as the shearing lengths of the head, tail, and sub-plate of the steel plate.
[0085] After outputting the shearing optimization results, the method may further include:
[0086] The shearing optimization results are sent to the fixed-length shearing system so that the fixed-length shearing system can perform fixed-length shearing based on the shearing optimization results.
[0087] In one embodiment, the order data includes the required subboard width, which may be referred to as the subboard order width, such as the subboard width in row 5 of Table 2;
[0088] Based on the global contour points before length determination, affine parameters, and the order data, a shearing optimization calculation is performed, and the shearing optimization result is output, which may include:
[0089] Obtain the length of the head and tail defects before segmented shearing, and determine the length of the head and tail defects before fixed length based on the length of the head and tail defects before segmented shearing and the scaling factor; the length of the head and tail defects includes the length of the steel plate with defects at the head and tail, and the unusable part with defects.
[0090] Based on the global contour points before length determination and the required subplate width in the order data, the head and tail lengths are optimized to obtain the head optimized length and tail optimized length. The head optimized length and tail optimized length represent the parts of the steel plate that are not used for shearing to form subplates, that is, the parts that cannot be used to form subplates after head and tail length optimization calculation.
[0091] The placement position of the sub-board is determined based on the optimized lengths of the head and tail and the global contour points before sizing, and the optimized length of the tail is updated.
[0092] In one embodiment, the head and tail defect lengths include the head defect length and the tail defect length;
[0093] Based on the length of the defects at the beginning and end before the specified length and the width of the sub-board required in the order data, the lengths at the beginning and end are optimized to obtain the optimized lengths at the beginning and end, including:
[0094] Starting from the head defect length or tail defect length on the steel plate, move it towards the plate body along the transmission direction and calculate point by point whether the width of the steel plate is greater than the sum of the required sub-plate width and the preset width margin. When the width of the steel plate is greater than the sum of the required sub-plate width and the preset width margin, stop moving to obtain the head optimized length and tail optimized length.
[0095] During implementation, shearing optimization calculations are performed based on the reconstructed global contour data before sizing (including global contour points before sizing) and the order data issued by the secondary system.
[0096] ① Coordinate transformation. Set the starting point of the global contour points before scaling to zero, and transform the coordinates of the other points together according to the offset of the starting point.
[0097] For example, set the leftmost X-coordinate of the reconstructed global contour to 0, and then shift all X-coordinates sequentially by offset. For instance, if the leftmost coordinate is -dmm, then all X-coordinates will be shifted to the right by dmm, where d is any real number.
[0098] ② Update the length of defects at the beginning and end of the steel plate before cutting to length.
[0099] Since the length change caused by the deflection angle is small and negligible, only the length change caused by scaling is considered. The length of the defects at the beginning and end of the fixed length = the length of the defects at the beginning and end of the segment × the scaling factor. That is, the length of the defects at the beginning and end of the fixed length is determined by multiplying the length of the defects at the beginning and end of the segment by the scaling factor.
[0100] ③ Optimize the head and tail lengths based on the updated head and tail defect lengths and the subboard width required by the order.
[0101] Starting from the length of the head defect, move towards the board body along the X-axis and calculate the width at each point to see if it is greater than the sub-board order width plus the width margin (the width margin can be set manually, such as 10mm). When the above width requirement is met, stop moving. At this time, the X coordinate of that point is the head length. Similarly, starting from the length of the tail defect, move towards the board body along the X-axis and calculate the width at each point to see if it is greater than the sub-board order width plus the width margin (the width margin can be set manually, such as 10mm). When the above width requirement is met, stop moving.
[0102] ④ Calculate the placement position of the sub-board based on the optimized length of the head and tail.
[0103] Pre-set the sub-board placement angle range, and determine the placement position of each sub-board based on the optimized length at the beginning and end, the global contour points before the fixed length, and the order data.
[0104] For example, starting from the top edge coordinates of the header, begin traversing and placing all sub-boards. Traverse the coordinates from top to bottom and left to right, with sub-board lengths ranging from the maximum to the minimum length required by the order, and placement angles from - Degree to Degree (Pre-set subboard placement angle) Degree to The process continues until a suitable solution is found. If no solution is found, the number of sub-boards is reduced, the calculation is repeated, and the process is recursively called until all sub-boards have been processed.
