钢板切割领域切板件智能套料算法数据

通过系统动态规划，套料算法可实现加工零件从设计到生产的一站式全自动标准化套料，无需人工排版，节约了人工成本，提高了钢板利用率，减少排版套料时间，并自动生成板材利用率清单、设备利用率清单、生产安排表、耗材定额表等管理表格，实现对生产流程的数字化管理。
遗传优化算法用于智能排版，通过交叉和变异，选择排布每个零件的位置和旋转角度，在下一代产生更好的解，通过适应度函数评估每一种材料的适应度，并且利用交叉和变异的过程不断地迭代寻找最优解。完成快速报价，提升报价速度和准确率，高效排板，提升材料利用率。对于一个配件的摆放位置，配件的边界标为1，边界内/外分别标为3和0，当有新的配件被放置，将像素值相加，就可以根据像素值分辨各种情况：0表示此位置无配件，1表示某个配件未重叠边界，2表示两个配件的边界重叠，4表示一个配件的边界和另一个配件的界内重叠，6表示两个配件界内部分重叠。对于配件的布置进行评分：
第一步：针对第一个配件，设置可选的范围（至少一个顶点贴边），第一代的解包括随机产生的M组坐标值；
第二步：按照每组坐标值，将第一个配件放入材料种，判定是否违规，如未违规，按照a中所述的综合评分，布置剩余的配件使得综合评分最高，如果当前型号配件放不下了，则布置小一号的配件，直到最小的也放不下了。
第三步：计算材料利用率作为绩效评估值J，并根据绩效评估值相对大小计算选择概率（绩效评估值则选择概率大）。
第四步：进行基因操作，遗传最佳（绩效最高）的那组坐标用于保底，并随机选择其他组进行基因交叉或变异，从而产生下一代的解，希望通过交叉或变异在下一代产生更好的解。
第五步：针对下一代的解，重复第二至四步直至绩效评估值高于阈值，或者最大迭代次数。

Using systematic dynamic programming, the nesting algorithm realizes a fully automated, standardized one-stop nesting workflow from part design to production. It eliminates the need for manual layout, reduces labor costs, improves steel plate utilization rate, shortens nesting time, and automatically generates management documents including material utilization list, equipment utilization list, production schedule, and consumable quota form, enabling digital management of the production process.

The genetic optimization algorithm is applied to intelligent layout: it selects the position and rotation angle of each part through crossover and mutation operations to generate better solutions in the next generation. The fitness function is used to evaluate the fitness of each material solution, and the processes of crossover and mutation are utilized to continuously iterate to find the optimal solution. This approach facilitates rapid quotation, improves the speed and accuracy of quotation, enables efficient plate layout, and enhances material utilization rate.

For the placement of a part, its boundary is marked as 1, while the interior and exterior of the boundary are marked as 3 and 0 respectively. When a new part is placed, the pixel values of the corresponding area are summed to distinguish various placement scenarios: 0 indicates no part at this position; 1 indicates the boundary of a single part without overlapping with others; 2 indicates overlapping boundaries of two parts; 4 indicates overlapping between the boundary of one part and the interior of another part; 6 indicates overlapping interiors of two parts.

Scoring for part arrangement is conducted as follows:
Step 1: For the first part, define the optional placement range (with at least one vertex attached to the edge of the stock material). The initial generation of solutions includes M randomly generated coordinate sets.
Step 2: For each coordinate set, place the first part into the stock material and check for placement violations. If no violation occurs, arrange the remaining parts to achieve the highest comprehensive score as mentioned in section a. If parts of the current model cannot be placed, switch to smaller-sized parts until even the smallest available part cannot be accommodated.
Step 3: Calculate the material utilization rate as the performance evaluation value J, and compute the selection probability based on the relative magnitude of the performance evaluation values (higher performance evaluation values correspond to higher selection probabilities).
Step 4: Perform genetic operations: retain the best-performing (highest performance) coordinate set as a safeguard, and randomly select other sets for genetic crossover or mutation to generate the next generation of solutions, with the goal of producing better solutions via crossover or mutation in the next iteration.
Step 5: Repeat Steps 2 to 4 for the next generation of solutions until the performance evaluation value exceeds the set threshold or the maximum number of iterations is reached.

搜集汇总

数据集介绍

特点

该数据集为钢板切割领域的智能套料算法数据，包含15081条记录，每日更新，用于实现加工零件的全自动标准化套料，提高钢板利用率和生产效率。数据集采用遗传优化算法进行智能排版，通过交叉和变异迭代寻找最优解。