A method and system for arranging irregular workpieces for surface scratched steel plates

Abstract

The present invention discloses an irregular workpiece arrangement method and system for surface scratched steel plates, comprising: using a scratch recognition method to detect scratches on the surface of a steel plate to obtain a scratch area; based on the non-scratched area on the surface of the steel plate, using a dynamic division method to divide the non-scratched area to obtain a plurality of area units; using an irregular workpiece arrangement method, arranging preset irregular workpieces in each of the area units to obtain an arrangement plan; according to the arrangement plan, cutting the steel plate to obtain a preset irregular workpiece. The present invention detects the scratch area by using a scratch recognition method, which can effectively avoid arranging irregular workpieces in the scratch area, thereby improving the quality and yield of the workpiece; adapting to the different shapes of the scratch area and irregular workpieces by a dynamic division method to meet the arrangement requirements of irregular workpieces; arranging the workpieces in each area unit, optimizing the arrangement plan, and improving the utilization rate of the steel plate.

Description

Technical Field

[0001] The invention relates to the technical field of workpiece arrangement optimization, and in particular to an irregular workpiece arrangement method and system for surface scratched steel plates. Background Art

[0002] In the metal processing and manufacturing industry, steel plates often have scratches on the surface during production, transportation and processing due to various reasons. These scratches will affect the appearance quality of the steel plates. If the scratches are ignored during the production and manufacturing of the workpiece, the mechanical properties and corrosion resistance of the workpiece will be affected. Therefore, a method is needed to avoid scratches on the surface of the steel plate and perform sample processing on the workpiece.

[0003] For example, a Chinese patent with authorization announcement number CN111275243B discloses a two-dimensional irregular layout method for single-specification plates, including the following processes: an assignment process, assigning all irregular components to rectangular plates respectively; a single plate layout process, obtaining an initial solution step, arranging all irregular components assigned to rectangular plates into corresponding rectangular plates, and calculating the current length of the rectangular band, which is the initial solution; a rearrangement step, solving the minimum value of the sum of overlapping values ​​between irregular components, and rearranging the irregular components according to the solution result; an exchange step, first calling a direct exchange step to exchange the irregular components in the rectangular plate, and when the exchange fails and exits the direct exchange step, calling a taboo search exchange step; a local optimization process, exchanging irregular components one-to-many between every two rectangular plates.

[0004] The above patents have the problems raised by the background technology: they do not take into account the possible scratches on the steel plate, which will affect the quality of the workpiece; the irregular workpieces are arranged on the entire steel plate, which has low efficiency; and the steel plate may not be fully utilized if the arrangement is performed directly on the entire plate; to solve the above problems, the present invention proposes a method and system for arranging irregular workpieces for steel plates with surface scratches. Summary of the invention

[0005] In view of the shortcomings of the prior art, the main purpose of the present invention is to provide an irregular workpiece arrangement method and system for surface scratched steel plates, which can effectively solve the problems in the background technology. The specific technical solutions of the present invention are as follows:

[0006] A method for arranging irregular workpieces for surface scratched steel plates, comprising:

[0007] The scratch recognition method is used to detect the scratches on the surface of the steel plate and obtain the scratch area;

[0008] Based on the non-scratched area on the surface of the steel plate, the non-scratched area is divided by a dynamic division method to obtain a plurality of area units;

[0009] Using the irregular workpiece packing method, the preset irregular workpieces are packed in each of the area units to obtain a packing plan;

[0010] According to the arrangement scheme, the steel plate is cut to obtain a preset irregular workpiece;

[0011] The non-scratched area based on the surface of the steel plate is divided into the non-scratched area by a dynamic division method to obtain a plurality of area units, including:

[0012] According to the shape and size of the scratches in each scratch area, the non-scratched area on the surface of the steel plate is divided by a mesh division method to obtain a plurality of mesh units;

[0013] According to the preset shape and size of the irregular workpiece, the plurality of grid units are adjusted to obtain a plurality of area units;

[0014] The method of arranging irregular workpieces by arranging preset irregular workpieces in each of the area units to obtain an arranging scheme includes:

[0015] In each area unit, an irregular workpiece arrangement model is constructed;

[0016] The irregular workpiece packing model is solved by using a particle swarm optimization method combined with a multi-swarm strategy to obtain a packing plan for each area unit;

[0017] After arranging each area unit, the remaining gaps between each area unit are arranged to obtain the optimal arrangement plan;

[0018] The irregular workpiece arrangement model is constructed in each area unit, including:

[0019] Minimizing the area occupied by the discharge of irregular workpieces is taken as the objective function;

[0020] Taking the non-overlapping of each irregular workpiece and the presence of each irregular workpiece within the area unit as constraint conditions, an irregular workpiece arrangement model is obtained.

[0021] Specifically, the scratch recognition method is used to detect scratches on the surface of the steel plate to obtain the scratch area, including:

[0022] Acquiring a surface image of the steel plate;

[0023] Extracting the grayscale feature of each pixel of the surface image;

[0024] Determining the position of each scratch in the surface image according to the grayscale feature of each pixel;

[0025] According to each scratch position in the surface image, a scratch area is identified.

[0026] Specifically, the method of detecting scratches on the surface of the steel plate by using a scratch recognition method to obtain a scratch area also includes:

[0027] Performing grayscale conversion on the surface image of the steel plate to obtain a grayscale image;

[0028] De-noising the grayscale image to obtain a first de-noised image;

[0029] Using a preset window, performing boundary expansion on the first denoised image to obtain an expanded image;

[0030] Normalizing the extended image according to a preset window size to obtain a normalized image;

[0031] Denoising the normalized image to obtain an updated image.