[0105] ⑤ Update the tail length data.
[0106] If the optimized board length is Length, the optimized head length is HL, and the sub-board length is SubLength, then the optimized tail length = Length - HL - SubLength × k, where k is the number of sub-boards.
[0107] Alternatively, other common optimization algorithms can be used for optimization calculations, such as traversal methods and support vector machine methods.
[0108] Finally, the fixed-length cutting system performs fixed-length cutting based on the optimization calculation results.
[0109] In summary, the embodiments of the present invention achieve global shearing optimization for fixed-length shearing by using a method of global contour reconstruction feature extraction before segmented shearing and global contour reconstruction before fixed-length shearing, when the global contour cannot be measured by the contour meter before fixed-length shearing.
[0110] After the steel plate passes through the pre-segment shearing profiler, the shearing optimization system extracts the key feature points for global contour reconstruction based on the global contour data sent by the pre-segment shearing profiler. When the steel plate passes through the pre-sizing shearing profiler, it first measures the head contour data, extracts the key feature points of the head contour, matches them with the global contour key feature points before segmentation, then estimates the affine parameters, reconstructs the global contour before the sizing shearing based on the global contour before segmentation and the affine parameters, and performs global shearing optimization calculation in combination with the order data sent by the secondary system. The sizing shearing performs shearing according to the optimization result.
[0111] A specific embodiment is given below.
[0112] 1) The shearing optimization system obtains the global contour data of the steel plate before the segment shearing.
[0113] The global contour data at least includes the steel plate Id, the head defect length, the tail defect length, and the global contour coordinate points of the steel plate, as shown in Table 1.
[0114] 2) Extract the global contour key feature points (A1, A2, A3......).
[0115] Select any point on the contour as the starting point, select m consecutive points, connect the first and last points of the m points to form a straight line, calculate the distances from the remaining m - 2 points to this straight line, select the points with distance > 0 and the maximum distance as the extreme points, then increase the starting point index by z steps. When p < m, select m points again and repeat the above operations until all points on the contour are traversed. If an extreme point is selected more than the preset number of times during the traversal process, then this extreme point is considered as a key feature point.
[0116] Referring to Table 3, let m = 10, p = 1, and the preset number = 2. Calculate the extreme points of 10 consecutive points starting from P1 to P3 in turn. The coordinates of each point are shown in Table 3 below. Taking P1 as the starting point, calculate the straight line formed by P1 and P10, with y = 0.55x + 714.59. The distances from the intermediate points P2 to P9 to this straight line are shown in the "Starting from P1" column. The calculations starting from P2 and P3 are the same as that starting from P1. It can be seen from each column that P7 is an extreme point, and P7 has appeared 3 times, which is greater than 2. Therefore, P7 can be regarded as a key feature point, represented by A1(700.00, 1650.00). In this example, 6 key feature points can be found, which are A1(700.00, 1650.00), A2(3.10, 576.70), A3(700.00, -650.00), A4(24900.00, -650.00), A5(25596.90, 576.70), A6(24900.00, 1650.00), as Figure 2 shown.
[0117] Table 3
[0118]
[0119] In this embodiment of the invention, considering that extreme points are less affected by small changes, and that the steel plate needs to undergo cooling deformation and rotation before reaching the fixed-length shearing, extreme points are required for positioning. Therefore, the above-mentioned method for quickly selecting feature points is proposed.
[0120] Optionally, key points of contour features can be extracted, and commonly used curve feature point selection methods such as curvature extremum method, shape context method, and chain code difference method can also be used.
[0121] Optionally, before extracting key feature points, the contour coordinate data can be filtered to remove noise. Common filtering methods such as three-point concyclic detection, curvature filtering, and statistical filtering can be used.
[0122] 3) The shearing optimization system obtains the segmented data of the segmented shearing.
[0123] Segmented data must include at least the segment position; if no segments are defined, the segment position is 0.
[0124] In this example, there are no segments; the segment position is 0.
[0125] 4) The steel plate is transported to the pre-cutting profiler. The shearing optimization system obtains the head profile data of the steel plate measured by the pre-cutting profiler and the order data issued by the secondary system.
