An AI-based optimization method and system for multi-layer palletizing of aluminized zinc-coated steel coils

Abstract

This invention relates to the field of artificial intelligence technology, specifically disclosing an AI-based method and system for optimizing the multi-layer stacking of galvanized steel coils. The method includes: extracting multi-dimensional features from the steel coil attribute information of the target object to obtain geometric adaptation parameters and stackable feature parameters; performing structured spatial analysis on the environmental information of the target object and constructing a stacking operation space map of the target object; performing spatial optimization configuration on the stacking operation space map to obtain an initial stacking sequence and initial spatial occupancy layout; performing force transmission analysis on the target object, and based on the analyzed contact force network, extrapolating the offset trajectory of the target object's center of gravity to identify overturning risk points; and verifying the support capacity of the target object's contact state to adjust the inter-layer misalignment angle and the coil arrangement order. This invention can improve the efficiency of AI-based optimization for multi-layer stacking of galvanized steel coils.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based method and system for optimizing the multi-layer stacking of aluminized zinc steel coils. Background Technology

[0002] Traditional methods for stacking aluminized zinc steel coils cannot extract multi-dimensional features of the coil's properties, making it difficult to accurately obtain geometric adaptation and stackable core parameters. Furthermore, they cannot perform structured spatial analysis of the working environment, resulting in a lack of scientific spatial configuration basis for the stacking layout.

[0003] Existing palletizing technologies lack systematic force transmission analysis and center of gravity offset trajectory deduction, making it impossible to accurately identify tipping risk points. Furthermore, they cannot perform support capacity verification and dynamic layout adjustment based on risk characteristics. Therefore, improving the spatial adaptability and structural stability of multi-layer palletizing of aluminized zinc steel coils and optimizing palletizing operation efficiency have become urgent problems to be solved. Summary of the Invention

[0004] This invention provides an artificial intelligence-based method and system for optimizing the multi-layer stacking of aluminized zinc steel coils, in order to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides an artificial intelligence-based multi-layer palletizing optimization method for aluminized zinc-coated steel coils, comprising: Multidimensional feature extraction is performed on the steel coil attribute information of the target object to obtain the geometric adaptation parameters and stackable feature parameters of the target object. The environmental information of the target object is analyzed in a structured spatial manner, and a palletizing operation spatial map of the target object is constructed based on the analyzed three-dimensional grid occupancy status and grid point load-bearing capacity threshold. Based on geometric adaptation parameters and stackable feature parameters, spatial optimization configuration is performed on the palletizing operation spatial map to obtain the initial stacking sequence and initial spatial occupancy layout of the target object. Based on the initial stacking sequence and initial spatial occupancy layout, force transmission analysis is performed on the target object, and based on the analyzed contact force network, the offset trajectory of the target object's center of gravity is deduced to identify the overturning risk points of the target object. Based on the distribution characteristics of overturning risk points, the support capacity of the target object's contact state is verified in order to adjust the interlayer misalignment angle and the arrangement order of the steel coils.

[0006] In a preferred embodiment, the step of performing multi-dimensional feature extraction on the steel coil attribute information of the target object to obtain the geometric adaptation parameters and stackable feature parameters of the target object includes: Acquire end-face image sequences and 3D point cloud data of the target object; Edge features are extracted from the end face image sequence, and arc segment tracking analysis is performed on the extracted end face edge points to obtain the end face curvature radius data and end face center coordinate data of the target object; Based on 3D point cloud data, surface integrity detection is performed on the end face area of ​​the target object to obtain end face flatness deviation data and end face bearing strength prediction data. The end face curvature radius data, end face center coordinate data, end face flatness deviation data, and end face bearing capacity prediction data are integrated into the geometric adaptation parameters and stackable feature parameters of the target object.

[0007] In a preferred embodiment, the structured spatial parsing of the environmental information of the target object includes: Spatial rasterization is performed on the scene point cloud data of the target object to obtain a three-dimensional spatial raster of the scene point cloud data; Density statistics are performed on the three-dimensional spatial grid to obtain the point cloud distribution of the three-dimensional spatial grid; Based on the point cloud distribution, determine the 3D grid occupancy status of the target object; Spatial registration analysis was performed on the ground bearing capacity data of the target object and the three-dimensional spatial grid to obtain the positional correspondence between the ground bearing capacity data and the three-dimensional spatial grid; Based on the location correspondence, the ground bearing capacity data is mapped to a three-dimensional spatial grid to obtain the load-bearing capacity threshold of the grid points of the target object.

[0008] In a preferred embodiment, the step of constructing a palletizing operation space map of the target object based on the parsed three-dimensional grid occupancy status and grid point load-bearing capacity threshold includes: Mark the grid cells marked as free in the 3D grid occupancy status as candidate placement grid cells; Based on the spatial relationship of the candidate placement grids, the candidate placement grids are aggregated by adjacency to obtain the continuous placeable area of ​​the candidate placement grids; Based on the neighborhood grid attributes of the continuous placeable region, the upper limit of the load-bearing capacity of the continuous placeable region is spatially weighted and corrected to obtain the corrected load-bearing threshold of the candidate placement grid. The occupancy identifier, the corrected load-bearing threshold, and the ownership identifier of the continuous placeable area of ​​the candidate placement grid are constructed using attribute tensors to obtain the multidimensional features of the candidate placement grid. Based on the grid spatial coordinates, spatial indexing and association are performed on the multidimensional features to obtain the palletizing operation spatial map of the target object.

[0009] In a preferred embodiment, the step of spatially optimizing the palletizing operation spatial map based on geometric adaptation parameters and stackable feature parameters to obtain the initial stacking sequence and initial spatial occupancy layout of the target object includes: Based on the mechanical bearing parameters in the stackable characteristic parameters, the target objects are arranged in a comprehensive mechanical hierarchy to obtain the stacking priority sequence of the target objects. Based on the stacking priority sequence, candidate placement grids that match the current steel coil geometry adaptation parameters are extracted sequentially from the palletizing operation space map; Calculate the initial contact matching degree of the candidate placement grid; Based on the end face flatness deviation data of the target object, the initial contact matching degree is weighted and corrected to obtain the comprehensive adaptation weight of the candidate placement grid. Based on the comprehensive adaptation weight and stacking priority sequence, the spatial occupancy fusion configuration of the palletizing operation spatial map is performed to obtain the initial stacking sequence and initial spatial occupancy layout of the target object.