[0032] Specifically, identifying a scratch area according to each scratch position in the surface image includes:

[0033] Based on the grayscale features, a threshold segmentation method is used to extract multiple scratch domains containing incomplete scratches;

[0034] The scratch domains belonging to the same scratch are connected by using a clustering fusion method to obtain a plurality of spliced ​​scratch domains of complete scratches, and the spliced ​​scratch domains constitute the scratch area.

[0035] Specifically, the clustering fusion method is used to connect the scratch domains belonging to the same scratch to obtain a plurality of spliced ​​scratch domains of complete scratches, including:

[0036] Clustering the multiple scratch domains containing incomplete scratches according to distance correlation to obtain multiple types of scratch domains;

[0037] Converting the scratch coordinate distribution in the scratch domain from the spatial coordinate system to the polar coordinate system to obtain a scratch sine curve of each scratch point in each scratch domain in the polar coordinate system;

[0038] Setting a first threshold, and taking the scratch domain in each category where the number of scratch sinusoidal curves is greater than the first threshold as the main scratch domain of each category;

[0039] Calculate the distance and angle between the main scratch domain and the nearby scratch domains;

[0040] Setting a distance threshold and an angle threshold, connecting the scratch domain whose distance value is less than the distance threshold and whose angle value is less than the angle threshold with the main scratch domain as a new main scratch domain;

[0041] The calculation process is repeated until each scratch domain is connected, and multiple spliced ​​scratch domains of complete scratches are obtained.

[0042] Specifically, the particle swarm optimization method combined with a multi-swarm strategy is used to solve the irregular workpiece packing model to obtain a packing scheme for each regional unit, including:

[0043] Set up multiple particle groups, initialize each particle group according to the preset area size of the irregular workpiece, set the number of particles, the maximum number of iterations, the initial position and initial speed of the particles;

[0044] Iteratively update the position and velocity of particles in each particle group, calculate the objective function value corresponding to the position of each particle, and update the individual optimal position of the particle and the global optimal position;

[0045] Set an information exchange period, during which each particle swarm exchanges the optimal particle information and updates the global optimal position of each particle swarm;

[0046] When the maximum number of iterations is reached, stop iterative updating;

[0047] The workpiece packing method corresponding to the solution with the best objective function value among the global optimal solutions of all particle swarms is selected as the packing scheme of the solved irregular workpiece packing model.

[0048] An irregular workpiece arranging system for a steel plate with surface scratches, used to implement the irregular workpiece arranging method for a steel plate with surface scratches, comprising:

[0049] The scratch recognition module detects the scratches on the surface of the steel plate using the scratch recognition method to obtain the scratch area;

[0050] A dynamic division module, based on the non-scratched area on the surface of the steel plate, divides the non-scratched area by a dynamic division method to obtain a plurality of area units;

[0051] An irregular workpiece arrangement module arranges preset irregular workpieces in each of the area units using an irregular workpiece arrangement method to obtain an arrangement plan;

[0052] A segmentation module, which cuts the steel plate according to the arrangement scheme to obtain a preset irregular workpiece;

[0053] The non-scratched area based on the surface of the steel plate is divided into the non-scratched area by a dynamic division method to obtain a plurality of area units, including:

[0054] According to the shape and size of the scratches in each scratch area, the non-scratched area on the surface of the steel plate is divided by a mesh division method to obtain a plurality of mesh units;

[0055] According to the preset shape and size of the irregular workpiece, the plurality of grid units are adjusted to obtain a plurality of area units;

[0056] The method of arranging irregular workpieces by arranging preset irregular workpieces in each of the area units to obtain an arranging scheme includes:

[0057] In each area unit, an irregular workpiece arrangement model is constructed;

[0058] The irregular workpiece packing model is solved by using a particle swarm optimization method combined with a multi-swarm strategy to obtain a packing plan for each area unit;

[0059] After arranging each area unit, the remaining gaps between each area unit are arranged to obtain the optimal arrangement plan;

[0060] The irregular workpiece arrangement model is constructed in each area unit, including:

[0061] Minimizing the area occupied by the discharge of irregular workpieces is taken as the objective function;

[0062] Taking the non-overlapping of each irregular workpiece and the presence of each irregular workpiece within the area unit as constraint conditions, an irregular workpiece arrangement model is obtained.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] The present invention detects the scratch area through the scratch recognition method, which can effectively avoid arranging irregular workpieces in the scratch area, thereby improving the quality and yield rate of the workpiece; the non-scratched area of ​​the steel plate excluding the scratch area is divided by the dynamic division method, which can adapt to the different shapes of the scratch area and irregular workpieces and meet the arrangement requirements of irregular workpieces, thereby improving the utilization rate of the steel plate space, reducing material waste, and providing a variety of arrangement space options for irregular workpieces; by arranging the irregular workpieces according to the required shape and size of the irregular workpieces in each area unit to obtain the local optimal arrangement plan, the arrangement plan can be optimized, the production efficiency can be improved, the waste generated in the cutting process can be reduced, and the cost can be reduced; the steel plate is cut according to the optimal arrangement plan to obtain the required irregular workpiece, which can ensure the quality of the workpiece, improve the production efficiency, shorten the production cycle, and improve the yield rate of workpiece production. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 This is a workflow diagram of an irregular workpiece arrangement method for a surface scratched steel plate in Example 1 of the present invention;