[0126] The steel plate head profile number includes at least the steel plate head profile coordinates and steel plate Id measured by the profiler before the sizing cutter when the steel plate reaches the sizing cutter.
[0127] The order data must include at least the cut-to-length plan data, such as the number of sub-boards, the minimum length of the sub-boards, the maximum length of the sub-boards, and the width of the sub-boards, as shown in Table 2.
[0128] 5) Extract the key points of the head contour (B1, B2, B3...), using the same method as step 2). The key points of the head contour of the steel plate are as follows: Figure 3 As shown, because the steel plate rotates during transportation, there will be a deflection angle relative to the global contour data. .
[0129] In this example, the key feature points of the head are B1 (624.80, 1353.90), B2 (21.60, 242.70), and B3 (803.05, -911.50).
[0130] 6) Perform feature point matching based on the global contour feature points (A1, A2, A3...) before segmentation and the key feature points of the head contour before length setting (B1, B2, B3...).
[0131] Assuming global contour feature points (A1~A6) before segmentation and key head contour feature points (B1~B3) before scaling, such as Figure 2 As shown in Figure 3, the matching process is as follows:
[0132] ① Determine the key feature points A1, A2, and A3 of the global contour located on the head before segmentation based on the contour length of the head contour data before the fixed length.
[0133] ② Calculate CA = (A1 + A2 + A3) / 3, CB = (B1 + B2 + B3) / 3, Cn = An - CA, Dn = Bn - CB, n = 1~3. The calculation results are shown in Table 4 below. Table 4 shows the calculation results.
[0134] Table 4
[0135]
[0136] Calculate the distance from C1 to Dn respectively, and select Dn with the smallest distance as the feature matching point of C1. Calculate the feature matching points of C2 and C3 in the same way.
[0137] In this example, the distance matrix is shown in Table 5 below. Table 5 shows the distance matrix results, and the feature matching points corresponding to C1~C3 are D1~D3 respectively.
[0138] Table 5
[0139]
[0140] 7) Perform affine parameter estimation based on the matched feature points, including scaling factor S, rotation radians R, and translation T, using the singular value decomposition algorithm.
[0141] The two point sets selected here are A1, A2, A3 and B1, B2, B3, respectively. The calculated rotation radians R = 0.0785275, the equivalent angle is 4.499 degrees, and the scaling factor S = 0.988.
[0142] Alternatively, affine parameter estimation can also employ RANSAC+SVD, neural network methods, etc.
[0143] 8) Reconstruct the global profile before length setting based on the affine parameters and the global profile before segmentation.
[0144] The global contour before segmentation is scaled and rotated according to affine parameters to obtain the global contour before scaling. If the global contour coordinate point P(X, Y), the corresponding coordinate point Q(X, Y) of the transformed global contour before scaling is as follows:
[0145] Q(X)=(P(X)×cos(R)-P(Y)×sin(R))×S;
[0146] Q(Y)=(P(X)×sin(R)+P(Y)×cos(R))×S;
[0147] For example: P(X) = 28.00, P(Y) = 730.00, then:
[0148] Q(X)=(28.00×cos(0.0785275)-730.00×sin(0.0785275))×0.988=-29.00;
[0149] Q(Y)=(28.00×sin(0.0785275)+730.00×cos(0.0785275))×0.988=721.19;
[0150] If the plate is segmented, the global outline of the entire steel plate is reconstructed based on the first segment. Then, the global outline of each segment is determined sequentially based on the position of the segment and the reconstructed global outline. In this example, the plate is not segmented.
[0151] 9) Perform shearing optimization calculations based on the reconstructed global contour data before sizing and the order data issued at the secondary level.
[0152] The optimized calculation is as follows:
[0153] ① Coordinate transformation: Set the leftmost X coordinate of the reconstructed global contour to 0, and shift all X coordinates sequentially according to the offset. For example, if the leftmost coordinate is -70mm, then shift all X coordinates to the right by 70mm.
[0154] ② Before cutting to length, the length of the steel plate at the head and tail is updated. Since the length change caused by the deflection angle is small, it can be ignored. Only the length change caused by scaling is considered. The length of the head and tail defects before cutting to length = the length of the head and tail defects before segmentation × scaling factor. Therefore, in this example, the updated head defect length = 542 × 0.988 = 535 mm, and the updated tail defect length = 621 × 0.988 = 614 mm.
