A panel unit division method and cutting method based on skeleton

Abstract

The present invention discloses a panel unit division method and blanking method based on a skeleton, the division method comprises the following steps: step 1: extracting each beam component of the skeleton, the types of beam components are divided into ring beams, main beams and secondary beams, and extracting the axis points and cross-sectional corner points of the beam components; step 2: extracting the beam intersections of each beam component projected on the surface of the skeleton, the beam intersections refer to the intersections formed by the intersection of a beam component and another beam component, and determining the types of the two beam components constituting the beam intersections, based on a plurality of adjacent beam intersections to form a polygon, each beam intersection is used as a vertex of the polygon, each polygon is used as a panel unit, the panel unit division is completed, and the panel unit division result is obtained. The present invention does not need to rely on manual division, the division efficiency is higher and more accurate, and the panel division result can also be given for blanking, which is conducive to saving material costs.

Description

Technical Field

[0001] The invention relates to the technical field of building engineering construction, and in particular to a panel unit division method and a material cutting method based on a skeleton. Background Art

[0002] In order to incorporate cultural elements and provide architectural visual effects, modern buildings, especially some large public buildings, will adopt curved skylights, curtain walls and other spatial curved structures. The curved skylights are usually made of glass panels, one piece of glass constitutes a panel unit, and multiple panel units are spread over the keel. In this type of building, after the construction of the keel (that is, the bracket) with a curved structure on the top of the building is completed, it is necessary to design the glass panel unit on the bracket, and lay the panel unit (and glass) on the bracket through the beam component. If the panel unit is manually extracted, the efficiency is very low, and the rationality of the panel unit cutting and arrangement plan directly affects the material cost. For this reason, a method is needed that can intelligently, efficiently and automatically complete the panel unit division and can complete the laying of the panel unit on the keel with a more reasonable cutting distribution plan based on the panel unit division result. Summary of the invention

[0003] In view of the deficiencies in the prior art, an object of the present invention is to provide a panel unit division method and a cutting method based on a skeleton, which can solve the technical problems described in the background technology.

[0004] The technical solution to achieve the purpose of the present invention is: a panel unit division method based on a skeleton, comprising the following steps:

[0005] Step 1: Obtain point cloud data of the target skeleton surface, extract the various beam components of the skeleton based on the point cloud data, and classify the various beam components into corresponding types. The types of beam components are divided into ring beams, main beams and secondary beams, and extract the axis points and cross-sectional corner points of the beam components. The axis points refer to the points obtained after the axis of the beam component is discretized, and the cross-sectional corner points refer to the vertices of the cross section obtained by slicing the beam component at the axis points in a direction perpendicular to the axis.

[0006] Step 2: Extract the beam intersections of each beam component projected on the skeleton surface. The beam intersection refers to the intersection formed by the intersection of a beam component and another beam component, and determine the types of the two beam components that constitute the beam intersection. A polygon is formed based on several adjacent beam intersections. Each beam intersection is used as a vertex of the polygon, and each polygon is used as a panel unit. The panel unit division is completed to obtain the panel unit division result.

[0007] Further, in Step 1, the beam member has a polygon cross-section. The beam member is sliced to obtain a polygon cross-section, and the vertices of the cross-section are cross-section intersection points. One axis point corresponds to one cross-section, and each cross-section includes multiple vertices.

[0008] Further, the specific implementation process of Step 2 includes the following steps:

[0009] Step 21: Traverse each secondary beam and each primary beam, and sort the axis points of the same secondary beam and the same primary beam respectively.

[0010] Step 22: Preset a global minimum distance threshold M and a maximum traversal times threshold N. M is an integer greater than 1, and N is the number of axis points of a single secondary beam.