[0066] Figure 2 This is a schematic diagram of the structure of the scratch domain connection in Example 1 of the present invention;

[0067] Figure 3 It is a schematic diagram of the structure of dynamically dividing the non-scratch area in Embodiment 1 of the present invention;

[0068] Figure 4 This is a schematic structural diagram of an irregular workpiece arrangement system for surface-scratched steel plates in Example 2 of the present invention. DETAILED DESCRIPTION

[0069] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0070] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0071] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0072] Example 1

[0073] This embodiment provides a method for arranging irregular workpieces for surface scratched steel plates, such as Figure 1 As shown, the irregular workpiece arrangement method for surface scratched steel plates comprises:

[0074] S101, detecting scratches on the surface of the steel plate using a scratch recognition method to obtain a scratch area;

[0075] S102, based on the non-scratched area on the surface of the steel plate, dividing the non-scratched area by using a dynamic division method to obtain a plurality of area units;

[0076] S103, using an irregular workpiece arrangement method, arranging preset irregular workpieces in each of the area units to obtain an arrangement plan;

[0077] S104, cutting the steel plate according to the layout plan to obtain a preset irregular workpiece.

[0078] Since some scratches may be generated on the surface of the steel plate due to friction during the production and transportation process, these scratches will affect the function of the workpiece when the workpiece is produced, resulting in a high defective rate of the workpiece. Therefore, the present invention avoids the scratch area to arrange and produce irregular workpieces.

[0079] In this embodiment, firstly, a scratch recognition method is used to detect scratches on the surface of the steel plate to obtain a scratch area. Specifically, a surface image of the steel plate is obtained by photographing with a camera, and the scratch area on the surface of the steel plate is identified and marked by combining image recognition and machine vision algorithms. By accurately identifying the scratch area, data support is provided for subsequent steps, which can effectively avoid arranging irregular workpieces in the scratch area, thereby improving the quality and yield of the workpiece; secondly, after identifying the scratch area, these scratch areas are removed from the steel plate, and the obtained scratch-free areas are used as non-scratch areas. The non-scratch areas are divided by a dynamic division method to obtain a plurality of area units of different shapes and sizes. Through dynamic area division, it is possible to adapt to different shapes of scratch areas and irregular workpieces and meet the arrangement requirements of irregular workpieces, thereby improving the utilization rate of the steel plate space, reducing material waste, and providing a variety of methods for irregular workpieces. Optimized nesting space selection; in each area unit, the preset irregular workpieces are nested using an irregular workpiece nesting method. The irregular workpiece nesting method can be a genetic algorithm, a simulated annealing algorithm, a dynamic programming method, etc. Specifically, it is a particle swarm optimization algorithm. According to the shape and size of the required irregular workpieces, nesting is performed in each area unit, with the goal of minimizing material waste and maximizing nesting density, optimizing the nesting scheme, improving production efficiency, reducing waste generation in the cutting process, and reducing costs; finally, according to the optimal nesting scheme, high-precision cutting equipment (such as laser cutting machine, water jet cutting machine, etc.) is used to cut the steel plate to obtain the preset irregular workpieces. The cutting process must be carried out strictly in accordance with the nesting scheme to ensure that the shape, size and position of each workpiece are accurate, achieve high-precision cutting, ensure workpiece quality, improve production efficiency, and shorten the production cycle.

[0080] On the whole, this embodiment uses an industrial camera to capture the surface image of the steel plate, transmits the surface image to a data processing unit, and pre-processes the surface image to remove background noise caused by reflections generated in the shooting environment; uses image recognition methods and machine vision algorithms to identify and mark scratched areas on the surface of the steel plate; constructs an irregular workpiece nesting model based on the position and shape of the non-scratched area and the shape and size of the required irregular workpiece, and uses a particle swarm optimization method to solve the irregular workpiece nesting model to obtain an optimal nesting solution; a cutting machine cuts the steel plate according to the placement position of each workpiece in the optimal nesting solution to obtain the required irregular workpiece.

[0081] The present invention detects the scratch area through the scratch recognition method, which can effectively avoid arranging irregular workpieces in the scratch area, thereby improving the quality and yield rate of the workpiece; the non-scratched area of ​​the steel plate excluding the scratch area is divided by the dynamic division method, which can adapt to the different shapes of the scratch area and irregular workpieces and meet the arrangement requirements of irregular workpieces, thereby improving the utilization rate of the steel plate space, reducing material waste, and providing a variety of arrangement space options for irregular workpieces; by arranging the irregular workpieces according to the required shape and size of the irregular workpieces in each area unit to obtain the local optimal arrangement plan, the arrangement plan can be optimized, the production efficiency can be improved, the waste generated in the cutting process can be reduced, and the cost can be reduced; the steel plate is cut according to the optimal arrangement plan to obtain the required irregular workpiece, which can ensure the quality of the workpiece, improve the production efficiency, shorten the production cycle, and improve the yield rate of workpiece production.

[0082] Furthermore, the scratch recognition method is used to detect scratches on the surface of the steel plate to obtain the scratch area, including:

[0083] S201, acquiring a surface image of the steel plate;

[0084] S202, extracting the grayscale feature of each pixel of the surface image;

[0085] S203, determining the position of each scratch in the surface image according to the grayscale feature of each pixel;

[0086] S204: Identify a scratch area according to each scratch position in the surface image.

