针对大规模两重互异性矩形件组批排样方法  

Batch Packing Method for Large Scale Rectangular Pieces with Twofold Reciprocity

作　　者：邵柏岩 叶伯生[1] 王宏磊 梁广 SHAO Baiyan;YE Bosheng;WANG Honglei;LIANG Guang(School of Mechanical Science&Engineering,Huazhong University of Science and Technology,Wuhan 430074,China;College of Mathematics and Physics,Beijing University of Chemical Technology,Beijing 100029,China;不详)

机构地区：[1]华中科技大学机械科学与工程学院,武汉430074 [2]北京化工大学数理学院,北京100029 [3]哈尔滨工业大学深圳校区机电工程与自动化学院,深圳518055

出　　处：《组合机床与自动化加工技术》2023年第11期187-192,共6页Modular Machine Tool & Automatic Manufacturing Technique

基　　金：湖北省重点研发计划项目(2021BAA197)。

摘　　要：针对智能优化算法求解大规模矩形排样问题时,求解速度慢、稳定性差的弊端,提出了一种启发式算法,即带对偶项的降序有限首次适应算法。经实验验证,算法在测试集上的求解时长在100 ms内,求解出的板材利用率可达95%左右,相较于现有启发式算法有2%的提升。此外,针对个性化订单矩形件材料强互异性的情况,提出了一种考虑材料相似度的二阶段贪心算法。定量描述订单间材料的相似度,最大程度的将相似度高的订单组批进同一生产批次来提高板材利用率。实验表明,用该算法对订单进行组批优化后再进行排样求解出的板材利用率相较于未考虑材料相似度有12%的提升。Aiming at the disadvantages of slow solution speed and poor stability when intelligent optimization algorithm solves large-scale rectangular layout problem,a descending finite first fit heuristic algorithm with dual term is proposed.It has been verified by experiments that the solution time of the algorithm on the test set is within 100 milliseconds,and the utilization rate of the plate can reach about 95%,which is 2%higher than the existing heuristic algorithm.In addition,a two-stage greedy algorithm considering the similarity of materials is proposed for the case of strong heterogeneity of materials for personalized order rectangular parts.Quantitatively describe the similarity of materials between orders,and batch the orders with high similarity into the same production batch to the greatest extent to improve the utilization rate of plates.Experiments show that using this algorithm to group and optimize orders,and then carry out layout to solve the board utilization rate has a 12%increase.

关 键 词：矩形排样优化 组批优化 整数规划 启发式算法 材料相似度 

分 类 号：TH162[机械工程—机械制造及自动化] TG502[金属学及工艺—金属切削加工及机床]