A three-dimensional nesting optimization method and system of block wall heuristic and tree search

Abstract

This invention discloses a 3D material cutting optimization method and system based on block wall heuristics and tree search, belonging to the field of nesting optimization technology. This method breaks through the traditional forward cutting approach, solving the problem from a 3D binning perspective. Specifically, it includes: first, packaging similar parts into simple blocks to reduce computational complexity; second, setting multiple sets of wall thickness parameters based on complete cutting constraints, and using blocks to construct layered wall structures of different specifications; then, attempting to embed each wall structure into the remaining space of the raw materials to form multiple intermediate solutions; next, filling the remaining space using a fast completion algorithm to generate a complete solution and evaluate material utilization; finally, iteratively exploring and retaining the better branches through a tree search strategy until the termination condition is met, selecting the optimal solution. This invention effectively reduces computational redundancy through "block-wall" two-level aggregation and tree search pruning, balancing solution efficiency and material utilization, and can provide an efficient and accurate material cutting optimization solution for industrial scenarios.

Description

Technical Field

[0001] This invention belongs to the technical field of nesting optimization, and more specifically, relates to a three-dimensional material cutting optimization method and system based on block wall heuristics and tree search. Background Technology

[0002] In the production of block materials in industries such as furniture manufacturing and machinery manufacturing, material preparation is the first step in the manufacturing process. Especially for the cutting of precious metal raw materials, improving the utilization rate of raw materials can fundamentally reduce resource waste in the manufacturing process, thereby achieving the goal of reducing production costs. The material preparation problem for these block materials cannot be simplified to the material preparation problem for two-dimensional sheets; therefore, corresponding algorithms are needed to handle this type of three-dimensional material preparation problem. Currently, there is limited research on three-dimensional material preparation algorithms. Some research on three-dimensional material preparation algorithms is categorized under the three-dimensional bin packing problem. The three-dimensional material preparation problem can be seen as a special type of three-dimensional bin packing problem, differing from three-dimensional bin packing in that it needs to satisfy the special constraint of complete cut constraint. That is, starting from one edge of the raw material, cutting is performed along a certain direction until another edge, dividing the raw material into two parts. All cuts must be performed according to this rule.

[0003] Currently, algorithms for solving 3D material cutting problems are mainly divided into three types: exact algorithms, heuristic algorithms, and metaheuristic algorithms (intelligent algorithms). Traditional exact algorithms often use branch and bound methods or model the problem as a (hybrid) integer programming model, but their limitation is that they are only applicable to small-scale instances. Heuristic algorithms, by setting rules, such as specifying the priority of cutting directions or specifying the cutting position based on the largest item size, can usually quickly find feasible solutions, but the quality of the solutions is often not very high. Metaheuristic algorithms, such as genetic algorithms and simulated annealing algorithms, usually need to be combined with specific heuristic rules, such as cutting rules or item matching rules, to solve the problem, and need to balance the quality and efficiency of the solution.

[0004] In general, current research on 3D blanking algorithms is limited, and related 3D bin packing studies rarely consider complete cutting constraints. 3D bin packing research focuses primarily on directional and stability constraints, and its solution methods are not directly applicable to 3D blanking problems. Furthermore, existing 3D blanking algorithms have limited effectiveness and insufficient material utilization, especially when dealing with a large number of part types. Therefore, there is an urgent need to develop a 3D blanking optimization method with higher material utilization. Summary of the Invention

[0005] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides a block-wall heuristic and tree-search-based three-dimensional material cutting optimization method and system. Its purpose is to maximize material utilization and reduce production costs, thereby solving the technical problems of limited effectiveness and insufficient raw material utilization in existing three-dimensional material cutting algorithms.