[0010] In a preferred embodiment, the initial contact matching degree is calculated using the following formula: ; In the formula, For the first The initial contact matching degree of each candidate placement grid. For the first The radius of curvature of the support profile corresponding to each candidate placement grid. The radius of curvature of the end face of the steel coil to be placed is [value]. This is the minimum spacing between the center of the candidate placement grid and the edge of the already placed steel coil. The preset maximum allowable spacing, This is the preset weight adjustment factor.

[0011] In a preferred embodiment, the step of performing force transmission analysis on the target object based on the initial stacking sequence and initial spatial occupancy layout includes: Spatial geometry analysis is performed on the initial stacking sequence and initial spatial occupancy layout to obtain the contact point coordinate sequence and contact type identifier of the target object; Based on the contact point coordinate sequence and contact type identifier, the direction of the supporting force at the contact point of the target object is determined; Based on the weight parameters of the target object and the direction of the supporting force at the contact point, the force value is allocated to the target object to obtain the supporting force vector at the contact point of the target object. By treating the target object as a node and the support force vector at the contact point as a connecting edge, a topological relationship between the node and the connecting edge is constructed to obtain the contact force network of the target object.

[0012] In a preferred embodiment, the step of extrapolating the offset trajectory of the target object's center of gravity based on the analyzed contact force network and identifying the overturning risk points of the target object includes: Based on the support force vectors and mass parameters of the nodes in the contact force network, determine the initial center of gravity position of the target object in the current layout; Virtual displacement perturbation is applied to the contact points in the contact force network, and the force vector offset of the contact points after perturbation is measured to obtain the center of gravity movement path of the target object. Extract the contact contour between the bottom steel coil and the underlying support of the target object; Spatial overlay analysis of the center of gravity movement path and contact contour is performed to identify the trajectory segments in the center of gravity movement path that exceed the contact contour. Mark the spatial location corresponding to the trajectory segment as the overturning risk point of the target object.

[0013] In a preferred embodiment, the step of verifying the support capacity of the target object's contact state based on the distribution characteristics of overturning risk points, in order to adjust the interlayer misalignment angle and the steel coil arrangement order of the target object, includes: Spatial clustering analysis was performed on the overturning risk points to obtain the clustering areas and dominant offset directions of the overturning risk points; Based on the clustering area and the dominant offset direction, and using the preset support force margin stability requirements as a benchmark, the requirement satisfaction status of the contact points within the clustering area is determined, and the support capacity verification conclusion of the target area is obtained. When the support capacity verification result is that the support force margin is insufficient, the target object is configured with an axis rotation angle according to the dominant offset direction to obtain the interlayer misalignment angle of the target object. When the support capability verification conclusion is that the support boundary is unstable, the steel coil pairs to be replaced in the stacking priority sequence are extracted according to the spatial location of the cluster area, and the replacement steel coil arrangement order is generated.

[0014] To address the aforementioned problems, this invention also provides an artificial intelligence-based multi-layer palletizing optimization system for aluminized zinc-coated steel coils, the system comprising: The steel coil feature extraction module is used to extract multi-dimensional features from the steel coil attribute information of the target object, and obtain the geometric adaptation parameters and stackable feature parameters of the target object. The spatial map construction module is used to perform structured spatial analysis of the environmental information of the target object, and construct the palletizing operation spatial map of the target object based on the analyzed three-dimensional grid occupancy status and grid point load-bearing capacity threshold. The initial layout generation module is used to perform spatial optimization configuration of the palletizing operation spatial map based on geometric adaptation parameters and stackable feature parameters, so as to obtain the initial stacking sequence and initial space occupancy layout of the target object. The stability simulation module is used to perform force transmission analysis on the target object based on the initial stacking sequence and initial spatial occupancy layout, and to simulate the offset trajectory of the target object's center of gravity based on the analyzed contact force network, thereby identifying the overturning risk points of the target object. The layout optimization and adjustment module is used to verify the support capacity of the target object's contact state based on the distribution characteristics of overturning risk points, so as to adjust the interlayer misalignment angle and the arrangement order of steel coils of the target object.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention accurately constructs a palletizing operation spatial map by extracting multi-dimensional features of steel coils and analyzing the structured environment space. Based on spatial optimization configuration, it generates an initial stacking sequence and spatial occupancy layout with higher matching degree, which significantly improves the accuracy of palletizing layout planning and operation efficiency.

[0016] 2. This invention uses a contact force network to deduce the trajectory of the center of gravity offset and identify the tipping risk points. By dynamically adjusting the interlayer misalignment angle and the arrangement order of steel coils through support capacity verification, it effectively enhances the stability of the stacking structure, reduces the risk of tipping, and improves the utilization rate of stacking space and operational safety. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an artificial intelligence-based multi-layer palletizing optimization method for aluminized zinc-coated steel coils, provided in an embodiment of the present invention. Figure 2 A functional block diagram of an artificial intelligence-based multi-layer palletizing optimization system for aluminized zinc-coated steel coils provided in an embodiment of the present invention; Figure 3 This is a diagram showing the relationship between interlayer misalignment angle and support force margin provided in an embodiment of the present invention; Figure 4 This is a diagram showing the relationship between spacing and initial contact matching degree provided in an embodiment of the present invention; Figure 5 This is a comparison diagram of the bearing capacity of the steel coil end face and the stacking priority provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides an AI-based method for optimizing the multi-layer palletizing of galvanized steel coils. The execution entity of this AI-based method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the AI-based method for optimizing the multi-layer palletizing of galvanized steel coils can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating an artificial intelligence-based multi-layer palletizing optimization method for galvanized steel coils according to an embodiment of the present invention. In this embodiment, the artificial intelligence-based multi-layer palletizing optimization method for galvanized steel coils includes: Multidimensional feature extraction is performed on the steel coil attribute information of the target object to obtain the geometric adaptation parameters and stackable feature parameters of the target object. In this embodiment of the invention, the multi-dimensional feature extraction of the steel coil attribute information of the target object to obtain the geometric adaptation parameters and stackable feature parameters of the target object includes: Acquire end-face image sequences and 3D point cloud data of the target object; Edge features are extracted from the end face image sequence, and arc segment tracking analysis is performed on the extracted end face edge points to obtain the end face curvature radius data and end face center coordinate data of the target object; Based on 3D point cloud data, surface integrity detection is performed on the end face area of ​​the target object to obtain end face flatness deviation data and end face bearing strength prediction data. The end face curvature radius data, end face center coordinate data, end face flatness deviation data, and end face bearing capacity prediction data are integrated into the geometric adaptation parameters and stackable feature parameters of the target object.