[0011] Step 221: Traverse the sorted axis points of the secondary beam, and search for the axis point of the primary beam that is closest to the currently traversed axis point of the secondary beam through the nearest neighbor algorithm. If the distance D between the currently traversed axis point of the secondary beam and the closest axis point of the primary beam is D < M, then update M with the value of D, that is, use the current D value as the updated M, and continue to traverse the axis points of the secondary beam until the distance D between the currently traversed axis point of the secondary beam and the closest axis point of the primary beam is D ≥ M. Exit the traversal, output the axis point of the secondary beam corresponding to the last updated M value, and output the previous axis point and the next axis point of this axis point of the secondary beam as an adjacent point pair. If the number of traversal times of the axis points of the secondary beam reaches N, that is, when all axis points of the secondary beam are traversed, and the M value has not been updated yet, it indicates that there are no potential intersection points between the current secondary beam and the current primary beam, and no output is made. If the exit traversal condition is met when traversing the second axis point of the secondary beam, output the first two axis points of the secondary beam as an adjacent point pair.

[0012] Step 222: For the axis points of each remaining primary beam, repeat Step 221 until the axis points of each primary beam are traversed, so as to find the adjacent point pairs between the current secondary beam and all primary beams.

[0013] Step 223: For the axis points of each remaining secondary beam, repeat Steps 221 - 222 until the axis points of each secondary beam are traversed, so as to find the adjacent point pairs between all secondary beams and all primary beams.

[0014] Step 224: According to the adjacent point pairs corresponding to each current axis point of the secondary beam, fit a first straight line. The XY coordinates of each adjacent point pair can fit a corresponding first straight line.

[0015] Step 23: For the current first straight line, traverse the main beam axis points, extract the adjacent point pairs of the main beam axis points located on both sides of the first straight line, so as to find the adjacent point pairs in the main beam axis points corresponding to the current first straight line, fit the second straight line corresponding to the current main beam axis point based on the adjacent point pairs in the current main beam axis points, and continue to traverse the remaining first straight lines, so as to fit the corresponding second straight line based on the adjacent point pairs in the main beam axis points for each straight line, that is, the second straight line is fitted based on the XY coordinates of the adjacent point pairs;

[0016] Step 24: traverse the adjacent point pairs of each secondary beam axis point, if the adjacent point pairs of the secondary beam axis point are located on both sides of the second straight line, the secondary beam where the current secondary beam axis point is located intersects with the main beam corresponding to the second straight line, and a third straight line is fitted based on the XY coordinates of the adjacent secondary beam axis points;

[0017] Step 25: Calculate the intersection of the second straight line and the third straight line, and interpolate the Z coordinate of the intersection based on the coordinates of the adjacent point pair of the main beam axis point in step 23, so as to obtain the three-dimensional coordinates of the intersection, which also includes the XYZ coordinates;

[0018] Step 26: Take the direction of the line between the adjacent point pairs of the main beam axis points as the direction vector of the three-dimensional intersection point, take the average value of the two cross-sectional dimensions at the positions of the adjacent point pairs of the main beam axis points as the cross-sectional dimension of the three-dimensional intersection point, the cross-sectional dimension is generated by the corresponding cross-sectional node, and according to the direction of the three-dimensional intersection point and the cross-sectional dimension, translate the three-dimensional intersection point to the upper surface of the main beam, and traverse each main beam axis point to obtain all the intersection points of the main beam and the secondary beam, and obtain the three-dimensional intersection points of all the intersection points and translate the three-dimensional intersection points to the upper surface of the main beam;

[0019] Step 27: Take two axis points from the starting point of the secondary beam axis point, the end point of the secondary beam axis point, the starting point of the main beam axis point, and the end point of the main beam axis point according to the size of the coordinate values, calculate the fourth straight line according to the XY coordinates of the taken axis points, calculate the nearest ring beam axis points located on both sides of the fourth straight line, calculate the fifth straight line according to the XY coordinates of the nearest ring beam axis point, and process the nearest ring beam axis point according to the method of steps 25-26, so as to obtain the intersection of the ring beam and the main beam projected to the upper surface of the ring beam, and the intersection of the ring beam and the secondary beam, so as to obtain polygons based on the intersections of the main beam and the secondary beam, the main beam and the ring beam, and the secondary beam and the ring beam, so as to obtain panel units and the panel unit division results.

[0020] Further, in step 21, the axis points of the same secondary beam are sorted from small to large according to the X coordinate, and the axis points of the same main beam are sorted from small to large according to the Y coordinate.

[0021] Furthermore, M=1000.

