微小卫星的自适应蚁群算法装配序列规划  

Assemble Sequence Planning Based on the Self Adaptive ant Colony Algorithm in the Small Satellite

作　　者：孙康[1] 廖文和[1] 

机构地区：[1]南京航空航天大学机电学院,江苏南京210096

出　　处：《机械制造与自动化》2007年第3期24-27,29,共5页Machine Building & Automation

摘　　要：介绍了复杂的卫星零件装配通过构造装配矩阵将改进蚁群算法用于生成装配序列的方法。但是蚁群算法容易陷入局部最优,为了避免算法陷入局部最优或者停滞,运用自适应伪随机比率选择规则,通过信息素的局部和全局更新,帮助"蚁群"快速的寻找全局最优解,最后通过实例验证了算法的实用性。For complicate assembling of satellite, an improved ant colony optimization algorithm is introduced, in order to generate assembly sequence through an assembly relation matrix. But it is easy to fall in local peak, and an adaptive pseudo random proportional action choice rule is used to avoid local peak. Though the methods, which update the local and full pheromone, we could help the ants search the optimiurn result in the whole range quickly. Finally, an example illustrates the effectiveness of the strategy.

关 键 词：卫星装配 虚拟装配 装配序列规划 蚁群算法 

分 类 号：TP391.73[自动化与计算机技术—计算机应用技术]