[0006] To achieve the above objectives, according to one aspect of the present invention, a three-dimensional material cutting optimization method based on block wall heuristics is provided, comprising the following steps: S1. Input parameters: Obtain part information and raw material information. The part information includes at least the specifications, quantity, and whether rotation is allowed. The raw material information includes at least the specifications of the raw material. S2. Block Packaging Processing: Combine parts of the same specifications into blocks according to preset packaging rules, and record the size information of each block and the information of the parts it contains; S3. Wall structure generation: Select different wall thicknesses, and according to these wall thicknesses and the dimensions of the remaining space of the current raw materials, select suitable blocks from the blocks generated in step S2, and fill them to generate several wall structures; S4. Scheme Construction: In each round, embed the wall with the highest fill rate among the currently generated wall structures into the remaining space of raw materials, and update the list of available parts and the remaining space of raw materials; S5. Iterative optimization and output: Repeat steps S3 and S4 until all parts can no longer be placed into the remaining space of raw materials, and obtain the final cutting scheme.

[0007] Preferably, the part information in step S1 specifically includes: the three-dimensional dimensions of the i-th type of part. ,Width ,high and the quantity required The raw material information specifically includes: the three-dimensional dimensions of the raw material. ,Width ,high Furthermore, there is no limit to the quantity of raw materials.

[0008] Preferably, the block packaging process in step S2 specifically includes the following steps: S21 Attitude Enumeration: For parts of the same specification, enumerate all placement attitudes that meet the placement constraints. S22 Block Assembly: For each placement orientation, enumerate all possible arrangements of parts in the length, width, and height directions to form a block; the size of the formed block is constrained by two main factors: the three-dimensional dimensional boundaries of the raw materials and the preset maximum wall thickness threshold. That is, the minimum value of the three-dimensional dimensions of the block must not exceed the maximum wall thickness threshold. ; S23 Block Information Recording: For all blocks that meet the constraints, record their core information. ,in, This represents the k-th block. These represent the blocks in Dimensions in the direction These respectively represent the parts in The number of permutations in the three directions.

[0009] Preferably, step S3 specifically includes the following steps: S31 Wall Thickness Selection: Select different values ​​from the set of dimensions of all parts as candidate wall thicknesses. ; S32 Wall Orientation and Dimensioning: Based on the dimensions of the remaining space of the current raw materials. , , and the selected wall thickness Determine the specific dimensions of walls in three directions: horizontal, vertical, or horizontal. , , These represent the length, width, and height of the remaining space from the raw materials, respectively, with the transverse wall dimensions being... The vertical wall dimensions are The dimensions of the flat wall are ; S33 Space Filling: Select the current space to be filled, filter out all blocks whose volume does not exceed the current space volume and sort them in descending order of volume. Check each block in turn from three dimensions to see if it can be embedded in the current space. If it can be embedded, place the block at the lower left corner of the current space and use the three-part division method to divide the remaining space after filling into upper space, front space, and right space, and enter the space list for later use. Repeat the current step until no block can be embedded in any space in the space list, thus completing the filling of the wall in this direction.

[0010] Preferably, the filling rate in step S4 The calculation formula is as follows:

[0011] in, The number of different types of parts placed in this wall. To be placed in the wall The number of parts , For the width and height of the wall, The wall's fill rate is the ratio of the sum of the volumes of all placed parts to the wall's volume, where the wall thickness is given. .

[0012] Preferably, the material utilization rate of the final feeding scheme in step S5 is [missing information]. The calculation formula is as follows:

[0013] in, This represents the number of part types in the scheme. For the first in this scheme The number of parts , , These are the length, width, and height dimensions of the raw material; the ratio of the sum of the volumes of all the placed parts to the volume of the raw material is the material utilization rate of this scheme. .

[0014] According to one aspect of the present invention, a three-dimensional blanking optimization method based on block wall heuristic and tree search is provided, comprising the following steps: A1. Input parameters: Obtain part information and raw material information. The part information includes at least the specifications, quantity, and whether rotation is allowed. The raw material information includes at least the specifications of the raw material. A2. Block Packaging Processing: Combine parts of the same specifications into blocks according to preset packaging rules, and record the size information of each block and the information of the parts it contains; A3. Wall structure generation: Select different wall thicknesses, and based on these wall thicknesses and the dimensions of the remaining space of the current raw materials, select suitable blocks from the blocks generated in step A2, and fill them to generate several wall structures; A4. Multi-branch solution construction: Try placing each wall structure generated in step A3 into the remaining space of the raw materials to form multiple different partial solutions; A5. Solution Completion and Evaluation: For each partial solution, the remaining space is filled using the block wall-based heuristic 3D cutting optimization method described above, generating the corresponding complete cutting solution; the space filling rate of the complete solution is used as the core evaluation criterion to evaluate the merits of the corresponding partial solutions, and all valid complete solutions are recorded. A6. Branch Selection and Iteration: In each round, select the better branch from all current partial solutions. Each selected scheme is retained for further exploration. This continued exploration involves repeating steps A3 to A5 for each retained scheme until no further wall structures can be incorporated into the retained schemes, at which point the algorithm terminates naturally. If the number of complete schemes reaches a preset threshold during the exploration process... If so, the algorithm terminates prematurely; A7. Optimal Solution Output: Select the solution with the highest material utilization rate from all generated complete solutions as the final material cutting solution.