[0021] Multiple frames of end-face images are continuously captured by an image acquisition device to form an end-face image sequence. The entire steel coil is then scanned by a 3D scanning device to obtain 3D point cloud data containing complete spatial coordinate information. During the acquisition process, it is ensured that the end-face image sequence completely covers the entire area of ​​the steel coil end-face, and the 3D point cloud data completely covers the steel coil end-face and surrounding related areas.

[0022] Edge features are extracted from each frame of the end face image sequence. The boundary points between the end face of the steel coil and the background area are identified frame by frame to form end face edge points. Arc segment tracking analysis is carried out sequentially according to the spatial arrangement of the end face edge points. The complete circular outline of the end face of the steel coil is determined along the continuous direction of the end face edge points. The end face curvature radius data of the target steel coil is determined based on the geometric features of the circular outline. The center coordinate data of the end face of the target steel coil is determined based on the center positioning result of the circular outline.

[0023] Surface integrity detection is carried out by selecting all point cloud information corresponding to the end face of the steel coil in the three-dimensional point cloud data. The actual spatial coordinates of each point cloud in the end face area are compared with the preset standard plane coordinates one by one. The coordinate differences of all point clouds are counted to form the end face flatness deviation data of the target steel coil. The load-bearing capacity of the end face is evaluated based on the distribution density and spatial uniformity of the point cloud in the end face area, and the estimated end face compressive strength of the target steel coil is obtained.

[0024] The end face curvature radius data, end face center coordinate data, end face flatness deviation data, and end face compressive strength prediction data are classified and integrated according to feature type. The end face curvature radius data and end face center coordinate data are integrated into the geometric adaptation parameters of the target steel coil, and the end face flatness deviation data and end face compressive strength prediction data are integrated into the stackable feature parameters of the target steel coil.

[0025] The beneficial effects are that by fully acquiring the image sequence and 3D point cloud data of the steel coil end face, the basic information of the steel coil end face can be comprehensively obtained. Through edge feature extraction and arc segment tracking analysis, the end face curvature radius data and end face center coordinate data can be accurately determined, ensuring the accuracy of geometric adaptation parameters. Based on the surface integrity detection of 3D point cloud data, the end face flatness deviation data and end face bearing capacity prediction data can be objectively obtained, ensuring the reliability of stackable feature parameters. The integrated geometric adaptation parameters and stackable feature parameters can fully reflect the geometric adaptation characteristics and stacking load-bearing capacity of the steel coil, providing accurate basic data support for the spatial planning and stability verification of subsequent palletizing operations.

[0026] The environmental information of the target object is analyzed in a structured spatial manner, and a palletizing operation spatial map of the target object is constructed based on the analyzed three-dimensional grid occupancy status and grid point load-bearing capacity threshold. In this embodiment of the invention, the structured spatial parsing of the environmental information of the target object includes: Spatial rasterization is performed on the scene point cloud data of the target object to obtain a three-dimensional spatial raster of the scene point cloud data; Density statistics are performed on the three-dimensional spatial grid to obtain the point cloud distribution of the three-dimensional spatial grid; Based on the point cloud distribution, determine the 3D grid occupancy status of the target object; Spatial registration analysis was performed on the ground bearing capacity data of the target object and the three-dimensional spatial grid to obtain the positional correspondence between the ground bearing capacity data and the three-dimensional spatial grid; Based on the location correspondence, the ground bearing capacity data is mapped to a three-dimensional spatial grid to obtain the load-bearing capacity threshold of the grid points of the target object.

[0027] The step of constructing a palletizing operation space map of the target object based on the parsed 3D grid occupancy status and grid point load-bearing capacity threshold includes: Mark the grid cells marked as free in the 3D grid occupancy status as candidate placement grid cells; Based on the spatial relationship of the candidate placement grids, the candidate placement grids are aggregated by adjacency to obtain the continuous placeable area of ​​the candidate placement grids; Based on the neighborhood grid attributes of the continuous placeable region, the upper limit of the load-bearing capacity of the continuous placeable region is spatially weighted and corrected to obtain the corrected load-bearing threshold of the candidate placement grid. The occupancy identifier, the corrected load-bearing threshold, and the ownership identifier of the continuous placeable area of ​​the candidate placement grid are constructed using attribute tensors to obtain the multidimensional features of the candidate placement grid. Based on the grid spatial coordinates, spatial indexing and association are performed on the multidimensional features to obtain the palletizing operation spatial map of the target object.

[0028] Using a standard fixed space size adapted to the palletizing operation of aluminum-zinc coated steel coils, the entire three-dimensional space covered by the scene point cloud data of the target steel coil in the entire operation scene is uniformly divided into cubic grid units with completely consistent side lengths. During the segmentation process, it is ensured that there is no spatial omission and no grid overlap. After the segmentation is completed, a three-dimensional spatial grid that completely covers the scene point cloud data of the entire operation scene is formed.

[0029] Following the spatial coordinate order of the 3D spatial raster, the number of scene point cloud data contained in each cubic raster unit is counted one by one. The total number of point clouds obtained is defined as the density feature of the raster unit. The density features of all raster units are arranged and integrated according to spatial coordinates to form the complete point cloud distribution of the 3D spatial raster.

[0030] In a scenario where aluminum-zinc coated steel coils are stacked, a minimum number of point clouds is pre-set as the minimum density standard for determining whether a grid cell is occupied by an entity. When the number of point clouds in a single grid cell is greater than or equal to the pre-set density standard, the grid cell is determined to be occupied. When the number of point clouds in a single grid cell is less than the pre-set density standard, the grid cell is determined to be idle. The occupancy and idle status determination results of all grid cells are summarized according to spatial coordinates to form the three-dimensional grid occupancy status of the target steel coil.

[0031] Extract the three-dimensional coordinates of the actual collection location corresponding to each set of ground bearing pressure data. Match and calibrate these coordinates one by one with the three-dimensional spatial coordinates of each grid cell in the three-dimensional spatial grid in the X, Y, and Z dimensions. When the coordinates match perfectly, the set of ground bearing pressure data is determined to belong to the corresponding grid cell. After all ground bearing pressure data matching and calibration are completed, the positional correspondence between the ground bearing pressure data and the three-dimensional spatial grid is accurately obtained.

[0032] Based on the one-to-one correspondence between ground bearing capacity data and three-dimensional spatial grids, the numerical value of each set of ground bearing capacity data is completely assigned to the three-dimensional spatial grid cell to which it belongs. The ground bearing capacity data assigned to each three-dimensional spatial grid cell serves as the maximum weight standard that the grid cell can withstand, which is the load-bearing capacity threshold of the grid point of the target steel coil.

