基于递归循环的层次化爆炸图自动生成方法

提出以递归循环和深度优先搜索层次化装配序列的策略自动生成复杂产品层次化爆炸图的方法。为使设计结构符合装配规划的要求,设计了可方便定义子装配体和编辑装配树的装配结构重构功能。为避免重复的干涉检测,研究干涉矩阵合成方法,通过对原始装配关系信息的整合,柔性化生成面向层次化结构的装配关系矩阵。为降低大规模序列规划问题的复杂性,将整体规划任务分解为若干个及若干层级子任务,子任务内采用多规则筛选算法,以基于干涉矩阵的几何可行性为前提,并行性、连续性、稳定性和方向性等为优化筛选指标,循环构建序列。提出递归爆炸算法,循环对层次化装配序列进行深度优先搜索,实时计算已爆炸零部件的累积包围盒,从而确定待爆炸零件位移矢量,自动生成间隔均匀有序、结构紧凑且空间层次感强的爆炸图。开发了基于Siemens NX的装配规划系统Auto Assem,以减速器及汽车部件实例验证了方法的有效性。

Automatic Generation of Hierarchical Exploded View Based on Recursive Iteration Method

A method for generating hierarchical exploded view automatically for complex product using the strategy of recursive iteration and depth-first searching hierarchical assembly sequences is proposed. To meet the requirements of assembly sequence planning (ASP), assembly reconstruction function that can conveniently define subassembly and edit assembly tree is designed. To avoid repetitive interference detection, the method of merging interference matrix based on assembly tree is studied, which can flexibly generate hierarchical structure-oriented assembly relation matrices by integrating the original ones. To reduce the complexity of large-scale ASP, the whole ASP task is divided into sub-tasks of hierarchies according to assembly tree. Within each sub-task, multi-rule screening (MRS) algorithm is applied to circularly construct a sequence, taking interference matrix-based geometric feasibility as precondition while parallelism, continuity, stability and directionality as optimization screening criterions. The proposed recursive explosion algorithm iterates depth-first searching hierarchical assembly sequences and real-time calculating the accumulative bounding box of exploded components to decide the displacement vector of each exploding component. The generated exploded view is full of sense of hierarchy, uniform in interval and compact in structure. An assembly planning system “AutoAssem” is developed on Siemens NX, and a gear reducer and big parts of a car are taken as examples to verify the effectiveness of the method and algorithm.