[0022] Further, in step 25, the intersection is on both the second straight line and the third straight line, and the Z coordinates of the adjacent point pairs of the main beam axis points are interpolated based on the intersection, so that the intersection and the adjacent point pairs are also on the same straight line in three-dimensional space.

[0023] A material cutting method based on the panel division unit result, after executing step 2, further includes the following steps:

[0024] Step 31: construct a first matrix with a dimension of A×B, where A represents the number of secondary beams in the keel, and B represents the number of main beams in the keel. The elements in the first matrix represent whether the main beam and the secondary beam intersect. If the main beam and the secondary beam intersect and have an intersection, the corresponding element is assigned a value of 1; if the main beam and the secondary beam do not intersect and have no intersection, the corresponding element is assigned a value of 0;

[0025] Step 32: Each row and column of the first matrix is ​​expanded by a matrix with an element of 0, that is, the expansion dimension is , and record it as the second matrix;

[0026] Step 33: Use the dimension A convolution kernel whose elements are all 1 performs a convolution operation on the second matrix to obtain a convolution result, and different types of polygons are determined according to the convolution result. Different types of polygons correspond to different panel units, and the type of polygon is determined by the number of edges of the polygon;

[0027] Step 34: Among the polygons determined in step 33, any polygon whose side length is less than 2 times the width of the beam component is merged with an adjacent panel unit to obtain a merged polygon. The merged polygon and the unmerged polygon each correspond to a panel unit. The minimum circumscribed rectangle of each panel unit is calculated, and the position that is 1 cm inward in the vertical direction of the minimum circumscribed rectangle is used as the position of the blanking element. The blanking element is the material that constitutes the panel unit.

[0028] Further, in step 33, various different convolution results are divided into corresponding panel units, wherein the convolution result is 1, that is, there is only 1 element value of 1, and the divided panel unit represents a triangle formed by a ring beam, a main beam and a secondary beam; the convolution result is 2, that is, the number of element values ​​1 is 2, and the divided panel unit represents a quadrilateral formed by a ring beam, two main beams and a secondary beam, or a ring beam, two secondary beams and a main beam; the convolution result is 3, and the divided panel unit represents a pentagon formed by a ring beam, two main beams and two secondary beams; the convolution result is 4, and the divided panel unit represents a quadrilateral formed by the intersection of two main beams and two secondary beams.

[0029] Furthermore, after step 34, step 35 is also included:

[0030] Step 35: Use a heuristic algorithm to generate a layout plan for blanking components, and calculate the waste rate of each layout plan for blanking components with different strategies under the heuristic algorithm, and perform construction blanking with the layout plan for blanking components with the smallest waste rate.

[0031] The beneficial effects of the present invention are as follows: the present invention can quickly and efficiently divide the panel units on the keel by software programs, without relying on manual division, with higher efficiency and accuracy, and can also cut materials based on the panel division results, which is conducive to saving material costs. Finally, the automatic division of the spatial curved surface structure panel and the intelligent generation of a reasonable cutting plan are realized, thereby improving the efficiency and intelligence of panel design. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of the process of the present invention;

[0033] Figure 2 Schematic diagram of how various convolution results are divided into corresponding panel units. DETAILED DESCRIPTION

[0034] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments:

[0035] like Figure 1-Figure 2 As shown, a skeleton-based panel unit division method includes the following steps:

[0036] Step 1: Obtain point cloud data of the target skeleton surface, extract the various beam components of the skeleton based on the point cloud data, and classify the various beam components into corresponding types, as well as extract the axis points and cross-sectional corner points of the beam components. The axis points refer to the points obtained after the axis of the beam components is discretized, and the cross-sectional corner points refer to the vertices of the cross-section obtained by slicing the beam components at the axis points in a direction perpendicular to the axis.

[0037] The cross section of a beam component is usually a polygon, so the cross section obtained by slicing the beam component is a polygon. A polygon has multiple vertices, which means that each axis point corresponds to a cross section, each cross section has multiple vertices, and each axis point naturally corresponds to multiple cross section corner points. Therefore, discretizing the axis will result in multiple discrete points, and each discrete point is an axis point. Therefore, a beam component will have an axis, and discretizing the axis will result in multiple axis points, and each axis point corresponds to multiple cross section corner points.