[0015] Preferably, step A5 specifically includes: A51. Exploration Process: Assume the total number of walls generated in step A3 of this round is... Then in this round, we will try to do this respectively. Place the wall into the remaining space of the raw materials to obtain One option; A52. Completion process: Check in sequence We examine each plan to see if it needs completion. If the current material space can no longer accommodate any more walls, it means we have reached a leaf node, i.e., a complete material cutting plan has been formed, and this plan is recorded. If walls can be further embedded, we use a fast completion algorithm to complete this part of the plan to obtain the corresponding complete plan, and calculate the material utilization rate of the complete plan. ; A53. Evaluation Process: Set the number of nodes at each level of the tree search to be... That is, each layer retains only the better ones. One solution will continue to be explored, while the remaining solutions will be discarded and no longer explored; if only one solution is explored at this layer... The solution is insufficient. If there are [number], then all [number] will be retained, among which [number] [number] [are retained]. The calculation formula is as follows: .

[0016] According to one aspect of the present invention, a three-dimensional material cutting optimization system based on block wall heuristic and tree search includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method.

[0017] According to another aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0018] In summary, compared with the prior art, the three-dimensional material cutting optimization method and system based on block wall heuristics and tree search provided by the present invention mainly have the following beneficial effects: 1. This invention provides a block-wall heuristic and tree-search-based 3D blanking optimization method. The designed "block-wall" two-level aggregation structure improves the regularity of part arrangement and computational efficiency through the "blocks," while the "walls" perfectly fit the complete cutting constraint of the 3D blanking problem. Specifically, firstly, a wall structure is designed that can well fit the complete cutting constraint; secondly, the introduction of blocks increases the neatness of part arrangement and improves the utilization rate of raw materials to a certain extent. Compared with considering each part individually, this method is more efficient in computational solution.

[0019] 2. The block-wall-based heuristic method can handle large-scale problems within minutes, achieving high material utilization and meeting the real-time requirements of industrial scenarios. Simultaneously, this method can also serve as a fast completion algorithm to evaluate the potential of certain solutions. Specifically, parts are first packaged into regular blocks, and then filling operations are performed based on these blocks, thereby reducing unnecessary computational steps. When filling the remaining space of raw materials with wall structures, this method directly selects the wall with the highest filling rate for the filling process. This heuristic method combines excellent material utilization with efficient solution performance, enabling the solution of large-scale 3D material preparation problems within minutes, adapting to the actual needs of industrial production scenarios. In the preferred embodiment of this invention, the block-wall heuristic method can also serve as a fast completion algorithm to quickly complete and optimize certain solutions, thereby achieving accurate evaluation of the value of those solutions.

[0020] 3. This invention provides a 3D material cutting optimization method based on block wall heuristics and tree search. It proposes a tree search optimization structure for wall placement schemes. The core of this method lies in intelligently selecting search branches by real-time evaluation of the merits of some schemes after wall embedding. Branches with poor performance are pruned, significantly reducing redundant computation; potentially superior branches are retained and explored in depth. By setting a reasonable number of retained nodes, a balance is achieved between algorithm accuracy and operational efficiency. The 3D material cutting optimization method based on block wall heuristics and tree search achieves better results than the block wall heuristic method.

