摘要
元件贴装顺序是决定贴片机生产效率的关键问题。针对拱架型贴片机,采用了一种遗传算法。该遗传算法有其独特的染色体编码解码方式和交叉算子。算法中的染色体根据被贴装的印刷电路板由一条或多条子链组成,染色体的一个基因代表一个取贴循环。实验结果表明,该算法可以有效解决元件贴装顺序问题。同时,分析比较了三种传统交叉算子和该交叉算子的优化结果,表明这些传统交叉算子不能有效解决该问题。
Component placement sequence problems are critical problems determining the surface mounting machine's efficiency. A genetic algorithm (GA) is applied to solve them for gantry machines. The GA has its special chromosome coding, chromosome encoding and crossover operator. The chromosome of individual has one or several links, which depends on the assembled printed circuit board. One gene of a chromosome represents one pick-and-place cycle. Experiment results show that the GA can resolve the problem effectively. At the same time, A performance comparison between three traditional crossover operaters and the crossover operator are presented and analyzed. The results show that the traditional crossover operators cannot solve the problems effectively.
基金
上海市科委重点科技资助项目(011111063)。~~
关键词
元件贴装顺序问题
贴片机
遗传算法
印刷电路板
component placement sequence problem
surface mounting machine
genetic algorithm
printed circuit board