[0087] In this embodiment, an image of the steel plate surface is captured by a camera device (such as an industrial camera) to obtain a clear image. The quality of the image directly affects the accuracy and efficiency of scratch detection. The industrial camera can be used to obtain a relatively clear steel plate image and provide raw data for scratch detection. By shooting with a camera, the non-contact nature of the scratch detection process can be ensured to avoid secondary damage to the steel plate. Since the steel plate metal material itself has a highly reflective property, there will be background noise in the collected image, which will affect the accuracy of scratch recognition. Therefore, in order to improve the efficiency and accuracy of scratch recognition, it is necessary to use a scratch enhancement method to preprocess the steel plate image to enhance the contrast between the scratch and the background to obtain a preprocessed image. By eliminating noise and interference in the image, the accuracy of scratch detection can be improved. By increasing the contrast of the scratch area, the scratch is easier to identify and locate.

[0088] Specifically, by calculating the statistical features and texture features of the preprocessed image, multiple features in the image, including grayscale features and position distribution features, are extracted to obtain a multi-feature set. The grayscale feature can reflect the grayscale difference between the scratch area and the background, and the position distribution feature can describe the position, shape, size and other information of the scratch in the image. By extracting multiple features of the image, rich information is provided for the recognition and positioning of the scratches. According to the extracted multi-feature set, a multi-feature fusion method is used to combine and fuse multiple features, and multiple features such as grayscale features and position distribution features are fused to form a comprehensive feature vector. The comprehensive feature vector is classified and identified, and the scratch area is determined. Through the multi-feature fusion method, the complementarity between different features is fully utilized to improve the accuracy of scratch recognition.

[0089] Furthermore, the method of detecting scratches on the surface of the steel plate by using a scratch recognition method to obtain a scratch area also includes:

[0090] S301, performing grayscale conversion on the surface image of the steel plate to obtain a grayscale image;

[0091] S302, denoising the grayscale image to obtain a first denoised image;

[0092] S303, using a preset window, performing boundary expansion on the first denoised image to obtain an expanded image;

[0093] S304, performing normalization processing on the extended image according to a preset window size to obtain a normalized image;

[0094] S305: De-noise the normalized image to obtain an updated image.

[0095] In this embodiment, the collected steel plate image is a color image. To facilitate subsequent processing, the steel plate image is gray-converted and converted from a color image to a gray-scale image. The color image is composed of three color channels: red, green, and blue, while the gray-scale image has only one brightness channel, which represents the brightness information of the image. Through gray-scale conversion, the image data can be simplified, the amount of calculation can be reduced, the color information in the image can be removed, and the scratch area can be made more prominent in the image. In the process of steel plate image acquisition, noise will be generated due to the acquisition environment and other reasons. The denoising method can be mean filtering, median filtering, Gaussian filtering, etc. Specifically, the gray-scale image is denoised through mean filtering to smooth the image and reduce noise, which can improve the image quality and make the scratches clearer.

[0096] Specifically, the boundary of the denoised image is expanded according to the window size k to protect the boundary information of the image from being destroyed. The window size can be set according to the actual calculation requirements. The matrix pixels within each window size of k×k in the expanded image are normalized. The normalization formula is:

[0097]

[0098] In the formula, is the normalized pixel value, is the original pixel value, is the average value of the matrix element pixels in the window, is the standard deviation of the matrix element pixels in the window. The pixels in each window are normalized to return the image pixel values ​​to a distribution with a mean of 0 and a variance of 1. Through normalization, the image data has a uniform scale, which is convenient for subsequent processing. Since new noise may be introduced or the original noise may be enhanced during the normalization process, denoising is required again after normalization. By further reducing the noise in the image and improving the image quality, high-quality image data is provided for subsequent scratch detection and analysis.

[0099] Further, the identifying the scratch area according to each scratch position in the surface image includes:

[0100] S401, extracting a plurality of scratch domains including incomplete scratches using a threshold segmentation method based on the grayscale feature;

[0101] S402, using a clustering fusion method, connecting scratch domains belonging to the same scratch to obtain a plurality of spliced ​​scratch domains of complete scratches, wherein the spliced ​​scratch domains constitute the scratch area.

[0102] In this embodiment, scratch recognition is performed on the image according to the window size. According to the grayscale features and position distribution features in the multi-feature set, by setting the grayscale threshold, the pixel points with grayscale values ​​higher than the grayscale threshold are classified as scratch areas, and the remaining pixel points are classified as non-scratch areas by using the threshold segmentation method, and multiple scratch domains containing scratches are identified. By using the threshold segmentation method, the area containing scratches can be accurately extracted according to the grayscale features and position distribution features. By performing scratch recognition in each window, the accuracy of scratch recognition can be improved. The scratches are generally distributed as irregular curves, and can be locally displayed as short straight lines. Scratches belonging to the same scratch The domains will be relatively close in distance. Clustering is used to obtain multiple scratch domains that are preliminarily identified as belonging to the same scratch. In each clustered category, the scratch domains belonging to the same scratch are connected by further calculating the distance similarity and angle similarity between the scratch domains. After connection processing, multiple complete scratch domains can be obtained. These scratch domains accurately describe the position and shape information of the scratches in the steel plate image. Through the clustering fusion method, the scratches divided into multiple parts can be connected to obtain a complete scratch domain, thereby improving the integrity of scratch recognition. The combination of clustering algorithm and connection processing can further screen the scratch domains belonging to the same scratch.