[0155] ③ Optimize the head and tail lengths based on the updated head and tail defect lengths and the sub-board width required by the order. Starting from the head defect length of 535mm, move towards the board body along the X-axis and calculate the width at each point to see if it is greater than the sub-board order width + width margin (the width margin can be set manually, such as 10mm). When the above width requirement is met, stop moving. At this time, the X coordinate of that point is the head length. In this example, the optimized head length is 595mm. Similarly, starting from the tail defect length of 614mm, move towards the board body along the X-axis and calculate the width at each point to see if it is greater than the sub-board order width + width margin (the width margin can be set manually, such as 10mm). When the above width requirement is met, stop moving. In this example, the width requirement is met starting from defect length 614mm, so the optimized tail length is 614mm.
[0156] ④ Calculate the placement position of the sub-boards based on the optimized lengths of the head and tail. Starting from the coordinates of the top edge of the head, begin traversing and placing all sub-boards. The order requires two sub-boards, with a minimum length of 10000×2=20000mm, a maximum length of 12000×2=24000mm, and a width of 2200mm. Traverse the coordinates from top to bottom and from left to right, with the sub-board length decreasing from the maximum to the minimum length and the placement angle decreasing from -5 degrees to 5 degrees, until a suitable solution is found. If no solution is found, reduce the number of sub-boards and recalculate. In this example, two sub-boards can be placed, starting at coordinates (595, 1632) with a placement angle of 0 degrees, requiring no segmentation.
[0157] ⑤ Update the tail length data: board length = 25598mm, head length = 595mm, sub-board length = (25598 - 595 - 614) / 2 = 12195mm. Take the maximum value of 12000mm as the sub-board length. Therefore, the optimized tail length = 25598 - 595 - 12000 × 2 = 1003mm. The optimized result is: head 595mm + sub-board 11100mm × 2 + tail 1003mm.
[0158] Alternatively, other common optimization algorithms can be used for optimization calculations, such as traversal methods and support vector machine methods.
[0159] 10) Fixed-length cutting: Fixed-length cutting is performed based on the optimized calculation results.
[0160] This invention also provides a steel plate shearing optimization device, as described in the following embodiments. Since the principle by which this device solves the problem is similar to that of the steel plate shearing optimization method, the implementation of this device can be referred to the implementation of the steel plate shearing optimization method, and repeated details will not be elaborated further.
[0161] Figure 4 This is a schematic diagram of the steel plate shearing optimization device in an embodiment of the present invention, as shown below. Figure 4 As shown, the device 400 may include:
[0162] The global contour acquisition and feature extraction module 401 for the segmented shearing stage is used to acquire global contour points of the steel plate for the segmented shearing stage through a pre-segmented contour instrument; extract feature points based on the global contour points and output the feature points of the segmented shearing stage; the feature points reflect the local contour changes of the steel plate.
[0163] The head contour acquisition module 402 before fixed-length shearing is used to acquire the head contour points of the steel plate and order data collected by the pre-fixed-length shearing contour instrument when the steel plate is transported to the pre-fixed-length shearing contour instrument; extract feature points based on the head contour points of the steel plate and output the feature points of the head contour; the order data includes the requirements data for shearing the steel plate.
[0164] The feature point matching module 403 is used to match the feature points of the head contour with the feature points of the segmented cutting stage to determine the affine parameters.
[0165] The global contour reconstruction module 404 before length setting is used to determine the global contour points before length setting based on the affine parameters and the global contour points in the segmented shearing stage.
[0166] The fixed-length shearing optimization module 405 is used to perform shearing optimization calculations based on the global contour points, affine parameters, and order data before fixed-length cutting, and output the shearing optimization results; the shearing optimization results include parameters for controlling the shearing of the head, sub-plate, and tail of the steel plate.
[0167] In one embodiment, the extraction of feature points includes:
[0168] Slide along all contour points with a fixed window size m, select m consecutive points in sequence, calculate the extreme point within each window, until all contour points have been traversed; wherein, calculating the extreme point within each window includes: connecting the first and last two points among the m points to form a straight line, calculating the distance from the remaining m-2 points to the straight line, and marking the point with the largest distance as the extreme point;
[0169] Points that are marked as extreme points more than a preset number of times are used as feature points.