[0033] Following the spatial arrangement order of the 3D spatial grid, all grid cells in the occupied state of the 3D grid are traversed one by one. All grid cells that are clearly identified as free by the occupied marker are uniformly marked as candidate placement grids. The marking operation retains the original 3D spatial coordinate information of each candidate placement grid throughout the process, without modifying any coordinate values ​​or spatial positions.

[0034] For each candidate placement grid, check its six orthogonal directions of up, down, left, right, front, and back. If the adjacent grid cells are also candidate placement grids, it is determined that there is a direct spatial adjacency connection between them. All candidate placement grids with this connection are merged into a complete spatial region. All merged independent spatial regions are the continuous placeable regions of candidate placement grids.

[0035] Extract the load-bearing capacity threshold of all directly adjacent neighboring grates around the continuous placeable area. Calculate the actual spatial distance between each neighboring grates and the geometric center of the continuous placeable area. Assign weighting coefficients from high to low according to the rule of spatial distance from near to far. Multiply the load-bearing threshold of each neighboring grates by the corresponding weighting coefficient and sum them up. Then, fuse them with the original load-bearing upper limit value of the continuous placeable area. After the calculation is completed, the corrected load-bearing threshold of the candidate placement grates is obtained.

[0036] The occupancy identifier of the candidate placement grid, the corrected load-bearing threshold, and the exclusive ownership identifier of the continuous placeable area to which the candidate placement grid belongs are combined in an orderly manner according to the three-dimensional spatial feature dimension to construct a feature set containing multiple attribute information. This feature set is the multidimensional feature of the candidate placement grid.

[0037] All candidate placement grids' multidimensional features are bound one-to-one with their unique three-dimensional spatial coordinates. Spatial index associations that enable rapid positioning and retrieval are established according to the axial order of the three-dimensional spatial coordinates. After integrating all index associations with multidimensional features, a complete spatial map of the target steel coil's stacking operation is formed.

[0038] The beneficial effects are as follows: through standardized 3D spatial grid division and precise point cloud density statistics, the occupancy status of 3D grids in the work scene can be objectively determined, providing a clear basis for palletizing space selection. Through precise registration and numerical mapping between ground bearing capacity data and 3D spatial grids, each grid unit can have a clear load-bearing capacity standard, ensuring the accuracy of subsequent palletizing load-bearing judgment. Through precise marking and adjacency aggregation of candidate placement grids, continuous placeable areas that meet the requirements for steel coil placement can be divided, avoiding the problem of unusable scattered spaces. By correcting the load-bearing threshold through neighborhood weighting, the load-bearing indicators of continuous placeable areas can be made more consistent with the actual working environment. The palletizing operation space map constructed through attribute tensor construction and spatial index association fully integrates core information such as space occupancy, load-bearing capacity, and area affiliation, without data loss or ambiguous descriptions. It provides reproducible and accurately callable complete spatial data support for subsequent optimal configuration of steel coil palletizing space, completely avoiding the problems of unreasonable space planning and load-bearing judgment errors.

[0039] Based on geometric adaptation parameters and stackable feature parameters, spatial optimization configuration is performed on the palletizing operation spatial map to obtain the initial stacking sequence and initial spatial occupancy layout of the target object. In this embodiment of the invention, the step of performing spatial optimization configuration of the palletizing operation spatial map based on geometric adaptation parameters and stackable feature parameters to obtain the initial stacking sequence and initial spatial occupancy layout of the target object includes: Based on the mechanical bearing parameters in the stackable characteristic parameters, the target objects are arranged in a comprehensive mechanical hierarchy to obtain the stacking priority sequence of the target objects. Based on the stacking priority sequence, candidate placement grids that match the current steel coil geometry adaptation parameters are extracted sequentially from the palletizing operation space map; Calculate the initial contact matching degree of the candidate placement grid; Based on the end face flatness deviation data of the target object, the initial contact matching degree is weighted and corrected to obtain the comprehensive adaptation weight of the candidate placement grid. Based on the comprehensive adaptation weight and stacking priority sequence, the spatial occupancy fusion configuration of the palletizing operation spatial map is performed to obtain the initial stacking sequence and initial spatial occupancy layout of the target object.

[0040] The formula for calculating the initial contact matching degree is as follows: ; In the formula, For the first The initial contact matching degree of each candidate placement grid. For the first The radius of curvature of the support profile corresponding to each candidate placement grid. The radius of curvature of the end face of the steel coil to be placed is [value]. This is the minimum spacing between the center of the candidate placement grid and the edge of the already placed steel coil. The preset maximum allowable spacing, This is the preset weight adjustment factor.

[0041] Based on the mechanical bearing capacity parameters in the stackable characteristic parameters, the estimated end-face compressive strength data of all target steel coils is first extracted. This data is used as the sole criterion for determining the mechanical hierarchy. All target steel coils are sorted one by one according to the numerical order of the estimated end-face compressive strength data from high to low. The sorting process strictly follows the mechanical bearing capacity specifications for multi-layer stacking of aluminized zinc steel coils to ensure that the estimated end-face compressive strength data of the steel coil to be placed in the lower layer after sorting is always greater than the estimated end-face compressive strength data of the steel coil to be placed in the upper layer. After the sorting operation of all steel coils is completed, the stacking priority sequence of the target steel coils is finally generated.

[0042] According to the order of steel coil placement determined by the stacking priority sequence, the target steel coil to be placed is selected one by one. The end face curvature radius data in the geometric adaptation parameters of the steel coil is retrieved. All grid cells in the palletizing operation space map are fully traversed. The support contour curvature radius corresponding to each grid is compared with the end face curvature radius data of the steel coil to be placed. Grid cells with completely equal values ​​are selected. The selection process fully preserves the original attribute information such as the spatial coordinates and the corrected load-bearing threshold of the grid. After the selection is completed, candidate placement grids that accurately match the geometric adaptation parameters of the current steel coil are obtained.