[0103] Furthermore, the clustering fusion method is used to connect the scratch domains belonging to the same scratch to obtain a plurality of spliced ​​scratch domains of complete scratches, including:

[0104] S501, clustering the plurality of scratch domains containing incomplete scratches according to distance correlation to obtain a plurality of types of scratch domains;

[0105] S502, converting the scratch coordinate distribution in the scratch domain from the spatial coordinate system to the polar coordinate system, and obtaining the scratch sine curve of each scratch point in each scratch domain in the polar coordinate system;

[0106] S503, setting a first threshold, and taking the scratch domain in each category where the number of scratch sinusoidal curves is greater than the first threshold as the main scratch domain of each category;

[0107] S504, calculating the distance value and angle value between the main scratch domain and the nearby scratch domains;

[0108] S505, setting a distance threshold and an angle threshold, connecting the scratch domain whose distance value is smaller than the distance threshold and whose angle value is smaller than the angle threshold to the main scratch domain, as a new main scratch domain;

[0109] S506, repeat the calculation process until each scratch domain is connected, and obtain a plurality of complete scratch splicing scratch domains.

[0110] In this embodiment, according to the distance correlation between each scratch domain, all the scratch domains identified to contain scratches are clustered, and each category obtained is a scratch domain that is close in distance. The scratch domains of each category are preliminarily identified as scratch domains belonging to the same scratch. By clustering, the data processing speed can be improved and the calculation complexity can be reduced; the scratch coordinate distribution in the scratch domain is converted from the spatial coordinate system to the polar coordinate system. The conversion formula is as follows:

[0111]

[0112] In the formula, and are the independent variables in the spatial rectangular coordinate system, and are the independent variables in the polar coordinate system, and In order to obtain the offset generated by transferring from the spatial rectangular coordinate system to the polar coordinate system, the scratch sine curve of each scratch point in each scratch domain in the polar coordinate system is obtained through coordinate transformation. The longer the straight line segment in the rectangular coordinate system, the more corresponding sine curves passing through the same point in the polar coordinate system. A first threshold is set, and the scratch domain in each category with the number of scratch sine curves greater than the first threshold is used as the main scratch domain of each category. The longer scratch is determined as the main scratch according to the scratch length. By screening out the main scratch domain and using the main scratch as a reference, the redundancy and error of subsequent processing can be reduced.

[0113] like Figure 2 , according to the position relationship of the scratches in the scratch domain, the distance and angle between the main scratch domain and the nearby scratch domain are calculated. The calculation formula is as follows:

[0114]

[0115]

[0116] In the formula, is the distance between the main scratch domain and the secondary scratch domain, is the angle value of the scratch between the main scratch domain and the secondary scratch domain, and are the horizontal and vertical coordinates of the scratch centroid in the secondary scratch domain, and The horizontal and vertical coordinates of the endpoints of the scratch frame of the main scratch domain are respectively. Set the distance threshold according to the actual calculation accuracy and requirements and angle threshold , when the calculated distance and angle values ​​between the secondary scratch domain and the primary scratch domain are both less than the distance threshold and angle threshold, that is, D When the secondary scratch domain and the main scratch domain are connected, the secondary scratch domain is considered to belong to the same scratch, and the secondary scratch domain is connected to the main scratch domain to form a new main scratch domain. By comparing the distance value and the angle value, the secondary scratch domain that meets the judgment condition is connected to the main scratch domain to form a new main scratch domain. The scratch domains belonging to the same scratch are merged to obtain a complete scratch domain range, thereby improving the accuracy and completeness of scratch recognition. Repeat the calculation process of the distance value and the angle value between the main scratch domain and the secondary scratch domain until each secondary scratch domain is connected to form multiple complete scratch domains. Through the iterative calculation process, all independent scratch domains are attributed to the scratch domain belonging to the same scratch, ensuring that all scratch domains are correctly connected, thereby improving the overall efficiency and accuracy of scratch recognition.

[0117] Furthermore, the non-scratched area based on the steel plate surface is divided into the non-scratched area using a dynamic division method to obtain a plurality of area units, including:

[0118] S601, dividing the non-scratched area on the surface of the steel plate by a mesh division method according to the shape and size of the scratches in each of the scratched areas to obtain a plurality of mesh units;

[0119] S602: According to a preset shape and size of the irregular workpiece, the plurality of grid units are adjusted to obtain a plurality of area units.

[0120] In this embodiment, the scratch domain is removed from the steel plate according to each identified scratch domain, and irregular workpieces are arranged in the area after the scratch domain is removed. By avoiding the scratch area for arrangement, the quality and yield of workpiece production can be improved. Since the shapes and sizes of scratches are not fixed, directly arranging irregular workpieces in the area where the scratches are removed may lead to reduced space utilization and fail to consider edge issues. Therefore, before arrangement, the non-scratched area is divided into areas to meet the requirements of different scratch shapes and sizes and irregular workpiece shapes and sizes, thereby improving the utilization of the steel plate.

[0121] Specifically, Figure 3 First, in the non-scratch area, the grid division method is used for division. The present embodiment adopts the pentahedral grid division to obtain multiple regular and complete pentahedral areas. The grid shape can be set according to the actual workpiece shape and calculation requirements. Generally, the smaller the grid size, the higher the calculation accuracy. The non-scratch area is divided into multiple regular area units by the grid division method. The workpieces are arranged in these regular areas to reduce the difficulty of arrangement and simplify the calculation process; the vertices of the pentahedral grid are connected to the end points of the scratch domain so that the irregular edges are only divided into one area grid, which can improve the efficiency of arrangement; the division results obtained by the preliminary division are dynamically adjusted according to the shape and size of the scratch domain and the shape and size of the required irregular workpiece. When the required workpiece size is large, two or more adjacent grid units can be merged to ensure that at least one irregular workpiece can be accommodated in each area unit, so as to realize the optimization process of area division; the shape, size and quantity of the required workpieces and their layout on the steel plate need to be considered in the dynamic optimization process of area division. By adjusting the grid unit, the layout of the irregular workpiece on the steel plate can be optimized, thereby improving the material utilization rate and reducing the waste generation during the cutting process.

