引用本文:陈铁梅,罗家祥,胡跃明.基于蚁群–混合蛙跳算法的贴片机贴装顺序优化[J].控制理论与应用,2011,28(12):1813~1820.[点击复制]
CHEN Tie-mei,LUO Jia-xiang,HU Yue-ming.Mounting sequence optimization on surface mounting machine using ant-colony algorithm and shuffled frog-leaping algorithm[J].Control Theory and Technology,2011,28(12):1813~1820.[点击复制]
基于蚁群–混合蛙跳算法的贴片机贴装顺序优化
Mounting sequence optimization on surface mounting machine using ant-colony algorithm and shuffled frog-leaping algorithm
摘要点击 2418  全文点击 1623  投稿时间：2010-07-30  修订日期：2011-05-16
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/j.issn.1000-8152.2011.12.CCTA100878
  2011,28(12):1813-1820
中文关键词  贴装顺序优化  蚁群算法  混合蛙跳算法
英文关键词  component mounting sequence optimization  ant-colony algorithm  shuffled frog-leaping algorith
基金项目  国家自然科学基金资助项目(60835001, 60804053, 61105081); 教育部重点科研基金资助项目(200805611065); 广东省科技攻关重大资助项目(912220500017).
作者单位E-mail
陈铁梅* 广东商学院 信息学院
华南理工大学 自动化科学与工程学院 精密电子制造装备教育部工程研究中心 		mei57726@tom.com 
罗家祥 华南理工大学 自动化科学与工程学院 精密电子制造装备教育部工程研究中心  
胡跃明 华南理工大学 自动化科学与工程学院 精密电子制造装备教育部工程研究中心  
中文摘要
      针对喂料器的位置确定的条件下, 研究拱架式贴片机的元器件贴装顺序优化问题. 建立了新的拱架式贴片机贴装顺序的数学模型. 针对问题的路径寻优特点, 把混合蛙跳算法与蚁群算法相融合, 实现对贴片机的元件贴装顺序优化问题的求解. 在算法中提出了适应于贴片机实际贴装情况的分段启发函数、分段信息素以及信息素的分段更新策略等多种改进方法. 为验证算法有效性, 以20块实际生产的PCB为实例进行了测试. 实验结果表明, 算法具有较好的求解精度和全局搜索能力, 与文献中的单一混合蛙跳算法相比, 平均效率提高了7.89%; 与蚁群算法相比, 平均效率提高了3.79%.
英文摘要
      The component mounting sequence optimization of the surface mounting machine is considered under the condition that the feeder allocations are known. A new mathematical model of mounting sequence is specifically built for arch surface mounting machines. Based on the characteristics of searching for the optimal path, a new hybrid algorithm of ant-colony algorithm merged with the shuffled frog-leaping algorithm is proposed to solve the problem. According to the actual mounting situation, a few improved methods are proposed in the algorithm, such as the segmented heuristic function, the segmented pheromone, and the pheromone update strategy. To verify the efficiency of the algorithm, component mounting experiments of 20 printed-circuit-boards(PCBs) are tested. The results show the algorithm has higher accuracy in solving the problem, and in searching the optimal path. It provides an improvement in average efficiency 7.89% over the single shuffled frog-leaping algorithm, and 3.79% over the single ant-colony algorithm.