[0038] In an optional embodiment, the beam components of the skeleton are divided into ring beams, main beams and secondary beams, that is, the types of beam components are divided into three types: ring beams, main beams and secondary beams.

[0039] Step 2: Extract the beam intersection points where each beam component projects onto the surface of the framework. A beam intersection point refers to the intersection point formed by a beam component intersecting with another beam component, and determine the types of the two beam components that constitute the beam intersection point. Based on several successively adjacent beam intersection points, enclose a polygon. Each beam intersection point serves as a vertex of the polygon, and each polygon serves as a panel unit to complete the division of panel units and obtain the result of panel unit division.

[0040] The specific implementation process of this step includes the following steps:

[0041] Step 21: Traverse each secondary beam and each main beam. The axis points of the same secondary beam are sorted in ascending order according to the X coordinate, and the axis points of the same main beam are sorted in ascending order according to the Y coordinate.

[0042] In this step, in the actual framework, almost all secondary beams are in the horizontal direction, and the main beams are in the front-back direction perpendicular to the horizontal direction. Therefore, the secondary beams are sorted according to the X coordinate, and the main beams are sorted according to the Y coordinate. Of course, in actual applications, the main beams can also be sorted according to the X coordinate, and the secondary beams can be sorted according to the Y coordinate, and the sorting can also be from large to small.

[0043] Step 22: Preset a global minimum distance threshold M and a maximum traversal times threshold N. M is an integer greater than 1, and N is the number of axis points of a single secondary beam. For example, M = 1000. Suppose the current secondary beam intersects 50 main beams, which means the number of axis points of the current secondary beam is 100, then N = 100.

[0044] Step 221: Traverse the sorted axis points of the secondary beam, and use the nearest neighbor algorithm to search for the axis point of the main beam that is closest to the currently traversed axis point of the secondary beam among the axis points of the current main beam. If the distance D between the currently traversed axis point of the secondary beam and the closest axis point of the main beam is D < M, then update M with the D value, that is, use the current D value as the updated M, and continue to traverse the axis points of the secondary beam until the distance D between the currently traversed axis point of the secondary beam and the closest axis point of the main beam is D ≥ M. Then exit the traversal, output the axis point of the secondary beam corresponding to the last updated M value, and output the previous axis point and the next axis point of this axis point as an adjacent point pair. If the number of traversal times of the axis points of the secondary beam reaches N, that is, when all the axis points of the secondary beam have been traversed and the M value has not been updated yet, it means that there are no potential intersection points between the current secondary beam and the current main beam, and no output is made. If the exit traversal condition is met when traversing the second secondary beam point, output the first two axis points of the secondary beam as an adjacent point pair;

[0045] Step 222: Repeat Step 221 for the axis points of each remaining main beam until the axis points of each main beam have been traversed. If there are adjacent point pairs between the current secondary beam and the main beams, the adjacent point pairs between the current secondary beam and all the main beams can be found;

[0046] Step 223: for the secondary beam axis point of each remaining secondary beam, repeat steps 221 and 222 until all the secondary beam axis points of each secondary beam are traversed. If there are adjacent point pairs between the secondary beam and the main beam, then the adjacent point pairs between all the remaining secondary beams except the current secondary beam and all the main beams can be found.

[0047] Step 224: Fit a first straight line according to each adjacent point pair corresponding to the current secondary beam axis point. The XY coordinates of each adjacent point pair can be fitted to a corresponding first straight line.

[0048] For the current secondary beam, a first straight line is fitted based on the adjacent point pairs corresponding to each current secondary beam axis point, so that a corresponding first straight line can be fitted based on the XY coordinates of the adjacent point pairs corresponding to each secondary beam axis point, so the number of first straight lines of the current secondary beam is the number of main beams that the current secondary beam intersects with all the main beams. Traverse all secondary beams to obtain the first straight lines corresponding to the axis points of each secondary beam on each secondary beam. For example, assuming that there are 100 secondary beams and 50 main beams, when the current secondary beam intersects with 50 main beams, that is, all main beams intersect with the current secondary beam, 50 first straight lines are fitted. If the number of main beams that the next secondary beam intersects with is only 30 among all the main beams, 30 first straight lines can be fitted for the next secondary beam.