[0122] Furthermore, the irregular workpiece arrangement method is used to arrange the preset irregular workpieces in each of the area units to obtain an arrangement plan, including:

[0123] S701, constructing an irregular workpiece arrangement model in each area unit;

[0124] S702, using a particle swarm optimization method combined with a multi-swarm strategy to solve the irregular workpiece packing model, and obtain a packing plan for each area unit;

[0125] S703, after arranging each area unit, arranging the remaining gaps between each area unit to obtain an optimal arrangement scheme.

[0126] In this embodiment, in each divided regional unit, an irregular workpiece packing model is constructed according to the relationship between the workpiece packings. Through the construction of the model, a data basis is provided for the subsequent solution of the optimal packing scheme, indicating the arrangement of irregular workpieces in the regional unit, which is helpful to optimize the packing scheme; in each regional unit, the constructed irregular workpiece packing model is solved, and the solution method used is a particle swarm optimization method combined with a multi-group strategy. The particle swarm optimization method simulates the foraging behavior of a flock of birds and seeks the optimal solution through information sharing between particles. In the irregular workpiece packing problem, each particle represents a packing scheme, the position of the particle represents the position and direction of the workpiece in the regional unit, and the speed of the particle represents the trend and speed of the workpiece movement. The optimal packing scheme is obtained by continuously iterating and updating the position and speed of the particles; the particle swarm is divided into multiple subgroups, each particle swarm can be solved independently, and the optimal solution between the overall particle swarm is obtained through information exchange between multiple particle swarms. The particle swarm optimization method combined with a multi-group strategy can enhance the search ability of the algorithm and avoid falling into a local optimal solution.

[0127] Specifically, after arrangement in each area unit, there will be some remaining gaps between adjacent area units, and these gaps can be further arranged. By analyzing the shape and size of the remaining gaps, the remaining gaps between each area unit are arranged, and irregular workpieces of suitable shapes and sizes are selected for filling. By arranging in the gaps between the area units, the remaining gaps between the area units are fully utilized, which improves material utilization and arrangement efficiency. By optimizing the workpiece arrangement plan, the optimal arrangement plan can be further approached, thereby reducing material waste and production costs.

[0128] Furthermore, in each area unit, constructing an irregular workpiece arrangement model includes:

[0129] S801, minimizing the area occupied by the discharge of irregular workpieces as an objective function;

[0130] S802: Taking the non-overlapping of each irregular workpiece and the presence of each irregular workpiece within the area unit as constraint conditions, a model for arranging irregular workpieces is obtained.

[0131] In this embodiment, minimizing the area occupied by the irregular workpieces is used as the objective function, and each irregular workpiece does not overlap with each other and each irregular workpiece is inside the area unit as a constraint condition, and an irregular workpiece arrangement model is obtained. Specifically, n irregular workpieces are arranged on a steel plate with a width of W and a length of L, and the area occupied by the irregular workpieces is minimized. The arrangement strategy is designed to minimize the steel plate length L occupied by the workpieces after arrangement. At the same time, it is required that the workpieces do not overlap, avoid the scratch area, and are completely inside the steel plate. Therefore, the irregular workpiece arrangement model is:

[0132]

[0133]

[0134]

[0135]

[0136]

[0137] In the formula, is the width of the steel plate, The length of the steel plate occupied by the workpiece after arrangement. is the coordinate set of the irregular workpiece, represents the coordinates of the i-th workpiece, represents the coordinates of the jth workpiece, To allow for a set of rotation angles, represents the placement angle of the i-th workpiece, is the total number of irregular artifacts, is the coordinate point set of the scratch domain, Represents the coordinates of the irregular workpiece, Represents the coordinate point set after the workpiece is rotated clockwise by an angle, represents the area unit range of the steel plate, Indicates that there is no overlap between the workpieces. It means that the workpiece is completely inside the steel plate. Indicates that the workpiece and the scratch area do not overlap, Indicates that the placement angle of the workpiece is within the allowed rotation angle set.

[0138] Furthermore, the particle swarm optimization method combined with a multi-swarm strategy is used to solve the irregular workpiece packing model to obtain a packing scheme for each area unit, including:

[0139] S901, setting multiple particle groups, initializing each particle group according to the preset area size of the irregular workpiece, setting the number of particles, the maximum number of iterations, the initial position and the initial speed of the particles;

[0140] S902, iteratively updating the position and velocity of particles in each particle group, calculating the objective function value corresponding to the position of each particle, and updating the individual optimal position of the particle and the global optimal position;

[0141] S903, setting an information exchange period, during which each particle swarm exchanges optimal particle information and updates the global optimal position of each particle swarm;

[0142] S904, when the maximum number of iterations is reached, stop iterative updating;

[0143] S905, selecting a solution with an optimal objective function value among the global optimal solutions of all particle swarms, and using the corresponding workpiece arrangement method as the arrangement solution of the solved irregular workpiece arrangement model.

