摘要
装配序列规划作为装配工艺中的核心内容,对实现生产自动化、装备装拆有着十分重要的意义,是目前国内外CIMS及装备维修领域的研究热点.提出并实现了一种用神经网络来实现自动装配规划的方法.该方法首先利用装配联系矩阵、装配联系图、惩罚矩阵来表达零件之间优先关系、装配代价,然后利用BP神经网络来求解满足此约束条件的最佳产品装配序列.
As the key of assembly process,Assembly Sequence Planning（ASP）plays an important role in the automatic production and assembly of the equipment.Moreover,it becomes the hotpot in the CIMS and the equipment maintenance field both home and aboard.An algorithm based on neural networks is presented to solve this NP-hard problem.Firstly the problem is defined by combining three techniques：assembly relation matrix,the assembly relation graph and the penalty matrix.And then, a modified Back-propagation Neural Networks（BPNN）algorithm is presented to optimize the solution process of ASP.Simulation results show significant performance in terms of sequence correctness subject to assembly constraints.
出处
《武汉理工大学学报（交通科学与工程版）》
2010年第5期1053-1056,共4页
Journal of Wuhan University of Technology（Transportation Science & Engineering）
基金
国家863计划项目资助(批准号:2002AA7170)
关键词
装配序列规划
CIMS
BP神经网络
assembly sequence planning
CIMS
back-propagation neural networks
作者简介
张晶（1980-）：男，博士生，主要研究领域为装备保障、计算机图形学