[0021] 4. In the preferred method of this invention, a standard size set is designed as the wall thickness to avoid meaningless enumeration and exploration of wall thickness. By defining a maximum wall thickness, the generated blocks do not exceed the maximum wall thickness, ensuring that block generation perfectly matches wall filling and preventing the generation of redundant and useless blocks. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of a three-dimensional material cutting optimization method based on block wall heuristics and tree search provided by the present invention; Figure 2 This is a schematic diagram of the wall structure generated in an embodiment of the present invention; Figure 3 This is a schematic diagram of the fully cut constraint cutting method provided in the embodiments of the present invention; Figure 4 This is a schematic diagram of the three-dimensional blanking result provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the tree search in the wall selection stage provided in an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0024] Example 1 This invention provides a block-wall-based heuristic method for optimizing 3D material cutting. Please refer to [link / reference]. Figure 1 This includes the following steps: S1. Input parameters: Obtain part information and raw material information. The part information includes at least the specifications, quantity, and whether rotation is allowed. The raw material information includes at least the specifications of the raw material. Specifically, the part information mentioned in step S1 includes: the three-dimensional dimensions of the i-th type of part. ,Width ,high and the quantity required The raw material information specifically includes: the three-dimensional dimensions of the raw material. ,Width ,high Furthermore, there is no limit to the quantity of raw materials.

[0025] S2. Block Packaging Processing: Combine parts of the same specifications into blocks according to preset packaging rules, and record the size information of each block and the information of the parts it contains; Specifically, the block packing process in step S2 includes the following steps: S21 Attitude Enumeration: For parts of the same specification, enumerate all placement attitudes that meet the placement constraints. Specifically, parts of the same specifications are placed in several different feasible orientations. For a cuboid, there are at most six possible orientations. Sometimes, the 3D cutting problem needs to consider factors such as texture direction, so not all six orientations are feasible. In addition, if two or three of the dimensions of the cuboid are the same, the number of possible orientations will also decrease.

[0026] S22 Block Assembly: For each placement orientation, enumerate all possible arrangements of parts in the length, width, and height directions to form a block; the size of the formed block is constrained by two main factors: the three-dimensional dimensional boundaries of the raw materials and the preset maximum wall thickness threshold. That is, the minimum value of the three-dimensional dimensions of the block must not exceed the maximum wall thickness threshold. ; Specifically, the dimension of the block in the x-direction must not exceed the dimension of the raw material in the x-direction. The dimensions in the y and z directions also do not exceed the corresponding dimensions of the raw material in the respective directions. , Meanwhile, to meet the requirements of subsequently assembling the blocks into the wall structure, the assembled blocks need to be "flat". Therefore, a clear constraint is imposed: the minimum value of any of the three-dimensional dimensions of the block must not exceed the preset maximum wall thickness threshold. .

[0027] S23 Block Information Recording: For all blocks that meet the constraints, record their core information. ,in, This represents the k-th block. These represent the blocks in Dimensions in the direction These respectively represent the parts in The number of permutations in the three directions.

[0028] S3. Wall structure generation: Select different wall thicknesses, and according to these wall thicknesses and the dimensions of the remaining space of the current raw materials, select suitable blocks from the blocks generated in step S2, and fill them to generate several wall structures; Specifically, step S3 includes the following steps: S31 Wall Thickness Selection: Select different values ​​from the set of dimensions of all parts as candidate wall thicknesses. ; Specifically, the three-dimensional dimensions of all parts can be obtained based on the part specification information input in step S1. Choose the wall thickness from these dimensions. ; S32 Wall Orientation and Dimensioning: Based on the dimensions of the remaining space of the current raw materials. , , and the selected wall thickness Determine the specific dimensions of walls in three directions: horizontal, vertical, or horizontal. , , These represent the length, width, and height of the remaining space from the raw materials, respectively, with the transverse wall dimensions being... The vertical wall dimensions are The dimensions of the flat wall are ; Specifically, the orientation and dimensions of the walls are determined, including horizontal walls (XZ direction), vertical walls (YZ direction), and horizontal walls (XY direction). The three-dimensional dimensions of the corresponding walls are determined by combining the actual dimensions of the remaining space in the raw material. In the initial state (when no walls are embedded), the dimensions of the remaining space are consistent with the original dimensions of the raw material, i.e. , , .