[0144] In this embodiment, a particle swarm optimization method combined with a multi-group strategy is used to solve the irregular workpiece packing model. Specifically, multiple particle swarms are set in each area unit. According to the preset area size of the irregular workpiece, the number of particles and the maximum number of iterations are set for each particle swarm, and the initial position and initial speed of the particle are randomly initialized. The position of the particle can represent a packing scheme, specifically including the position coordinates and angle of each workpiece, and the speed represents the changing trend of the packing scheme. By setting multiple particle swarms, the problem is decomposed into multiple sub-problems, and the optimal solution can be searched in parallel to improve the search efficiency; in each particle swarm after initialization, according to the principle of the particle swarm optimization algorithm, the position and speed of the particle are iteratively updated, the speed of the particle needs to be calculated, the position of the particle is updated, and the objective function value is calculated according to the new position. The quality of the scheme is evaluated by the calculated objective function value. In each iteration, the individual optimal position and the global optimal position of the particle are updated according to the objective function value. Through iterative updating, the particle can approach the optimal solution and improve the search accuracy.

[0145] Specifically, during the solution process, a fixed information exchange cycle is set to exchange the optimal particle information between each particle swarm. In each information exchange cycle, the particle swarms will exchange information such as the position of the optimal particle and the objective function value. By exchanging information, each particle swarm will update the global optimal position of the particle swarm again according to the search progress and optimal solution of other particle swarms. The optimal solution information can be shared by each particle swarm to accelerate the convergence process. Through information exchange, each particle swarm can share the optimal solution information to avoid falling into the local optimal solution and accelerate the convergence process. When the number of iterations reaches the preset maximum number of iterations, the iterative update is stopped. By setting the maximum number of iterations, it is ensured that the algorithm completes the search process within a limited time to avoid an infinite loop. After the iterative process stops, the solution with the optimal objective function value is selected from the global optimal solutions of all particle swarms as the optimal result of the solved irregular workpiece arrangement model to ensure that the result obtained is the optimal solution among all particle swarm results. The particle position corresponding to the optimal solution, that is, the placement position and angle of the irregular workpiece, is used as the optimal result of the solved irregular workpiece arrangement model to achieve efficient and accurate model solution.

[0146] Example 2

[0147] In this embodiment, if Figure 4 , providing an irregular workpiece arranging system for surface scratched steel plates, used to implement the irregular workpiece arranging method for surface scratched steel plates, comprising:

[0148] The scratch recognition module detects the scratches on the surface of the steel plate using the scratch recognition method to obtain the scratch area;

[0149] A dynamic division module, based on the non-scratched area on the surface of the steel plate, divides the non-scratched area by a dynamic division method to obtain a plurality of area units;

[0150] An irregular workpiece arrangement module arranges preset irregular workpieces in each of the area units using an irregular workpiece arrangement method to obtain an arrangement plan;

[0151] The segmentation module cuts the steel plate according to the layout plan to obtain a preset irregular workpiece.

[0152] In this embodiment, the scratch recognition module uses scratch recognition technology to detect scratches on the surface of the steel plate, and determines the specific location and range of these scratches, thereby dividing the scratch area. By accurately identifying the scratches, it is possible to avoid placing the workpiece on the scratches during layout, thereby ensuring the quality of the workpiece; the dynamic division module divides the non-scratch area of ​​the steel plate with the scratch area removed, and uses the dynamic division method to further subdivide these areas into multiple area units. Through dynamic division, it can adapt to irregular workpieces of different sizes and shapes, flexibly respond to complex conditions on the surface of the steel plate, improve the flexibility and efficiency of layout, and ensure that each area unit can accommodate the workpiece to the maximum extent, reducing material waste.

[0153] Specifically, the irregular workpiece nesting module uses the irregular workpiece nesting method to nest the preset irregular workpieces in each area unit and find the optimal nesting scheme, that is, to minimize material waste while meeting the workpiece size, shape and quantity requirements. By optimizing the nesting scheme, the material utilization rate can be significantly improved, the production cost can be reduced, the space between the workpieces can be effectively utilized, and the error and waste in the cutting process can be reduced; the segmentation module cuts the steel plate according to the final nesting scheme to obtain the preset irregular workpiece. The quality and dimensional accuracy of the workpiece are ensured through precise cutting. Through the efficient cutting process, the production cycle can be shortened and the production efficiency can be improved.

[0154] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for arranging irregular workpieces for surface scratched steel plates, characterized in that: include: The scratch recognition method is used to detect the scratches on the surface of the steel plate and obtain the scratch area; Based on the non-scratched area on the surface of the steel plate, the non-scratched area is divided by a dynamic division method to obtain a plurality of area units; Using the irregular workpiece packing method, the preset irregular workpieces are packed in each of the area units to obtain a packing plan; According to the arrangement scheme, the steel plate is cut to obtain a preset irregular workpiece; The non-scratched area based on the surface of the steel plate is divided into the non-scratched area by a dynamic division method to obtain a plurality of area units, including: According to the shape and size of the scratches in each scratch area, the non-scratched area on the surface of the steel plate is divided by a mesh division method to obtain a plurality of mesh units; According to the preset shape and size of the irregular workpiece, the plurality of grid units are adjusted to obtain a plurality of area units; The method of arranging irregular workpieces by arranging preset irregular workpieces in each of the area units to obtain an arranging scheme includes: In each area unit, an irregular workpiece arrangement model is constructed; The irregular workpiece packing model is solved by using a particle swarm optimization method combined with a multi-swarm strategy to obtain a packing plan for each area unit; After arranging each area unit, the remaining gaps between each area unit are arranged to obtain the optimal arrangement plan; The irregular workpiece arrangement model is constructed in each area unit, including: Minimizing the area occupied by the discharge of irregular workpieces is taken as the objective function; Taking the non-overlapping of each irregular workpiece and the presence of each irregular workpiece within the area unit as constraint conditions, an irregular workpiece arrangement model is obtained.