[0043] First, calculate the sum of the radius of curvature of the support profile corresponding to the candidate placement grid and the radius of curvature of the end face of the steel coil to be placed. Then, calculate the difference between the two and take the absolute value. Divide the absolute value by the sum to obtain the relative difference value. Divide 1 by 1 and the sum of the relative difference value to obtain the result of the curvature radius matching degree. The smaller the relative difference value, the higher the curvature radius matching degree. Next, calculate the ratio of the minimum distance between the center of the candidate placement grid and the edge of the placed steel coil to the preset maximum allowable distance. Multiply this ratio by the preset weight adjustment factor to obtain the distance influence value. Perform natural exponential operation on the negative distance influence value to obtain the result of the distance fit degree. The smaller this ratio, the higher the distance fit degree. Multiply the curvature radius matching degree calculation result with the distance fit degree calculation result to finally obtain the initial contact matching degree of the candidate placement grid. The closer the radius of curvature of the support profile and the radius of curvature of the end face of the steel coil are and the smaller the minimum distance between the candidate placement grid and the placed steel coil, the higher the initial contact matching degree value.

[0044] The end-face flatness deviation data is accurately extracted from the stackable characteristic parameters of the target steel coil. This data is used as a fixed benchmark for weighted correction. The smaller the value of the end-face flatness deviation data, the higher the placement fit and stability of the steel coil end face. The actual value of the end-face flatness deviation data is used as the weighted adjustment coefficient to accurately calculate the initial contact matching degree. The adjustment process ensures that the correction result directly reflects the actual flatness state of the steel coil end face. After the adjustment is completed, the comprehensive adaptation weight of the candidate placement grid is obtained.

[0045] All candidate placement grids that match the current steel coil are sorted in descending order of their comprehensive adaptation weight values. After sorting, the candidate placement grid with the highest comprehensive adaptation weight value is selected as the placement position of the steel coil. Combined with the steel coil placement order determined by the stacking priority sequence, each target steel coil is bound to the selected candidate placement grid in turn, completing the spatial occupancy allocation and position confirmation of all steel coils. The spatial occupancy fusion configuration of the palletizing operation spatial map is completed. After the configuration is completed, the initial stacking sequence and initial spatial occupancy layout of the target steel coil are finally obtained.

[0046] The beneficial effect is that the stacking priority sequence generated based on mechanical bearing parameters and estimated end-face compressive strength data perfectly matches the mechanical bearing logic of multi-layer stacking of aluminized zinc steel coils, ensuring the basic mechanical stability of the stacking structure from the root. By accurately matching candidate placement grids with end-face curvature radius data, it ensures a perfect geometric fit between the steel coil and the placement position, avoiding placement offset problems caused by geometric mismatch. The initial contact matching degree is calculated through two dimensions: curvature radius matching degree and spacing adaptation degree. This objectively quantifies the placement adaptability of candidate placement grids. The calculation process is clear, reproducible, and without ambiguity. Weighted correction of end-face flatness deviation data allows the comprehensive adaptation weight to fully combine the actual surface condition of the steel coil, improving the authenticity and accuracy of the adaptation judgment. The spatial occupancy fusion configuration based on the comprehensive adaptation weight sorting and stacking priority sequence can realize the efficient use of palletizing space and the optimal allocation of steel coil placement positions. The generated initial stacking sequence and initial spatial occupancy layout data are complete and well-founded, providing accurate and reliable basic layout support for subsequent force transmission analysis, overturning risk identification and layout optimization adjustment, and comprehensively improving the accuracy, scientificity and operational safety of multi-layer palletizing planning of aluminized zinc steel coils.

[0047] Based on the initial stacking sequence and initial spatial occupancy layout, force transmission analysis is performed on the target object, and based on the analyzed contact force network, the offset trajectory of the target object's center of gravity is deduced to identify the overturning risk points of the target object. In this embodiment of the invention, the step of performing force transmission analysis on the target object based on the initial stacking sequence and the initial spatial occupancy layout includes: Spatial geometry analysis is performed on the initial stacking sequence and initial spatial occupancy layout to obtain the contact point coordinate sequence and contact type identifier of the target object; Based on the contact point coordinate sequence and contact type identifier, the direction of the supporting force at the contact point of the target object is determined; Based on the weight parameters of the target object and the direction of the supporting force at the contact point, the force value is allocated to the target object to obtain the supporting force vector at the contact point of the target object. By treating the target object as a node and the support force vector at the contact point as a connecting edge, a topological relationship between the node and the connecting edge is constructed to obtain the contact force network of the target object.

[0048] Based on the analyzed contact force network, the trajectory of the target object's center of gravity shift is deduced, and the overturning risk points of the target object are identified, including: Based on the support force vectors and mass parameters of the nodes in the contact force network, determine the initial center of gravity position of the target object in the current layout; Virtual displacement perturbation is applied to the contact points in the contact force network, and the force vector offset of the contact points after perturbation is measured to obtain the center of gravity movement path of the target object. Extract the contact contour between the bottom steel coil and the underlying support of the target object; Spatial overlay analysis of the center of gravity movement path and contact contour is performed to identify the trajectory segments in the center of gravity movement path that exceed the contact contour. Mark the spatial location corresponding to the trajectory segment as the overturning risk point of the target object.

[0049] The spatial arrangement relationship and spatial location information of all steel coils in the initial stacking sequence are fully decomposed. The contact positions between each pair of adjacent steel coils and between the steel coil and the lower support are located one by one. The three-dimensional spatial coordinates of each contact position are obtained and arranged in order to form the contact point coordinate sequence of the target steel coil. The contact form is determined to be line contact or surface contact according to the contact surface shape of the contact position. The determination result is marked as the contact type identifier of the target steel coil.

[0050] The spatial orientation of each contact point is determined based on the contact point coordinate sequence. The force characteristics of line contact and surface contact are distinguished by the contact type identifier. In the case of line contact, the direction of the support force is perpendicular to the tangent of the contact line. In the case of surface contact, the direction of the support force is perpendicular to the contact plane and upward. This determines the unique direction of the support force at each contact point of the target steel coil.

[0051] The weight parameters of the target steel coil are retrieved. According to the contact area ratio and spatial force ratio of each contact point, the total weight of the steel coil is distributed to each contact point to obtain the magnitude of the support force at each contact point. The magnitude of the support force is combined with the corresponding support force direction to generate the support force vector at each contact point of the target steel coil.

[0052] Each target steel coil in the initial stacking sequence is defined as an independent node, and the support force vector at the contact point corresponding to each node is defined as the connecting edge connecting adjacent nodes. A one-to-one topological relationship between nodes and connecting edges is established according to the stacking level and contact relationship of the steel coils. All topological relationships are combined to form the contact force network of the target steel coil.

[0053] Extract the support force vector and the mass parameters of the target steel coil corresponding to each node in the contact force network. Based on the position, direction and magnitude of each support force vector, and combined with the mass distribution characteristics of the steel coil, calculate and determine the overall three-dimensional center of gravity coordinates of the target steel coil under the current initial spatial occupancy layout. These coordinates are the initial center of gravity position of the target steel coil.