[0029] S33 Space Filling: Select the current space to be filled, filter out all blocks whose volume does not exceed the current space volume and sort them in descending order of volume. Check each block in turn from three dimensions to see if it can be embedded in the current space. If it can be embedded, place the block at the lower left corner of the current space and use the three-part division method to divide the remaining space after filling into upper space, front space, and right space, and enter the space list for later use. Repeat the current step until no block can be embedded in any space in the space list, thus completing the filling of the wall in this direction.

[0030] S4. Scheme Construction: In each round, embed the wall with the highest fill rate among the currently generated wall structures into the remaining space of raw materials, and update the list of available parts and the remaining space of raw materials; Specifically, the fill rate mentioned in step S4 The calculation formula is as follows:

[0031] in, The number of different types of parts placed in this wall. To be placed in the wall The number of parts , For the width and height of the wall, The wall's fill rate is the ratio of the sum of the volumes of all placed parts to the wall's volume, where the wall thickness is given. .

[0032] S5. Iterative optimization and output: Repeat steps S3 and S4 until all parts can no longer be placed into the remaining space of raw materials, and obtain the final cutting scheme.

[0033] Specifically, the material utilization rate of the final feeding scheme in step S5. The calculation formula is as follows:

[0034] in, This represents the number of part types in the scheme. For the first in this scheme The number of parts , , These are the length, width, and height dimensions of the raw material; the ratio of the sum of the volumes of all the placed parts to the volume of the raw material is the material utilization rate of this scheme. .

[0035] Example 2 This invention provides a block wall heuristic and tree search-based 3D material cutting optimization method, the flowchart of which is shown below. Figure 1As shown, its first three steps are the same as the block wall-based heuristic 3D material cutting optimization method mentioned above, specifically including the following steps: A1. Input parameters: including the three-dimensional dimensions of various parts, the required quantity, and the three-dimensional dimensions of the raw materials, specifying whether rotation of various parts is allowed.

[0036] Specifically, enter the specifications and quantities of the three types of parts: Part 1: 10×10×4, 12 pieces in total; Part 2: 8×6×6, 7 pieces in total; Part 3: 4×4×4, 13 pieces in total. All three types of parts can be rotated arbitrarily. Enter the specifications of the raw material: 20×20×20.

[0037] A2. Block Packaging Processing: Pack parts of the same specifications into "simple blocks" in a reasonable arrangement, and record the three-dimensional dimensions of each block, the number of parts contained in it, and the corresponding part types, to ensure that the block size does not exceed the specifications of the raw materials.

[0038] Specifically, taking part 2 as an example, the enumeration process of packaging into blocks is demonstrated. Part 2 has a specification of 8×6×6, totaling 7 pieces, and the raw material specification is 20×20×20. From the part list data, we know that the maximum wall thickness is 10, and the minimum dimension of the block cannot exceed 10. Since part 2 has the same dimension in two dimensions, there are 3 possible placement postures for part 2, namely: Posture 1: <8, 6, 6> means the dimension is 8 in the x-direction, 6 in the y-direction, and 6 in the z-direction; Posture 2: <6, 8, 6>; Posture 3: <6, 6, 8>. If placed in Posture 1, at most 20÷8=2.5 pieces can be placed in the x-direction, which is rounded down to 2. Similarly, at most 3 pieces can be placed in the y-direction and at most 3 pieces in the z-direction. By forming blocks according to posture 1, a total of 13 blocks can be obtained: [1,1,1], [1,1,2], [1,1,3], [1,2,1], [1,2,2], [1,2,3], [1,3,1], [1,3,2], [2,1,1], [2,1,2], [2,1,3], [2,2,1], [2,3,1]. (Note: [nx,ny,nz] means placing nx blocks along the x-direction, ny blocks along the y-direction, and nz blocks along the z-direction.) Due to the special characteristic that the raw material in this embodiment is a cube, posture 2 can also produce 13 blocks, and posture 3 can also produce 13 blocks. A total of 39 blocks are generated for part 2.