2. The irregular workpiece arrangement method for surface scratched steel plates according to claim 1, characterized in that: The method of detecting scratches on the surface of the steel plate by using the scratch recognition method to obtain the scratch area includes: Acquiring a surface image of the steel plate; Extracting the grayscale feature of each pixel of the surface image; Determining the position of each scratch in the surface image according to the grayscale feature of each pixel; According to each scratch position in the surface image, a scratch area is identified.

3. The irregular workpiece arrangement method for surface scratched steel plates according to claim 2, characterized in that: The method of detecting scratches on the surface of the steel plate by using a scratch recognition method to obtain a scratch area also includes: Performing grayscale conversion on the surface image of the steel plate to obtain a grayscale image; De-noising the grayscale image to obtain a first de-noised image; Using a preset window, performing boundary expansion on the first denoised image to obtain an expanded image; Normalizing the extended image according to a preset window size to obtain a normalized image; Denoising the normalized image to obtain an updated image.

4. The irregular workpiece arrangement method for surface scratched steel plates according to claim 2, characterized in that: The step of identifying a scratch area according to each scratch position in the surface image comprises: Based on the grayscale features, a threshold segmentation method is used to extract multiple scratch domains containing incomplete scratches; The scratch domains belonging to the same scratch are connected by using a clustering fusion method to obtain a plurality of spliced ​​scratch domains of complete scratches, and the spliced ​​scratch domains constitute the scratch area.

5. The irregular workpiece arrangement method for surface scratched steel plates according to claim 4, characterized in that: The clustering fusion method is used to connect the scratch domains belonging to the same scratch to obtain multiple spliced ​​scratch domains of complete scratches, including: Clustering the multiple scratch domains containing incomplete scratches according to distance correlation to obtain multiple types of scratch domains; Converting the scratch coordinate distribution in the scratch domain from the spatial coordinate system to the polar coordinate system to obtain a scratch sine curve of each scratch point in each scratch domain in the polar coordinate system; Setting a first threshold, and taking the scratch domain in each category where the number of scratch sinusoidal curves is greater than the first threshold as the main scratch domain of each category; Calculate the distance and angle between the main scratch domain and the nearby scratch domains; Setting a distance threshold and an angle threshold, connecting the scratch domain whose distance value is less than the distance threshold and whose angle value is less than the angle threshold with the main scratch domain as a new main scratch domain; The calculation process is repeated until each scratch domain is connected, and multiple spliced ​​scratch domains of complete scratches are obtained.

6. The irregular workpiece arrangement method for surface scratched steel plates according to claim 1, characterized in that: The method of solving the irregular workpiece packing model by using the particle swarm optimization method combined with a multi-swarm strategy to obtain a packing scheme for each regional unit includes: Set up multiple particle swarms, initialize each particle swarm according to the preset area size of the irregular workpiece, and set the number of particles, the maximum number of iterations, the initial position and initial velocity of the particles; Iteratively update the position and velocity of particles in each particle group, calculate the objective function value corresponding to the position of each particle, and update the individual optimal position of the particle and the global optimal position; Set an information exchange period, during which each particle swarm exchanges the optimal particle information and updates the global optimal position of each particle swarm; When the maximum number of iterations is reached, stop iterative updating; The workpiece packing method corresponding to the solution with the best objective function value among the global optimal solutions of all particle swarms is selected as the packing scheme of the solved irregular workpiece packing model.

7. An irregular workpiece arrangement system for surface scratched steel plates, characterized in that: A method for arranging irregular workpieces for surface scratched steel plates according to any one of claims 1 to 6, comprising: The scratch recognition module detects scratches on the surface of the steel plate using a scratch recognition method to obtain a scratch area; A dynamic division module, based on the non-scratched area on the surface of the steel plate, divides the non-scratched area by a dynamic division method to obtain a plurality of area units; An irregular workpiece arrangement module arranges preset irregular workpieces in each of the area units using an irregular workpiece arrangement method to obtain an arrangement plan; A segmentation module, which cuts the steel plate according to the layout plan to obtain a preset irregular workpiece; The non-scratched area based on the surface of the steel plate is divided into the non-scratched area by a dynamic division method to obtain a plurality of area units, including: According to the shape and size of the scratches in each scratch area, the non-scratched area on the surface of the steel plate is divided by a mesh division method to obtain a plurality of mesh units; According to the preset shape and size of the irregular workpiece, the plurality of grid units are adjusted to obtain a plurality of area units; The method of arranging irregular workpieces is used to arrange preset irregular workpieces in each of the area units to obtain an arrangement plan, including: In each area unit, an irregular workpiece arrangement model is constructed; The irregular workpiece packing model is solved by using a particle swarm optimization method combined with a multi-swarm strategy to obtain a packing plan for each area unit; After arranging each area unit, the remaining gaps between each area unit are arranged to obtain the optimal arrangement plan; The irregular workpiece arrangement model is constructed in each area unit, including: Minimizing the area occupied by the discharge of irregular workpieces is taken as the objective function; Taking the non-overlapping of each irregular workpiece and the presence of each irregular workpiece within the area unit as constraint conditions, an irregular workpiece arrangement model is obtained.

Images