[0170] In one embodiment, the feature point matching module 403 is specifically used for:
[0171] Based on the starting and ending points of the steel plate head contour points collected by the pre-cutting contour instrument, determine the head contour length collected by the pre-cutting contour instrument.
[0172] Based on the head contour length, feature points located within the head contour length are selected from the feature points in the segmented cutting stage.
[0173] After datafication of feature points within the length of the head contour, the feature points of each head contour are matched one by one with the feature points of the corresponding segmented pruning stage based on the nearest neighbor matching method to obtain matching feature point pairs.
[0174] Affine parameters are determined based on matching feature point pairs.
[0175] In one embodiment, the feature point matching module 403 is specifically used for:
[0176] One of the following methods—singular value decomposition, random sample consensus, or neural network—is used to determine the affine parameters based on matching feature point pairs.
[0177] In one embodiment, the affine parameters include a scaling factor, rotation radians, and translation amount; the order data includes the required sub-board width;
[0178] The fixed-length shearing optimization module 405 is specifically used for:
[0179] Obtain the lengths of the head and tail defects before segmented shearing, and determine the lengths of the head and tail defects before fixed length based on the lengths of the head and tail defects before segmented shearing and the scaling factor; the lengths of the head and tail defects include the lengths of the steel plate with defects at the head and tail.
[0180] Based on the global contour points before length determination and the required subplate width in the order data, the head and tail lengths are optimized to obtain the head optimized length and tail optimized length; the head optimized length and tail optimized length represent the parts of the steel plate that are not used for shearing to form the subplate, respectively;
[0181] The placement position of the sub-board is determined based on the optimized lengths of the head and tail and the global contour points before sizing, and the optimized length of the tail is updated.
[0182] In one embodiment, the head and tail defect lengths include the head defect length and the tail defect length;
[0183] The fixed-length shearing optimization module 405 is specifically used for:
[0184] Starting from the head defect length or tail defect length on the steel plate, move it towards the plate body along the transmission direction and calculate point by point whether the width of the steel plate is greater than the sum of the required sub-plate width and the preset width margin. When the width of the steel plate is greater than the sum of the required sub-plate width and the preset width margin, stop moving to obtain the head optimized length and tail optimized length.
[0185] In one embodiment, the global contour points include global contour point coordinate data;
[0186] The global contour acquisition and feature extraction module 401 in the segmented cutting stage is also used to: filter the global contour point coordinate data; the filtering includes curvature filtering and / or statistical filtering.
[0187] In one embodiment, the shearing optimization result includes the shearing lengths of the steel plate head, tail, and sub-plate;
[0188] The device 400 may also include:
[0189] The shearing optimization result distribution module is used to distribute the shearing optimization result to the fixed-length shearing system after the fixed-length shearing optimization module 405 outputs the shearing optimization result, so that the fixed-length shearing system can perform fixed-length shearing according to the shearing optimization result.
[0190] Figure 5 This is a schematic diagram of a computer device in an embodiment of the present invention, such as... Figure 5As shown, this embodiment of the invention also provides a computer device 500, including a processor 501, a memory 502, and a computer program 503 stored in the memory 502 and executable on the processor 501. When the processor 501 executes the computer program 503, it implements the above-mentioned steel plate shearing optimization method.
[0191] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described steel plate shearing optimization method.
[0192] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described steel plate shearing optimization method.
[0193] This invention proposes a method for segmented pre-cutting global contour reconstruction feature extraction + fixed-length pre-cutting global contour reconstruction, ensuring high precision and quality in cutting optimization.
[0194] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0195] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0196] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1The function specified in one or more boxes.
[0197] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0198] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for optimizing steel plate shearing, characterized in that, include: The global contour points of the steel plate are obtained by a pre-segmentation contouring instrument; feature points are extracted based on the global contour points, and the feature points of the segmented shearing stage are output; the feature points reflect the local contour changes of the steel plate. When the steel plate is transported to the pre-cutting profiler, the profile points of the steel plate head and order data collected by the pre-cutting profiler are obtained. Feature points are extracted based on the head contour points of the steel plate, and the feature points of the head contour are output; the order data includes the requirements data for steel plate shearing. Feature points of the head contour are matched with feature points of the segmented cutting stage to determine affine parameters; Determine the global contour points before length setting based on the affine parameters and the global contour points in the segmented cutting stage; Based on the global contour points, affine parameters, and order data before length determination, shearing optimization calculations are performed, and shearing optimization results are output. The shearing optimization results include parameters for controlling the shearing of the head, sub-plate, and tail of the steel plate.