[0049] Step 23: For the current first straight line, traverse the main beam axis points, extract the adjacent point pairs of the main beam axis points located on both sides of the first straight line, so as to find the adjacent point pairs in the main beam axis points corresponding to the current first straight line, and fit the second straight line corresponding to the current main beam axis points based on the adjacent point pairs in the current main beam axis points. Continue to traverse the remaining first straight lines, so as to fit the corresponding second straight line based on the adjacent point pairs in the main beam axis points for each straight line, and fit the second straight line based on the XY coordinates of the adjacent point pairs.

[0050] Step 24: Traverse the adjacent point pairs of each secondary beam axis point. If the adjacent point pairs of the secondary beam axis point are located on both sides of the second straight line, the secondary beam where the current secondary beam axis point is located intersects with the main beam corresponding to the second straight line, and a third straight line is fitted based on the XY coordinates of the adjacent secondary beam axis points.

[0051] Step 25: Calculate the intersection of the second straight line and the third straight line, and interpolate the Z coordinate of the intersection based on the coordinates of the adjacent point pair of the main beam axis point in step 23 to obtain the three-dimensional coordinates of the intersection, which also includes the XYZ coordinates.

[0052] In this step, the intersection is on both the second and third straight lines, but the second and third straight lines are calculated through XY coordinates, and only have XY coordinates, but no Z coordinates. Therefore, it is necessary to interpolate the Z coordinates of the adjacent point pairs of the main beam axis points based on the intersection, and make the intersection and the adjacent point pairs (i.e., the two main beam axis points) also on the same straight line in three-dimensional space. For example, the coordinates of the two main beam axis points of the adjacent point pair are (0,0,0) and (3,3,2), respectively, and the coordinates of the intersection are (1,1). Through interpolation, the coordinates of the intersection are (1,1,2 / 3), so that the intersection and the adjacent point pair are also on a straight line in three-dimensional space.

[0053] Step 26: Take the direction of the line between the adjacent point pairs of the main beam axis points as the direction vector of the three-dimensional intersection, and take the average value of the two cross-sectional dimensions at the adjacent point pairs of the main beam axis points as the cross-sectional dimension of the three-dimensional intersection. The cross-sectional dimension is generated by the corresponding cross-sectional node. According to the direction of the three-dimensional intersection and the cross-sectional dimension, translate the three-dimensional intersection to the upper surface of the main beam, and traverse each main beam axis point to obtain all the intersections of the main beam and the secondary beam, and obtain the three-dimensional intersection of all the intersections and translate the three-dimensional intersection to the upper surface of the main beam.

[0054] Step 27: According to the size of the coordinate values, two axis points are taken from the starting point of the secondary beam axis point, the end point of the secondary beam axis point, the starting point of the main beam axis point, and the end point of the main beam axis point, and the fourth straight line is calculated according to the XY coordinates of the taken axis points, and the nearest ring beam axis points located on both sides of the fourth straight line are calculated, and the fifth straight line is calculated according to the XY coordinates of the nearest ring beam axis points, and the nearest ring beam axis points are processed according to the method of steps 25-26, so as to obtain the intersection of the ring beam and the main beam projected to the upper surface of the ring beam, and the intersection of the ring beam and the secondary beam. Therefore, polygons are obtained based on the intersections of the main beam and the secondary beam, the main beam and the ring beam, and the secondary beam and the ring beam, so as to obtain panel units and the panel unit division results.

[0055] The above steps can complete the panel unit division without relying on manual division, which is more efficient and the divided panel unit results are more accurate.

[0056] The present invention also provides a blanking method based on the panel division unit result, which, after executing step 2, further includes the following steps:

[0057] Step 31: Construct a dimension The first matrix, A represents the number of secondary beams in the keel, B represents the number of main beams in the keel, and the elements in the first matrix represent whether the main beam and the secondary beam intersect. If the main beam and the secondary beam intersect and have an intersection, the corresponding element is assigned a value of 1; if the main beam and the secondary beam do not intersect and have no intersection, the corresponding element is assigned a value of 0.

