基于离散布谷鸟搜索算法的拆卸序列规划方法

A Disassembly Sequence Planning Method Based on Discrete Cuckoo Search Algorithm

文章提出一种基于离散布谷鸟搜索算法的拆卸序列规划性方法,该方法以赋权拆卸混合图模型为理论基础,并以此为基础建立了可拆卸条件。首先,建立离散布谷鸟搜索算法和拆卸序列规划之间的映射关系;其次,基于最优拆卸方向分层筛选顶点集合组成初始种群;再次,利用离散Levy飞行对鸟巢位置进行变换,利用离散巢寄生行为进行鸟巢的局部调整。在达到最大迭代次数后,对属性值比较好的几个鸟巢进行解码并判断可行性,最终得到最优可行拆卸序列。以管路模型为例,分别利用离散布谷鸟搜索算法、遗传蝙蝠算法和粒子群优化算法进行了拆卸序列规划。经过比较后发现,该方法求解出的拆卸序列质量较高。

A new disassembly sequence planning(DSP)method based on discrete cuckoo search(DCS)algorithm is proposed.This method is based on the disassembly weighted hybrid graph model,which the detachable conditions are established on.Firstly,the mapping relationship between the DCS algorithm and the disassembly sequence planning is established.Secondly,the initial population is composed of a hierarchical selection of vertex sets based on the optimal disassembly direction.Thirdly,the nest position is transformed by discrete Levy flight and the nest is locally adjusted by the discrete nest parasitism.After reaching the maximum number of iterations,the birds’nest with better attribute value are decoded and their feasibilities are judged.Finally,the optimal feasible disassembly sequence is obtained.Taking the pipeline model as an example,the DCS algorithm,genetic bat algorithm and particle swarm optimization algorithm are used to plan the disassembly sequence.It is found that this method is more efficient.