2. The method as described in claim 1, characterized in that, The extracted feature points include: Slide along all contour points with a fixed window size m, select m consecutive points in sequence, calculate the extreme point within each window, until all contour points have been traversed; wherein, calculating the extreme point within each window includes: connecting the first and last two points among the m points to form a straight line, calculating the distance from the remaining m-2 points to the straight line, and marking the point with the largest distance as the extreme point; Points that are marked as extreme points more than a preset number of times are used as feature points.
3. The method as described in claim 1, characterized in that, Feature point matching is performed between the feature points of the head contour and the feature points of the segmented slicing stage to determine the affine parameters, including: Based on the starting and ending points of the steel plate head contour points collected by the pre-cutting contour instrument, determine the head contour length collected by the pre-cutting contour instrument. Based on the head contour length, feature points located within the head contour length are selected from the feature points in the segmented cutting stage. After datafication of feature points within the length of the head contour, the feature points of each head contour are matched one by one with the feature points of the corresponding segmented pruning stage based on the nearest neighbor matching method to obtain matching feature point pairs. Affine parameters are determined based on matching feature point pairs.
4. The method as described in claim 3, characterized in that, Affine parameters are determined based on matching feature point pairs, including: One of the following methods—singular value decomposition, random sample consensus, or neural network—is used to determine the affine parameters based on matching feature point pairs.
5. The method as described in claim 1, characterized in that, The affine parameters include scaling factor, rotation radians, and translation amount; the order data includes the required sub-board width; Based on the global contour points before length determination, affine parameters, and the order data, a shearing optimization calculation is performed, and the shearing optimization results are output, including: Obtain the lengths of the beginning and end defects before segmented shearing, and determine the lengths of the beginning and end defects before fixed length based on the lengths of the beginning and end defects before segmented shearing and the scaling factor. The head and tail defect lengths include the lengths of the steel plate where defects exist at the head and tail. Based on the global contour points before length determination and the required subplate width in the order data, the head and tail lengths are optimized to obtain the head optimized length and tail optimized length; the head optimized length and tail optimized length represent the parts of the steel plate that are not used for shearing to form the subplate, respectively; The placement position of the sub-board is determined based on the optimized lengths of the head and tail and the global contour points before sizing, and the optimized length of the tail is updated.
6. The method as described in claim 5, characterized in that, The head and tail defect lengths include the head defect length and the tail defect length. Based on the length of the defects at the beginning and end before the specified length and the width of the sub-board required in the order data, the lengths at the beginning and end are optimized to obtain the optimized lengths at the beginning and end, including: Starting from the head defect length or tail defect length on the steel plate, move it towards the plate body along the transmission direction and calculate point by point whether the width of the steel plate is greater than the sum of the required sub-plate width and the preset width margin. When the width of the steel plate is greater than the sum of the required sub-plate width and the preset width margin, stop moving to obtain the head optimized length and tail optimized length.
7. The method as described in claim 1, characterized in that, The global contour points include global contour point coordinate data; Before extracting feature points based on the global contour points, the method further includes: Filter the global contour point coordinate data; The filtering includes curvature filtering and / or statistical filtering.
8. The method as described in claim 1, characterized in that, The shearing optimization results include the shearing lengths of the steel plate head, tail, and sub-plate; After outputting the shearing optimization results, the following are also included: The shearing optimization results are sent to the fixed-length shearing system so that the fixed-length shearing system can perform fixed-length shearing based on the shearing optimization results.