[0039] A3. Wall Structure Generation: Set multiple sets of different wall thickness parameters (wall thickness is the thickness dimension along a certain dimension). For each set of wall thickness, combined with the three-dimensional dimensions of the currently available space, select suitable blocks from the generated simple blocks and fill them in sequence to generate several "wall structures".

[0040] Specifically, based on the input parts list, the standard size set is {10, 8, 6, 4}. The initial remaining space of the raw materials is 20×20×20. Wall thicknesses of 4, 6, 8, and 10 are selected respectively, and walls are generated along the horizontal, vertical, and planar directions, resulting in a total of 12 walls. Figure 2 All wall structures are generated, and all components within them satisfy the complete cut constraint.

[0041] For a visual explanation of the complete cut constraint, please refer to [link / reference]. Figure 3 , Figure 3 (a) in the text represents a cutting scheme that does not meet the complete cutting constraint, meaning that several small cuboids cannot be separated by a through-cut. Figure 3 In the diagram, (b) represents the cutting scheme that satisfies the complete cutting constraint. First, longitudinal cutting is used to obtain a blue cuboid, then transverse cutting is used to obtain a red cuboid, and finally longitudinal cutting is used to obtain a yellow and a green cuboid.

[0042] A4. Multi-branch solution construction: In each iteration round, all wall structures generated in step A3 are attempted to be embedded into the remaining space of the raw materials to form multiple different partial solutions (i.e., intermediate solutions that do not completely fill the raw materials).

[0043] Specifically, in the first round, the bottom left corner of the 12 walls generated above is placed into the bottom left corner of the raw material space, resulting in 12 partial solutions. Correspondingly, the list of available parts and the remaining space of the raw material are updated. For example, wall ① has dimensions of 4×20×20, a wall thickness of 4, and a orientation of YZ. It contains 4 parts (part 1). After placement, the raw material space is updated from 20×20×20 to 16×20×20, and the list of available parts is updated as follows: Part 1: 10×10×4, 8 remaining; Part 2: 8×6×6, 7 remaining; Part 3: 4×4×4, 13 remaining.

[0044] A5. Complete Solution Completion and Evaluation: For each partial solution, the remaining space is filled using the block wall-based heuristic 3D cutting optimization method described above, generating the corresponding complete cutting solution; the space filling rate of the complete solution is used as the core evaluation criterion to evaluate the merits of the corresponding partial solutions, and all valid complete solutions are recorded.

[0045] Specifically, such as Figure 4 (a) shows the complete solution obtained after quickly completing the corresponding partial solution for wall ①, with a material utilization rate of 0.92. The remaining 11 partial solutions were also completed using the same method to obtain their corresponding complete solutions, and these complete solutions will all be recorded.

[0046] A6. Iterative Exploration and Termination Criteria: From all current partial solutions, select the ones with the highest fill rates. The better solutions are retained and the exploration continues. That is, for each retained partial solution, steps A3 to A5 are repeated, and the remaining inefficient solutions are discarded. Iteration continues until all retained solutions can no longer be embedded in any new wall structure, at which point the algorithm naturally terminates. If the number of recorded valid complete solutions reaches a preset threshold during the exploration process... The algorithm terminates prematurely.

[0047] Specifically, the tree search process can be achieved through... Figure 5 The simplified diagram shown is for illustrative purposes only and is intended to visually illustrate the core logic of tree search. It does not represent the specific tree search process in this example. The diagram uses parameter settings... As clearly observed in the diagram, in each iteration of the tree search, three relatively optimal partial solutions (marked with gray circles) are selected. These solutions become the nodes to be expanded in the next level of the tree search. For these three retained relatively optimal partial solutions, operations such as wall structure generation, wall embedding into the remaining space of raw materials, rapid solution completion, and merit evaluation are performed. The number of branches below each node corresponds to the number of wall structures currently generated. The remaining unselected partial solutions (marked with white circles) are directly discarded and will not participate in subsequent iterations. The squares in the diagram represent complete material cutting solutions that have completed filling of the remaining space, meaning that the branch has reached the leaf node and the search process has terminated.