[0054] Virtual displacement disturbances are applied sequentially to all contact points in the contact force network according to a preset small displacement amount to simulate the slight positional shift that may occur during the placement of the steel coil. After the disturbance, the magnitude and direction of the support force vector of each contact point are re-measured. The center of gravity position is updated sequentially according to the force vector change. All updated center of gravity positions are connected in chronological order to form the complete center of gravity movement path of the target steel coil.

[0055] Locate the contact area between the bottom layer steel coil and the lower support in the initial spatial layout, collect the three-dimensional coordinates of the outer boundary of the contact area, and stitch the outer boundary coordinates in a continuous sequence to form a complete contact profile between the bottom layer steel coil and the lower support.

[0056] All coordinate points along the center of gravity movement path are compared one by one with the boundary coordinates of the contact contour in three-dimensional space to determine whether each coordinate point on the center of gravity movement path falls within the boundary range of the contact contour. Continuous trajectory segments in the center of gravity movement path where the coordinate points fall completely outside the boundary of the contact contour are selected, and these trajectory segments are the trajectory segments that exceed the contact contour.

[0057] All three-dimensional spatial positions corresponding to the trajectory segments that exceed the contact contour are marked one by one and their coordinate information is recorded. All marked spatial positions are the overturning risk points of the target steel coil.

[0058] The beneficial effects are as follows: by conducting precise spatial geometric analysis of the initial stacking sequence and initial spatial occupancy layout, the coordinate sequence of contact points and contact type identification can be completely obtained, providing accurate basic data for force transmission analysis. Based on the contact characteristics, the direction of the support force can be determined and the force value allocation can be completed, which can realistically restore the stress state after the steel coils are stacked. The constructed contact force network can clearly present the force transmission relationship between steel coils, making the force analysis more logical and accurate. Based on the contact force network and mass parameters, the initial center of gravity position can be determined to ensure the authenticity of the center of gravity calculation. The center of gravity movement path obtained by virtual displacement disturbance and force vector offset measurement can simulate the changes in the center of gravity in actual operation. By extracting the contact contour and conducting spatial stacking analysis, the trajectory segment where the center of gravity exceeds the support range can be accurately identified, thereby accurately locating the overturning risk point. This provides a clear risk basis for the stability verification and optimization adjustment of the subsequent stacking layout, effectively improving the structural safety and stability of multi-layer stacking of galvanized steel coils.

[0059] Based on the distribution characteristics of overturning risk points, the support capacity of the target object's contact state is verified in order to adjust the interlayer misalignment angle and the arrangement order of the steel coils.

[0060] In this embodiment of the invention, the step of verifying the support capacity of the target object's contact state based on the distribution characteristics of overturning risk points, in order to adjust the interlayer misalignment angle and the steel coil arrangement order of the target object, includes: Spatial clustering analysis was performed on the overturning risk points to obtain the clustering areas and dominant offset directions of the overturning risk points; Based on the clustering area and the dominant offset direction, and using the preset support force margin stability requirements as a benchmark, the requirement satisfaction status of the contact points within the clustering area is determined, and the support capacity verification conclusion of the target area is obtained. When the support capacity verification result is that the support force margin is insufficient, the target object is configured with an axis rotation angle according to the dominant offset direction to obtain the interlayer misalignment angle of the target object. When the support capability verification conclusion is that the support boundary is unstable, the steel coil pairs to be replaced in the stacking priority sequence are extracted according to the spatial location of the cluster area, and the replacement steel coil arrangement order is generated.

[0061] All the marked overturning risk points are summarized in three-dimensional space coordinates. The overturning risk points are grouped according to the preset spatial distance judgment threshold. Overturning risk points with a spatial distance less than the preset spatial distance judgment threshold are divided into the same group. The continuous spatial range corresponding to each group is the cluster area of ​​overturning risk points. The trend of the change of the center of gravity offset coordinate of all overturning risk points in each cluster area is statistically analyzed. The offset direction with the highest frequency is determined as the dominant offset direction corresponding to the cluster area.

[0062] A minimum support force value that must be met by the contact points of the stacked aluminum-zinc coated steel coils is preset as the preset support force margin stability requirement. The actual support force value of all contact points within the cluster of overturning risk points is tested one by one. Contact points with actual support force values ​​greater than or equal to the preset minimum support force value are judged as meeting the requirements, and contact points with actual support force values ​​less than the preset minimum support force value are judged as not meeting the requirements. Based on the proportion of contact points in the cluster that meet the requirements, a support capacity verification conclusion for the target area is formed.

[0063] When the support capacity verification result indicates that the support force margin is insufficient, the main offset direction corresponding to the overturning risk point is used as a reference. The target steel coil is rotated and adjusted in the opposite direction to the main offset direction along the central axis. The rotation adjustment angle is determined by ensuring that the support force at the contact point meets the preset support force margin stability requirements. After the rotation adjustment is completed, the interlayer misalignment angle of the target steel coil is determined.

[0064] When the support capacity verification conclusion determines that the support boundary is unstable, locate the stacking level and spatial position of the steel coils corresponding to the area where the overturning risk point is concentrated. Select the steel coils whose bearing capacity in the stacking level does not match the current position from the stacking priority sequence as the steel coils to be replaced. Then select the steel coils in the sequence whose bearing capacity meets the load-bearing requirements of the position to form the steel coil pairs to be replaced. After swapping the arrangement positions of the steel coil pairs to be replaced, the steel coil arrangement order after replacement is generated.

[0065] The beneficial effects are as follows: by conducting spatial clustering analysis on overturning risk points, the risk clustering area can be accurately divided and the dominant offset direction can be determined, providing a clear analytical object and directional basis for support capacity verification. Verification based on the preset support force margin stability requirements can objectively determine the support status of the contact points, clearly distinguish between the two core problem types of insufficient support force margin and support boundary instability. For insufficient support force margin, the interlayer misalignment angle obtained by configuring the axis rotation angle can accurately supplement the support force margin of the contact points and quickly enhance the local support stability. For support boundary instability, the steel coil arrangement order can be replaced to optimize the mechanical matching relationship of the stacking layers and effectively repair the instability state of the support boundary. Targeted adjustments are carried out throughout the process based on risk characteristics to achieve dynamic optimization of the stacking layout, significantly reduce the overturning probability of multi-layer stacking of galvanized steel coils, and comprehensively improve the overall stability of the stacking structure and the safety level of on-site operations.

