基于改进自适应和声遗传算法的装配序列优化研究  

Research of optimized assembly sequence based on improved adaptive harmony genetic search algorithm

作　　者：赵鸿飞[1] 张琦[1] 朱春生[1] 王海涛[1] 李焕良[1] 

机构地区：[1]解放军理工大学野战工程学院,南京210007

出　　处：《计算机应用研究》2013年第8期2357-2359,2363,共4页Application Research of Computers

基　　金：江苏省青年科学基金资助项目(BK2012061)

摘　　要：在考虑装配工具及零件装配方向改变对装配成本影响的同时,增加装配体约束稳定性影响作为装配序列优化评价要素,构建装配成本模型。基于装配体几何约束关系,建立装配方向约束矩阵和连接关系矩阵,并依据专家知识得到零件所对应工具的映射表,通过矩阵扫描得到装配成本惩罚次数,并与惩罚系数加权得到装配成本函数值。在和声算法中引入自适应的参数调节方式及遗传算子,使和声算法能够解决离散型数学问题,并较好地改善了和声算法方向性差的缺点,通过求解惩罚函数最小值,得到其所对应的最优装配序列解,最后结合实例对比,验证了该方法有的效性。The mathematical model of assembly cost is constructed by take not only the change of tool and assembly direction as the assessment elements of optimized assembly sequence, but also consider the stability effect of assembly. The constraint matrix of assembly direction and matrix of component＇ s connection relation are constructed based on the geometrical- restric- tion- relation between the parts. According to the expert knowledge, this paper constituted the mapping table between tool and parts, and got the number of penalty by matrix scan, then by weighting with the quotiety of penalty, achived the function-val- ued of assembly cost. The harmony search algorithm could solve discrete math problems by introduces the adjust method of a- daptive parameter and genetic operator, and improved the directionality of harmony search. By contrast with the other algo- rithms, it proved the efficiency.

关 键 词：自适应和声算法 装配成本模型 序列优化 遗传算子 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]