[0058] Step 32: Expand the left and right sides of the rows and columns of the first matrix by a matrix with zero elements, that is, the expansion dimension is , and record it as the second matrix.

[0059] Step 33: Use the dimension A convolution kernel whose elements are all 1 is used to perform a convolution operation on the second matrix to obtain a convolution result. Different types of polygons are determined according to the convolution result. Different types of polygons correspond to different panel units, and the type of the polygon is determined by the number of edges of the polygon.

[0060] refer to Figure 2 , Figure 2 A schematic diagram showing how various convolution results are divided into corresponding panel units. Figure 2 The a) part indicates that the convolution result is 1, that is, there is only one element value of 1, and the divided panel unit represents a triangle surrounded by a ring beam, a main beam and a secondary beam; the b) part indicates that the convolution result is 2, that is, the number of element values ​​1 is 2, and the divided panel unit represents a quadrilateral surrounded by a ring beam, two main beams and a secondary beam, or a ring beam, two secondary beams and a main beam; the c) part indicates that the convolution result is 3, and the divided panel unit represents a pentagon surrounded by a ring beam, two main beams and two secondary beams; the d) part indicates that the convolution result is 4, and the divided panel unit represents a quadrilateral formed by the intersection of two main beams and two secondary beams.

[0061] Step 34: Among the polygons determined in step 33, any polygon whose side length is less than 2 times the width of the beam component is merged with an adjacent panel unit to obtain a merged polygon. The merged polygon and the unmerged polygon each correspond to a panel unit. The minimum circumscribed rectangle of each panel unit is calculated, and the position of a preset value inwardly along the vertical direction of the minimum circumscribed rectangle is used as the position of the blanking element. The blanking element is the material that constitutes the panel unit (such as glass), and the preset value is 1 / 2 of the panel unit seam. For example, 1 / 2 of the panel unit seam is 1 cm, then the position is retracted inward by 1 cm.

[0062] In an optional implementation, after step 34, the method further includes step 35:

[0063] Step 35: Use a heuristic (such as Guillotine) algorithm to generate a blanking component arrangement plan, and calculate the waste rate of each blanking component arrangement plan under different strategies of the Guillotine algorithm, and perform construction blanking with the blanking component arrangement plan with the smallest waste rate. Among them, the strategy includes the motherboard selection strategy and the residual material division strategy, which can be referred to in the following table:

[0064]

[0065] Among them, the waste rate is defined as the ratio of the residual material area to the total area of ​​the parent material.

[0066] The heuristic algorithm may also be any one of a Shelf algorithm, a maximum rectangle algorithm, a skyline algorithm, a genetic algorithm, and a particle swarm algorithm.

[0067] The present invention can rely on software programs to quickly and efficiently divide panel units on the keel, without relying on manual division, with higher efficiency and accuracy, and can also cut materials based on the panel division results, which is beneficial to saving material costs. Ultimately, the automatic division of spatial curved surface structure panels and the intelligent generation of reasonable cutting plans are realized, improving the efficiency and intelligence of panel design.

[0068] The embodiment disclosed in this specification is only an example of a unilateral feature of the present invention, and the protection scope of the present invention is not limited to this embodiment, and any other functionally equivalent embodiments fall within the protection scope of the present invention. For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present invention.