9. A steel plate shearing optimization device, characterized in that, include: The global contour acquisition and feature extraction module for the segmented shearing stage is used to acquire the global contour points of the steel plate for the segmented shearing stage through the pre-segmented contour instrument; extract feature points based on the global contour points, and output the feature points of the segmented shearing stage; the feature points reflect the local contour changes of the steel plate. The pre-cutting head contour acquisition module is used to acquire the steel plate head contour points and order data collected by the pre-cutting head contour instrument when the steel plate is transported to the pre-cutting head contour instrument. Feature points are extracted based on the head contour points of the steel plate, and the feature points of the head contour are output; the order data includes the requirements data for steel plate shearing. The feature point matching module is used to match the feature points of the head contour with the feature points of the segmented cutting stage to determine the affine parameters. The global contour reconstruction module before length setting is used to determine the global contour points before length setting based on affine parameters and global contour points in the segmented shearing stage. The fixed-length shearing optimization module is used to perform shearing optimization calculations based on the global contour points, affine parameters, and order data before fixed-length cutting, and output the shearing optimization results; the shearing optimization results include parameters for controlling the shearing of the head, sub-plate, and tail of the steel plate.
10. The apparatus as claimed in claim 9, characterized in that, The extracted feature points include: Slide along all contour points with a fixed window size m, select m consecutive points in sequence, calculate the extreme point within each window, until all contour points have been traversed; wherein, calculating the extreme point within each window includes: connecting the first and last two points among the m points to form a straight line, calculating the distance from the remaining m-2 points to the straight line, and marking the point with the largest distance as the extreme point; Points that are marked as extreme points more than a preset number of times are used as feature points.
11. The apparatus as claimed in claim 9, characterized in that, The feature point matching module is specifically used for: Based on the starting and ending points of the steel plate head contour points collected by the pre-cutting contour instrument, determine the head contour length collected by the pre-cutting contour instrument. Based on the head contour length, feature points located within the head contour length are selected from the feature points in the segmented cutting stage. After datafication of feature points within the length of the head contour, the feature points of each head contour are matched one by one with the feature points of the corresponding segmented pruning stage based on the nearest neighbor matching method to obtain matching feature point pairs. Affine parameters are determined based on matching feature point pairs.
12. The apparatus as claimed in claim 11, characterized in that, The feature point matching module is specifically used for: One of the following methods—singular value decomposition, random sample consensus, or neural network—is used to determine the affine parameters based on matching feature point pairs.
13. The apparatus as claimed in claim 9, characterized in that, The affine parameters include scaling factor, rotation radians, and translation amount; the order data includes the required sub-board width; The fixed-length shearing optimization module is specifically used for: Obtain the lengths of the head and tail defects before segmented shearing, and determine the lengths of the head and tail defects before fixed length based on the lengths of the head and tail defects before segmented shearing and the scaling factor; the lengths of the head and tail defects include the lengths of the steel plate with defects at the head and tail. Based on the global contour points before length determination and the required subplate width in the order data, the head and tail lengths are optimized to obtain the head optimized length and tail optimized length; the head optimized length and tail optimized length represent the parts of the steel plate that are not used for shearing to form the subplate, respectively; The placement position of the sub-board is determined based on the optimized lengths of the head and tail and the global contour points before sizing, and the optimized length of the tail is updated.
14. The apparatus as claimed in claim 13, characterized in that, The head and tail defect lengths include the head defect length and the tail defect length. The fixed-length shearing optimization module is specifically used for: Starting from the head defect length or tail defect length on the steel plate, move it towards the plate body along the transmission direction and calculate point by point whether the width of the steel plate is greater than the sum of the required sub-plate width and the preset width margin. When the width of the steel plate is greater than the sum of the required sub-plate width and the preset width margin, stop moving to obtain the head optimized length and tail optimized length.
15. The apparatus as claimed in claim 9, characterized in that, The global contour points include global contour point coordinate data; The global contour acquisition and feature extraction module in the segmented cutting stage is also used to: filter the global contour point coordinate data; the filtering includes curvature filtering and / or statistical filtering.
16. The apparatus as claimed in claim 9, characterized in that, The shearing optimization results include the shearing lengths of the steel plate head, tail, and sub-plate; Also includes: The shearing optimization result distribution module is used to distribute the shearing optimization results to the fixed-length shearing system after the fixed-length shearing optimization module outputs the shearing optimization results, so that the fixed-length shearing system can perform fixed-length shearing based on the shearing optimization results.
17. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 8.
18. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method of any one of claims 1 to 8.
19. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the method of any one of claims 1 to 8.