[0066] like Figure 2 The diagram shown is a functional block diagram of an artificial intelligence-based multi-layer palletizing optimization system for aluminum-zinc coated steel coils provided in an embodiment of the present invention.

[0067] The AI-based multi-layer palletizing optimization system 100 for galvanized steel coils described in this invention can be installed in an electronic device. Depending on the functions implemented, the AI-based multi-layer palletizing optimization system 100 for galvanized steel coils may include a steel coil feature extraction module 101, a spatial map construction module 102, an initial layout generation module 103, a stability deduction module 104, and a layout optimization adjustment module 105. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and is stored in the memory of the electronic device.

[0068] In this embodiment, the functions of each module / unit are as follows: The steel coil feature extraction module 101 is used to extract multi-dimensional features from the steel coil attribute information of the target object to obtain the geometric adaptation parameters and stackable feature parameters of the target object. The spatial map construction module 102 is used to perform structured spatial analysis of the environmental information of the target object, and construct the palletizing operation spatial map of the target object based on the analyzed three-dimensional grid occupancy status and grid point load-bearing capacity threshold. The initial layout generation module 103 is used to perform spatial optimization configuration of the palletizing operation spatial map based on geometric adaptation parameters and stackable feature parameters, so as to obtain the initial stacking sequence and initial space occupancy layout of the target object. The stability simulation module 104 is used to perform force transmission analysis on the target object based on the initial stacking sequence and initial spatial occupancy layout, and to simulate the offset trajectory of the target object's center of gravity based on the analyzed contact force network, thereby identifying the overturning risk points of the target object. The layout optimization and adjustment module 105 is used to verify the support capacity of the target object's contact state based on the distribution characteristics of the overturning risk points, so as to adjust the interlayer misalignment angle and the arrangement order of the steel coils of the target object.

[0069] like Figure 3 The diagram shown illustrates the relationship between interlayer misalignment angle and support force margin in an embodiment of the present invention. It visually presents the dynamic correlation between interlayer misalignment angle and support force margin in multi-layer stacking of aluminized zinc steel coils. The interlayer misalignment angle is the rotation angle parameter of the upper and lower layers of the steel coil along the axis, and the support force margin is the difference between the actual support force at the contact point and the safety threshold, which is the core indicator for determining the stability of the stacking.

[0070] like Figure 4 The diagram shown illustrates the relationship between spacing and initial contact matching degree according to an embodiment of the present invention. It clearly reveals the quantitative correlation between spacing and initial contact matching degree, providing core data support for the design of contact-type structures. From the overall trend, all four sets of experimental data show a monotonically decreasing characteristic of initial contact matching degree as the spacing increases. This indicates that spacing expansion is the key factor leading to a decrease in the degree of contact fit. The closer the matching degree value is to 1, the better the fit and stress distribution uniformity of the contact interface.

[0071] like Figure 5 The figure shown is a comparison chart of the end-face bearing strength and stacking priority of steel coils according to an embodiment of the present invention. It intuitively presents the difference in the distribution of the original end-face bearing strength and the stacking priority after sorting: the blue original curve fluctuates violently, with the bearing strength oscillating greatly in the range of 1100N~1300N, reflecting the disorder of load distribution when stacking randomly, which easily leads to overloading of low-strength steel coils and waste of the bearing capacity of high-strength steel coils; the red sorted curve shows a monotonically decreasing trend, which is the result of re-sorting the steel coils according to their bearing strength from high to low. This figure clearly verifies the value of orderly stacking. By distributing the bearing load in a gradient manner, the risk of stress concentration can be eliminated, the bearing potential of high-strength steel coils can be maximized, and low-strength steel coils can be protected at the same time, significantly improving the stability and safety of stacking, and providing an intuitive quantitative basis for optimizing the storage and stacking of steel coils.

[0072] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0073] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0074] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0075] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0076] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for multi-layer palletizing optimization of galvannealed steel coils based on artificial intelligence, characterized in that, The method includes: Multidimensional feature extraction is performed on the steel coil attribute information of the target object to obtain the geometric adaptation parameters and stackable feature parameters of the target object. The environmental information of the target object is analyzed in a structured spatial manner, and a palletizing operation spatial map of the target object is constructed based on the analyzed three-dimensional grid occupancy status and grid point load-bearing capacity threshold. Based on geometric adaptation parameters and stackable feature parameters, spatial optimization configuration is performed on the palletizing operation spatial map to obtain the initial stacking sequence and initial spatial occupancy layout of the target object. Based on the initial stacking sequence and initial spatial occupancy layout, force transmission analysis is performed on the target object, and based on the analyzed contact force network, the offset trajectory of the target object's center of gravity is deduced to identify the overturning risk points of the target object. Based on the distribution characteristics of overturning risk points, the support capacity of the target object's contact state is verified in order to adjust the interlayer misalignment angle and the arrangement order of the steel coils.

2. The artificial intelligence-based multi-layer palletizing optimization method for galvannealed steel coils according to claim 1, characterized in that, The multi-dimensional feature extraction of the steel coil attribute information of the target object yields the geometric adaptation parameters and stackable feature parameters of the target object, including: Acquire end-face image sequences and 3D point cloud data of the target object; Edge features are extracted from the end face image sequence, and arc segment tracking analysis is performed on the extracted end face edge points to obtain the end face curvature radius data and end face center coordinate data of the target object; Based on 3D point cloud data, surface integrity detection is performed on the end face area of ​​the target object to obtain end face flatness deviation data and end face bearing strength prediction data. The end face curvature radius data, end face center coordinate data, end face flatness deviation data, and end face bearing capacity prediction data are integrated into the geometric adaptation parameters and stackable feature parameters of the target object.

3. The artificial intelligence-based multi-layer palletizing optimization method for galvannealed steel coils according to claim 1, characterized in that, The structured spatial parsing of the environmental information of the target object includes: Spatial rasterization is performed on the scene point cloud data of the target object to obtain a three-dimensional spatial raster of the scene point cloud data; Density statistics are performed on the three-dimensional spatial grid to obtain the point cloud distribution of the three-dimensional spatial grid; Based on the point cloud distribution, determine the 3D grid occupancy status of the target object; Spatial registration analysis was performed on the ground bearing capacity data of the target object and the three-dimensional spatial grid to obtain the positional correspondence between the ground bearing capacity data and the three-dimensional spatial grid; Based on the location correspondence, the ground bearing capacity data is mapped to a three-dimensional spatial grid to obtain the load-bearing capacity threshold of the grid points of the target object.