Claims

1. A panel unit division method based on skeleton, characterized in that: It includes the following steps: Step 1: Obtain the point cloud data of the target skeleton surface, extract each beam component of the skeleton based on the point cloud data, divide each beam component into corresponding types, the types of beam components are divided into ring beams, main beams and secondary beams, and extract the axis points and section corner points of the beam components. The axis points refer to the points obtained by discretizing the axis of the beam component, and the section corner points refer to the vertices of the section obtained by slicing the beam component along the direction perpendicular to the axis at the axis points; Step 2: Extract the beam intersection points where each beam component projects onto the skeleton surface. The beam intersection points refer to the intersection points formed by the intersection of a beam component and another beam component, and determine the types of the two beam components that make up the beam intersection point. Based on several successively adjacent beam intersection points, enclose a polygon. Each beam intersection point serves as a vertex of the polygon, and each polygon serves as a panel unit to complete the panel unit division and obtain the panel unit division result. The specific implementation process of Step 2 includes the following steps: Step 21: Traverse each secondary beam and each main beam, and sort the axis points of the same secondary beam and the same main beam respectively; Step 22: Preset the global minimum distance threshold M and the maximum traversal times threshold N. M is an integer greater than 1, and N is the number of axis points of a single secondary beam. Step 221: Traverse the sorted axis points of the secondary beam, and search for the axis point of the main beam that is closest to the currently traversed axis point of the secondary beam among the axis points of the current main beam through the nearest neighbor algorithm. If the distance D between the currently traversed axis point of the secondary beam and the nearest axis point of the main beam is less than M, then update M with the value of D, that is, the current D value is used as the updated M, and continue to traverse the axis points of the secondary beam until the distance D between the currently traversed axis point of the secondary beam and the nearest axis point of the main beam is greater than or equal to M. Exit the traversal, output the axis point of the secondary beam corresponding to the last updated M value, and output the previous axis point and the next axis point of this axis point as an adjacent point pair. If the number of traversal times of the axis points of the secondary beam reaches N, that is, when all the axis points of the secondary beam are traversed, and the M value has not been updated yet, it means that there are no potential intersection points between the current secondary beam and the current main beam, and no output is made. If the exit traversal condition is met when traversing the second axis point of the secondary beam, output the first two axis points of the secondary beam as an adjacent point pair; Step 222: For the axis points of the main beam of each remaining main beam, repeat Step 221 until the axis points of the main beam of each main beam are traversed, so as to find the adjacent point pairs between the current secondary beam and all main beams; Step 223: For the axis points of the secondary beam of each remaining secondary beam, repeat Steps 221 - 222 until the axis points of the secondary beam of each secondary beam are traversed, so as to find the adjacent point pairs between all secondary beams and all main beams; Step 224: According to the adjacent point pair corresponding to each current axis point of the secondary beam, fit a first straight line. The XY coordinates of each adjacent point pair can fit a corresponding first straight line; Step 23: For each first straight line, extract the nearest points of the corresponding main beam axis points located on both sides of the first straight line as the main beam adjacent point pairs, so as to find the adjacent point pairs in the main beam axis points corresponding to the current first straight line, and fit the second straight line based on the adjacent point pairs in the current main beam axis points. When the corresponding main beam axis points are all located on the same side of the first straight line, it indicates that the corresponding main and secondary beams do not intersect, and no output is performed, and continue to traverse the remaining first straight lines, so as to fit the corresponding second straight line based on the adjacent point pairs in the main beam axis points for each straight line, and the second straight line is fitted based on the XY coordinates of the adjacent point pairs; Step 24: traverse the adjacent point pairs of each secondary beam axis point, if the adjacent point pairs of the secondary beam axis point are located on both sides of the second straight line, the secondary beam where the current secondary beam axis point is located intersects with the main beam corresponding to the second straight line, and a third straight line is fitted based on the XY coordinates of the adjacent secondary beam axis points; Step 25: Calculate the intersection of the second straight line and the third straight line, and interpolate the Z coordinate of the intersection based on the coordinates of the adjacent point pair of the main beam axis point in step 23, so as to obtain the three-dimensional coordinates of the intersection, which also includes the XYZ coordinates; Step 26: Take the direction of the line between the adjacent point pairs of the main beam axis points as the direction vector of the three-dimensional intersection point, take the average value of the two cross-sectional dimensions at the positions of the adjacent point pairs of the main beam axis points as the cross-sectional dimension of the three-dimensional intersection point, the cross-sectional dimension is generated by the corresponding cross-sectional node, and according to the direction of the three-dimensional intersection point and the cross-sectional dimension, translate the three-dimensional intersection point to the upper surface of the main beam, and traverse each main beam axis point to obtain all the intersection points of the main beam and the secondary beam, and obtain the three-dimensional intersection points of all the intersection points and translate the three-dimensional intersection points to the upper surface of the main beam; Step 27: Take two axis points from the starting point of the secondary beam axis point, the end point of the secondary beam axis point, the starting point of the main beam axis point, and the end point of the main beam axis point according to the size of the coordinate values, calculate the fourth straight line according to the XY coordinates of the taken axis points, calculate the nearest ring beam axis points located on both sides of the fourth straight line, calculate the fifth straight line according to the XY coordinates of the nearest ring beam axis point, and process the nearest ring beam axis point according to the method of steps 25-26, so as to obtain the intersection of the ring beam and the main beam projected to the upper surface of the ring beam, and the intersection of the ring beam and the secondary beam, so as to obtain polygons based on the intersections of the main beam and the secondary beam, the main beam and the ring beam, and the secondary beam and the ring beam, so as to obtain panel units and the panel unit division results.