[0048] Specifically, algorithm parameter settings Regarding the 12 partial schemes generated, Therefore, all were retained, and walls were continued to be generated. Attempts were made to place walls and evaluate them. After a total of 7 levels of tree search, a total of 353 complete solutions were obtained. Algorithm parameter settings... Since the number of solutions did not exceed 500, the solution with the highest material utilization rate was selected as the final result. Figure 4 As shown in (b) of the diagram.

[0049] The results of this design show that three walls were installed, all YZ walls. The first wall measures 10×20×20 mm and contains 10 parts (part 1). The second wall measures 4×20×20 mm and contains 2 parts (part 1) and 10 parts (part 3). The third wall measures 6×20×20 mm and contains 7 parts (part 2) and 3 parts (part 3). The final material utilization rate is 0.956.

[0050] A7. Optimal Solution Output: Select the solution with the highest material utilization rate from all generated complete solutions as the final material cutting solution.

[0051] In summary, the method proposed in this application breaks through the traditional forward cutting approach, solving the problem from a 3D packing perspective: the parts to be cut are equivalent to "boxes to be packed into," and the raw materials are equivalent to "target storage space." By reverse-engineering the combination and arrangement of parts, the optimal forward cutting solution is finally obtained. Through the "block-wall" two-level aggregation strategy and reverse-solving logic, the computational redundancy of large-scale 3D cutting problems is effectively reduced, balancing solution efficiency and material utilization, and providing an efficient and accurate cutting optimization solution for industrial scenarios.

[0052] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A three-dimensional nesting optimization method based on block wall heuristics, characterized in that: Includes the following steps: S1. Input parameters: Obtain part information and raw material information. The part information includes at least the specifications, quantity, and whether rotation is allowed. The raw material information includes at least the specifications of the raw material. S2. Block Packaging Processing: Combine parts of the same specifications into blocks according to preset packaging rules, and record the size information of each block and the information of the parts it contains; S3. Wall structure generation: Select different wall thicknesses, and according to these wall thicknesses and the dimensions of the remaining space of the current raw materials, select suitable blocks from the blocks generated in step S2, and fill them to generate several wall structures; S4. Scheme Construction: In each round, embed the wall with the highest fill rate among the currently generated wall structures into the remaining space of raw materials, and update the list of available parts and the remaining space of raw materials; S5. Iterative optimization and output: Repeat steps S3 and S4 until all parts can no longer be placed into the remaining space of raw materials, and obtain the final cutting scheme.

2. The three-dimensional material cutting optimization method based on block wall heuristic as described in claim 1, characterized in that: The part information mentioned in step S1 specifically includes: the three-dimensional dimensions of the i-th part. ,Width ,high and the quantity required The raw material information specifically includes: the three-dimensional dimensions of the raw material. ,Width ,high Furthermore, there is no limit to the quantity of raw materials.

3. The three-dimensional material cutting optimization method based on block wall heuristic as described in claim 1, characterized in that, Step S2, the block packing process, specifically includes the following steps: S21 Attitude Enumeration: For parts of the same specification, enumerate all placement attitudes that meet the placement constraints. S22 Block Assembly: For each placement orientation, enumerate all possible arrangements of parts in the length, width, and height directions to form a block; the size of the formed block is constrained by two main factors: the three-dimensional dimensional boundaries of the raw materials and the preset maximum wall thickness threshold. That is, the minimum value of the three-dimensional dimensions of the block must not exceed the maximum wall thickness threshold. ; S23 Block Information Recording: For all blocks that meet the constraints, record their core information. ,in, This represents the k-th block. These represent the blocks in Dimensions in the direction These respectively represent the parts in The number of permutations in the three directions.

4. The three-dimensional material cutting optimization method based on block wall heuristic as described in claim 1, characterized in that, Step S3 specifically includes the following steps: S31 Wall Thickness Selection: Select different values ​​from the set of dimensions of all parts as candidate wall thicknesses. ; S32 Wall Orientation and Dimensioning: Based on the dimensions of the remaining space of the current raw materials. , , and the selected wall thickness Determine the specific dimensions of walls in three directions: horizontal, vertical, or horizontal. , , These represent the length, width, and height of the remaining space from the raw materials, respectively, with the transverse wall dimensions being... The vertical wall dimensions are The dimensions of the flat wall are ; S33 Space Filling: Select the current space to be filled, filter out all blocks whose volume does not exceed the current space volume and sort them in descending order of volume. Check each block in turn from three dimensions to see if it can be embedded in the current space. If it can be embedded, place the block at the lower left corner of the current space and use the three-part division method to divide the remaining space after filling into upper space, front space, and right space, and enter the space list for later use. Repeat the current step until no block can be embedded in any space in the space list, thus completing the filling of the wall in this direction.