4. The artificial intelligence-based multi-layer palletizing optimization method for galvannealed steel coils according to claim 1, wherein, The step of constructing a palletizing operation space map of the target object based on the parsed 3D grid occupancy status and grid point load-bearing capacity threshold includes: Mark the grid cells marked as free in the 3D grid occupancy status as candidate placement grid cells; Based on the spatial relationship of the candidate placement grids, the candidate placement grids are aggregated by adjacency to obtain the continuous placeable area of ​​the candidate placement grids; Based on the neighborhood grid attributes of the continuous placeable region, the upper limit of the load-bearing capacity of the continuous placeable region is spatially weighted and corrected to obtain the corrected load-bearing threshold of the candidate placement grid. The occupancy identifier, the corrected load-bearing threshold, and the ownership identifier of the continuous placeable area of ​​the candidate placement grid are constructed using attribute tensors to obtain the multidimensional features of the candidate placement grid. Based on the grid spatial coordinates, spatial indexing and association are performed on the multidimensional features to obtain the palletizing operation spatial map of the target object.

5. The artificial intelligence-based multi-layer palletizing optimization method for galvannealed steel coils according to claim 1, wherein, The process of spatially optimizing the palletizing operation spatial map based on geometric adaptation parameters and stackable feature parameters to obtain the initial stacking sequence and initial spatial occupancy layout of the target object includes: Based on the mechanical bearing parameters in the stackable characteristic parameters, the target objects are arranged in a comprehensive mechanical hierarchy to obtain the stacking priority sequence of the target objects. Based on the stacking priority sequence, candidate placement grids that match the current steel coil geometry adaptation parameters are extracted sequentially from the palletizing operation space map; Calculate the initial contact matching degree of the candidate placement grid; Based on the end face flatness deviation data of the target object, the initial contact matching degree is weighted and corrected to obtain the comprehensive adaptation weight of the candidate placement grid. Based on the comprehensive adaptation weight and stacking priority sequence, the spatial occupancy fusion configuration of the palletizing operation spatial map is performed to obtain the initial stacking sequence and initial spatial occupancy layout of the target object.

6. The artificial intelligence-based multi-layer palletizing optimization method for aluminized zinc-coated steel coils as described in claim 5, characterized in that, The formula for calculating the initial contact matching degree is as follows: ； In the formula, For the first The initial contact matching degree of each candidate placement grid. For the first The radius of curvature of the support profile corresponding to each candidate placement grid. The radius of curvature of the end face of the steel coil to be placed is [missing information]. This is the minimum spacing between the center of the candidate placement grid and the edge of the already placed steel coil. The preset maximum allowable spacing, This is the preset weight adjustment factor.

7. The artificial intelligence-based multi-layer palletizing optimization method for aluminized zinc-coated steel coils as described in claim 1, characterized in that, The step of performing force transmission analysis on the target object based on the initial stacking sequence and initial spatial occupancy layout includes: Spatial geometry analysis is performed on the initial stacking sequence and initial spatial occupancy layout to obtain the contact point coordinate sequence and contact type identifier of the target object; Based on the contact point coordinate sequence and contact type identifier, the direction of the supporting force at the contact point of the target object is determined; Based on the weight parameters of the target object and the direction of the supporting force at the contact point, the force value is allocated to the target object to obtain the supporting force vector at the contact point of the target object. By treating the target object as a node and the support force vector at the contact point as a connecting edge, a topological relationship between the node and the connecting edge is constructed to obtain the contact force network of the target object.

8. The artificial intelligence-based multi-layer palletizing optimization method for aluminized zinc-coated steel coils as described in claim 1, characterized in that, Based on the analyzed contact force network, the trajectory of the target object's center of gravity shift is deduced, and the overturning risk points of the target object are identified, including: Based on the support force vectors and mass parameters of the nodes in the contact force network, determine the initial center of gravity position of the target object in the current layout; Virtual displacement perturbation is applied to the contact points in the contact force network, and the force vector offset of the contact points after perturbation is measured to obtain the center of gravity movement path of the target object. Extract the contact contour between the bottom steel coil and the underlying support of the target object; Spatial overlay analysis of the center of gravity movement path and contact contour is performed to identify the trajectory segments in the center of gravity movement path that exceed the contact contour. Mark the spatial location corresponding to the trajectory segment as the overturning risk point of the target object.

9. The artificial intelligence-based multi-layer palletizing optimization method for aluminized zinc-coated steel coils as described in claim 1, characterized in that, The method of verifying the support capacity of the target object's contact state based on the distribution characteristics of overturning risk points, in order to adjust the interlayer misalignment angle and the steel coil arrangement order of the target object, includes: Spatial clustering analysis was performed on the overturning risk points to obtain the clustering areas and dominant offset directions of the overturning risk points; Based on the clustering area and the dominant offset direction, and using the preset support force margin stability requirements as a benchmark, the requirement satisfaction status of the contact points within the clustering area is determined, and the support capacity verification conclusion of the target area is obtained. When the support capacity verification result is that the support force margin is insufficient, the target object is configured with an axis rotation angle according to the dominant offset direction to obtain the interlayer misalignment angle of the target object. When the support capability verification conclusion is that the support boundary is unstable, the steel coil pairs to be replaced in the stacking priority sequence are extracted according to the spatial location of the cluster area, and the replacement steel coil arrangement order is generated.

10. An AI-based multi-layer palletizing optimization system for aluminized zinc-coated steel coils, characterized in that: The system for implementing the AI-based multi-layer palletizing optimization method for aluminized zinc-coated steel coils as described in claim 1 includes: The steel coil feature extraction module is used to extract multi-dimensional features from the steel coil attribute information of the target object, and obtain the geometric adaptation parameters and stackable feature parameters of the target object. The spatial map construction module is used to perform structured spatial analysis of the environmental information of the target object, and construct the palletizing operation spatial map of the target object based on the analyzed three-dimensional grid occupancy status and grid point load-bearing capacity threshold. The initial layout generation module is used to perform spatial optimization configuration of the palletizing operation spatial map based on geometric adaptation parameters and stackable feature parameters, so as to obtain the initial stacking sequence and initial space occupancy layout of the target object. The stability simulation module is used to perform force transmission analysis on the target object based on the initial stacking sequence and initial spatial occupancy layout, and to simulate the offset trajectory of the target object's center of gravity based on the analyzed contact force network, thereby identifying the overturning risk points of the target object. The layout optimization and adjustment module is used to verify the support capacity of the target object's contact state based on the distribution characteristics of overturning risk points, so as to adjust the interlayer misalignment angle and the arrangement order of steel coils of the target object.

Images