2. The skeleton-based panel unit division method according to claim 1, characterized in that: In step 1, the cross section of the beam component is a polygon, and the beam component is sliced ​​to obtain a polygonal cross section, where the vertices of the cross section are the cross section intersections, one axis point corresponds to one cross section, and each cross section includes multiple vertices.

3. The skeleton-based panel unit division method according to claim 1, characterized in that: In step 21, the axis points of the same secondary beam are sorted from small to large according to the X coordinate, and the axis points of the same main beam are sorted from small to large according to the Y coordinate.

4. The skeleton-based panel unit division method according to claim 1, characterized in that: M=1000。 5. The skeleton-based panel unit division method according to claim 1, characterized in that: In step 25, the intersection point is on both the second straight line and the third straight line, and the Z coordinates of the adjacent point pairs of the main beam axis points are interpolated based on the intersection point, so that the intersection point and the adjacent point pairs are also on the same straight line in three-dimensional space.

6. A cutting method based on the panel division unit result obtained by the skeleton-based panel unit division method according to any one of claims 1 to 5, characterized in that: After executing step 2, the following steps are also included: Step 31: construct a first matrix with a dimension of A×B, where A represents the number of secondary beams in the keel, and B represents the number of main beams in the keel. The elements in the first matrix represent whether the main beam and the secondary beam intersect. If the main beam and the secondary beam intersect and have an intersection, the corresponding element is assigned a value of 1; if the main beam and the secondary beam do not intersect and have no intersection, the corresponding element is assigned a value of 0; Step 32: Expand the left and right sides of the rows and columns of the first matrix by a matrix with zero elements, that is, the expansion dimension is , and record it as the second matrix; Step 33: Use the dimension A convolution kernel whose elements are all 1 performs a convolution operation on the second matrix to obtain a convolution result, and different types of polygons are determined according to the convolution result. Different types of polygons correspond to different panel units, and the type of polygon is determined by the number of edges of the polygon; Step 34: Among the polygons determined in step 33, any polygon whose side length is less than 2 times the width of the beam component is merged with an adjacent panel unit to obtain a merged polygon. The merged polygon and the unmerged polygon each correspond to a panel unit. The minimum circumscribed rectangle of each panel unit is calculated, and the position of the blanking element that is retracted inward by a preset value in the vertical direction of the minimum circumscribed rectangle is used as the position of the blanking element. The blanking element is the material constituting the panel unit, and the preset value is 1 / 2 of the panel unit joint. Also includes step 35: Step 35: Use a heuristic algorithm to generate a layout plan for blanking components, and calculate the waste rate of each layout plan for blanking components with different strategies under the heuristic algorithm, and perform construction blanking with the layout plan for blanking components with the smallest waste rate.

7. The blanking method according to claim 6, characterized in that: In step 33, various different convolution results are divided into corresponding panel units, where the convolution result is 1, that is, there is only 1 element value of 1, and the divided panel unit represents a triangle surrounded by a ring beam, a main beam and a secondary beam; the convolution result is 2, that is, the number of element values ​​1 is 2, and the divided panel unit represents a quadrilateral surrounded by a ring beam, two main beams and a secondary beam, or a ring beam, two secondary beams and a main beam; the convolution result is 3, and the divided panel unit represents a pentagon surrounded by a ring beam, two main beams and two secondary beams; the convolution result is 4, and the divided panel unit represents a quadrilateral formed by the intersection of two main beams and two secondary beams.

Images