5. The three-dimensional material cutting optimization method based on block wall heuristic as described in claim 1, characterized in that, The fill rate mentioned in step S4 The calculation formula is as follows: in, The number of different types of parts placed in this wall. To be placed in the wall The number of parts , For the width and height of the wall, The wall's fill rate is the ratio of the sum of the volumes of all placed parts to the wall's volume, where the wall thickness is given. .

6. The three-dimensional material cutting optimization method based on block wall heuristic as described in claim 1, characterized in that, Material utilization rate in the final material cutting scheme in step S5 The calculation formula is as follows: in, This represents the number of part types in the scheme. For the first in this scheme The number of parts , , These are the length, width, and height dimensions of the raw material; the ratio of the sum of the volumes of all the placed parts to the volume of the raw material is the material utilization rate of this scheme. .

7. A three-dimensional material cutting optimization method based on block wall heuristics and tree search, characterized in that: Includes the following steps: A1. Input parameters: Obtain part information and raw material information. The part information includes at least the specifications, quantity, and whether rotation is allowed. The raw material information includes at least the specifications of the raw material. A2. Block Packaging Processing: Combine parts of the same specifications into blocks according to preset packaging rules, and record the size information of each block and the information of the parts it contains; A3. Wall structure generation: Select different wall thicknesses, and based on these wall thicknesses and the dimensions of the remaining space of the current raw materials, select suitable blocks from the blocks generated in step A2, and fill them to generate several wall structures; A4. Multi-branch solution construction: Try placing each wall structure generated in step A3 into the remaining space of the raw materials to form multiple different partial solutions; A5. Solution Completion and Evaluation: For each partial solution, the block wall-based three-dimensional cutting optimization method described in any one of claims 1 to 6 is used as a fast completion algorithm to fill the remaining space and generate the corresponding complete cutting solution; Using the space filling rate of the complete solution as the core evaluation criterion, we assess the merits and demerits of the corresponding partial solutions and record all valid complete solutions. A6. Branch Selection and Iteration: In each round, select the better branch from all current partial solutions. Each selected scheme is retained for further exploration. This continued exploration involves repeating steps A3 to A5 for each retained scheme until no further wall structures can be incorporated into the retained schemes, at which point the algorithm terminates naturally. If the number of complete schemes reaches a preset threshold during the exploration process... If so, the algorithm terminates prematurely; A7. Optimal Solution Output: Select the solution with the highest material utilization rate from all generated complete solutions as the final material cutting solution.

8. The 3D material cutting optimization method based on block wall heuristic and tree search as described in claim 7, characterized in that: Step A5 specifically includes: A51. Exploration Process: Assume the total number of walls generated in step A3 of this round is... Then in this round, we will try to do this respectively. Place the wall into the remaining space of the raw materials to obtain One option; A52. Completion process: Check in sequence We examine each plan to see if it needs completion. If the current material space can no longer accommodate any more walls, it means we have reached a leaf node, i.e., a complete material cutting plan has been formed, and this plan is recorded. If walls can be further embedded, we use a fast completion algorithm to complete this part of the plan to obtain the corresponding complete plan, and calculate the material utilization rate of the complete plan. ; A53. Evaluation Process: Set the number of nodes at each level of the tree search to be... That is, each layer retains only the better ones. One solution will continue to be explored, while the remaining solutions will be discarded and no longer explored; if only one solution is explored at this layer... The solution is insufficient. If there are [number], then all [number] will be retained, among which [number] [number] [are retained]. The calculation formula is as follows: 。 9. A three-dimensional material cutting optimization system based on block wall heuristics and tree search, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method according to any one of claims 1 to 8.

Images