论文部分内容阅读
基于矩形件三阶段约束排样问题(CTDC),提出基于束搜索的启发式算法优化排样方式、以快速生成同质块三阶段排样方式。采用动态规划确定段的价值。束搜索是一种剪枝的分支定界算法,节点用局部排样方式和余料来表示,对节点的分支,即填充余料。在每一层上选择高潜力的节点作为精英节点做进一步分支,其他节点直接删除不再回溯,这有利于提高算法效率。实验结果表明:算法生成的三阶段排样方式,排样价值高,切割工艺相对简单,且时间相对合理。
Based on the rectangular three-stage constrained nesting problem (CTDC), a heuristic algorithm based on beam search is proposed to optimize the nesting mode to rapidly generate the homogeneous nesting three-stage nesting mode. Use dynamic programming to determine the value of the segment. Bundle search is a pruning branch-and-bound algorithm. The nodes are represented by the local nesting method and the residual material. The branch of the node, that is, the filling of residual material. Select high potential nodes on each layer as the elite nodes for further branching. The other nodes are deleted directly and no longer backtracking, which is beneficial to improve the efficiency of the algorithm. The experimental results show that the three-stage nesting method generated by the algorithm has high nesting value and the cutting process is relatively simple, and the time is relatively reasonable.